In recent years, natural antimicrobials, such as bacteriocins, have received a good deal of attention  over a number of microorganism control issues (Maldonado-Barragán  et  al.,  2016;  Gao  et  al.,  2016),   ribosomally synthesized antimicrobial peptides and proteins that usually only show inhibitory activity to closely related bacterial species. A number of bacteriocins produced by LAB  have  potential  as  novel   antimicrobial   agents   invarious practical applications from food to medicine (Nes et al., 2014; Gao et al, 2016). The interest of the food industry in bacteriocin compounds has increased in recent years, as they can be used as natural preservatives, replacing some chemicals, and increasing food shelf life (Yang et al., 2008). However, identifying new substances that have a broad spectrum of action, stable in the food, innocuous, while retaining the food’s color properties, texture, taste and aroma, is difficult (de la Fuente-Salcido et al., 2013).

Several classification systems used to group bacteriocins has been developed, and most of this information is based on findings from LAB. Two major classes, the Class I constitutes the lantibiotics and Class II constitutes the unmodified non-lantibiotics, are well defined (Nes et al., 2014). Class II bacteriocins are commonly found and characterized in enterococci. According to Nes et al. (2014), Class II can be further divided into four subgroups: Class IIa (the pediocin-like and strong antilisterial enterocins); Class IIb (the two-peptide bacteriocins); the circular bacteriocins; and the leaderless bacteriocins (synthesized without a leader peptide) Bacteriocinogenic enterococci strains have been found in foods, such as fresh milk, and fecal samples from humans and animals. The bacteriocino-genic strains have been characterized, especially in Enterococcus faecalis and E. faecium, however these peptides have also been isolated from E. mundtii, E. avium, E. hirae and E. durans. The enterococcal bacteriocins proved to be active against gram-positive foodborne pathogens, such as Listeria monocytogenes, Staphylococcus aureus and Clostridium spp. (Nes et al., 2014).

Enterococcus mundtii has been found to produce class IIa bacteriocins. Mundticin ATO6 was purified from E. mundtii isolated from chicory endive (Bennik et al., 1998), mundticin KS was identified in E. mundtii NFRI 7393 isolated from grass silage in Thailand (Kawamoto et al., 2002) and E. mundtii CRL35 from regional Argentinian cheese produced enterocin CRL35 (Saavedra et al., 2004). On the other hand, the structures of bacteriocins produced by E. mundtii have not yet been determined (De Vuyst et al., 2003).

The intestinal microbiota is considered a rich source of bacteriocin-producing strains. Though there has been considerable research on LAB bacteriocins to date, there are few studies on the screening for enterococci bacteriocin-producing strains isolated from fecal samples of wild animals (Poeta et al., 2007; Poeta et al., 2008; Brandão et al., 2010; Almeida et al., 2011). The aims of this study were to identify bacteriocinogenic enterococci strains isolated from fecal samples of wild South American (Arctocephalus australis) and Subant-arctic fur seals (Arctocephalus tropicalis), to determine the physicochemical characteristics and the spectrum of antimicrobial compounds and to evaluate the presence of enterocin genes by PCR.

Enterococci strains

Thirteen enterococci strains (three E. faecalis, one E. faecium, four E. hirae, two E. casseliflavus, two E. gallinarum and one E. mundtii) identified by biochemical and molecular methods in a previous study from fecal samples of wild young South American and Subantarctic fur seals were selected (Table 1) (Santestevan et al., 2015).  The colonies were preserved in a 10% (w/v) skimmed milk Molico® (Nestlé) solution supplemented with 10% (v/v) glycerol (LabSynth®) and frozen at -20°C. Prior to each experiment, an aliquot of frozen bacterial cells was recovered on Brain Heart Infusion Agar (BHIA, Oxoid) and incubated at 37°C for 24 h.

