Identification of lactobacilli with inhibitory effect on biofilm formation by pathogenic bacteria on stainless steel surfaces

International Journal of Food Microbiology 191 (2014) 116–124

Contents lists available at ScienceDirect

International Journal of Food Microbiology

j ourna l homepage: www.e lsev ie r .com/ locate / i j foodmicro

Identification of lactobacilli with inhibitory effect on biofilm formation bypathogenic bacteria on stainless steel surfaces

Fatma Ait Ouali a, Imad Al Kassaa b,c, Benoit Cudennec b, Marwan Abdallah b, Farida Bendali a, Djamila Sadoun a,Nour-Eddine Chihib b, Djamel Drider b

a Laboratoire de Microbiologie Appliquée, Faculté des Sciences de la Nature et de la Vie, Université de Bejaia, Bejaia, Algeriab Laboratoire Régional de Recherche en Agroalimentaire et Biotechnologies: Institut Charles Viollette, Bâtiment Polytech'Lille, Université Lille 1, Avenue Paul Langevin, Cité Scientifique,59655 Villeneuve d'Ascq Cedex, Francec Centre AZM de Biotechnologie, EDST-Université Libanaise Tripoli-Lebanon, Faculté de santé publique section 3, Université Libanaise, Tripoli, Lebanon

E-mail address: [email protected] (D. Drider

http://dx.doi.org/10.1016/j.ijfoodmicro.2014.09.0110168-1605/© 2014 Elsevier B.V. All rights reserved.

a b s t r a c t

Article history:Received 17 March 2014Received in revised form 8 September 2014Accepted 14 September 2014Available online 18 September 2014

Keywords:BiofilmAntagonismLactic acid bacteriaS. aureusStainless steelCytotoxicity

Two hundred and thirty individual clones of microorganisms were recovered from milk tanks and milkingmachine surfaces at two distinct farms (Bejaja City, Algeria). Of these clones, 130 were identified as lactic acidbacteria (LAB). In addition Escherichia coli, Salmonella, Staphylococcus aureus and Pseudomonas aeruginosa specieswere identified in the remaining 100 isolates—spoilage isolate. These isolates were assayed for ability to formbiofilms. S. aureus, Lactobacillus brevis strains LB1F2, LB14F1 and LB15F1, and Lactobacillus pentosus strainsLB2F2 and LB3F2were identified as the best biofilm formers. Besides, these LAB isolateswere able to producepro-teinaceous substances with antagonism against the aforementioned spoilage isolates, when grown in MRS orTSB-YE media. During the screening, L. pentosus LB3F2 exhibited the highest antibacterial activity when grownin TSB-YE medium at 30 °C. Additionally, L. pentosus LB3F2 was able to strongly hamper the adhesion ofS. aureus SA3 on abiotic surfaces as polystyrene and stainless steel slides. LAB isolates did not show any hemolyticactivity and all of themwere sensitive to different families of antibiotic tested. It should be pointed out that LB3F2isolate was not cytotoxic on the intestinal cells but could stimulate their metabolic activity. This report unveiledthe potential of LB1F2, LB14F1, LB15F1, LB2F2, and LB3F2 isolates to be used as natural barrier or competitiveexclusion organism in the food processing sector as well as a positive biofilm forming bacteria.

© 2014 Elsevier B.V. All rights reserved.

1. Introduction

Biofilms are sustainable sources of contaminations that are responsi-ble of food-related illness and important economic losses. This lifestyleconfers protection to bacterial cells and decreases the efficiency ofcleaning and disinfection procedures. Consequently, the need for thedevelopment of new strategies and anti-biofilms agents allowing theprevention and control of biofilm-formation by various pathogens is ofmajor importance in order to reduce the risk of contamination in foodsector.

The process of bacterial biofilm formation is occurring in fourdependent stages (Donlan, 2002). The bacterial adhesion stage is as-sociated with the production of exopolyscharides, DNA and proteins.The initial stage of bacterial adhesion was reported to be reversiblebecause of the weakness of the interactions between bacteria andsurfaces, however this stage becomes irreversible as a result of anchor-ing by appendages and/or production of extracellular polymers mainlyexopolysaccharides. Industrial formulations acting on the bacterial

).

adhesion to food contact surfaces such as stainless steel are needed toreduce cross-contamination, food spoilage and transmission of diseasesdue to biofilms. Stainless steel is largely used in the food industriessector because of its mechanical strength, resistance to corrosion, andlongevity (Marques et al., 2007).

Staphylococcus aureus is among themost commonpathogenic bacte-ria isolated from different surfaces such as stainless steel in foodprocessing plants (Pastoriza et al., 2002), where it can adhere andforms biofilms (Kunigk and Almeida, 2001; Archer et al., 2011; Brooksand Jefferson, 2012; Abdallah et al., 2014). Foodborne disease causedby S. aureus is typically intoxication due to the ingestion of enterotoxinspreformed in food by enterotoxigenic strains (Normanno et al.,2007). The development of concepts and the discovery of novelanti-staphylococcal biofilm agents are expected to be promising issuesin food safety. Related to this topic, differentmetabolites such as oreganoessential oil, bacteriophage-derived peptidase, 3-[(3-cholamidopropyl)dimethylammonio] propanesulfonic staphopains, β-2,2-amino acidsderivatives, phenolic compounds and flavonoids have recently beendescribed as potent anti-staphylococcal biofilm agents (Schillaci et al.,2013; Fenton et al., 2013; Luis et al., 2013;Manner et al., 2013). Strategiesusing biofilms produced by the competitive exclusionmicroorganisms to

Table 1Primers used in this work.

Primers Sequences 5′-3′ References

Primer S1Primer S2

5′-AGAGTTTGATC(A,C)TGGCTCAG-3′)5′-GG(A,C)TACCTTGTTACGA(T,C)TTC-3′

Dubernet et al., 2002 /Messaoudi et al., 2011

LbLMA1R16-1

5′-CTCAAAACTAAACAAAGTTTC-3′5′-CTTGTACACACCGCCCGTCA-3′)

Dubernet et al., 2002 /Messaoudi et al., 2011

(GTG)5 5′-GTGGTGGTGGTGGTG-3′ Al kassaa et al., 2014

117F. Ait Ouali et al. / International Journal of Food Microbiology 191 (2014) 116–124

inactivate foodborne pathogens in the food processing environments areof major importance. Kim et al. (2013) showed the inactivation ofEscherichia coliO157:H7 on stainless steel upon exposure to Paenibacilluspolymyxa biofilms. Zhao et al. (2013) reported the reduction ofListeria monocytogenes in a ready-to-eat poultry processing plantby LAB. Pérez-Ibarreche et al. (2014) reported that lactobacilli withbiofilm forming aptitudes were able to control L. monocytogenesbiofilms.

