Preliminary in vitro assessment of probiotic properties of Bacillus subtilis GM1, a spore forming bacteria isolated from goat milk

Abstract

Background:

Species of the Bacillus genus have a long history of use in biotechnology. Some Bacillus strains have recently been identified for food applications and industrial as safe bacteria, which mostly have been recognized as probiotic strains.

Aims:

The primary purpose of the current study was to evaluate the probiotic characteristics of Bacillus subtilis strains isolated and identified from the goat milk samples.

Methods:

After sampling from 40 goat milk and cultivation, suspected colonies were subjected to biochemical and molecular identification. Then, the confirmed isolate was assessed for in vitro probiotic tests, including hemolysis and lecithinase properties, bile salt, acid, and artificial gastric juice resistance, antioxidant activity, antibiotics susceptibility, enterotoxin genes detection, and attachment capacity to the HT-29 cells.

Results:

Among 11 suspected isolates evaluated, only one isolate was identified as B. subtilis. In vitro tests for this strain showed similar results to other probiotic strains. The B. subtilis strain was susceptible to various antibiotics. The enterotoxin genes were not detected based on PCR assay. Concerning its probiotic characteristics assessment, especially tolerance to bile salts and acidic conditions, the Bacillus strain could have the potential to consider as a probiotic.

Conclusion:

Goat milk can be recommended as a source of Bacillus isolates. Also, the isolated strain showed high adaptability to the gastrointestinal environment, relatively equal percentages of adhesion properties, and some safety aspects, having the potential to be considered as an appropriate probiotic.

Introduction

Today, foods, in addition to satisfying hunger, can also improve the mental and physical functions of consumers, well-being (Ismail et al., 2018), so over the last two decades, scientific interest in probiotics has received to increasing level (Seo et al., 2011; Sridevi et al., 2015). Also, consumer demands in food production, especially dairy-containing probiotics, have changed considerably (Ismail et al., 2018) owing to increasing knowledge about their benefits on immunity and gastrointestinal health (Abid et al., 2019).

Live microorganisms that, if administered in adequate amounts, could provide health benefits to the host are defined as probiotics (Olmos and Paniagua-Michel, 2014; Ranadheera et al., 2019). These microorganisms could affect the host beneficially by improving the balance and restoration of the gut microbiota, immunomodulation, managing lactose intolerance, reducing blood pressure and cholesterol, lowering allergic, preventing urogenital symptoms, and so on (Seo et al., 2011; Sridevi et al., 2015; Pique et al., 2019). Common probiotic bacteria are Lactobacillus and Bifidobacteria strains (Igbal et al., 2014) since they are indigenous microflora of the animals and many studies represent their beneficial properties (Lee et al., 2012), however manufacturing formulations of these probiotics have some challenges. Both of them are slow growers and sensitive to temperature, having short shelf-life in products. Moreover, these bacteria are strict anaerobes or microaerophiles; hence their handling and production require complicated procedures (Suva et al., 2016). Thus, researchers and food industries are constantly looking for new strains with better properties, such as more acid and bile tolerance, and longer shelf-life (Abid et al., 2019).

A group of highly diverse microorganisms are Bacillus species (Suva et al., 2016) that contain industrial grade strains, some of which are recognized as safe and known as probiotics (Hanafy et al., 2016). Bacillus spp. is closely related to Lactobacillus spp., and both are categorized in the same class (Elshaghabee et al., 2017). Bacillus strains have the stability against harsh conditions due to spore formation (Lee et al., 2019), which can then germinate and exert probiotic activities in the gastrointestinal tract (Kim et al., 2014). Bacillus spp. form resistant spores under environmental stress with the ability to resist physical and chemical conditions such as high temperature or pressure, air-drying activity, and UV light (Elisashvili et al., 2019). In addition to spore formation, some properties such as fast growth rate, diversity of nutrients consumption, growth in anaerobic and aerobic conditions, and high level of enzyme production give unique characteristics to B. subtilis (Sorokulova, 2013; Suva et al., 2016). Also, B. subtilis genome is sequenced, generating great amount of essential knowledge and developments of molecular techniques (Olmos and Paniagua-Michel, 2014). This bacterium has been used in fermented foods for several centuries without harmful effects, generally recognized as safe (GRAS) (Kim et al., 2014). Taking into account the beneficial properties of B. subtilis, these strains are potential probiotic candidate, and recent decade studies in which the strains were isolated from various sources have shown several proprietary strains, marketed as supplements for human consumption in Asia, Europe, and the US (Elshahabee et al., 2017). Therefore, isolations of more new strains of B. subtilis as good probiotics have received growing attention.