Screening for antilisterial activity by the double-agar layer test

Antibacterial activity of the strains was detected by the double-agar layer test (Lewus and Montville, 1991). To determine the inhibitory spectrum of the isolates against L. monocytogenes ATCC 35152, all enterococci strains were inoculated with a sterile toothpick on Trypticase Soy Agar (TSA, Himedia, Mumbai, India) and incubated for 18 h at 37°C. Simultaneously, an indicator organism was grown at 37°C for 18 h and 100 µL of this culture with 10⁶ CFU mL^-1 was inoculated in 10 mL Trypticase Soy Broth (TSB, Himedia, Mumbai, India) containing 0.7% agar. This culture was equilibrated at 45°C, mixed and then poured as an overlay onto the plate with growth of enterococci. The plates were incubated for 18 h at 37°C. Antilisterial activity was visually detected by observing clear inhibition zones around the producer strain and those strains in which antimicrobial activity was observed.

Antimicrobial activity of crude extraction from Enterococcus spp.

The crude extract bioassay of five strains, which showed antimicrobial activity in the double-agar layer test against L. monocytogenes, was performed according to Motta and Brandelli (2002). The strains were grown in 100 mL TSB at 37°C for 18 h, and the cells were harvested by centrifugation at 3,680 x g for 15 min at 4°C. The pH of the cell-free supernatants was adjusted to 6.5 with sterile 1 M NaOH to exclude acid production as the inhibitory mechanism, heated at 90°C for 10 min to inactivate remaining cells, and then filtered through a 0.22 µm pore-size nylon syringe filter (Chromafil). The antimicrobial substance was kept at 4°C for further characterization.

In the agar spot test, 20 µL of crude extract were dripped onto TSA plates previously inoculated with 100 µL of L. monocytogenes ATCC 35152 at 10⁶ CFU mL^-1. The plates were incubated at 37°C for 18 h, and examined for clearance zones of the bacterial lawn in the agar overlay. The assays were performed in triplicate.

Determination of growth curve of selected strain and analysis of antimicrobial activity

The selected producer strain was submitted to growth curve to evaluate the maximum production of the antimicrobial substance. The strain was grown in TSB, and 1 mL of the adjusted bacterial inoculum (approximately 10⁷ CFU mL^-1) was added into 100 mL of TSB and incubated at 35°C under shaking (180 rev min^-1). Aliquot samples were taken at specific time intervals(1, 4, 8, 16, 32, 40 and 48 h) to determine the number of viable cells (CFU mL^-1),  the

Santestvan et al., 2015. *Antibiotics: ciprofloxacin (CIP), erythromycin (ERI), nitrofurantoin (NIT), norfloxacin (NOR) and tetracycline (TET). ** Sensitive (S): resistance (R): positive (+); and negative (-). 2. Inhibition zone (mm). the pH and antimicrobial activity in millimeters (mm) these parameters were determined as described elsewhere (Motta and Brandelli, 2002). The cell-free supernatant pH was adjusted (pH 6.5) to verify the antimicrobial activity of L. monocytogenes ATCC 35152. Growth curve was performed in triplicate.

Arbitrary units assay

Antimicrobial activity was quantified by arbitrary units per mL at the time interval in which the highest antimicrobial activity (greatest halo) was observed in mm. The activity unit of the antimicrobial substance was determined in appropriate twofold serial dilutions on microplates. The dilutions were tested against L. monocytogenes ATCC 35152 with 10⁶ CFU mL^-1, using the agar spot test. The arbitrary units per milliliter (AU mL^-1) were defined as a reciprocal of the highest dilution factors generating an area of inhibition indicating the potency of the antimicrobial activity (Bigwood et al., 2012).