Lactobacilli is an important part of the natural LAB consortium. Theyare considered as potentially antagonistic to spoilage and pathogenicbacteria because of their capabilities to produce an array of inhibitorysubstances such as organic acids, H2O2 and bacteriocins. Lactobacillihave been reported to compete with pathogens in urogenital and intes-tinal tracts (Reid et al., 1988), and interfere with their adhesion oncatheters device (Hawthorn and Reid, 1990; Rodrigues et al., 2004).The use of lactobacilli as anti-biofilm strategies is a promising hope infood safety. Here, two groups of microorganisms were isolated fromthe industrial samples: LAB and non-LAB. Considering LAB natural andfriendly environmental microorganisms, we term the remaining non-LAB as “spoilage,” bearing in mind their possible unfriendly natureand recognizing that some of themmay in fact be foodborne pathogens,such as Salmonella.

The objectives of this study were to explore novel and atypical eco-logical niches, isolate LAB with potential applications as hurdles againstbiofilm forming pathogens and spoilage strains.

2. Materials and methods

2.1. Isolation of lactic acid bacteria and spoilage bacteria

Onehundred and twenty of samples (materials)were collected fromthe inner surfaces of milk tanks and milking machines in two distinctfarms located at Bejaia city (Algeria). The samples were serially diluted(0.85% m/v NaCl) and plated onto the surface of MRS medium (de Manet al., 1960) (Merck, Germany), andM17 agar (Merck, Germany) plates,respectively. Plates were then aerobically and anaerobically incubatedat 30 °C for 24–48 h. Gram-positive and catalase negative isolateswere assumed as LAB. Basic microbiological methods were employedfor characterization of non-LAB isolates. The presence of E. coliwas test-ed on Eosin Ethylene Bleu agar plates (Merck), Salmonella clones wereidentified using Hektoen agar plates (Merck), and S. aureus isolateswere selected on Chapman and Baird Parker media (Merck), followedby the assay for coagulase and DNase activities. Pseudomonas isolateswere identified on nutritive agar (Fluka, Spain) and characterized fortheir pigmentation.

To keep the viability of the newly isolated strains, all of them wereperiodically streaked in MRS or Nutrient agar (Fluka). All isolates werekept at −80 °C in the presence of 20% glycerol (Sigma, France).

2.2. Assessment of LAB antagonism and preliminary of characterization ofthe inhibitory substances

The antagonism of 130 LAB isolates were tested against the afore-mentioned “spoilage” isolates using the spot method (Jacobsen et al.,1999) with some modifications. Briefly, 5 μl of overnight MRS culturesof LAB isolates grown in anaerobic jars (GasPak, BBL) to avoid the effectof H2O2, were spotted onto 1.5%MRS agar. Plates were dried for 30 minat room temperature, incubated under anaerobic conditions at 37 °C for18 h; and overlaid with 10 ml of Brain Heart Infusion (BHI) (Merck)medium containing 0.75% agar seeded with 1% (v/v) of overnightculture of the indicator strain, leading to a final concentration of about106 cells/ml. The incubation was carried out aerobically at 37 °C for18 h. The sterile MRS broth was tested as a negative control.

To identify the inhibitory substances secreted into the growthmedium, LAB isolates presenting antagonism were grown overnight at30 °C in 20 ml MRS broth. The cell-free supernatant (CFS) obtained by

centrifugation (8000 ×g, 20 min, 4 °C), filtered with 0.22 μm-pore-sizeAcrodisc® syringe filters (Pall Gelman Laboratory, USA) was separatedinto three samples named 1, 2 and 3. Sample 1 was directly tested;sample 2 was adjusted to pH 6.5 with 1 N NaOH (Merck-eurolab, BriareLe Canal, France) to rule out the hypothesis of acid inhibition. Sample 3was treated with catalase (300 U/ml) (C-3515, Sigma-Aldrich Chemie,Steinheim, Germany) to rule out the hypothesis of inhibition by theH2O2. The antagonistic activities of these three sampleswere determinedat least in triplicate for each LAB isolate using the well diffusion assay(Ennahar et al., 1998). The absence or presence of any inhibitory zonewas recorded after 18 h of incubation at 37 °C. The CFS obtainedfrom the antagonistic LAB isolates were treated with proteases such asα-chymotrypsin, proteinase K and papain (Sigma-Aldrich Chemie,Steinheim, Germany). Each CFS was adjusted to pH 6.5 with 1.0 mol/lNaOH (Sigma), filter sterilized (0.22 μm), treated with proteases at1.0 mg/ml and left for 1 h at 30 °C. The treated and untreated CSF(control samples)were heated at 100 °C for 5min and then immediatelycooled in ice to inactivate the proteases. The residual activity of treatedand untreated samples was determined by measuring the diameter ofinhibition zones according to the above cited methods.

2.3. Biomolecular typing of the antagonistic LAB isolates

The identification of the antagonistic LAB isolates was carried out bythe API 50 CHL system (Biomérieux, France), and the 16S rDNA se-quence analysis. For the last method, total DNA was extracted fromLAB isolates with theWizard® Genomic DNA purification Kit (PromegaCorp., France). The amplification of 16S rDNA was done with primersS1 and S2 (Table 1), and the following PCR program: denaturation at95 °C/3 min, 29 cycles at 94 °C/40 s, annealing at 55 °C/50 s andextension at 72 °C/1 min, followed up by a final extension cycle at72 °C/10min. The PCR ampliconswere separated on 0.6% (w/v) agarosegel upon electrophoresis carried out at 100 V for 1 h. The ampliconswere purified, cloned into the pGEM®-T Easy Vector System (PromegaCorp., France), and transferred to E. coli JM109 strain. The recombinantplasmids containing the 16S rDNA were extracted by GeneJET plasmidMiniprep (Fermentas), and sequenced at Eurofins MWG Operon(Germany). The resulting sequences were assembled into a uniquecontig with BioEdit sequence alignment software and analyzed usingthe NCBI-Standard Nucleotide BLAST (http://blast.ncbi.nlm.nih.gov).Alignment with known 16S rDNA sequences in the NCBI database(http://www.ncbi.nlm.nih.gov/BL ASTU) was done with the basic localalignment search tool, an online software. The sequences have beendeposited in gene banks and were assigned the following accessionnumbers: KF923749, KF923750, KF923751, KF923752 and KF923753.

LAB isolates with the highest antibacterial activity were character-ized at the molecular level using a specific Lactobacilli PCR previouslydescribed by Dubernet et al. (2002), and a REP-PCR fingerprintingtechnique (Al Kassaa et al., 2014). The specific Lactobacilli PCR requiredthe use of primers LbLMA1 and R16-1 (Table 1) and the following PCRprogram: (i) denaturation at 95 °C/5min, (ii) 34 cycles of denaturationat 94 °C/1 min, annealing at 55 °C/1 min and extension at 72 °C/1 min,followed by (iii) a final extension at 72 °C/5 min. The 50 μl reactionmixture contained 25 μl of DreamTaq™ Green PCR Master Mix 2X(dNTPs, MgCl2 1.5 mM), reaction buffer DreamTaq™ polymerase(Fermentas), 2 μl of each primer (20 mM) and 5 μl of DNA sample of

118 F. Ait Ouali et al. / International Journal of Food Microbiology 191 (2014) 116–124

cellular suspension from a single colony grown overnight and freeRNase and DNase water (16 μl).

