Determination of competition and adhesion abilities of lactic acid bacteria against gut pathogens in a whole-tissue model

Abstract

In an intestinal system with a balanced microbial diversity, lactic acid bacteria (LAB) are the key element which prevents the colonization and invasion of gut pathogens. Adhesion ability is important for the colonization and competition abilities of LAB. The aim of this study was to determine the adhesion and competition abilities of LAB by using a whole-tissue model. Indigenous strains were isolated from spontaneously fermented foods like cheese and pickles. The aggregation and competition abilities of the isolates were determined, as well as their resistance to gastrointestinal conditions. Four Lactobacillus strains and one Weissella strain were found to be highly competitive against three major gut pathogens, namely Clostridium difficile, Listeria monocytogenes and Salmonella Enteritidis. Tested strains decreased the number of pathogens to below their disease-causing levels. According to the results, the numbers of C. difficile and L. monocytogenes bacteria decreased by an average of 3 log, and their adhesion rates decreased by approximately 50%. However, the number of S. Enteritidis bacteria was decreased by only 1 log compared with its initial number. We thought that the weak effect on Salmonella was due to its possession of many virulence factors. The results showed that natural isolates from sources other than human specimens like the Weissella strain in this study were quite competent when compared with the human isolates in terms of their adhesion to intestines and resistance to gastrointestinal tract conditions. It was also revealed that a whole-tissue model with all-natural layers can be successfully used in adhesion and competition tests.

INTRODUCTION

The human gut is a highly complex ecosystem in which nutrients, microbiota, and host cells interact extensively [1, 2]. The total absorption area of the mucus membranes in the human digestion system is approximately 30–40 m² [3]. These membranes also have an important role in the immune system. This ecosystem has a diverse and abundant microbial community [4]. The total number of microorganisms in the human colon is about 10¹¹ to 10¹⁴/g [5, 6]. The microbial population in the colon is called the microbiota [7], and it is quite variable in terms of microbial species. The most common microbial genera in the human microbiota include Bacteroides, Lactobacillus, Enterococcus, Clostridium, Fusobacterium, Bifidobacterium, Eubacterium, Peptococcus, Peptostreptococcus, Escherichia, Veillonella and Candida [8]. In recent years, this rich microflora is believed to be the first line of defense of humans against pathogens. It is a well-known fact that the colonization of gut microbiota is critical for the normal development of host defense [9]. Intestines show a multilayer structure, with each layer having its own characteristics. These layers (from outside to inside) are the serosa, muscularis, submucosa and mucosa. In addition, the innermost mucosa consists of three layers, the muscularis mucosa, lamina propria and epithelium [8,9,10]. The muscularis mucosa, which is found in the deepest layer, contains plasma cells, lymphocytes, macrophages and fibroblasts. The lamina propria is located between the muscularis mucosa and epithelium and plays a part in the defense by fighting against invading microorganisms which pass through the epithelium layer [8]. Pathogens may produce toxins, invade the gut mucosa, and disseminate infections by translocation, thus cause systemic infections [2,3,4,5,6,7,8,9,10,11]. Studies in the last three decades have shown that some members of the microbiota play a key role as promoters of the intestinal defense against pathogenic microorganisms. In vitro studies have demonstrated the ability of such bacteria to inhibit the growth of several foodborne pathogens by competition and displacement [12]. Such strains, which are also thought to possess probiotic properties as a result of variety of tests, may have a beneficial effect, have no effect, or have adverse effects on human health. However, in order to call a strain a “probiotic”, the strain must be proven to have beneficial effects. Also, it should not be assumed that strains considered to be probiotics will be effective or safe under all conditions of use [13, 14]. Among the bacteria used for this purpose, the most encountered are Bifidobacterium (adolescentis, animalis, bifidum, breve and longum) and Lactobacillus (acidophilus, casei, fermentum, gasseri, johnsonii, paracasei, plantarum, rhamnosus and salivarius) [15, 16]. In particular, Lactobacillus species were determined to be more active against pathogens in the microbiota. They help to maintain the microbial balance by reducing the pathogenic microorganisms and thereby are associated with a good health status of the host [17]. Because of these health benefits of probiotic strains, many researchers have performed studies to determine these features using in vivo and in vitro models [14, 16, 18, 19]. Currently, in vitro studies focused on the relationship between the host and microbiota are dealing with the limited properties of transformed intestinal cancer cell lines like Caco-2 cells. Despite their myriad advantages, such cells fail to recapitulate the normal physiology and lineage development of the native intestinal epithelium [20]. The primary objective of the current work was to select natural strains to pass through the digestive system successfully. For this purpose, the resistance of isolates to pepsin, bile salts and pancreatin was determined. Subsequently, the adhesion ability of selected strains to the intestinal epithelium and their competitiveness against three major gut pathogens were determined. Sheep intestinal tissue was employed as the model for the measurement of their ability to bind to intestinal cells. The results showed that lactic acid bacteria can compete with pathogens in the complex natural environment of real intestinal tissue and prevent their attachment. Furthermore, more extensive results were obtained from studies with conventional cell culture models.