Effects of proteolytic enzymes, pH, temperature and chemicals on antimicrobial compounds stability

To gain insight into the nature of the antimicrobial compounds (ACs) of the selected strain, the following treatments were used: (a) the sensitivity to proteolytic enzymes was evaluated by addition of enzymes trypsin (Sigma), papain (Merck) and proteinase K (Merck) at a final concentration of 2 mg ml^-1} to the antimicrobial substance. Incubations were performed at 35°C for trypsin and papain and 37°C for proteinase K for 1 h. To inactivate enzymes, the treatments were subjected to heating at 100°C for 3 min. The AC was used as untreated control together with the enzyme controls; (b) The effect of pH on the antimicrobial activity of the ACs was investigated, and aliquots of ACs were subjected to different conditions of pH 1.0 to 12.0 (pH increases by 1.0 unit), adjusted with 1 M HCl or 1 N NaOH and incubated at 37°C for 2 h; (c) The effect of different temperatures was determined by incubating the AC at 30 to 100°C (in 10°C increments) for  30 min,  and  121°C  for 15 min; (d) The effects of chemicals on the ACs were also evaluated by treatments with 50% butanol (Vetec), 95% ethanol (Vetec), acetone (Dynamic), chloroform (Vetec), methanol (Vetec) and DMSO (Nuclear) at a proportion of 1:1, and 10% Tween 80 (Vetec) at a proportion of 9:1 with incubations at 37°C for 1 h. After the different treatments the antimicrobial activities were evaluated by the agar spot method. The indicator microorganism used was L. monocytogenes culture ATCC 35152.

Determination of antimicrobial spectrum of antimicrobial compounds

The following cultures were selected to determine the antimicrobial spectrum of antimicrobial compounds: Bacillus cereus (ATCC 14579), Staphylococcus aureus (ATCC 190506), Enterococcus faecalis (ATCC 29212), Streptococcus agalactiae (ATCC 13813), Klebsiella pneumoniae (ATCC 700603), Escherichia coli (ATCC 25922) and Pseudomonas aeruginosa (ATCC 27853). Four different serotypes of L. monocytogenes isolated from cheese (Mello 2007) were also evaluated by agar spot test with 20 µL of the ACs. The antimicrobial activity was measured in millimeter and the cultures were standardized with 10⁶ CFU mL^-1.

Detection of enterocin genes by PCR

Genomic DNA was extracted by the physicochemical method of Donato (2007). To identify the genes encoding the enterocins, PCR amplification was performed using specific primers to well-known structural enterocin genes: enterocin A (entA), enterocin B (entB), enterocin P (entP) and mundticin KS (munKS) (Park et al., 2003; Foulquié et al., 2003; Gutierrez et al., 2005; Zendo et al., 2005). Amplification was performed in a 25 µL mixture, containing 100 ng of DNA, 1.5 mM of MgCl₂, 200 µM of each dNTP, 1X reaction buffer, 1 U of Taq polymerase and 0.5 µM of each primer. Amplification was carried out in an Eppendorf Mastercycler Personal 5332 Thermocycler (Eppendorf^®) according to the following program: 1 cycle at 94°C for 5 min; 35 cycles at 94°C for 1 min, at 52°C for entA, or 47°C for entB, or 55°C for ent P and mun KS for 1 min, at 72°C for 2 min; 1 cycle at 72°C for 15 min. Finally, PCR products were resolved using electrophoresis on 1.5% (w/v) agarose gel, visualized on a UV transilluminator and photographed.

Detection of plasmid in selected producer strain

Plasmid DNA extraction was performed according to Sambrook and Russel (2001). The presence of plasmid DNA was observed using electrophoresis on 0.7% agarose gel. The Escherichia coli strain TOP10 DH10 BETA (Invitrogen^®), containing a plasmid, was used as positive control. This assay was performed in duplicate.

Statistical analysis

Statistical analysis was performed with the data obtained in millimeters in the antimicrobial activity tests at different pH and temperatures. ANOVA followed by Tukey with a 95% confidence level was performed on the data and considered significant where the P-value is equals to or less than 0.05. Testing was completed with Statistica software version 12.

Antimicrobial activity-producing strains

Among the 13 enterococci strains selected, five (36.46%) showed activity against L. monocytogenes ATCC 35152 in the double-agar layer test (Table 1). Of these five strains selected, only the crude extract bioassay of the E. mundtii J5 strain showed an antilisterial activity and was therefore chosen for subsequent tests.

