Introduction

Helicobacter pylori strain has been first successfully cultivated by Warren and Marshall in 1983 [1]. This bacterium is now recognized as a major cause of chronic gastritis and peptic ulcer and is a risk factor for gastric carcinoma [2]. H. pylori strains were found in the human stomach of about half of the world populations. The prevalence of infection of developing countries was higher (> 80%) than the developed ones (< 40%) [3]. Currently, the standard triple and sequential therapies have been widely used for H. pylori eradication. These therapies required proton pump inhibitor (PPI) plus antibiotics including amoxicillin, metronidazole, tinidazole, and clarithromycin [4]. However, antibiotic-based therapy had the significant effect on the composition of intestinal microbiota, would not be cost-effective, causes side effects, and in particular, encouraging widespread antibiotic resistance [5, 6]. Hence, the alternative or adjunctive therapy has been proposed.

Probiotics are defined as “live microorganisms, which when administered in adequate amounts, confer a health benefit to the host” [7]. The use of probiotics to improve intestinal health of humans has been purposed for many years [8]. Lactic acid bacteria (LAB; Lactobacillus, Enterococcus, etc.) and Bifidobacterium are known to be commonly used as probiotics [9]. The studies involving the antimicrobial activity against H. pylori in vitro and in vivo of probiotic LAB such as Lb. acidophilus LB [10], Lactobacillus johnsonii La1 [11], Lb. casei Shirota [12], Lb. rhamnosus GG [13], and Lb. gasseri OLL2716 have been reported [14].

LAB could inhibit the growth of H. pylori by production of organic acids, competitive adhesion, and reduction of cytokines [15, 16]. Moreover, antimicrobial peptide or bacteriocin secretion is one of the most interesting mechanisms which suppress the growth of pathogenic bacteria because it acts specifically to target bacteria only at nonamolar level [17]. Several investigations found that antimicrobial peptides produced by LAB exhibited antimicrobial activity against H. pylori; however, there are rarely [18,19,20,21,22]. Furthermore, there are only few reports on screening of LAB isolated from Thai traditional food products that have the ability to combat H. pylori clinical isolates and no one has studied on bacteriocin. Thus, the objective of this study was to isolate and identify LAB presented in Khanom-jeen, a Thai traditional fermented rice noodle, and to screen their ability to inhibit the growth of H. pylori in vitro.

Materials and Methods

Sample Collection, Isolation, Indicator Strains, and Cultivation

Thai fermented rice noodle, also called Khanom-jeen, was produced by a traditional lactic acid fermentation of rice flour [23]. Sixteen samples were collected from traditional Khanom-jeen manufacturer in Songkhla and Nakhonratchasima provinces, Thailand (Table 1). LAB were isolated by pour plate technique using MRS [24] agar containing 0.3% CaCO3 and incubated at 30 °C for 2–3 days. The number of viable LAB found in samples was determined by counting with the colony-forming unit. Colonies surrounded by a clear zone were selected for purification. Pure cultures were maintained in MRS broth with 20% glycerol at − 80 °C and lyophilized.

Table 1 Source, LAB count, strain number and number of strains isolated from fermented rice noodle
Full size table

H. pylori strains 2649, MS83, and BK364 isolated from a patient suffering from gastritis, peptic ulcer diseases, and gastric cancer, respectively were used as the indicator strains. These three clinical H. pylori strains were obtained from the Faculty of Medicine, Thammasat University, Pathum Thani, Thailand. H. pylori strains were cultivated on Columbia blood agar supplemented with 7% sheep blood and incubated at 37 °C for 3 days under microaerobic conditions using a gas-generating kit (Anaero Pack-MicroAero, Mitsubishi Gas Chemical, Japan). They were subcultured twice on the same medium before experimental use.

