The following Lactobacillus pentosus and Weissella confusa isolates used in this study to evaluate their probiotic potential were already characterized using 16S rRNA gene sequence analysis and their sequences were submitted to GenBank [18] : Lb. pentosus AJ7 (JX683265), Lb. pentosus AJ9 (JX683266), Lb. pentosus AJ15 (JX683267), Lb. pentosus AJ82 (JX683270), Lb. pentosus AJ94 (JX683271), Lb. pentosus AJ99 (JX683272), Lb. pentosus AJ112 (JX683273), Lb. pentosus AJ116 (JX683274), Lb. pentosus AJ125 (KC533686), W. confusa AJ53 (KC533685), W. confusa AJ67 (JX683268) W. confusa AJ79 (JX683269). The indicator strains Listeria monocytogenes MTCC657, Staphylococcus aureus subsp. aureus MTCC737, Lactobacillus rhamnosus MTCC1408, Lactobacillus plantarum 6160 were procured from MTCC (Microbial Type Culture Collections), Institute of Microbial Technology, Chandigarh, India. The microbial culture media and analytical grade chemicals were procured from HiMedia, (Mumbai, India) and Animal cell culture medium and Histopaque from Sigma-Aldrich, (USA).

The effect of pH and bile on the growth of lactobacilli was studied by growing the isolates in the de Mann Rogosa & Sharpe (MRS) broth adjusted to different pH (3.0, 3.5 and 7.0) using 2 mol∙L⁻¹ HCl or NaOH and MRS broth supplemented with 0.05%, 0.1%, 0.3% (w/v) of bile and MRS broth without bile at pH 6.5 with initial inoculum of overnight culture of isolates to obtain the final optical density (OD) 0.2 at 600 nm, incubated at 37˚C. The OD readings were taken every hour for eight hour then at twenty-fourth hour in a sterile flat bottom ninety- six well microtitre plate using 200 microliters of culture at 600 nm using ELISA reader (Molecular Devices) [19] .

The effect of gastric juice on the survival of the isolates was checked at 0, 2 and 4 h because the time from entrance to release from the stomach is 3 - 4 h and similarly, for the intestinal juice also the survivability of isolates was checked at 0, 2 and 4 h [20] . The isolates were propagated overnight in MRS broth, cells were harvested by centrifugation and 10⁹ - 10¹⁰ CFU mL⁻¹ were suspended in artificial gastric fluid (NaCl, 0.72 g∙L⁻¹; KCl, 0.05 g∙L⁻¹; NaHCO₃ g∙L⁻¹; pepsin, 0.3 g∙L⁻¹) adjusted to pH 2.0, 2.5 and 3.0 and incubated for 0, 2 and 4 h. The intestinal juice resistance was studied by exposing the isolates (10⁹ - 10¹⁰ CFU mL⁻¹) to artificial intestinal fluid (0.1% w/v pancreatin and 0.3% w/v bile salts, pH 8.0) for 0, 2 and 4 h [20] [21] . Samples from all trials plated in MRS agar to count colony forming units. The sterile saline (0.85% w/v NaCl) adjusted to pH 7.0 was used as control.

The bile salt hydrolase (BSH) activity of LAB isolates was checked by streaking the fresh culture of LAB on BSH medium consisting of MRS Agar supplemented with 0.5% (w/v) sodium salt of TDCA (taurodeoxycholic acid) and 0.37 g∙L⁻¹ CaCl₂. The plates were incubated in an anaerobic jar at 37˚C for 36 - 48 h. Precipitation zones around colonies were considered as positive test for the BSH activity [22] .

The antibiotic susceptibility of the lactobacilli isolates was checked using antibiotic octadiscs (HiMedia, Mumbai, India) towards various antibiotics such as aminoglycosides, penicillins, tetracyclines, cephalosporins, inhibitors of urinary tract antiseptics, nucleic acid synthesis and cytoplasmic membrane functions [23] . The antibiotic discs were placed on the MRS agar media seeded with the LAB isolates and incubated anaerobically at 37˚C for overnight. The susceptibility was recorded by measuring diameter of inhibition zones around the discs.

