Probiotic potential of ‘lactobacilli’ isolated from traditionally fermented Ethiopian honey wine (Tej)

Abstract

Probiotics are live microorganisms with health benefits. Lactic acid bacteria (LAB) are among the well-known probiotic formulations in various products intended for human or animal consumption. Prior researches on Tej have thus far looked into its physicochemical compositions, microbial load and starter cultured development. This work on screening of probiotic from Tej is novel because there hasn’t been any report on screening of probiotics for improvement of the beverage. The aim of this research is therefore to investigate the probiotic properties of LAB isolated from Tej. Samples were collected from Addis Ababa, and LAB were identified following morphological, physiological, and Matrix Assisted Laser Desorption Ionization Time of Flight Mass spectrometer (MALDI-TOF). A total of 300 isolates were purified and characterized among which 280 of them were LAB. All the 280, LAB isolates were screened for their antibacterial properties out of which 200 showed antibacterial activities, 158 were tolerant to low pH values (pH 2 and 3), 134 survived the bile concentrations of 0.3%, 0.4%, and 1%, and 64 had better adhesion properties. The LAB isolates with adhesion properties were tested for their safety (hemolytic activity) and 18 isolates that had no hemolytic activity were used for further analysis. Each probiotic feature was assessed using standard protocols. The percentage survival (mean ± SD) of the LAB ranged from 5.04 ± 0.05 to 65.47 ± 0.08 and 11.05 ± 0.06 to 70.48 ± 0.04% at pH 2.0 and 3.0, respectively. A statistically significant difference (p < 0.05) was observed in the mean percentage survival at pH 2 and pH 3. The percentage survival (mean ± SD) of the LAB isolates to bile concentrations of 0.3%, 0.5%, and 1% ranged from 21.74 ± 0.08 to 92.48 ± 0.03, 23.84 ± 0.05 to 72.02 ± 0.03 and 30.22 ± 0.16 to 65.37 ± 0.05, respectively. For 77% of the isolates the mean survival was higher at the lowest bile concentration compared to the highest bile concentration and the differences were statistically significant (p < 0.05). The mean cell surface hydrophobicity of the LAB was in the range of 24.11 ± 0.85 to 76.71 ± 0.66%. The antioxidant ability of the LAB based on DPPH (2,2-Diphenyl-1-picrylhydrazyl) radical scavenging ranged from 0.4 to 68.9%, and cholesterol assimilation from 1.6 to 19.4%. Variabilities in the patterns of resistance to antibiotics were observed among the strains of LAB with the most frequent resistance to vancomycin. The LAB isolates with in vitro probiotic potentials that didn’t exhibit hemolytic activities were identified as Levilactobacillus brevis (ST1, ST2, ST17), Lecticaseibacillus paracasei (ST3, ST8, ST10, ST11, ST12, ST13, ST14, ST15, ST16, ST18), Lentilactobacillus hilgardii (ST4, ST5, ST6) and Lentilactobacillus parabuchneri (ST7) using the MALDI-TOF MS. In conclusion, LAB with potential probiotic properties have been identified from the Ethiopian traditionally fermented honey wine Tej, further study on the in vivo probiotic potential and whole genome analysis needs to be done to ascertain the safety of the identified potential probiotics.

Introduction

Traditionally fermented beverages, foods, and condiments are produced by people of a particular area by applying aged techniques from locally available substrates and using rudimentary apparatus. Tej (Ethiopian honey wine) is among the most widely consumed indigenous traditionally fermented alcoholic beverages of Ethiopia [1, 2], and also the most preferred traditionally fermented alcoholic beverage for international visitors and tourists. It is consumed during social events such as marriage, special ceremonies, and holidays all over the country [3, 4]. It is produced from a mixture of honey, water, and stems or leaves of Rhamnus prenoides (Gesho) [5]. Fully matured Tej has yellow color, sweet, cloudy appearance and effervescent nature, and its fermentation time differ depending on the weather conditions [6, 7]. Tej is reported to contain various nutrients such as proteins, ash, fat, carbohydrate and high moisture content [6–8].

Hazardous metabolites and microorganisms that can potentially pose health risks were reported from traditionally fermented alcoholic beverages [9]. Consumption of alcoholic beverages has also been associated with an increased risk of breast cancer [10]. However, traditionally fermented products such as Tej are versatile sources of bioactive microbes and molecules [1, 11]. Some of the bioactive microbes from fermented foods are probiotics “Live microorganisms that confer a health benefit on the host when administered in adequate amounts” [12]. The health benefits of probiotics include prevention and treatment of Clostridium difficile associated diarrhea [13], Helicobacter pylori (H. pylori) infection [14, 15]; blocking adherence of pathogens to epithelial cells [16], immune modulation [17], preventing communication between pathogenic microorganisms [18]. Lactococcus and Bifidobacterium spp. were also reported to metabolize cholesterol [19], and increase the quality and quantity of spermatozoa [20]. Monoculture or mixed cultures of probiotics were found to enhance the quality of the indigenous microflora in the gastrointestinal tract [21, 22] and improve cognitive functions [23]. Immediate recovery of patients suffering from chronic irritable bowel syndrome was also reported after ingestion of probiotic-supplemented fermented milk [24].

Probiotic microorganisms are conventionally added to milk and other dairy products to make them functional foods [18]. Milk and its products play a significant role as carriers for probiotic microorganisms. However, consumption of probiotics through such carriers has limitations due to lactose intolerance, allergy to milk proteins, and saturated fat contents. Thus, there has been a shift from the consumption of dairy-based probiotics to non-dairy probiotics [25]. Substrates used for Tej production such as Ethiopian honey was reported to have antimicrobial activities [26]. Honey has also antioxidant and prebiotic properties [27]. Thus, at the start of the research, the investigators hypothesized that Tej could be a potential source of unique non–dairy probiotic microorganisms with beneficial health effects and can be used as a potential non-dairy carrier substrate for probiotic microorganisms.

The most common Tej fermenting microorganisms reported are the genus Lactobacillus from lactic acid bacteria and Saccharomyces from the yeasts [3, 7, 8]. A few researches screened and evaluated the starter culture potential of lactic acid bacteria and yeasts from Tej [28], and microbial dynamics during Tej fermentation [29, 30]. There are no studies regarding probiotics from Tej, this is the first attempt to uncover the probiotic potential of lactic acid bacteria isolated from Tej. Therefore, this research aimed to investigate the probiotic properties of LAB isolated from Tej samples.

Materials and methods

Sample collection

A stratified random sampling technique was used to collect Tej (250–500 mL) samples. A total of fifty-five samples were collected from the different sub-cities in Addis Ababa. The numbers of samples collected per sub-city were, 4 (from Addis Ketema, Gulele, Ledeta and Nefas Silk Lafto), 5 (from Arada, Bole, Kirkos and Yeka), 6 (from Akaki Kality and Kolfe Keranio), and 7 (from Lemi Kura) (Fig. 1). The Tej samples were collected from different Tej producers (Fig. 2). The collected Tej samples were transported to the microbiology laboratory of the Department of Microbial, Cellular and Molecular Biology (DMCMB), Addis Ababa University (AAU), and Bio and Emerging Technology Institute (BETin), Addis Ababa, Ethiopia using ice box. The collected Tej samples were stored at 4 °C in the refrigerator till analysis.

Fig. 1.

Map of the study area. Study country, Ethiopia (left corner), Federal state of Addis Ababa pinned with study areas (right corner) and figure captions (bottom right). The MAP was created using ArcGIS version 10.5 software (ESRI, California, USA)

Fig. 2.