Identification of Strains

Phenotypic Characterization

Phenotypic characteristics of LAB strains were performed as described previously [25]. Cell morphology and colony appearance were determined after cultivating the strains on MRS agar plate at 30 °C for 48 h. Catalase and nitrate reduction, gas production, slime formation, hydrolysis of arginine and starch, acid production from carbohydrates, growth at temperature of 45 °C and at various pH values (3.5, 9.0, and 9.6), and NaCl tolerance (6% NaCl) were investigated. Meso-diaminopimelic acid (DAP) was determined as reported previously [26]. Lactic acid isomers were analyzed by the enzymatic method using D- and L-lactate dehydrogenases (Boehringer, Germany) [27].

Statistical Analysis

The LAB strains were grouped based on their phenotypic characteristics by cluster analysis. Hierarchical cluster analysis was performed by using SPSS for Windows version 16.0.

Genotypic Characterization

The 16S rRNA gene was amplified by polymerase chain reaction (PCR) using the primers 20F (5′-AGTTTGATCCTGGCTC-3′) and 1530R (5′-AAGGAGGTGAT CCAGCC-3′). Amplification reaction was carried out in 50 μl volume, using Taq DNA polymerase and buffer system (Vivantis Technologies, Malaysia). The final PCR mixture consisted of 10× ViBuffer A (without MgCl2; 50 mM KCl, 50 mM Tris-HCl (pH 9.1 at 20 °C) and 0.1% Triton™ X-100), 5 mM MgCl2, 0.2 μM concentration of each deoxynucleoside triphosphate, 0.2 μM concentration of each primer, one unit of Taq DNA polymerase, and 50 ng of template DNA. Amplification was carried out in a thermocycler (C1000 Touch™, Bio-Rad Laboratories Inc., Berkeley, California) with the following cycling program: initial denaturation at 94 °C for 5 min, followed by 40 cycles of 94 °C for 2 min, annealing at 55 °C for 1 min, and extension at 72 °C for 2 min, and a final extension step at 72 °C for 5 min [28]. Three microliters of the PCR product was electrophoresed for 40 min at 100 V on a 1.5% (w/v) agarose gel in 1× TAE buffer (40 mM Tris base (pH 7.6), 20 mM acetic acid, 1 mM EDTA, pH 8.0). A 10-kb DNA molecular weight marker (Vivantis Technologies, Malaysia) was included as standard for the calculation of the fragments. After staining with ethidium bromide (0.5 μg ml−1) for 5 min, the fingerprint patterns were visualized under ultraviolet light followed by digital capture using the Gel Documentation system (Bio-Rad Laboratories Inc., Berkeley, California). PCR products were sequenced (Macrogen, Seoul, Korea) by using the universal primers 27F (5′-AGAGTTTGATCMTGGCTCAG-3′) and 1492R (5′-TACGGYTACCTTGTTACGACTT-3′) [29]. The 16S rRNA gene sequences were multiple-aligned with the selected sequences obtained from the EzTaxon-e database [30] by using the CLUSTAL_X program [31]. Gaps and ambiguous bases were eliminated before reconstructing the phylogenetic tree by the neighbor-joining method [32] in the MEGA 6 program [33]. Confidence values of the branches of phylogenetic tree were evaluated by using the bootstrap resampling method with 1000 replications [34].

Screening of Anti-H. pylori Activity

Cells of LAB strain cultivated in MRS broth at 30 °C for 18 h were removed by centrifugation at 14,000 rpm for 10 min at 4 °C. The supernatant was collected and then filtered through a 0.2-μm pore-size filter. The antimicrobial activity was tested by the spot-on-lawn method as previously described with minor modification [35]. This assay was performed by spotting 10 μl of cell-free supernatant (CFS) onto a double layer including 5 ml of brain heart infusion (BHI) agar (0.7% agar) containing 107 cfu/ml of H. pylori culture which overlaid on the BHI agar medium (1.5% agar). After incubation under microaerobic conditions at 37 °C for 72 h, the inhibition zone (mm) on H. pylori lawn was measured.