Milk is a highly nutritious food with different commercial origins, such as cow, sheep, camel, and goat, produced worldwide for human consumption (Hernandez-Saldana et al., 2016). The valuable nutritious contents of milk, as well as high-water activity and natural pH, provide a suitable environment for supporting the growth of different microorganisms. The easily digested goat’s milk could be an alternative dairy product to cow’s milk, today. It has gained much attention and interest (Quigley et al., 2013; Zhang et al., 2017a). It possesses other beneficial characteristics such as low allergenicity, smaller fat globules, a higher concentration of fatty acids, higher levels of iron bioavailability, and buffering capacity that potentially make it useful in medicine and human nutrition (Quigley et al., 2013; Hernandez-Saldana et al., 2016; Ranadheera et al., 2019). Moreover, goat’s milk has been interested as a suitable source for isolating microorganisms with potentially desired characteristics, so current research aimed to isolate and identify B. subtilis strains from raw goat’s milk and then screening them for probiotic traits.

Materials and Methods

Bacterial isolation from goat’s milk

Under aseptic conditions, 40 milk samples were firstly collected from goats of Varamin in Tehran province of Iran. Desired dilutions were prepared in sterile saline buffer, the samples were pretreated for 15 min at 80°C for isolation of the spore-forming bacteria. Then, 100 µL of each dilution was streaked on nutrient agar (Merck, Germany). Different colonies of bacteria based on their morphologies were selected, and sub-cultured onto nutrient agar to pure the colonies (Ragul et al., 2017; Talebi et al., 2018).

Identification of Bacillus isolates

Preliminary screening of Bacillus species was determined based on morphological and biochemical analysis. The polymerase chain reaction was done to bacterial identification by 27F (5´-AGA GTT TGA TCC TGG CTC AC-3´) and 1492R (5´-CGG TTA CCT TGT TAC GAC TT-3´) as forward and reverse primers, respectively. The obtained sequence was submitted to the National Center for Biotechnology Information (NCBI) (Davati et al., 2015; Zulkhairi Amin et al., 2020).

Screening of probiotic properties of the isolate

Hemolysis assay

For determination of hemolysis pattern, Staphylo-coccus aureus ATCC 25923 as a control strain and GM1 isolate were streaked on blood agar (Merck, Germany) plates with 5% (W/V) defibrinated blood, and incubated at 37°C for 24 h (Manhar et al., 2016).

Lecithinase activity

A loop-full of an overnight culture of selected bacteria and Bacillus cereus ATCC14579 as positive control were streaked on egg yolk agar (Biomark, India) and incubated at 37°C. The plates were checked for lecithinase activity after 24-48 h (Lakshmi et al., 2017).

Bile salts tolerance assay

Bile salts (Sigma-ldrich, USA) tolerance of GM1 isolate was evaluated by inoculating 100 μL of bacterial suspension in nutrient broth (Biolife, Italy) comprising 0.3% salts and incubated for 8 h at 37°C. Before and after the incubation period, the absorbance at the wavelength of 600 nm was recorded. The absorbance of a medium without bile salts was measured as a control sample, and the rate of inhibitory was calculated as follows:

$\mathrm{C\; inh}=\frac{\left(\mathrm{T8}-\mathrm{T0}\right)\mathrm{control}-\left(\mathrm{T8}-\mathrm{T0}\right)\mathrm{treatment}}{\left(\mathrm{T8}-\mathrm{T0}\right)\mathrm{control}}$

T₀: The OD600 at time zero

T₈: The OD600 after 8 h incubation

For a suitable probiotic candidate, C_inh (inhibitory) of less than 0.4 was considered significant (Ebnetorab et al., 2020).

