Structure-Function Analysis of Immunity Proteins of Pediocin-Like Bacteriocins: C-Terminal Parts of Immunity Proteins Are Involved in Specific Recognition of Cognate Bacteriocins

Abstract

The immunity proteins of pediocin-like bacteriocins show a high degree of specificity with respect to the pediocin-like bacteriocin they recognize and confer immunity to. The aim of this study was to identify regions of the immunity proteins that are involved in this specific recognition. Six different hybrid immunity proteins were constructed from three different pediocin-like bacteriocin immunity proteins that have similar sequences but confer resistance to different bacteriocins. These hybrid immunity proteins were then tested for their ability to confer immunity to various pediocin-like bacteriocins. The specificities of the hybrid immunity proteins proved to be similar to those of the immunity proteins from which the C-terminal halves were derived, thus revealing that the C-terminal half of immunity proteins for pediocin-like bacteriocins contains a domain that is involved in specific recognition of the bacteriocins they confer immunity to. Moreover, the results also revealed that the effectiveness of an immunity protein is strain dependent and that its functionality thus depends in part on interplay with strain-dependent factors. To further investigate the structure-function relationship of these immunity proteins, the enterocin A and leucocin A immunity proteins (EntA-im and LeuA-im) were purified to homogeneity and structurally analyzed under various conditions by Circular dichroism (CD) spectroscopy. The results revealed that both immunity proteins are α-helical and well structured in an aqueous environment, the denaturing temperature being 78.5°C for EntA-im and 58.0°C for LeuA-im. The CD spectra also revealed that there was no further increase in the structuring or α-helical content when the immunity proteins were exposed to dodecylphosphocholine micelles or dioleoyl-l-α-phosphatidyl-dl-glycerol (DOPG) liposomes, indicating that the immunity proteins, in contrast to the bacteriocins, do not interact extensively with membranes. They may nevertheless be loosely associated with the membrane, possibly as peripheral membrane proteins, thus enabling them to interact with their cognate bacteriocin.

Gram-positive bacteria produce ribosomally synthesized antimicrobial peptides, often termed bacteriocins. A dominant and important group of these bacteriocins consists of the pediocin-like bacteriocins produced by lactic acid bacteria (LAB) (12, 22, 26). The pediocin-like bacteriocins are cationic, display antilisteria activity, and kill target cells by permeabilizing the cell membrane (8, 21). They all contain between 37 and 48 residues, and they have very similar primary structures. The C-terminal parts of these bacteriocins are important determinants of their target cell specificities (13). At least 20 pediocin-like bacteriocins have been characterized (12, 22, 26; see reference 15 for original references), and the three-dimensional structures of some of these have been analyzed by nuclear magnetic resonance spectroscopy and mutagenesis (14, 15, 17, 33, 38).

Genes encoding pediocin-like bacteriocins are cotranscribed with and/or in close vicinity to a gene encoding a cognate immunity protein which protects the bacteriocin producers from their own bacteriocin (2, 10, 19, 27, 28, 37). Immunity genes that are not associated with bacteriocin genes have also been described (8). The primary structures of at least 17 immunity proteins of pediocin-like bacteriocins have been deduced from DNA sequences (see reference 16 for original references). These immunity proteins consist of 88 to 115 amino acid residues, and they display 5 to 85% sequence similarities (11). The proteins show remarkably high degrees of specificity with respect to the bacteriocins they recognize, although some immunity proteins may recognize and render cells immune to a few pediocin-like bacteriocins in addition to their cognate bacteriocin (16). The mode of action of immunity proteins is not known. Functional analysis of the two immunity proteins conferring resistance to mesentericin Y105 and carnobacteriocin B2 has shown that these proteins are localized intracellularly (10, 27) and are therefore not likely to act by directly preventing the extracellular bacteriocin from binding to the membrane (27). It has been suggested that immunity proteins may act by disturbing the interaction between the bacteriocin and a (putative) membrane-located bacteriocin receptor (25, 27, 36). Whereas it is known that the C-terminal part of the pediocin-like bacteriocins is an important determinant of the target cell specificity, the basis of the immunity proteins' specificity is unknown.