Characterization of Antimicrobial Compound

LAB strain which showed a highest anti-H. pylori activity was selected to determine the nature of compounds responsible for antimicrobial activity. To exclude the effect of organic acids, CFS collected from the selected strain was neutralized to pH 6.5 by using 3 M Tris base. In addition, CFS was treated with different enzymes including pepsin, trypsin, α-chymotrypsin, proteinase K, and catalase at final concentration of 1 mg/ml. After incubation for 3 h at optimal pH and temperature for each specific enzyme, the antimicrobial activity of enzyme-treated samples was tested by critical dilution spot-on-lawn assay. The antimicrobial activity was expressed as arbitrary activity units (AU) per milliliters which were defined as the reciprocal of the highest dilution causing a clear zone of inhibition in the indicator lawn according to the following formula [35];

$$ \mathrm{The}\ \mathrm{antibacterial}\ \mathrm{activity}\ \left(\mathrm{AU}/\mathrm{ml}\right)={2}^N\times \kern0.5em 100 $$

where, AU is the arbitrary unit and N is the highest two-fold serial dilution showing an inhibition zone of the bacterial indicator strain.

Probiotic Properties Assays

Acid Tolerance

Acid tolerance was determined as a minor modification of the previous report [36]. Single colony of LAB strain was anaerobically cultivated in MRS broth at 37 °C for 18 h. Cells (4.0 × 109 cfu/ml) were centrifuged (8000×g for 5 min at 4 °C), washed twice with sterile normal saline (0.85% NaCl), and re-suspended in 10 ml of MRS broth. Cell suspension (1%) was inoculated into 10 ml of MRS with adjusting pH to 2.0 and 3.0 using 6 N hydrochloric acid (HCl). The cultures were then anaerobically incubated at 37 °C. At 0, 1.5, and 3 h of incubation, viable cell counts were determined on a MRS agar plate using the spread plate technique.

Bile Tolerance

Bile tolerance was evaluated with some modifications [37]. One hundred microliters of cell suspension was inoculated into 10 ml of MRS broth supplemented with 0.3 and 0.5% of oxgall (Sigma-Aldrich, MD, USA). The cultures were then anaerobically incubated at 37 °C. After incubation at 0, 3, and 6 h, viable cell counts were examined on a MRS agar plate using the spread plate technique.

Results

The total number of LAB found in samples collected from Songkhla and Nakhonratchasima was ranged from 3.5 × 106 to 4.0 × 107 cfu/g and 1.2 × 104 to 4.0 × 1014 cfu/g, respectively. Thirty-two strains of LAB were isolated from 16 samples of the fermented rice noodle as shown in Table 1.

Identification of Strains

From the total 32 LAB strains, they were 14 rods, 6 coccobacilli, and 12 cocci and they appeared singly, in pair or in chain. All LAB strains were Gram-positive, catalase-negative, non-motile, non-spore forming bacteria. They were divided into nine groups based on their phenotypic characteristics as shown in Fig. 1. Based on the 16S rRNA gene sequence, LAB strains belonged to the genus Lactobacillus, Weissella, Enterococcus, Lactococcus, and Leuconostoc (Fig. 2).

Fig. 1
figure 1

Dendrogram using average linkage (between groups) showing the hierarchical cluster of the LAB strains based on their phenotypic characteristics

Full size image
Fig. 2
figure 2

Neighbor-joining tree based on 16S rRNA gene sequences showing the relationship among the representative strains in group 1 to group 9 and related species. Bootstrap percentages > 50%, based on 1000 replications, are given at nodes. Bar 0.01 substitutions per nucleotide position

Full size image

Group 1 included two coccal strains (P4-6 and P4-6A). All produced gas from glucose. They grew at pH 9.0 and 9.6 and in 6% NaCl but did not grow at 45 °C and pH 3.5. The strains did not contain meso-DAP in the cell wall. They produced D-lactic acid. All strains could ferment tested carbohydrates but did not ferment glycerol, lactose, and raffinose (Table 2). The strains P4-6 and P4-6A showed 99.64% 16S rRNA gene sequence similarity to Ln. pseudomesenteroides JCM 9696T (Fig. 2).