Resistance to artificial gastric juice

Gastric juice resistance of the selected isolate was determined using nutrient broth with 0.3% w/v pepsin (Sigma-Aldrich, USA) and 0.5% w/v sodium chloride (Merck, Germany) adjusted to pH 2.5 with HCl (Merck, Germany) 5N and to pH 7 as control. The cell pellet of the fresh culture of B. subtilis was washed with phosphate buffered saline (PBS), inoculated in artificial gastric juice and incubated at 37°C for 4 h. The viability percentage of cells was determined by culturing on nutrient agar and calculated as the log CFU/ml before (0 h) and after (4 h) incubation.

N₀: The initial viable cells

N_t: The final viable cells (Lavanya and Dayakar, 2017)

Acid tolerance

The tolerance of B. subtilis strain to acid was assessed in several pH solutions. The strain was grown in nutrient broth, and the cell pellet of an overnight culture was harvested by centrifugation, washed, and added to PBS with pH values of 2 and 4. The plate count method was used for the determination of viable cells after 4 h incubation at 37°C, and the viability rate was calculated as follows:

Where,

N₀: The counts of viable bacteria at time zero

N_t: The counts of viable bacteria after 4 h treatment in acidic conditions (Thirabunyanon and Thongwittaya, 2012)

Auto-aggregation

Auto-aggregation characteristic of bacteria was measured based on the method of Shivangi et al. (2020) with some modifications. Cell pellet of overnight culture was harvested, washed, and resuspended in PBS until the optical density of suspension reached 0.3±0.05 at 600 nm. The cell suspension vortexed for 10 s and incubated for 4 h and 24 h at 37°C. After incubation time, optical density at 600 nm was recorded to calculate auto-aggregation as follows:

$\mathrm{Auto}-\mathrm{aggregation}\left(\mathrm{\%}\right)=\left(1-\frac{\mathrm{At}}{\mathrm{A0}}\right)\times 100$

Where,

A₀: The absorbance at time zero

A_t: The absorbance at the indicated time (4 h or 24 h)

Hydrophobicity assay

The selected strain was grown in nutrient broth (18 h, 37°C). Then, harvested pellets were washed with PBS, resuspended in 2 ml of the same buffer, and their optical density was measured at 600 nm (A0). About 2 ml of bacterial suspension was added to 2 ml of chloroform (Merck, Germany), ethyl acetate (Merck, Germany), and toluene (Merck, Germany), then mixed for 5 min. After the separation of a mixture into two phases, the optical density of the aqueous phase was measured at 600 nm (A1). Hydrophobicity (%) was estimated by the following equation (Kuebutornye et al., 2020):

$\mathrm{Hydrophobi\; city}\left(\mathrm{\%}\right)=\left(1-\frac{\mathrm{A1}}{\mathrm{A0}}\right)\times 100$

Adherence assay

The attachment ability of the selected isolate to the epithelial cells was performed with human colon adenocarcinoma cells obtained from the Iranian Biological Resource Center (HT-29 IBRC C10097). First, 1 10⁵ cells/ml were seeded in a 24-well plate and grown at 37°C (humidified atmosphere, 5% CO₂) till to 80-90% confluency. Before adding the bacterial suspension, the cells of each well were washed with PBS (Bioidea, Iran) to remove antibiotics. Then, B. subtilis GM1 was inoculated to cells in each well and incubated for 3 h (37°C, 5% CO₂). Finally, the non-adherent bacteria were removed by washing with buffer, and a 0.25% trypsin-EDTA (Bioidea, Iran) solution was used to detach the adherent cells. The colony count method was used for the enumeration of the attached bacteria.