For the aggregation between same cells (auto-aggregation), LAB isolates were grown in MRS broth for overnight at 37˚C followed by harvesting (centrifugation, 10,000 g), washing and re-suspension of pellet in sterile PBS (pH 7.0) and adjusted to 1OD at 600 nm. The cell suspension left for 1 h at room temperature followed by centrifugation at 300 g for 2 min at 20˚C and its OD was read at 600 nm. Auto-aggregation was determined using the following equation.

OD₀ and OD₆₀ refers to the initial OD and OD of the supernatant determined after 60 min, respectively [19] .

The co-aggregation (aggregation between genetically different microorganisms) of LAB isolates was performed with the indicator strains Lactobacillus plantarum MTCC 6160 and Lactobacillus rhamnosus MTCC 1408 which were grown in 10 mL of MRS broth and Listeria monocytogenes MTCC 657 and Staphylococcusaureus subsp. aureus MTCC 737 grown in 10 mL of Tryptic soya broth at 37˚C [19] . Briefly, the cells were harvested by centrifugation (10,000 g) after overnight incubation, washed and resuspended in sterile PBS and adjusted to 1 OD at 600 nm. One mL of each cell suspension (i.e., indicator strain and LAB strain to be evaluated for probiotic potential) was transferred to a tube and OD recorded at 600 nm. The mixed cell suspension left for 1h at room temperature followed by centrifugation at 300 g for 2 min at 20˚C and its OD was read at 600 nm. The co-ag- gregation was determined using the following equation.

OD_tot and OD_S refers to the initial OD taken immediately after the relevant strains mixed and OD of the supernatant after 60 min respectively.

The hydrophobicity assay is an index to measure the ability of isolates to adhere the mucosal surfaces of epithelial lining [23] . The LAB isolates were grown in MRS broth for overnight at 37˚C followed by harvesting, washing and re-suspension of pellet cells in sterile PBS (pH 7.0) and adjusted to OD of 1 at 600 nm. A sample of 1.5 ml LAB suspension was mixed to 1.5 ml of n-hexadecane by vortex and incubated 30 min at room temperature for the separation of aqueous and organic phases. The OD of aqueous phase was read at 600 nm [22] . The hydrophobicity percentage was calculated as follows

where OD1 refers to the initial OD of the cell suspension and OD2 refers to the OD of the cell suspension in aqueous phase after incubation of 30 min.

A probiotic strain must adhere the intestinal mucosal surfaces to act in situ and provide gut health benefits. To evaluate the adhesion ability of isolates to intestinal mucosa, epithelial intestinal HCT-15 cell line was chosen [24] . The adhesion of LAB isolates and inhibition of pathogen adhesion to HCT-15was evaluated. Additionally, it is recognized that the application of probiotics can go beyond gut health and it can be used as suppository for vaginal health and thereby may reduce the risk of urinary tract infections, bacterial vaginosis and yeast vaginosis by balancing the microflora [25] - [27] . To study the adhesion ability of isolates at vaginal mucosa, vaginal HeLa cell line was chosen [26] [27] . Both the cell lines were maintained in the DMEM medium supplemented with 10% of fetal bovine serum (FBS), 2 5 mM HEPES, 50 μg∙mL⁻¹ penicillin, 50 μg∙mL⁻¹gentamicin, 50 μg∙mL⁻¹ streptomycin and 1.25 μg∙mL⁻¹ amphotericin-B. The cell lines were used for assay when reached to 90% to 95% confluences. For the LAB adhesion and inhibition of pathogen adhesion to the cell lines, monolayers of cell lines were maintained in the six well tissue culture plates (Tarson, India) and the LAB isolates and pathogen were suspended in serum and antibiotic free DMEM medium. The monolayers of cell lines were washed with the PBS thrice to remove antibiotics followed by the addition of LAB suspensions (10¹⁰ - 10¹¹ CFU mL⁻¹) and kept for 90 min at 37˚C [20] . After incubation, wells were washed gently with PBS thrice and the non-adherent recovered LAB cells were plated over MRS agar. The adhesion of LAB isolates were calculated as the amount of adhered bacterial cells, compared to the initial amount of LAB cells added to the cell lines with the following formula,

Amount of adherent LAB cells = log (A) − log (B)

Where A is the No. of viable LAB cells added to the well for adhesion and B is the No. of viable LAB cells non-adhered to the cell lines in the well.