Tej samples collected from different producers from left to right Tej collected from, Tej bet (household Tej producers) (1,5 and 7), small scale Tej producers (2, 3, 4, 6) (The picture was taken during sample collection)

Isolation of lactic acid bacteria (LAB)

To isolate LAB, a volume of 0.1mL of the appropriate dilutions was spread-plated onto De Man, Rogosa, and Sharpe agar (MRS agar) after each sample was serially diluted in test tubes containing 9 mL of sterile phosphate buffered saline. The plates were then incubated at 37 °C for 48–72 h anaerobically. About 10–15 colonies showing distinct cultural characteristics were randomly selected from the countable MRS plates and transferred into MRS broth for purification. Purification was done by repeated sub-culturing on MRS Agar plates. The inoculated agar plates were incubated at 37 °C for 72 h in an anaerobic jar. Microscopic characterization was done for pure cultures of the presumptive lactic acid bacteria. Overnight cultures were wet mounted on microscopic slides for the morphology, cell arrangement, and motility of the isolated bacteria. The isolates were maintained at − 80 °C in MRS broth containing 15% (v/v) glycerol (Fine chemicals (Eth) till further analysis [31, 32].

Biochemical and physiological tests for identification of LAB

KOH-test: The KOH (Potassium hydroxide) test was done for the pure isolates using 3% KOH solution. For the KOH test, bacterial cultures were grown on MRS agar at 37 °C for 24 h under anaerobic conditions and were separately placed on the microscope glass slide. The culture was then mixed with KOH solution. A bacterial colony that didn’t produce a viscid-like structure was selected as LAB [33].

Catalase-test: For the catalase test, 2–3 mL of 3% hydrogen peroxide (H₂O₂; Fine chemicals (Eth)) solution was placed on a clean slide, and using a sterile wooden stick a 24 h fresh colony was mixed with 3% H₂O₂ solution and the colony that produced air bubbles within 5–10 s was considered catalase positive and otherwise it was considered catalase negative [34]. Colony that didn’t produce air bubbles was taken as presumptive lactic acid bacteria.

Oxidase-test: For the oxidase test, 2–3 drops of 1% (w/v) aqueous tetramethyl-p-phenylene-diamine dihydrochloride (Himedia, India) was applied onto sterile filter paper, and a pure colony of bacteria was rubbed on the filter paper. The development of purple color within 10 s was recorded as positive, 10–60 s as delayed positive, and the absence of color or development during later stages was considered negative [34].

Acid and gas production from glucose

A pure colony of each LAB isolate grown for 48 h was inoculated into test tubes containing 5 mL MRS broth, 5% glucose, and 0.01% bromophenol blue (LOB CHEMICAL, India) and adjusted to pH 7.4. The tubes were incubated at 37 °C for 48 h. The presence of free air space above the inverted Durham tube was considered positive for gas production and the absence was considered negative for gas production. In addition, the change in color of the medium from red to yellow was considered an indication of acid production [35].

Oxidation-fermentation (OF)-test

The OF test is used to test whether a bacterium is oxidative or fermentative. Fresh culture was inoculated by stabbing the OF medium (HIMEDIA, India) approximately 1/4 inch from the bottom and conducted [36].

Growth of the isolate at different temperatures

The effects of 25, 30, 35, 40, and 45 °C incubation temperatures against the LAB isolates were investigated [37].

Growth of the isolate at different NaCl concentrations

The effects of 2%, 4%, and 6.5% (w/v) NaCl (CARE, India) concentrations against the LAB isolates were investigated [37].

In-vitro probiotic assay for the LAB isolates

Antibacterial activity

Antibacterial activity was conducted for all (280) LAB isolates. The test organisms; Pseudomonas aeruginosa ATCC 27,853, Staphylococcus aureus ATCC 25,923, and Escherichia coli ATCC 25,922 were obtained from the Ethiopian Public Health Institute (EPHI), Salmonella sp, and Shigella sp previously isolated from food sources were obtained from Bio and Emerging Technology Institute (BETin), Addis Ababa, Ethiopia. Prior to inoculation of the test pathogens to the Muller Hinton Agar, the overnight culture of each test organism was adjusted to 0.5 McFarland standards (approximately 10⁸ CFU/mL). Then, sterile cotton swab was used to uniformly distribute the test organisms over dry surfaces of pre-sterilized Muller Hinton Agar plates, and wells were punched into the inoculated agar plates using sterile cork borer. The antibacterial activities of the identified LAB isolates were determined using the agar-well diffusion method. The overnight active broth culture of each LAB isolate was centrifuged (DR AWELL, U.S.A), separately at 5000 rpm for 10 min. The cell-free supernatant (CFS) from each culture was collected, sterilized by membrane filter (0.22 μm) and placed in the wells made on the inoculated Muller Hinton agar plates for determination of the antagonistic activity against the test organisms [38]. Uninoculated MRS broth and cefoxitin were used as the negative and positive controls, respectively.

Low pH

A total of 200 LAB that showed antibacterial activity were assessed for low pH tolerance, and 158 were able to survive at the low pH values tested. To determine pH tolerance, the LAB isolates were cultured in 10 mL MRS broth separately for 16–18 h at 37 °C, then centrifuged at 6000 rpm for 10–15 min at 4 °C. A sterile test tube was used to collect the pellet, and the pellet was washed with phosphate-buffered saline (PBS, pH 7.0) before inoculation into previously prepared MRS broth at pH values of 2 and 3 using 1 N HCl. One mL of the diluted cell suspension (10⁷ CFU/mL) was inoculated at the pH adjusted above. The cell viability was determined by the plate count method on an MRS agar after an initial 0 and 90 min of incubation at 37 °C for 24–48 h [39].

Bile tolerance

Bile tolerance was conducted on 158 LAB isolates that survived the low pH values, and a total of 134 isolates were able to survive up to 1% bile salt concentrations. For the bile tolerance test, 20 µL of overnight cultures aligning with 0.5 MacFarland (10⁸ CFU/mL) were inoculated onto sterile MRS agar plates supplemented with bile salt (0.3%, 0.5%, 1%, 2%, and 3%, w/v ox gall) (Sigma Chemical Co. St Louis, Missouri, USA). Plates were incubated for 5 days at 37 °C [40]. To assess survival rate, LAB isolates that grew in any of the indicated bile salt concentrations were grown overnight (16–18 h) at 37 °C in 10 mL of MRS broth. The culture was centrifuged at 6000 rpm at 4 °C for 10 min, and the pellet was collected in a sterile tube and washed with phosphate-buffered saline (pH 7.0) before inoculation in MRS broth previously adjusted to 0.3%, 0.5%, and 1% bile salts (Oxoid, UK). The cells were re-suspended in phosphate-buffered saline and the cell density was adjusted to 0.5 MacFarland (10⁸ CFU/mL). Finally, 1 mL of the diluted cell suspension was inoculated into the bile salt adjusted to the above concentrations, and incubated for 6 h. After 6 h of incubation at 37 °C serial dilutions were made in sterile buffered peptone water and 1 mL of the diluted samples were spread plated on MRS agar. The inoculated MRS agar plates were incubated at 37 °C for 48 h and cells were counted. The survival rate was calculated as the percentage of LAB colonies produced on MRS agar compared to the initial LAB colonies [39].

Cell surface hydrophobicity

The binding capacity of the LAB isolates was conducted for the 134 isolates that survived up to 1% bile salt concentration, and 64 LAB isolates showed better adhesion properties. Microbial adhesion to hydrocarbons was assessed using in vitro cell surface hydrophobicity according to the method described by [32, 41]. The cultures grown overnight in MRS broth were centrifugation at 8,000 rpm for 10 min, washed, and re-suspended in a PBS buffer followed by absorbance (A0) measurement at 600 nm using UV-visible spectrophotometer (Jenway, 6405, Felsted, Dunmow, UK). About 3 mL of the cell suspension was mixed with 1 mL of hydrocarbon (p-xylene) after 10 min of pre-incubation at room temperature (23 °C); the two-phase system was mixed on a vortex (LAB STAC United Kingdom) for 2 min. After 15 min the aqueous phase was carefully removed and incubated at 37 °C for 30 min after which its absorbance was measured at 600 nm (A). The percentage of cell surface hydrophobicity of the strain adhering to p-xylene was calculated using the equation:

where A0 and A are the absorbances before and after extraction with p-xylene.