Where,

N₀: Initial colony counts

N_t: Final colony counts after 3 h (Kim et al., 2014)

Antibiotic resistance

Susceptibility of the GM1 strain to antibiotics was characterized according to the recommendation of the Clinical and Laboratory Standards Institute (CLSI) for the disc diffusion method. The antibiotic discs (Padtan Teb Co., Iran) used in this assay were tetracycline (30 μg), chloramphenicol (30 μg), vancomycin (30 μg), gentamycin (10 μg), erythromycin (15 μg), streptomycin (10 μg), clindamycin (2 μg), and penicillin (10 μg). Cells from bacterial culture were seeded on Mueller-Hinton agar, discs were placed, and the diameter of the inhibition zone was determined after incubation at 37°C for 24 h (Talebi et al., 2018).

Detection of toxic genes

First, a simple boiling method was used for DNA extraction. The fragments of the enterotoxin genes non-hemolytic gene (nhe), and hemolysin BL (hbl) were amplified from the isolated samples, and B. cereus ATCC14579 was used as a positive control. The sequences of the used primers are provided in Table 1 (Kim et al., 2011).

Table 1.

Primers and PCR conditions used in this study

Primer	Target size (bp)	Sequences (5´-3´)	Reaction conditions	References
hblA F hblA R	1154	AAGCAATGGAATACAATGGG AGAATCTAAATCATGCCACTGC	94°C, 120 s → (94°C, 1 min → 56°C, 1 min → 72°C, 2 min) 35 cycles → 72°C, 300 s	Kim et al. (2011)
hblC F hblC R	740	GATACCAATGTGGCAACTGC TTGAGACTGCTCGCTAGTTG	94°C, 120 s → (94°C, 1 min → 58°C, 1 min → 72°C, 2 min) 35 cycles → 72°C, 300 s	Kim et al. (2011)
hblD F hblD R	829	ACCGGTAACACTATTCATGC GAGTCCATATGCTTAGATGC	94°C, 120 s → (94°C, 1 min → 58°C, 1 min → 72°C, 2 min) 35 cycles → 72°C, 300 s	Kim et al. (2011)
nheA F nheA R	499	TACGCTAAGGAGGGGCA GTTTTTATTGCTTCATCGGCT	94°C, 120 s → (94°C, 1 min → 56°C, 1 min → 72°C, 2 min) 35 cycles → 72°C, 300 s	Kim et al. (2011)
nheB F nheB R	769	CTATCAGCACTTATGGCAG ACTCCTAGCGGTGTTCC	94°C, 120 s → (94°C, 1 min → 54°C, 1 min → 72°C, 2 min) 35 cycles → 72°C, 300 s	Kim et al. (2011)
nheC F nheC R	581	CGGTAGTGATTGCTGGG CAGCATTCGTACTTGCCAA	94°C, 120 s → (94°C, 1 min → 58°C, 1 min → 72°C, 2 min) 35 cycles → 72°C, 300 s	Kim et al. (2011)

DPPH scavenging activity assay

Antioxidant activity of bacterial isolate was carried out by mixing 100 μL of the sample (cell-free supernatant) with an equal volume of DPPH solution (Merck, Germany) (0.2 mM) in a 96-well plate and left in darkness for 30 min at 30°C. The absorbance of samples was determined at 517 nm using Synergy HTX multimode reader. Control was a sample containing deionized water with DPPH, and the scavenging ability was measured as the following equation:

$\mathrm{DPPH\; scavenging\; ability}\left(\mathrm{\%}\right)=\left(\frac{\mathrm{Ac}-\mathrm{As}}{\mathrm{Ac}}\right)\times 100$

A_s: Absorbance of the sample

A_c: Absorbance of the control (Xing et al., 2015; Talebi et al., 2018)

Statistical analysis

IBM SPSS Statistics Software (ver. 26.0, SPSS Inc., USA) was used to analyze results that were presented as mean±SD. Analysis of variance followed by the Post Hoc method (Duncan) was used to find the differences between means (significant difference, P<0.05). All experiments were performed in triplicates.