The culture wells, subsequent to incubation with LAB isolates and removing the non-adhered bacterial cells with PBS wash, were fixed with 2% formaldehyde followed by the staining with 0.1% acridine orange dye and observed under fluorescent microscope. For the pathogen (Listeria monocytogenes MTCC 657) exclusion assay, the monolayers were first treated with the LAB isolates (5 ´ 10¹⁰ - 10¹¹ CFU mL⁻¹) for 90 min, washed with PBS to remove non-adhered LAB followed by addition of Listeria cells (2 ´ 10¹⁰ CFU mL⁻¹) suspended in serum and antibiotic free medium and incubated for 60 min [20] . The excluded Listeria cells were recovered by washing with the PBS and were plated on Listeria selective agar [28] . The amount of excluded Listeria cells were calculated as the amount of Listeria cells recovered after incubation compared to initial amount added to the cell lines with the following formula,

Amount of excluded Listeria cells = log (X) − log (Y)

where X is the no. of viable Listeria cells added to the well for adhesion and Y is the no. of viable Listeria cells adhered to the cell lines in the well.

The hPBMCs were isolated from the blood of healthy human donor. Briefly, 3.0 mL of Histopaque 1077 (Sigma Aldrich, USA) was placed in a 15 mL centrifuge tube and 3.0 mL of heparinized whole blood was layered on the top followed by centrifugation (400 g, 30 min at room temperature). The upper plasma layer was removed and discarded. The cells separated at the interface was taken up by aspiration and washed with PBS. The cell viability test was determined by trypan blue exclusion and the cell density was adjusted to 10⁷ cells mL⁻¹ in RPMI-1640 medium supplemented with 10% autologous serum, 50 μg∙mL⁻¹ penicillin, 50 μg∙mL⁻¹ gentamicin, 50 μg∙mL⁻¹ streptomycin and 1.25 μg∙mL⁻¹ amphotericin-B and incubated at 37˚C for overnight in a suspension flask. The hPBMCs were incubated with the LPS (1.0 μg∙mL⁻¹) for 30 min before exposed to LAB isolates. Fresh bacterial suspensions of LAB isolates (10⁸ CFU mL⁻¹) were added to hPBMCs (10⁷ cells mL⁻¹) resulting in bacteria-to-cell ratio 10:1 and co-incubated for 24 h in antibiotic and serum free RPMI-1640 medium at 37˚C in a 12-well tissue culture plates [29] . After incubation the supernatants were collected and used immediately for estimation of cytokines. Because, Lipopolysaccharide (LPS) provokes host cells to produce pro-inflammatory mediators including TNF-α, IL-1α, IL-6 by activating several types of transcription factors, LPS stimulated hPBMCs culture was used to study the anti-inflammatory potential of probiotic strains for reduction in TNF-α. The anti-inflammatory cytokine IL-10 was also measured in the supernatant of hPBMCs. The assays for both the pro-inflammatory (TNF-α) and anti-inflammatory (IL-10) cytokines were performed twice independently in duplicates [29] . TNF-α and IL-10 cytokines were measured using anti-human TNF-α and anti-human IL-10 monoclonal antibodies respectively by sandwich ELISA kit (Diasource) as per manufacturer’s recommendations.

In situ inhibition of Listeria by LAB isolates was studied by co-culturing both LAB and Listeria for different time intervals and subsequently their CFU were calculated. The LAB isolates were grown in nutrient broth till the OD reached 0.8 at 600 nm. Overnight culture of Listeria monocytogenes MTCC657 grown in nutrient broth was added to the LAB culture and incubated at 30˚C under shaking condition (180 rpm) in a shaker incubator. The CFU in the co-culture were calculated for LAB and Listeria by selectively plating over MRS agar and Listeria selective agar respectively at 0, 3, 6, and 9 h of intervals. A control LAB and control Listeria culture were also processed in similar way.

