Technological and Probiotic Traits of the Lactobacilli Isolated From Vaginal Tract of the Healthy Women for Probiotic Use

1. Background
 Urogenital tract infections (UGTI) and sexually transmitted infections are the two major medical problems, affecting millions of women every year (1-3). Although, antibiotic and drug therapy can be effective for the eradication of the urinary tract infection, however several problems have emerged, such as multiresistant bacteria (4-6). For the prevention and treatment of UGTI, probiotic lactobacilli were proposed as an alternative to the antibiotics (7, 8). Moreover, vaginal lactobacilli have a number of properties which render them highly suitable for probiotic (5, 9). A healthy vaginal women microbiota is dominated by lactobacilli. They play a significant role in the maintenance of ecological balance (10, 11). The protective role of lactobacilli is based on several mechanisms; specific adherence to the vaginal epithelium and inhibition of the pathogens adhesion to the vaginal epithelial surface, coaggregatin with some uropathogenic bacteria , competition for nutrients, production of the active metabolites including organic acid, mainly lactic acid that contributes in the maintenance of the low vaginal pH (4-4.5), hydrogen peroxide (H2O2) is usually generated by lactobacilli present in a healthy vagina, production of bacteriocins (5, 10-12), but there are significant differences between strains (14). However, there are only few reports concerning probiotics residence in the female reproductive tract, although the presence of lactobacilli in vagina was observed by Doderlein in 1841 (6).
 The aim of this study was to get a better knowledge on the potential probiotic properties of the vaginal lactobacilli of the healthy Jijelian women (Algeria) in order to gain new indications on how to improve their exploitation in new the technological applications in probiotic production.

2. Objectives
 The main goal of the present work was to isolate lactobacilli strains from vaginal samples and to investigate their probiotic properties.

3. Materials and Methods

3.1. Vaginal Samples
 Vaginal specimens were obtained from 60 women (19-46 years) with healthy vaginal ecosystem of Jijel region. From vaginal fluid, samples were collected with sterile cotton swabs inserted into the vagina, rotated a few turns along the vaginal sidewall, and allowed to absorb for few seconds (15). The swabs were immersed in the sterile normal saline, and used for anaerobic (BD GaspakTM Anaerobic system, USA) culture preparation by streaking on de Man Rogasa Sharpe (MRS) Agar (CONDA, Pronadisa, Spain) plates (16). The same swab was used to prepare a smear, underwent Gram staining, and was evaluated by using Nugent criteria. The flora was interpreted as normal (score 0-3), as intermediate (score 4-6), and interpreted as consistent with BV (score 7-10) (17).

3.2. Isolation and Identification of Microorganisms
 The swabs were vortexed for 1 min at maximum speed, the resultants suspensions were diluted 7 times with potassium phosphate buffer (pH 7.4), and 50 mL of each dilution was seeded on MRS agar plate (pH 5.5). The plates were incubated for 48 h at 37ºC under anaerobic condition (16, 18). To identify strains, each colony was subjected to colony morphology, Gram coloration, catalase activity test, motility, growth at various temperatures, and homofermentative/heterofermentative test. They were further characterized by the carbohydrate fermentation, then stored at -20ºC in MRS broth supplemented with glycerol (30% v/v) (13, 19).

3.3. Determination of Hydrogen Peroxide Produced by Lactobacilli
 Isolates were cultured into MRS agar medium supplemented with 250 mg.mL-1 3,3´,5,5´-tetramethylbenzidine (Sigma-Aldrich, USA) and 0.01 mg.mL-1 horseradish peroxidase (Sigma-Aldrich, USA), incubated in an anaerobic condition at 37ºC for 48 h. The plates were exposed to aerobic environment for 30 min. The color intensity was graded the potential of H2O2 production (20, 21).

3.4. Determination of Acid and Bile Salt Tolerance
 Fourteen (14) isolates (H2O2+) were cultured for 16-18 h and the cells were collected by centrifugation (30000 ×g, 15 min). The pellet were washed twice and finally suspended in phosphate buffer saline (PBS). Approximately 108 CFU.mL-1 of each isolate was inoculated into the acidified PBS pH 2.0 and incubated for up 2 h. The viable cells were counted at 0 h and after 2 h (22). The bile tolerance of each isolate was determined by comparing the count after 8 h of exposure to 0.3% bile salt  Oxgall (Sigma-Aldrich, USA), (w/.v) with initial count at 0 h (22).