The aim of this study was to identify regions of the immunity proteins of pediocin-like bacteriocins that are, directly or indirectly, specifically recognized by (cognate) bacteriocins. This was done by combining the N-terminal and C-terminal halves of three different immunity proteins that confer immunity to different pediocin-like bacteriocins and thereafter testing the ability of these hybrid immunity proteins to confer immunity to various pediocin-like bacteriocins. To further investigate the structure-function relationship of immunity proteins, the cognate immunity proteins for enterocin A (EntA-im) and leucocin A (LeuA-im) were purified to homogeneity and, by using Circular dichroism (CD) spectroscopy, structurally analyzed under various environmental conditions.

MATERIALS AND METHODS

Bacterial strains, plasmids, and media.

The bacteriocin producer strains used in this study were Lactobacillus curvatus LTH1174, producing curvacin A (31), Enterococcus faecium CTC492, producing enterocin A and enterocin B (4), Leuconostoc mesenteroides 6, producing leucocin A and C (35), and Pediococcus acidilactici LMG 2351, producing pediocin PA-1 (23). Sakacin P was produced by the two-plasmid Lactobacillus sake LB790(pSAK20/pSSP2) heterologous expression system described earlier (3).

The indicator strains used for bacteriocin activity measurements were L. sake NCDO 2714 and Enterococcus faecalis NCDO 581 containing a plasmid with or without an inserted (hybrid) immunity gene. All LAB strains were grown in MRS broth (Oxoid) at 30°C. Agar plates for all LAB transformants were made by adding 1.5% agar to MRS broth. Selective antibiotic concentrations used for L. sake were 2 μg of erythromycin/ml for initial selection and 5 μg/ml for normal growth, whereas they were 5 and 10 μg/ml, respectively, for E. faecalis. The LAB-Escherichia coli shuttle vector pMG36e (34) containing the strong constitutive P32 promoter was used to express the (hybrid) immunity genes.

Plasmid isolation and transformation.

All plasmid isolations from both E. coli and LAB were done using the QIAprep spin miniprep kit (Qiagen). Lysis of LAB strains was performed by adding lysozyme to resuspension buffer (Qiagen) to a final concentration of 5 mg/ml. The Bio-Rad gene pulser was used for all electroporations. L. sake was made electrocompetent and transformed as described previously (1). After growth from an optical density at 610 nm (OD₆₁₀) of 0.1 to 0.6 in the presence of 2% (wt/vol) glycine, the cells were washed in 1 mM MgCl₂ followed by a wash in 30% (wt/vol) polyethylene glycol 1500 (molecular mass range, 1,300 to 1,600 Da) prior to electroporation. Electrocompetent E. faecalis was produced by growing them overnight in M17 containing 7% (wt/vol) glycine and 0.5 M sucrose at 37°C. The cells were thereafter washed twice in 0.5 M sucrose containing 10% (vol/vol) glycerol (30). The amount of expression plasmid in transformed LAB strains was determined by quantifying spectrophotometrically the amount of plasmid in plasmid preparations from the LAB strains by use of the Bio Photometer instrument (Eppendorf). The plasmid copy number was estimated to be between 30 and 80 copies per cell. E. coli DH5α cells were made competent by using the CaCl₂ method (29) and transformed by heat shock (42°C, 2 min).

PCR and DNA sequencing.

Eurogentec (Belgium) produced all oligonucleotide primers used in this study. The hybrid immunity genes were amplified by PCR by following a standard amplification protocol on a PTC-200 Peltier thermal cycler (MJ Research), using Vent DNA polymerase (New England Biolabs). PCR mixtures contained 50 ng of plasmid DNA, 125 ng of each primer, deoxynucleoside triphosphates (Amersham Biosciences) to a final concentration of 0.05 M of each nucleotide, 10× reaction buffer (New England Biolabs), and 2 U of Vent DNA polymerase (New England Biolabs). Amplification proceeded through 30 cycles after a 2-min hot start at 95°C with a program including denaturation (1 min at 92°C), primer annealing (45 s at 52°C), and polymerization (1 min at 72°C).

