Purification and characterization of the bacteriocin produced by Lactobacillus sakei MBSa1 isolated from Brazilian salami

Authors

  • M.S. Barbosa,

    Corresponding author
    1. Departamento de Alimentos e Nutrição Experimental, Faculdade de Ciências Farmacêuticas, Universidade de São Paulo, São Paulo, Brazil
    • Correspondence

      Matheus de Souza Barbosa, Department of Food and Experimental Nutrition, Faculty of Pharmaceutical Sciences, University of São Paulo, 580, Professor Lineu Prestes, 13 B, Sao Paulo, SP, 05508-000, Brazil.

      E-mail: matheusbarbosa@usp.br

    Search for more papers by this author
  • S.D. Todorov,

    1. Departamento de Alimentos e Nutrição Experimental, Faculdade de Ciências Farmacêuticas, Universidade de São Paulo, São Paulo, Brazil
    Search for more papers by this author
  • Y. Belguesmia,

    1. Institut National de la Recherche Agronomique, UR 1268 Biopolymères Interactions Assemblages, Equipe Fonctions et Interactions des Protéines, Nantes, France
    Search for more papers by this author
  • Y. Choiset,

    1. Institut National de la Recherche Agronomique, UR 1268 Biopolymères Interactions Assemblages, Equipe Fonctions et Interactions des Protéines, Nantes, France
    Search for more papers by this author
  • H. Rabesona,

    1. Institut National de la Recherche Agronomique, UR 1268 Biopolymères Interactions Assemblages, Equipe Fonctions et Interactions des Protéines, Nantes, France
    Search for more papers by this author
  • I.V. Ivanova,

    1. Institut National de la Recherche Agronomique, UR 1268 Biopolymères Interactions Assemblages, Equipe Fonctions et Interactions des Protéines, Nantes, France
    2. Department of Microbiology, Sofia University, Sofia, Bulgaria
    Search for more papers by this author
  • J.-M. Chobert,

    1. Institut National de la Recherche Agronomique, UR 1268 Biopolymères Interactions Assemblages, Equipe Fonctions et Interactions des Protéines, Nantes, France
    Search for more papers by this author
  • T. Haertlé,

    1. Institut National de la Recherche Agronomique, UR 1268 Biopolymères Interactions Assemblages, Equipe Fonctions et Interactions des Protéines, Nantes, France
    Search for more papers by this author
  • B.D.G.M. Franco

    1. Departamento de Alimentos e Nutrição Experimental, Faculdade de Ciências Farmacêuticas, Universidade de São Paulo, São Paulo, Brazil
    Search for more papers by this author

Abstract

Aims

The study aimed at determining the biochemical characteristics of the bacteriocin produced by Lactobacillus sakei MBSa1, isolated from salami, correlating the results with the genetic features of the producer strain.

Methods and Results

Identification of strain MBSa1 was performed by 16S rDNA sequencing. The bacteriocin was tested for spectrum of activity, heat and pH stability, mechanism of action, molecular mass and amino acid sequence when purified by cation-exchange and reversed-phase HPLC. Genomic DNA was tested for bacteriocin genes commonly present in Lact. sakei. Bacteriocin MBSa1 was heat-stable, unaffected by pH 2·0 to 6·0 and active against all tested Listeria monocytogenes strains. Maximal production of bacteriocin MBSa1 (1600 AU ml−1) in MRS broth occurred after 20 h at 25°C. The molecular mass of produced bacteriocin was 4303·3 Da, and the molecule contained the SIIGGMISGWAASGLAG sequence, also present in sakacin A. The strain contained the sakacin A and curvacin A genes but was negative for other tested sakacin genes (sakacins T-α, T-β, X, P, G and Q).

Conclusions

In the studied conditions, Lact. sakei MBSa1 produced sakacin A, a class II bacteriocin, with anti-Listeria activity.

Significance and Impact of the Study

The study covers the purification and characterization of the bacteriocin produced by a lactic acid bacteria isolated from salami (Lact. sakei MBSa1), linking genetic and expression information. Its heat-resistance, pH stability in acid conditions (pH 2·0–6·0) and activity against L. monocytogenes food isolates bring up a potential technological application to improve food safety.

Introduction

Several preservation technologies can be used to ensure that foods maintain an acceptable level of quality from manufacture until consumption (Zhou et al. 2010). Fermentation is a millennial process used to extend the shelf life of easily perishable products such as raw meat (Rantsiou and Cocolin 2006). The manufacturing process of many meat products includes a fermentation step, performed under conditions that inhibit the growth of several spoilage and pathogenic bacteria. However, few pathogens, such as Listeria monocytogenes, can survive in fermented products and become a health hazard (Thévenot et al. 2005).

The use of natural antimicrobials as food preservatives is receiving increased attention, because they are a promising tool for improvement in food safety and may replace or reduce the use of chemical additives (Juneja et al. 2012). Among these antimicrobial compounds, bacteriocins produced by lactic acid bacteria (LAB) that target pathogenic bacteria without toxic or other adverse effects for consumers are under intensive investigation (O'Shea et al. 2013). Many bacteriocins produced by LAB were already described, and they vary in spectrum of their activities (narrow or broad), modes of action, molecular masses and genetic and biochemical properties (Nishie et al. 2012).

Fermented sausages contain many species of LAB, and several studies have shown that some of them may produce bacteriocins (Table 1). However, according to our knowledge, there is no report on the occurrence of such type of LAB in similar fermented meat products in Brazil. This survey aimed at isolating bacteriocin-producing LAB strains in salami samples collected on the Brazilian market and determining the biochemical characteristics of the bacteriocin produced by the isolate Lactobacillus sakei MBSa1, correlating the results with the genetic features of the producer strain.