MATERIALS AND METHODS

Isolation and identification of lactic acid bacteria

Naturally fermented cheeses, pickles and fresh fruits obtained from farmers’ markets were used in order to obtain a wide variety of isolates belonging to different genera of lactic acid bacteria. All isolation and selection procedures were done according to the methods used by Schillinger and Lücke [21].

Resistance to gastrointestinal conditions

Pepsin resistance of isolates was tested in pH 2.0 and 3.0 phosphate buffered saline solution (PBS) containing 3 mg/mL pepsin enzyme (Sigma-Aldrich, Germany). Overnight cultures were inoculated (1%) into PBS-containing tubes and incubated at 37°C for 3 hr under anaerobic conditions. Samples were taken at 0, 1, and 3 hr, and the viability of each culture was determined by serial plating on MRS agar [22,23,24]. The percent cell survival was calculated according to the formula below (1) [25].

$\text{\% survival =}\frac{\text{log cfu of viable surviving cells}}{\text{log cfu of initial viable inoculated cells}}\times 100$ (1)

Bile salt resistance of isolates was determined in PBS solution containing 1% bile salts (Merck, Germany). Freshly prepared cultures were inoculated (1%) into PBS tubes and incubated at 37°C for 4 hr under anaerobic conditions. Aliquots of samples were taken at 0 and 4 hr. Serial dilutions were done on MRS agar plates [24, 26, 27]. The percent cell survival was calculated as mentioned above. Pancreatin resistance of isolates was determined in PBS solution (at pH 8.0) containing 1 mg/mL pancreatin (Sigma-Aldrich, Germany). Freshly prepared cultures were inoculated (1%) into PBS tubes and incubated at 37°C for 4 hr under anaerobic conditions. Aliquots of samples were taken at the beginning and at the end of 4 hr of incubation. Enumeration of the cultures was done by plating on MRS agar plates [28].

Aggregation assays

The autoaggregation and coaggregation abilities of cultures were determined by the method described by Kos et al. [29]. Initially, the cultures were grown in MRS broth for 18 hr at 30°C and centrifuged at 4,000 rpm for 15 min. The pellets were washed twice in PBS solution, and the optical densities of the cultures were adjusted to 0.6 absorbance at 600 nm in PBS by spectrophotometer. After adjustment of cell density, the cultures were incubated for 5 hr without the presence of any other microorganisms for the autoaggregation assay. The coaggregation test was done by incubating the same cultures together with Salmonella enterica subspecies enterica serovar Enteritidis (ATCC 13076) and Escherichia coli type 1 (LMG 2093). At the end of the 5 hr of incubation, the aggregation rate was calculated by using equations 2 and 3, as described by Collado et al. [30].

$\text{\% Autoaggregation = }\left(\text{1}-\left[{\text{A}}_{\text{t}}{\text{ / A}}_{\text{0}}\right]\right)\text{×}100$ ,(2)

where A₀ is the first optical density, and At is the optical density after 5 hr.

$\text{\% Coaggregation}=\frac{\left(\frac{\left(A_{pat}+A_{pro}\right)}{2}\right)-A_{mix}}{\frac{A_{pat+A_{pro}}}{2}}\times 100$ ,(3)

where A_pat and A_pro represent the A₆₀₀ of each culture individually, and A_mix> represents the combined absorbance of the two cultures at each time interval.

Adhesion and competition abilities of the cultures

Sheep intestine was used as the whole-tissue model in the determination of the ability of lactic acid bacteria to bind to intestinal epithelial cells and in the determination of the competition ability against gut pathogens. Fresh intestines of three healthy animals were collected from a commercial slaughterhouse and transferred to the laboratory within an hour. The integrity of the intestines was maintained during transport to the laboratory in order to protect them against external contamination.