3.5. Survival in Simulated Human Intestinal Fluid
 Simulated intestinal fluid (SIF) was prepared with 9 g.L-1  NaCl, 10 g.L-1  of pancreatin, 10 g.L-1  trypsin and 3 g.L-1 of bile salts (pH 6.5). The cultures were incubated in this solution for 180 min at 37ºC. The number of viable cells was counted at 0 h, 90 min and after 180 min (23).
 
3.6. Adhesion to Vaginal Epithelial Cells (VECs)
 VECs were collected from healthy premenopausal volunteers’ women by sterile cotton swabs, immersed in 0.04 M citric acid -Na2HPO4 buffer pH 4.5 and stored at 4ºC for less than 3 h until use. The VECs were washed using the same buffer, centrifuged at 800×g for 4 min and resuspended to a concentration of 1×106 VEC.mL-1 (24). An overnight culture of the lactobacilli was suspended to reach 108 CFU.mL-1 in normal saline. Equal volumes of the bacterial suspension and the vaginal cells were mixed and incubated for 1h at 37ºC and the cells with adherent bacteria were collected and washed three times in citric-acid- Na2H2PO4 buffer (800 ×g, 7min). Bacterial adhesion to VECs was assessed by microscopy (×100) after staining with 1% of crystal violet. The number of bacteria attached to 50 consecutive VECs smears was counted and VECs Controls smears were made to confirm that the presence of native bacteria was negligible (15).

3.7. Growth Inhibition of Vaginal and Urogenital Pathogens
 Indicator strains used were E. coli ATCC25922, Staphylococcus aureus ATCC25923, Klebsiella pneumonia ATCC700603 (UHC, Constantine, Algeria), vaginal strains (E. coli, Staphylococcus aureus, Candida albicans) and urogenital strains (E. coli, Klebsiella. sp).
 Inhibitory activity was determined by well diffusion test with minor modifications (17). The surface of Muller-Hinton agar plate was spread with a standardized suspension of each pathogen microorganisms (107CFU.mL-1). Culture supernatants of the lactobacilli isolates were filtered through 0.45 mm cellulose filter (Sartorius, Germany). 25 mL of the cultures and supernatants of lactobacilli were placed into wells in the pathogens inoculated plates and were incubated for 24 h at 37ºC. The diameter of inhibition zones was measured. Control assays of MRS medium with pH 6.5 and pH 4.0 were also performed (5).

3.8. Autoaggregation and Coaggregation Assay
 Bacterial cells of overnight culture were harvested by centrifugation at 5000 ×g for 15 min, washed twice in PBS pH 6.0 to give viable counts of 108 CFU.mL-1. Four ml of cell suspension were mixed by vortexing for 10 s and autoaggregation was determined during 5 h of incubation at room temperature. 0.1 mL of the upper suspension was transferred to another tube containing 3.9 mL PBS and the absorbance was measured at 600 nm (25). The percentage of autoaggregation was calculated by the following expression:

Autoaggregation (%) = [ODi - ODf / ODi] ×100

 Where ODi is the OD at initial time (t=0 h) of autoaggregation assay, and ODf is the OD at t=1 h, 2 h, 3 h, 4 h and 5 h.
 For the co-aggregation experiment, equal volumes (2 mL) of the lactobacilli and indicator strain cultures were mixed together by vortexing for 10 s. 4 mL of each bacterial suspension were used as a control. The absorbance was measured at 600 nm after mixing and 5 h of incubation at room temperature (25). The percentage of coaggregation was calculated using the following equation:

                              [(Ax+Ay)/2] - A (x+y)
% Coaggregation =                                    ×100   
                                     [Ax+Ay] /2

x and y represent strains in the control tube and (x+y) the mixture.