The DNA sequences of the cloned entA-im and leuA-im genes and the hybrid immunity genes were verified by automated DNA sequence determination, using a MegaBase DNA analysis system and the DYEamic ET dye terminator cycle sequencing kit (Amersham Biosciences).

Construction of the entA-im and leuA-im expression plasmids and of plasmids containing hybrid immunity genes.

PCR fragments containing the entA-im and leuA-im gene with flanking SapI and PstI sites were cloned into the vector, pTYB11 (New England Biolabs). Specific primers for PCR amplification were synthesized (listed in Table 1), and the plasmids pEAIM and pLIM were used as templates in the PCRs. The amplification products were ligated into the SapI and PstI sites of pTYB11. These constructs allowed the fusion of the target protein's N terminus to the intein tag, and the proteins are obtained without any vector-derived residues.

TABLE 1.

Oligonucleotide primers used in the PCRs

Name	Primer sequence
PmgsekF	5′-GAAATGGCAATCGTTTCAG-3′
PmgsekR	5′-CTTTACCAACTGTCTTGG-3′
EntF	5′-TGGTTGCTCTTCCAACATGAAAAAAAATGCTAAGC-3′
EntR	5′-GACTGCAGCATTAAAATTGAGATTTATCTCCATAAT-3′
LeuF	5′-GGTGGTTGCTCTTCCACCTTGAGAAAAAATAACAAT-3′
LeuR	5′-GGTGGTCTGCAGTCATTATCTTTCAAAGATACTATAAAC-3′
Entleuimma	5′-ATTCGTATATAATTGACCAATCTATTTATTAAACGACTG-3′
Leuentimma	5′-AAATATAGATAATTCACCAATCGGTTAATCAATGCTT-3′
Orfyleuimma	5′-ATTCGTATATAATTGACCAAGCGGTTGACTAACACTTC-3′
Leuorfyimma	5′-CTATAAATATAATTTGCGAGTCGGTTAATCAATGCTTC-3′
Entorfyimma	5′-CTATAAATATAATTTGCGAGTCTATTTATTAAAGGACTG-3′
Orfyentimma	5′-AAATATAGATAATTCACCAAGCGGTTGACTAACACTTC-3′

^aOligonucleotides primer containing the site of recombination for the hybrid immunity genes.

PCR fragments containing the hybrid immunity genes, entA/orfY-im, orfY/entA-im, entA/leuA-im, leuA/entA-im, orfY/leuA-im, and leuA/orfY-im (see Fig. 1), with flanking SacI and XbaI sites, were constructed by using the PCR megaprimer method and oligonucleotides listed in Table 1 and cloned into the SacI and XbaI sites of pMG36e. The plasmids used as templates in the PCRs, pEAIM, pLIM, and pORFY, are described in reference 16.

FIG. 1.

Sequence alignment of leucocin A immunity protein (LeuA-im) (18), enterocin A immunity protein (EntA-im) (4), and OrfY immunity protein (OrfY-im) (7). Regions of sequence similarity are indicated by black boxes. The conserved region is underlined, and the arrows (between the arginine and leucine residues) show the division point between N-terminal and C-terminal parts.

Expression and purification of the enterocin A and leucocin A immunity proteins.

