Inhibition of Listeria monocytogenes Growth, Adherence and Invasion in Caco-2 Cells by Potential Probiotic Lactic Acid Bacteria Isolated from Fecal Samples of Healthy Neonates

Lactic acid bacteria (LAB) isolated from healthy humans may prove an effective tool against pathogen growth, adherence and invasion in intestinal epithelial cells. This study aimed to evaluate the antilisterial properties of LAB isolated from fecal samples of healthy neonates. Forty-five LAB strains were tested for their antimicrobial activity against ten Listeria monocytogenes strains with spot-on-lawn and agar-well diffusion assays, and ten lactobacilli strains were further assessed for their inhibitory effect against adherence and invasion of Caco-2 cells by L. monocytogenes EGDe. Inhibition was estimated in competition, exclusion or displacement assays, where lactobacilli and L. monocytogenes were added to Caco-2 monolayers simultaneously or 1 h apart from each other. Inhibition of L. monocytogenes growth was only displayed with the spot-on-lawn assay; cell-free supernatants of lactobacilli were not effective against the pathogen. Lactobacillus (L.) paragasseri LDD-C1 and L. crispatus LCR-A21 were able to adhere to Caco-2 cells at significantly higher levels than the reference strain L. rhamnosus GG. The adherence of L. monocytogenes to Caco-2 cells was reduced by 20.8% to 62.1% and invasion by 33.5% to 63.1% during competition, which was more effective compared to the exclusion and displacement assays. These findings demonstrate that lactobacilli isolated from neonatal feces could be considered a good candidate against L. monocytogenes.


Introduction
The gastrointestinal (GI) microbiota of healthy humans consists of bacteria, archaea, fungi and viruses that coexist in a mutualistic association with the host, a condition described as eubiosis [1,2]. Gut microbiota contributes to human health, where besides food digestion, it also regulates nutrient metabolism, stimulates the immune system, maintains the integrity of the mucosal barrier and protects the host against pathogens [3]. The microbial colonization of the human gut begins at birth and takes almost three years to reach the complexity and diversity of the adult's gut microbiota [4]. The colonizing bacteria are derived from the mother, breast milk and surrounding environment, and the colonization is influenced by extrinsic and intrinsic factors, which include geographic area, mode of delivery, feeding habits and genetics [5]. The microbial balance can be disturbed by environmental factors, lifestyle or diseases, resulting in ratio alterations of beneficial and potentially harmful microorganisms that can lead to dysbiosis [6]. Probiotics are living microbial strains that are able to exclude or inhibit pathogens and restore the balance in the GI tract to enhance the function of the intestinal epithelial barrier and to modulate host immune responses [7,8].

Antimicrobial Activity Assay
The antimicrobial activity of the LAB strains against L. monocytogenes was assessed with two methods: (i) the spot-on-lawn assay and (ii) the agar-well diffusion assay, according to Toure et al. [31]. In brief, for the spot-on-lawn assay, an aliquot of 3 µL of active MRS culture for each LAB strain was spotted on 45 mL MRS with 1.2% agar (Agar No2 Bacteriological, LabM) in a Petri dish (150 mm) and incubated anaerobically at 37 • C for 24 h. Following incubation, the Petri dishes were overlaid with 20 mL brain heart infusion (BHI; LabM) broth with 0.7% agar pre-warmed at 45 • C and inoculated with 0.3 mL of an overnight culture of L. monocytogenes. The Petri dishes were incubated aerobically for 24 h.
For the agar-well diffusion assay, activated LAB cultures were centrifuged at 3600 rpm for 10 min at 4 • C (Eppendorf Refrigerated Centrifuge, Hamburg, Germany), and the cell-free supernatant (CFS) of each LAB strain was adjusted to pH 6.5 with NaOH (4 M) and filtered through a 0.22 µm pore-size filter (Millex-GP Filter Unit, Merck Millipore Ltd., Cork, Ireland). Petri dishes (150 mm) that contained 20 mL solidified 1.2% agar were overlaid with 45 mL of BHI broth containing 0.7% agar and inoculated with 0.7 mL of an overnight L. monocytogenes culture. The plates were allowed to solidify for 1 h at ambient temperature (approx. 22 • C) and for 1 h at 4 • C. Wells of 5 mm diameter were cut in the solidified agar using a sterile metal-cork borer and filled with 40 µL of the CFS. The Petri dishes were incubated for 2 h at ambient temperature and for 16 h at 37 • C. For both methods, the presence of inhibitory zones around the spots or the agar wells were considered antimicrobial activity against L. monocytogenes. The antimicrobial assays were performed in duplicates.