3.9. Hydrophobic Partition
 Bacterial pellets were obtained from overnight cultures, washed, and resuspended in urea magnesium phosphate buffer (pH 6.0). The absorbance of the cell suspension was measured at 450 nm to obtain approximately 1.0. Three mL of the bacterial suspensions were put in contact with 0.6 mL of xylene and vortexed for 2 min. The lower aqueous layer was carefully removed, transferred to the clean tubes, and absorbance was measured as described before (26). The percentage of hydrophobicity was obtained from the following calculation:

% hydrophobicity= [(ODbefore-ODafter)]/ODbefore ×100

3.10. Acid Production
 Lactobacilli strains were inoculated (1%) in MRS broth (pH 6.5) and the amount of acids produced was indirectly determined by measuring the pH of culture supernatants with a pH meter (HANNA, HI 2211/ PH/ ORP Meter). The experiments were performed in duplicate and the mean pH±S.D was calculated (9).

3.11. Antibiotic Resistance
 The antibiotics used were inhibitors of the cell wall synthesis such as Ampicillin (10 mg), Cefotaxime (30 mg), inhibitor of protein synthesis (Gentamicin (10 mg), Erythromycin (15 mg), Chloramphenicol (30 mg), Tetracycline (30 mg) and inhibitor of nucleic acid synthesis (Ofloxacin, (5 mg), Trimethoprim-sulphamethoxazole (25 mg), Ciprofloxacin (5 mg), Nitrofurantoin (200 mg). Antibiotic discs were obtained from (Bioanalyse®). The disk diffusion method was used. After incubation the diameters of the inhibition zone were measured (5, 27).

3.12. Resistance to Spermicides
 Suspensions of 14 isolates were adjusted to 108 CFU.mL-1 and then were inoculated by swabbing onto MRS agar. 10 mL of each concentration (0.1%, 0.25%, 1%, and 2.5%) of nonoxynol-9 (N-9) cream was deposited on each disk. After incubation, the strain was considered sensitive if the diameter of the inhibition zone is more than 9 mm (20).

3.13. Genetic Identification of the Best Lactobacilli
 To identify the most efficient isolates, 16S Ribosomal DNA (16S-rDNA) sequencing was performed.  Genomic DNA was extracted with the Wizard® genomic DNA purification kit (Promega, Madison, USA) according to the manufacturer instruction. To amplify 16S-rDNA, polymerase chain reaction (PCR) was used with the following primers 16SP0: 5´-GAAGAGTTTGATCCTGGCTCAG-3´ and 16SP65´-CTACGGCTACCTTGTTACGA-3´. Fragment of about 1500 bp in size was excised from a 1% (w.v) agarose gel staining and purified with Microcon YM-100 kit (Bedford, MA, USA). The BigDye Terminator sequence was performed using the Vector NTI (Version 8) software package (BD Biosciences, San Jose, USA). The sequences were compared with sequences deposited in the Gen Bank database of the National Center for Biotechnology Information using the BLAST program. The phylogenetic tree was generated by Tree Dyn program (v198.3) proposed by Methods and Algorithms for Bioinformatics LIRMM. (http://phylogeny.lirmm.fr/phylo_cgi/simple_ phylogeny.cgi).

3.14. Statistical Analysis
 Results are expressed as the mean±standard (SD). Statistical differences were analyzed by one way analysis of variance (ANOVA) using SPSS software (version 13) and p values < 0.05 were considered to be statistically significant.

4. Results
 
4.1. Selection of Vaginal Lactobacilli and H2O2 Production
 Using Nugent Scoring System, twenty four women (40%) were classified as having normal vaginal flora, 16 (26.66%) were intermediate and 20 (33.33%) were BV. 40 samples that have score  In this study we used the capacity of producing H2O2 as the first criterion of selection. As shown in (Figure 1), potential production of H2O2 capacity was divided into 4 groups according to the intensity of the color. Among the 70 isolates, 57 (81.42%) were found to be able to produce H2O2 and 26 (37.14%) were hyperproducers. In this study, isolates producing H2O2 were selected to evaluate their potential traits.