Table 1. Bacteriocinogenic lactic acid bacteria isolated from fermented meat products
StrainSourceReference

Lactobacillus sakei ST22Ch

Lact. sakei ST153Ch

Lact. sakei ST154Ch

SalpicaoTodorov et al. (2013)
Enterococcus faecium ST211CHLomboTodorov et al. (2012)
Pediococcus acidilactici LAB 5 Vacuum-packed fermented meat productMandal et al. (2011)
Lactobacillus plantarum LP 31Argentinian dry-fermented sausageMüller et al. (2009)

Pediococcus acidilactici HA-6111-2

Pediococcus acidilactici HA-5692-3

Portuguese fermented sausageAlbano et al. (2007)
Lact. plantarum N014Thai fermented porkPhumkhachorn et al. (2007)
Lact. sakei I151Fermented sausagesUrso et al. (2006)
Lactococcus lactis WNC 20Thai fermented sausageNoonpakdee et al. (2003)
Lact. sakei CTC494Fermented sausageAymerich et al. (2000)

Lactobacillus curvatus LTH 1174

Lact. sakei LTH 673

Fermented sausageTichaczek et al. (1992)

Material and methods

Search for LAB with anti-Listeria activity in salami

Salami samples (50 g), collected on local markets in the city of Sao Paulo (Brazil), were homogenized in a stomacher (Seward 400, London, UK) with 450 ml of 0·1% sterile peptone water (Difco, Detroit, MI) and submitted to subsequent decimal dilutions in 0·1% sterile peptone water (Difco). Each dilution was plated on MRS agar (Oxoid) in duplicates and incubated 48 h at 30°C. Growing colonies were randomly selected and tested for inhibitory activity against L. monocytogenes Scott A by the triple-layer method (Todorov and Dicks 2005). In this method, plates of MRS agar presenting isolated colonies are overlaid with approximately 5 ml of semi-solid BHI medium (BHI broth (Oxoid) supplemented with 0·75% bacteriologic agar (Oxoid)) containing L. monocytogenes Scott A (105–106 CFU ml−1) and incubated for 24 h at 37°C. Colonies presenting growth inhibition zones around them were transferred to MRS broth (Difco), incubated for 24 h at 30°C and then plated on MRS agar (Oxoid) and incubated for 24 h at 30°C. Isolated colonies were submitted to Gram staining and tested for catalase production using 3% hydrogen peroxide (v/v). Gram-positive and catalase-negative cultures presenting anti-Listeria activity were freeze-dried and stored at −20°C.

Strains presenting anti-Listeria activity were grown in MRS broth (Difco) for 24 h at 30°C and submitted to centrifugation at 4000 g for 15 min at 4°C (model Mikro 22R; Hettich Zentrifugen, Tuttlingen, Germany). The pH of the obtained cell-free supernatant (CFS) was adjusted to 6·0–6·5 with 1 mol l−1 NaOH (Synth, Sao Paulo, Brazil), heated 30 min at 70°C and sterilized by filtration (Millex GV 0·22 μm; Millipore, Billerica, MA). Anti-Listeria activity of the CFS was tested by the spot-on-the-lawn method (van Reenen et al. 1998) with modifications. An aliquot of 10 μl of CFS was spotted onto the surface of a plate containing 10–12 ml of 1·5% bacteriologic agar (Difco), overlaid with 5 ml of BHI semi-solid agar (BHI broth (Oxoid) added of 0·85% (w/v) bacteriological agar (Oxoid)) containing L. monocytogenes Scott A (105–106 CFU ml−1). The plates were incubated at 37°C for 12 h and observed for the formation of clear zones of inhibition around the spotted CFS. Bacteriocin production was confirmed by testing the proteinaceous nature of the antimicrobial compound. For this test, the CFS was treated (1 h at 37°C) with the following proteolytic enzymes (0·1 mg ml−1): α-chymotrypsin from bovine pancreas type II, Streptomyces griseus protease type XIV, trypsin and proteinase K (all from Sigma-Aldrich, St. Louis, MO) solubilized in 20 mmol l−1 phosphate buffer pH 7 (Noonpakdee et al. 2003). After treatment, CFS was heated at 90°C for 5 min for enzyme inactivation and tested for residual antimicrobial activity by the spot-on-the-lawn method (van Reenen et al. 1998). Absence of zone of inhibition after enzymatic treatment indicated the presence of bacteriocin(s). Control tests with nontreated CFS were also performed.