The REP-PCR was carried out using the primer (GTG)5 (Table 1) andthe following PCR program: (i) denaturation at 95 °C/7 min, (ii) 29 cy-cles of denaturation at 94 °C/1min, annealing at 40 °C/1min and exten-sion at 72 °C/8 min, followed by (iii) a final extension at 72 °C/16 min.The 50 μl reaction mixture contained 25 μl of DreamTaq™ Green PCRMaster Mix 2× (dNTPs, MgCl2 1.5 mM), reaction buffer DreamTaq™polymerase (Fermentas), 5 μl of each primer (20 mM) and 5 μl of DNAsample of cellular suspension from a single colony grown overnightand free RNase and DNase water (15 μl).

The amplicons gathered from the specific Lactobacilli PCR andREP-PCR were separated on 1.2% (w/v) and 0.7% (w/v) agarose gels,respectively. In both cases, the electrophoresis was carried out at100 V for 1 h and the gels were stained with 0.5 μg/ml gel red.

2.4. Adhesion of the LAB and spoilage isolates to polystyrene tissue cultureplates (TCP)

The semi-quantitativemethod of adhesion to polystyrene tissue cul-ture plates described by O'Toole and Kolter (1998) was used with somemodifications. Briefly, 100 μl of each culture (LAB isolates, 108 CFU/mland spoilage isolates, 106 CFU/ml) in TSB-YE (TSB supplemented with0.6% yeast extract, Difco, France) were added to the wells of sterile 96-well polystyrene tissue culture plates (Nunc®, polystyrene, France),and incubated for 24 h at 30 °C in a humid atmosphere created byplacing the microplate in a box containing four beakers filled withwater. Cultures were decanted and wells were washed twice with ster-ile distilled water to remove the non adherent cells. The adherent cellsin each well were fixed with 200 μl of 99% ethanol (Sigma, France),and after 15 min the plates were emptied and left to dry, then theywere stained for 45 min with 200 μl of 0.1% crystal violet (BiochemChemopharma, Québec, Canada). The stained biofilms were rinsedthree times with 200 μl of distilled water and extracted with 200 μl of33% (v/v) glacial acetic acid (Sigma, France). The content of each well(125 μl) of each well was then transferred to new sterile microplate,and the amount of biofilm was quantified by measuring the OD595nm

using a microplate reader (ELX800. Bio-Tek, USA).

2.5. Inhibition of the spoilage isolates by the LAB forming biofilms

LAB and spoilage isolates at 108 CFU/ml, and 106 CFU/ml, respective-ly were grown in TSB-YE, as appropriate medium for biofilm formation,and incubated at 37 °C for 18 h. Afterwhich, CFS, obtained by centrifuga-tion (8000 × g, 20 min, 4 °C) were filtered with 0.22 μm-pore-sizeAcrodisc® syringe filters. The 96-well polystyrene microplates were in-oculated with 50 μl of overnight spoilage cultures and 50 μl of the LABneutralized CFS. The microplates were left stirring for 15 min before tobe incubated for 24 h at 37 °C in a humid atmosphere. Afterwards, themicroplates were rinsed three times with 200 μl of distilled water. Theadherent cells in each well were fixed with 200 μl of 99% ethanol(Sigma, France). After 15 min, the plates were emptied and left to dry,then were stained for 20 min with 200 μl of 0.1% crystal violet (BiochemChemopharma, Québec, Canada). The stained biofilms were rinsed threetimeswith 200 μl of distilledwater and extractedwith 200 μl of 33% (v/v)glacial acetic acid (Sigma, France). The content of each well (125 μl) wastransferred to new sterile microplate. The amount of biofilmwas quanti-fied by measuring the OD595nm using a microplate reader (ELX800.Bio-Tek, USA) as described above. Each experiment was performed intriplicate using independently grown cultures.

2.6. Inhibition of the biofilm formation on stainless steel AISI 304L

2.6.1. Stainless steel cleaning treatmentBefore carrying the adhesion assays, the stainless steel slides were

cleaned according to the procedure described by Abdallah et al. (2014).

They were kept for a night in 70% ethanol (Sigma) for disinfection, andthen washed twice with sterile distilled water until complete removalof the ethanol. Cleaning procedure of the stainless steel slides was pur-sued for 20 min by immersion, under agitation conditions, in 200 ml ofan alkaline detergent 1% (v/v) TFD4 solution (initial temperature20 °C). The slides were rinsed with sterile distilled water followed byfive successive immersions of 1 min each in sterile ultra pure water(milliQwater,Millipore)with agitation. Finally, theywere dried, individ-ually enveloped and autoclaved at 120 °C for 20 min.

2.6.2. S. aureus SA3 strain cultureOne milliliter of S. aureus SA3 culture at 105 CFU/ml was seeded in

50 ml of TSB and incubated for 15 h at 37 °C with agitation(160 rpm). The cells pellet recovered by centrifugation (5000 ×g,20 min), was washed twice with 20 ml of PBS buffer [phosphate buff-ered saline; 8 g NaCl, 0.2 g KCl, 1.44 g Na2HPO4, and 0.27 g KH2PO4

per liter; pH 7 corrected with HCl/NaOH, and resuspended in the samebuffer. Serial dilutions were made in (0.85% w/v) NaCl (Sigma-AldrichChemie, Steinheim, Germany) in order to have an inoculum of107 CFU/ml.

2.6.3. Kinetic of S. aureus SA3 biofilm development on AISI 304LTwenty AISI 304L stainless steel slides prepared as been described

before were used in this study. Two milliliters of S. aureus SA3 suspen-sion at 107 CFU/ml (as described before) were deposited on each slidesurface and maintained for 1 h at 37 °C. After the initial adhesion timeof 1 h, the bacterial suspension was removed and slides were washedtwice with PBS to remove loosely attached cells. The adhered cellswere observed using epifluorescence microscopy or detached for cellcounting on Tryptic Soy Agar (TSA) plates. The bacterial counts, andmicroscopic observations were presented as a 0 h of biofilm formation.For the biofilm formation assays, washed slides were placed in sterilePetri plates, the upper face was covered with 2 ml of TSB and slideswere incubated, at 37 °C. The biofilm formation was monitored bycounting sessile cells, and biofilm observation using epifluorescencemi-croscopy, after 3, 6, and 24 h. The counts of S. aureus SA3 adherent bac-teria to the slides surfaces was performed as follow: The TSB mediumwas removed, the slides were washed twice with PBS phosphate buffer(pH 7), and immersed in 30ml of the same buffer before to be subjectedto sonication for 5 min at 50 kHz.

2.6.4. Epifluorescence microscopyBiofilm cells were stained with a BacLight LIVE/DEAD bacterial viabil-

ity staining kit according to the manufacturer's instructions (MolecularProbes, Invitrogen, France). In brief, 1.5 μl of each reagent were dilutedin 1ml of physiological water (0.85%m/v NaCl). Then 1ml of themixturewas gently deposited on the upper face of the slide (i.e. on the face of thebiofilm development). After 15 min incubation of the slides in the dark,the staining solution was aspirated and biofilms were observed usingan epifluorescence microscope (Nikon Optiphot-2 EFD3, Japan).