4.2. Survival Under Acidic, Bile and SIF Conditions
 Most isolates show considerable tolerance to acid condition with different percentage, where strains B6, B10, and B9 had high tolerance with 82.23%; 79.98%, and 69.08% survival rates respectively (Table 1).
 Concerning the resistance of the same isolates to 0.3% bile-salts for 2 h, an important viability was noted. It was 71.84% for strain A3, 77.38% for B6, and 59.54% for B10 (Table 1). The results showed the high differences between strains after incubation for 180 min in SIF. Isolate B10 represent 71.93% survival rate after 90 min of incubation and reached to 60.59% after 180 min (Table 1). It’ is appearing that strains B6 and B10 have the ability to resist to hostiles conditions.

4.3. Adhesion to Vaginal Epithelial Cells, Aggregation, Hydrophobicity, and Coaggregation
 The highest levels of adhesion was observed for the strains B6 (Figure 2), and C3 with the means of 76, and 80 bacteria per VEC, respectively, while other strains such as A1, A3, B9, B10, C5, and W5 adhered less well to VEC with a mean adherence of 60 bacteria at most (Table 2).
 The results showed that the isolates adhered to VEC with varying degrees, where isolates C3, B6, and B10 were the most adhesives. Bacterial aggregation between microorganisms is one of the important defense mechanism against infection of UGT. Strongly autoaggregating strains A1, A3, B6, B3, B4, B9, B 10, and C3 showed a high autoaggregation percentage between 53 and 73% (Table 2).
 The percentages of hydrophobicity toward xylene are shown in (Table 2). Hydrophobic cell surface was demonstrated by high adherence to xylene, an apolar solvent. Our results demonstrated that all tested strains had lower hydrophobicity, where the highest percentage is represented by strain B6 with 34.06%.
 In the present study, no correlation was observed between cell surface hydrophobicity, ability to adhere to the vaginal cell and auto aggregation because all strains have low hydrophobicity while some isolates have a very strong ability of adhesion.
 The results of co aggregation showed that the highest S. aureus coaggregation percentage were obtained with strains B3 (54.02%) and B6 (52.12%) after 5 h of incubation, respectively. The co aggregation with E. coli is lower, where 38.05% is the highest percentage with strain B10 (Table 2).

4.4. Growth Inhibition of Vaginal and Urogenital Pathogens
 The results showed that our isolates have exhibited inhibitory activity against pathogenic strains with a significant difference (p<0.05) (Table 3). The smallest diameter of inhibition zone was 9 mm and the widest was 20.5 mm. Most strains present no inhibitory activity against C. albicans except strains B6, B1 B4, B10, C3 and C5. However the majority of lactobacilli have antagonistic activity against other pathogens.
 The results showed that Klebsiella.sp from urinary origin was inhibited by all cultures and their supernatants except supernatants of B1, C3, C5, and the culture of strain C3. Inhibition zones were shown to be produced by lactobacilli, as they disappeared when we used the supernatant of the most isolates with the exception for supernatant obtained from a number of strains such as B6 and B10 which have the ability of inhibiting vaginal and urinary pathogens.

4.5. Antibiotic Resistance
 The results showed that all strains were resistant to Ofloxacin, Gentamicin, and Ciprofloxacin; with almost all strains were sensitive to Trimethoprim-Sulphamethoxazole, Ampicillin, Erythromycin, Cefotaxime, Chloramphenicol, Tetracycline and Nitrofurantoin (Table 4).

4.6. Resistance of Vaginal Lactobacilli to Spermicides and Acid Production
 The results of the test are summarized in Table 5. We found that all strains show not an inhibition zone with the first two concentrations (0.1%, 0.2%), while when a concentration increases to 1%, inhibition zones between 3.5 and 7.5 mm could be measured for 12 strains, except B10 and B6 isolates. A significant inhibition was noted with 2.5% where zones are in majority > 6.5 mm. Through these results, we  conclude that most of the strains are resistant to the tested spermicide and are able to be co-administered with this spermicide for the treatment of vaginal infections.
 The results showed that all lactobacilli have an acidifying capacity with a significant difference between strains (p <0.05). Measuring the initial pH, and the pHs after 4 h, 6 h, and 24 h of incubation shows a decrease in the pH of the each sample and development of an acid environment with a pH~ 4.0 that contributes to the maintenance of the high redox potential which can protect the vagina against the invasion of the undesirable microorganisms (Table 5).

