Screening and characterization of LAB-produced bacteriocin-like substances from the intestine of grey mullet (Mugil cephalus L.) as potential biocontrol agents in aquaculture

Authors


  • The first two authors contributed equally to this work.

Correspondence

Yi-Sheng Chen, Department of Biotechnology, Ming Chuan University, No. 5 De-Ming Road, Gui-Shan, Taoyuan 333, Taiwan. E-mail: yisheng@mail.mcu.edu.tw

Abstract

Aim

To isolate and characterize bacteriocin-like inhibitory substance (BLIS)-producing lactic acid bacteria from the intestine of grey mullet.

Methods and Results

Inhibitory activity against at least one or more indicator strains was observed in one Enterococcus thailandicus, one Enterococcus faecium and two Lactococcus garvieae strains. Enterococcus faecium B3-8 and Ent. thailandicus B3-22 showed the greatest inhibitory activities against Listeria monocytogenes ATCC 19111 and were therefore further characterized. The results suggested that the inhibitory substances from the two strains showed similar characteristics with respect to sensitivity to heat and proteolytic enzymes. BLIS from Ent. thailandicus B3-22 was characterized by a broader inhibitory spectrum than that from Ent. faecium B3-8. SDS-PAGE revealed that the molecular size of partially purified BLISs from Ent. faecium B3-8 and Ent. thailandicus B3-22 was c. 5 and 3 kDa, respectively. The molecular mass of purified bacteriocin from Ent. thailandicus B3-22 was further determined to be 6319 Da by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS). The results indicated that BLIS from Ent. thailandicus B3-22 can effectively inhibit the growth of all tested L. garvieae strains.

Conclusions

The findings obtained in this study suggest the potential use of Ent. thailandicus B3-22 as a biocontrol agent against pathogenic L. garvieae in the aquaculture.

Significance and Impact of the Study

This is the first report describing the characteristics of BLIS from Ent. thailandicus that showed potential for use as a biocontrol agent in the aquaculture.

Introduction

Grey mullet (Mugil cephalus L.) is an important economic fish species in Taiwan. Mullet roe is the most famous food product made from grey mullet. It is a high-priced product that is known as black gold in Taiwan. Aquaculture farms of grey mullet are increasing as a result of the progress of aquaculture technologies and the decrease in wild grey mullet.

A number of phytopathogenic and spoilage bacteria cause economic losses of aquatic products. Lactococcus garvieae is considered to be the aetiological agent of lactococcosis in various fish species worldwide (Eldar et al. 1999; Chen et al. 2002; Vendrell et al. 2006; Jung et al. 2010). In a previous study, Chen et al. (2002) indicated that L. garvieae can cause outbreaks of disease in grey mullet in Taiwan and is a major pathogen in mullet aquaculture in Taiwan. Efforts to prevent and control invasion by disease-causing agents have concentrated on good husbandry techniques and/or the use of vaccines and antibiotics (Vendrell et al. 2006).

An alternative approach to disease prevention in aquaculture is the use of bacteriocin-producing lactic acid bacteria (LAB). LAB have been commonly used as probiotics in aquaculture. Administration of probiotics was reported to competitively exclude pathogenic bacteria through the production of inhibitory compounds, improved water quality or enhanced immune response of host species (Verschuere et al. 2000; Farzanfar 2006; Gatesoupe 2008; Pirarat et al. 2011).

In previous studies, various LAB-produced bacteriocins have been applied as biopreservative agents in seafood products and have been shown to be effective in the control of pathogenic and spoilage micro-organisms (Calo-Mata et al. 2007; Yin et al. 2007; Diop et al. 2009; Chahad et al. 2012). A number of studies suggest the potential use of bacteriocin-producing LAB as biocontrol agents against pathogens in aquaculture (Gatesoupe 2008; Gillor et al. 2008; Desriac et al. 2010; Satish Kumar et al. 2011; Heo et al. 2012). However, a similar application in the aquaculture of grey mullet has not previously been studied.