Identification of bacteriocin-producing LAB isolates

Bacteriocin-producing LAB isolated from the salami samples were submitted to 16S rDNA sequence analysis, by amplification of genomic DNA with primers 8f (5′-CAC GGA TCC AGA CTT TGA T(C/T)(A/C) TGG CTC AG-3′) and 1512r (5′- GTG AAG CTT ACG G(C/T)T AGC TTG TTA CGA CTT-3′) as described by Felske et al. (1997). The 20 μl reaction volume contained 100 pmol l−1 each primer, 1X PCR buffer (New England BioLabs, Ipswich, MA), 24 μmol l−1 dNTP (Fermentas, Hanover, MD), 2 mmol l−1 MgCl2 (Fermentas) and 0·0125 U Taq DNA polymerase (New England BioLabs). Amplification was carried out in a DNA MasterCycler® (Eppendorf Scientific, Hamburg, Germany). PCR conditions included denaturation at 94°C for 5 min, followed by 35 cycles of denaturation at 94°C for 10 s, primer annealing at 61°C for 20 s and polymerization at 68°C for 2 min and then at 72°C for 7 min. PCR-amplified DNA fragments were separated by 0·8% (w/v) agarose gel electrophoresis and visualized by staining with ethidium bromide (0·1 mg ml−1). Fluorescent bands of approximately 831 bp were made visible using an UVP BioImaging System (DIGIDOC-IT System, Upland, CA). The bands were purified with QIAquick® PCR Purification Kit (Qiagen, Hilden, Germany) following the manufacturer's instructions and submitted to amino acid sequencing at the Center for Human Genome Studies, Institute of Biomedical Sciences, University of Sao Paulo, Brazil. The sequences were compared to those deposited in GenBank, using the blast algorithm (http://www.ncbi.nlm.nih.gov/BLAST). The identifications of species were confirmed by species-specific PCR amplification assays as described by Berthier and Ehrlich (1998), using primers Ls-F (ATG AAA CTA TTA AAT TGG TA) and Ls-R (GCT GGA TCA CCT CCT TTC C). The PCRs were performed with 1X PCR buffer (New England BioLabs), 25 μmol l−1 dNTP (Fermentas), 100 μmol l−1 MgCl2 (Fermentas) and 0·025 U Taq DNA polymerase (New England BioLabs). PCR conditions were denaturation at 94°C for 5 min followed by 35 cycles of denaturation at 94°C for 1 min, annealing at 36°C for 30 s, polymerization at 72°C for 1 min and a final polymerization at 72°C for 5 min. PCR-amplified DNA fragments were separated by 2% (w/v) agarose gel electrophoresis and visualized by treatment with ethidium bromide (0·1 mg ml−1) and made visible using a UVP BioImaging System (DIGIDOC-IT System). Strain MBSa1, identified as Lact. sakei, presented a good anti-Listeria activity and therefore was selected for genetic and biochemical characterization of the bacteriocin.

Titration of the bacteriocin produced by strain MBSa1

The amount of bacteriocin produced by strain MBSa1 was determined using twofold dilutions and the spot-on-the-lawn method described by van Reenen et al. (1998). One arbitrary unit (AU) was defined as the reciprocal of the highest dilution that resulted in production of a clear zone of inhibition of L. monocytogenes Scott A. Results were expressed in AU ml−1 (Kaiser and Montville 1996).

Effect of pH and temperature on activity of bacteriocin MBSa1

Both tests were carried out as described by Albano et al. (2007). The pH of the CFS was adjusted to 2·0, 4·0, 6·0, 8·0 and 10·0 with concentrated phosphoric acid (Synth) or 1 mol l−1 NaOH (Synth) and tested for activity against L. monocytogenes Scott A after 1 h at 25°C. For the antilisterial tests, the pH of the CFS was adjusted to 6·0–6·5 with 1 mol l−1 NaOH (Synth) or concentrated phosphoric acid (Synth). The effect of temperature on the activity of the bacteriocin was evaluated by keeping the CFS at 4, 25, 30, 37, 45, 60, 80 and 100°C for 60 min and at 121°C for 15 min and then testing for activity against L. monocytogenes Scott A.

Spectrum of activity of bacteriocin MBSa1

The antimicrobial activity of the CFS containing the bacteriocin produced by strain MBSa1 was determined against the Gram-negative and Gram-positive bacteria isolated from foods, listed in Table 2, using the spot-on-the-lawn test (van Reenen et al. 1998). Lactobacilli and enterococci were grown in MRS broth (Difco) at 30°C for 24 h, and the other strains were grown in BHI broth (Oxoid) at 37°C for 24 h.

Table 2. Spectrum of activity of the bacteriocin produced by Lactobacillus sakei MBSa1
Indicator microorganismSourceActivitya
  1. a

    – no inhibitory activity; + inhibition halo diameter 1–10 mm; ++ inhibition halo diameter 11–15 mm; +++ inhibition halo diameter >20 mm.

  2. b

    Food Microbiology Laboratory, Faculty Pharmaceutical Science, University of Sao Paulo (USP), Sao Paulo, Brazil.

  3. c

    Bacterial Zoonoses Laboratory, Oswaldo Cruz Institute (FIOCRUZ), Rio de Janeiro, Brazil.

  4. d

    Department for Research in Animal Production, AGRIS, Sardegna, Olmedo, Italy.

  5. e

    Science and Food Technology Institute, School Biology, Central University of Venezuela (UCV), Caracas, Venezuela.