Results

Bacterial isolation and identification from goat milk

Among 11 pure colonies of bacteria, only 1 isolate, due to the results of Gram-staining besides the biochemical tests, was generally indicated as B. subtilis strain. This isolate was further analyzed by sequencing of 16S rRNA gene. So, this strain was finally identified with biochemical and 16S rRNA partial sequencing tests as B. subtilis GM1. It was deposited under accession number MK818228.1 in the GenBank database (https://www. ncbi.nlm.nih.gov/nuccore/MK818228.1) and then was characterized for its probiotic capacity. As a control, B. subtilis G2 recovered from the commercial probiotic product was used to compare the results. Table 2 shows characteristics of B. subtilis GM1 and B. subtilis G2.

Table 2.

Biochemical characterization of Bacillus isolated from goat milk and commercial product

Tests	GM1	G2	Tests	GM1	G2
Gram-staining	+	+	Glucose	+	+
Spore formation	+	+	Maltose	+	+
Starch hydrolysis	+	+	Rhamnose	+	+
Simon’s citrate	+	+	Nitrate reduction	+	+
Methyl red	-	-	VP	+	+
Catalase	+	+	Growth at 50°C	+	+
Urease	-	-	SIM	-/-/+	-/-/+

Plus sign (+) indicates the positive, and minus sign (-) indicates the negative results of reaction/test

Hemolysis and lecithinase activity

B. subtilis GM1 showed α-hemolytic activity. For lecithinase production assay, there was no precipitation around the colony of bacterial strain even after 48 h of incubation, indicating that B. subtilis GM1 is lecithinase negative, while B. cereus ATCC14579 was lecithinase positive (Table 3).

Table 3.

Hemolysis, lecithinase activities, and the results of the viability of Bacillus species isolated in different conditions

Bacteria	Growth at (%)			Biochemical activities
	pH 2	pH 4	Gastric juice tolerance	Bile 0.3% (rate of inhibition)	Lecithinase activity	Hemolytic activity
B. subtilis strain GM1	80.12±0.24a	89.82±0.84a	75.29±1.25a	0.33±0.02a	-	α
B. subtilis strain G2	73.88±0.80b	72.15±0.61b	90.71±0.85b	0.36±0.02a	-	α

Data are presented as mean±SD, and n=3. Mean within the same column followed by different superscript letters differ significantly (P<0.05). Minus sign indicates the negative results. Alpha symbol (α) indicates partial hemolysis

Resistance of bacteria to acid and bile

To evaluate the tolerance of isolate to acid and bile salts, B. subtilis GM1 was subjected to two ranges of acidic pH (2 and 4) and bile salts (0.3% concentration) for 4 h and 8 h, respectively. The results showed that more than 80% of the bacterial cells survived at two pH values for 4 h, and the Bacillus strain was also resistant to bile salts after 8 h of exposure (Table 3).

Gastric juice tolerance

Exposure of the selected isolate to artificial gastric juice (pH 2.5) revealed a viability count of more than 70%. The result showed the ability of the strain to pass through the stomach conditions (Table 3).

Hydrophobicity and aggregation assays

To get further insight into the probiotic characteristics of the isolate, its surface properties were evaluated. The results of using chloroform, toluene, and ethyl acetate for evaluation of the hydrophobic potential of B. subtilis strain are shown in Table 4. Also, auto-aggregation which is strongly correlated with adherence to the gastrointestinal system, showed that the selected isolate had about 42% cell adhesion ability at the first 4 h; However, cell adhesion increased to 90.25% after 24 h of incubation.

Table 4.