The objective of this study was to screen high-activity anti-L. garvieae LAB strains isolated from grey mullet and characterize the inhibitory substances produced by these strains.

Materials and methods

Grey mullet samples

Six grey mullet samples (c. 40–45 cm), including one diseased sample, were collected from two different fishing ports located in Keelung City, Taiwan. The disease and causative agent of the diseased grey mullet were not known; however, symptoms such as darkening, enlargement of the spleen and a strong stench were observed during the dissection. The intestines of all grey mullet samples were removed and then homogenized for the isolation of LAB.

Isolation of lactic acid bacteria

Isolation was mainly performed following the procedures described by Chen et al. (2012). Instead of distilled water, artificial seawater (Tetra Marine Salt Pro, Tetra, Germany) was used for preparing the isolation medium. Briefly, a 1-g portion of crushed intestine mixture from each grey mullet sample was respectively mixed with 4 ml of 7·5 g l−1 NaCl solution. Dilutions of the resultant solution (10- to 1000-fold) were spread directly onto the surface of MRS (De Man Rogosa and Sharpe) (De Man et al. 1960) agar plates containing 10 g l−1 CaCO3. The plates were incubated under anaerobic conditions (Mitsubishi AnaeroPak™ System, Pack-Anaero; Mitsubishi Gas Chemicals, Tokyo, Japan) at 30°C for 3–5 days. Colonies of acid-producing bacteria, identified by a clear zone around each colony, were randomly selected from MRS agar plates and purified by replating on MRS agar plates. Colonies were reselected and initially examined for Gram staining and the production of catalase. Only Gram-positive, catalase-negative strains were selected. The selected strains were stored at −80°C in 100 g l−1 skim milk.

Screening for BLIS-producing strains

An agar spot test (Schillinger and Lücke 1989) and agar well diffusion assay (Srionnual et al. 2007) were used to detect and determine the antibacterial activities of isolates. Lactobacillus sakei ssp. sakei JCM 1157T and Listeria monocytogenes ATCC 19111 were used as indicator strains. In addition, the supernatant fluid of bacteriocin-like inhibitory substance (BLIS)-producing strains was obtained by centrifugation at 4480 g at 4°C for 30 min. The fluid was filtered through 0·45-μm membrane filters (Toyo Roshi Kaisha Ltd., Tokyo, Japan). Antibacterial activity of cell-free supernatants was further confirmed by pH adjustment and proteinase K treatment (Srionnual et al. 2007).

Identification of BLIS-producing strains

Sequence analysis of 16S rDNA was used to identify the bacterial isolates. A colony PCR method described by Sheu et al. (2000) was performed in this study. PCR were carried out using a Genomics Taq gene amplification PCR kit (Genomics, Taipei, Taiwan) and performed on a Gene Amp PCR System 9700 (PerkinElmer Inc., Boston, MA, USA). PCR products were purified with a Clean/Gel Extraction Kit (BioKit, Miaoli, Taiwan) and then sequenced with the following primer: 5′-AGAGTTTGATCCTGGCTCAG-3′ (27F). DNA sequencing was performed using an ABI 3730 DNA Analyzer (Applied Biosystems, Foster City, CA, USA). Partial sequencing, c. 900 bp ahead, was performed and manually aligned using the software Genetyx-Win (ver. 5.1; Genetyx, Tokyo, Japan). Sequence homologies were examined by comparing the obtained sequences with those in the DNA Data Bank of Japan (DDBJ; http://www.ddbj.nig.ac.jp/) using Blast.

Optimum temperature for BLIS production

To study the optimum temperature for BLIS production, 100 μl of overnight bacterial culture was inoculated into 5 ml MRS broth and then incubated at 22, 25, 30, 33 or 37°C for 12 h. After incubation, BLIS activity was determined using the size of the inhibition zone around the well (8 mm diameter) in the agar well diffusion assay. Lactobacillus sakei ssp. sakei JCM 1157T was used as the indicator strain.