Adhesion Assay
Ten lactobacilli strains with strong adhesive abilities to Caco-2 cells [26] were selected for the competition assays against the L. monocytogenes EGDe strain. The adhesion assays were performed according to Moroni et al. [24]. Briefly, single overnight cultures of L. monocytogenes and lactobacilli were harvested by centrifugation at 3600 rpm for 10 min at 4 • C and washed twice with 1 4 -strength Ringer's solution (LabM). The washed bacterial pellets were resuspended in DMEM with 0.1% FBS at a final concentration of~10 6 CFU/mL for L. monocytogenes EGDe and 10 6 -10 7 CFU/mL for lactobacilli. Caco-2 cell monolayers developed in 24-well plates were inoculated with 500 µL of a single bacterial culture for the individual adhesion assays or with 250 µL of L. monocytogenes EGDe and 250 µL of each Lactobacillus strain for the inhibition adhesion assays. For the inhibition assays, three different treatments were followed: (i) bacterial cultures were added to Caco-2 monolayers simultaneously (competition); (ii) Lactobacillus spp. cells were added to Caco-2 monolay-ers 1 h before L. monocytogenes EGDe (exclusion); and (iii) L. monocytogenes EGDe cells were added to Caco-2 monolayers 1 h before Lactobacillus spp. (displacement). The plates were incubated for 1 h at 37 • C and 5% CO 2 (CO2 Incubator CO2CELL 170 STD, Munich, Germany). Following incubation, each well was washed carefully twice with 500 µL phosphate buffered saline (PBS; TaKaRa, Kusatsu, Japan), and the adhered bacterial cells were harvested with 1 mL trypsin (Trypsin-EDTA 10X; Biosera Europe, Nuaillé, France). The L. monocytogenes EGDe cells were enumerated on Listeria Isolation Medium (Oxford Formulation) plates (LabM) after incubation for 48 h at 37 • C in aerobic conditions, and the lactobacilli were enumerated on MRS agar plates after 72 h at 37 • C under anaerobic conditions. The adhesion capacity of each bacterial strain was estimated as the number of adherent cells divided by the total cells added multiplied by 100. The inhibition of L. monocytogenes adhesion was calculated as follows: inhibition rate = 100(1 − T Lb /T), where T Lb and T are the numbers of adherent L. monocytogenes cells (CFU/well) in the presence and absence of lactobacilli, respectively. The experiments were performed in three biological and two technical replicates.

Invasion Assay
The inhibitory effect of lactobacilli against Caco-2 invasion by L. monocytogenes EGDe was estimated using the gentamicin-based assay according to Moroni et al. [24] and Zilelidou et al. [21]. Briefly, Caco-2 cell monolayers in 24-well plates were inoculated with L. monocytogenes EGDe or lactobacilli in single or dual cultures, as described above for the adhesion assays. The plates were incubated for 1 h at 37 • C and 5% CO 2 . The monolayers were washed twice with PBS to remove the non-adherent bacteria and then incubated in 0.5 mL/well DMEM with 0.1% FBS and 150 µg/mL gentamicin (Gentamicin sulfate, 10 mg/mL, Biosera Europe) for 45 min to kill the non-invaded L. monocytogenes EGDe cells. The Caco-2 monolayers were washed twice with 500 µL PBS and lysed with 1 mL ice-cold 0.1% (v/v) Triton™ X-100 (Fischer Scientific, Geel, Belgium). The population of the invaded L. monocytogenes EGDe cells were enumerated on TSA-YE after incubation for 48 h at 37 • C in aerobic conditions. The inhibition of invasion was calculated as follows: inhibition rate = 100(1 − T Lb /T), where T Lb and T are the numbers of invaded L. monocytogenes cells (CFU/well) in the presence and absence of lactobacilli, respectively. The experiments were performed in three biological and two technical replicates.

