Antagonistic Activity of Lactobacillus reuteri Strains on the Adhesion Characteristics of Selected Pathogens

Adhesion ability of probiotics is the key factor that decides their colonization in the gastrointestinal tract and potential to inhibit pathogens. Therefore, adhesion ability can be considered as a key determinant for probiotic efficacy. Presents study documents the antagonistic activity of viable/untreated, Lithium chloride (LiCl) treated or heat-killed forms of eight probiotic Lactobacillus reuteri strains on the adhesion characteristics of selected pathogens. All strains investigated were able to adhere to Caco-2 cells. L. reuteri strains tested were able to inhibit and displace (P < 0.05) the adhesion of Escherichia coli ATCC25922, Salmonella typhi NCDC113, Listeria monocytogenes ATCC53135, and Enterococcus faecalis NCDC115. The probiotic strain L. reuteri LR6 showed the strongest adhesion and pathogen inhibition ability among the eight L. reuteri strains tested. In addition, the abilities to inhibit and to displace adhered pathogens depended on both the probiotic and the pathogen strains tested suggesting the involvement of various mechanisms. The adhesion and antagonistic potential of the probiotic strains were significantly decreased upon exposure to 5 M LiCl, showing that surface molecules, proteinaceous in nature, are involved. The heat-killed forms of the probiotic L. reuteri strains also inhibited the attachment of selected pathogens to Caco-2 cells. In conclusion, in vitro assays showed that L. reuteri strains, as viable or heat-killed forms, are adherent to Caco-2 cells and are highly antagonistic to pathogens tested in which surface associated proteins play an important role.


INTRODUCTION
Globally, the market of probiotics is growing faster as they have claimed to exert several health promoting effects, including interaction with the immune system, production of antimicrobial substances, enhancement of the mucosal barrier function and competition with enteropathogens for adhesion sites (Boesten and de Vos, 2008;Papadimitriou et al., 2015). There are numerous probiotic genera and species including lactobacilli and bifidobacteria which have been implicated in a number of health promoting functions that affect general health and well-being of the host. Adhesion is considered as a potential biomarker for selection of potential probiotics; as their colonization with extended transit time is extremely crucial for optimal expression of their general as well as specific physiological functions (Duary et al., 2011). Several reports have given special attention to the protective role of probiotics against enteropathogens and the underlying mechanisms (Salminen et al., , 1999Ouwehand et al., 2002;Ouwehand and Salminen, 2003;Collado et al., 2007). Some of these possible protective mechanisms include competition for nutrients and adhesion sites  or immune modulation (Schiffrin et al., 1997;Salminen et al., 1998). Thus, the probiotics intervention may provide significant protection against gastrointestinal infection and this would enhance human health.
Lactobacilli have been shown to possess surface adhesins similar to those on bacterial pathogens (Neeser et al., 2000). Several surface-located molecules such as lipoteichoic acid, lectin-like molecules and proteins have been identified as adhesins which interact with their specific receptors displayed on the host cell surface (Martinez et al., 2000;Beganović, 2008;Beganović et al., 2010). Due the importance of probiotics in the prevention of infections, the aim of this study was to assess the antagonistic properties of probiotic strains derived from breast fed human infant feces (Singh et al., 2012). Earlier, we reported that the cell surface proteins play an important role in probiotic activities of the Lactobacillus reuteri strains (Singh et al., 2016). In the present study, the probiotic L. reuteri strains were evaluated for their adhesion and abilities to exclude, displace and compete with selected pathogens using Caco-2 as an experimental model. These experiments were also conducted with LiCl treated and heat-killed forms of L. reuteri strains to check their functional interest and the importance of probiotic cell surface integrity.

Bacterial Strains and Culture Conditions
Eight L. reuteri strains viz., LR5, LR6, LR9, LR11, LR19, LR20, LR26, and LR34, of fecal origin were selected for this study. The Lactobacillus strains were grown in MRS broth (deMan, Ragosa and Sharp broth; Himedia, Mumbai, India) at 37 • C for 18-24 h and maintained as glycerol stocks until further use. From the stock cultures, working cultures were prepared and were propagated twice prior to use by sub-culture in MRS broth. The bacterial pathogens used in this study were Escherichia coli ATCC25922, Salmonella typhi NCDC113, Listeria monocytogenes ATCC53135, Enterococcus faecalis NCDC115 which were maintained in BHI (Brain Heart Infusion) broth.