Table 2 Characteristics of LAB strains isolated from fermented rice noodle
Full size table

Group 2 contained three coccal strains (P42-1, P42-1A, and P41-3). These strains produced gas from glucose. All strains grew in 6% NaCl but did not grow at 45 °C. Variable pH growth was found from 3.5 to 9.6. They did not have meso-DAP in the cell wall. Tested strains produced D-lactic acid. All strains produced acid from tested carbohydrates except for glycerol, maltose, D-mannitol, ribose, and xylose (Table 2). The strains P42-1, P41-3, and P42-1A were closely related to Ln. citreum ATCC 49370T with 99.78, 99.86, and 99.85% 16S rRNA gene sequence similarity, respectively (Fig. 2).

Group 3 consisted of three coccal strains (P3-1, P4-1, and P32-9). All strains produced gas from glucose. They grew at pH 3.5 and 9.0 and in 6% NaCl. Variable growth was observed at 45 °C and pH 9.6. These strains did not have meso-DAP in the cell wall. They produced only D-lactic acid. All strains fermented D-amygdalin, L-arabinose, cellobiose, D-galactose, gluconate, maltose, D-mannose, melibiose, raffinose, salicin, sucrose, trehalose, and xylose (Table 2). Strains P3-1, P4-1, and P32-9 were closely related to Ln. lactis DSM 20202T with 99.78% 16S rRNA gene sequence similarity (Fig. 2).

Group 4 included two rod-shaped strains (P2-4 and HB3-4). They did not produce gas from glucose. These strains grew at pH 3.5 and 9.0, 45 °C, and in 6% NaCl but did not grow at pH 9.6. They did not contain meso-DAP in the cell wall. They produced only L-lactic acid. All strains could produce acid from tested carbohydrate except for glycerol and rhamnose (Table 2). Strains P2-4and HB3-4 were closely related to Lb. paracasei subsp. tolerans JCM 1171T with 99.57 and 100% similarity based on their 16S rRNA gene sequence (Fig. 2).

Group 5 contained three coccal strains (N21-9, SC3-4, N11-7). All strains could hydrolyze arginine. They grew at 45 °C, pH 9.0, and in 6%NaCl but did not grow at pH 3.5. These strains did not contain meso-DAP in the cell wall and tested strains produced L-lactic acid. All strains produced acid from tested carbohydrates but did not produce acid from gluconate, glycerol, α-methyl-D-glucoside, and xylose. Variable fermentation was observed from D-mannitol, rhamnose, ribose, and sucrose (Table 2). The strains (N21-9, SC3-4, and N11-7) were closely related to E. faecium CGMCC 1.2136T with 100, 99.93, and 99.93% 16S rRNA gene sequence similarity (Fig. 2).

Group 6 contained one coccal strain (P43-1A). It could hydrolyze arginine. The strain grew at pH 9.0 and 9.6 and in 6% NaCl but did not grow at 45 °C and pH 3.5. It did not have meso-DAP in the cell wall. This strain produced L-lactic acid. It produced acid from D-amygdalin, L-arabinose, cellobiose, D-galactose, lactose, maltose, D-mannitol, D-mannose, α-methyl-D-glucoside, salicin, sucrose, and trehalose (Table 2). This strain was closely related to Lc. lactis subsp. lactis JCM 5805T with 99.93% 16S rRNA gene sequence similarity (Fig. 2).

Group 7 consisted of six coccobacillary strains (P1-1, P1-2A, P23-13, P1-3, P2-2A, and P4-3). All strains produced gas from glucose. They did not grow at pH 3.5. Variable growth was observed at 45 °C, pH 9.0 and 9.6, and in 6% NaCl. Tested strains did not have meso-DAP in cell wall and produced DL-lactic acid. All strains produced acid from D-galactose. Variable acid production was found in gluconate, maltose, D-mannose, ribose, salicin, sucrose, and xylose (Table 2). The strains P1-1, P1-2A, P4-3, P23-13, P1-3, and P2-2A were closely related to W. confusa JCM 1093T with similarity 99.78, 100, 100, 100, 99.64, and 99.64% based on their 16S rRNA gene sequences (Fig. 2).