Bacillus cereusATCC 1178
Staphylococcus aureusATCC 29213
Staph. aureusATCC 25923
Staph. aureusATCC 6538
Listeria welshimeriUSPb+
Listeria seeligeriUSP
Listeria ivanovii ssp. ivanoviiATCC 19119++
Listeria innocuaATCC 33090++
L. innocua 225/07 serovar 6aFIOCRUZc+
L. innocua 224/07 serovar 6aFIOCRUZ+
L. innocua 047/07 serovar 6aFIOCRUZ+
L. innocua 588/08 serovar 6aFIOCRUZ+
Listeria monocytogenes Scott AUSP+
L. monocytogenes 602/08 serovar 1/2aFIOCRUZ+
L. monocytogenes 046/07 serovar 1/2cFIOCRUZ+
L. monocytogenes 103 serovar 1/2aUSP+
L. monocytogenes 106 serovar 1/2aUSP+
L. monocytogenes 104 serovar 1/2aUSP+
L. monocytogenes 409 serovar 1/2aUSP+
L. monocytogenes 506 serovar 1/2aUSP+
L. monocytogenes 709 serovar 1/2aUSP+
L. monocytogenes 607 serovar 1/2bUSP+
L. monocytogenes 603 serovar 1/2bUSP+
L. monocytogenes 426 serovar 1/2bUSP+
L. monocytogenes 637 serovar 1/2cUSP+
L. monocytogenes 422 serovar 1/2cUSP+
L. monocytogenes 712 serovar 1/2cUSP+
L. monocytogenes 408 serovar 1/2cUSP+
L. monocytogenes 211 serovar 4bUSP+
L. monocytogenes 724 serovar 4bUSP+
L. monocytogenes 101 serovar 4bUSP+
L. monocytogenes 703 serovar 4bUSP+
L. monocytogenes 620 serovar 4bUSP+
L. monocytogenes 302 serovar 4bUSP+
Escherichia coliATCC 8739
E. coli O157:H7ATCC 35150
Enterobacter aerogenesATCC 13048
Salmonella TyphimuriumATCCC 14028
Salmonella EnteritidisATCC 13076
Enterococcus faecalisATCC 12755+
Enterococcus hirae D105AGRISd+
Enterococcus faecium S5AGRIS
Ent. faecium S154AGRIS
Ent. faecium S100AGRIS+
Ent. faecium ST62AGRIS
Ent. faecium ST211AGRIS
Ent. faecium ET 12UCVe
Ent. faecium ET 88UCV
Ent. faecium ET 05UCV
Lactobacillus sp. V94USP
Lactobacillus fermentum ET35UCV
Pediococcus pentosaceus ET 34UCV
Lactobacillus curvatus ET 06UCV
Lact. curvatus ET 31UCV
Lact. curvatus ET 30UCV
Lact. sakei ssp. sakei 2aUSP
Lact. sakeiATCC 15521+
Lactobacillus plantarum V69USP
Lactobacillus delbrueckii B5USP
Lact. delbrueckii ET32UCV
Lactobacillus acidophilus La14DuPont
Lact. acidophilus Lac4DuPont
Lact. acidophilus La5Chr. Hansen
Lactococcus lactis B16USP
Lc. lactis ssp. lactis MK02RUSP
Lc. lactis ssp. lactis D2USP
Lc. lactis ssp. lactis B1USP
Lc. lactis ssp. lactis D4USP
Lc. lactis ssp. lactis B2USP
Lc. lactis ssp. lactis B15USP
Lc. lactis ssp. lactis D3USP
Lc. lactis ssp. lactis D5USP
Lc. lactis ssp. lactis B17USP
Lc. lactis ssp. lactis R704Chr. Hansen

Effect of temperature on growth and bacteriocin production by strain MBSa1

Growth and production of bacteriocin by strain MBSa1 in MRS Broth (Difco) were evaluated at 25, 30 and 37°C, using the spot-on-the-lawn method (van Reenen et al. 1998). Growth was monitored every 2 h up to 24 h, measuring absorbance at 600 nm (Ultrospec 2000; Pharmacia Biotech, Little Chalfont, UK).

Search for bacteriocin genes

The MBSa1 strain was investigated for the presence of known sakacin and curvacin A genes using PCR and the primers listed in Table 3. Total DNA was extracted and submitted to amplification in a reaction mixture (20 μl) containing approximately 25 ng μl−1 of extracted DNA, 1X PCR buffer (New England BioLabs), 100 μmol l−1 MgCl2 (Fermentas), 200 μmol l−1 dNTPs (Fermentas), 0·025 U Taq polymerase (New England BioLabs) and 1 pmol l−1 each primer. Amplification was achieved in 35 cycles using a DNA thermocycler MasterCycler® PCR (Eppendorf Scientific). PCR conditions are show in Table 3. PCR-amplified DNA fragments were separated by 2% (w/v) agarose gel electrophoresis, stained with ethidium bromide (0·1 mg ml−1) and observed using the UVP BioImaging System (DIGIDOC-IT System). For each primer, the corresponding bands (sizes described in Table 3) were purified with QIAquick® PCR Purification Kit (Qiagen) according to the manufacturer's instructions and submitted to sequencing at the Center for Human Genome Studies, Institute of Biomedical Sciences, University of Sao Paulo, Brazil. The sequences were compared to those deposited in GenBank, using the blast algorithm (http://www.ncbi.nlm.nih.gov/BLAST).

Table 3. Targeted bacteriocin genes and primers used in the study
Target bacteriocinPrimeraSequence (5′–3′)bSize (bp)ReferenceInitial denaturationDenaturationAnnealingElongation
  1. a

    F- Forward; R- Reverse.

  2. b

    deoxyinosine; R, T/A; W, A/G; S, C/A; Y, A/C; D, C/T; N, A/C.

Sakacin ASakA-FGAAWTRMMANCAATTAYMGGTGG150Dortu et al. (2008)95°C, 15 min95°C, 30 s55°C, 1 min72°C, 1 min
SakA-RCAGCCGCTAATCATACCACC
Sakacin T-αSakT-α -FTCGGTGGCTATACTGTCTAAACA160Macwana and Muriana (2012)95°C, 15 min95°C, 30 s58°C, 1 min72°C, 1 min
SakT-α -RTGTCCTAAAAATCCACCAATGC
Sakacin T-βSakT-β -FAAGAAATGATAGAAATTTTTGGAGG151Macwana and Muriana (2012)95°C, 15 min95°C, 30 s56°C, 1 min72°C, 1 min
SakT-β -RTGTGAAATCCAATCTTGTCCTG
Sakacin QSakQ-FGAA RTW SYA NCA ATT ADN GGT GG130Dortu et al. (2008)95°C, 15 min95°C, 30 s53°C, 1 min72°C, 1 min
SakQ-RTAC CAC CAG CAG CCA TTC CC
Sakacin XSakX-FAGCTATGAAAGGTATTGTCGGG156Macwana and Muriana (2012)95°C, 15 min95°C, 30 s58°C, 1 min72°C, 1 min
SakX-RTAAGATTTCCAGCCAGCAGC
Sakacin PSakP-FATG GAA AAG TTT ATT GAA TTA186Reminger et al. (1996)94°C, 3 min94°C, 30 s40°C, 1 min72°C, 1 min
SakP-RTTA T TT ATT CCA GCC AGC GTT
Sakacin GSakGA1-FTTA GAA CTA CAC TGA TCG TG Todorov et al. (2011)94°C, 4 min94°C, 30 s38°C, 30 s72°C, 30 s
 SakGA1-RTGG AAG AAT GAG TAC TTG TT
Sakacin GSakGA2-FCGT TAC AAC AGA ACT TCA AG Todorov et al. (2011)94°C, 4 min94°C, 30 s38°C, 30 s72°C, 30 s
SakGA2-RTGG AAG AAT GAG TAC TTG TT
Curvacin ACurA-FGTA AAA GAA TTA AGT ATG ACA171Reminger et al. (1996)94°C, 3 min94°C, 30 s40°C, 1 min72°C, 1 min
CurA-RTTA CAT TCC AGC TAA ACC ACT