Surface characteristics of the Bacillus isolates

Strain	Autoaggregation (%)		Hydrophobicity (%)			HT-29 attachment
	4 h	24 h	Chloroform	Toluene	Ethyl acetate
B. subtilis GM1	41.97±2.26a	90.25±1.04a	67.21±1.10a	62.69±2.08a	50.39±1.08a	49.00±0.70a
B. subtilis G2	48.83±0.84b	72.46±0.98b	61.32±1.76b	60.89±4.95a	62.19±1.02b	50.42±0.70a

Cell surface characteristics are presented as mean±SD, and n=3. Mean within the same column followed by different superscript letters differ significantly (P<0.05)

Adhesion to HT-29 cell line

The adhesion of B. subtilis GM1to HT-29 cells was studied. The results demonstrated that B. subtilis G2 (commercial probiotic strain) and B. subtilis GM1 presented relatively equal percentages of adhesion properties (Table 4).

Safety assessment

Results of the antibiotic resistant test showed the susceptibility of B. subtilis GM1 to all antibiotics using different mechanisms, including cell wall inhibitors and protein synthesis inhibitors (Table 5). Also, 44.80 ± 1.68 and 41.42 ± 2.11 antioxidant activity was recorded as a result of radical scavenging DPPH assay for cell-free supernatant of B. subtilis GM1 and G2, respectively. In addition, all six enterotoxin genes were detected in B. cereus ATCC14579 as a reference strain with the PCR method but were not observed in the isolated strain (Fig. 1).

Table 5.

Antibiotic-resistant of the B. subtilis isolates

Strain	Chloramphenicol	Tetracycline	Erythromycin	Streptomycin	Vancomycin	Gentamycin	Clindamycin	Penicillin
B. subtilis GM1	32.00±3.00S	25.00±3.00S	27.33±2.52S	20.33±3.05S	23.00±2.00S	24.00±1.73S	20.33±1.53S	32.00±2.65S
B. subtilis G2	31.00±1.00S	30.67±0.58S	27.33±1.15S	17.33±1.15S	22.33±0.58S	24.67±0.58S	18.67±0.58S	29.00±1.73S

Data are presented as mean±SD for the zone of inhibition diameter (mm), three replications. S: Sensitive, and R: Resistance (No resistant isolate was detected)

Fig. 1.

PCR products of enterotoxin genes from the isolated Bacillus subtilis strain and Bacillus cereus. Lane 7: 1 kb DNA ladder; Lanes 1, 3, and 5: hblA, hblB, and hblC; Lanes 2, 4, and 6: Negative results of hbl genes from isolated strain; Lanes 8, 10, and 12: nheA, nheB, and nheC; and Lanes 9, 11, and 13: Negative results of nhe genes from isolated strain

Discussion

Recently, consumers’ interest in products containing probiotics has risen owing to their beneficial effects (Yadav et al., 2015; Abid et al., 2019). So, the natural products and food industries are constantly looking out for strains with probiotic potential (Abid et al., 2019). Among the microorganisms, the species of genus Bacillus, for their varied range of physiological properties and capability to the production of different types of metabolites, antibiotics, enzymes, and so on, are utilized in most the pharmaceutical, agricultural, and industrial processes (Celandroni et al., 2019); Hence isolation of Bacillus strains and selection of them according to their ability, are carried out in several laboratories all over the world (Milian et al., 2014).

Goat milk, as a valuable natural product, is known to have a buffering capacity and better digestibility than cow and human milk (Atanasova et al., 2014). It contains complex and rich autochthonous microbiota that, the important genera of microbiota in goat milk belonged to lactic acid bacteria (Quigley et al., 2013; Hernandez-Saldana et al., 2016; Makete et al., 2016; Pisano et al., 2019), Although, there are few studies about the microbial community of milk in different lactation phases (McInnis et al., 2015; Niyazbekova et al., 2020). Also, it should be noted that geographic locations, genetic specificity, feedings, milking equipment, milk transportation, and storage are essential factors that could affect the microbial composition of milk (Niyazbekova et al., 2020). The predominant bacterial flora reported from the farms of the Languedoc Roussillon and Midi-Pyrenees regions in France belonged to Staphylococcus, Serratia, and Arthrobacter genera (Tormo et al., 2011). Also, the cell-free supernatant of all eight strains of Bacillus amyloliquefaciens isolated from Aloqt (goat milk-based product) possesses inhibition properties against Escherichia coli and S. aureus (Hanafy et al., 2016). Obtained results by Abid et al. (2019) support that Bacillus tequilensis GM, isolated from Tunisian spontaneously fermented goat milk, could be a probiotic candidate strain. To the best of our knowledge, there is no research on the isolation of B. subtilis with probiotic characteristics from goat milk. Thus, the main objective of the current study was to assay the probiotic properties of the newly isolated Bacillus strain via several in vitro tests.

