Previous Article | Next Article ![]()
Applied and Environmental Microbiology, July 2006, p. 4761-4766, Vol. 72, No. 7
0099-2240/06/$08.00+0 doi:10.1128/AEM.00436-06
Copyright © 2006, American Society for Microbiology. All Rights Reserved.
Department of Microbiology, Stellenbosch University, Stellenbosch, South Africa
Received 23 February 2006/ Accepted 30 April 2006
|
|
|---|
|
|
|---|
The majority of class IIa bacteriocins have a double-glycine sequence at the N terminus, which serves as a recognition signal for peptide procession and secretion (3, 16, 27, 29). ATP-binding cassette (ABC) transporters translocate the bacteriocin across the cell membrane (9). A few class IIa bacteriocins make use of a signal peptide instead of a double-glycine leader sequence (5, 28). The leader peptide is usually positively charged and has a hydrophobic core and cleavage region. The peptide is processed by a signal peptidase during translocation across the cell membrane (30). Enterocin P (5) and bacteriocin 31 (28), produced by E. faecium P13 and Enterococcus faecalis Y1717, respectively, are examples of sec-dependent class IIa bacteriocins. Other examples of sec-dependent peptides are the large heat-labile class III peptide enterolysin A, produced by E. faecalis LMG 2333 (20), and lactococcin 972, a two-peptide bacteriocin classified as class IIb (15). Two nonlantibiotics, divergicin A (31) and acidocin B (14), are also secreted by a signal peptide.
The purpose of this study was to characterize the bacteriocin produced by E. faecium T8 and compare it with other bacteriocins of the same class.
|
|
|---|
were cultured in Luria-Bertani (LB) broth (Biolab Diagnostics) on a rotating wheel at 37°C. Other strains included in the test panel were obtained from Laboratorium voor Microbiologie (LMG), University of Ghent, Ghent, Belgium, and our own culture collection. E. faecalis BFE1071 was isolated from the feces of minipigs, Lactobacillus salivarius 241 was isolated from the ileum of a piglet, Lactobacillus curvatus DF38 was isolated from salami, and Lactobacillus casei LHS3 was isolated from fortified wine. Bacteria were stored at 80°C in 40% (vol/vol) glycerol.
Identification of strain T8.
Strain T8 was isolated from children infected with human immunodeficiency virus. Strain T8 was identified to the genus level by its phenotypic characteristics (10) and to the species level by 16S rRNA gene sequencing. DNA was isolated using a High Pure PCR preparation kit from Roche Diagnostics (Indianapolis, Ind.) and amplified by PCR using the 16S rRNA gene primers and conditions described by Felske et al. (8). Amplified fragments were cloned into the pGEM-T Easy vector system (Promega, Madison, Wis.). Constructs were transformed into E. coli DH5
. Plasmid DNAs were isolated using a QIAprep Spin miniprep kit (QIAGEN, Valencia, Calif.). DNAs were sequenced at the Central Analytical Facility, Stellenbosch University, on an automatic sequencer (ABI Genetic Analyzer 3130Xl; Applied Biosystems SA), using BigDye Terminator chemistry (Biosystems, Warrington, England). All ligation reactions and transformations were performed according to standard procedures (2). Homology with sequences in GenBank was determined by using the BLAST program (1).
Antimicrobial spectrum of bacteriocin T8.
Enterococcus faecium T8 was cultured in 500 ml MRS broth (Biolab) for 24 h at 37°C. Cells were harvested (10,000 x g, 10 min, 4°C), and proteins were precipitated from the cell-free supernatant with 80% ammonium sulfate (22). The precipitate was collected by centrifugation (10,000 x g, 1 h, 4°C), and the pellet was resuspended in 10 ml ammonium acetate buffer (pH 6.5). The concentrated bacteriocin was dialyzed against 4 liters sterile distilled water, using a Spectra-Por membrane with a 1,000-Da cutoff (Spectrum Inc., CA). Antimicrobial activity was determined by using the agar spot test method (25) and the well diffusion method (26). The activity of the crude extract was expressed in arbitrary units per ml (AU/ml). One AU is defined as the reciprocal of the highest serial twofold dilution showing a clear zone of inhibition of the indicator strain (29). The crude extract was used to determine the spectrum of antimicrobial activity and to characterize bacteriocin T8.
Characterization of bacteriocin T8.
One milliliter of crude extract containing bacteriocin T8 was incubated for 2 h at 30°C in the presence of 0.5 mg ml1 (final concentration) amylase (Sigma Diagnostics, St. Louis, Mo.), proteinase K (Roche Diagnostics Corporation, Mannheim, Germany), pronase (Boehringer Mannheim GmbH, Germany), or pepsin (Boehringer Mannheim GmbH). The enzymes were inactivated (10 min at 80°C). A bacteriocin T8 crude extract not treated with enzymes was used as a control (11).