Purification of bacteriocin MBSa1

Bacteriocin MBSa1 was purified according to Batdorj et al. (2006), with modifications. MRS broth (Biokar, Beauvais, France) was inoculated with a 1% (v/v) overnight culture of MBSa1 strain, and after 18 h at 25°C, cells were removed by centrifugation at 6000 g for 15 min at 4°C (Centrifuge GR 2022, Jouan, France). The pH of the CFS was adjusted to 6·8 with 10 mol l−1 NaOH (Euromedex, Souffelweyersheim, France) and loaded into a SP-Sepharose Fast Flow cation-exchange column (GE Healthcare, Amersham, Uppsala, Sweden) equilibrated with 20 mmol l−1 phosphate (Sigma-Aldrich) buffer pH 6·8 (buffer A). The column was washed with buffer A, and the absorbed substances were eluted with a linear gradient from 0 to 100% buffer B (20 mmol l−1 sodium phosphate + 1 mol l−1 NaCl (Euromedex) pH 6·8). The fractions were collected and tested for anti-Listeria activity using the spot-on-the-lawn test, and Listeria ivanovii ssp. ivanovii ATCC 19119 as indicator of activity.

Active fractions were pooled and loaded into a reversed-phase (RP) column (SOURCE15RPC 10 ml; GE Healthcare) equilibrated with solvent A (0·05% trifluoroacetic acid (TFA) (Sigma-Aldrich), 95% H2O and 5% solvent B (80% acetonitrile (Biosolve, Valkenswaard, Netherlands), 10% isopropanol (Sigma-Aldrich), 10% H2O, 0·03% TFA)). Elution was performed with solvent B with a linear gradient from 0 to 100% for 25 min, at a flow rate of 5 ml min−1. After drying under reduced pressure (Speed-Vac, SC110A, Savant, Holbrook, NY), each fraction was tested for anti-Listeria activity using the spot-on-the-lawn test, using L. ivanovii ssp. invanovii ATCC 19119 as indicator strain.

Fractions presenting activity were pooled and submitted to another purification step by RP-high performance liquid chromatography (RP-HPLC) using Unicorn 3.21 software (Amersham Pharmacia Biotech). The pool was loaded into a preparative C18 column (Symmetry 300™ C18, 5 μm 4·6 × 50 mm Waters, Hertfordshire, UK) equilibrated with solvent C (0·05% TFA, 5% solvent D (80% acetonitrile, 20% H2O, 0·03% TFA), 95% H2O). Elution was performed with solvent D using a linear gradient from 25 to 60% in 35 min, at a flow rate of 6 ml min−1. Peaks were detected by monitoring absorbance at 220 nm. Fractions were collected, dried under vacuum, dissolved in sterile ultra-pure water (Milli-Q; Millipore) and tested for anti-Listeria activity. The protein concentration in this material, corresponding to purified bacteriocin MBSa1, was measured in microtitre plates using Pierce® BCA protein assay kit (Thermo Fisher Scientific, Schwerte, Germany), with albumin (Sigma-Aldrich) as standard.

The molecular mass of the purified bacteriocin MBSa1 was determined in a quadrupole-time-of-flight hybrid mass spectrometer (Q-TOF Global, Waters), equipped with an electrospray ionization (ESI) source and operated in the positive ion mode. Fractions collected from the HPLC chromatography were diluted in a mixture of water and acetonitrile (1 : 1, v/v) acidified with 0·1% formic acid and infused into the mass spectrometer at a continuous flow rate of 5 μl min−1. Following parent mass determination, ions were fragmented in the collision cell of the mass spectrometer and the obtained MS/MS spectra were interpreted to reconstruct the sequence tag of the peptide. This tag was further searched against NCBI databank using the blast software.

Test for disulphide bonds in bacteriocin MBSa1 activity

The presence of disulphide bonds in bacteriocin MBSa1 was checked according to Joerger and Klaenhammer (1986) with modifications. The dried purified bacteriocin MBSa1 was resuspended in 50 mmol l−1 Tris–HCl buffer pH 8·0 and divided into four portions of 100 ml: an aqueous solution of 100 mmol l−1 dithiothreitol (DTT) (Sigma-Aldrich) was added to the first, trypsin (0·1 mg ml−1) was added to the second, proteinase K (0·1 mg ml−1) (controls of proteic character of the studied substance) was added to the third, and the last portion was used as positive control. The mixtures were incubated 1 h at 37°C and checked for anti-Listeria activity by the agar diffusion method.