Production of BLIS by selected strains

To study BLIS production, 100 ml of overnight bacterial culture was inoculated into 900 ml of MRS medium. At specific time intervals, 1-ml samples were removed, and the optical density at 600 nm of the culture and the arbitrary activity units (AU) ml−1 (reciprocal of the highest dilution at which activity was still obtained) of the BLIS were determined according to the method of Srionnual et al. (2007). Lactobacillus sakei ssp. sakei JCM 1157T was used as the indicator strain. The incubation temperature was set based on the results obtained from the optimum growth temperature analysis.

Partially purified BLIS solution

Cell-free culture supernatant (1 l) was prepared and then purified by ammonium sulfate precipitation (40%) and Sep-Pak C18 cartridges (Waters, Milford, MA, USA). Eluted BLIS fractions from Sep-Pak C18 cartridges were freeze-dried for 48 h using an EYELA FDU-2200 freeze-dryer (Tokyo Rikakikai, Tokyo, Japan), and inhibition activity of the freeze-dried BLIS powders was determined using the agar spot test. Lactobacillus sakei ssp. sakei JCM 1157T was used as the indicator strain. The freeze-dried BLIS powder was weighed and then thousand-fold (w/v) diluted by resuspending in 20% acetonitrile solution. In this study, the arbitrary activity unit (AU ml−1) of tested partially purified BLIS solution (PPBS) was previously evaluated and both PPBSs from strains B3-8 and B3-22 were c. 1280 AU ml−1.

Molecular size approximation

The molecular size of BLIS was determined using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), following the method described by Srionnual et al. (2007). A 4–20% Mini-Protean TGX precast gel (Bio-Rad Laboratories, Hercules, CA, USA) was used and BLIS molecular size was estimated using Precision Plus Protein Dual Xtea Standards (Bio-Rad).

Effects of enzymes and heat on BLIS activity

To evaluate heat stability, the PPBS of selected strains was incubated at 80°C for 30 min, 90°C for 30 min or 121°C for 15 min. To analyse sensitivity to various enzymes, PPBSs were treated with proteinase K (Merck, Darmstadt, Germany), trypsin (Wako, Osaka, Japan) or catalase (Wako) (30 IU mg−1, 37°C for 5 h). Next, the inhibitory activity of individual 2 μl PPBS treatments was determined using the agar spot test. All experiments were carried out in triplicate, and the results are presented as mean ± SD (standard deviation). Data generated from the experiments were analysed for significance using the Student's t-test.

Inhibition spectrum of PPBSs

Partially purified BLIS solutions from selected strains, Enterococcus faecium B3-8 and Enterococcus thailandicus B3-22, were tested against the Gram-positive and Gram-negative bacteria listed in Table 1. The inhibitory activity assay was performed using the agar spot test method with previously prepared PPBSs from selected strains. Duplicate tests were performed for each PPBS sample, and mean values were calculated.

Table 1. Inhibition spectra of the BLISs produced by Enterococcus faecium B3-8 and Enterococcus thailandicus B3-22
Indicator strainsMediumIncubation temp. (°C)Diameter of the inhibition zone (mm)aOrigin
Ent. faecium B3-8Ent. thailandicus B3-22
  1. –, no inhibitory zone observed; BLIS, bacteriocin-like inhibitory substance; PABS, partially purified BLIS solution.

  2. a

    Determined using the agar spot test. PPBS treatments (2 μl) were spotted.

  3. b

    Type strain obtained from BCRC.

  4. c

    Strain obtained from this study.