The procedure for the preparation of intestinal tissues was adapted from those explained by Sellwood et al. [31] and Evans et al. [32]. Feed residues in the intestines were removed mechanically, and the inner sides of the tissues were prewashed with 0.15 M NaCl. Then the intestines were turned inside out and washed in a solution at pH 7.4 (0.096 M NaCl, 0.008 M KH₂PO₄, 0.0056 M Na₂HPO₄, 0.0015 M KCl and 0.01 M EDTA) for 20 minutes at room temperature. After the washing steps were completed, the tissues were immersed in a solution (0.096 M NaCl, 0.008 M KH₂PO₄, 0.0056 M Na₂HPO₄, 0.0015 M KCl and 0.3 M sucrose) to prevent them from drying. Intestine tissue was cut into 2 cm-wide ring-shaped parts with a sterile scalpel. Then, one side of the rings was cut in order to obtain flat rectangular pieces. Finally, the sizes of the pieces were adjusted to 2 cm².

Before adhesion tests, the pieces were placed in petri dishes with the inner epithelium side on top. Petri dishes were filled with 5 mL PBS suspension containing 3 × 10⁸ cfu/mL lactic acid bacteria and 15 mL Dulbecco’s Modified Eagle Medium (DMEM, Biowest, France). The dishes were incubated in a shaker at 37°C for 60 min at 40 rpm. At the end of incubation, the tissue was taken with forceps and washed 3 times in PBS solution. After washing, the intestine pieces were homogenized in 1/10 (w/v) PBS and serially diluted. The number of bacteria adhered to the intestinal tissue was determined by plating on MRS agar [31, 33, 34].

The same tissue model was used for the determination of competition ability of the lactic acid bacterial cultures against S. Enteritidis, Clostridium difficile and Listeria monocytogenes. Because the disease-causing levels of pathogens are generally between 10³ and 10⁶, the pathogenic bacteria concentration was adjusted to 1.5 × 10⁶ cfu/mL. The cultures of lactic acid bacteria (1.2 × 10⁹ cfu/mL) were inoculated simultaneously on the pieces of intestinal epithelium. Then the plates were incubated at 37°C for 60 min at 40 rpm on a shaker. At the end of incubation, the intestinal tissue was removed with forceps and washed 3 times in PBS. After washing, the intestinal fragments were homogenized in 1/10 (w/v) PBS and diluted serially [31, 33, 34]. At the end of the adhesion test, the adhesion rate was calculated by using equation 4. [12]. The test procedure for the competition tests is presented in Fig. 1.

Fig. 1.

Test procedure for the competition tests.

$\text{\% Adhesion =}\frac{\text{log cfu of viable surviving cells}}{\text{log cfu of initial viable inoculated cells}}\times 100$ (4)

Scanning electron microscopy

A scanning electron microscopy (SEM) technique was used for visualization of adhered cells on intestinal tissue samples. The preparation of intestinal samples was done as described previously. Sections of intestinal tissue (10 mm²) were partially dehydrated on filter paper for a short time. Images of the samples were taken by QUANTA 250 FEG scanning electron microscope at the YETEM (Energy Technologies Research Unit) center of Süleyman Demirel University.

Statistical analysis

All experiments were done in triplicate, and the data are presented as means ± standard deviation. The Minitab 16 statistical software (Minitab, Inc, State College, PA, USA) was used. Differences were determined by using one-way ANOVA with a significance level of p<0.05.

RESULTS

Identification and characterization

A total of 137 natural strains were isolated from various fermented foods. Of them, 77 were isolated from 44 different cheeses, 20 were isolated from fermented sausage (sucuk), 21 were isolated from 9 different pickles, 5 were isolated from lyophilized red pepper, and 17 were isolated from 16 different fresh fruits and herbs [35]. All isolates were Gram-positive, catalase-negative and non-spore-forming. The isolates were also tested for their ability to grow in 6.5% NaCl and to not grow at pH 9.6 in MRS medium. After preliminary eliminations, 69 isolates were selected, identified by PCR analysis, and then subjected to further analyses, such as analysis of their resistance to conditions mentioned in the Materials and Methods section. After all properties of isolates were determined, the 5 most successful strains were selected (Table 1) and used for the intestinal adhesion and competitive exclusion experiments.

Table 1. Identification of the five lactic acid bacterial isolates using 16S rRNA gene.

