To characterize novel multiple bacteriocins produced by Lactobacillus sakei D98.
To characterize novel multiple bacteriocins produced by Lactobacillus sakei D98.
Lactobacillus sakei D98 isolated from Shubo (rice malt) produced at least three bacteriocins. Using three purification steps, three novel antimicrobial peptides termed sakacin D98a, sakacin D98b and sakacin D98c were purified from the culture supernatant. Amino acid and DNA sequencing analysis revealed that the sakacins D98a, D98b and D98c are novel class IIa-like or class IId bacteriocins. In particular, sakacin D98b has a variant pediocin-box sequence, YANGVXC (with Ala instead of Gly), and a different location for the disulfide bridge (Cys11 and Cys18) from that found in other class IIa bacteriocins.
Three novel bacteriocins were identified from Lactobacillus sakei D98. Their antimicrobial spectra and intensities indicate that these sakacins would have different modes of action. In addition, sakacin D98b showed low inhibitory activity against Listeria, probably due to the differences in amino acids and position of the disulfide bridge compared with the other class IIa bacteriocins.
Sakacins D98a and D98c are novel bacteriocins belonging to class IId bacteriocins. On the other hand, sakacin D98b, a class IIa-like bacteriocin, has a unique internal structure and activity spectrum.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
Lactic acid bacteria (LAB) are widely disseminated in nature and used in a variety of fermentation processes. Accordingly, LAB, such as those belonging to the genera Lactococcus and Lactobacillus, are employed in food preservation and often isolated from traditional fermented foods such as soy sauce, Miso, sake (Japanese rice wine) and Nukadoko in Japan (Ennahar et al. 1999; Ishizaki et al. 2001; Nakayama et al. 2007). Sake is brewed in a unique fermentation process that uses Shubo (rice malt) as the starter culture of yeasts and LAB. Additionally, LAB are popularly used as starter cultures in other fermented foods (Leroy and de Vuyst 1999; Benkerroum et al. 2000; Rilla et al. 2003) and are described as potential food preservatives.
The major role of the LAB in Shubo is to produce lactic acid and other antimicrobial substances, probably including bacteriocins, which protect the yeasts from contamination and help their growth (Taniguchi et al. 2010). In particular, Lact. sakei, Lact. plantarum and Leuconostoc mesenteroides are used to achieve controlled acidification and to inhibit spoilage bacteria in Shubo (Hammes et al. 2005).
Many strains of LAB in fermented foods and starter cultures were found to produce bacteriocins, ribosomally synthesized antimicrobial peptides with inhibitory activity against bacteria closely related to producer strains and against gram-positive foodborne spoilage bacteria. In fact, several bacteriocin-producing strains, isolated from fermented foods including the above-mentioned ones, have already been reported by our laboratory (Sawa et al. 2009, 2010, 2012; Wilaipun et al. 2004; Masuda et al. 2011, 2012).
Bacteriocins are commonly divided into several groups (Cotter et al. 2005). Class I bacteriocins (lantibiotics), including nisin, are characterized by the presence of post-translationally modified amino acid residues such as lanthionine (Asaduzzaman and Sonomoto 2009). Class II bacteriocins contain no lanthionine and are further divided into four subgroups. Class IIa bacteriocins are pediocin-like bacteriocins containing the N-terminal region motif YGNGVXC, called the pediocin box, and have high activity against Listeria monocytogenes (Drider et al. 2006). Class IIb bacteriocins such as lactococcin Q are two-peptide bacteriocins, whose complete antimicrobial activities require the presence of both the peptides in equal amounts (Zendo et al. 2006). Class IIc bacteriocins such as enterocin AS-48 (Galvez et al. 1986) and lactocyclicin Q (Sawa et al. 2009) are circular bacteriocins. Class IId bacteriocins include other class II bacteriocins such as lacticin Q (Fujita et al. 2007).