Live Cells
The Lactobacillus strains were grown overnight (16-18 h) in MRS broth at 37 • C and harvested at 5000 g for 10 min. The cells were washed twice in phosphate buffer saline (PBS) and OD 600 was adjusted to 1.5 which corresponds to 10 9 cfu/ml based on calibration curve performed.

LiCl Treatment
The Lactobacillus strains were harvested by centrifugation at 5000 g for 10 min, washed with sterile distilled water and then resuspended in 5 M LiCl for 30 min (Zhang et al., 2013). After LiCl treatment, the cells were washed twice in PBS and OD 600 was adjusted to 1.5 which corresponds to 10 9 cfu/ml.

Heat Killed Cells
The Lactobacillus strains were grown overnight (16-18 h) in MRS at 37 • C and harvested by centrifugation at 5000 g for 10 min. Then, the cells were washed twice with PBS and OD 600 was adjusted to 1.5 which corresponds to 10 9 cfu/ml. The bacterial suspension was heat killed at 80 • C for 10 min in a water bath and stored at −70 • C until further use (Ouwehand et al., 2000).

Caco-2 Cell Culture and Experiment Design
Caco-2 Cell Culture The Caco-2 cell line was procured from the National Center of Cell Science, Pune, India. Cells were routinely grown in Dulbecco's modified eagle's minimal essential medium (DMEM; Sigma, USA), supplemented with 10% fetal bovine serum (FBS; Sigma, USA), 100 µg streptomycin per ml (Sigma, USA) and 100 U penicillin per ml (Sigma, USA) at 37 • C in a 5% CO 2 atmosphere. For adhesion and inhibition assays, Caco-2 monolayers were prepared in 6-well tissue culture plates. Cells were inoculated at a concentration of 7 × 10 4 cells per well to obtain confluence and allowed to differentiate. The culture medium was changed on alternate days, and the last two media changes were without antibiotics.

In vitro Adherence Assay
A 1.0 ml aliquot of the bacterial suspension (viable, heat killed, and LiCl treated lactobacilli; 10 9 cells) was added to confluent Caco-2 monolayer and incubated for 2 h in a 5% CO 2 atmosphere. Following incubation, the Caco-2 monolayers were washed with sterile PBS (pH 7.4), Giemsa-stained and examined microscopically under oil immersion, as described previously by Duary et al. (2011).

Inhibition of Pathogen Adherence to Caco-2 Cells
The inhibition ability of viable/untreated, LiCl treated or heatkilled forms of L. reuteri strains against pathogens adherence was performed according to procedure described by Zhang et al. (2013) with some modifications. Three different protocols were followed to evaluate the ability of L. reuteri strains (viable, heat inactivated, and LiCl treated) to inhibit pathogen (E. coli ATCC25922, L. monocytogenes ATCC5313, S. typhi NCDC113, and E. faecalis NCDC115) adhesion to Caco-2 cells.
For competition assays, Lactobacillus (live, heat killed, and LiCl treated; approximately 10 8 -10 9 cfu/ml) and pathogens (approximately 10 7 cfu) were co-incubated with Caco-2 monolayer for 2 h. For exclusion assays, Lactobacillus (live, heat killed, and LiCl treated; approximately 10 8 -10 9 cfu/ml) was cultured with Caco-2 monolayer for 1 h. Following 1 h incubation, Caco-2 monolayer was washed three times with PBS (pH 7.4); pathogens (approximately 10 7 cfu) were added and incubated for another 1 h. For displacement assays, pathogens (approximately 10 7 cfu) were cultured with Caco-2 monolayer for 1 h, and then the Lactobacillus (live, heat killed, and LiCl treated) were added and cultured for another 1 h. The monospecies cultures of pathogenic bacteria were used as the controls.
In all the above treatments, non-adhered bacterial cells were removed by washing with PBS (pH 7.4). After washing, the Caco-2 cells were detached by addition of 0.25% (v/v) Trypsin-EDTA solution at 37 • C for 5 min and the number of viable adhering E. coli, L. monocytogenes, S. typhi, and E. feacalis were determined by plating on EMB, PALCAM, XLD, and CA agar plates after serial dilutions, respectively.