Isolate	Source	Name	BLAST accession no
DA4	Tulum cheese	Lactobacillus casei	MK161059
DA28	Tulum cheese	Weisella cibaria	MK161060
DA100	Pickle	Lactobacillus plantarum	MK161061
DA140	Tulum cheese	Lactobacillus plantarum	MK161062
DA263	Bitter orange	Lactobacillus coryniformis	MK161063

Resistance to gastrointestinal conditions

Before reaching the colon, probiotic bacteria are expected to survive through the human gastrointestinal system. In order to promote health effects, cells must be metabolically active when they arrive in the colon [36]. Microorganisms should be resistant to acids in the stomach, bile salts, and pancreatic fluids. Resistance to such conditions in the GI tract is important for the prediction of potential probiotic strains [25]. In the current study, the strains were tested for resistance to pepsin enzyme at two different pH values. While most strains showed high resistance when exposed to pepsin enzyme at pH 3 for 3 hr (Table 2), no or very limited resistance was found at pH 2.0 and thus is not presented. Weissella cibaria DA28 and Lactobacillus plantarum DA100 were the most stable isolates at pH 3. Similar results were obtained in the study of Teneva et al. [37], who recorded that the viability of Lactobacillus bulgaricus strains decreased by 6 logN against pepsin enzyme at pH 2.0 after 2 hr of incubation. Some other studies also emphasized that Lactobacillus strains lost their viability at pH 2.0 and that under more acidic conditions, 3 hours of incubation could significantly reduce their growth and viability [26, 38, 39].

Table 2. The viability results of microorganisms at pH 3.0 (log CFU/mL).

Microorganism	Viability counts at pH 3			% Viability
	0 hr	1 hr	3 hr	0–1 hr	0–3 hr
L. casei DA4	8.22 ± 0.36a	8.29 ± 0.26b	7.89 ± 0.19b	100.85 ± 1.82b	95.91 ± 1.55b
W. cibaria DA28	8.11 ± 0.25a	8.83 ± 0.15a	8.78 ± 0.13a	108.92 ± 1.41a	108.22 ± 1.35a
L. plantarum DA100	8.15 ± 0.24a	8.02 ± 0.15b	8.90 ± 0.12a	98.40 ± 1.40b	109.25 ± 1.30a
L. plantarum DA140	8.01 ± 0.33a	7.42 ± 0.23c	7.77 ± 0.17b	92.59 ± 1.68c	97.00 ± 1.48b
L. coryniformis DA263	7.27 ± 0.12b	5.87 ± 0.06d	5.73 ± 0.05c	80.73 ± 1.14d	78.85 ± 1.13c

Different letters in the same column indicate significant difference (p<0.05).

Another important factor which affects the survival of probiotic strains in the GI tract is bile salt resistance. Although the most widely used bile salt concentration is between 0.3 to 0.5% w/v in experiments simulating the human gastrointestinal system, in our study, all strains showed a high survival rate (96.26–99.16%) against 1% (w/v) ox bile after 4 and 24 hr of incubation (Table 2). This could be related to natural adaptation of the Lactobacillus casei DA4, W. cibaria DA28 and L. plantarum DA140 strains, which were isolated from tulum cheese with a high salt content. Sahadeva et al. [26] explained that the resistance of the bacteria to high bile salt concentrations (2%) could be a result of a stress adaptation mechanism developed by pre-exposure to acidic conditions.

Resistance to pancreatin is another important factor which determines the fate of microorganisms passing through the gastrointestinal system. Pancreatin is usually secreted in humans immediately after the stomach and it can result in a sharp change in pH from 3 to 8. In the current study, the strains were found to be resistant to pancreatin at a concentration of 1 mg/mL (Table 3). The minimum and maximum survival rates of isolates were 96.00% and 102.06%, respectively. Tokatlı et al. [25] determined the resistance of L. plantarum, Pediococcus ethanolidurans and Lactobacillus brevis strains isolated from mixed pickles to 1 mg/mL pancreatin, and their strains showed a maximum survival rate of 63%. Resistance to pepsin, bile salt and pancreatin provides good foresight into the selection of probiotic strains regarding their chance to survive against the metabolic activities in the human GI tract. Our strains were resistant to gastric conditions. The L. plantarum DA100 strain was the most resistant to the gastric conditions overall, and it particularly showed the highest resistance to against pepsin enzyme. Its resistance to acidic conditions could be explained by the natural adaptation of LAB strains of pickle origin to such conditions. Tokatlı et al. [25] similarly reported that Lactobacillus strains isolated from pickles had higher survival rates in acidic conditions compared with other intestinal hurdles.

Table 3. Results against 1 mg/mL pancreatin and %1 bile salt (log CFU/mL).