Group 8 included seven rod-shaped strains (P23-10, P22-7, HB3-3, P22-9, P43-01, N11-5, and P22-7A). All strains produced gas from glucose. Some strains could hydrolyze arginine. They grew at pH 3.5 but did not grow at pH 9.0 and 9.6. Variable growth was observed at 45 °C and in 6% NaCl. These strains did not contain meso-DAP in cell wall. They produced DL-lactic acid. All strains did not produce acid from D-amygdalin, cellobiose, rhamnose, salicin, and sorbitol. Variable acid production was found in L-arabinose, D-galactose, gluconate, glycerol, lactose, maltose, D-mannitol, D-mannose, melibiose, α-methyl-D-glucoside, raffinose, ribose, sucrose, trehalose, and xylose (Table 2). Based on 16S rRNA gene sequences, these strains (P23-10, P22-7, HB3-3, P22-9, P43-01, N11-5, and P22-7A) were closely related to Lb. fermentum CECT 562T with 100, 99.93, 99.78, 99.57, 99.86, 99.64, and 99.64% similarity, respectively (Fig. 2).

Group 9 comprised five rod-shaped strains (SB1-2, P42-4, P42-4A, P43-1, and SC2-1). They did not produce gas from glucose. All strains grew at pH 3.5 and in 6% NaCl. Variable growth was found at 45 °C and at pH 9.0 and 9.6. These strains contained meso-DAP in the cell wall and produced DL-lactic acid. All strains produced acid from tested carbohydrates but did not produce acid from α-methyl-D-glucoside and rhamnose. Variable acid production was found in L-arabinose, D-galactose, glycerol, lactose, melibiose, raffinose, sorbitol, and xylose (Table 2). Strain SB1-2 was closely related to Lb. pentosus DSM 20314T with 99.93% similarity of 16S rRNA gene sequence. Strains P42-4, P42-4A, P43-1, and SC2-1 were closely related to Lb. plantarum subsp. plantarum ATCC 14917T with 100% similarity of 16S rRNA gene sequence (Fig. 2).

Antimicrobial Activity Against H. pylori Strains

Thirty-two strains of LAB isolated from the fermented rice noodle were tested for their anti-H. pylori activity by the spot-on-lawn method. Fifteen LAB strains exhibited different levels of H. pylori inhibition as shown in Table 3. Eleven strains of LAB (P42-1, P42-1A, P32-9, N21-9, P4-3, P23-13, HB3-3, P22-7, P22-7A, P22-9, and P43-01) could suppress the growth of H. pylori BK364, which were in the range of 6.0 to 11.5 mm. Four strains (HB3-4, P43-1A, P43-01, and SB1-2) displayed the inhibition against H. pylori MS83, which were in the range of 5.8 to 9.0 mm. Only one strain (SC3-4) showed the suppression against H. pylori 2649 (5.8 mm). H. pylori strain BK364 was the most sensitive whereas strain 2649 was less sensitive to CFSs collected from LAB. CFS from strain P43-01 showed the highest activity which was able to inhibit the growth of two clinical strains of H. pylori (strains MS83 and BK364) whereas CFSs from other strains inhibited only one strain of H. pylori. Strain P43-01 was selected and identified. Based on 16S rRNA gene sequence similarity, the selected strain was identified as Lb. fermentum P43-01 as describe previously.

