To purify and primarily characterize an anti-Alicyclobacillus bacteriocin produced by Bifidobacterium animalis subsp. animalis CICC 6165, suggested to be named bificin C6165.
To purify and primarily characterize an anti-Alicyclobacillus bacteriocin produced by Bifidobacterium animalis subsp. animalis CICC 6165, suggested to be named bificin C6165.
During purification of the bificin C6165, optimal recovery was achieved with ammonium sulfate precipitation followed by two chromatographic steps. Mass spectrometry analyses revealed a distinctive peak corresponding to a molecular mass of 3395·1 Da. This bacteriocin was heat stable, effective after refrigerated storage and freeze–thaw cycles. The primary mode of action of bificin C6165 is most probably due to pore formation, as indicated by the efflux of K+ from metabolically active cells of Alicyclobacillus acidoterrestris. In the presence of 10 mmol l−1 gadolinium, bificin C6165 did not affect cells of Alicyclobacillus acidoterrestris. This suggests that the mode of action of bificin C6165 relies on a net negatively charged cell surface.
Bificin C6165 is indeed a novel bacteriocin and it exhibited remarkable potency for Alicyclobacillus control.
Application of bacteriocins in preservation of fruit juices has seldom been studied. Bificin C6165 may be an alternative method to control juice spoilage by this Alicyclobacillus acidoterrestris and meet increasing consumer demand for nature and artificial chemical additive-free food products.
Ribosomally synthesized peptides with antimicrobial properties are produced by a broad variety of living organisms ranging from prokaryotes to higher eukaryotes (Papagianni 2003). Such antimicrobial peptides are referred to as bacteriocin when they are produced by bacteria. The use of bacteriocins for food preservation has been a matter of extensive work in recent years (Todorov et al. 2011; Ansari et al. 2012; Cosentino et al. 2012; Santagati et al. 2012). Bifidobacteria are normal inhabitants of the human gastrointestinal tract throughout life. They are considered to be probiotics because they exert health effects beyond basic nutrition when ingested in sufficient numbers. This concept has raised a growing interest for their application in food and medicine (Cheikhyoussef et al., 2009).
Bacteriocins are divided into three classes based on their structure (Klaenhammer 1993). The lantibiotics (class I) are small heat-resistant peptides that undergo post-translational modifications (Moll et al. 1999). The lantibiotics are further divided into two subgroups, namely the elongated and amphipathic pore-forming type A peptides (e.g. nisin), and the globular peptides of the type B category (e.g. mersacidin and actagardine) (Bauer et al. 2005). Class II is comprised of nonmodified, heat-stable bacteriocin, which are divided into subclasses IIa (strong anti-Listeria bacteriocins), IIb (two-peptide bacteriocins) and IIc (other peptide bacteriocins). The majority of bacteriocins produced by the Bifidobacteria group belong to class IIa (Cheikhyoussef et al., 2009). Class III is represented by heat-labile proteins with sizes in excess of 15 kDa.
The mode of action of bacteriocins is not yet fully understood. Model membrane studies with nisin have shown that lipid II acts as a docking station (Wiedemann et al. 2001). After binding, nisin inserts itself into the cell membrane to form short-lived pores, which disturb the integrity of the cytoplasmic membrane and cause the efflux of ions and other cell components (Sablon et al. 2000). At high concentrations of nisin, pore formation may occur in the absence of lipid II, provided the cell membrane contains at least 50% negatively charged phospholipids (Wiedemann et al. 2001). Under these conditions, the positively charged C-terminus of nisin is important for initial binding and antimicrobial activity. Mersacidin and the antibiotic vancomycin also bind to lipid II, but to a different part of the molecule (Brotz et al. 1997).
Alicyclobacillus acidoterrestris is a novel Gram-positive thermo-acidophilic spore former that has been recognized as a spoilage micro-organism (Cerny et al. 1984). This micro-organism is considered to be one of the most important target organisms in quality control of fruit juices and fruit juice–containing drinks, because of its ability to germinate and outgrow spores under high acid conditions (Silva and Gibbs 2001; Grande et al. 2005). Therefore, developing an effective method for inhibiting or controlling the germination and outgrowth of A. acidoterrestris in such products is an urgent need of the beverage industries. Application of bacteriocins in preservation of fruit juices has seldom been studied (Komitopoulou et al. 1999). Bacteriocins produced by Bifidobacteria are generally stable at high temperature and low pH value; and hence, it might be suitable for inhibiting the growth of food-borne pathogens and spoilage bacteria in acidic fruit juices drinks.