Microorganism	Viability counts against pancreatin		% Viability
	0 hr	4 hr	0–4 hr
L. casei DA4	8.23 ± 0.07b	8.17 ± 0.25ab	99.35 ± 1.79ab
W. cibaria DA28	8.47 ± 0.15a	8.32 ± 0.07a	98.46 ± 1.18a
L. plantarum DA100	8.39 ± 0.09ab	8.41 ± 0.32a	100.36 ± 2.07a
L. plantarum DA140	8.51 ± 0.11a	8.17 ± 0.18ab	96.08 ± 1.50ab
L. coryniformis DA263	7.76 ± 0.04c	7.92 ± 0.08b	102.02 ± 1.19b
Microorganism	Viability counts against bile salt		% Viability
	0 hr	4 hr	0–4 hr
L. casei DA4	7.46 ± 0.38a	7.33 ± 0.60ab	98.39 ± 3.96ab
W. cibaria DA28	7.34 ± 0.06a	7.34 ± 0.13a	100.00 ±1.36a
L. plantarum DA100	7.05 ± 0.40a	6.93 ± 0.25b	98.32 ± 1.77b
L. plantarum DA140	7.18 ± 0.10a	7.18 ± 0.16ab	100.07 ± 1.45ab
L. coryniformis DA263	7.36 ± 0.09a	7.18 ± 0.02ab	97.55 ± 1.71ab

Different letters in the same column indicate significant difference (p<0.05).

Auto- and coaggregation assays

Ferreira et al. [40] and Tuo et al. [41] reported that autoaggregation and coaggregation abilities can be used for preliminary selection of probiotic strains. The strains in this study showed autoaggregation and coaggregation properties at various levels (Table 4). The autoaggregation abilities of isolates ranged between 48.70% to 94.16% after 5 hr of incubation. While all strains retained their aggregation ability after 2 and 5 hr, there was a significant decrease in the aggregation ability of W. cibaria DA28. Anandharaj et al. [42] showed that the autoaggregation abilities of Lactobacillus and Weissella strains varied between 18% to 79% after 4 hr of incubation. However, Pessoa et al. [43] determined that the autoaggregation abilities of two L. plantarum strains, isolated from cocoa fermentation, were between 18.08 and 20.94% after 5 hr of incubation. The different levels of autoaggregation ability could be strain dependent, or physicochemical characteristics of the cell surface may affect the autoaggregation ability. Aggregation ability is related to cell adhesion properties, and it also influences the ability to survive and persist in the gastrointestinal system. Coaggregation ability may play an important role in the elimination of pathogens from the gastrointestinal system [41]. Lactobacillus strains can form a barrier by coaggregation that prevents colonization of pathogenic bacteria [40, 41]. In the present work, the tested strains showed coaggregation with E. coli type 1 and S. Enteritidis at various levels (Table 4). L. plantarum DA100 showed the highest coaggregation ability (59.49%) with E. coli type 1 after both 2 and 5 hr. The coaggregation ability of the same strain with S. Enteritidis was 57.30% after 2 hr and was 62.40% after 5 hr. Rajoka et al. [44] reported that L. rhamnosus showed the highest coaggregation capability with E. coli (49%), Salmonella (52%) and Staphylococcus (29%). Li et al. [45] determined the ability of 18 LAB strains to aggregate and adhere to Caco-2 intestinal cells. In the same study, although Enterococcus faecalis and Lactobacillus fermentum strains showed low rates of auto- and coaggregation with Salmonella spp., both strains showed high rates of adhesion to intestinal cells (104.13 and 84.43%, respectively). We observed a very similar result, with the W. cibaria DA28 strain having a rate of adhesion to intestinal cells of 84.70%.

Table 4. Coaggregation percentages of Lactic Acid Bacteria at the end of 2nd and 5th hours.

Microorganism	Coaggregation % E. coli		Coaggregation % Salmonella		Autoaggregation %		Adhesion %
	2 hr	5 hr	2 hr	5 hr	2 hr	5 hr
L. casei DA4	44.26 ± 4.6b	49.17 ± 2.3b	49.11 ± 2.4d	75.43 ± 0.0b	52.65 ± 5.4d	64.20 ± 0.0c	76.48 ± 0.38d
W. cibaria DA28	39.95 ± 5.0b	33.30 ± 7.5c	32.14 ± 1.2e	37.47 ± 3.3e	59.25 ± 5.8c	48.70 ± 1.8d	84.70 ± 0.40a
L. plantarum DA100	59.49 ± 3.5a	59.49 ± 3.5a	57.30 ± 2.1b	62.40 ± 2.3c	72.85 ± 1.2b	74.60 ± 1.2b	78.94 ± 0.16c
L. plantarum DA140	34.61 ± 1.8c	53.72 ± 10.7ab	77.41 ± 0.0a	83.87 ± 0.0a	75.00 ± 0.0b	75.00 ± 0.0b	79.87 ± 0.30b
L. coryniformis DA263	21.95 ± 0.2d	51.21 ± 0.1b	52.56 ± 1.8c	57.69 ± 1.8d	92.00 ± 0.0a	94.16 ± 1.1a	76.37 ± 0.34d

Mean values and standard deviations of three replicates are presented. Values with different letters within the same row differ significantly (p<0.05).