2.6.5. Conditioning of slides with the LAB neutralized supernatantTwo milliliters of neutralized CFS prepared as above described were

deposited on the surface of each AISI 304L slide placed in sterile Petriplates, andmaintained for 2 h at 37 °C. This step may lead to the forma-tion of a conditioning film on the AISI surface. In parallel, sterile TSB-YEwas used as control for the conditioning step of the stainless steel. Afterthis period, the CFS and the TSB-YE were removed and replaced by 2mlof S. aureus A3 suspension at 107 CFU/ml prepared as described inSection 2.6.2. At the end of 1 h of incubation time, the buffer containingthe non-adherent S. aureus SA3 cells was aspirated and 2 ml of sterileTSB medium were deposited on the surface of each coupon for 0, 3, 6,and 24 h incubation to monitor the installation of S. aureus A3 biofilm.After each time of incubation, the slides were washed twice with30 ml of PBS buffer (pH 7.0). Finally, the slides were immersed individ-ually in 30 ml phosphate buffer and sonicated. The detached S. aureus

Table 2API 50 CH and 16S rDNA sequencing identifications of the selected LAB strains.

LAB strains API 50 CH 16S rDNA

LB1F2 L. brevis L. brevisLB2F2 L. plantarum L. pentosusLB3F2 L. plantarum L. pentosusLB14F1 L. brevis L. brevisLB15F1 L. brevis L. brevis

119F. Ait Ouali et al. / International Journal of Food Microbiology 191 (2014) 116–124

SA3 cells were enumerated by plating the bacteria on TSA after growthat 37 °C for 24 h. Additional slides were prepared and served to theepifluorescence observation after their staining with live/dead compo-nents as described in Section 2.6.4.

2.6.6. Antibiotics susceptibilityThe susceptibility of the antagonistic LAB isolates was tested against

nearly all antibiotic classes encompassing penicillin, cephalosporins,aminosides, tetracycline, macrolides, glycopeptides, and polypeptide(MAST, UK). The bacterial suspensions of about 107 CFU/ml were seed-ed onto MRS agar plates by the flooding technique. The plates were airdried for 15 min, and then disks impregnated with antibiotics weredeposited on the plates. The formation of inhibition zones around thedisks was determined after 24 h of incubation at 37 °C. The susceptibil-ity to these antibiotics was determined based on the recommendationsof the Antibiogram Committee of the French Microbiology Society.

2.6.7. Crude measure of hemolytic activityTo test the crude measure hemolytic activity, fresh cultures of iso-

lates were streaked on blood agar (Columbia base, Biokar Diagnostics),containing 5% (w/v) sheep blood and incubated for 48 h at 30 °C. Afterthis period, blood agar plates were examined for the presence of thehemolytic zones around colonies (Ghrairi et al., 2008).

2.6.8. Cytotoxicity assessmentCytotoxicity of Lactobacillus pentosus LB3F2 against intestinal cells

was investigated using Cell Proliferation Kit II (XTT) (Roche AppliedScience, USA). This assay was based on the reduction of a tetrazoliumsalt (XTT) into yellow formazan salt by active mitochondria. STC-1 cellline was a gift gratefully received from Dr. C. Roche (INSERM U865,Lyon, France). This cell line is derived from an endocrine tumor devel-oped in the small intestine of double transgenic mice (Rindi et al.,1990). All chemicals for the cell culture were from PAN-Biotech GmbH(Germany). Cells were routinely grown in 75 cm2

flasks at 37 °C, 5%CO2 atmosphere in Dulbecco's modified Eagle's Medium (DMEM) sup-plemented with 4.5 g/l glucose, 5% of fetal calf serum, 2 mM glutamine,100 U/ml penicillin and 100 μg/ml of streptomycin.

When 80% confluence were reached, cells were trypsinized andseeded into 96-well plate at a density of 7000 cells per well in 100 μlof DMEM. After 48 h of incubation at 37 °C, and under 5% CO2 atmo-sphere, confluent cells were washed twice with PBS. In parallel,L. pentosus LB3F2 cells were harvested by centrifugation (8000 ×g,10 min, 4 °C) from MRS broth 18 h old cultures and washed twicewith 2ml ofminimummedium(DMEMwithout serumand antibiotics).Absorbance (490nm vs 655nm reference) was measured in a microplatereader spectrophotometer (ELx808, BioTek, USA). For each condition, ab-sorbance of wells without STC-1 cells and with bacteria was subtractedfrom absorbance of wells containing STC-1 cells and bacteria. Resultswere expressed as percentage of proliferation.

2.6.9. Statistical analysisAll data in this study represented the mean of three experimental

replicates. Statistical comparisons among the different results obtainedby the different tests were performed by one-way analysis of varianceusing XL-STAT with Student–Newman–Keuls method (version 2009).

Significant differences among adherent cell counts on differentstainless steel slides surfaces, untreated, treated with TSB or with theCFS, were determined by one-way analysis of variance using XL-STATsoftware version 2009 to determine differences among various treat-ments at a 95% confidence level.

3. Results

Of 400 bacterial isolates recovered from the milk tanks and milkingmachine surfaces at Bejaia city (Algeria), 130 of them were consideredas presumptive LAB because of the positive Gram staining and the

absence of catalase activity, whereas 100 isolates were identified asE. coli (32%), Salmonella sp., (22%), S. aureus (20%) and Pseudomonasaeruginosa (22%) and considered thereof as “spoilage” isolates. Theother 170 strains were other than the aforementioned spoilageisolates.

Notably, the antagonistic isolates were identified by the API 50 CHLsystem and 16S rDNA sequence analysis. These two independentmethods of identification fully reliable as only three species of fivewere identified similarly (Table 2). The use of genus-specific lactobacilliPCR proposed by Dubernet et al. (2002) allowed the amplification of250 bp DNA fragment from total DNA extracted from LAB isolates argu-ing on their belonging to the Lactobacillus genus (Fig. 1A). The related-ness between LB1F1, LB2F2, LB3F2, LB14F1 and LB15F1 isolates asstudied by the REP-PCR (GTG)5 fingerprinting approach and interpretedby the numerical analysis permitted three groups based on a distancecut-off of 70% relatedness (Fig. 1B). The Lactobacillus brevis isolateswere not closely related each to other. LB14F1 and LB15F1 (L. brevis),LB2F2 and LB3F2 (L. pentosus) were contained in two distinct groups;whereas LB1F2 (L. brevis) was found in the third group that was morelinked to L. brevis (Fig. 1B).