In this study, we described our data on the purification, characterization and mode of action of a novel anti-Alicyclobacillus bacteriocin produced by Bifidobacterium animalis subsp. animalis CICC 6165, suggested to be named bificin C6165.
The bacteriocin producer strain B. animalis subsp. animalis CICC 6165 (China Center of Industrial culture Collection, China) was anaerobically cultivated in MRS broth (Oxoid ltd, Basingstoke, UK) at 37°C and routinely stored at 4°C on MRS agar. For long-time storage, strain was maintained as frozen stocks at −80°C in 20% glycerol.
Cell-free supernatant (CFS) preparation was given as follows: 24-h bacterial culture was harvested by centrifugation (8000 × g, 20 min) to obtain a CFS followed by a filtration through 0·2-μm-pore-size filters (Millex GV, Millipore, MA, USA).
Antimicrobial activity was tested using the spot-on-lawn method, as described by Ennahar et al. (2001). Inhibition was recorded as negative if no inhibition zone was observed around the agar well. Antimicrobial activity was expressed as arbitrary units (AU) per ml, and one AU was defined as the reciprocal of the highest dilution showing a clear zone of growth inhibition (Ennahar et al. 2001).
The ability of the bacteriocin to adsorb to the producer cells was studied as described by Todorov et al. (2011). Antimicrobial activity was tested as described above. Alicyclobacillus acidoterrestris DSM 3922 was used as indicator strain.
One litre of CFS (adjusted to pH 6·0 with 4 mol l−1 NaOH) of B. animalis subsp. animalis CICC 6165 was prepared as described above. Ammonium sulfate was gently added to the CFS maintained at 4°C to obtain 80% saturation, and the mixture was stirred overnight at 4°C. After centrifugation for 1 h at 20000 × g at 4°C, the resulting pellet was resuspended in 20 mmol l−1 sodium phosphate buffer (pH 6·0, buffer A) (Fraction I). Fraction I was applied onto a SP-Sepharose Fast Flow (15 mm internal diameter, 100 mm length, Sigma-Aldrich, St Louis, MO, USA) cation-exchange column equilibrated with buffer A. After the column was washed with 300 ml buffer A, the bacteriocin was eluted with linear NaCl gradient (0–0·5 mol l−1) at pH 6·0 (fraction II). Fraction II was loaded on a Sep-Pak C18 cartridge (Waters, Milford, USA). The cartridge was washed with 0·05% trifluoroacetic acid (TFA), and the bacteriocin was eluted with 100% acetonitrile in 0·05% TFA. After drying under reduced pressure (SIM Int. Co., Beijing, China), the bacteriocin fraction was dissolved in 0·05% TFA (Fraction III). Fraction III was used for final purification by reverse-phase HPLC on a C 18 reverse-phase column (ODS-80Ts, 4·6 × 150 mm, Tosho, Japan). Elution was performed using a linear gradient from 100% 0·05% TFA to 100% acetonitrile in 60 min at a flow rate of 1·0 ml ml−1 (fraction IV). Protein concentrations were determined using Pierce BCA Protein Assay Reagent (Thermo Fisher Scientific, Waltham, MA, USA).
Lyophilized samples of the active fraction collected from the C 18 reverse-phase column were resuspended in distilled water and subjected to a MALDI-TOF mass spectrometry in Beijing protein research and Development Center. Sequencing was performed on a Shimadzu PPSQ-21A protein sequencing apparatus (Shimadzu, Kyoto, Japan).
An 18-h-old culture of strain B. animalis subsp. animalis CICC 6165 was inoculated (2%, v v−1) into MRS broth and incubated at 25°C, 37°C, 40°C and 45°C, respectively. Antimicrobial activity of the bacteriocin and modifications in pH and optical density (OD600) of the culture were determined at regular intervals for 36 h. A. acidoterrestris DSM 3922 was used as sensitive strain.