The enterocin A and leucocin A immunity proteins were overexpressed and purified by using the IMPACT-CN system (New England Biolabs) according to the manufacturer's protocol. The E. coli DH5α strain was used for preparation of plasmids and cloning, and the E. coli strain ER2566 (New England Biolabs) transformed with either pTYB11/entA-im or pTYB11/leuA-im was used for expression of the fusion proteins. E. coli ER2566 containing pTYB11/entA-im or pTYB11/leuA-im was grown at 37°C in Luria-Bertani medium containing 100 μg of ampicillin/ml to an OD₆₀₀ of 0.6 to 0.8. Isopropyl-β-d-thiogalactoside was added to a final concentration of 0.5 mM, and induction was conducted at 16°C overnight. The cells were harvested by centrifugation (5,000 × g for 10 min), resuspended in column buffer (20 mM Tris-HCl, 500 mM NaCl, 1 mM EDTA) containing 0.1% (vol/vol) Triton X-100, and broken by sonication (10 pulses at 10 s with amplitude 100, intervals of 20 s). The clarified cell extract obtained by centrifugation was directly applied onto a chitin column equilibrated with 10 volumes of the column buffer. The column was subsequently washed with 10 volumes of the column buffer. Induction of the on-column cleavage was conducted by quickly flushing the column with three volumes of the column buffer containing 50 mM dithiothreitol. The column was left at 18°C for 40 h before elution of the target protein in three volumes of column buffer.

The protein was concentrated by using ultrafiltration Centricon concentrators (cutoff, 3,000; Millipore) and further purified by chromatography on a μRPC SC 2.1/10 C₂C₁₈ column (Amersham Biosciences) using the SMART chromatography system (Amersham Biosciences) and water-2-propanol, containing 0.1% trifluoroacetic acid, for the mobile phase. The primary structure was confirmed by mass spectrometry on a Voyager-DE RP matrix-assisted laser desorption time-of-flight mass spectrometer (PerSeptive Biosystems); 3,5-dimethoxy-4-hydroxy-cinnamic acid was used as the matrix.

Production, synthesis, and purification of bacteriocins.

Enterocin A, leucocin A, leucocin C, pediocin PA-1, curvacin A, and sakacin P were purified to homogeneity from 0.5- or 1.0-liter cultures of their producer strains by cation-exchange chromatography and reverse-phase chromatography, as described previously (32). The purity of the bacteriocins was analyzed by analytical chromatography on a μRPC SC 2.1/10 C₂C₁₈ column (Amersham Biosciences) using the SMART chromatography system (Amersham Biosciences) and water-2-propanol containing 0.1% (vol/vol) trifluoroacetic acid for the mobile phase. Primary structure was confirmed by mass spectrometry on a Voyager-DE RP matrix-assisted laser desorption time-of-flight mass spectrometer (PerSeptive Biosystems); 3,5-dimethoxy-4-hydroxy-cinnamic acid was used as the matrix.

Bacteriocin assay.

Bacteriocin activity was measured by using a microtiter plate assay system, essentially as described previously (24). Each well of a microtiter plate contained 200 μl of culture medium with bacteriocin fraction at twofold dilutions and an indicator strain at an OD₆₁₀ of about 0.01 (inoculated from a 16- to 20-h overnight culture at 30°C). The microtiter plate cultures were incubated overnight (14 to 16 h) at 30°C, after which growth of the indicator strain was measured spectrophotometrically at 610 nm with a microtiter plate reader. The MIC was defined as the concentration of bacteriocin that inhibited growth of the indicator strain by 50%. The MICs that are presented are the results of at least five independent measurements and had standard deviations of less than 50% of the value. Transformed indicator strains were grown in the presence of 5 μg of erythromycin per ml to ensure that the plasmid containing the immunity gene was maintained in the cells. The protection pattern of each of the (hybrid) immunity proteins was judged by comparing MICs for a strain expressing a hybrid immunity gene with MICs for the same strain containing the unmodified pMG36e plasmid.

Liposome preparation.

Single-bilayer phospholipid vesicles were prepared essentially as described previously (5). Eight micromoles of DOPG (Sigma) dissolved in chloroform was carefully dried under a stream of ultrapure nitrogen. The dried lipids were redissolved in 1 volume of absolute ethanol and dried again. Subsequently, the lipids were redissolved in 200 μl of absolute ethanol and slowly (about 100 μl/min) and at constant speed injected into 4 ml of 10 mM potassium phosphate (pH 7.4) at room temperature. The ethanol was removed by dialysis against 10 mM potassium phosphate (pH 7.4).