Determination of Minimal Inhibitory Concentration (MIC) and Minimal Killing Concentration (MKC) of the purified bacteriocin MBSa1

MIC was determined as described by Nielsen et al. (1990) with modifications. The dried purified bacteriocin MBSa1 was re-suspended in 50 mmol l−1 Tris–HCl buffer pH 8·0 and submitted to serial twofold dilutions in 96-well microtitre plates (TPP, Trasadingen, Switzerland) containing 100 μl of BHI broth (Oxoid) in each well. In the next step, 20 μl of an overnight culture of L. monocytogenes Scott A obtained in BHI broth at 37°C was added to each well, achieving 102–103 CFU ml−1 in the wells. For determination of MIC, the microtitre plates were incubated 24 h at 37°C and observed for turbidity in the wells. For determination of MKC, the content of each well was plated on TSA-YE agar plates and checked for growth of colonies. MIC was recorded as the lowest concentration of bacteriocin that resulted in absence of turbidity in the well and MKC was recorded as the lowest concentration of bacteriocin that resulted in absence of growth of L. monocytogenes Scott A in the TSA-YE agar plates in 24 h.

In vitro anti-Listeria activity of the purified bacteriocin MBSa1

The anti-Listeria activity of the purified bacteriocin MBSa1 was tested according to Todorov and Dicks (2004). A 24-h culture of L. monocytogenes Scott A in BHI broth was transferred to fresh BHI broth, and purified bacteriocin MBSa1 at concentration corresponding to the MIC was added to the culture at times 0 h, 6 h (early exponential phase) and 8 h (late exponential phase) and incubated at 37°C. Absorbance measurements (Thermo Fisher Scientific Multiskan®FC) were taken at 595 nm every hour up to 24 h. A culture of L. monocytogenes Scott A without addition of the bacteriocin MBSa1 was used as control.

Results

Several LAB isolated from the studied salami samples presented anti-Listeria activity, indicating that this meat product is a good source for new strains with potential application in the control of undesired micro-organisms in foods. One isolate (MBSa1) was especially active against most tested Listeria strains, mainly L. monocytogenes belonging to different serotypes and isolated from a variety of foods (Table 2). This strain was inactive against the tested Gram-negative bacteria (Salmonella, Escherichia coli and Enterobacter), Bacillus cereus and Staphylococcus aureus. Three of ten strains of Enterococcus spp. were inhibited, and when tested against other species of LAB, a limited antimicrobial activity was observed.

The bacteriocin produced by strain MBSa1 was heat-resistant (Table 4). Full residual activity was observed even after autoclaving during 15 min at 121°C. Frozen storage did not affect activity as well (data not shown). The bacteriocin remained stable at pH 2·0 to 6·0, but lost part of the activity at pH 8·0 and 10·0, with residual activity of 41·6 and 33·6%, respectively. Treatment with proteinase K, trypsin, pepsin, α-chymotrypsin and protease type XIV resulted in total loss of activity (Table 4).

Table 4. Effect of proteolytic enzymes, temperature and pH on activity of the bacteriocin produced by Lactobacillus sakei MBSa1
TreatmentResidual bacteriocin activity (%)
Enzyme
Proteinase K0
Trypsin0
Pepsin0
α-Chymotrypsin0
Protease Type XIV0
Temperature, °C (min)
4 –100 (60)100
121 (15)100
pH
2·0 – 6·0100
8·041·6
10·033·3

Identification based on 16S rDNA sequencing, confirmed by amplification with the species-specific primers, indicated that MBSa1 strain is Lact. sakei (GenBank access number is AB593361.1).

Lactobacillus sakei MBSa1 grew well in MRS broth at 25, 30 and 37°C, causing similar decrease of pH of the medium (Fig. 1). For all tested temperatures, bacteriocin production started in the early exponential growth phase (4 h of incubation). The optimum condition for bacteriocin production (1600 AU ml−1) was 25°C and 20 h of incubation time (Fig. 1).

Figure 1.

Growth (OD 600 nm), bacteriocin production (AU ml−1) and pH reduction of Lactobacillus sakei MBSa1 in MRS broth at 25, 30 and 37°C.

When the DNA extracted from Lact. sakei MBSa1 was tested for bacteriocin genes using primers CurA-F/CurA-R, flanking the curvacin A structural gene (curA) and primers SakA-F/SakA-R, flanking the sakacin A structural gene (sakA), only DNA fragments of 171 bp and 150 bp length were obtained, respectively (Fig. 2). No other structural sakacin genes (Table 3) were detected.

Figure 2.

DNA fragments obtained after PCR with genomic DNA from Lactobacillus sakei MBSa1 using curvacin A-specific primers (CurA-F/CurA-R) (a) and sakacin A-specific primers (SakA-F/SakA-R) (b). Lane 1, molecular weight marker (100 bp); lane 2, amplicon obtained using genomic DNA; lane 3, amplicon obtained using sterile water (control).

The purification sequence, that is, cation exchange followed by sequential hydrophobic-interaction and reversed-phase chromatography, resulted in a stepwise increase in the specific activity (Table 5). The chromatogram of the bacteriocin at the final step of purification (C18 RP-HPLC) presented only one peak at 13 min retention time (Fig. 3). When tested against L. ivanovii, the purified bacteriocin presented a high specific activity (74949·6 AU mg−1). The molecular mass of bacteriocin MBSa1, determined by Q-TOF-MS, was 4303·3 Da. The amino acid sequencing by MS/MS indicated that the molecule contained the SIIGGMISGWASGLAG sequence (Table 6).

Table 5. Purification of bacteriocin produced by Lactobacillus sakei MBSa1
Purification stepVolume (ml)Activity (AU ml−1)Total Activity (AU)Yield (%)Protein (mg ml−1)Specific activity (AU mg−1)Purification factor
Supernatant40064002·56 × 1061003·421871·341·00
Cation exchange19032006·08 × 10523·751·861720·430·92
SOURCE™15RPC70128008·96 × 105351·966530·613·49
C18 RP-HPLC18192008·19 × 1053210·9374949·640·05
Table 6. Amino acid sequence and molecular mass of bacteriocin MBSa1 and other bacteriocins produced by Lactobacillus sakei
BacteriocinSequenceMolecular Mass
  1. The consensus sequence are bold letters.