Bacteriocin-producing strains of Lact. sakei were previously isolated from various foods and identified (Schillinger and Lucke 1989; Sobrino et al. 1992; Diep et al. 2000; Skaugen and Nes 2000; Dortu et al. 2008). In particular, Lact. sakei 5 produces three bacteriocins belonging to different classes (Vaughan et al. 2001). Notably, sakacin 5× has a variant pediocin box with YGNGLXC in the N-terminal region. In addition, bacteriocin 31 produced by Enterococcus faecalis (Tomita et al. 1996), plantaricin C19 produced by Lact. plantarum C19 (Atrih et al. 2001), piscicocin CS526 produced by Carnobacterium piscicola CS526 (Suzuki et al. 2005; Yamazaki et al. 2005) and enterocin NKR-5-3C produced by Ent. faecium NKR-5-3 (Ishibashi et al. 2012; Himeno et al. 2012) also have unique pediocin-box sequences, but show high anti-Listeria activity as in the case of the typical class IIa bacteriocins.
In this report, we describe the structural analysis and characterization of multiple bacteriocins produced by Lact. sakei D98 isolated from Shubo. We identified a novel class IIa-like bacteriocin with a variant pediocin box, and two class IId bacteriocins, termed sakacin D98b, and sakacins D98a and D98c, respectively.
The Shubo sample was obtained from a sake brewery, Inoue Goumei Co. Ltd., in Fukuoka prefecture, Japan. A bacteriocin-producing strain, Lactobacillus sakei D98, was isolated from Shubo and identified by sugar fermentation pattern analysis (API50 CHL; BioMérieux, Marcy l'Etoile, France) and 16S rRNA gene sequencing as described previously (Sawa et al. 2010). Lactobacillus sakei D98 was stored at −80°C in MRS medium (Oxoid, Basingstoke, UK) with 15% glycerol. Before use, it was propagated in MRS medium at 30°C for 18 h. Bacterial strains used as indicator strains for the bacteriocin assay were propagated for 18 h at the temperature (30 or 37°C) recommended by the culture collections. Tryptic soy broth (BD Biosciences, Franklin Lakes, NJ, USA) supplemented with 0·6% yeast extract (Nacalai Tesque, Kyoto, Japan; TSBYE) was used to culture indicator strains.
The bacteriocin activity assay was performed using the spot-on-lawn method (Sawa et al. 2010). After overnight incubation at temperatures appropriate for the indicator strains, the bacterial lawns were checked for inhibition zones. At the purification step, the activity titre, expressed in arbitrary activity units (AU) per millilitre of bacteriocin preparation, was defined as the reciprocal of the highest dilution causing a clear zone of growth inhibition in the indicator lawn. The minimum inhibitory concentrations (MICs) of the bacteriocins against various indicator strains were determined using the spot-on-lawn method with the purified bacteriocin solutions. The MICs were defined as the minimum concentrations of the bacteriocins that yielded clear zones of growth inhibition in the indicator lawns.