Table 3 Anti-H. pylori activity of CFSs collected from LAB
Full size table

Characterization of Anti-H. pylori Substances

Effect of neutralization, catalase, and protease treatment on inhibitory activity of CFS obtained from Lb. fermentum P43-01 was presented as Table 4. The results showed that the anti-H. pylori activity of untreated CFS against H. pylori MS83 and BK364 was 200 AU/ml, and after CFS neutralization, their antimicrobial activity did not change. In addition, the inhibitory activity of CFS against two clinical H. pylori strains did not alter when treated with catalase. However, CFS was inactivated after incubation with α-chymotrypsin and reduced from 200 to 100 AU/ml when treated with pepsin. Trypsin and proteinase K were not effective on the anti-H. pylori activity of CFS. These results indicated that the inhibitory substance might be the proteinaceous compounds or bacteriocins.

Table 4 Inhibitory activity against H. pylori of CFS collected from Lb. fermentum P43-01 after neutralization and enzyme treatments
Full size table

Probiotic Properties of Selected Strain

Survival of Lb. fermentum P43-01 after exposures to low pH (2.0 and 3.0) and bile salts (0.3 and 0.5%) was shown in Table 5. Decreases in viable cells were observed through the incubation period at pH 2.0. The survival of a selected strain was decreased about 2 log cycles after incubation at pH 2.0 for 3 h. At pH 3.0, the viability of a selected strain was constant at 7.6 log10 cfu/ml during 3 h of incubation. A strain was able to grow in the presence of 0.3 and 0.5% of oxgall. There were increases in viable cell count around 1.2 and 1 log cycles after exposures to 0.3 and 0.5% of oxgall for 6 h.

Table 5 Acid and bile tolerances of Lb. fermentum P43-01
Full size table

Discussion

This study focused on isolation and identification of lactic acid bacteria presented in fermented rice noodle produced in Thailand. The number of LAB presented in fermented rice noodle samples obtained from two producers in Thailand ranged from 104 to 1014 cfu/g (Table 1). Samples collected from Songkhla had lower LAB counts than that obtained from Nakhonratchasima. LAB that involved in Khanom-jeen process were Lb. plantarum, Lb. fermentum, and Pediococcus acidilactici [23]. In this study, the differentiation of LAB in each group was carried out based on their phenotypic characteristics such as cell form; gas production; arginine hydrolysis; growth at 45 °C, at pH 3.5, 9.0, and 9.6, and in 6% NaCl; cell wall type; lactic acid isomer; and acid production from carbohydrates (Table 2). The coccal strains belonged to the genera Leuconostoc, Enterococcus, Lactococcus, and Weissella whereas rod-shaped strains belonged to the genus Lactobacillus based on their 16S rRNA gene sequence and phylogenetic analysis (Fig. 2) [38]. Two heterofermentative cocci that produced D-lactic acid in group 1 were identified as Ln. pseudomesenteroides while three heterofermentative cocci in group 2 were identified as Ln. citreum and three heterofermentative cocci in group 3 were identified as Ln. lactis [39]. Two heterofermentative rods in group 4 which did not have meso-DAP in the cell wall were identified as Lb. paracasei subsp. tolerans [40]. Three homofermentative cocci in group 5 were identified as E. faecium while one homofermentative coccus in group 6 was Lc. lactis subsp. lactis [41, 42]. Six heterofermentative coccobacilli in group 7 were identified as W. confusa [43]. Seven heterofermentative rods in group 8 were identified as Lb. fermentum while five homofermentative rods in group 9 that contained meso-DAP were identified as Lb. plantarum subsp. plantarum and Lb. pentosus [40].

H. pylori infection remains a worldwide spread disease with a definite morbidity and mortality. Antibiotic-based therapy has been applied to treat patients who were infected with H. pylori. Unfortunately, no current therapy is able to achieve a 100% success rate due to the antibiotic resistance of H. pylori [4, 44]. Thus, the need for alternative or adjunctive therapy such as the use of probiotics for the H. pylori eradication has claimed by many researches [2, 45].