Antimicrobial compounds produced by E. mundtii J5 and quantification of antimicrobial activity

The growth curves of the E. mundtii J5 strain related to antimicrobial compounds (ACs-J5) activity and pH are presented in Figure 1. The strain reached the mid exponential phase after 6 h of incubation and the highest (16 to 18 mm) inhibition zones were observed in the stationary phase (12 to 40 h). The initial pH of the culture was 6.5; after 12 h it passed to 5.0, reaching 4.0 at 48 h growth.  The antimicrobial activity of ACs-J5 was 400 AU mL^-1 against L. monocytogenes ATCC 35152.

Stability of ACs-J5

The effect of different treatments on the antimicrobial activity of ACs-J5 is presented in Table 2. The activity of ACs-J5 was lost after proteolytic enzyme treatment, suggesting its proteinaceous nature. However, ACs-J5 tolerance to pH treatments (1 to 12).  The greater antimicrobial activity of AC-J5 was observed at pH 4.0 to 7.0, with the highest antilisterial activity at pH 6.0 (p < 0.05).  At  Ph  1.0  to  3.0,  the AC-J5 activity remained, in media, at 63% (p < 0.05); and at pH 8.0 to 12.0 the activity was around 66%, compared to the controls (p < 0.05). As shown in Table 2, ACs-J5 was resistant towards the heat treatments. The only significant reduction in activity was seen at 121°C for 15 min where the activity decreased to 75% (p < 0.05). There was a significant difference at 121°C (p = 0.03). Methanol, DMSO, acetone and ethanol did not inactivate the antilisterial activity of ACs-J5, while butanol and chloroform did. On the other hand, the antilisterial activity was increased by 75% after the treatment of ACs-J5 with Tween 80 (Table 2).

Determination of antimicrobial activity spectrum of ACs-J5

All L. monocytogenes strains tested showed sensitivity to AC-J5, but B. cereus, S. aureus, E. faecalis, S. agalactiae, K. pneumoniae, E. coli and P. aeruginosa strains were not affected by the ACs-J5 (Table 3). Interestingly, L. monocytogenes serotype 1c presents the largest halo (45 mm) and serotype 4b the smallest (15.8 mm).

Identification of the enterocin genes and plasmid in E. mundtti J5 strain

The mundticin KS gene was detected in the E. mundtii J5 strain. The other enterocin genes tested were negative. No plasmids were observed in the E. mundtii J5 strain.

L. monocytogenes has been recognized as one of the most relevant foodborne bacteria pathogens (Scallan et al., 2011). During this study, five (38.5%) enterococci strains isolated from fecal samples of wild fur seals showed antilisterial activity. Bacteriocinogenic enterococ-cal strains isolated from fecal samples of wild animals were previously identified by Almeida et al. (2011), Brandão et al. (2010), and Poeta et al. (2008, 2007). Poeta et al. (2008) found bacteriocinogenic isolates in 17.85% of the enterococci isolated from fecal samples of wild animals. Brandão et al. (2010) also found a similar number (45.6%) of bacteriocinogenic enterococcal isola-tes of fecal origin from humans, pets, wild animals and birds against L. monocytogenes CECT4032. Only E. mundtii J5 produced antilisterial activity after the pH was adjusted. Other studies evaluating the bacteriocin-like substances from enterococci showed similar data to this (Birri et al., 2010; Adeniyi et al., 2015). Birri et al. (2010) worked with the supernatant of 104 LAB study from the gastrointestinal tracts of children, and only one isolate of E.  avium  (0.96%)   was   a   producer   of  an antagonistsubstance. Many substances produced by microorga-nisms have been identified as inhibitors of pathogenic bacteria. However, the inhibition caused by enterococci may not  be related  to  bacteriocins,  but  other  inhibitory substances, such as organic acids resulting from their metabolism (Gaamouche et al., 2014).