Further, the cultivation of the LAB isolates in TSB-YE permitted toshow their adhesive and biofilm-forming properties particularly forLB1F2, LB2F2, LB3F2, LB14F1 and LB15F1 isolates. LB2F2 and LB3F2 iso-lates were classified as strong biofilm producers with an OD595 nm of0.78 and 0.72, respectively (Fig. 2). Regarding, the spoilage isolates,S. aureus SA3 with an OD595nm of 0.6 was considered as well as astrong biofilm producer at 37 °C, whereas, the other ones includingS. aureus, E. coli, P. aeruginosa and Salmonella sp. showed a moderate(0.22≥ OD≥ 0.12) even a weak potential (0.12≥ OD≥ 0.08) of bio-film formation at 37 °C. Remarkably, the identification of S. aureus SA3was recently confirmed by pyrosequencing (unpublished results).Statistical analysis revealed significant differences (P b 0.05) of theaptitudes of the biofilm formation for the different LAB isolates, exceptfor LB2F2 and LB3F2 isolates (P N 0.05). These data underline the highdiscrepancy in terms of the biofilm formation of LAB isolated from thesame niche, even between LAB belonging to the same genus andspecies. LB1F1, LB2F2, LB3F2, LB14F1 and LB15F1 isolates exhibitedthe highest biofilm-forming properties.

The LAB isolates with the highest adhesion properties were testedfor their antagonism against the 100 spoilage isolates obtained fromthe same ecological niche. Although, all the LAB isolateswere antagonis-tic, the potency of their inhibition was exerted in strain dependentmanner. The antagonismof this LABwas confirmed by the agar-well dif-fusion method using the native and neutralized supernatants (Table 3).The best antagonismwas observed for LB1F2, LB2F2, LB3F2, LB14F1 andLB15F1 isolates which meanwhile were the best biofilm-formingisolates. Among these five LAB isolates, the highest antagonism wasregistered for LB3F2 isolate. The experimental conditions allowing opti-mization of the antibacterial activity of LB3F2 isolate were obtainedwith the culture supernatant (pH 5.5) after its growth in TSB-YE brothfor 15 h at 30 °C.

To identify the metabolites responsible of such antagonism, theneutralized CSF (pH 6.5) gathered from LB1F2, LB2F2, LB3F2, LB14F1and LB15F1 LAB isolates were tested after treatment with variousmetabolite-specific enzymes. Afterwards, the inhibitory activity was

Fig. 1. (A) Specific Lactobacilli PCR amplification. Lanes 1 to 5 correspond to amplicons gathered from LB15F1, LB14F1, LB3F2, LB2F2 and LB1F2 isolates respectively. Lane C corresponds tothe negative control and laneM to DNAmolecular weight markers. The arrow indicates the PCR product of 250 bp usually obtained for LAB of Lactobacillus genus. (B) Pattern of Rep-PCRfingerprinting technique of the selected LAB isolates.

120 F. Ait Ouali et al. / International Journal of Food Microbiology 191 (2014) 116–124

recorded after addition of catalase to the CFS excluding therefore the in-volvement of hydrogen peroxide.

However, the addition of proteases as papain, α-chymotrypsin andproteinase K abolished the antagonism indicating the proteinaceous na-ture of the secreted inhibitory substances. Additionally, the inhibitorysubstances were insensitive to the heat treatment at 100 °C for 5 min.All these data let us to conclude that LB1F2, LB2F2, LB3F2, LB14F1 andLB15F1 LAB isolates were able to produce bacteriocins or bacteriocins-like inhibitory substances (BLIS).

Further, the CFS from LAB isolates LB1F2, LB14F1, LB15F1, LB2F2 andLB3F2 were able to impede the adhesion and subsequently the biofilmformation of S. aureus SA3 based on the data obtained with the semi-quantitative TCP method (Fig. 3). These data were strengthened bythe P-value recorded (P b 0.05) for the adhesion levels in the absenceand presence of CFS. The conditioning of AISI 304L stainless steel slidesfor 2 h with the neutralized CFS from L. pentosus LB3F2 permitted a de-crease along time of the viable S. aureus SA3 adherent cells arguingclearly the effectiveness of the bacteriocin(s) or BLIS secreted into thegrowth medium.

Fig. 2. Biofilm formation of LB2F2, LB3F2, LB15F1, LB1F2 and LB14F1 isolates. ODwas usedto quantify the biofilm formation. The data are themeans of at least three independent ex-periments. C designates the negative control, which corresponds to sterile TSB-YE. Theerror bars represent the standard deviations.

The effects of L. pentosus LB3F2 on the intestinal STC-1 cell metabolicactivity were assayed using XTT test (Fig. 4). The results showed thatafter 3.5 h and 4 h of contact LB3F2 was not toxic for STC-1 cells at theconcentrations of 105, 106 and 107 CFU/ml. Moreover, after 4 h of con-tact, LB3F2 at 107 CFU/ml exerted a positive effect on the intestinalcell metabolism activity. The data depicted in Fig. 5 showed the numberof S. aureus SA3 cells adhering on the stainless steel slides at 37 °C. Thisnumber appeared to increase from4 to 7 Log CFU/ml after 6 h of contact,and to 8 Log CFU/ml after 24 h of contact.

The DNA binding dye Syto-9 labeled viable cells (green) allowed totrack the biofilm formation by S. aureus SA3 on the untreated stainlesssteel slides surfaces after a contact of 0, 3, 6 and 24 h at 37 °C (Fig. 6A,B, C, D). This experiment was conducted on stainless steel slides treatedwith neutralized supernatants. Untreated slides and TSB treated oneswere used as negative controls. The results showed that, after 24 h ofbiofilm development, S. aureus SA3 adhered much less on the condi-tioned slides (Fig. 6H) (i.e. treated with native supernatant), than onuntreated, and TSB treated, slides (Fig. 6D). In addition, our dataunderlined that, in the first 6 h of biofilm formation, most of attachedcells on conditioned slides were predominantly stained with thepropidium iodide (Fig. 6E, G), which underlines the antibiofilm efficacyof such treatment.

The hardy S. aureus SA3 adheredmuch less on the conditioned slides(Fig. 6E) and an important number of cells labeled in red showing notonly the inhibition of adherence but also the induction of cell death(Fig. 6E, F, G, H).

Notably, while no crudemeasure of hemolytic activity was observedafter 48 h of incubation, the aforementioned isolates appeared to besensitive to all the antibiotics tested, except for ciprofloxacin, fosfomycin,and vancomycin (Table 4).

4. Discussion

It is important to study the interactions between bacteria and thesurfaces in a specific food processing environment in order to providemore effective measures for prevention of biofilm formation and forits removal. The aim of this study was to examine themicrobial consor-tium cohabiting in the dairy environment such as the milk tanks andmilking machines used in two traditional and distinct farms located atBejaia City, Algeria. After describing the lactobacilli and spoilage

Table 3Antagonism of the selected lactobacilli strains before and after pH neutralization of the culture supernatant. The data are the means of at least three independent experiments. The targetstrains usedwere isolated from the same ecological niche as the lactobacilli producing inhibitory substances. In each column there are two data (before neutralization “BN” and after neu-tralizing “AN” of the supernatant).