Safety of the LAB isolatesthe quotation mark in Lactobacillus

Hemolytic ability

Hemolytic activity was done for 34 LAB isolates that showed adhesion property above 20%. For hemolytic assay, cultures were streaked onto blood agar plates (HIMEDIA, India) supplemented with 5% (v/v) sheep blood and incubated at 37 °C for 24 h to observe the pattern of hemolysis [42].

Antibiotic susceptibility tests

Antibiotic susceptibility was examined for the 18 LAB isolates with non-hemolytic properties using the standard disc diffusion technique [43]. The standard antimicrobial discs used in the study were Ampicillin (AM, 10 µg), Cefoxitin (FX, 30 µg), Vancomycin (VA, 5 µg), Ciprofloxacin (CIP, 5 µg), Doxycycline (DOX, 30 µg), Gentamicin (GM, 30 µg), Penicillin (PEN, 10 IU), Streptomycin (STR, 10 µg), Tetracycline (TE, 30 µg), and Erythromycin (E RY, 15 µg). The inoculum density of the LAB isolates was separately adjusted to 0.5 McFarland (about 10 ⁸ CFU/mL) turbidity standard. The adjusted bacterial inoculum was swabbed over the surface of sterile Mueller-Hinton agar plates using sterile cotton swab. Then, the antibiotic discs were dispensed onto the surface of the inoculated agar plates and the plates were incubated at 37 °C for 24 h. Finally, the diameters of the zones of inhibitions were measured in millimeters (mm) and the results were interpreted following [43, 44].

Identification of the LAB isolates using MALDI-TOF MS

Identification of the LAB isolates were done using matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF MS) server version 4.1.100(PYTH) 174 2019-06-158-01-16-19. Sample preparation and the MALDI-TOF MS (Bruker MALDI Biotyper; with instrument ID of 8604832.05381) analysis were carried out as reported in [45]. Briefly, a fresh pure colony of each of the LAB isolates was picked with sterile wooden application sticks and smeared onto the MALDI target plate (Bruker Daltonics GmBH, Germany), each plate was overlaid with 1 µL of 70% formic acid and was air dried. All smears on the target plates were overlaid by α-cyano-4-hydroxycinnamic acid in 50% acetonitrile and 2.5% trifluoroacetic acid, and the smear was overlaid with 70% formic acid and allowed to dry at room temperature. Mass spectra were analyzed using Flexcontrol 3.0 software and the MALDI Biotyper database was used for identification. Samples with Biotyper ID score of ≥ 1.7 were accepted for identification, while ID scores < 1.7 were considered unreliable and those samples were excluded from the analysis.

DPPH (2,2-Diphenyl-1-picrylhydrazyl) radical scavenging activity of cell-free supernatant of LAB isolates

DPPH radical scavenging properties was done for the 18 LAB isolates that were identified by the MALDI-TOF MS. Cultures of the LAB were adjusted to approximately 10⁹ CFU/mL. The culture was centrifuged (10,000 x g, 10 min, 4 °C). The pH value of the supernatant was neutralized using 1 M NaOH. The resulting supernatant was filtered (0.22 μm pore size) and mixed with 1 mL of fresh DPPH solution (0.2 mM) and stored for 30 min in darkness at room temperature and observed using spectrophotometer at 517 nm [46].The scavenging effect of the cell-free supernatant (CFS) of the identified LAB isolates on the free radical DPPH (1, 1-diphenyl-2-picrylhydrazyl (DPPH) was measured in accordance with the slightly modified method of Lin and Chang [47]. PBS was used as blank and ascorbic acid (1 mg/mL) was used as positive control.

Calculation of the radical scavenging activity was done using the following formula:

A0 and A are the absorbance of the blank and the sample.

Cholesterol assimilation by the LAB isolates

Cholesterol assimilations of the LAB isolates were assessed for the 18 isolates identified by MALDI-TOF MS. Cholesterol (Sigma-Aldrich Chemicals Ltd, USA) was added to MRS broth at a final concentration of 100 µg/mL. Each overnight culture of the lactic acid bacteria was inoculated at 1% (v/v) into the MRS broth containing cholesterol at 37 °C for 24 h. The LAB suspensions were centrifuged at 4000 rpm for 10 min and the supernatant containing non-assimilated cholesterol was collected. Briefly, 500 µL of 33% (w/v) KOH and 1 mL absolute ethanol (Fine chemicals (Eth)) were added to 500 µL of the supernatants. The solutions were then vortex for 1 min and incubated at 37 °C for 15 min. For phase separation, 1 mL of deionized water and 1.5 mL of hexane were added to the solutions and vortexed for 1 min. The phases were then allowed to separate at RT (23 °C). Subsequently, 500 µL of the hexane layer was transferred into a glass tube and the solvent was evaporated at RT (23 °C). 1.5 mL of FeCl₃ working solution was added to the tube containing the sample. After thorough mixing, the solution was allowed to stand for 10 min. Then, concentrated sulfuric acid (1 mL) was added and the solution was vortexed and placed in the dark. Forty-five minutes later absorbance was determined at 517 nm using a UV spectrophotometer (Jenway, 6405, Felsted, Dunmow, UK) [47, 48]. The activity of cholesterol-lowering (%) was calculated as follows:

where cholesterol A and cholesterol R are the absorbance of the cholesterol added and cholesterol recovered at 517 nm, respectively.

Data analysis

Data analysis was done using IBM SPSS (version 22. SPSS Inc. Chicago. IL. USA). All results were presented as a mean of a triplicate experiment and mean comparison was done using one way ANOVA. Independent sample, t-test was used to assess differences in the survival of the ‘lactobacilli’ at different pH values. Multiple comparisons were done using Tukey HSD. P-value < 0.05 was considered statistically significant.

Results

Isolation, biochemical, and physiological characterization of LAB

A total of 280 LABs were isolated, purified, and characterized. The LAB isolates were characterized by pinpointed white to cream-colored colonies (Fig. 3). All of them were Gram-positive, catalase-negative, oxidase negative non – motile, and included both hetero/homo fermentative. They survived incubation temperatures of 25 °C to 45 °C, and grew at salt concentrations of 2% and 4%, but didn’t survive a salt concentration of 6.5%.

Fig. 3.

The colony morphologies of ‘lactobacilli’ isolated from Tej bet (household Tej producers (above) and small scale Tej producers (below) (Pictures were taken during purification of the isolates)

Antibacterial properties

Variations in the antibacterial activities of the cell-free supernatant (CFS) of the LAB isolated from Tej samples was observed. The antibacterial activities differed between similar and different LAB species, as well as within similar LAB species against different test pathogens (Table 1). The highest zone of inhibition (19.00 ± 0.30 mm) was observed for the cell-free supernatant of Lentilactobacillus hilgardii ST 6 against P. aeruginosa ATCC 27,853, Lentilactobacillus hilgardii ST 9 (18.00 ± 0.02 mm) against S. aureus ATCC 25,923, Lentilactobacillus hilgardii ST 6 (17.00 ± 0.23 mm) against Shigella sp., Lentilactobacillus hilgardii ST 4 (16.00 ± 0.33 mm) against Salmonella sp., and Lecticaseibacillus paracasei ST 18 (15.00 ± 0.01) against E. coli ATCC 25,922. Statistically significant (p < 0.05) inhibition against the test bacteria was observed for most of the ‘lactobacilli’ against the test bacteria. However, about 33% (6/18) ‘lactobacilli’ species had no antimicrobial activity against E. coli ATCC 25,922. The clinical isolates of Salmonella and Shigella were sensitive to the cell free supernatants of over 94% of the ‘lactobacilli’ species.

