Lactobacillus fermentum MTCC 8711 and MRSA ATCC 43300 were grown in de Man, Rogosa and Sharpe (MRS) broth (pH 6.5) (Himedia, Mumbai, India) and brain heart infusion (BHI) broth (pH 7.4) (Himedia), respectively at 37°C. Both strains were maintained at -80°C in the appropriate cultivation broth containing 20% (v/v) glycerol.

The Caco-2 human colon adenocarcinoma cell line was obtained from the American Type Culture Collection. Cells were routinely maintained in DMEM F12 Ham (pH 7.6) (Himedia) supplemented with non-essential amino acids, L-glutamine, sodium pyruvate, antibiotics, and 20% of heat-inactivated FBS at 37°C in a humidified chamber with 5% CO₂ supply. The cryopreserved cells were thawed rapidly by placing the vial at 37°C. Cells were transferred to a T₂₅ flask containing 5 ml of pre-warmed complete culture media. The cells were seeded at a density of 4 × 10⁵ cells/ml. The cells were grown at 37°C in a 95% air-5% CO₂ for 10–15 days till they form a monolayer with 70% confluency. At 70–80% confluence, cells were trypsinized for 3–5 min. Complete culture medium was added to each flask and mixed with the trypsinized cells to inactivate trypsin. Cells were centrifuged at 1000 rpm for 2 min at room temperature in a 15-ml centrifuge tube. The supernatant was removed, and the pellet was resuspended into 8 ml of culture medium. Following the cell counts by trypan blue exclusion, subsequently, cells were subcultured into T25/T75 flasks. Media was replaced routinely every 2 days.

The adhesive property of L. fermentum MTCC 8711 was evaluated using the Caco-2 cell line model. In vitro adhesion assays were performed as described by Bouchard et al. (2013). Caco-2 monolayer cells were prepared in 6-well plates, washed twice with phosphate-buffered saline (PBS, pH 7.4) to remove the unattached cells. An overnight culture (16 h) of L. fermentum MTCC 8711 (10⁷ CFU/ml) was harvested and co-cultured with the monolayer of Caco-2 cells for 1 h at 37°C in antibiotic- and serum-free DMEM-F12 HAM. After 1 h of incubation, the monolayers were washed three times with sterile PBS to remove non-adherent bacteria. The adhered bacteria were detached by adding 1 ml of 0.04% Tween 80, serially diluted with PBS, and appropriated dilution was plated onto MRS agar plates. The CFUs were calculated after overnight growth at 37°C to determine the number of adhered bacteria on Caco-2 cell monolayer.

For microscopic observation, an overnight culture of L. fermentum 8711 grown in MRS broth was pelleted and suspended in 1 ml of 0.1 M sodium carbonate buffer. Fluorescein isothiocyanate (FITC) at a final concentration of 2 μg/ml was added to the suspension and incubated at 20°C for 60 min. After incubation, the bacterial cells were washed three times with PBS to remove the excess stain and suspended in 1 ml of antibiotic-free DMEM. The Caco-2 cells were stained with Mitotracker Deep Red (1 μg/ml) and Hoechst (1 μg/ml), incubated for 30 min at 37°C in a 5% CO₂. Bacterial suspension (L. fermentum) was added with MOI of 1:100 to Caco-2 cells monolayer. The plate was incubated for 1 h at 37°C in 5% CO₂–95% air. The cells were washed twice with PBS. The cells were carefully fixed to a glass slide using methanol and glycerol. The fixed cells were observed, and images were captured using high content screening system-Operetta (Perkin Elmer).

Also, adhesion efficiency of L. fermentum was evaluated by flow cytometry analysis. Briefly, Caco-2 cells were grown in 6-well plates containing complete culture medium at 37°C with 5% CO₂ until they reach 70% confluency. After achieving the confluency, cells were harvested by trypsinization and centrifuged at 1500 rpm for 5 min. Then the Caco-2 cells were stained with Mitotracker Deep red and subsequently treated with FITC stained L. fermentum and incubated for 1 h. After incubation, treated cells were centrifuged at 1000 rpm for 2 min and washed twice and resuspended in sterile PBS and subjected to flow cytometry analysis using FACSAria III (Becton Dickinson, San Jose, CA, United States).