Probiotics, for exerting their effects, have to remain alive in the host during a pass from the stressful conditions of the upper intestine that contains bile and the acidic pH of the stomach (Makete et al., 2016). Because of the toxic effect of bile salts on living cells, tolerance to these salts is a primary characteristic of probiotic isolates (Ragul et al., 2017). In this study, the isolated strain showed tolerance to two sets of acidic pH values 2 and 4 after 4 h. Also, it revealed viability through the artificial gastric juice that showed strain tolerance to stomach conditions. In addition, the results showed the resistance of isolate to bile salts. Our findings are similar to the reports of Wang et al. (2010), Lee et al. (2017), and Zulkheiri-amin et al. (2020), who noted the ability of Bacillus species to remain viable in the digestive tract.

The potency of probiotics to the attachment to the target sites for expressing optimal functionality is another important property. The adherent probiotic strains can inhibit the colonization of pathogens and establish competition in the gastrointestinal tract (Mahmoudi et al., 2019). Several in vitro tests, such as the hydrophobicity of the cell surface, auto-aggregation properties, and Caco-2 or HT-29 adhesion assays, have been considered to evaluate the colonization ability of a strain to interact with the host (Sagheddu et al., 2019). To find the adhesion ability of bacteria, the cell surface hydrophobicity assay could offer a competitive advantage to probiotic strains during the adhesion, which helps them colonize in the gastrointestinal tract (Shivangi et al., 2020). Here, the adhesion capacity was observed 67.21 ± 1.10% in chloroform, 62.69 ± 2.08% in toluene, and 50.39 ± 1.08% in ethyl acetate for B. subtilis GM1. The adhesion capacity of the isolate was approximately in line with Fc6 and much higher than Fc3 and Fs1 when compared to the report of Kavitha et al. (2018). Therefore, the isolated strain showed good affinity to different solvents, implying that it possesses hydrophobic cell surface characteristics. In contrast, in a study by Mohkam et al. (2016), most of the isolates showed good affinity to xylene, still, they were weak electron donors (basic characteristic) and weak electron acceptors (acidic characteristic), which indicated they had low hydrophilic surface. Auto-aggregation, the bacterial aggregation between microorganisms of the same strain, is important in the human gut (Lim et al., 2021). B. subtilis GM1 showed auto-aggregation ability of 90.25 ± 1.04 after 24 h. The auto-aggregation percentages of B. subtilis MKHJ 1-1, B. subtilis p223, B. clausi ATCC 700160 were reported to be 91.32 ± 0.74%, 86.03 ± 2.46%, and 93.42 ± 0.86% after 24 h incubation, respectively. So, our finding is nearly in line with the auto-aggregation ability of previously reported Bacillus strains (Jeon et al., 2017; Lim et al., 2021).