Adhesion and competition abilities of the cultures

Another important criterion for probiotic lactic acid bacteria is their ability to bind to the intestinal epithelium. Adhesion to the surface of intestinal epithelial cells is a priority for the colonization of probiotic bacteria in the GI system [34]. The results concerning the adhesion abilities of isolates with intestine cells are presented in Table 4. The adhesion abilities of cultures with intestinal cells were determined to be between 76.33% and 84.65% after 1 hr of incubation. The highest adhesion rate was achieved by W. cibaria DA28, which showed a rate of 84.65% at the end of the 1 hr incubation. Similarly, Kılıc et al. [46] studied the adhesion ability of 20 L. plantarum strains isolated from human fecal samples with respect to Caco-2 cells. They determined variable rates of adhesion to Caco-2 cells ranging between 48.75% and 75.18%. Sharma and Kanwar [47] also determined the adhesion abilities of lactic acid bacteria isolated from traditional fermented Himalayan to be between 2.45 and 9.55%. Similar studies showed that Lactobacillus spp. may have different levels of adhesion, indicating that the binding property may be strain specific [48, 49]. An SEM technique was used for the visualization of adhesion of strains (Fig. 2).

Fig. 2.

Scanning electron microscopy (SEM) image of the Lactobacillus plantarum DA140 strain in intestinal cells.

In the competition experiments, our strains showed different percentages of adherence to intestinal cells in the presence of pathogens. Furthermore, they managed to reduce both C. difficile and L. monocytogenes adhesion to intestinal epithelial cells significantly. Table 5 shows the competition results of the tested strains and pathogens. Competition rates were calculated by comparison of the initial and final numbers of pathogenic bacteria in the intestine in the presence of lactic acid bacteria. The results were evaluated according to the results obtained in previous studies [12, 50,51,52]. At the end of the competition assay, there was a 3 log (cfu/mL) average reduction of pathogens when compared with their initial numbers. Although the same strains adhered to epithelial cells at high rates, their competition abilities were limited against S. Enteritidis. On the other hand, Lau and Chye [12], Potočnjak et al. [52] and Feng et al. [53] reported that Lactobacillus strains prevented Salmonella spp. from sticking to Caco-2 cells. Our results showed that the employment of whole intestinal tissue in adhesion assays may provide different results from those of experiments done by using monolayered cell cultures. In vitro assessments at the cellular level alone may not be able to mimic the actual in-situ conditions in the gut ecosystem; however, they remain a practical method for pre-screening of potential strains.

Table 5. Competition results of the strains.