Inhibitory activity of bificin C6165 produced by B. animalis subsp. animalis CICC 6165 was determined against selected indicator strains with purified bificin C6165 at concentration of 5 mg l−1. Indicator strains used in this study are listed in Table 2.
The purified bificin C6165 was treated with different enzymes such as trypsin (Wolsen, Xian, China), proteinase K (Wolsen), pepsin (Wolsen), papain (Wolsen), a-chymotrypsin (Wolsen), catalase (Wolsen), lipase (Wolsen) and α-amylase (Wolsen). The enzymes were diluted in sterile phosphate buffer (100 m mol l−1, pH 6·5) to a concentration of 0·2 mg ml−1. Bificin C6165 was added to the enzyme solutions at concentration of 5 mg l−1. Control samples that contained only buffer solution were used. The samples were incubated at 37°C for 2 h, and residual activities were determined as before (Ennahar et al. 2001).
The pH ranges tested were chosen so as to imitate fruit juices environmental conditions. The effect of pH on bacteriocin activity was determined by adjusting the CFS or purified bificin C6165 in distilled water (5 mg l−1) between pH 3·5 and 7·5 with sterile 4 mol l−1 HCl or 4 mol l−1 NaOH. After 2 h of incubation at 37°C, samples were neutralized to pH 6·5, and the activities were determined as described before.
The effect of temperature on the bacteriocin activity was tested by heating the CFS and purified bificin C6165 to 60°C, 80°C and 100°C for 30 min. After each treatment, samples were immediately cooled under refrigeration and residual activities were determined as previously described. To test the stability of bificin C6165 during three freeze–thaw cycles, it was frozen at −20°C during 24 h and thawed for 20 min at 25°C. The effect of extended storage at low refrigerated temperature on bificin C6165 stability was also evaluated by placing bificin C6165 at 4°C for 14 days.
In a separate experiment, the effect of food additives such as EDTA (1% m v−1) (Solarbio, Beijing, China), Tween 20 (1% m v−1) (Solarbio), ascorbic acid (1% m v−1) (Solarbio), potassium sorbate (1% m v−1) (Solarbio), sodium citrate (1% m v−1) (Solarbio) and β-Carotene (1% m v−1) (Solarbio) on the activity of bacteriocin in CFS and purified bificin C6165 was determined as described before.
A 20-ml aliquot of bacteriocin-containing CFS was added to 100 ml culture of A. acidoterrestris DSM 3922, Enterococcus faecium CICC 20420 and Staphylcoccus aureus CICC 10201 in early and middle exponential phases. Optical density readings (at 600 nm) were recorded at convenient intervals.
Early stationary-phase cultures of A. acidoterrestris DSM 3922 and E. faecium CICC 20420 and S. aureus CICC 10201 were harvested (8000 × g, 20 min), washed twice with sterile saline water and resuspended in 10 ml of this solution. Equal volumes of the cell suspensions and bacteriocin-containing CFS were mixed, and viable cell numbers were determined before and after incubation for 4 h at 37°C by plating onto AAM agar or BHI agar. Cell suspensions without addition of the bacteriocin-containing CFS served as controls.
Alicyclobacillus acidoterrestris DSM 3922 cells were grown to mid-exponential phase (OD600 = 1·0) at 45°C, centrifuged (8000 × g, 20 min) and resuspended to 1010 CFU ml−1 in AAM broth. Then, serial twofold dilutions of bificin C6165 were added. After the incubation period (at 45°C for 24 h), the cell viability was determined on agar plates after 48-h incubation at 45°C. The most diluted concentration of bificin C6165 that inhibited the growth of A. acidoterrestris DSM3922 was recorded as 1 MIC.
In separate experiment, endospore suspensions of A. acidoterrestris DSM 3922 were incubated at 40°C for 24 h with increasing bacteriocin concentrations. Cell viability was determined by plating onto AAM agar. Preparation of spore suspensions was described by Grande et al. (2005).