Purification of the sakacins from Lactobacillus sakei D98 was performed according to a previously described protocol with some modifications (Sawa et al. 2012). Using a 3-step procedure, the bacteriocins were purified from a 1-l culture of Lact. sakei D98 grown to the stationary phase in MRS medium at 30°C for 16 h. The cells were removed by centrifugation at 8000 × g for 20 min, and 20 g of activated Amberlite XAD-16 resin (Sigma-Aldrich, St. Louis, MO, USA) was added to the supernatant. The resin matrix was shaken at slow speed for 3 h. The resin matrix was washed with 50 ml of Milli-Q water and 100 ml of 30% (v/v) ethanol. An active fraction was eluted with 150 ml of 70% (v/v) isopropanol containing 0·1% trifluoroacetic acid. To remove isopropanol, the eluted active fraction was evaporated to 45 ml. Then, the sample was diluted with an equal volume of 25 mmol l−1 sodium phosphate buffer (pH 5·7, PB). The sample was applied to an SP Sepharose Fast Flow cation exchange chromatography column (length, 70 mm, and internal diameter, 10 mm; GE Healthcare, Uppsala, Sweden) equilibrated with 50 ml of PB. After the column was washed with 50 ml PB, the active fraction containing bacteriocins was eluted with 70 ml of 0·5 mol l−1 NaCl in PB. This active fraction was applied to a RESOURCE RPC 3-ml column (GE Healthcare) in an LC-2000Plus high-performance liquid chromatography (HPLC) system (JASCO, Tokyo, Japan). Active fractions were eluted in the following manner with a gradient of Milli Q-acetonitrile containing 0·1% trifluoroacetic acid at a flow rate of 1·0 ml min−1: 0–5 min, 0–20% (v/v); 5–30 min, 20–60% (v/v); 30–35 min, 60–100% (v/v); 35–40 min, 100% of acetonitrile. Purified active fractions were stored at −30°C. The antibacterial activity of the fractions obtained in each purification step was determined as described above using Lact. sakei ssp. sakei JCM 1157T as the indicator strain. The peptide concentration of each fraction was estimated by the Pierce® BCA™ Protein Assay Kit (Takara Bio, Shiga, Japan). For MIC determination and characterization, purified bacteriocins were concentrated using a SpeedVac concentrator (Savant, Farmingdale, NY, USA) and dissolved at appropriate concentrations with 0·1% (v/v) Tween 80 unless mentioned otherwise.
The molecular masses of bacteriocins were analysed by electrospray ionization time-of-flight mass spectrometry (ESI-TOF MS) with a JMS-T100LC mass spectrometer (JEOL, Tokyo, Japan). The amino acid sequences were determined based on Edman degradation with a PPSQ-21 protein sequencer (Shimadzu, Kyoto, Japan). To cleave disulfide bonds and detect cysteine residues in Edman degradation, the peptide was chemically modified according to the methods of Friedman et al. (1970) and Katsumi et al. (2000), as described by Sawa et al. (2012).
To obtain the genes encoding bacteriocins, PCR and sequencing analysis were performed using primers listed in Table 1. To determine the genes encoding sakacins D98a, D98b and D98c, short degenerated primers (DA.F1-R3, DB.F1-R3 and DC.F1-R3) for nested anchored rapid PCR (NAR-PCR) were designed on the basis of the amino acid sequences obtained (Sawa et al. 2012).
|Primer name||Corresponding amino acid sequence||Sequence (5′-3′)|
Total DNA was extracted from Lact. sakei D98 cells treated with lysozyme (Seikagaku, Tokyo, Japan) and cetyltrimethylammonium bromide (Wako, Osaka, Japan) according to previously described procedures (Sawa et al. 2010). Total DNA of Lact. sakei D98 was digested with BamHI, EcoRI, HindIII, KpnI, SacI or XbaI (Nippon Gene, Tokyo, Japan), and the digested DNA was ligated into pUC18 cloning vector (Toyobo, Osaka, Japan) treated with the corresponding restriction enzymes and dephosphorylated. Each ligation product was then used as a template for NAR-PCR using short degenerate primers (DA, DB and DC) and vector-specific primers (Mup13 primers) with TaKaRa Ex Taq (Takara Bio; Sawa et al. 2012). This yielded partial sequences of the structural genes of the sakacins. Moreover, to amplify the upstream and downstream regions of the structural genes of sakacins D98a, D98b and D98c, anchored PCR was performed using the above-mentioned ligated products, sakacin-specific primers (SA primers) and vector-specific primers (M13 primers) with Taq DNA polymerase (Promega, Madison, WI, USA; Sambrook et al. 1989).
The fragments obtained were purified using a QIAquick PCR purification kit (Qiagen, Hilden, Germany) and sequenced after TA cloning (Sawa et al. 2010). Fragments encoding each entire sakacin (D98a, D98b and D98c) were amplified by PCR using newly designed specific primers (SAA.F1-R2, SAB.F1-R2 and SAC.F1-R2, respectively) based on the obtained nucleotide sequences shown previously.