Microoroganisms	Lactic acid Bacteria			Pathogens
	Initial	Final	Adhesion %	Initial	Final	Adhesion %
Salmonella enterica subspecies enterica serovar Enteritidis
L. casei DA4	9.07 ± 0.01	7.20 ± 0.02	79.34 ± 0.04d	6.17 ± 0.01	4.84 ± 0.01	78.42 ± 0.02c
W. cibaria DA28	9.07 ± 0.01	7.55 ± 0.01	83.21 ± 0.02c	6.17 ± 0.01	4.95 ± 0.00	80.19 ± 0.04a
L. plantarum DA100	9.07 ± 0.01	7.74 ± 0.00	85.32 ± 0.01b	6.17 ± 0.01	4.84 ± 0.01	78.42 ± 0.02c
L. plantarum DA140	9.07 ± 0.01	7.95 ± 0.01	87.63 ± 0.02a	6.17 ± 0.01	4.69 ± 0.02	75.99 ± 0.01d
L. coryniformis DA263	9.07 ± 0.01	6.84 ± 0.00	75.41 ± 0.01e	6.17 ± 0.01	4.90 ± 0.01	79.40 ± 0.02b
Clostridium difficile
L. casei DA4	9.07 ± 0.01	7.77 ± 0.01	85.64 ± 0.02a	6.17 ± 0.01	3.45 ± 0.00	55.21 ± 0.02b
W. cibaria DA28	9.07 ± 0.01	7.69 ± 0.00	84.89 ± 0.01a	6.17 ± 0.01	3.00 ± 0.01	48.60 ± 0.02d
L. plantarum DA100	9.07 ± 0.01	7.77 ± 0.01	85.63 ± 0.03a	6.17 ± 0.01	2.89 ± 0.01	46.47 ± 0.06e
L. plantarum DA140	9.07 ± 0.01	7.63 ± 0.01	84.07 ± 0.06ab	6.17 ± 0.01	3.47 ± 0.02	56.20 ± 0.03a
L. coryniformis DA263	9.07 ± 0.01	7.35 ± 0.00	81.80 ± 2.29b	6.17 ± 0.01	3.05 ± 0.01	49.30 ± 0.02c
Listeria monocytogenes
L. casei DA4	9.07 ± 0.01	8.30 ± 0.00	91.49 ± 0.01a	6.17 ± 0.01	3.00 ± 0.01	48.49 ± 0.03d
W. cibaria DA28	9.07 ± 0.01	7.00 ± 0.00	77.14 ± 0.03d	6.17 ± 0.01	3.45 ± 0.02	55.08 ± 0.07b
L. plantarum DA100	9.07 ± 0.01	6.07 ± 0.01	66.90 ± 0.01e	6.17 ± 0.01	3.30 ± 0.01	53.44 ± 0.04c
L. plantarum DA140	9.07 ± 0.01	7.74 ± 0.01	85.30 ± 0.02b	6.17 ± 0.01	3.00 ± 0.01	48.60 ± 0.01d
L. coryniformis DA263	9.07 ± 0.01	7.44 ± 0.01	82.01 ± 0.01c	6.17 ± 0.01	4.00 ± 0.02	64.80 ± 0.02a

Different letters in the same column indicate significant difference (p<0.05).

DISCUSSION

The goal of this study was to investigate lactic acid bacteria isolated from different sources in terms of their abilities to adhere to intestinal epithelial cells and competition abilities, as well as to determine their resistance to various conditions in the gastrointestinal system. Survival of passage through the challenging conditions in the digestive tract and adhesion to human cells are important for strains to show their health effects in the human gut. The resistance to such conditions is crucial, as exposure to bile salts causes dissociation of the lipid bilayer of cell membranes and disruption of the integrity of cells [45,46,47,48,49,50,51,52,53,54]. Consumption of approximately 1.0 × 10⁶ to 1.0 × 10¹⁰ viable cells per day is generally accepted as the minimum limit required in order to gain satisfactory probiotic functions in the intestine [15, 55,56,57]. The strains in this study were able to pass through the gastrointestinal tract within this viability range. Although the strains showed a high resistance at pH 3.0, the same rate of viability was not achieved at pH 2.0. According to Liong and Shah [58], pH 3.0 is the actual limit in determining the acid resistance of probiotic cultures. Argyri et al. [59] stated that a pH value of 2.5 should be used for the selection of potential probiotic strains because lower pH values are not common in the human stomach. Selected strains showed strong resistance to 1% ox bile and 1 mg/mL pancreatin application after 4 hours of incubation. Our results demonstrated that some strains can survive or even grow under such a concentration of ox bile. Argyri et al. [59] proposed that the hydrolyzation process reduces the inhibitory effects of bile salt and that some components in foods may also play a protective role against bile salts. We believe that this resistance could be related to the activity of bile salt hydrolase enzymes produced by microorganisms. This resistance may also be related to the activity of the bile salt hydrolase enzymes produced by microorganisms.

Both L. plantarum strains in this study demonstrated high rates of auto- and coaggregation and a high percentage of adhesion to the mucosal surfaces of intestinal tissue. Compared with the Lactobacillus strains, W. cibaria showed the highest adhesion to the intestinal tissue, but its aggregation ability was the lowest among all strains. Tuo et al. [41] and Li et al. [45] obtained similar results in their studies, and this suggested that there may not be a positive correlation between a high auto/coaggregation ability and adhesion properties. The results of the adhesion experiments in our study also showed that some strains with low aggregation and coaggregation properties may show unexpected adhesion properties with real intestinal tissue, which is contrary to general opinion. Supporting our findings, various studies concluded that cell surface structures found on both intestinal and bacterial cells may play a role in adhesion to intestinal tissue. Studies also concluded that coaggregation, autoaggregation and adhesion abilities are strain specific. This specificity is determined by surface proteins, glycoproteins, teichoic acids and lipoteichoic acids, as well as S-layer proteins [41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60]. The strains used in our study showed high rates of adhesion to intestinal epithelial cells (Table 4).