The efflux of K+ from cells of A. acidoterrestris DSM 3922 at different pH values was recorded. Alicyclobacillus acidoterrestris DSM 3922 was grown in AAM broth at 45°C to mid-exponential phrase (OD600 = 1·0). The culture was divided into three equal volumes, harvested (8000 × g, 20 min) and washed with sterile saline water of pH 4·0, 5·0 and 6·5, respectively. The cells were resuspended in sterile saline water of corresponding pH, supplemented with 0·2% (m v−1) glucose and 0·5 mmol l−1 KCl and adjusted to 1010 CFU ml−1. The level of K+ in the medium (K+out) was recorded with a valinomycin-based potassium-selective electrode (PHG, Netherland). As soon as a steady state of K+out was maintained for at least 5 min, purified bacteriocin (1 MIC) was added to each of the cell suspensions. Changes in K+out were recorded for a period of 1 h. The intracellular level of K+ (K+in) was calculated from the increase recorded in the K+out level, and results were expressed in μmol K+in per number of cells. The total content of intracellular K+ was determined by exposing the cells to lysozyme (Solarbio) (10 mg ml−1, 30 min, 37°C).
In a separate experiment, the efflux of K+ from cells of A. acidoterrestris DSM 3922 grown at different temperatures was recorded. Strain A. acidoterrestris DSM 3922 was grown at 30°C, 37°C and 45°C to mid-exponential phrase, washed with sterile saline water (pH 4·0) and resuspended in the same buffer, supplemented with 0·2% (m v−1) glucose and 0·5 mmol l−1 KCl, as described before. K+out readings were recorded for 1 h, and the K+in levels were determined as described before.
The BMM medium was used to study the efflux of cytoplasmic content from A. acidoterrestris DSM 3922 cells treated with purified bificin C6165 (1 MIC). Cultures of A. acidoterrestris DSM 3922 were harvested by centrifugation (8000 × g, 20 min) and resuspended in the BMM medium. After incubated for 30 min at 45°C with 1 MIC bificin C6165, the cultures were centrifuged, and the concentration of cytoplasmic content was measured, in the remaining supernatant, spectrophotometrically by the determination of the absorbance at 260 nm (Bendali et al. 2008).
Alicyclobacillus acidoterrestris DSM 3922 was grown in AAM broth at 45°C to mid-exponential growth (OD600 = 1·0). The cell suspensions were treated with 10 mmol l−1 Gd3+ and 2 mmol l−1 Gd3+, respectively. Cell suspensions with Gd3+ were allowed 30-min contact time at 45°C before the addition of bificin C6165. Viable cells numbers were determined every hour of incubation at 45°C by plating onto AAM agar. Cell density was measured at 600 nm over 4 h. Cell suspensions treated with only Gd3+ were prepared as negative control.
To determine whether bificin C6165 production is plasmid linked in B. animalis subsp. animalis CICC 6165, the strain was cultured in MRS broth supplement with acriflavine (Wolsen, China) (10–50 ug ml−1) as described by Bauer et al. (2005) and transferred in the same media on successive 5 days. Dilutions of the treated cells were placed onto MRS agar and incubated at 37°C. Colonies were selected at random and inoculated onto a second set of MRS agar plates (10–30 CFU plate−1, in duplicate). After overnight incubation at 37°C, one set of plates was overlaid with AAM soft agar (1%, m v−1, agar) and seeded with viable cells of A. acidoterrestris DSM 3922 (106 CFU ml−1). The plates were incubated for 48 h at 40°C and examined for colonies of B. animalis subsp. animalis CICC 6165 that failed to produce zones of growth inhibition (bificin C6165−). Corresponding colonies for the second set of plates were selected for plasmid isolations. Plasmid DNA was isolated using the lysozyme–mutanolysin lysis method (Burger and Dicks 1994), followed by CsCl density gradient centrifugation and agarose gel electrophoresis, as described by Sambrook et al. (1989).
An 18-h-old culture of strains B. animalis subsp. animalis CICC 6165 was inoculated (2%, v v−1) into MRS broth, skim milk medium (SMM), BHI broth and M17 broth. Incubation was at 37°C for 30 h. Samples were taken every five hour and examined for bacterial growth (OD at 600 nm) and antimicrobial activity against A. acidoterrestris DSM 3922.
The influence of pH on the level of bacteriocin production was studied in a 2·5-L fermentor (Minifors zc2454; INFORS-HT, Doetinchem, Switzerland). The pH of culture broth was adjusted initially to predetermined values (5·0, 5·5, 6·0, 6·5 and 7·0) and maintained constantly during fermentation by the automatic addition of 1 mol l−1 NaOH solution. Samples were aseptically withdrawn from the fermentor at 5-h intervals and tested for cell growth and bacteriocin level. All cultivations were performed in duplicate.