The adhesion property of MRSA was evaluated using Caco-2 cell line. Caco-2 monolayer cells were prepared in 6-well plates, washed twice with phosphate-buffered saline (PBS, pH 7.4) to remove the unattached cells. An overnight culture (16 h) of MRSA was harvested and infected with the monolayer of Caco-2 cells for 1 h at 37°C in antibiotic and serum free DMEM-F12 HAM. After 1 h of incubation, the monolayers were washed three times with sterile PBS to remove unbound bacteria. The adhered bacteria were detached by adding 0.1X Triton-X, serially diluted with PBS, and appropriated dilution was plated on Mannitol Salt Agar agar (pH 7.4) (Himedia). The plates were incubated overnight at 37°C to determine the number of adhered bacteria on Caco-2 cell monolayer.

For microscopic observation, an overnight culture of MRSA grown in BHI broth was centrifuged, and the cells were suspended in 0.1 M sodium bicarbonate buffer. The bacterial suspension was stained with FITC (25 μg/ml) for 30 min at 37°C. The Caco-2 cells monolayer was stained with Mitotracker Deep Red (1 μg/ml) and Hoechst (1 μg/ml), incubated for 30 min at 37°C in a 5% CO₂. MRSA with MOI of 1:10 was incubated with Caco-2 cells for 2 h at 37°C in a 5% CO₂. After incubation, the cell monolayer was washed twice with PBS. Cells were observed and examined using high content screening (Operetta- Perkin Elmer).

Also, adhesion efficiency of MRSA was analyzed by flow cytometry. Caco-2 cells were grown in 6-well plates containing complete culture medium at 37°C with 5% CO₂ until they reach 70% confluency. After achieving the confluency, cells were trypsinized and centrifuged at 1500 rpm for 5 min. Then, the Caco-2 cells were stained with Mitotracker Deep Red and hochest. Caco-2 cells were infected with acridine orange (AO) stained MRSA and incubated for 1 h. After incubation, infected cells were centrifuged at 1000 rpm for 2 min and washed and resuspended in sterile PBS and subjected flow cytometry analysis using FACSAria III.

Anti-adhesion property L. fermentum against MRSA on Caco-2 cells was evaluated using three independent assays: (i) inhibition of adhesion, (ii) competitive adhesion, and (iii) displacement. For microscopic observation, the Caco-2 cells and MRSA were stained with Hoechst (1 μg/ml), and L. fermentum was stained with FITC and observed using high content screening system, Operetta (Perkin Elmer).

The adhesion inhibition assay was performed according to the method described by Gueimonde et al. (2006) with slight modifications. We used 1X PBS to wash the cells instead of HH buffer used by Gueimonde et al. (2006). The overnight culture of MRSA and L. fermentum MTCC 8711 were grown in TSB and MRS broth respectively. L. fermentum MTCC 8711 with MOI of 1:100 was incubated with Caco-2 monolayer separately for 1 h at 37°C in a 5% CO₂–95% air. After incubation, unbound L. fermentum cells were removed by washing twice with 1X PBS followed by infection with MRSA (MOI 1:10) for 2 h at 37°C in a 5% CO₂–95% air and number of bacteria adhering to the Caco-2 cells were determined by plating on MRS plates for L. fermentum, and MSA plates for MRSA after serial dilution followed by the incubation at 37°C for 24 h. Each experiment was repeated three times.

The competitive adhesion assay was performed as described previously (Jankowska et al., 2008) with slight modifications. They have used the probiotic and pathogenic bacteria in equal numbers with MOI of 1:100. Based on the preliminary experimental results, we have used MRSA with MOI of 1:10 and L. fermentum with MOI of 1:100. The overnight culture of MRSA and L. fermentum MTCC 8711 were grown in their specific broth at 37°C. MRSA and L. fermentum cells were co-incubated with Caco-2 cells monolayer for 2 h at 37°C in a 5% CO₂–95% air. After incubation, Caco-2 cells monolayer was washed twice with sterile PBS. Bacteria were detached from Caco-2 cells by Triton-X (0.1X) treatment, and the number of bacteria adhering to the Caco-2 cells were determined by plating on MRS and MSA plates and incubated at 37°C for 24 h.