In a separate experiment, the crude extract of bacteriocin T8 was adjusted to pH values ranging from 2.0 to 12.0 (increments of 2 pH units) with sterile 1 M NaOH or 1 M HCl and incubated for 2 h at 37°C. The pH was then neutralized by slowly adding 1 M NaOH or 1 M HCl (11).
The cell-free supernatant of E. faecium T8 was incubated at 60°C or 100°C for 10, 30, and 90 min. The control was a cell-free supernatant of E. faecium T8 not subjected to heat treatment and kept at 8°C (11). All treated supernatants were tested for antimicrobial activity as described previously.
E. faecalis MDK2, isolated from patients diagnosed with vaginosis, was used as a sensitive strain in all bacteriocin activity tests.
The molecular size of bacteriocin T8 was determined by Tricine-sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (24). One half of the gel was stained with Coomassie blue R250 (Saarchem, Krugersdorp, South Africa). The other half was prewashed with sterile distilled water and overlaid with E. faecalis MDK2 (106 CFU ml1) to determine the position of bacteriocin T8 (29). A Rainbow protein molecular mass marker with fragments of 2.5 to 45.0 kDa (Amersham International, United Kingdom) was used.
Mode-of-action studies were performed to determine if the bacteriocin is bactericidal. Actively growing cells of E. faecalis MDK2 were inoculated (0.2% [vol/vol]) into BHI broth and incubated for 5 h at 37°C to mid-exponential growth phase. Bacteriocin T8 crude extract (6,400 AU ml1) was added to the culture, and cell density readings were recorded hourly at 600 nm for 8 h (11). The control was autoclaved (15 min at 121°C) bacteriocin T8.
Plasmid curing.
Actively growing cells of E. faecium T8 were inoculated into MRS broth and supplemented with acridine orange (Merck), novobiocin (Sigma), or SDS (Sigma) (21). The components were added at 0.5, 1.0, 2.0, 3.0, 4.0, and 5.0% (wt/vol). After 18 h of incubation at 37°C, cells from each treatment were plated onto MRS agar and incubated for a further 18 h. Colonies were replica plated onto MRS agar, and one set was overlaid with E. faecalis MDK2 embedded in 1% (wt/vol) soft agar (106 CFU ml1). The plates were incubated for 18 h at 37°C. Colonies with no inhibition zones were identified. Corresponding colonies were selected from replica plates, inoculated into MRS broth, and incubated for 18 h at 37°C. The cells were harvested (10,000 x g, 10 min, 4°C), and plasmid DNAs were isolated using a QIAGEN Plasmid Midi kit. Bacteriocin-producing cells of E. faecium T8 were used as a control.
Sequencing of plasmid T8.
Plasmid DNA isolated from E. faecium T8 was digested with HindIII (Roche) and ligated with T4 ligase (Roche) into pBluescript SK(+/) (Invitrogen, Paisley, United Kingdom). The construct was transformed into E. coli DH5
according to the method of Ausubel et al. (2). DNA was isolated from the transformants by using a QIAprep Spin miniprep kit (QIAGEN). After sequencing of the pBluescript construct, primers were designed to sequence the rest of the plasmid. Primers designed from sequenced fragments (Table 1, MDK1a and MDK1b) were used to amplify the DNA, which was then cloned into the pGEM-T Easy vector system (Promega). The constructs were transformed into E. coli DH5
. The DNA was isolated as described previously. Subsequent primers (Table 1, MDK2a and MDK2b) were designed from amplified DNA fragments, and the plasmid was sequenced by primer walking.
|
View this table: [in a new window] |
TABLE 1. Primers used in this study
|
|
|
|---|
|
View this table: [in a new window] |
TABLE 2. Spectrum of antimicrobial activity of bacteriocin T8
|
Characterization of bacteriocin T8 and mode-of-action studies.
Bacteriocin T8 was sensitive to proteinase K, pepsin, and pronase (Table 3). Treatment with amylase had no effect on bacteriocin activity. No activity was recorded at pHs 2 and 12 and after 90 min at 100°C. The addition of bacteriocin T8 to actively growing cells of E. faecalis resulted in a rapid decrease in cell density (Fig. 1). According to Tricine-SDS-polyacrylamide gel electrophoresis, the molecular size of bacteriocin T8 is between 3.5 and 6.5 kDa (not shown).
|
View this table: [in a new window] |
TABLE 3. Effects of enzymes, pH, and temperature on the activity of bacteriocin T8
|
![]() View larger version (5K): [in a new window] |
FIG. 1. Effect of bacteriocin T8 on cell growth of E. faecalis MDK2. Bacteriocin T8 was added to mid-logarithmic-phase cells (after 5 h of growth). Symbols: , growth in the absence of bacteriocin T8; , growth in the presence of bacteriocin T8. OD, optical density.