The products were purified and directly sequenced to confirm the sequences obtained. DNA sequencing was carried out by FASMAC Co., Ltd (Kanagawa, Japan). The DNA and amino acid sequences obtained were analysed using GENETYX-WIN software (GENETYX, Tokyo, Japan). Database searches were performed using the National Center for Biotechnology Information BLAST tool (http://blast.ncbi.nlm.nih.gov/Blast.cgi). Alignments of amino acid sequences were analysed using ClustalW2 (http://www.ebi.ac.uk/Tools/msa/clustalw2/). The putative secondary structures were analysed using The PSIPRED Protein Structure Prediction Server (http://bioinf.cs.ucl.ac.uk/psipred/).
Bacteriocins from Lact. sakei D98 were purified in three steps as shown in Table 2. In the final step of reverse-phase HPLC, antimicrobial activities were detected at retention times of about 21, 22 and 27 min, indicated as peaks A, B and C in Fig. 1, respectively. Finally, three active peptides were purified with a total yield of 1·7% (Table 2). The molecular masses of the purified bacteriocins were determined to be 3660·3, 4692·8 and 3558·3 Da for peaks A, B and C by ESI/MS, respectively (Fig. 2).
|Step||Volume (ml)||Total activity (AU)a||Total protein (mg)b||Specific activity (AU mg−1)||Purification (Fold)||Yield (%)|
|Supernatant||1000||5·12 × 107||2800||1·83 × 104||1||100|
|Amberlite XAD-16||150||4·10 × 107||820||5·00 × 104||2· 73||80|
|SP-Sepharose||70||3·28 × 107||34||9·65 × 105||52·7||64|
|Reverse-phase HPLC (total)c||20||8·80 × 105||0·7||1·26 × 106||68·9||1·7|
The sequences of 34 and 33 N-terminal amino acid residues were obtained by Edman degradation from peaks A and C as follows: S1FVGTTIGKIIGHSITHYGNMVSQQHLHKSPING34 and A1AKGFIGWLIGESDDIAKGIVSAMNSDNKKKHR33, respectively. The calculated molecular masses of the sequences from peaks A and C were 3660·1 and 3558·4 Da, respectively. The calculated sizes agreed well with the observed ones. These peptides from peaks A and C, termed sakacin D98a and sakacin D98c, respectively, belong to class IId bacteriocins. In addition, the amino acid sequences of sakacins D98a and D98c showed 39 and 56% identity with those of leucocin K and a putative peptide produced by Lact. paraplantarum (Lee et al., unpublished observations, 2006, accession No. AAL77871.2 and No. AAL77871.1), respectively.
The peptide in peak B was treated with reduction and pyridylethylation, and sequenced. A part of the N-terminal amino acid sequence of the peptide in peak B was obtained by Edman degradation as follows: K1ITHYANGVSCSTKTGTCSVNWNQAASGIGKIIV.34 The calculated molecular weight of this amino acid sequence was 3509·9 Da, which indicated that the C-terminal part was not sequenced. This peptide, termed sakacin D98b, had an N-terminal region containing the sequence YANGVSC, similar to the pediocin box, and a database search of the sequence proved sakacin D98b is a new bacteriocin.