An overnight culture of MRSA and L. fermentum MTCC 8711 were grown in their specific broth. MRSA with MOI of 1:10 was incubated with Caco-2 cells monolayer for 2 h at 37°C in a 5% CO₂–95% air. After incubation, cells were washed twice with sterile PBS. The Caco-2 cells monolayers were then treated with L. fermentum MTCC 8711 with MOI of 1:100 and incubated further for 1 h under a 5% CO₂ atmosphere. The cells were washed twice with PBS. Adhered bacteria were detached and serially diluted. Appropriate dilutions were plated in MRS and MSA plates and CFUs were estimated.

Caco-2 cells were seeded at a density of 10⁵ cells in a 12 well plate containing complete culture media and incubated at 37°C in a 95% air–5% CO₂. The Caco-2 monolayer was washed twice with DMEM. The Caco-2 cells were infected with MRSA (10⁶ cells) for 2 h and incubated at 37°C in a 95% air–5% CO₂. After incubation, the monolayer was washed twice with 1X PBS and 1 ml of complete culture media with 10 μg/ml lysostaphin was added. The plate was incubated at 37°C for 24 h. Subsequently, the Caco-2 cells monolayer was treated with L. fermentum with MOI of 1:100 in DMEM for 1 h at 37°C. After 1 h of incubation, the cell monolayer was washed twice with 1X PBS. The plate was incubated for 24 h at 37°C in a 95% air–5% CO₂. Untreated Caco-2 cells were maintained as a control. After incubation, Caco-2 cells medium was removed from both MRSA, and L. fermentum treated cells and incubated in 0.5 mg/ml methylthiazolyldiphenyltetrazolium bromide (MTT) for 4 h at 37°C. After incubation, the solution was removed, and 500 μl of DMSO was added to dissolve the formazon crystals. Absorbance was read at 560 and 630 nm in Multimode Plate Reader (Perkin Elmer). Uninfected cells were used as a control. A subtraction analysis of the dual wavelength (D550 to D690) was performed to increase the accuracy of cytotoxicity measurement.

Live and dead cell assay was performed to determine the viability of MRSA infected Caco-2 cells. Caco-2 cell monolayer was washed twice with 1X PBS, and the cells were suspended in DMEM medium. The cells were infected with MRSA at an MOI of 1:10 and further incubated for various time points (4, 16, and 24 h) at 37°C. At each time point, the cells were trypsinized using Trypsin-EDTA (HiMedia) for 3 min followed by addition of 20% FBS for inactivation. The trypsinized cells were centrifuged and suspended in 1X PBS. Propidium iodide (PI) (5 mg/ml) was added to each sample, and live/dead cells were quantified using FACSAria III. Uninfected cells were used as a control.

Caco-2 cell monolayer was washed twice with 1X PBS and the cells were suspended in DMEM. The Caco-2 cell monolayer was infected with MRSA (MOI 1:10) and incubated at 37°C for 2 h with 5% CO₂–95% air. After 2 h, the cells monolayer was washed and treated with L. fermentum (MOI 1:100) for 1 h at 37°C. After treatment; the cells monolayer was washed and incubated for 24 h at 37°C with 5% CO₂–95% air. After incubation, the Caco-2 cells were trypsinized using Trypsin- EDTA (HiMedia) for 3 min and inactivated by adding 20% FBS. The trypsinized cells were centrifuged and suspended in 1X PBS. PI (5 mg/ml) was added to each sample, and cell viability was analyzed using FACSAria III.

All experiments were repeated three times, and the data are expressed as Mean ± SD. Data were analyzed for statistical significance using Student’s t-test and a p-value of <0.05 was used as a threshold for significance.

Lactobacillus fermentum MTCC 8711 (Accession number: AVAB00000000) genome sequence was retrieved from NCBI. Adhesins of L. fermentum from the genome sequence were predicted using a Software program for Prediction of Adhesins and Adhesin-like proteins using Neural networks (SPAAN; Sachdeva et al., 2005). SPAAN uses neural networks integrated with five different attributes like amino acid frequencies, multiplet frequencies, dipeptide frequencies, charge composition, and hydrophobic composition. Proteins with P_ad (protein being an adhesin) score greater than 0.7 were considered as potential adhesins. Cellular localization of the potential adhesins was predicted using CELLO v.2.5 (Yu et al., 2006).