Statistical Analysis
The results for adhesion and pathogen inhibition are expressed as the mean ± SD of three independent experiments. Statistical analysis was done by StatGraphicPlus software. Data were subjected to a one-way analysis of variance (ANOVA) followed by a Tukey's post hoc test. Differences were considered statistically significant when P < 0.05.

Adhesion Assay
All the L. reuteri strains adhered to Caco-2 cells albeit at different levels. However, on comparative evaluation, L. reuteri strains LR6, LR20, and LR34 were found to be the most adhesive strains based on their respective adhesion scores, with LR6 being the most adhesive strain among all the strains tested. The adhesion score for other strains tested, i.e., LR5, LR9, LR11, LR19, and LR26 differed significantly. All the L. reuteri test strains were found to be highly adhesive (>100 bacteria/20 microscopic fields) when assessed in Caco-2 cell lines. In comparison, it was also observed that the heat inactivation and LiCl treatment had a marked effect on the adhesion ability of the strains as the adhesion of the strains was significantly (P < 0.05) reduced as shown in Figure 1.

Competition Assay
Competition assay explained the ability of probiotic strains to compete with pathogens for the adhesion site on epithelial cells. Among the L. reuteri strains tested; LR6, LR9, and LR11 exhibited the maximum inhibition of E. coli ATCC25922. For S. typhi NCDC113, the strains mainly L. reuteri LR6, LR9, LR11, LR19, and LR26 showed the maximum inhibition. Among the tested probiotic L. reuteri strains; LR5, LR6, LR9, LR20, and LR34 inhibited the adhesion of L. monocytogenes ATCC53135 to Caco-2 cells to significant (P < 0.05) levels. For E. feacalis NCDC115, the significant inhibition of adhesion to Caco-2 cells was seen for the strain LR6 and LR9. From the results of competitive assay shown in Table 1, we can conclude that the strain LR6 is the most competitive probiotic strain which can compete strongly with the selected pathogens for the adhesion to epithelial cells.
In competition inhibition assay, it was observed that heat inactivation decreases the ability of L. reuteri strains to compete with selected pathogens for adhesion to Caco-2 cells as compared with their untreated/live forms. The inhibition ability of the heat inactivated forms of L. reuteri strains showed the variability in results ranging from 6.7 ± 1.13% to 37.6 ± 1.07% for E. coli ATCC25922, 18.9 ± 1.32% to 57.4 ± 2.32% for S. typhi NCDC113, 8.3 ± 0.61% to 20.4 ± 1.17% for L. monocytogenes ATCC53135, and 8.9 ± 1.27% to 25.8 ± 1.46% for E. feacalis NCDC115 as shown in Table 1. For heat inactivated forms, the strains LR6, LR9, and LR11 showed the maximum inhibition to adhesion of S. typhi NCDC113. In case of E. coli ATCC25922, the maximum inhibition was exhibited by heat inactivated forms of LR6, LR9, LR11, and LR19, respectively. The strain LR6 also showed the highest inhibition of L. monocytogenes ATCC53135 FIGURE 1 | Adhesion of differently treated Lactobacillus reuteri strains to Caco-2 cells. * Significantly different (P < 0.05) from the untreated control.
Frontiers in Microbiology | www.frontiersin.org and E. feacalis NCDC115 adhesion to Caco-2 cells. From the results given in Table 1, it is also evident that the heat inactivated forms of L. reuteri strains were able to compete with pathogens as the results on comparison were found statistically insignificant to their viable forms. However, the ability of the L. reuteri strains to compete with pathogens assayed for adhesion site on Caco-2 decreases significantly (P < 0.05) on LiCl treatment (meant for removal of surface proteins) as depicted in Table 1.