4.7. Identification of the Selected Isolates
 Two strains were retained as they display a high potential of probiotic profile, they were identified by 16S rDNA gene sequences and showed a high similarity to Lb. gasseri (B6: 99%) and to Lb. plantarum (B10: 99%). The GenBank accession numbers of the isolates B6 and B10 are assigned as KF739067 and KJ920199, respectively (Figure 3).

5. Discussion
 In our study, 14 isolates producing H2O2 were chosen to evaluate their technological and probiotic properties. The production of H2O2 by lactobacilli is considered as a non specific antimicrobial defense mechanism of the normal vaginal ecosystem (20) since women colonized by H2O2 producing strains had a protective effect against bacterial vaginosis (28-30). Aroutcheva et al (17) reported that 81.80% of the isolated strains produce H2O2 where 38.90% had a high H2O2 production. Several studies have demonstrated that Lactobacillus species can be administered orally or vaginally resulting in colonization in vagina, reduction in vaginal coliform counts and even reduction in UGTI (31). Therefore, we studied their viability under acidic, bile and SIF conditions since these abilities are important if the strains are recommended to be used orally as therapeutic probiotic. To exert their beneficial effect, lactobacilli need to resist human gastric transit conditions (22, 32). Our results showed that strains B6 and B10 have the highest tolerance to the gastro-intestinal condition. The results of a similar study have also shown the tolerance of vaginal Lb. fermentum SK5 to stimulated human gastrointestinal tract conditions (22).
 The ability to adhere to the epithelial cells and coaggregation ability are considered as important criterions for in vitro probiotic selection (15, 33, 34). Our results are in agreement with those of several authors (10, 14). Similarly, Strus et al (10) showed that from 111 isolates, 50% have a high adherence to vaginal mucus.
 The relationship between autoaggregation, adhesion and hydrophobicity has been reported by several authors (7, 22, 25, 35), however these correlations have not been found or reported by the others (36, 37). The obtained results were in agreement with those reported by Blakrishma (36); as he showed that there is no correlation between these parameters. In another study, Lp9 isolate has surface hydrophobicity of 37- 38%, suggesting its adhesiveness (37). Furthermore, adhesion and coaggregation of lactobacilli probiotic may inhibit the adherence of pathogens to the tissue receptors on the vaginal epithelial cells (12, 38).
 Our isolates present antagonistic activity against indicator bacteria. H2O2, lactic acid and other organic acids produced by LAB are frequently associated with this activity in vitro (15). The inhibition of urogenital infections increases the relevance of these wild strains for use in probiotic products (39). In another study different inhibitory ability between vaginal lactobacilli strains against C. albicans were obtained (40).
 It is very important to know the susceptibility of vaginal LAB to antibiotics in order to understand their behavior with respect to antibiotics in pharmaceutical preparations which were used for restoration of unbalanced vaginal flora (16). Probiotic strains’ resistance antibiotic may be an advantage in the case of co-administrations, but not all lactobacilli have intrinsic resistance (22).
      According to our results, the majority of strains are resistant to the tested spermicide. Pascual et al. (20) have studied the resistance and susceptibility of 62 strains of LAB to spermicide and have shown that 19.4% were resistant, whereas 80.6% were sensitive to nonoxynol-9.
 The result showed that Lb. gasseri B6 produces a high quantity of the lactic acid. It was reported that production of organic acid by LAB in the vaginal environment is responsible for the change in the pH and when the vaginal pH was lower than 4.5 it reflects a healthy vaginal ecosystem (9, 17, 41). According to Aslim and Kilic (19) lactobacilli produce lactic acid to maintain vaginal pH£4.5; this mechanism is used against pathogenic microorganisms.
5. Conclusions
 In conclusion, Lb. gasseri B6 and Lb. plantarum B10 possessed desirable technological and probiotic properties. These Lactobacillus isolates are the best candidate for preventing vaginal and urogenital infections.

Acknowledgments
 This work was supported by a grant allocated by the “Ministère de l’Enseignement Supérieur et de la Recherche Scientifique” of Algeria (Project code: F01720120001).