Treatment of the B. animalis subsp. animalis CICC 6165 with 0·1 mol l−1 NaCl (pH 2·0) for 6 h caused adsorption of the bificin C6165 to these cells at 320 AU ml−1 activity level. As the activity of the bificin C6165 in the culture medium was much higher than that detected in the surface of the bacteriocin-producing B. animalis subsp. animalis CICC 6165 (320 AU ml−1), CFS from 24-h cultures in MRS broth at 37°C was used for bacteriocin purification. The purification procedure for bificin C6165 is summarized in Table 1. Eighty two per cent of bacteriocin activity was recovered after ammonium sulfate precipitation of the CFS of B. animalis subsp. animalis CICC 6165. Further, the precipitate was subjected to SP-Sepharose cation-exchange chromatography. The specific activity of studied bacteriocin increased 438-fold and the recovery was 63·6%. The eluted active fractions were pooled and further separated using reverse-phase HPLC column. The main peaks were collected and assayed for bacteriocin activity. The results indicated that this activity corresponded to the peak at 34 min. The next HPLC reverse-phase chromatography was applied to the active peak for final purification. At this stage, the purification factor reached 1169 and the recovery was 25%.
|Purification stage||Volume (ml)||Total protein concentration (mg)||Total activity (106 AU)||Specific activity (AU mg−1)||Increase in specific activity||Act yield (%)|
|Ammonium sulfate precipitation||100||625||2·1||3360||8||82|
Mass spectrometry analyses revealed a distinctive peak corresponding to a molecular mass of 3395·1 D (Fig. 1). The purified bificin C6165 was directly subjected to amino acid sequencing, and the first 13 amino acids were successfully identified from the N-terminus before the reaction was blocked. Alternatively, bificin C6165 was reduced with the method by Meyer et al. 1994, and performed in the same way, Edman degradation proceeded to the end, and 33 amino acids (KKISGXTLTS DXISLSIXTV SKDXKLATAT XSI) were identified. This sequence of bificin C6165 was compared to deposited sequences, but no significant similarities were found, showing that bificin C6165 indeed is a novel peptide.
Similar growth rates of B. animalis subsp. animalis CICC 6165 and production of bificin C6165 (2560 AU ml−1) were recorded when the strain was cultivated either at 37°C or 40°C in MRS broth for 24 h. When cultivation temperature was at 45°C, after 24 h only 640 AU ml−1 bacteriocin activity was recorded. A very low growth rate was observed at 25°C cultivation temperature where a low level of bacteriocin activity (100 AU ml−1) was detected after 5 days of incubation. Bacteriocin activity was determined at the beginning of the exponential phase at all temperatures used. Based on these results, all further experiments were conducted at 37°C. The maximal bificin C6165 production was recorded after 24–30 h of growth in MRS broth (Fig. 2), with a decrease in activity to 1600 AU ml−1 observed after 33 h of growth. During the same period of growth, the pH of the medium decreased from 6·52 to 3·51. The cell density (OD600 nm) increased from 0·05 to 2·58 in 24 h and stayed more or less constant during the following 12 h (Fig. 2).
When bificin C6165 was tested against twenty Alicyclobacillus Spp. strains used in this study, all of them were found to be sensitive. Bificin C6165 also had inhibitory activities towards species belong to the same genus but not towards the producer strain, towards LAB strains and Gram-positive bacteria including S. aureus and E. faecium. In contrast, other Gram-positive and Gram-negative bacteria were not inhibited by bificin C6165 (Table 2).