Lactobacillus sakei ssp. sakei JCM 1157TMRS3055·5 b
Lact. delbrueckii ssp. bulgaricus ATCC 11842TMRS372·5 b
Lact. plantarum 1511MRS30Chang et al. (2011)
Lact. animalis C060203MRS375·5Chen and Yanagida (2006)
L. lactis ssp. lactis 1203MRS307Chang et al. (2011)
L. lactis ssp. lactis 1205MRS307Chang et al. (2011)
L. lactis ssp. lactis 1216MRS304·5Chang et al. (2011)
Lactococcus garvieae E-9MRS308 c
L. garvieae E-10MRS307·5 c
L. garvieae 4103MRS307·5Chang et al. (2011)
L. garvieae 1–4MRS308Chen et al. (2012)
L. garvieae cb3-4MRS308·5Chen et al. (2012)
Ent. faecium D081821MRS374·5Chen et al. (2007)
Ent. durans C102901MRS305Yanagida et al. 2006
Leuconostoc mesenteroides 1112MRS305Chang et al. (2011)
Leuc. citreum 1512MRS302Chang et al. (2011)
Weissella cibaria 110MRS305Srionnual et al. (2007)
W. cibaria 1521MRS305Chang et al. (2011)
W. hellenica ATCC 51523TMRS308 b
W. paramesenteroides ATCC 33313TMRS308 b
Streptococcus thermophilus ATCC 19258TMRS372·5 b
Listeria monocytogenes ATCC 19111BHI374·54 b
Staphylococcus aureus ssp. aureus ATCC 12600TLB372·5 b
Acinetobacter baumannii ATCC 19606TLB37 b
Bacillus subtilis ssp. subtilis ATCC 6051TLB374 b
Escherichia coli ATCC 11775TLB37 b
Vibrio vulnificus ATCC 27562TTSB37 b

Mass spectrometry

The PPBS solution was subsequently placed on a C18 reverse-phase column (BioBasic-18; Thermo Scientific, Rockford, IL, USA) integrated into a high-performance liquid chromatography (HPLC) system (L-2130, L-2450; Hitachi, Tokyo, Japan). The bacteriocin was eluted with a gradient of MilliQ water–acetonitrile containing 1 ml l−1 trifluoroacetic acid, at a flow rate of 1 ml min−1, as follows: 0–5 min, 30–50% (v/v); 5–45 min, 50–100% (v/v); and 45–50 min, 100% (v/v) acetonitrile (Fig. 3). The bacteriocin activity of the fractions obtained from the purification procedure was determined against Lact. sakei ssp. sakei JCM 1157T. The active eluted solution was then freeze-dried on an EYELA FDU-2200 freeze-dryer (Tokyo Rikakikai). The molecular mass of the purified bacteriocin was determined by matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry (MS) using a mass spectrometer (SCIEX QSTAR Elite; Applied Biosystems).

Analysis using enterococcal bacteriocin-specific primers

Several primer pairs described by Marciňáková et al. (2005) were used for PCR amplification of structural genes of enterocins A, P, L50B and B. The sequences were as follows: ent A forward primer 5′-GGTACCACTCATAGTGGAAA-3′, reverse primer 5′-CCCTGGAATTGCTCCACCTAA-3′; ent P forward primer 5′-GCTACGCGTTCATATGGTAAT-3′, reverse primer 5′-TCCTGCAATATTCTCTTTAGC-3′; ent L50B forward primer 5′-ATGGGAGCAATCGCAAAATTA-3′, reverse primer 5′-TAGCCATTTTTCAATTTGATC-3′; ent B forward primer 5′-CAAAATGTAAAAGAATTAAGATCG-3′, reverse primer 5′-AGAGTATACATTTGCTAACCC-3′. The PCR was performed by following the reaction conditions described by Marciňáková et al. (2005).