Partial sequences of the structural genes of each sakacin were obtained by NAR-PCR. Based on the DNA sequences obtained, specific primers were designed and employed for ligation-anchored and nested PCR to obtain the adjacent sequences. As a result, the DNA sequences of the genes of sakacins D98a, D98b and D98c were obtained (Fig. 3). The structural genes, termed saaA, saaB and saaC, encoded 57, 63 and 51 amino acid residues of prepeptides whose leader peptides contained double glycine cleavage sites to produce 34, 45 and 33 amino acid mature peptides, respectively (Table 3). In addition, based on the DNA sequence obtained, the calculated molecular mass of sakacin D98b was 4695·2 Da (Table 3). The observed molecular mass was about 2 Da less than this calculated value. This result suggested that sakacin D98b possesses a disulfide bridge between Cys11 and Cys18. The amino acid sequence of the sakacin D98b mature peptide showed 53·3% (24/45) identity to that of the duracin GL mature peptide produced by Ent. faecalis YI717 (Du et al. 2012). Furthermore, the structural genes of sakacins D98a and D98c were encoded in adjacent open reading frames (ORFs) in the same operon, whereas sakacin D98b was located far from them (Fig. 3).
|Bacteriocin||MW (Da)||Amino acid sequence of prepeptide|
|Observed molecular weight||Calculated molecular weight||Leader peptide||Mature peptide|
The inhibitory spectra of the culture supernatant of Lact. sakei D98 and purified sakacins are presented in Table 4. Among the three novel sakacins, sakacin D98b showed the strongest and broadest inhibitory activity against indicator strains. Sakacin D98a showed notable inhibitory activity against Ent. faecalis and Lact. sakei. Sakacin D98c showed inhibitory activity only against Lact. sakei. In addition, these three purified novel bacteriocins did not show any synergistic activity in any combination (equimolar mixtures of two peptides within 10 μmol l−1) against indicator strains in Table 4, and they exhibited no inducing activity for bacteriocin production (data not shown).
|Indicator strain||Activity in culture supernatant||MIC (μmol l−1)|
|(AU ml−1)||Sakacin D98a||Sakacin D98b||Sakacin D98c|
|Bacillus coagulans JCM 2257T||0||N.A.||N.A.||N.A.|
|B. circulans JCM 2504T||100||49·1||0·49||N.A.|
|Kocuria rhizophila NBRC 12708||0||N.A.||N.A.||N.A.|
|Listeria innocua ATCC 33090T||200||N.A.||0·49||N.A.|
|L. monocytogenes ATCC BAA-679||200||N.A.||0·49||N.A.|
|Pediococcus pentosaceus JCM 5885||0||12·3||0·25||N.A.|
|Enterococcus faecalis JCM 5803T||400||6·15||0·25||N.A.|
|Lactococcus lactis ssp. lactis ATCC 19435T||0||N.A.||N.A.||N.A.|
|Lactobacillus plantarum ATCC 14917T||0||N.A.||N.A.||N.A.|
|Lact. sakei ssp. sakei JCM 1157T||12800||0·31||0·03||2·65|
|Leuconostoc mes. ssp. mes.JCM 6124Ta||0||N.A.||31·9||N.A.|
LAB are important micro-organisms in the food industry, being common starter organisms used in fermented products. Lactobacillus sakei D98 was isolated from rice malt, Shubo, which forms the starter culture of yeasts and LAB in the production of sake, Japanese traditional rice wine. This strain produced three novel bacteriocins, sakacin D98b with a pediocin box-like sequence, and sakacins D98a and D98c belonging to class IId bacteriocins, which are translated as prepeptide substrates with conserved double glycine-type cleavage sites. Lactobacillus sakei D98 might have an important role in controlling microflora by the production of organic acids and/or bacteriocins in Shubo.
Sakacins D98a and D98c belonging to class IId bacteriocins showed 39 and 56% identity with leucocin K and a putative peptide, respectively. They have some characteristics of class IIb bacteriocins, such as the tandem arrangement in the same direction of their structural genes, and the amphiphilic α-helix regions, which were predicted to lie between Ile16 and Ser26 of sakacin D98a and between Gly19 and Glu24 of sakacin D98c by PSIPRED. However, sakacins D98a and D98c did not show any synergistic activity, which is observed in class IIb bacteriocins. They also did not exhibit any bacteriocin production inducing activity. In addition, sakacins D98a and D98c showed significantly different inhibitory spectra, indicating that these sakacins have different modes of action (Table 4).