|Indicator strains||Sources||Growth conditions||Sensitive/total strains|
|Alicyclobacillus acidoterrestris||DSM (1), Our lab collections (19)||AAM, 45°C||20/20|
|Bifidobacterium animalis subsp. animalis||CTCC||MRS, 37°C||0/1|
|B. bifidum||CTCC||MRS, 37°C||2/2|
|B. adolescentis||CTCC||MRS, 30°C||1/1|
|Lactobacillus rhamnosus||CTCC||MRS, 37°C||0/3|
|L. brevis||CTCC||MRS, 37°C||0/1|
|L. curvatus||CTCC||MRS, 37°C||1/1|
|L. helveticus||CTCC||MRS, 30°C||1/1|
|L. paracasei||CTCC||MRS, 37°C||0/1|
|L. delbrueckii subsp. lactis||CTCC||MRS, 30°C||0/2|
|Clostridium butyricum||ATCC||RCM, 25°C||0/1|
|Enterococcus faecium||CTCC||MRS, 37°C||2/2|
|Bacillus subtilis||CTCC||TSA, 30°C||0/1|
|Staphylcoccus aureus||CTCC||TSA, 30°C||1/1|
|Salmonella enterica spp. enterica||CTCC||TSA, 30°C||0/1|
|Escherichia coli||CTCC||TSA, 30°C||0/2|
Treatment of the CFS with proteinase K, pepsin, a-chymotrypsin, papain or trypsin resulted in complete inactivation of antimicrobial activity (Table 3). Similar results were observed when CFS and purified bificin C6165 were tested. Treatment with catalase did not affect the activity against the target strains, showing that H2O2 was not involved. Moreover, treatment with α-amylase and lipase did not affect the antimicrobial activity, suggesting that bificin C6165 does not belong to the controversial group IV of the bacteriocins, which contain carbohydrates or lipids in the active molecule structure.
|Alicyclobacillus acidoterrestris DSM 3922||E. faecium CICC 20420||S. aureus CICC 10201|
|Proteinase K, trypsin, pepsin, papain, a-chymotrypsin||−||−||−|
|Extended storage at low refrigerated temperature||+||+||+|
|pH 3·5-pH 6·5||+||+||+|
Treatment of bificin C6165 at either 60°C for 30 min or 80°C for 30 min did not elicit any loss of antimicrobial activity. However, such activity was destroyed at 100°C for 30 min (Table 3). The results of bacteriocin stability throughout the refrigerated storage time showed that the maximum inhibitory activity remained constant up to 14 days. In addition, 100% of the initial activity was observed after the three freeze–thaw cycles. The pH stability of the bacteriocin activity was studied in the pH range 3·5–7·5. Although bificin C6165 was effective within the pH range 3·5–6·5, it became completely ineffective at pH 7·5 (Table 3).
Various food additives in fruit juices were tested for their influence on bificin C6165 activity. The presence of the different additives tested had several effects, which were also dependent upon the sensibility of each indicator micro-organism. The presence of ascorbic acid enhanced bificin C6165 activity against E. faecium CICC 20420 and S. aureus CICC 10201 (Table 4). On the contrary, the presence of the potassium sorbate decreased bacteriocin activity against E. faecium CICC 20420 and S. aureus CICC 10201 (Table 4). Moreover, Tween 20 increased bificin C6165 activity against all indicator strains. EDTA only affected the activity of bificin C6165 against S. aureus CICC 10201 (Table 4).
|Alicyclobacillus acidoterrestris DSM 3922||Enterococcus faecium CICC 20420||Staphylcoccus aureus CICC 10201|
|Control system (without additives)||2560||1280||2560|
Addition of CFS of B. animalis subsp. animalis CICC 6165 to an early exponential-phase culture of A. acidoterrestris DSM 3922, E. faecium CICC 20420 or S. aureus CICC 10201 repressed cell growth of the indicator strains (Fig. 3). When the supernatant was added to the mid-exponential-phase culture, a similar inhibition was observed (Fig. 3). Treatment of cells of A. acidoterrestris DSM 3922, E. faecium CICC 20420 or S. aureus CICC 10201 (108–109 CFU ml−1) at stationary phase with bificin C6165 resulted in the complete loss of cellular viability (death), whereas the counts were not modified when these micro-organisms were incubated with the control samples (data not shown).
The decrease in the A. acidoterrestris DSM 3922 cell viability depends on the bacteriocin concentration expressed in AU ml−1 as shown in Fig. 4(a). At the lowest concentration (8 and 16 AU ml−1), no effect was observed. At 32–64 AU ml−1, the effect was bacteriostatic, whereas the bactericidal effect was only observed at concentrations higher than 256 AU ml−1 (Fig. 4a). The MIC was determined equal to 256 AU ml−1. For endospore suspensions, cell viability decreased in proportion to the bacteriocin concentration added (Fig. 4b). A bacteriocin concentration of 512 AU ml−1 reduced the viable cell counts below the detection limit.