CD.

CD spectra were recorded by using a Jasco J-810 spectropolarimeter (Jasco) calibrated with ammonium d-camphor-10-sulfonate (Icatayama Chemicals). Measurements were performed at 23°C by using a quartz cuvette (Starna) with a path length of 0.1 cm. All the measurements were performed with a protein concentration of 0.10 mg/ml in either 10 mM potassium phosphate buffer (pH 7.4), 16 mM dodecylphosphocholine (Sigma) micelles, or 1.4 mM DOPG liposomes. Samples were scanned five times at 20 nm/min with a bandwidth of 1 nm and a response time of 1 s, over the wavelength range 190 to 260 nm. The data were averaged, and the spectrum of a protein-free control sample was subtracted. The α-helical content was calculated after smoothing (means-movement method; convolution width, 13) from ellipticity data, using the neural network program CDNN version 2.1 and the supplied neural network based on the 33-member basis set (6). Thermal denaturation curves were determined by recording the change in CD signal at 222 nm during heating. The temperature was controlled using a Peltier-type temperature control system (TPC-423S/L; Jasco) and a heating rate of 1°C/min. The protein concentration was 0.10 mg/ml in 10 mM potassium phosphate buffer (pH 7.4), and the path length was 0.1 cm. After baseline correction, the unfolding curve was smoothed (means-movement method; convolution width, 13) and normalized, and the apparent melting temperature (T_m) was determined from the transition midpoint visible in the first derivative of the unfolding curve, using the computer program Origin 7.0 (OriginLab Corporation). All measurements were conducted at least twice.

RESULTS AND DISCUSSION

Hybrid immunity proteins confer resistance to bacteriocins.

Six different hybrid immunity proteins were constructed from three different immunity proteins (EntA-im, LeuA-im, and OrfY-im) that have similar sequences (i.e., belong to the same group) but confer resistance to different bacteriocins (16). EntA-im and LeuA-im are the cognate immunity proteins for enterocin A and leucocin A, respectively, whereas OrfY-im is an immunity protein for which a cognate bacteriocin has not been identified (7, 16). All three immunity proteins have a conserved sequence of eight amino acid residues (L-X-N-R-L-X-N-Y; six residues are identical) situated approximately in the middle of the proteins (Fig. 1). Each of the six hybrid proteins was constructed by combining the N-terminal half of one immunity protein with the C-terminal half of another. The N-terminal half starts with residue 1 and ends with the arginine residue in the conserved region. The C-terminal half starts with the leucine residue that succeeds the arginine residue in the conserved region and extends to the C terminus of the protein. The following six hybrid proteins were constructed: EntA/LeuA-im, EntA/OrfY-im, LeuA/EntA-im, LeuA/OrfY-im, OrfY/EntA-im, and OrfY/LeuA-im. For these hybrids, the immunity protein from which the N-terminal half is derived is indicated before the slash, and the immunity protein from which the C-terminal half is derived is indicated after the slash.

In order to determine if the hybrid proteins are active and display specificity with respect to the bacteriocins they confer immunity to, EntA-im, LeuA-im, OrfY-im, and the six hybrid immunity proteins were expressed in bacteriocin-sensitive L. sake NCDO 2714 and E. faecalis NCDO 581. Subsequently, the bacteriocin sensitivities of the two strains containing plasmid with an inserted (hybrid) immunity gene were tested and compared to the bacteriocin sensitivities of the corresponding strains that contained plasmid without immunity gene. Two different strains were used in these experiments, since it has been shown that the effectiveness of some immunity proteins may be strain dependent (16).