LAB strain Diameter of the inhibition zones (mm)

E. coli S. aureus P. aeruginosa

E1 E2 E3 SA1 SA2 SA3 SA4 P1 P2 P3

BN AN BN AN BN AN BN AN BN AN BN AN BN AN BN AN BN AN BN AN

L. pentosus LB3F2 10 05 10 03 11 05 17 08 17 07 19 09 18 06 08 03 09 03 08 02L. pentosus LB2F2 09 03 10 03 10 04 15 04 16 05 15 05 15 05 09 03 10 04 11 03L. brevis LB1F2 10 04 11 04 11 03 12 04 14 04 13 05 14 04 08 03 09 03 10 03L. brevis LB14F1 10 04 10 03 09 03 15 05 15 03 13 04 13 03 10 03 10 03 09 04L. brevis LB15F1 08 03 09 03 08 03 12 03 15 03 15 03 14 03 09 03 10 03 10 03

150

Ctrl

121F. Ait Ouali et al. / International Journal of Food Microbiology 191 (2014) 116–124

bacteria contained in this consortium, our investigations focused on thelactobacilli because of their potential of applications.

This study permitted the recovery of at least 130 LAB isolates amongwhich LB1F2, LB2F2, LB3F2, LB14F1 and LB15F1 were able to formbiofilms, and to produce bacteriocins like inhibitory substances thatalter the formation of S. aureus SA3 biofilm on stainless steel slides.

Bacteriocins are small antimicrobial peptides ribosomally synthe-sized by Gram positive and Gram negative bacteria (Drider andRebuffat, 2011) exhibiting thereof safe (Belguesmia et al., 2011), andpossibilities of biotechnological applications (Yang et al., 2014). Of par-ticular interest the bacteriocins produced by lactobacilli which are ableto influence septoformation, peptidoglycan and protein synthesis, affectcytoplasmic membranes leading to their destabilization and conse-quently to cell death (Rybal'chenko et al., 2013). Bacteriocins producedby L. pentosus and L. brevis species designed as pentocins and brevicins,respectively are not very abundant in the literature. Pentocin TV35b andpentocin 31-1with antagonism against Gram positive bacteria havebeen reported (Okkers et al., 1999; Zhang et al., 2009). We hypothesizethat BLIS produced by LB15F1, LB2F2 and LB3F2 isolates are differentfrom pentocins TV35b and 31 previously reported because of thediscrepancies in their spectra. Further, the potential of the bacteriocinproduced by L. brevis species is not very well documented as well. Todate only few bacteriocins including brevicin 27 and brevicin 286 havebeen reported by Benoit et al. (1997) and Coventry et al. (1996)respectively.

The level of adhered S. aureus SA3 cells on AISI 304 has increased,within 24 h, from 4.7 to 8 log CFU/ml. In this process, the temperatureof 37 °C appeared to play a positive role on the adhesion process.Morton et al. (1998) pointed out the role of the optimum growth tem-perature on the adhesion process. According to de Oliviera et al.(2010) the initial inoculum size might influence the bacterial adhesionprocess as well. Related to this last point, Zottola and Sasahara (1994)

Fig. 3. Adhesion inhibition of S. aureus SA3 to polystyrene microplates by LAB strains cellfree supernatants (CFS): CFS1: LB2F2, CSF2: LB3F2, CSF3: LB15F1, CFS4: LB1F2, CFS5:LB14F1. The positive control (PC) was the untreated S. aureus culture, and the negativecontrol (NC) was the TSB-YE alone. The data are the means of at least three experiments.The error bars represent the standard deviations.

suggested earlier that the initial number of bacterial cells for biofilmformation should be 106 to 107 CFU/ml.

The anti-adhesive and anti-biofilm-forming properties of lactobacillihave been assessed in different circumstances. Indeed, Abedi et al.(2013) showed the good anti-adhesive properties of Lactobacillusdelbrueckii against E. coli in different competitive conditions. Vuottoet al. (2013) reported that L. brevis CD2 prevents the biofilm formationby Prevotella melaninogenica.

Based on the data obtained in this study, LB1F2, LB14F1, LB15F1,LB2F2 and LB3F2 isolates are considered as good candidates for applica-tions as natural barriers or competitive-exclusion microorganisms tocontrol staphylococcal biofilm formation. To strengthen this optionof application, the antibiotic susceptibility and hemolytic activity ofLB1F2, LB14F1, LB15F1, LB2F2 and LB3F2 isolates were investigated.They resulted sensitive to the most antibiotics tested except forfosfomycine, vancomycin and ciprofloxacin. Resistance of Lactobacillusspp. to antimicrobial agents was reported to be species-dependent(Danielsen and Wind, 2003). Related to this, resistance to vancomycinand ciprofloxacin was reported by Vay et al. (2007) and Danielsen andWind (2003) for the homofermentative lactobacilli group. Temmermanet al. (2003) examined the resistance of bacterial isolates recoveredfrom 55 European probiotic products using the disc diffusion method.According to these authors, 79% of the isolates were resistant tokanamycin, 65% to vancomycin, 26% to tetracycline, 23% to penicillinG, 16% to erythromycin and 11% to chloramphenicol. It should bepointed out that the natural bacterial resistance to antibiotics is notas amajor risk to animal or humanwelfare, contrarily to the acquiredresistance which is propagated by plasmids and transposons elements(Gueimonde et al., 2013). Moreover, the isolates LB1F2, LB14F1,

Time of incubation (h)3.5 4.0

Prol

ifera

tion

(% o

f con

trol

)

50

75

100

125

105

106

107

b

a aa

a

aa a

Fig. 4. Effects of L. pentosus LB3F2 on the STC-1 cell mitochondrial activity. XTT's responseafter 3.5 and 4 h of incubation with LB3F2 at 105, 106 and 107 CFU/ml. Data are themean ± SD of six values. Means without a common letter are different (p b 0.05) usingone way ANOVA with Student–Newman–Keuls method.

Time (hours)

log

CFU

/ml

2

4

6

8

10

aa

bb

ccd d

ef

g g

h

0 2463

Fig. 5. Adhesion of S. aureus SA3 cells (Log CFU/ml) at different incubation times, to stain-less steel surfaces without (■) and after conditioning with sterile TSB medium ( ) orL. pentosus LB3F2 supernatant ( ) for 2 h. The error bars represent the standard deviationson bacteria counting. The data are the means of at least three independent experiments.Means without a common letter are different (p b 0.05) using one way ANOVA withStudent–Newman–Keuls method.

Fig. 6. Epifluorescence images of S. aureus SA3 adhered cells on slides treated with TSB (A, B, C, D) and neutralized CFS (E, F, G, H). Sessile cells were stained with live/dead BacLight™bacterial viability kit after 1 h (A, E), 3 h (B, F), 6 h (C, G) and 24 h (D, H).

Table 4Antibiotic susceptibility of the studied LAB strains.

Antibiotic Concentration LB1F2 LB14F1 LB15F1 LB2F2 LB3F2

SXT:TS:cotrimoxadole(trimethoprim/sulphamethoxazole)

25 μg S S S S S

Tetracycline 30 μg S S S S SRifampin 30 μg S S S S SCefepine 30 μg S S S R RCefotaxime 30 μg S S S S SAmpicilin 10 μg S S S S SPenicillin G 10 UI S S S S SOxacillin 5 μg R R R S SCefoxitin 30 μg S S S S SKanamycin 1 μg S S S S SLincomycin 5 μg R S S R RCiprofloxacin 5 μg R R R R RGentamycin 500 μg S S R S SStreptomycin 500 μg S R S S SErythromycin 15 μg S S S S SVancomycin 30 μg R R R R RFosfomycin 50 μg R R S R R

R = resistant, S = sensitive.