  2. a

    Vaughan et al. (2001).

  3. b

    Tichaczek et al. (1994).

  4. c

    Holck et al. (1994).

  5. d

    Simon et al. (2002).

  6. e

    Mathiesen et al. (2005).

  7. f

    Kaiser and Montville (1996).

  8. g

    Larsen et al. (1993).

  9. h

    Aymerich et al. (2000).

  10. i

    Tichaczek et al. (1992).

  11. j

    Holck et al. (1992).

Sakacin 5TaKTNWGSVVGSCVAGGLVGALGGTPIWIGAGPLVGAGQDAISQ4083
Sakacin 5XaKYYGNGLSCNKSGCSVDWSKAISIIGNNAVANLTTGGAAGWKS4365
Sakacin Pb,cKYYGNGVHCGKHSCTVDWGTAIGNIGNNAAANWATGGNAGWNK4437
Sakacin GdKYYGNGVSCNSHGCSVNWGQAWTCGVNHLANGGHGVC3834
Sakacin QeMQNTKELSVVELQQLLGGKR ASFGKCVVGAWGAGAAGLGAGVSG4486
Bavaricin MNfTKYYGNGVYXNSKKXWVDWGQAAGGIGQTVVXGWLGGAIPGK4769
Bavaricin AgKYYGNGVHXGKHSXTVDWGTAIGNIGNNAAANXATGXNAGG4002
Sakacin KhARSYGNGVYCNNKKCWVNRGEATQSIIGGMISGWASGLAGM4308
Curvacin AiARSYGNGVYCNNKKCWVNRGEATQSIIGGMISGWASGLAGM4308
Sakacin AjARSYGNGVYCNNKKCWVNRGEATQSIIGGMISGWASGLAGM4308
MBSa1SIIGGMISGWASGLAG4303
Figure 3.

Chromatogram of the purified bacteriocin produced by Lactobacillus sakei MBSa1 (C18 reversed-phase HPLC).

The amplification of DNA of Lact. sakei MBSa1 with specific primers targeting six different sakacin genes (sakacin Tα, Tβ, Q, X, P and G) generated negative results, but when PCR was performed with primers for sakacin A (SakA-F/SakA-R) and curvacin A (CurA-F/CurA-R), homologous fragments for the two bacteriocin genes were obtained (GenBank accession numbers AB292465.1 and MSUXNA4Z015, respectively).

Treatment with DTT resulted in mild change in antimicrobial activity, indicating that disulphide bonds are not essential for the antimicrobial activity of bacteriocin MBSa1 (Fig. 4).

Figure 4.

Activity of the purified bacteriocin MBSa1 against Listeria monocytogenes Scott A, after treatment with Tris-HCl (50 mmol l−1) at pH 8·0, proteinase K (1 mg ml−1), trypsin (1 mg ml−1) and dithiothreitol (100 mmol l−1).

Addition of the purified bacteriocin at the determined MIC/MKC values (3497 AU mg−1 for both MIC and MKC) at times 0 h and 8 h inhibited completely the growth of L. monocytogenes, indicating a bacteriostatic effect (Fig. 5). However, when the bacteriocin was added after 6 h (early exponential phase), the inhibitory effect was observed only until 20 h of incubation.

Figure 5.

Growth of Listeria monocytogenes Scott A in BHI broth at 37°C after addition of the purified bacteriocin MBSa1, added at time 0 h (■), 6 h (▲) and 8 h (●). Control curve, without addition of bacteriocin (♦).

Discussion

Bacteria belonging to Lactobacillus species are common in fermented and nonfermented foods such as dairy (Morales et al. 2011) meat products (Castro et al. 2011) and vegetables (Chen et al. 2010). Lactobacillus sakei is specially adapted to the meat environment and has been widely used as a starter culture for the manufacture of a variety of meat products (Carr et al. 2002). Chaillou et al. (2005) determined the complete genome sequence of the French sausage isolate Lact. sakei 23K, showing that this strain has a specialized metabolic repertoire that may contribute to its competitive ability in these foods.

Due to production of antimicrobial compounds, such as lactic and acetic acids, diacetyl, hydrogen peroxide and bacteriocins, some Lact. sakei strains possess interesting biotechnological potential application for food biopreservation (Carr et al. 2002). Several bacteriocins produced by Lact. sakei have been identified, such as sakacin A (Schillinger and Lucke 1989; Holck et al. 1992), sakacin M (Sobrino et al. 1992), bavaricin A (Larsen et al. 1993; Messens and de Vuyst 2002), sakacin P (Holck et al. 1994; Tichaczek et al. 1994; Vaughan et al. 2001; Urso et al. 2006; Carvalho et al. 2010), sakacin K (Hugas et al. 1995), bavaricin MN (Kaiser and Montville 1996), sakacins 5T and 5X (Vaughan et al. 2001), sakacin G (Simon et al. 2002), sakacin Q (Mathiesen et al. 2005), sakacin C2 (Gao et al. 2010) and sakacin LSJ618 (Jiang et al. 2012). It was observed that bacteriocin MBSa1 shares several properties with several bacteriocins produced by other Lact. sakei. Bacteriocin MBSa1 presents the same heat stability as sakacins M, C2, P and LSJ618 (Sobrino et al. 1992; Carvalho et al. 2010; Gao et al. 2010; Jiang et al. 2012). The resistance to pH in the range 2·0–6·0 is similar to that of sakacin LSJ618 (Jiang et al. 2012). However, sakacin C2 (Gao et al. 2010) and sakacin P (Carvalho et al. 2010) are stable at high pH (pH > 8·0), which was not observed for bacteriocin MBSa1.