The Lb. pentosus isolates were growing in the MRS broth containing 0.3% of bile (normal concentration in the intestine of healthy man) [4] [30] whereas no growth was observed in case of W. confusa isolates AJ53, AJ67 and AJ79 (Table 1) and were excluded for the further studies. Their resistance to simulated gastric and intestinal juices is shown in Table 2. Based on the growth profile in acidic conditions and presence of bile salts (Table 1) and survivability in the simulated gastric and intestinal juices (Table 2), six isolates AJ7, AJ9, AJ15, AJ82, AJ99 and AJ112 were chosen for the cell line adhesion studies.

The isolates AJ9, AJ99 and AJ116 were BSH negative, while others were positive (Table 3).

Legends: −, no growth; +, very weak growth, ++, weak growth, +++, good growth, ++++, luxurious growth.

The values are mean of two independent experiment performed in duplicates, Legends: +, positive; −, negative; ND, not determined, ^*calculated by formula: amount of adhered LAB cells = log (A) ? log (B), where A is the number of viable LAB cells added to the well and B is the number of viable non-adhered LAB cells, ^**calculated by formula: amount of excluded Listeria by LAB cells = log (X) ? log (Y), where X is the number of viable Listeria cells added to the well and Y is the number of viable adhered Listeria cells.

The LAB isolates were found to be susceptible to most of the antibiotics screened (Table S1). All lactobacilli isolates were sensitive to penicillin, ampicillin, erythromycin, ciprofloxacin, gentamicin, lincomycin, chloramphenicol, streptomycin, amikacin, cephaloridine and tetracycline. However, all the isolates were resistant to colistin.

The auto-aggregation between lactobacilli isolates varied from 34% to 63%, while co-aggregation between LAB and non-LAB strains varied from lowest 24% to highest 57% (Table 3). The hydrophobicity of LAB isolates were ranged from least 19% (AJ15) to maximum 90% (AJ82) (Table 3).

The ability of six selected isolates to adhere to epithelial cell lines and pathogen exclusion is shown in Table 3. The isolateAJ15 neither adhered to intestinal HCT-15 cell line and nor inhibited the adhesion of Listeria pathogen, however, strains AJ7, AJ9, AJ82, AJ99 and AJ112 were able to adhere to both the epithelial cell lines HCT-15 and HeLa (Figure 1(A) & Figure 1(B)) considerably better or similar percentage than other lactobacilli strains evaluated in literature [26] as well as inhibited the adhesion of Listeria cells greatly (Table 3).

The production of cytokines TNF-α and IL-10 by four selected LAB isolates (AJ7, AJ82, AJ99 and AJ112) was used as an index of immunomodulation (Table 4). The TNF-α induction was higher in control sample of LPS treated hPBMCs than the hPBMCs incubated with LAB strains. However, there was weak induction of TNF-α in the case of hPBMCs incubated with both the LPS and LAB together (Table 4). It indicates that LAB isolates AJ7, AJ82, AJ99 and AJ112 were able to reduce induction of pro-inflammatory cytokine TNF-α in LPS treated hPBMCs. The stimulation of anti-inflammatory cytokine IL-10 was observed in case of hPBMCs incubated with both the LPS and LAB strains independently and together. The induction of IL-10 in hPBMCs incubated with the LPS was weak in comparison to the hPBMCs incubated with LAB and hPBMCs incubated with LPS and LAB strains both (Table 4). The results show that these isolates AJ7, AJ82, AJ99 and AJ112 induce low TNF-α and high IL-10.

The two potent isolates AJ7 and AJ82 displayed a very good antagonistic activity against Listeria monocytogenes MTCC657 in nutrient broth when co-cultured (Figure 2(A) & Figure 2(B)). The isolates AJ7 and AJ82 competitively inhibited Listeria. The individual control cultures of both the LAB and Listeria had grown exponentially whereas the number of viable Listeria cells when co-cultured with LAB isolates were dramatically decreased to zero at 9 h of co-incubation probably owing to the effects of various antimicrobial substances released by LAB [6] . On the other hand, the LAB isolates co-cultured with Listeria grew similar to control LAB culture.