122 F. Ait Ouali et al. / International Journal of Food Microbiology 191 (2014) 116–124

123F. Ait Ouali et al. / International Journal of Food Microbiology 191 (2014) 116–124

LB15F1, LB2F2 and LB3F2 did not show any hemolytic sign. Concretely,this study unveiled the potential of the LB1F2, LB14F1, LB15F1, LB2F2and LB3F2 isolates to inhibit the growth of both Gram positive andGram negative bacteria isolated from the same ecological niche. Addi-tionally, the aforementioned LAB isolateswere able to strongly interfereon the staphylococcal biofilm formation. Of particular interest LB3F2that was subjected to additional studies. Indeed, this isolate was notonly non cytotoxic on the intestinal STC-1 cells but stimulated theirmetabolic activity. Regarding this last point, more experiments areneeded to confirm the ability of LB3F2 to stimulate proliferation of in-testinal cells. To attribute nutraceutical applications to LB3F2 isolate,its ability to reduce pathogen adhesion and associated cytotoxicityagainst STC-1 cells remain to be done.

The presence of foodborne pathogens in milk and in the dairy envi-ronment could be originated by several outbreaks including the inges-tion of contaminated feed followed by their amplification in bovinehosts and fecal dissemination in the farm environment (Oliver et al.,2005) As for the Brazilian diary farms (Lee et al., 2014), the prevalenceand persistence of S. aureus biofilm forming bacteria is of publichealth concern because unprocessed milk is regularly consumed bythe Algerian population.

This work highlighted the impact of lactobacilli LB14F1, LB15F1,LB2F2 and LB3F2 isolated fromdairy environment on the staphylococcalbiofilm formation by altering their adhesion on the steel stainless slides.Overall LB1F2, LB14F1, LB15F1, LB2F2 and LB3F2 offer pertinent aca-demic research and real biotechnological applications.

Acknowledgments

The authors are grateful for the farmers (Akbou and Tazmalt, Bejaia,Algeria) who permitted the sampling in their farms, and rendering thisproject possible. The authors are indebted to Dr. Andres Hidalgo (CNIC,Madrid) and Pr. Mike Chikindas (Rutgers University, NJ) for Englishediting and critical reading of the manuscript.

References

Abdallah, M., Chataigne, G., Ferreira-Theret, P., Benoliel, C., Drider, D., Dhulster, P., Chihib,N.E., 2014. Effect of growth temperature, surface type and incubation time on the re-sistance of Staphylococcus aureus biofilms to disinfectant. Appl. Microbiol. Biotechnol.98, 2597–2607.

Abedi, D., Feizizadeh, S., Akbari, V., Jafarian-Dehkordi, A., 2013. In vitro anti-bacterial andanti-adherence effects of Lactobacillus delbrueckii subsp bulgaricus on Escherichia coli.Res. Pharm. Sci. 8, 260–268.

Al Kassaa, I., Hamze, M., Hober, D., Chihib, N.E., Drider, D., 2014. identification of vaginallactobacilli with potential probiotic properties isolated from women in NorthLebanon. Microb. Ecol. 67, 722–734.

Archer, N.K., Mazaitis, M.J., Costerton, J.W., Leid, J.G., Powers, M.E., Shirtliff, M.E., 2011.Staphylococcus aureus biofilms: properties, regulation, and roles in human disease.Virulence 2, 445–459.

Belguesmia, Y., Madi, A., Sperandio, D., Merieau, A., Feuilloley, M., Prévost, H., Drider, D.,Connil, N., 2011. Growing insights into the safety of bacteriocins: the case ofenterocin S37. Res. Microbiol. 162, 159–163.

Benoit, V., Lebrihi, A., Millière, J.B., Lefebvre, G., 1997. Purification and partial amino acidsequence of brevicin 27, a bacteriocin produced by Lactobacillus brevis SB27. Curr.Microbiol. 34, 173–179.

Brooks, J.L., Jefferson, K.K., 2012. Staphylococcal biofilms: quest for the magic bullet. Adv.Appl. Microbiol. 81, 63–87.

Coventry, M.J., Wan, J., Gordon, J.B., Mawson, R.F., Hickey, M.W., 1996. Production ofbrevicin 286 by Lactobacillus brevis VB286 and partial characterization. J. Appl.Bacteriol. 80, 91–98.

Danielsen, M., Wind, A., 2003. Susceptibility of Lactobacillus spp. to antimicrobial agents.Int. J. Food Microbiol. 82, 1–11.

de Man, J.C., Rogosa, M., Sharpe, M.E., 1960. A medium for the cultivation of lactobacilli. J.Appl. Bacteriol. 23, 130–135.

de Oliviera, M.M., Brugnera, D.F., Alves, E., Piccoli, R.H., 2010. Biofilm formation by Listeriamonocytogenes on stainless steel surface and biotransfer potential. Braz. J. Microbiol.41, 97–106.

Donlan, R.M., 2002. Biofilms: microbial life on surfaces. Emerg. Infect. Dis. 8, 881–890.Drider, D., Rebuffat, S., 2011. Prokaryotic Antimicrobial Peptides: From Genes to Applica-

tions. Springer, (ISBN-13: 978-144197691).Dubernet, S., Desmasures, N., Guéguen, M., 2002. A PCR-based method for identification

of lactobacilli at the genus level. FEMS Microbiol. Lett. 214, 271–275.

Ennahar, S., Aoude-Werner, D., Assobhei, O., Hasselmann, C., 1998. Antilisterial activity ofenterocin 81, a bacteriocin produced by Enterococcus faecium WHE 81 isolated fromcheese. J. Appl. Microbiol. 85, 521–526.

Fenton,M., Keary, R.,McAuliffe, O., Ross, R.P., O'Mahony, J., Coffey, A., 2013. Bacteriophage-Derived Peptidase CHAP(K) Eliminates and Prevents Staphylococcal Biofilms. Int. J.Microbiol. 2013, 625341 (8 pp.).

Ghrairi, T., Frère, J., Berjeaud, J.M., Manai, M., 2008. Purification and characterization ofbacteriocins produced by Enterococcus faecium from Tunisian rigouta cheese. FoodControl 19, 162–169.

Gueimonde, M., Sánchez, B.G., de Los Reyes-Gavilán, C., Margolles, A., 2013. Antibiotic re-sistance in probiotic bacteria. Front. Microbiol. 4, 1–6.

Hawthorn, L.A., Reid, G., 1990. Exclusion of uropathogen adhesion to polymer surfaces byLactobacillus acidophilus. J. Biomed. Mater. Res. 24, 39–46.

Jacobsen, C.N., Rosenfeldt, N.V., Hayford, A.E., Moller, P.L., Michaelsen, K.F., Paerregaaerd,A., Sandstrom, B., Tvede, M., Jakobsen, M., 1999. Screening of probiotic activities offorty-seven strains of Lactobacillus spp. by in vitro techniques and evaluation of thecolonization ability of five selected strains in humans. Appl. Environ. Microbiol. 65,4949–4956.