Pathogens can only colonize and cause infection when they are resilient enough to pass through the conditions in the gastrointestinal system. Such pathogens may cause various gastrointestinal illnesses or infections resulting even in death. Many scientists believe that beneficial bacteria found in the intestines play important roles in preventing the attachment and proliferation of pathogens. Given this situation, colonization and persistence of probiotic bacteria have become important criteria in preventing or reducing pathogen adhesion to epithelial cells [12]. Five strains of lactic acid bacteria were tested against three major foodborne bacteria to determine their competition abilities. L. monocytogenes was chosen because of its ability to grow at refrigeration temperatures and because it is a cause of infections in a wide variety of foods, including ready-to-eat products [61]. Another pathogen chosen was C. difficile, which produces toxins and is considered to be a major agent of colitis and antibiotic-associated diarrhea. C. difficile infections generally develop when the stability of the native intestinal flora is impaired and colonization resistance decreases [62]. Various studies have shown that the indigenous microbiota can inhibit the persistence of C. difficile in the intestines and thus prevent its infections [63]. Similarly, our strains well competed with both and significantly prevented their adhesion. All strains reduced the number of both pathogens by 3 log (cfu/mL), reducing their numbers to below a disease-causing level.

Among Gram-negative pathogens, S. Enteritidis and Salmonella enterica serotype Typhimurium are the two major foodborne pathogens that cause acute gastroenteritis and systemic infections [64]. Unlike the other two pathogens mentioned above, our strains were less effective in preventing adhesion of S. Enteritidis. At the end of the competition tests, the number of Salmonella decreased by only 1 log compared with the initial number. This can be explained by the adhesion and invasion mechanisms of Salmonella, which are unlike those many other pathogenic bacteria. In addition, the penetration of Salmonella into the subepithelial layer occurs very rapidly [65]. This feature gives Salmonella an advantage in settling and multiplying in biological niches [66]. In our study, lactic acid bacteria were introduced to the intestinal tissue simultaneously with Salmonella for testing their competition ability, but they did not show the same success against Salmonella when compared with the other two pathogens. Zhao et al. [67] reported that insufficient immune response, impaired mucosal barrier function and lack of microbiota can increase susceptibility to salmonella infections. We thought that the weak effect on Salmonella was due to two reasons. The first was the use of real multilayered tissue instead of monolayered cells like Caco-2 in our study, and the second was the absence of flora already in the intestinal tissue. Considering the gastrointestinal risks arising from such pathogens, the colonization of lactic acid bacteria in the intestinal system to support antibiotiv treatments is of increases importance.

In conclusion, this study reveals the significant role of lactic acid bacteria in competition with pathogens and the prevention of their adhesion to intestinal structures. Also, the studied strains adhere to the requirements in probiotic guidelines [15, 18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68]. Furthermore, it was shown that employment of real intestinal tissues may provide distinct and even accurate results due to the presence of multilayers rather than single layered cell cultures.

Contrary to the common thinking, some strains of lactic acid bacteria that originate from sources other than the human body may also provide satisfactory properties in terms of their resistance to GI tract conditions and adhesion to intestinal structures. In addition, we determined that the Weissella genus, which has not been studied much, may show a variety of desired functions like other lactic acid bacteria, and thus its importance/functions in the intestine should be considered.

In this study, a comprehensive approach was used in the evaluation of bacteria to select effective probiotic strains. Hill et al. [15] indicated that probiotic microorganisms may show multiple mechanisms such as production of specific bioactives, vitamin synthesis and regulation of intestinal transit. The probiotic properties of all microorganisms can be divided into a variety of groups, rather than a single large cluster. While some properties are widespread among many strains, some properties are less common, and some can even be strain specific, like S-layer proteins. Although multiple mechanisms may exist in a single strain, no probiotic strain can be expected to possess all properties at the same time [15, 18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68]. Considering that each strain has different probiotic properties, to achieve effective results, approaches such as personalized/individualized strains in the microbiota of the gut or other regions are of increased importance .

Finally, the current study is a first step in this direction, and it has shown how important strain-specific properties can be in the selection of probiotics and the importance of the use of real intestine.

CONFLICT OF INTEREST

The authors declare that there are no conflicts of interest.

ETHICAL APPROVAL

This article does not contain any studies with human participants or animals performed by any of the authors.