Results

A total of 194 acid-producing bacteria were isolated; 148 were from the normal grey mullet samples and 46 were from the diseased grey mullet sample. Inhibitory activities against at least one indicator strain were observed with the following four isolated cultures: B3-8, B3-22, E-9 and E-10. 16S rRNA gene sequences indicated that B3-8 and B3-22 belonged to the genus Enterococcus and were closely related to Ent. faecium, Ent. durans, Ent. hirae and Ent. thailandicus, with homology values from 98·5 to 99·9%. Therefore, a supplementary identification test was performed by analysing the rpoA gene sequences of these two strains (Tanasupawat et al. 2008). Based on the sequencing results, strain B3-8 was identified as Ent. faecium (99·4% homology to Ent. faecium and 96·8% to Ent. thailandicus) and B3-22 was identified as Ent. thailandicus (99·9% homology to Ent. thailandicus and 96·5% to Ent. faecium). Besides the Enterococcus strains, the remaining two selected strains, E-9 and E-10, were identified as L. garvieae based on their 16S rRNA gene sequences. Specifically, strains E-9 and E-10 were isolated from the diseased grey mullet sample, while strains B3-8 and B3-22 were isolated from the other collected samples.

As previously described, L. garvieae is considered to be the aetiological agent of lactococcosis in various fish species; therefore, further characterization of BLISs from L. garvieae E-9 and E-10 was terminated.

The results of the initial screening for optimal incubation temperature showed that maximum inhibitory activity from Ent. faecium B3-8 and Ent. thailandicus B3-22 against the indicator strain Lact. sakei ssp. sakei JCM 1157T was observed at 37 and 30°C, respectively. Based on these results, more detailed experiments to investigate BLIS production at optimal incubation temperatures were performed. The production of BLIS from Ent. faecium B3-8 reached 1280 AU ml−1 after 16 h of incubation at 37°C. During the 16 h of growth, the pH value decreased from 6·07 to 4·97 and the OD600 of the culture increased from 0·446 to c. 1·572 (Fig. 1a). On the other hand, the highest production of BLIS from Ent. thailandicus B3-22 reached 1280 AU ml−1 after 6 h of incubation at 30°C. The OD600 of the culture increased from 0·389 to c. 1·308, and the pH value decreased from 6·1 to 5·1 after 16 h of incubation (Fig. 1b).

Figure 1.

Production of bacteriocin-like inhibitory substances during the growth of Enterococcus faecium B3-8 (a) and Enterococcus thailandicus B3-22 (b). Symbols: black square, pH value; black circle, cell number at OD600; Grey bar, arbitrary activity (AU ml−1).

The molecular size of partially purified BLISs from strains B3-8 and B3-22 was analysed by SDS-PAGE, and the specific band associated with the antibacterial activity was c. 5 and 3 kDa, respectively (Fig. 2).

Figure 2.

SDS-PAGE analysis of partially purified bacteriocin-like inhibitory substances (BLIS) from Enterococcus faecium B3-8 and Enterococcus thailandicus B3-22. (a) Ent. faecium B3-8 and (b) Ent. thailandicus B3-22. L: Coomassie Brilliant Blue–stained gel, R: gel placed onto MRS agar surface overlaid with Lactobacillus sakei ssp. sakei JCM 1157T. Lane M, Precision Plus Protein Dual Xtra standards (Bio-Rad Laboratories); lanes 1 and 2, partially purified BLIS B3-8; lanes 3 and 4, partially purified BLIS B3-22.

The effects of various enzymes on the inhibitory agent were studied. The PPBSs from strains B3-8 and B3-22 were completely inactivated by the treatment with proteinase K and trypsin, whereas catalase had no effect (Table 2). On the other hand, the PPBS from strain B3-8 maintained antibacterial activity after heat treatment at 80°C for 30 min. An obvious decrease in antibacterial activity was observed after heat treatment at 90 or 121°C (Student's t-test; < 0·05, Table 2). The PPBS from strain B3-22 showed lower heat sensitivity, as indicated by the obvious decrease in antibacterial activity after heat treatment at 121°C for 15 min (Student's t-test; < 0·05, Table 2).