The hybrid immunity proteins EntA/LeuA-im, LeuA/EntA-im, and OrfY/LeuA-im functioned in both strains, whereas the hybrid OrfY/EntA-im functioned only in the E. faecalis strain (Table 2). The two hybrids that contained the C-terminal half from OrfY-im (EntA/OrfY-im and LeuA/OrfY-im) did not function in either strain (Table 2). They may perhaps be inactive per se, or they may possibly be poorly expressed in the transformed cells despite the fact that all transformed strains contained approximately equivalent amounts of the expression plasmid (quantified spectrophotometrically as described in Materials and Methods) with an inserted immunity gene. Nevertheless, these two hybrids could not as a consequence be included in the subsequent analysis for identifying regions in the immunity proteins involved in recognition of cognate bacteriocins.

TABLE 2.

Bacteriocin sensitivities^a of strains expressing various (hybrid) immunity genes

Immunity protein	Fold increase in MIC of:
	Bacteriocin tested against L. sake NCDO 2714		Bacteriocin tested against E. faecalis NCDO 581
	Enterocin A	Leucocin A	Enterocin A	Leucocin A
EntA-im	250	2	>250	4
LeuA-im	2	32	10	>500
OrfY-im	125	32	65	>500
EntA/LeuA-im	0	16	2	16
LeuA/EntA-im	65	2	32	0
OrfY/LeuA-im	0	16	4	>500
OrfY/EntA-im	0	0	>250	0
LeuA/OrfY-im	0	0	0	0
EntA/OrfY-im	0	0	0	0

^aThe results are presented as the fold increase in MIC observed for strains (L. sake NCDO2714 and E. faecalis NCDO581) expressing a hybrid immunity gene, relative to MICs for strains containing only the control plasmid (pMG36e).

Interestingly, OrfY-im rendered the L. sake strain markedly more immune to enterocin A than to leucocin A, whereas the opposite was the case when OrfY-im was expressed in the E. faecalis strain (Table 2). This indicates, as has been noted earlier in reference 16, that the effectiveness of an immunity protein is strain dependent and that the functionality of immunity proteins thus depends in part on interplay with strain-dependent factors. This might explain why the hybrid immunity protein OrfY/EntA-im functioned in the E. faecalis strain but not in the L. sake strain (Table 2). An alternative and perhaps more likely explanation is that there is a much higher expression of this hybrid immunity protein in the E. faecalis strain than in the L. sake strain, although the two strains contained about equal amounts of the expression plasmid with the inserted immunity gene (estimated copy number per cell was in both cases approximately 80). No protection was observed against curvacin A, sakacin P, pediocin PA-1, or leucocin C (results not shown).

The C-terminal halves of the immunity proteins recognize cognate bacteriocins.

The hybrid immunity protein EntA/LeuA-im conferred immunity to leucocin A but not to enterocin A (Table 2) and was thus similar to the parental immunity protein LeuA-im. In contrast, the hybrid immunity protein LeuA/EntA-im conferred immunity to enterocin A but not leucocin A (Table 2) and was thus similar to the parental immunity protein EntA-im. The same trend, i.e., that the specificity of the hybrid immunity proteins is similar to that of the immunity proteins from which the C-terminal region is derived, was also seen for the hybrid immunity proteins OrfY/LeuA-im and OrfY/EntA-im. OrfY/LeuA-im conferred immunity to leucocin A but little or no immunity to enterocin A, whereas OrfY/EntA-im conferred immunity in E. faecalis to enterocin A but not to leucocin A (OrfY/EntA-im was inactive in the L. sake strain [Table 2]). Taken together, the results show that the C-terminal halves of immunity proteins for pediocin-like bacteriocins contain a domain that is involved in specific recognition of the bacteriocins they confer immunity to.

Purification of EntA-im and LeuA-im.