Table 1.

Antibacterial properties of ‘lactobacilli’ against selected test bacteria

Strain no.	LAB strains	Zone of inhibition (mean ± SD) against the test organisms
		E. coli ATCC 25,922	P.aeruginosa ATCC 27,853	S. aureus ATCC 25,923	Salmonella sp.	Shigella sp.
ST1	Levilactobacillus brevis	13.00 ± 0.04bcA	11.00 ± 0.03dB	11.00 ± 0.03eB	13.00 ± 0.00dA	13.00 ± 0.03dA
ST2	Levilactobacillus brevis	15.00 ± 0.32aB	14.00 ± 0.31cC	10.00 ± 0.07fD	16.00 ± 0.03aA	15.00 ± 0.00bB
ST3	Lecticaseibacillus paracasei	12.00 ± 0.20cA	12.00 ± 0.41dA	11.00 ± 0.31eB	11.00 ± 0. 33fB	11.00 ± 0. 23fB
ST4	Lentilactobacillus hilgardii	0.00 ± 0.00eD	12.00 ± 0.01dC	13.00 ± 0.22cB	16.00 ± 0. 33aA	12.00 ± 0.00eC
ST5	Lentilactobacillus hilgardii	15.00 ± 0.58aA	12.00 ± 0.20dC	11.00 ± 0.04eC	15.00 ± 0.16bB	0.00 ± 0.00iD
ST6	Lentilactobacillus hilgardii	14.00 ± 0.34abE	19.00 ± 0.30aA	18.00 ± 0.21aB	15.00 ± 0.00bD	17.00 ± 0.23aC
ST7	Lentilactobacillus parabuchneri	0.00 ± 0.00eC	15.00 ± 0.00cA	15.00 ± 0.14bA	14.00 ± 0.06cB	15.00 ± 0.0bA
ST8	Lecticaseibacillus paracasei	14.00 ± 0.01abB	17.00 ± 0.20bA	10.00 ± 0.22fC	14.00 ± 0.06cB	14.00 ± 0.30cB
ST9	Lentilactobacillus hilgardii	14.00 ± 0.10abC	14.00 ± 0.03cC	18.00 ± 0.02aA	13.00 ± 0.22dD	15.00 ± 0.13bB
ST10	Lecticaseibacillus paracasei	14.00 ± 0.35abC	17.00 ± 0.32bA	15.00 ± 0.02bB	15.00 ± 0.33bB	14.00 ± 0.00cC
ST11	Lecticaseibacillus paracasei	0.00 ± 0.00eD	8.00 ± 0.00eC	10.00 ± 0.03fB	12.00 ± 0.11eA	10.00 ± 0.00gB
ST12	Lecticaseibacillus paracasei	10.00 ± 0.03dB	0.00 ± 0.00fD	10.00 ± 0.00fB	7.00 ± 0.21iC	13.00 ± 0.23dA
ST13	Lecticaseibacillus paracasei	0.00 ± 0.00eE	12.00 ± 0.00dA	11.00 ± 0.00eB	10.00 ± 0.04gC	8.00 ± 0.10hD
ST14	Lecticaseibacillus paracasei	12.00 ± 0.15cA	0.00 ± 0.00fE	11.00 ± 0.00eB	8.00 ± 0.00hD	10.00 ± 0.00gC
ST15	Lecticaseibacillus paracasei	0.00 ± 0.00eD	0.00 ± 0.00fD	10.00 ± 0.10fC	14.00 ± 0.12cA	12.00 ± 0.01eB
ST16	Lecticaseibacillus paracasei	10.00 ± 0.00dB	0.00 ± 0.00fC	10.00 ± 0.00fB	14.00 ± 0.02cA	10.00 ± 0.03gB
ST17	Levilactobacillus brevis	0.00 ± 0.00eC	8.00 ± 0.00eB	12.00 ± 0.0dA	0.00 ± 0.00jC	12.00 ± 0.21eA
ST18	Lecticaseibacillus paracasei	15.00 ± 0.01aA	12.00 ± 0.14dC	10.00 ± 0.02fC	13.00 ± 0.21dB	13.00 ± 0.32dB

Zone of inhibition values represent the means of three replicates. Values with different letters (a–j) in the same column and different superscripts with capital letters across the row are significantly different. (Tukey’s HSD, p < 0.05)

Low pH tolerance

The mean survival rate of the ‘lactobacilli’ to low pH showed variability between different and within similar species. Significantly better survival (p < 0.05) was recorded for the ‘lactobacilli’ at pH 3 compared to pH 2 (Table 2). The highest mean survival 65.47 ± 0.08% at pH 2 and 70.48 ± 0.04% were recorded for Lecticaseibacillus paracasei ST 8 at pH 3, and the lowest survival of 5.04 ± 0.05% and 11.05 ± 0.06% were recorded for Lecticaseibacillus paracasei, respectively.

Table 2.

Survival of ’lactobacilli’ at different pH and bile concentrations

Strains No.	Organism best match	Acid tolerance (%)		Bile tolerance (%)
		pH 2	pH 3	0.3	0.5	1
ST1	Levilactobacillus brevis	47.33 ± 0.1dB	69.34 ± 0.07 bA	81.96 ± 0.06 cA	47.22 ± 0.11fB	47.34 ± 0.05 dB
ST2	Levilactobacillus brevis	45.92 ± 0.13 eB	63.86 ± 0.11 cA	92.48 ± 0.03 aA	47.28 ± 0.04 fB	47.22 ± 0.12 dB
ST3	Lecticaseibacillus paracasei	9.98 ± 0.05 lB	14.95 ± 0.06 mA	92.45 ± 0.08 aA	47.37 ± 0.09 fB	47.34 ± 0.06 dB
ST4	Lentilactobacillus hilgardii	23.77 ± 0.08 jB	57.08 ± 0.17 eA	77.88 ± 0.04 dA	23.84 ± 0.05 oB	23.48 ± 0.03 lB
ST5	Lentilactobacillus hilgardii	47.12 ± 0.17 dB	61.54 ± 0.04 dA	83.98 ± 0.03 bA	40.76 ± 0.08 gB	40.65 ± 0.07fB
ST6	Lentilactobacillus hilgardii	54.79 ± 0.05 bB	69.36 ± 0.17bA	81.94 ± 0.08cA	64.83 ± 0.04cB	54.78 ± 0.04cC
ST7	Lentilactobacillus parabuchneri	47.63 ± 0.08dB	61.47 ± 0.89dA	73.65 ± 0.06fA	65.43 ± 0.04bB	65.37 ± 0.05aB
ST8	Lecticaseibacillus paracasei	65.47 ± 0.11aB	70.46 ± 0.06aA	73.68 ± 0.11fA	64.45 ± 0.06dB	55.34 ± 0.06cC
ST9	Lentilactobacillus hilgardii	45.77 ± 0.04eB	53.55 ± 0.06fA	66.43 ± 0.04gA	40.64 ± 0.05gB	38.33 ± 0.04gC
ST10	Lecticaseibacillus paracasei	49.78 ± 0.05cB	52.69 ± 0.08gA	76.22 ± 0.02eA	72.02 ± 0.03aB	60.85 ± 0.07bC
ST11	Lecticaseibacillus paracasei	43.26 ± 0.15fB	69.55 ± 0.08bA	64.75 ± 0.07hA	25.54 ± 0.09mB	24.69 ± 0.30kC
ST12	Lecticaseibacillus paracasei	42.19 ± 0.11gB	47.64 ± 0.06hA	32.34 ± 0.05mC	39.13 ± 0.04iB	41.43 ± 0.04eA
ST13	Lecticaseibacillus paracasei	17.08 ± 0.02kB	23.00 ± 0.04lA	42.12 ± 0.03jA	39.95 ± 0.07hB	38.51 ± 0.70gC
ST14	Lecticaseibacillus paracasei	5.04 ± 0.07mB	11.02 ± 0.08nA	21.74 ± 0.08oC	28.22 ± 0.02lB	33.48 ± 0.17iA
ST15	Lecticaseibacillus paracasei	34.15 ± 0.11iB	41.18 ± 0.04jA	34.13 ± 0.04lB	32.26 ± 0.08jC	35.58 ± 0.40hA
ST16	Lecticaseibacillus paracasei	31.49 ± 0.08jB	32.93 ± 0.04kA	30.77 ± 0.05nA	30.13 ± 0.04kB	30.22 ± 0.16jB
ST17	Levilactobacillus brevis	41.91 ± 0.13gB	48.02 ± 0.11hA	37.35 ± 0.07kA	25.23 ± 0.04nC	30.92 ± 0.17jB
ST18	Lecticaseibacillus paracasei	36.44 ± 0.06hB	43.79 ± 0.10iA	58.95 ± 0.07iA	49.66 ± 0.08eB	40.36 ± 0.06fC