Table 2. Effects on BLISs produced by Enterococcus faecium B3-8 and Enterococcus thailandicus B3-22
 Diameter of inhibition zone (mm)c,d
Ent. faecium B3-8Ent. thailandicus B3-22
  1. –, no inhibitory zone observed; BLIS, bacteriocin-like inhibitory substance; PABS, partially purified BLIS solution.

  2. a

    Significant (< 0·05).

  3. b

    Highly significant (< 0·01).

  4. c

    Lactobacillus sakei ssp. sakei JCM 1157T was used as the indicator strain.

  5. d

    Mean and standard deviation for = 3.

  6. e

    Determined using the agar well diffusion assay. Wells (8 mm diameter) were filled with 100 μl of culture supernatant.

  7. f

    Determined using the agar spot test. PPBS treatments (2 μl) were spotted.

  8. g

    The experimental data were evaluated using Student's t-test.

Growth temperature (°C)e
229·3 ± 0·5
259·0 ± 0·011·0 ± 0·0
3010·7 ± 0·511·3 ± 0·5
3311·7 ± 0·59·3 ± 0·5
3712·0 ± 0·08·3 ± 0·5
Enzymesf,g
Control5·3 ± 0·56·0 ± 0·0
Proteinase Kbb
Trypsinbb
Catalase5·3 ± 0·56·0 ± 0·0
Heatf,g
Control5·3 ± 0·56·0 ± 0·0
30 min at 80°C5·0 ± 0·05·3 ± 0·5
30 min at 90°C3·3 ± 0·5a5·3 ± 0·5
15 min at 121°C1·7 ± 0·5a3·7 ± 0·5a

The PPBS from strain B3-8 showed inhibitory activity against only three indicator strains, indicating a narrow antibacterial spectrum. On the other hand, the PPBS from B3-22 showed inhibitory activity against at least 22 indicator strains, indicating a broad antibacterial spectrum. Besides L. garvieae E-9 and E-10, another three L. garvieae strains, which were previously isolated from wild clams (Chen et al. 2012), fermented clams (Chen et al. 2012) and fermented ginger (Chang et al. 2011), were also used as indicator strains. Obvious inhibitory activity against all tested L. garvieae strains was observed (Table 1).

Studies of bacteriocin from Ent. thailandicus remain scarce; therefore, PPBS from Ent. thailandicus B3-22 was further purified and its molecular mass was determined. The reverse-phase chromatography revealed one well-separated peak with bacteriocin activity (Fig. 3). The molecular mass of the purified bacteriocin was 6319 Da (Fig. 4).

Figure 3.

Elution pattern of bacteriocin from Enterococcus thailandicus B3-22 by reverse-phase high-performance liquid chromatography with a C18 reverse-phase column. Broken line, concentration of acetonitrile. Arrow, active peak.

Figure 4.

MALDI mass spectrum of bacteriocin from Enterococcus thailandicus B3-22 (mass = 6320 Da).

Several enterococcal bacteriocin-specific primers were used to identify the bacteriocin-producing strains B3-8 and B3-22. The presence of amplified PCR product for ent P structural gene was found in strain B3-8 (c. 90 bp, data not shown). However, no amplified PCR product was observed in strain B3-22.

Discussion

Several bacteriocins associated with Enterococcus isolates have been previously described, including enterocins (Casaus et al. 1997; Cintas et al. 1998; Ennahar and Deschamps 2000; Jennes et al. 2000; Chen et al. 2007; Satish Kumar et al. 2011; Sawa et al. 2012), durancin L28-1A (Yanagida et al. 2005) and mundticin (Kawamoto et al. 2002). Enterococcus thailandicus was initially found in fermented sausage (‘mum’) in Thailand (Tanasupawat et al. 2008), and related strains were previously found in human blood and clinical specimens (Shewmaker et al. 2011). However, studies of BLIS from Ent. thailandicus remain scarce. To our knowledge, this is the first study of a BLIS from Ent. thailandicus.