Statistical Analysis
The statistical analysis was performed using the SPSS Statistics for Windows version 16.0 (SPSS Inc., Chicago, IL, USA). For the pairwise comparison of the adhered bacterial populations on the Caco-2 cells at 1 h and 2 h, the Student's t-test was used. For the comparison of the inhibition rates among the different strains, Tukey's honestly significant differences (HSD) test was performed. The significance level was set at a p-value < 0.05.

Antimicrobial Activity
The results of the antimicrobial assays are shown in Table 3. The spot-on-lawn assay showed that 21 out of 45 LAB strains were able to inhibit the growth of all ten L. monocytogenes strains by presenting an inhibitory zone around the lactobacilli spots. The most pronounced antilisterial effect was displayed by L. rhamnosus strains (LR-B19, LR-10, LR-B5, LR-B20, LR-A3, LA-A20, LR-C44, LR-52), L. paragasseri (LA-B17, LDD-C1), L. crispatus LC-C1 and the strain L. gasseri LG-C45. However, when the pH-neutralized CFS of the LAB strains were used in the agar-well diffusion assay, no growth inhibition of L. monocytogenes was observed.

Adhesion of Single Bacterial Strains on Caco-2 Cells
Following 1 h of contact of the single bacterial isolates with the Caco-2 cells, the adhesion efficiency of the lactobacilli ranged from 0.6% to 21% ( Figure 1A). The best adhesion efficiency was observed for L. paragasseri LDD-C1 (21%) and L. crispatus LCR-A21 (19.3%), which were significantly higher compared to the reference strain L. rhamnosus GG (LGG, p < 0.05). L. monocytogenes EGDe adhered to the Caco-2 cells at a level of 1.1%, significantly lower compared to the aforementioned Lactobacillus strains (p < 0.05). Following 2 h of Caco-2 contamination, the numbers of adhered bacterial cells increased significantly (p < 0.05) and ranged from 0.8% to 77.5% ( Figure 1B). Similar to 1 h, the best adhesion efficiency was observed for L. paragasseri LDD-C1 (51.1%) and L. crispatus LCR-A21 (77.5%). The strains L. acidophilus DSM20079 and LGG had no significant increase in adherence to Caco-2 during the 2 h of incubation. L. monocytogenes EGDe increased from 1.1% to 5.9% (p < 0.05).
When the Caco-2 cells were pre-incubated for 1 h with Lactobaciillus spp. is fore the addition of L. monocytogenes EGDe (exclusion), the inhibition rates ran 13.3% to 49.7% (Figure 3). The isolates L. gasseri LG-7528 (49.7%) and L. rhamnosu (44.7%) exhibited the highest rates of pathogen exclusion. In the displacement assay, where lactobacilli were added to the Caco-2 cells contamination with L. monocytogenes EGDe, the adherence was limited by 13.4% (Figure 4). The greatest inhibition of L. monocytogenes adhesion (38.0%) was disp the isolates L. gasseri LG-7528 and L. pentosus LP-A22. In the displacement assay, where lactobacilli were added to the Caco-2 cells 1 h after contamination with L. monocytogenes EGDe, the adherence was limited by 13.4% to 38.0% (Figure 4). The greatest inhibition of L. monocytogenes adhesion (38.0%) was displayed by the isolates L. gasseri LG-7528 and L. pentosus LP-A22.