Analyses of amino acid and DNA sequences showed that sakacin D98b has the pediocin box-like sequence YANGV and a disulfide bridge at a different position from those found in general class IIa bacteriocins. Recently, several class IIa bacteriocin mutants with changed consensus sequences were investigated for their modes of action. For example, the sakacin G2 mutant, a mutated sakacin P with the sequence YANGV (Ala instead of Gly), exhibited only a slight reduction in activity compared with other mutant forms of sakacin P (Fimland et al. 2006). The inhibitory activities of the original sakacin P and the sakacin G2 mutant were detected at 0·33 and 0·64 nmol l−1, respectively, against Lact. sakei NCDO 2714T (identical to JCM 1157T). This G2 mutant was predicted to be located near the membrane surface when the N-terminal domain interacts with the target membrane according to the model (Fimland et al. 2006). These results suggested that position 2 of the pediocin box was not significantly effective for inhibitory activity. Furthermore, the C-terminal region of sakacin D98b showed similarity to some class IIa bacteriocins with a YGNGV or YGNGL motif. In particular, the whole sequences of the three bacteriocins, bacteriocin 31 (bacA) (Tomita et al. 1996), durancin GL (Du et al. 2012) and divercin V41 (Metivier et al. 1998), showed significant identity (51·1, 53·3 and 28·9%, respectively) to sakacin D98b (Fig. 4).
In addition to differences in the pediocin box, sakacin D98b has a single disulfide bond at a position different from that seen in class IIa bacteriocins. For example, a typical class IIa bacteriocin, mesentericin Y 10537, contains a single intramolecular disulfide bridge that includes 4 amino acid residues between the 2 Cys residues (CXXXXC; Fleury et al. 1996; Derksen et al. 2008). On the N-terminal side of the putative amphipathic helix, the 6-membered disulfide loop linking Cys9 and Cys14 contributes to the compact structure of mesentericin Y 10537. The helical domain of the peptide interacts with the lipid bilayer, leading subsequently to the disruption of the membrane, and residues 1–14 form part of a recognition structure for a membrane-bound receptor, which may be critical for peptide targeting.
On the other hand, sakacin D98b had a single disulfide bridge that includes six amino acid residues between the 2 Cys residues (C11STKTGTC18) (Fig. 4). In addition, sakacin D98b has a putative amphipathic helix at the N-terminal region between K1 and S10, as predicted by PSIPRED. However, the inhibitory activity of sakacin D98b against Listeria was lower than that of mesentericin Y 10537 and leucocin A (Fleury et al. 1996; Sawa et al. 2010), whereas sakacin D98b showed higher inhibitory activity against Lact. sakei and Pediococcus pentosaceus than did leucocin A (Sawa et al. 2010). These differences in antimicrobial activity and spectra, especially against Listeria, are probably attributed to differences in the amino acid sequences as well as the disulfide bridge positions.
The results presented indicate that these newly discovered class IIa-like and IId bacteriocins likely have different antimicrobial mechanisms. In addition, the different amino acid sequences and disulfide bridge positions could cause the observed differences in antibacterial spectra and intensities. These indicate that the use of bacteriocins from strain D98 in combination results in a wide antimicrobial spectrum and reduces the appearance of bacteriocin-resistant bacteria. Strain D98 produces lactic acid and bacteriocins, which prevent the growth of undesired bacteria in Shubo. These beneficial characteristics may extend the further possibility of utilizing bacteriocins as biopreservatives and bacteriocin-producing LAB as starter cultures in various fermented foods.
This work was partially supported by Grants-in-Aid for Scientific Research from the Japan Society for the Promotion of Science. The authors thank Inoue Goumei Co. Ltd. for providing Shubo, the isolation source, and also thank Takahiro Oba and Kyoko Ueda of Biotechnology and Food Research Institute, Fukuoka Industrial Technology Center for their assistance in isolating the strain.