Efflux of potassium (K+) occurred rapidly and immediately after the addition of bificin C6165, followed by a slow release (Fig. 5). Samples treated with 1 MIC bificin C6165 for 1 h displayed greater changes in the extracellular K+ concentration than untreated controls. Cells incubated at pH5·0 lost approximately 0·65 μmol K+ 1010 cells−1 during the first 30 s in the presence of bificin C6165, followed by a short recovery period and slow K+ efflux, reaching an intracellular K+ concentration of approximately 0·2 μmol K+1010 cells−1 over the next hour (Fig. 5a). Cells incubated at pH 4·0 lost only 0·3 μmol K+ 1010cells−1 during the 30 s, but reached an intracellular K+ concentration of below 0·05 μmol K+ 1010 cells−1 over the next hour. The rate of efflux decreased dramatically when the extracellular pH was raised to pH 6·5 (Fig. 5a).
The rate of K+ efflux from cells incubated at 45°C was much higher, as recorded by a decrease in extracellular K+ from 1·3 μmol 1010 cells−1 to 0·65 μmol 1010 cells−1 within the first 30 s, and a further decrease to approximately 0·1 μmol 1010 cells−1 over the next 60 min. The initial rate of K+ was less at 30°C and 37°C (Fig. 5b).
It was observed that the substance makes the membranes of sensitive cells permeable, allowing the leakage of K+ ions but not larger compounds during the first 30 min. The loss of larger cytoplasmic content, indicated by the release of UV-absorbing material at 260 nm (A = 1·2) compared to the control (A = 1·0), was only observed after further incubation (after 30 min) (data not shown).
Gadolinium at 10 m mol l−1 completely inhibited the activity of bificin C6165 as detected by cell viability recorded for A. acidoterrestris DSM 3922 (results not presented). The inhibition of bacteriocin activity decreased with decreased concentration of Gd3+. The cell numbers decreased from approximately 108 CFU ml−1 to 105 CFU ml−1 over 4 h in the absence of Gd3+, compared to a decrease from approximately 108 CFU ml−1 to 106 CFU ml−1 in the presence of 2 mmol l−1 Gd3+ (Fig. 6a). Cell death in the presence of bificin C6165 is accompanied with cell lysis as revealed by a decrease in turbidity (Fig. 6b). At 2 mmol l−1 Gd3+, bacteriocin-induced lysis was not inhibited (Fig. 6b).
The MIC value was determined as the lowest concentration of acriflavine that prevented the appearance of turbidity (Yildirim et al. 1999). The MIC of acriflavine for B. animalis subsp. animalis CICC 6165 was found to be 40 ug ml−1. Bificin C6165− colonies that failed to produce zones of growth inhibition did not differ from the parent strain in growth rate. They remained immune to bificin C6165. B. animalis subsp. animalis CICC 6165 harboured one plasmid smaller than 10 kbp. The variant bificin C6165− derivatives did not have the <10 kbp plasmid that was present in the parent strain as it was shown by agarose gel electrophoresis (results not shown).
Four different culture media were compared for bificin C6165 production (Table 5). MRS medium is a better medium for cell growth and bacteriocin production than other media. Bificin C6165 production with regulated pH (pH 6·0) was higher than the production with nonregulated pH. The pH-controlled fermentations indicate that the production of bificin C6165 by the strain B. animalis subsp. animalis CICC 6165 is strongly dependent on the pH of culture broth. The highest level of bacteriocin activity was observed at pH 6·0. However, low bacteriocin activities were detected at pH 5·5 and 6·5 although the growth patterns were only slightly different from that at pH 6·0. The bacteriocin production was negligible when pH of the broth was maintained at 5·0 and 7·0 (data not shown).
|Culture media||OD600||Activity (AU ml−1)|
The sequence of bificin C6165 was compared to deposited sequences, but no significant similarities were found, showing that bificin C6165 indeed is a novel peptide. Two class I type B lantibiotics, mersacidin and actagardine, contain a conserved sequence motif (amino acids 12–18 and 6–12, respectively) that comprises one entire ring structure (Brotz et al. 1997) and this conserved motive (CTLTSEC) is shared with bificin C6165 (6–12). Mersacidin exerts its bactericidal action by the inhibition of peptidoglycan biosynthesis through interaction with lipid II (Brotz et al. 1997). The conserved sequence motif shared by mersacidin, actagardine and bificin C6165 may be involved in the binding of these bacteriocins to lipid II.