|
Two plasmids were isolated from E. faecium T8. Growth in the presence of 3% (wt/vol) SDS resulted in the loss of the smaller, 7-kb plasmid (plasmid T8, not shown) and a loss of bacteriocin activity. Sequencing of plasmid T8 revealed four open reading frames (ORFs). ORF 1 encodes a 74-amino-acid peptide, the bacteriocin precursor. As shown in Fig. 2, the start codon (ATG) is preceded 9 bp upstream by a potential Shine-Dalgarno ribosome-binding site (AAAGGA; underlined) and 68 bp upstream by a potential 10 consensus promoter region (Pribnow box) (TATAAT; underlined). The first 222 bp in ORF 1 encode the prebacteriocin (leader peptide and probacteriocin). The leader peptide contains a potential signal peptidase-processing site (VDA) from positions 82 to 90 (Fig. 2, double underlining). A conserved YGNG sequence is located at positions 100 to 111 (Fig. 2, box). Two cysteine residues (positions 118 to 120 and 133 to 135) and a valine residue (position 139 to 141) are also conserved (Fig. 2). The center of the signal peptide is hydrophobic (Fig. 3A), conforming to that of typical signal peptides (30). The mature peptide is hydrophilic overall, with minor hydrophobic peaks (Fig. 3B).
![]() View larger version (14K): [in a new window] |
FIG. 2. Nucleotide sequence of the bacteriocin region and deduced amino acid sequence. The 10 promoter box and the Shine-Dalgarno (S.D.) ribosome-binding sequences are underlined. ORF 1 encodes the structural gene of bacteriocin T8, and ORF 2 encodes the potential immunity gene.
|
![]() View larger version (12K): [in a new window] |
FIG. 3. Hydrophobicity profiles of the leader peptide (A) and the mature bacteriocin (B).
|
|
|
|---|
Bacteriocin T8 differs from the majority of class IIa bacteriocins in having a sec-dependent leader signal peptide (Fig. 2 and 4). As far as we could determine, this is the fourth class IIa sec-dependent bacteriocin described and the fifth sec-dependent bacteriocin described for a species of the genus Enterococcus. The three class IIa sec-dependent bacteriocins described thus far are enterocin P, produced by E. faecium P (5); bacteriocin 31, produced by E. faecalis Y1717 (28); and an enterocin P-like bacteriocin produced by E. faecium GM-1. The last bacteriocin differs from enterocin P and bacteriocin 31 in being active against gram-positive and gram-negative bacteria (12). The fourth sec-dependent bacteriocin belongs to class III and is produced by E. faecalis LMG 2333 (20).
![]() View larger version (11K): [in a new window] |
FIG. 4. Comparison of amino acid sequences of the N termini of class IIa sec-dependent bacteriocins. Data for enterocin P and bacteriocin 31 were obtained from the work of Cintas et al. (5) and Tomita et al. (28), respectively.
|
Loss of the 7-kb plasmid resulted in a loss of bacteriocin activity, indicating that the genes encoding bacteriocin T8 are located on the plasmid. The structural gene encodes a sec-dependent leader peptide with three positively charged amino acids, a number of hydrophobic amino acids, and a cleavage region (Fig. 2 and 4). Two small hydrophobic amino acids are positioned adjacent to the cleavage site (VDA), conforming to the definitions of a typical signal peptide and sec-dependent bacteriocins (30). Bacteriocin T8 has a conserved YGNGX2CX4CXV motif (Fig. 2 and 4), which is typical of class IIa bacteriocins. As deduced from the DNA sequence, the molecular mass of bacteriocin T8 is 5.1 kDa, similar to the 5.0 kDa recorded for bacteriocin 31 (5) but larger than the 4.6 kDa reported for enterocin P (28). At the amino acid level, bacteriocin T8 is 69% homologous to bacteriocin 31 and 47% homologous to enterocin P (Fig. 4).
The second ORF encodes a potential immunity protein that has a calculated molecular mass of 10.89 kDa (Fig. 2). The gene is located immediately downstream of the structural gene. This is a common feature of lactic acid bacterial bacteriocins (19). The protein is only 50% homologous to the immunity protein of bacteriocin 31. The immunity protein of bacteriocin 31 consists of 94 amino acids and has a molecular mass of 11.0 kDa (5). No homology has been detected with any other immunity proteins listed in GenBank.
The mobilization protein and relaxase mobilization nuclease are presumably involved in the transfer of the plasmid during conjugation. Mobilization genes in enterococci are often located on the same gene cluster as the structural genes of their bacteriocins, as observed for enterocins 1071A and 1071B (4). The genes encoding bacteriocins of enterococci are known to be associated with pheromone-responsive conjugative plasmids (17, 18, 23). Secretion of specific peptide sex pheromones by recipients initiates cell aggregation and conjugation (6, 7). The genes encoding bacteriocin 31 are also located on a pheromone-responsive conjugative plasmid (28).
Although bacteriocin T8 and bacteriocin 31 share certain characteristics, the two peptides are structurally different (69% homology at the amino acid level). The low homology recorded between the signal peptides (27%) and immunity proteins (50%) of the two strains is further proof that bacteriocin T8 is different from bacteriocin 31. Furthermore, the fact that no homology could be detected with any other immunity protein besides that of bacteriocin 31 confirms that bacteriocin T8 is unique.
|
|
|---|
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»