EntA-im and LeuA-im were purified to homogeneity, with about 1 mg of each protein being obtained from 1 liter of culture. For both immunity proteins, only one major symmetrical optical density peak eluted, at approximately 30 and 50% isopropanol for EntA-im and for LeuA-im, respectively, in the final reverse-phase column chromatography step. Analysis of the proteins obtained in this final step by mass spectrometry revealed molecular weights of 12,217 for EntA-im and 12,974 for LeuA-im. The molecular weights were in both cases identical to the theoretical molecular weights for these proteins. For EntA-im, the identity of the protein was also confirmed by sequencing (automated Edman degradation) the 26 N-terminal amino acid residues.

CD structural analysis: immunity proteins are well structured α-helical proteins in aqueous environments.

CD structural analysis of EntA-im and LeuA-im revealed that both proteins are α-helical (an α-helical content of about 50%) and well structured in an aqueous environment (10 mM potassium phosphate buffer, pH 7.4) (Fig. 2). The results are consistent with computer analysis (Predator) of the protein sequences, which indicates an α-helical content of 50% for EntA-im and 37% for LeuA-im. The computer analyses suggest, moreover, that at least two amphiphilic helical regions are present. The apparent melting temperatures of 78.5°C for EntA-im and 58.0°C for LeuA-im, determined from the transition midpoint in the CD spectrum (Fig. 3), indicate that the immunity proteins are structurally stable despite their small size (103 residues in EntA-im and 113 residues in LeuA-im).

FIG. 2.

CD spectra of the enterocin A (smooth line) and leucocin A (dashed line) immunity proteins (0.10 to 0.15 mg/ml) in 10 mM potassium phosphate buffer (pH 7.4 at 22°C). MRE, mean residual ellipticity.

FIG. 3.

Thermal denaturation curves of the enterocin A (smooth line) and leucocin A (dashed line) immunity proteins. The apparent fraction of folded proteins obtained by monitoring the CD values at 222 nm is shown as a function of temperature. The apparent melting temperatures, 78.5°C for EntA-im and 58.0°C for LeuA-im, were determined from the transition midpoints in the CD spectrum. MRE, mean residual ellipticity.

The fact that purified EntA-im readily forms crystals that diffract to better than 1.7 Å resolution (9) is consistent with the immunity proteins being well structured in an aqueous environment and indicates that they are not integral membrane proteins. Interestingly, the CD spectra revealed that there was no further increase in the structuring or α-helical content when EntA-im was exposed to dodecylphosphocholine micelles or DOPG liposomes (results not shown), also indicating that the immunity proteins, in contrast to the bacteriocins (17, 33, 38), do not interact extensively with membranes. This is consistent with their largely hydrophilic character and the finding that only a minor fraction (less than 1%) of the immunity protein for the pediocin-like bacteriocin carnobacteriocin B2 is associated with the cell membrane (27). This is in contrast, however, with the extensive membrane-interaction of the immunity protein for the non-pediocin-like class II bacteriocin lactococcin A (25, 36).

Conclusions.

The immunity proteins for the pediocin-like bacteriocins show a high degree of specificity in that they largely recognize and confer immunity only to their cognate bacteriocins and in some cases to a few bacteriocins that are closely related to the cognate bacteriocin (16). The extent to which an immunity protein specifically recognizes, interacts with (directly or indirectly), and confers immunity to a bacteriocin seems to depend on the C-terminal half of the bacteriocin (16). The C-terminal part of the pediocin-like bacteriocins is also the region that interacts with the hydrophobic part of the membrane (13, 14, 15, 20, 33), which suggests that immunity proteins and bacteriocins directly or indirectly interact in this part of the membrane. The present study shows that the C-terminal halves of the immunity proteins contain a domain that is involved in specific recognition of the cognate and related bacteriocins, most likely through direct or indirect interaction with the C-terminal halves of these bacteriocins, an interaction that presumably also involves the cell membrane, most likely the hydrophobic part. Although the immunity proteins do not appear to be integrated membrane proteins involved in extensive membrane interactions (in contrast to the immunity protein for the non-pediocin-like bacteriocin lactococcin A (25, 36), they may nevertheless be loosely associated with the membrane, possibly as peripheral membrane proteins, thus enabling them to interact with their cognate and related bacteriocins.