We found that the L. fermentum 8711 has a potential adhesion property to the Caco-2 cells and the estimated adhered bacterial cells were 1.0 × 10⁷ CFU/ml. Similarly, we have studied the adhesion property of MRSA on Caco-2 cells and found that MRSA also potentially adhered to the surface of Caco-2 cells. CFU estimation revealed that the number of adhered MRSA cells was 4.7 × 10⁶ CFU/ml. Thus, comparatively, L. fermentum has stronger ability to adhere to Caco-2 cells than MRSA (Figure 1). Figures 2, 3 shows the microscopic observation of adhesion of FITC stained L. fermentum 8711 and MRSA, respectively, on the surface of Caco-2 cells. Flow cytometry analysis revealed that L. fermentum cells adhered to 39.1% of the Caco-2 cells and MRSA adhered to 35% of the Caco-2 cells (Figure 4).

Inhibition of adhesion, competition, and displacement assays were performed to evaluate the ability of L. fermentum to interfere with the adhesion of MRSA on Caco-2 cells.

Caco-2 cells were pre-incubated with L. fermentum for 2 h with the MOI of 1:100, followed by the addition of MRSA at an MOI of 1:10. The ability of L. fermentum to inhibit the adhesion of MRSA to Caco-2 cells is shown in Figure 5A. A significant reduction in the adhesion of MRSA was observed, and the mean CFU value (2.0 ± 0.35 × 10⁶) was 63.2% lower when compared to the treatment with MRSA alone. Further, we noted that the density of Caco-2 cell monolayer was not disturbed by MRSA as evident by direct microscopic examination and cell counting. Thus, L. fermentum 8711 could efficiently inhibit the adhesion of MRSA and protect the integrity of the Caco-2 cells.

We tested competitive adhesion of L. fermentum and MRSA on Caco-2 cells by co-incubation experiment. When L. fermentum was co-incubated with MRSA and Caco-2 cells for 2 h, adhesion of MRSA was significantly decreased with a CFU of 1.67 ± 0.13 × 10⁶ CFU/ml. The adhered CFU value was 71.1% lesser when compared to the MRSA infection alone (Figure 5B).

Pre-infection of Caco-2 cells with MRSA for 2 h and subsequent addition of L. fermentum resulted in an efficient displacement of adhered MRSA from the Caco-2 cells. The adhered CFU value of MRSA was 96.5% lesser when compared to the MRSA infection alone (Figure 5C). These results suggested that L. fermentum 8711 has potential anti-adhesion property against MRSA.

Effects of L. fermentum 8711 and MRSA on cell viability of Caco-2 cells were determined at 24 h of post-infection using MTT assay. L. fermentum did not affect the Caco-2 cells, and the viability was ∼99% similar to that of untreated cells. MRSA induced a significant cytotoxic effect in Caco-2 cells, and only 24% cells were viable after 24% of infection. When the MRSA infected Caco-2 cells were treated with L. fermentum (displacement assay), the cell viability was increased to a level of 86% (Figure 6). Thus, L. fermentum has reduced the MRSA-induced cytotoxicity.

To study the cytotoxicity, Caco-2 cells infected with MRSA for various time points were subjected to live-dead assay using flow cytometry. At 4 h post-infection with MRSA, 31% of dead cells were identified, and after 24 h post-infection, dead cells were increased to 70% (Figure 7). L. fermentum also inhibited the cytotoxic effect of MRSA in Caco-2 cells. As shown in Figure 7, L. fermentum showed a protective effect against cytotoxicity induced by MRSA in Caco-2 cells. The dead cells were reduced to a level of 30.9% from 70%. This result suggests that L. fermentum effectively displaces the adhered MRSA from Caco-2 cells and further reduces the MRSA-induced cytotoxicity.

A total of 2266 protein sequences of L. fermentum MTCC 8711 were retrieved from NCBI, and the possible adhesins were predicted using SPAAN. Out of 2266 proteins, SPAAN predicted 98 proteins as putative adhesins or adhesin like proteins with Pad value > 0.7 (Supplementary Table S1). Of these, CELLO predicted 77 proteins to be localized in the extracellular space. Other 12 proteins were localized in the inner membrane and 9 proteins were cytoplasmically localized. Among the predicted adhesins, the majority of them were hypothetical proteins, and their roles are yet to be identified.