The production of bificin C6165 reached maximum activity after 24–30 h, and the highest bacterial count was observed after 24 h of incubation (Fig. 2), which is in agreement with the bacteriocin production data from other LAB (Cheikhyoussef et al. 2009; Todorov et al. 2011). This observation leads to the idea of a production dependent upon the cell number (Anastasiadou et al. 2008; Cheikhyoussef et al. 2009; Todorov et al. 2011). The bificin C6165 at 33 h decreased by 60%, which is in keeping with bacteriocins production by Bifidobacterium (Cheikhyoussef et al. 2009) and bacteriocin ST16 Pa (Todorov et al. 2011). The decrease in activity of bacteriocin at the end of the monitored period could be explained by proteolytic degradation or bacteriocin aggregation (Todorov et al. 2011).
For the food-processing industry, endospores represent the most difficult life form to inactivate because they show a higher resistance not only to heat treatments but also to food treatments like ultra-high pressure or pulsed electric fields of high intensity (Grande et al. 2005). As A. acidoterrestris endospores require intense heat treatments for inactivation, application of bificin C6165 may represent a technological advantage to currently used inactivation processes.
The ability of bificin C6165 to form pores in sensitive cells of A. acidoterrestris, as observed by K+ loss, is pH dependent. Although the rate of bacteriocin-induced K+ loss was highest at pH 4, initial K+ loss was highest at pH 5. This suggests that, under physiological conditions, the △pH may contribute to bacteriocin action as observed for nisin (Moll et al. 1999). A short period of reaccumulation of K+ could be observed after the initial rapid release following bificin C6165 addition. These results may suggest that bificin C6165 treatment results in an efflux of ions, large cytoplasmic content, such as amino acids and ATP remain intracellular and are utilized to reaccumulate K+., and that the peptide may act in a way similar to that suggested for mersacidin (Brotz et al. 1997). The decrease in K+ efflux was observed at lower temperatures may thus be due to a melting temperature or activation energy affect.
Cations such as Mg2+, Ca2+ and Gd3+ bind to negatively charged head groups of phosphatidylglycerol and cardiolipin in the cytoplasmic membrane of L. monocytogenes, rendering nisin Z less antimicrobial (Abee et al. 1994). Neutralization of the negative charges leads to condensation of lipids, resulting in a more rigid membrane. In our study, treating with Gd3+ (10 mmol l−1) prevented bificin C6165 activity against A. acidoterrestris. The inhibition of bacteriocin activity decreased with decreased concentration of Gd3+. Gd3+ at mmol l−1 did not inhibit the activity of bificin C6165 as well as uncouple cell lysis from cell death. The same result appears to apply to pediocin PD-1, which was reported by Bauer et al. (2005).
Curing studies with acriflavine permitted us to obtain a mutant strain (bificin C6165−), which was unable to produce any inhibitory activity. Plasmid analysis of the bificin C6165− variant revealed a loss of a plasmid (<10 kbp), and the bificin C6165− variant remained immune to bificin C6165. Therefore, this plasmid may encode bificin C6165 production but does not seem to carry immunity. Similar results were reported for bifidiocin B from B. bifidum NCFB (Yildirim et al. 1999).
A number of papers have reported that MRS medium is a better medium for cell growth and bacteriocin production than other media (Avonts et al. 2004; Todorov et al. 2011). Several studies reported that the optimum pH for bacteriocin production was usually 5·5–6·0 (Cheigh et al. 2002), often lower than the optimal pH for growth. However, the optimum pH of some bacteriocins has been reported to be lower than 5·0 (Aasen et al. 2000). In this study, the optimal pH for bacteriocin production by strain B. animalis subsp. animalis CICC 6165 was observed at 6·0 where cell growth was optimal.
This study has been supported by China State ‘12th Five-Year Plan’ scientific and technological support scheme (2012BAD31B01); National Natural Science Foundation of China (31071550, 31171721); and ‘948’ project of the Ministry of Agriculture of China (2011-G8-3).