Moreover, in vitro studies using cell lines such as HT-29 and Caco-2 have helped evaluate the efficacy of gastrointestinal conditions and probiotic bacteria on the adhesive ability of strains (Monteagudo-Mera et al., 2019). The adhesion ability of the commercial and selected strain was assessed using the HT-29 cell lines. Based on the adhesion score percentage, B. subtilis GM1 isolate was a good adhesive strain, and there was no significant difference with B. subtilis G2 as the control strain. Still, the HT-29 cell adhesion rate of B. subtilis GM1 strain was higher than that of Bacillus isolates reported on Caco-2 cell line by Talebi et al. (2018). Based on the results emerging from their study, 31.8 ± 0.5, 27.4 ± 0.3, and 23.9 ± 0.7% of 437F, 1630F, and 1020G strains can adhesion to the Caco-2 cell line, respectively. The type of cell line used (Caco-2 or HT-29), and the mechanisms by which the bacteria interact with the superficial components of intestinal cells could affect the adhesion capacities of strains (Fonseca et al., 2021).

Moreover, medical and food industries are common purposes for probiotics usage, so evaluating bacterial safety for selecting one strain as a probiotic is necessary (Deng et al., 2021). In vitro tests were usually used for the safety assessment of probiotics such as hemolytic activity, the synthesis of certain enzymes, production of enterotoxin genes and biogenic amines, as well as transferability of antibiotic resistance genes (AlGburi et al., 2016; Sagheddu et al., 2019). Lecithinase production and hemolytic activity are associated with the virulence of bacterial strains (Lakshmi et al., 2017). The current report confirmed that the selected strain did not have lecithinase activity. Also, the isolate showed α-hemolytic activity that γ- and α-hemolysis is considered safe (Zulkhairi Amin et al., 2020). The current results correlate with reports from Sorokulova et al. (2008), Kuebutoraye et al. (2020), and Park et al. (2020). In comparison, Duc et al. (2004) showed lecithinase and hemolysis activity in three products based on B. cereus probiotic (Bactisubtil, Subtyl, and BioSubtyl ^DL). Even though the isolates with hemolytic or lecithinase activity may be considered unsafe for personal healthcare and food applications but it is essential to know that these products have not been documented as probiotics until elimination or modification of these factors have done and confirmed that there are no harmful effects for eukaryotic cells (AlGburi et al., 2016).

Identification of antibiotic susceptibility characteris-tics is another factor that should be considered for the safety of probiotics. The transferability of resistance gene(s) from probiotic bacteria to other ones, including pathogenic or commensal bacteria, is also essential (Zhang et al., 2017b). However, resistance to the antibiotics was reported for some probiotics (AlGburi et al., 2016; Imperial and Ibana, 2016; Li et al., 2020), but in our study, strain GM1 showed sensitivity to all antibiotics, as listed in Table 5, which results were similar to the research of Thirabunyanon et al. (2012), Thankappan et al. (2015), and Lefevre et al. (2017) which they reported some Bacillus strains as a probiotic candidate that be sensitive to antibiotics.

The occurrence of enterotoxin genes was done by the PCR method, and results showed that B. subtilis isolate could not produce selected toxins. Our findings are the same as those obtained by Ouoba et al. (2008), Lee et al. (2016), and Mohkam et al. (2019). So, the absence of enterotoxin genes, lecithinase, and hemolytic activities suggests that the GM1 strain has no risk factors for human health.

Membrane phospholipids of intestinal epithelial cells could be affected by high levels of free radicals that antioxidant potential of probiotics or inducer characteristics of them for signaling intrinsic antioxidant defense could be affecting the oxidative status of the gut (Zolotukhin et al., 2017). The Bacillus strain represented the antioxidant activity, and the result agrees with that found for Bacillus atrophaeus and Bacillus safensis by Talebi et al. (2018).

In conclusion, although previous studies have carried out the characterization of raw goat’s milk that the most prevalent genera in raw goat milk have belonged to lactic acid bacteria, to the best of our knowledge, no study has investigated the probiotic potential of B. subtilis isolated from goat milk. So, the present study has revealed that B. subtilis GM1, isolated from goat milk, displayed potential probiotic characteristics. Other in vitro and in vivo properties of this strain such as enzymatic activity, co-aggregation, antimicrobial activity, biofilm formation, cholesterol reduction, and animal models must be evaluated in future for the final decision about its application as probiotic strain.

Conflict of interest

The authors declare that they have no competing interests.