The maximum production of bacteriocin MBSa1 in lactobacilli MRS broth occurred in the late logarithmic phase of growth (20 h at 25°C). Bacteriocin activity was first detected after 4 h of incubation at 25°C (late lag phase), which is similar to that found for sakacin A produced by Lact. sakei Lb796 (Schillinger and Lucke 1989) and sakacin P produced by Lact. sakei (Urso et al. 2006). However, maximum production of sakacin P by another Lact. sakei strain (Lact. sakei CCUG 42687) was reported at 20°C (Aasen et al. 2000).

As most bacteriocins produced by Lact. sakei, bacteriocin MBSa1 was inactive against Gram-negative bacteria. Until now, only two sakacins (C2 and LSJ618) are known for this activity: sakacin C2 inhibits E. coli ATCC 25922, Salmonella typhimurium CMCC 47729 and Shigella flexneri CMCC 51606 (Gao et al. 2010); and sakacin LSJ618 inhibits E. coli ECX4 and Proteus sp. (Jiang et al. 2012). However, the capability of bacteriocin MBSa1 to inhibit all tested foodborne strains of L. monocytogenes, besides L. monocytogenes Scott A, is remarkable. L. monocytogenes is a foodborne pathogen able to survive during manufacture of dry sausages, and its control is of great importance for the food industry. Bacteriocin MBSa1 did not inhibit the tested probiotic strains (e.g. Lactobacillus acidophilus La5) nor starter cultures (e.g. Lact. acidophilus La-14), suggesting an interesting potential for technological application in fermented foods for control of Listeria, without affecting starter or probiotic cultures.

Most bacteriocins produced by LAB contain positively charged amino acid residues and present hydrophobic characteristics (Nishie et al. 2012), and most bacteriocin purification strategies are based on ion-exchange and hydrophobic-interaction chromatographies. These techniques were successfully applied for purification of the bacteriocin produced by Lact. sakei MBSa1 strain and for previously reported bacteriocins, including sakacin A (Holck et al. 1992), bavaricin A (Larsen et al. 1993) and sakacins P, 5X and 5T (Vaughan et al. 2001).

Lactobacillus sakei MBSa1 presented the same C-terminal amino acid partial sequence and molecular mass as sakacin A and presented both sakacin A and curvacin A genes. The strain was negative for all other tested sakacin genes (Tα, Tβ, Q, X, P and G). The simultaneous positivity for both sakacin A and curvacin A genes is not surprising, because many similarities between different bacteriocins have been already reported (Schved et al. 1994; Bhugaloo-Vial et al. 1996). Noteworthy is that sakacin A produced by Lact. sakei Lb706 (Axelsson and Holck 1995) and curvacin A produced by Lactobacillus curvatus LTH1174 (Tichaczek et al. 1992) contain identical gene sequences for bacteriocin production and regulation (Eijsink et al. 1998; Aymerich et al. 2000).

Class II bacteriocins are known for having at least one disulphide bridge in the molecule. These bridges influence the antimicrobial activity (Ennahar et al. 2000), and bacteriocins with more than one disulphide bridge have higher activity than those with only one (Rihakova et al. 2009). Holck et al. (1992) and Tichaczek et al. (1992) have shown that sakacins A and P contain one single disulphide bond, and when treated with dithiothreitol (DTT), only part of the activity is lost, indicating that this bond is important but not essential for antimicrobial activity. Similarly, the antimicrobial activity of bacteriocin MBSa1 was only moderately reduced when treated with DTT.

When bacteriocin MBSa1 was added to a culture of L. monocytogenes Scott A to achieve the concentration corresponding to the MIC/MKC values, the growth of the pathogen was inhibited regardless the growth phase (lag phase or exponential phase), indicating a bacteriostatic activity. Sakacins produced by other Lact. sakei presented similar activities against Listeria spp. (Sobrino et al. 1992; Trinetta et al. 2008).

The control of L. monocytogenes in meat products is essential, as this pathogen causes outbreaks with high fatality rates (20–30%), especially among high-risk groups, such as pregnant women, neonates, elderly and immuno-compromised persons (Zunabovic et al. 2011). L. monocytogenes is a ubiquitous pathogen and may persist in the food industry environment due to its capability to produce resistant biofilms on equipment surfaces and premises (Carpentier and Cerf 2011). The entrance or recontamination of L. monocytogenes in the processing plants can have multiple sources, mainly raw ingredients, and Good Hygiene Practices and HACCP systems may be inefficient to avoid persistence in the processing environment and presence of Listeria in the final product (Tompkin 2002). Therefore, application of antimicrobial compounds may be necessary to inhibit the growth of pathogen. In this context, bacteriocins and bacteriocinogenic LAB can be explored as technological alternatives or ingredients for increasing the safety of the products manufactured in such conditions.

To conclude, Lact. sakei MBSa1 produces a bacteriocin that presents similarity with previously described sakacins. Its heat-resistance, pH stability in acid conditions, anti-Listeria activity and bacteriostatic action when applied in the concentration correspondent to the MIC value bring up a potential technological application for control of L. monocytogenes in foods. Further work is necessary to definitively prove that MBSa1 is identical or not to sakacin A and to evaluate its in situ activity against different L. monocytogenes strains in meat products and other types of foods.

Acknowledgements

The authors thank Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) (Project 08/58841-2), Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES-COFECUB Process: 3592-11-1) and Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) for financial supports. Prof. Iskra Ivanova thanks Région Pays de la Loire, France, for financial support as a Foreign Senior Scientist (contract 2011-12689).

Conflict of Interest

The authors declare no conflict of interest.

Ancillary