Inhibition of Invasion
The inhibition of invasion of the Caco-2 cells by L. monocytogenes EGDe v pending on the Lactobacillus spp. strain and the treatment method. In the compe say, where L. monocytogenes EGDe and Lactobacillus spp. were added to the Ca simultaneously, the invasion was reduced by 33.5% to 63.1% ( Figure 5). The high bition of invasion was recorded by the two reference strains, L. acidophilus D (61.8%) and LGG (63.1%). Among the strains isolated from the neonatal feces, LG-7528 resulted in a decrease in invasion by 58.8%, L. crispatus LCR-A21 by paracasei subsp. tolerans LPP-A16 by 49.2% and L. rhamnosus LR-B5 by 47.7%.

Inhibition of Invasion
The inhibition of invasion of the Caco-2 cells by L. monocytogenes EGDe varied depending on the Lactobacillus spp. strain and the treatment method. In the competition assay, where L. monocytogenes EGDe and Lactobacillus spp. were added to the Caco-2 cells simultaneously, the invasion was reduced by 33.5% to 63.1% ( Figure 5). The highest inhibition of invasion was recorded by the two reference strains, L. acidophilus DSM20079 (61.8%) and LGG (63.1%). Among the strains isolated from the neonatal feces, L. gasseri LG-7528 resulted in a decrease in invasion by 58.8%, L. crispatus LCR-A21 by 51.7%, L. paracasei subsp. tolerans LPP-A16 by 49.2% and L. rhamnosus LR-B5 by 47.7%. bition of invasion was recorded by the two reference strains, L. acidoph (61.8%) and LGG (63.1%). Among the strains isolated from the neonatal LG-7528 resulted in a decrease in invasion by 58.8%, L. crispatus LCR-A paracasei subsp. tolerans LPP-A16 by 49.2% and L. rhamnosus LR-B5 by 47.7 Pretreatment of the Caco-2 cells with lactobacilli for 1 h (exclusion as an inhibition of invasion at levels that ranged from 26.8% to 52.6%. The gasseri LDD-C1, L. acidophilus DSM20079 and L. rhamnosus LR-B20 were th in inhibiting L. monocytogenes EGDe to invade the Caco-2 cells at levels 5 46.6%, respectively ( Figure 6). Pretreatment of the Caco-2 cells with lactobacilli for 1 h (exclusion assay) resulted in an inhibition of invasion at levels that ranged from 26.8% to 52.6%. The isolates L. paragasseri LDD-C1, L. acidophilus DSM20079 and L. rhamnosus LR-B20 were the most efficient in inhibiting L. monocytogenes EGDe to invade the Caco-2 cells at levels 52.6%, 47% and 46.6%, respectively ( Figure 6).  During the exclusion assay, where Lactobacillus spp. were ad L. monocytogenes EGDe with Caco-2, the inhibition ranged from 2 The highest displacement rates were observed for L. paracasei s (45.6%) and L. pentosus LP-A22 (40.6%). In the displacement assa monocytogenes EGDe was increased by 4.5% with the addition of L During the exclusion assay, where Lactobacillus spp. were added 1 h after contact of L. monocytogenes EGDe with Caco-2, the inhibition ranged from 22.3% to 45.6% (Figure 7). The highest displacement rates were observed for L. paracasei subsp. tolerans LPP-A16 (45.6%) and L. pentosus LP-A22 (40.6%). In the displacement assay, Caco-2 invasion by L. monocytogenes EGDe was increased by 4.5% with the addition of LGG.
(SEM) of three biological and two technical replicates.