The pH values and bile concentrations represent the means of three replicates. Student t-test was used for comparison of the mean survival at pH 2 and 3. Values with different superscript letters (a-o) in the same column and different superscripts with capital letter across the row are significantly different (p < 0.05). Multiple comparisons for bile tolerance were done using Tukey’s HSD

Bile salt tolerance

Variabilities in tolerances to the different concentrations of bile tested were observed with a general decrease in survival rate with an increase in the bile concentrations. The highest mean bile tolerance of 92.48 ± 0.03% was recorded for Lecticaseibacillus paracasei ST 3 followed by 92.45 ± 0.03% for Levilactobacillus brevis ST 2 to bile salt concentration of 0.3%. The highest mean bile salt tolerances of 72.02 ± 0.03% and 65.37 ± 0.05% were recorded for Lecticaseibacillus paracasei ST 10 to bile salt concentration of 0.5% and Lentilactobacillus parabuchneri ST 7 to bile salt concentration of 1%, respectively. Statistically higher (p = 0.001) survival of the ‘lactobacilli’ was observed with a bile salt concentration of 0.3% compared to 0.5% and 1% bile concentrations. However, statistically significant differences were not observed (p > 0.05) between the overall survival of the Lactobacillus spp. at bile salt concentrations of 0.5% and 1%. The survival of Levilactobacillus brevis ST 2 and Lecticaseibacillus paracasei ST 3 to the bile salt concentration of 0.3% was significantly (p < 0.05) higher than the other LAB tested. The survival of Lecticaseibacillus paracasei ST 10 and Lentilactobacillus parabuchneri ST7 at bile salt concentrations of 0.5% and 1% was significantly (p < 0.05) higher than the other ‘lactobacilli’ (Table 2).

Cell-surface hydrophobicity of the LAB isolates

The ability of the LAB isolates to adhere to the gastrointestinal wall was measured using bacterial adhesion to hydrocarbon. A total of 64 LAB isolates were verified for their adhesion property. The highest cell surface hydrophobicity of 76.71 ± 0.66% demonstrated by Levilactobacillus brevis ST 1 was significantly (p < 0.05) higher than the cell surface hydrophobicity of most of the Lactobacillus spp tested. The lowest cell surface hydrophobicity of 24.11 ± 0.85% was recorded for Lecticaseibacillus paracasei ST 18 (Table 3). Most of the Lactobacillus spp in the current study demonstrated high level of adhesion property to hydrophobic surfaces simulated by cell – surface hydrophobicity.

Table 3.

Cell surface hydrophobicity, DPPH radical scavenging, and cholesterol assimilation activities of Lactobacillus spp

Strains No.	LAB species	Cell surface hydrophobicity (%)	DPPH radical (%)	Cholesterol assimilation (%)
ST1	Levilactobacillus brevis	76.71 ± 0.66a	68.90 ± 3.00a	31.80 ± 1.46a
ST 2	Levilactobacillus brevis	76.61 ± 1.00a	61.60 ± 1.43b	19.40 ± 1.10b
ST3	Lecticaseibacillus paracasei	76.50 ± 1.70ab	60.00 ± 0.83bc	18.60 ± 1.48b
ST4	Lentilactobacillus hilgardii	75.70 ± 1.56ab	59.30 ± 2.46bc	12.20 ± 0.68cd
ST5	Lentilactobacillus hilgardii	75.51 ± 0.86abc	58.20 ± 1.99bc	5.40 ± 0.55cd
ST6	Lentilactobacillus hilgardii	75.30 ± 2.33abcd	57.20 ± 1.02bc	4.90 ± 1.39cd
ST7	Lentilactobacillus parabuchneri	75.00 ± 1.10abcd	56.40 ± 1.73bc	4.90 ± 3.11cd
ST8	Lecticaseibacillus paracasei	74.00 ± 0.55abcd	55.00 ± 1.99c	4.80 ± 1.46cd
ST9	Lentilactobacillus hilgardii	73.81 ± 1.20abcde	29.30 ± 1.88d	4.50 ± 1.97d
ST10	Lecticaseibacillus paracasei	72.20 ± 0.66bcdef	27.10 ± 0.83d	4.31 ± 2.84d
ST11	Lecticaseibacillus paracasei	71.31 ± 0.45cdef	27.10 ± 0.47d	4.10 ± 0.96d
ST12	Lecticaseibacillus paracasei	71.00 ± 0.66def	25.00 ± 1.99d	4.01 ± 2.90d
ST13	Lecticaseibacillus paracasei	69.50 ± 0.49ef	15.70 ± 0.47e	3.71 ± 1.61 d
ST14	Lecticaseibacillus paracasei	68.50 ± 1.10f	12.10 ± 0.64e	3.60 ± 1.44d
ST15	Lecticaseibacillus paracasei	43.21 ± 0.65g	11.11 ± 0.44e	3.01 ± 0.85d
ST16	Lecticaseibacillus paracasei	40.00 ± 0.55gh	10.40 ± 1.43e	2.80 ± 1.91d
ST17	Levilactobacillus brevis	36.31 ± 0.45h	1.52 ± 1.98f	2.01 ± 2.84d
ST18	Lecticaseibacillus paracasei	24.11 ± 1.20i	1.73 ± 0.57f	1.60 ± 2.46d

Values with different letters within the column are statistically significant at Tukey’s HSD 0.05

Antioxidant activities

The highest DPPH radical scavenging activity (68.90%) was observed for Levilactobacillus brevis ST 1 and the lowest (0.40%) value was recorded for Lecticaseibacillus paracasei. The DPPH radical scavenging ability of Levilactobacillus brevis ST1 was significantly (p < 0.05) higher than all of the Lactobacillus spp in the current study. In contrast, Levilactobacillus brevis ST 17 and Lecticaseibacillus paracasei ST 18 showed the lowest radical scavenging activity (Table 3). This shows that radical scavenging activity could differ between different strains of the same species.

Cholesterol assimilation

The percentage cholesterol reduction by the LAB isolates in this study ranged from 1.60 to 31.80% for Lecticaseibacillus paracasei ST 18 and Levilactobacillus brevis ST 1, respectively. The differences in cholesterol assimilation were statistically significant (p < 0.05). Variability in cholesterol assimilation was observed between as well as within the species of the ‘lactobacilli’ (Table 3).