Kim, S., Bang, J., Kim, H., Beuchat, L.R., Ryu, J.H., 2013. Inactivation of Escherichia coli O157:H7 on stainless steel upon exposure to Paenibacillus polymyxa biofilms. Int. J. FoodMicrobiol. 167, 328–336.

Kunigk, L., Almeida, M.C.B., 2001. Action of peracetic acid on Escherichia coli and Staphylo-coccus aureus in suspension or settled on stainless steel surfaces. Braz. J. Microbiol.38–41.

Lee, S.H., Mangolin, B.L., Gonçalves, J.L., Neeff, D.V., Silva, M.P., Cruz, A.G., Oliveira, C.A.,2014. Biofilm-producing ability of Staphylococcus aureus isolates from Braziliandairy farms. J. Dairy Sci. 97, 1812–1816.

Luis, A., Silva, F., Sousa, S., Duarte, A.P., Domingues, F., 2013. Antistaphylococcal andbiofilm inhibitory activities of gallic, caffeic and chlorogenic acids. Biofouling30, 69–79.

Manner, S., Skogman, M., Goeres, D., Vuorela, P., Fallarero, A., 2013. Systematic explora-tion of natural and synthetic flavonoids for the inhibition of Staphylococcus aureusbiofilms. Int. J. Mol. Sci. 14, 19434–19451.

Marques, S.C., Rezende, J.G.O.S., de Freitas-Alves, L.A., Silva, B.C., Alves, E., de Abreu, L.R.,Picolli, R.H., 2007. Formation of biofilms by Staphylococcus aureus on stainless steeland glass surfaces and its resistance to some selected chemical sanitizers. Braz. J.Microbiol. 38, 538–543.

Messaoudi, S., Kergourlay, G., Rossero, A., Ferchichi, M., Prévost, H., Drider, D., Manai, M.,Dousset, X., 2011. Identification of lactobacilli residing in chicken ceca with antago-nism against Campylobacter. Int. Microbiol. 14, 103–110.

Morton, L.H.G., Greenway, D.L.A., Gaylarde, C.C., Surman, S.B., 1998. Consideration of someimplications of the resistance of biofilms to biocides. Int. Biodeterior. Biodegrad. 41,247–259.

Normanno, G., La salandra, G., Dambrosio, A., Quaglia, N.C., Corrente, M., Parisi, A., et al.,2007. Occurrence, characterization and antimicrobial resistance of enterotoxigenicStaphylococcus aureus isolated from meat and dairy products. Int. J. Food Microbiol.115, 290–296.

O'Toole, G., Kolter, R., 1998. Flagellar and twitchingmotility are necessary for Pseudomonasaeruginosa biofilm development. Mol. Microbiol. 30, 295–304.

Okkers, D.J., Dicks, L.M., Silvester, M., Joubert, J.J., Odendaal, H.J., 1999. Characterization ofpentocin TV35b, a bacteriocin-like peptide isolated from Lactobacillus pentosuswith afungistatic effect on Candida albicans. J. Appl. Microbiol. 87, 726–734.

Oliver, S.P., Jayarao, B.M., Almeida, R.A., 2005. Foodborne pathogens in milk and the dairyfarm environment: food safety and public health implications. Foodborne Pathog. Dis.2, 115–129.

Pastoriza, L., Cabo, M.L., Bernárdez, M., Sampedro, G., Herrera, J.R., 2002. Combinedeffects of modified atmosphere packaging and lauric acid on the stability of pre-cooked fish products during refrigerated storage. Eur. Food Res. Technol. 215,189–193.

Pérez-Ibarreche, M., Castellano, P., Vignolo, G., 2014. Evaluation of anti-Listeria meatborne Lactobacillus for biofilm formation on selected abiotic surfaces. Meat Sci. 96,295–303.

Reid, G., Bruce, A., Smeianov, V., 1988. The role of Lactobacilli in preventing urogenital andintestinal infections. Int. Dairy J. 8, 555–562.

Rindi, G., Grant, S.G., Yiangou, Y., Ghatei, M.A., Bloom, S.R., Bautch, V.L., et al., 1990.Development of neuroendocrine tumors in the gastrointestinal tract of transgen-ic mice. Heterogeneity of hormone expression. Am. J. Pathol. 136, 1349–1363.

Rodrigues, L., van der Mei, H.C., Teixeira, J., Oliveira, R., 2004. Influence of biosurfactantsfrom probiotic bacteria on formation of biofilms on voice prostheses. Appl. Environ.Microbiol. 70, 4408–4410.

Rybal'chenko, O.V., Orlova, O.G., Bondarenko, V.M., 2013. 2013. Antimicrobial peptides oflactobacilli. Zh. Mikrobiol. Epidemiol. Immunobiol. 4, 89–100.

Schillaci, D., Napoli, E.M., Cusimano, M.G., Vitale, M., Ruberto, A., 2013. Origanum vulgaresubsp. hirtum essential oil prevented biofilm formation and showed antibacterialactivity against planktonic and sessile bacterial cells. J. Food Prot. 76, 1747–1752.

Temmerman, R., Pot, B., Huys, G., Swings, J., 2003. Identification and antibiotic suscep-tibility of bacterial isolates from probiotic products. Int. J. Food Microbiol. 81,1–10.

Vay, C., Cittadini, R., Barberis, C., Hernán-Rodríguez, C., Perez-Martínez, H., Genero, F.,Famiglietti, A., 2007. Antimicrobial susceptibility of non-enterococcal intrinsicglycopeptide-resistant Gram-positive organisms. Diagn. Microbiol. Infect. Dis. 57,183–188.

Vuotto, C., Barbanti, F., Mastrantonio, P., Donelli, G., 2013. Lactobacillus brevis CD2 inhibitsPrevotella melaninogenica biofilm. Oral Dis. http://dx.doi.org/10.1111/odi.12186.

Yang, S.C., Lin, C.H., Sung, C.T., Fang, J.Y., 2014. Antibacterial activities of bacteriocins:application in foods and pharmaceuticals. Front. Microbiol. 26 (5), 241.

124 F. Ait Ouali et al. / International Journal of Food Microbiology 191 (2014) 116–124

Zhang, J., Liu, G., Shang, N., Cheng, W., Chen, S., Li, P., 2009. Purification and partial aminoacid sequence of pentocin 31-1, an anti-Listeria bacteriocin produced by Lactobacilluspentosus 31-1. J. Food Prot. 72, 2524–2529.

Zhao, T., Podtburg, T.C., Zhao, P., Chen, D., Baker, D.A., Cords, B., Doyle, M.P., 2013.Reduction by competititive bacteria of Listeria monocytogenes in biofilms and

Listeria in floor drains in a ready-to-eat poultry processing plant. J. Food Prot. 74,601–607.

Zottola, E.A., Sasahara, K.C., 1994. Microbial biofilms in the food processing industry–Should they be a concern. Int. J. Food Microbiol. 23, 125–148.