Discussion
Lactic acid bacteria predominate the GI tract of healthy humans and confer numerous beneficial health effects, also providing protection against pathogens. In this study, we aimed to investigate lactobacilli strains previously isolated from the feces of healthy neonates for their inhibitory effect against the foodborne pathogen L. monocytogenes.
The antimicrobial activity against L. monocytogenes growth was estimated with two methods: the spot-on-lawn and the agar-well diffusion assays. In the latter, pH-neutralized cell-free supernatants were used to estimate the antimicrobial effect of lactobacilli metabolic substances. Antilisterial activity was observed with the spot-on-lawn method in contrast to the agar-well diffusion assay, where no CFS was able to inhibit L. monocytogenes growth. These findings are in accordance with other studies where pH-neutralized CFS presented limited antimicrobial activity compared to acid CFS [32]. Numerous studies have investigated the antimicrobial activity and effectiveness of lactic acid bacteria, and specifically lactobacilli, against food-borne pathogens [33][34][35]. The source of isolation, bacterial species and the methods used are factors that affect the estimation of

Discussion
Lactic acid bacteria predominate the GI tract of healthy humans and confer numerous beneficial health effects, also providing protection against pathogens. In this study, we aimed to investigate lactobacilli strains previously isolated from the feces of healthy neonates for their inhibitory effect against the foodborne pathogen L. monocytogenes.
The antimicrobial activity against L. monocytogenes growth was estimated with two methods: the spot-on-lawn and the agar-well diffusion assays. In the latter, pH-neutralized cell-free supernatants were used to estimate the antimicrobial effect of lactobacilli metabolic substances. Antilisterial activity was observed with the spot-on-lawn method in contrast to the agar-well diffusion assay, where no CFS was able to inhibit L. monocytogenes growth. These findings are in accordance with other studies where pH-neutralized CFS presented limited antimicrobial activity compared to acid CFS [32]. Numerous studies have investigated the antimicrobial activity and effectiveness of lactic acid bacteria, and specifically lactobacilli, against food-borne pathogens [33][34][35]. The source of isolation, bacterial species and the methods used are factors that affect the estimation of antimicrobial activity of lactobacilli. In this study, antilisterial activity was mainly exhibited by the strains of the species L. rhamnosus. Syrokou et al. [36] investigated the antimicrobial activity of 207 LAB isolates and found that 23 strains were active against L. monocytogenes serotype 4b; however these strains belonged to the L. plantarum species, which was not included in our study, and were isolated from sourdough. Limited or no antimicrobial activity against L. monocytogenes was also observed for L. paracasei and L. rhamnosus, isolated from infant fecal samples, and L. acidophilus, isolated from pickled cabbage, examined with the agar-well diffusion method [37].
The ability of lactobacilli to adhere to epithelial cells is an important aspect of the probiotic that may confer a competitive advantage over pathogens for adhesion and invasion sites on epithelial cells. Among the ten Lactobacillus sp. strains tested for adherence to Caco-2 cells, the greatest adhesiveness was observed by L. paragasseri LDD-C1 and L. crispatus LCR-A21. The adhesion efficiency for these strains was greater than the reference strain LGG, which is considered highly adhesive [16,[38][39][40]. The lower adhesive capacity of LGG compared to other commercial and potential probiotic bacteria has been previously reported [41]. L. paragasseri LDD-C1 and L. crispatus LCR-A21 were more adherent than L. monocytogenes EGDe. After 2 h of contact with the Caco-2 cells, the percentages of adhered bacteria increased significantly compared to 1 h for all the strains except L. paragasseri LDD-C1 and the two reference strains L. acidophilus DSM20079 and L. rhamnosus GG. The highest adhesion efficiency was achieved by L. paragasseri LDD-C1 and L. crispatus LCR-A21, while other lactobacilli and L. monocytogenes EGDe remained at low adherence levels. This is indicative of the good adhesive properties of the specific isolates. In a previous study, it was shown that incubation for 2 h and 4 h resulted in higher levels of adherence for Salmonella enterica but not for L. paracasei, showing that a longer contact time does not result in higher adhesion rates for all bacteria [15]. They also showed that the strongest inhibition of adhesion was observed for the shortest time of contact (2 h vs. 4 h).