Hemolytic ability

The entire identified LAB isolates by the MALDI-TOF MS showed no hemolytic activity.

Antibiotic susceptibility

Most of the LAB isolates were sensitive to six to eight antibiotics out of ten antibiotics used in this study. The highest resistance was recorded against vancomycin followed by ciprofloxacin and ampicillin (Table 4). The MAR index of was; 0.1 (ST4), 0.2 for ST6 and ST11), 0.3 (ST1, ST3, ST5, ST7, ST9, ST14 and ST15), 0.4 (ST10, ST16, ST18), 0.5 (ST13), 0.6 (ST17), 0.8 (ST8) and 1 (ST12).

Table 4.

Antibiotic susceptibility profile of the Lactobacillus spp

Strain No.	LAB species	Antibiotic susceptibility
		PEN	AM	FX	VA	GM	STR	TE	ERY	CIP	DOX
ST1	Levilactobacillus brevis	MS	R	MS	R	S	S	MS	MS	S	R
ST2	Levilactobacillus brevis	R	R	MS	R	S	R	MS	S	R	R
ST3	Lecticaseibacillus paracasei	MS	S	S	R	S	R	S	S	R	MS
ST4	Lentilactobacillus hilgardii	MS	S	S	S	S	MS	S	S	R	S
ST5	Lentilactobacillus hilgardii	MS	S	S	R	S	R	S	S	R	MS
ST6	Lentilactobacillus hilgardii	R	S	S	S	S	S	MS	R	S	MS
ST7	Lentilactobacillus parabuchneri	MS	S	S	R	S	R	S	S	R	S
ST8	Lecticaseibacillus paracasei	R	R	R	R	R	R	MS	R	R	MS
ST9	Lentilactobacillus hilgardii	MS	R	MS	R	S	R	S	S	S	MS
ST10	Lecticaseibacillus paracasei	MS	R	MS	R	S	R	S	S	R	S
ST11	Lecticaseibacillus paracasei	R	S	S	R	S	S	MS	S	S	MS
ST12	Lecticaseibacillus paracasei	R	R	R	R	R	R	R	R	R	R
ST13	Lecticaseibacillus paracasei	S	R	R	R	S	S	R	S	S	R
ST14	Lecticaseibacillus paracasei	MS	R	R	S	S	S	R	S	S	S
ST15	Lecticaseibacillus paracasei	S	R	R	R	S	S	R	S	R	R
ST16	Lecticaseibacillus paracasei	R	MS	MS	R	MS	R	R	S	MS	MS
ST17	Levilactobacillus brevis	MS	R	R	MS	S	R	R	S	R	R
ST18	Lecticaseibacillus paracasei	S	R	R	MS	S	MS	MS	S	R	R

Ampicillin (Amp 10 µg), Cefoxitin (CK 30 µg), Ciprofloxacin (CIP 5 µg), Doxycycline (DO 30 µg), Gentamicin (GME 30 µg), Penicillin (P 10 IU), Streptomycin (S 10 µg), Tetracycline (TE 30 µg), Erythromycin (E 15 µg) and Vancomycin (VA 30 µg) antimicrobial discs. R (resistant), MS (moderately susceptible), or S (susceptible)

MALDI-TOF MS identification of the LAB

Lactic acid bacteria were further identified using MALDI-TOF MS. The LAB were identified as Levilactobacillus brevis (L. brevis), Lecticaseibacillus paracasei (L. paracasei), Lentilactobacillus hilgardii (L. hilgardii) and Lentilactobacillus parabuchneri (L. parabuchneri) (Table 5). More than 55% of the identified LAB was Levilactobacillus paracasei and the least frequent group was Lentilactobacillus parabuchneri. The main source of the ‘lactobacilli’ was Tej collected from Tej bet (household Tej producers) (Table 5).

Table 5.

Identified ‘lactobacilli’ using MALDI-TOF MS analysis

Strain No.	Organism best mach	Score value	Sample sources
ST1	Levilactobacillus brevis	2.33	Hotels and restaurants
ST2	Levilactobacillus brevis	2.26	Hotels and restaurants
ST3	Lecticaseibacillus paracasei	2.05	Tej bet (household Tej producers)
ST4	Lentilactobacillus hilgardii	1.95	Hotels and restaurants
ST5	Lentilactobacillus hilgardii	2.23	Tej bet (household Tej producers)
ST6	Lentilactobacillus hilgardii	2.04	Small-scale Tej producers
ST7	Lentilactobacillus parabuchneri	1.86	Small-scale Tej producers
ST8	Lecticaseibacillus paracasei	2.01	Small-scale Tej producers
ST9	Lentilactobacillus hilgardii	2.33	Tej bet (household Tej producers)
ST10	Lecticaseibacillus paracasei	2.19	Small-scale Tej producers
ST11	Lecticaseibacillus paracasei	2.33	Tej bet (household Tej)
ST12	Lecticaseibacillus paracasei	2.34	Tej bet (household Tej)
ST13	Lecticaseibacillus paracasei	2.43	Tej bet (household Tej)
ST14	Lecticaseibacillus paracasei	1.97	Tej bet (household Tej)
ST15	Lecticaseibacillus paracasei	2.17	Small-scale Tej producers
ST16	Lecticaseibacillus paracasei	2.21	Hotels and restaurants
ST17	Levilactobacillus brevis	2.21	Small-scale Tej producers
ST18	Lecticaseibacillus paracasei	2.00	Hotels and restaurants

Discussion

Tej is one of the traditionally fermented Ethiopian alcoholic beverages consumed on different occasions and ceremonies. The product is the source of income for several households in the country especially for females. Most of the researches to date dealt with the microbiological quality, chemical compositions of Tej, and defined starter culture development. There has been no report on the screening of lactic acid bacteria from Tej for potential use as probiotics. The study aimed to isolate lactic acid bacteria and characterize their probiotic potentials. A total of 55 Tej samples were collected from hotels and restaurants, small scale, and Household Tej producers.

Characterization of the LAB isolates revealed that the majority were whitish or cream-colored in their appearance, heterofermentative, and the most prevalent LAB in the Tej samples were 'lactobacilli'. The finding of this study is in line with reports by Bahiru et al. [3] from Tej samples. This shows that 'lactobacilli' were the most prevalent among Tej samples collected at different timelines. Such dominance might be due to differences in the adaptation mechanisms to the alcohol and acid produced during the fermentation of Tej samples. The MALDI-TOF MS identification grouped the “'lactobacilli' from Tej into four species namely L. paracasei, L. brevis, L. hilgardii, and L. parabuchneri.

The antagonistic capacity against pathogenic bacteria is an important attribute during the selection of potential probiotics because the ultimate goal of probiotics is to defend hosts against disease-causing agents. The antagonistic mechanisms against pathogens include competition for colonization of the host tissues, productions of organic acids, hydrogen peroxide, secondary metabolites with antimicrobial properties, and production of signaling molecules that can strengthen the host immune system [48]. Thus, antagonistic properties have applications for post antibiotics adjuvant therapy and correction of diarrhea-induced gastrointestinal microbiota abnormalities [15, 49, 50]. The antagonistic activities of the ‘lactobacilli’ in the current study showed variability between similar and different ‘lactobacilli’ species, and against the tested pathogens. Differences in the degrees of inhibition of E. coli, P. aeruginosa and S. aureus by lactic acid bacteria were also reported [51]. The inhibitory activity of the ‘lactobacilli’ in the current study were in agreement with the findings of Reuben et al. [52]., but higher than a report from other types of fermented Ethiopian beverages [53]. A study by Adongo [54] also reported variations in the level of inhibition of test bacteria by L. brevis with the highest zone of inhibition against S. Typhimurium and the lowest against E. coli. Lactobacilli have different antagonistic mechanisms such as production of organic acids and antibiotic peptides [55]. A whole genome sequence and comparative genome analysis of L. paracasei DTA93 revealed 14 genes linked to bacteriocin production [56]. However, bacteriocine encoding genes were not detected in Lactobacillus parabuchneri [57].