Displacement Assay
Displacement assay exhibits the potential of the probiotic strains to remove/displace the already adhered pathogen from the epithelial cells. The data depicted that amongst the L. reuteri strains tested, the strain LR6 showed maximum inhibition of E. coli ATCC25922, S. typhi NCDC113, L. monocytogenes ATCC53135, and E. feacalis NCDC115. After 5 M LiCl treatment, the displacement ability of L. reuteri strains against test pathogens were significantly reduced (P < 0.05), shown in Table 2.
The heat inactivated forms of L. reuteri strains showed reduced ability to displace the tested pathogens as compared to their untreated/viable forms. The heat inactivated forms of L. reuteri strains showed the variability in results for percentage displacement ranging from 6.5 ± 1.16% to 40.5 ± 0.99% for E. coli ATCC25922, 10 ± 0.16% to 50.5 ± 2.52% S. typhi NCDC113, 11.5 ± 1.03% to 40.5 ± 1.03% L. monocytogenes ATCC53135, and 12 ± 0.93% to 35.5 ± 0.89% for E. feacalis NCDC115 as shown in Table 2. For heat inactivated forms, the strains LR6 and LR9 showed the maximum displacement for S. typhi NCDC113. In case of E. coli ATCC25922, the maximum displacement was exhibited by heat inactivated forms of LR6. The strain LR6 also showed the highest inhibition of L. monocytogenes ATCC53135 and E. feacalis NCDC115 to Caco-2 cells.

Exclusion Assay
Exclusion assay explains that once the adhesion site is occupied by the probiotic bacteria it becomes unavailable for pathogen. It is evident from the results that the tested strains LR5, LR6, LR9, LR20, and LR26 were able to exclude E. coli ATCC25922 adhesion to significant levels. The significant reduction in E. coli ATCC25922 adhesion to Caco-2 cells was observed for LR5, LR6, LR9, LR20, and LR26. In case of S. typhi NCDC113, the maximum exclusion was showed by strains LR5, LR6, LR9, LR19, and LR26. On the other hand, only LR6 showed the maximum exclusion of L. monocytogenes ATCC53135 from adhesion to caco-2 cells. Similarly, LR6 and LR11 were the only strains which were able to exclude the E. feacalis NCDC115 to significant levels. The data is depicted in Table 3.