Optimizing growth temperature has an influence on the production of bacteriocins (Drosinos et al. 2005; Chen and Yanagida 2006; Van den Berghe et al. 2006). In this study, Ent. faecium B3-8 showed increased BLIS production when incubated at higher temperatures, such as 33 and 37°C. However, Ent. thailandicus B3-22 showed increased BLIS production when incubated at lower temperatures, such as 25 or 30°C (Table 2). This might be a favourable characteristic for the application of Ent. thailandicus B3-22 to aquaculture.

In both strains B3-8 and B3-22, complete inactivation was observed after the PPBSs were treated with proteinase K, which indicated the proteinaceous nature of the active agent. In addition, no effect was observed after the treatment with catalase, indicating that the active agent did not originate from H2O2. In determining the approximate molecular size of BLISs using SDS-PAGE, similar results were observed for strains B3-8 and B3-22. However, it was quite obvious that BLISs from B3-8 and B3-22 showed different characteristics regarding heat sensitivity and antibacterial spectra (Tables 1 and 2). BLISs from B3-8 and B3-22 were therefore considered as two different inhibitory substances in this study.

When detecting the bacteriocin gene by using several enterocin-specific primers, the structural gene for ent P was detected in strain B3-8, while none was detected in strain B3-22. BLISs from B3-8 and B3-22 were further confirmed as two different inhibitory substances. However, more characteristics such as molecular weight and amino acid sequences of the bacteriocin from strain B3-8 need to be confirmed before it can be proposed as enterocin P.

In contrast to BLIS from Ent. faecium B3-8, BLIS from Ent. thailandicus B3-22 showed a broad antibacterial spectrum against at least 22 indicator strains used in the current study. BLIS from Ent. thailandicus B3-22 showed great inhibitory activities against all tested L. garvieae strains, regardless of the isolation source (Table 1). It is therefore considered that Ent. thailandicus B3-22 can be applied as a potential biocontrol agent against pathogenic L. garvieae in aquaculture. Although L. garvieae E-9 and E-10 were isolated from the diseased grey mullet sample, the pathogenicity of these two strains remains unclear.

Besides L. garvieae, BLIS from Ent. thailandicus B3-22 also showed inhibition activities against L. monocytogenes and Staphylococcus aureus. The potential use of Ent. thailandicus B3-22 as a biopreservative is also suggested. In the current study, the initial selection was performed by determining the inhibitory activities of isolates against Lact. sakei ssp. sakei JCM 1157T and L. monocytogenes ATCC 19111.

With the exception of the four selected strains (B3-8, B3-22, E-9 and E-10), the possibility of the remaining 190 isolates exhibiting inhibitory activity against other pathogenic bacteria remains unclear. In addition, differences in the intestinal microflora between healthy and diseased grey mullet were not studied. Further studies on these points will be the important tasks in our future research.

The result of MALDI-TOF MS analysis revealed that the molecular mass of the purified bacteriocin from Ent. thailandicus B3-22 was 6319 Da. This result was compared with known bacteriocins from enterococci and other LAB (Jack et al. 1995; Franz et al. 2007; Nes et al. 2007; Zendo et al. 2008; Javed et al. 2011; Masuda et al. 2012). No identical result was found, indicating the possible discovery of a novel bacteriocin.

In conclusion, this is the first report of the bacteriocin from Ent. thailandicus. Because different species generally produce different bacteriocins, this implies that the bacteriocin from Ent. thailandicus B3-22 is a novel one. Our findings suggest that Ent. thailandicus B3-22 has obvious potential for practical application as a biopreservative in the aquaculture, due to its excellent inhibitory activity against L. garvieae.

Acknowledgements

The authors would like to thank Mr Chi-huan Chang for his kind cooperation in sampling and technical assistance.

Ancillary