The presence of lactobacilli reduced the adhesive ability of L. monocytogenes EGDe to the Caco-2 cells in a strain and treatment-dependent manner. The inhibition varied among the Lactobacillus strains, and there were isolates with greater blocking effect against pathogens compared to LGG. The results revealed that adhesion efficiency and inhibition of adhesion were two distinct events. For instance, L. gasseri LG-7528, a strain with low adhesion efficiency, exhibited the highest inhibitory capacity against L. monocytogenes EGDe in the competition and exclusion assays in contrast to the highly adhesive strains L. paragasseri LDD-C1, L. crispatus LCR-A21 and L. paracasei subsp. tolerans LPP-A16. Similar results were observed previously, where the low-adhesive lactobacilli strains were the most effective against L. monocytogenes [25,42].
Regarding the impact of treatment on the inhibition efficiency of lactobacilli, in the present study, inhibition was greater when the antagonistic bacteria were added to the Caco-2 cells simultaneously. The least inhibition rate was recorded in the displacement assay, where the lactobacilli were added to 1 h pre-contaminated Caco-2 cells with the pathogen. Previously, it has been suggested that the mechanisms of competition and exclusion are similar to each other and differ from displacement [25,43]. The displacement of GI bacteria including pathogens is probably a slow process with many of them needing 2 h to achieve increased degrees of displacement [43]. Pre-treatment of epithelial cells with probiotics was shown to result in increased inhibition of L. monocytogenes adherence, probably attributed to a mechanism related to co-aggregation of probiotic and pathogen cells [16]. Further studies are needed to elucidate these phenomena.
L. monocytogenes is an invasive pathogen and, once it is adhered to the eukaryotic cell surface, penetrates into the host cells beginning its intracellular lifecycle [44]. Probiotics that are able to block this internalization will be promising for a protective effect for human health. The presence of lactobacilli was able to reduce Caco-2 invasion by L. monocytogenes EGDe. The levels of inhibition varied among the different strains and treatments. Similar to the adherence process, competition resulted in a greater inhibition of invasion than exclusion and displacement. During competition, the reference strains L. acidophilus DSM20079 and L. rhamnosus GG were the most effective, resulting in higher inhibition rates than the other lactobacilli strains, followed by L. gasseri LG-7528. Although not highly adhesive, these isolates could effectively block Caco-2 invasion by the pathogen, thus demonstrating distinct mechanisms underlying these two processes. Moreover, the lactobacilli isolates exhibited different inhibitory effects in adhesion and invasion, which is in accordance with the results of Moroni et al. [24] who similarly reported different patterns by bifidobacterial strains against L. monocytogenes. Interestingly, the addition of LGG to the Caco-2 cells that were pre-contaminated with L. monocytogenes EGDe for 1 h increased the invasion of the pathogen in the epithelial cells. Similar results were observed in other studies in the process of adherence, where lactobacilli strains resulted in an increase in the adherence of pathogenic bacteria, such as Clostridium difficile, Escherichia coli, L. monocytogenes and Salmonella Typhimurium [16,25]. This could be of great concern; therefore, case-by-case studies on the interaction of potential probiotics with pathogenic bacteria should be conducted.

Conclusions
In conclusion, lactobacilli strains with probiotic properties isolated from the fecal samples of healthy neonates were able to inhibit L. monocytogenes adherence to and invasion in Caco-2 cells at variable levels. The inhibitory effect was strain and treatment dependent with competition resulting in greater inhibition compared to the exclusion and displacement assays. For each strain, the adhesive ability and inhibition of adhesion were distinct events, as well as the inhibition of adhesion and invasion. The results are indicative of the strainspecific properties of the lactobacilli; however, it is evident that some of these strains, including L. gasseri LG-7528, L. paragasseri LDD-C1, L. crispatus LCR-A21 and L. paracasei subsp. tolerans LPP-A16, could be further investigated for their potential probiotic use and their antimicrobial activity against other pathogens at growth, adherence and invasion in epithelial cells levels.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author. The data are not publicly available due to privacy restrictions.