After evaluation of the cell free supernatants (CFS) of lactic acid bacteria, Arrioja-Breton et al. [58] found that the main inhibitory components were organic acids. A comparative genome analysis demonstrated the presence of specific region coding for bacteriocin and type III poly ketide synthase in L. brevis MYSN105 [59, 60]. A study by Rushdy and Gomma [61] reported that the antimicrobial activity of L. brevis was not due to acidity or hydrogen peroxide but it was due to bacteriocin-like substance. In our study, the highest inhibition of Pseudomonas aeruginosa was observed for Lentilactobacillus hilgardii. The susceptibility of Pseudomonas aeruginosa observed in this study might be due to the production of hydrogen peroxide and organic acid production. Similarly, the inhibitory effect of hydrogen peroxide on S. aureus was reported to be due to the expression of specific genes [61]. Salmonella sp. though resistant to hydrogen peroxide, were reported to be sensitive to acetic acid produced by Lentilactobacillus hilgardii [62]. Shigella sp. was also sensitive to CFS of Lentilactobacillus hilgardii and Levilactobacillus brevis, this might be due to the combination of bacteriocin, hydrogen peroxide and other bioactive compounds in the CFS. The CFS contains diverse classes of chemicals such as, antimicrobial substances and antioxidants [63]. The CSF of ‘lactobacilli’ from vaginal sources also antagonized S.aureus and P.aeruginosa [64]. The CSF of lactic acid bacteria was also reported to have strong inhibitory effect against β-lactamase producing microorganisms, and the inhibitory activity of the CFS was strain dependent [65]. The concerted effect of organic acids and secondary metabolites might be responsible for the inhibitory activity of the CFS of the current study. Thus, the nature of the compounds responsible for the antimicrobial activity needs to be assessed through neutralization of CFS with NaOH, treatment with proteinase and catalase.

To reach the small intestine the potential probiotic lactic acid bacteria need to tolerate pH values of 3 or below for three hours [66]. The LAB isolated and identified from Tej were able to tolerate pH 2 and 3 with different degrees of tolerance during the four hours of incubation. Generally, a reduction in viability was observed at pH 2 compared to pH 3. This result is in agreement with reports by Tigu et al. [32]. and Garedew & Ashenafi [55]. Other reports also indicate survival of L. brevis and other LAB species at different pH values during incubation for 2 and 4 h though survival rate reduction was observed at the lowest pH value [39, 52, 67, 68]. Better acid tolerance of L. brevis BBE-Y52 compared to L. paracasei was also reported by Fang et al. [69]. In our study, the highest survival to pH 2 was recorded for L. paracasei ST8 and the lowest was also recorded for L. paracasei ST14. Various mechanisms were reported for acid tolerance in LAB such as neutralization of acid products using alkaline metabolites such as urea, ammonia and arginine [70, 71], using proton pumps [72], malolactic fermentation which decarboxylates L-malate to L-lactic acid producing carbon dioxide that consumes intracellular protons ultimately decreasing intracellular proton concentrations [73, 74]. Pre-adaptation to lethal or sub-lethal pH conditions for a limited time was also associated with better survival of the LAB to low pH conditions [75]. Genomic approaches also revealed the importance of bgi G in regulating acid stress responses of Lecticaseibacillus paracasei L9. The bgiG gene helps manage the phenotypic heterogeneity that arises during low pH stress, and acts through activation of genes like hsp20 responsible for enhancing acid tolerance [76]. Thus, the Lecticaseibacillus paracasei species with high resistance to the low pH might contain gene bgiG encoding acid stress responses.

Another key criterion for the evaluation of probiotic microorganisms that pass through the gastrointestinal tract is tolerance to bile salt. Due to its detergent properties’ bile can disrupt membranes, damage DNA, denature proteins, and chelate calcium ions leading to reduced survival of microorganisms in the GI tract [77]. The small intestine has an average 10 mM bile concentration during digestion and the time taken during digestion in the small intestine is 6 h on average [66, 78]. In our study, the highest bile tolerance was recorded for Levilactobacillus brevis followed by Lecticaseibacillus paracasei. Similarly, survival of LAB including L. brevis to bile concentration from 0.3 to 2% with gradual reduction as the incubation period increased from 0 to 6 h was reported previously [40, 68, 79]. High bile tolerance of Lecticaseibacillus paracasei was also reported from Artisanal fermented pickles [80], and strains of Lecticaseibacillus paracasei from wine were found to be better tolerant to 1.5% bile salt [81]. A study by Ma et al. [82], reported up-regulation of the gene mleS encoding malolactic enzyme in Lactobacillus paracasei L9 which is thought to be responsible for bile tolerance through maintaining membrane balance. A research conducted by Erturkmen et al. [82]. also reported high expression of bsh3 genes in Lacticaseibacillus paracasei strains correlating with bile salt tolerance. Similarly, Levilactobacillus brevis was reported to have bile salt hydrolase gene responsible for bile salt tolerance [83]. In general, intrinsic bile tolerances by lactic acid bacteria include efflux of bile salts [84], hydrolysis of bile salts using bile salt hydrolase enzyme [85, 86], and using bile acids as nutrients and electron acceptor [87]. Chen et al. [88]., employed genomic approach for assessing the bile salt tolerance mechanisms of ‘lactobacilli’ and found that the genes responsible for bile tolerance were the ones associated with the two-component and phosphotransferase systems (PTS), rather than the bile salt hydrolase genes.

Adhesion helps probiotic microorganisms to colonize the intestinal epithelium, competitively exclude pathogenic microorganisms, multiply, modulate the immune system of the host, prevent cells from being washed out, and produce essential substances such as enzymes, vitamins, and secondary metabolites [89, 90]. In this study, all LAB isolates showed adhesion to xylene with differences in the degree of adhesion. This is in agreement with [32, 39, 41, 48]. However, the percentage adhesion to xylene reported in our study is lower than those reported by Tigu et al. [32]. Adhesion of L. brevis was the highest compared to other members of the tested Lactobacillus spp [53, 67]. The cell surface hydrophobicity of some of the ‘lactobacilli’ in our study is higher than the values reported for ‘lactobacilli’ isolated from Artisanal fermented pickles [80]. This shows that adhesion to cell lines or hydrophobic surfaces is strain or species-specific.

Oxidative stress arises from the excessive production of reactive oxygen and nitrogen species and can cause an imbalance in the body by damaging cells and tissues [79]. In the current study, 90% of LAB isolates exhibited 0.4 to 63.2% ability to scavenge DPPH radicals. The highest values of DPPH radical scavenging activities of the cell free supernatant of L.brevis in our study were higher than the values reported by Kim et al. [79] for L. brevis KU15147 and L. brevis KU15154, and Yang et al. [90] for L. brevis KU15151. DPPH antioxidant values higher than the value for one of the L. brevis [79, 90] and Lecticaseibacillus paracasei [80] strains in our study were also reported.

Probiotic bacteria counteracts the effect of oxidative stress by producing metabolites with antioxidant properties, such as butyrate, glutathione (GSH), and foliate [47, 91]. The main antioxidant system in lactobacillus strains was found to be thioredoxin according to a recent study [92]. In addition, bacteriocins [93], bio surfactants [94] and exopolysaccharides [95] were reported to have strong antioxidant properties.