DISCUSSION
Probiotics efficacy is highly dependent on their survival and persistence in gastrointestinal tracts. Therefore, adhesion ability can be considered as a standard biomarker for selecting a potential probiotic (Duary et al., 2011). In the present investigation, we evaluated eight probiotic strains of L. reuteri, previously isolated from breast fed infant feces (Singh et al., 2012), for their potential to adhere Caco-2 cells. The results pertaining to adhesion were recorded in terms of number of bacteria adhering to Caco-2 cell line. On comparative evaluation based on adhesion score, L. reuteri strains LR6, LR20, and LR34 were found to be the most adhesive strains. Adhesion score for all the L. reuteri strains were more than 100 and, therefore, can be regarded as a strongly adhesive to Caco-2 cell lines as per the classification by Jacobsen et al. (1999). Also, the variation observed in adhesion abilities of L. reuteri strains suggests that the trait varies among probiotic strains. This is in complete agreement with the other researchers who also reported that the probiotics ability to adhere is very much strain, species and genus specific (Collado et al., 2007). Several studies have reported that probiotics compete with pathogens for the adhesion sites, as both probiotics and pathogens possess similar kind of adhesins on their surfaces. Also, the inhibition specifically depends on the probiotic strains and pathogens used as well as the methods of assessment (Chen et al., 2007;Gueimonde et al., 2007). In this study, the probiotic L. reuteri strains were evaluated for their abilities to exclude, compete and displace selected pathogens using Caco-2 as an experimental model. The pathogen adhesion inhibition by probiotic L. reuteri strains showed a high variability and indicated that it was clearly a strain dependent property. The L. reuteri strains tested did not show the same level of inhibition capacity against the pathogens, but they efficiently inhibited the adhesion of pathogenic bacteria to Caco-2 cell in all three assays. The strain LR6 with highest adhesion ability generally showed much higher inhibition of pathogen adhesion to Caco-2 cells, indicating that the pathogen inhibition capacity of probiotic strains may be related to their adhesion ability. Similarly, other workers have also reported the competitive exclusion of enteropathogens by bifidobacteria and lactobacilli (Bernet et al., 1993;Forestier et al., 2001;Lee et al., 2003;Collado et al., 2005Collado et al., , 2007Weizman et al., 2005;Pham et al., 2009;Wine et al., 2009;Zhang et al., 2013). Meanwhile, the pathogen inhibition ability of the L. reuteri strains did not correlate with the adhesive ability of the strains. In our results, the profile of competition, exclusion and displacement of pathogens by L. reuteri strains were different confirming that the mechanisms of competition, exclusion and displacement might be different. Therefore, it was believed that the differences in competitive exclusion between the strains correlate with the variations in their adhesion ability, possibly due to differences in their surface characteristics. This suggests that the mechanism involved in inhibition is complicated and many factors may be involved.
Generally, adhesion involves the interaction between bacterial associated molecular patterns such as; lipoteichoic acid (Granato et al., 1999), surface layer protein (Chen et al., 2007;Johnson-Henry et al., 2007), peptidoglycan (Van Tassell and Miller, 2011) and their pattern recognition receptors on the host epithelial cells. The surface associated proteinaceous components mediating bacterial adhesion to intestinal epithelial cells have been demonstrated for many Lactobacillus species (Rojas et al., 2002;Roos and Jonsson, 2002;Frece et al., 2005;Chen et al., 2007). In the present study, a significant (P < 0.05) difference was observed on comparing the adhesion ability of untreated and LiCl treated forms of L. reuteri strains, suggesting the importance of the surface associated proteins in adhesion. Also, the ability of the L. reuteri strains to displace, compete and exclude the pathogens from adhesion to caco-2 cells was significantly (P < 0.05) decreased on LiCl treatment. The results are in complete agreement with other workers who reported reduction in binding and adhesion ability of lactobacilli on removal/disruption of surface associated proteins (Sillanpää et al., 2000;Buck et al., 2005;Frece et al., 2005;Chen et al., 2007;Johnson-Henry et al., 2007;Wang et al., 2008;Li et al., 2011;Zhang et al., 2013).
By definition probiotics should be viable in order to exert health benefits. Many researchers have suggested that certain probiotic effects can also be obtained with non-viable probiotics (Ouwehand and Salminen, 1998). Evidences also suggested that non-viable probiotics are less effective which may be attributed to their reduced binding ability than viable probiotics (Conge et al., 1980;De Simone et al., 1987;Kato et al., 1994;Kaila et al., 1995;Perdigon et al., 1995). In this study, a significant reduction in the adhesion and pathogen inhibition abilities of the probiotic L. reuteri strains was observed in heat inactivated forms compared to their viable forms. This suggests that heat treatment inactivate the micro-organisms and also alters their physicochemical properties (El-Nezami et al., 1998). The reduction of adhesion and pathogen inhibition can be explained by the heat sensitive proteinaceous nature of the molecules involved. In contrast, Tareb et al. (2013) reported that heatkilled forms of both Lb. rhamnosus 3698 and Lb. farciminis 3699 exhibited higher adhesion and higher pathogen exclusion potential.
Probiotics intervention is more cost effective and natural approach to preserve intestinal homeostasis and restore the pathogenesis related dysbiosis than antibiotics. The results of this study demonstrate that probiotic strains of L. reuteri tested can exclude, displace and compete with enteropathogens. However, it is important to take into account that these processes studied are highly specific to probiotic and pathogenic strains. This study indicates that strong adhesion ability means greater inhibition activity for probiotic lactobacilli against pathogen, in which surface associated proteins play an important role which further need to be identified and studied. This study also supports the need for further investigations to demonstrate the potential benefits of L. reuteri strains, particularly strain LR6, live or heat-killed, in the food chain.