In the present study, the greatest antilisterial activity was observed  in  the  stationary  phase.  This is in partial agreement with the results obtained by Leroy and De Vuyst (2002), in which production of enterocin RZC5 (from E. faecium) occurred in the early growth phase. The production of enterocin HJ35 by E. faecium HJ35 also occurred in the mid-log growth phase, reaching maximum production during the late stationary phase (Yoon et al., 2005). Moshood and Tengku Haziyamin (2012) showed that maximum bacteriocin production of E. faecium B3L3 occurs at the end of the exponential phase.

Bacteriocins produced by LAB generally have a low antimicrobial activity in arbitrary units per milliliter (AU mL-1). However, the low values of ACs-J5 (400 AU mL-1) found in this study could be related to the technique used to determine the activity. Some studies carried out to show the enterocin production occurring in fermenters controlled conditions such as pH, for example, which is an important factor in the production of peptides. This can interfere with the antimicrobial activity obtained in arbitrary units per milliliter. Under these conditions, generally, the antimicrobial activity values obtained are much higher than those found in this work. Mundticin KS showed an antimicrobial activity of 6,400 to 12,800 AU/mL-1 when produced in a fermenter (Kawamoto et al., 2002).

 Xiaoyuan et al. (2014) showed the highest antilisteria activity reaching 51,200 AU/mL-1 from enterocin A after 24 h of induction in a 5 L fermenter. The ACs-J5 were stable within a wide range of pH and heat treatment and sensitive to all proteolytic enzymes tested, suggesting their proteinaceous nature. These features were typical of class IIa bacteriocins (Nes et al., 2014). The thermal and pH stability of ACs-J5 may constitute an advantage for potential use as a biopreservative in combination with thermal processing used to preserve food products.The stability of enterocins against treatment with chemicals has great importance from the technological point of view, since many organic and inorganic compounds are used for food processing, by incorporating the ingredients. In this study, the ACs-J5 were sensitive to chloroform and to a lesser extent butanol. Motta et al (2007) also observed that the activity of bacteriocin produced by Bacillus spp. was affected by butanol and to a lesser extent by acetone and methanol. Using the treatment with Tween 80, an increase in antimicrobial activity was observed. Similar results were observed for the mundticin CRL1656 (Espeche et al., 2014). The antimicrobial compounds produced by E. mundtii J5 did not inhibit the growth of other gram-positive and gram-negative strains. Paschoalin et al. (2011) reported antimicrobial activity of Enterococcus spp. for the production of enterocins capable of inhibiting the development of L. monocytogenes, besides inhibiting gram-negative bacteria, such as E. coli and S. paratyphi.

Only the mundticin KS gene was detected in the J5 strain. Mundticin KS is a positively charged, hydrophobic, 43-amino-acid peptide that contains the highly conserved YGNGV motif found in the N-terminal part of many class IIa bacteriocins in the classification described by Klaenhammer (1993).

In the present study, we did not detect plasmid in E. mundtii J5 strains. This is in agreement with Criado et  al.(2008), who observed the presence of enterocin P located in a chromosomal fragment. In a similar development, Moshood et al. (2015) were able to detect the enterocin gene in the chromosomal DNA of Enterococcus mundtii C4L10.

In this study a bacteriocinogenic enterococci strain (E. mundtii J5 isolated from the fecal samples of wild South American fur seal) was identified. The Antimicrobial compounds produced by E. mundtii J5 produced should belong to the class IIa, since were heat-stable and strongly inhibits the growth of L. monocytogenes strains. These properties make ACs-J5 a valuable biopreservative in food industries for avoiding pathogenic microorganisms and should be a good candidate for probiotic application. However, further studies on the purification and characterization of the intended antibiotic would be necessary.

The authors have not declared any conflict of interest.

The authors thank the government agencies Conselho Nacional de Desenvolvimento Científico e Tecnológico do Brasil (CNPq - #444335/2014-5 and #303251/2014-0) and the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) of the Brazilian government. They are grateful to biologist Mauricio Tavares from Grupo de Estudos de Mamíferos Aquáticos do Rio Grande do Sul and all staff from Ceclimar- UFRGS.