This study reported the anti-H. pylori activity of LAB strains isolated from Khanom-jeen. Fifteen LAB strains exhibited different spectrum of antimicrobial activity against clinical strains of H. pylori. The susceptibility of clinical H. pylori isolates to CFSs collected from LAB seemed to be strain-dependent sensitivity. H. pylori 2649 isolated from patient with gastritis was more tolerant than strains MS83 and BK364. It was well known that antimicrobial metabolites produced by LAB included organic acids (mainly lactic and acetic acid), diacetyl, hydrogen peroxide, and bacteriocin [46]. The broad antimicrobial spectrum that showed inhibition on H. pylori strains MS83 and BK364 was found in CFS produced by Lb. fermentum P43-01. Although, CFS collected from selected strain was neutralized by adjusting pH to 6.5, but the anti-H. pylori activity was still stable. Moreover, catalase treatment did not change its inhibitory activity. The results indicated that antimicrobial activity did not form organic acids and H2O2. However, the inhibitory activity of CFS from selected strains was destroyed by α-chymotrypsin and reduced by pepsin. Antimicrobial metabolites secreted by Lb. fermentum P43-01 which suppressed the growth of clinical isolates of H. pylori might be bacteriocins.

Anti-H. pylori activity from LAB has been reported in many investigations. In addition to lactic acid and low pH, bacteriocin-like substances produced by LAB seem to be an important factor which exerts the inhibition of H. pylori growth. E. faecium TM39 produced bacteriocin-like substance with antagonistic activity against H. pylori, in addition to lactic acid and pH [18]. Bacteriocin secreted by W. confusa PL9001 was the major substance which suppressed the growth of H. pylori [19]. Six strains of Bifidobacterium from human feces showed inhibitory activity against clinical H. pylori isolates by producing heat-stable proteinaceous compound [20]. Bulgaricin BB18 secreted by Lb. bulgaricus BB18 showed the inhibition against H. pylori [21]. Lb. brevis BK11 and E. faecalis BK61 produced bacteriocin which displayed strong inhibition on the growth of H. pylori [47]. Among bacteriocins tested (nisin A; lacticins A164, BH5, JW3, and NK24; pediocin PO2; and leucocin K), lacticins A164 and BH5 produced by Lc. lactis A164 and Lc. lactis BH5 showed the strongest inhibition on the growth of H. pylori [48]. The effects of antimicrobial peptides on the growth of H. pylori have been less infrequently reported when compared with those of other pathogenic bacteria such as Listeria monocytogenes, Staphylococcus aureus, Clostridium difficile, Esherichia coli, and Salmonella spp. More researches that involved antimicrobial peptides with anti-H. pylori activity which produced by LAB were thus needed. However, this is the first report regarding the antimicrobial peptides produced by Lb. fermentum P43-01 which showed the inhibition against H. pylori. To evaluate their possible use in the treatment of H. pylori infection, bacteriocin purification and their characterization should be performed in further studies.

Prior to passing through the intestinal tract, probiotics must survive under severe conditions in the stomach with very low pH. Moreover, they have to resist bile acids when they reach the intestine [49]. According to acid and bile tolerance, the viable cells of selected strain were substantially declined during incubation at pH 2.0, but the survived cells was remained at about 5.7 log cycles. A strain showed the resistance when it was exposed to pH 3.0. In addition, it could grow in the medium supplemented with 0.3 and 0.5% of bile salts. In agreement with the study of [50], most Lb. fermentum isolates exhibited high tolerance to low pH as well as high bile concentration with minimum cell count loss. These results revealed that Lb. fermentum P43-01 displayed an ability to tolerate an acidic conditions and bile salts, thus, it could be a potential probiotic candidate.

Conclusions

This study provides the distribution of Lactobacillus, Weissella, Enterococcus, Lactococcus, and Leuconostoc presented in Khanom-jeen, a Thai fermented rice noodle. Fifteen strains exhibited the antimicrobial activity against clinical H. pylori isolates. Lb. fermentum P43-01 is a candidate strain to use as probiotics for alternative or adjunctive therapy in patient infected with H. pylori. This strain could resist bile acids and produce antimicrobial peptides against H. pylori in vitro.