One of the main causes of metabolic disease is excessive cholesterol. Currently, there are insufficient choices available to regulate cholesterol levels. However, study findings reported that probiotic microorganisms’ lower cholesterol levels through various mechanisms such as deconjugation of bile which increases the demand of cholesterol for synthesis of bile acid, reduce absorption of cholesterol in the intestine by binding and incorporating into cell membrane [48, 96, 97]. In the current study cholesterol reduction by the LAB varied between 1.6% and 38.1%, and even between similar species of ‘lactobacilli’. The maximum value of percentage cholesterol reduction in the current study is lower than the values reported by Castorena-Alba et al. [96, 98]., but higher than the value reported by Kathade et al. [99]. Differences in cholesterol reduction capability are dependent on the presence of specific cholesterol reduction mechanisms by the probiotic LAB.

Antibiotic-susceptible probiotic is preferable to reduce the risk of horizontal gene transfer to pathogenic microorganisms. Most of the LAB isolates were sensitive to six to eight antibiotics out of ten antibiotics used in this study. Some of the LABs were resistant to streptomycin and vancomycin. Different level of antibiotic susceptibility was also reported by Akalu et al. [53]. and Garedew & Ashenafi [100]. Wang et al. [101] also reported resistance of L. brevis strains to ciprofloxacin, chloramphenicol, gentamycin, kanamycin, and streptomycin. Dose-dependent sensitivity and resistance to ampicillin, chloramphenicol, tetracycline, kanamycin, and streptomycin were reported for L. brevis BBE-Y52 [69]. The whole genome sequence of L.brevis HQ1-1 revealed 11 antibiotic resistance genes namely rpoB, mfd, desR, mprF, gyrB, taeA, lmrB, emeA, murA, ef-tu and arlR. The strains were resistant to norfloxacin, ciprofloxacin, fosfomycin, vancomycin and polymixin B. Howver, they were not resistant to ampicillin, amoxicillin, rifampicin, erythromycin and clarithromycin [102]. Lee et al. [103]. found susceptibility of L. brevis KU15006 to ampicillin, gentamycin, kanamycin, streptomycin, erythromycin, clindamycin, tetracycline and chloramphenicol. A study of genome and pangenome analysis of Lactobacillus hilgardii FLUB by Gustaw et al. [104] revealed the presence of five plasmids but classified one gene as antibiotic target modifying enzyme (RimA II), besides the absence of gidB genes encoding 16 S rRNA methyltransferase, provided resistance to amino glycoside antibiotic such as streptomycin. They also reported three genes encoding proteins (GdpD, MprF and PgsA) altering the cell wall charges, and genes conferring resistance to fluroquinolones (gyrA and gyrB). However, all except (rimA (II)) are chromosomally encoded. A research by Qureshi et al. [105]. reported the absence of antibiotic resistance gene in Lactobacillus paracasei ZFM54. However, a whole-genome sequence of Lactobacillus paracasei DTA93, revealed the presence of genes related to resistance to fluroquinolones, beta lactam antibiotics and multi-drug resistance efflux pumps [56, 106]. Similarly, Pei et al. [106] reported resistance of all strains of L. paracasei to vancomycin. From a comparative genomic analysis of L. parabuchneri, antibiotic resistance genes were not detected [57].

Pabari et al. [107] also reported the resistance of lactic acid bacteria to vancomycin. Resistance of probiotics to antibiotics is also considered beneficial for patients undertaking antibiotic therapy [79]. The MAR index of the potential probiotic microorganisms ranged between 0.1 and 1, showing unsuitability of majority of them as probiotics. Three ‘lactobacilli’ strains namely Lactobacillus hilgardii ST4, Lactobacillus hilgardii ST6 and Lecticaseibacillus paracasei ST11 with MAR index of ≤ 0.2 can be considered for potential probiotics. The MAR index of > 0.2 was associated with unrestricted use of antibiotics [108]. According to a report by Reuben et al. [109]., all the ‘lactobacilli’ isolated from raw milk showed MAR index of above the threshold limit of 0.2.

Genomic analysis of lactic acid bacteria and bifidobacteria revealed the presence of acquired antibiotic resistance genes and also wide spread existence of intrinsic antibiotic resistance genes [110]. Common prevalence of tet(W) gene in lactobacillus was also reported [111]. In LABs, vancomycin resistance is frequently associated with the vanA gene or similar genetic elements. The vanA gene encodes enzymes that alter cell wall components, making vancomycin less effective by reducing its ability to bind and inhibit cell wall synthesis [112]. In addition, some LABs developed mutations in genes that encode DNA gyrase (gyrA and gyrB) and topoisomerase IV, reducing ciprofloxacin’s binding and effectiveness. They also use efflux pumps to expel ciprofloxacin from the cell or have altered membranes that limit the drug’s entry [113]. Moreover, some LABs produce bla genes (β-lactamases) that break down ampicillin, while others have altered penicillin-binding proteins (PBPs) that reduce the drug’s effectiveness. Their unique cell wall structures also contribute to their lower susceptibility to ampicillin [114]. In Lactic Acid Bacteria (LAB), resistance to streptomycin often results from mutations in the rpsL gene, which codes for ribosomal protein S12, or the rrs gene, which codes for ribosomal RNA. These mutations prevent streptomycin from binding effectively to the ribosome, making the drug less effective. Additionally, LABs may use efflux pumps to remove streptomycin from the cell, which also helps them resist the drug. Knowing these genetic factors is crucial for managing antibiotic resistance in LABs [115]. Understanding these resistance mechanisms is crucial for developing effective strategies to manage and overcome antibiotic resistance in LAB.

Hemolytic activity is the virulence factor of pathogenic microorganisms which is not desirable for probiotic microorganisms intended for use in food products. All of the tested LABs in the current study (Lecticaseibacillus paracasei, Levilactobacillus brevis, Lentilactobacillus hilgardii, and Lentilactobacillus parabuchneri) were non-hemolytic. Jose et al. [36] and Hameed et al. [116] also reported that, all the tested probiotic lactic acid bacteria were non-hemolytic. This implies that selected ‘lactobacilli’ with the probiotics properties and MAR index of ≤ 0.2 are safe to be used as probiotics.

Conclusion

This study revealed that the lactic acid isolated from the Tej samples demonstrated good probiotic potential under the in vitro conditions tested. They were able to survive at pH 2 and 3, and bile salt concentrations of 0.3, 0.5, and 1%. The LAB species also exhibited adherence, antioxidant, cholesterol assimilation, and antibacterial activity against pathogenic bacteria. The LAB species are considered safe under the in vitro conditions as they didn’t show hemolytic activity when grown on media supplemented with sheep blood. Among the 18 ‘lactobacilli’ based on their MAR index only Lactobacillus hilgardii ST4, Lactobacillus hilgardii ST6 and Lecticaseibacillus paracasei ST11 with MAR index of ≤ 0.2 are suitable for further probiotic potential. Whole genome analysis of the probiotic LAB from Tej needs to be assessed to ascertain the absence of transferrable antibiotic resistance determinants, and also the presence of virulence factors.

Author contributions

TG: Methodology, Validation and Statistical analysis, Investigation, Data curation, Writing – original draft. AD: Conceptualization, Methodology, Supervision, Review & editing, writing original draft, funding acquisition, and project administration MA: Conceptualization, Methodology, Supervision, Funding acquisition, Writing – review & editing. FT: Conceptualization, Methodology, Supervision, Writing – review & editing, funding acquisition. DJ: Conceptualization, Methodology, Supervision, and funding acquisition. AN: MALDI-TOF analysis, review and editing.

Funding

This research was supported by the Addis Ababa University 7th Round Thematic Research Fund [RD/LT076, 2019].

Data availability

All the necessary data have been included in the manuscript.

Declarations

Ethics approval and consent to participate

Ethical approval was obtained from the Institutional Review Board of the Addis Ababa University College of Natural and Computational Sciences.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.