Coalho cheese as source of probiotic lactic acid bacteria

The aim of this study was to characterize the probiotic potential of 24 lactic acid bacteria (LAB) strains isolated from artisanal Coalho cheese from Pernambuco, Brazil by in vitro tests. The gastrointestinal tract (GIT) resistance, antimicrobial activity against intestinal pathogens, autoaggregation and coaggregation capacity, cell hydrophobicity, ß-galactosidase activity, deconjugate bile salt activity for the production of bile salt hydrolase (BSH), and the sensitivity to antibiotics were evaluated. Of the 24 strains, 22 remained viable to a simulated GIT. Two LAB inhibited the growth of Listeria monocytogenes and two inhibited Escherichia coli. The autoaggregation rate ranged from 27% to 96%, and the strains were able to coaggregate with Staphylococcus aureus and E. coli reaching levels between 58% and 47%, respectively. The hydrophobicity percentage ranged from 5% to 57%. Four strains were able to produce BSH. One LAB was able to produce up to 604 Miller units of ß-galactosidase. All strains were sensitive to five antibiotics and only two were resistant to vancomycin (30μg) and norfloxacin (10g). LAB strains which were able to overcome all barriers with a reduction of only one log cycle and LAB strains which were able to produce ß-galactosidase were identified by 16S rRNA sequence analysis as Lactococcus lactis subsp. Lactis, Enterococcus durans, and Enterococcus faecium. The evaluated LAB showed promising probiotic characteristics. Strains identified as L. lactis subsp. Lactis were selected for studies involving Research, Society and Development, v. 9, n. 8, e266984958, 2020 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v9i8.4958 3 their technological potential to investigate the possible use of these microorganisms into a functional product.

their technological potential to investigate the possible use of these microorganisms into a functional product.

Introduction
The word 'probiotics' originates from the Greek word 'for life'. According to FAO/WHO guidelines, probiotics are defined as live organisms which confer a health benefit on the host when administered in adequate amounts. Probiotics are also defined as live bacteria which contribute to the regulation of immune responses and have beneficial effects on the host. Resistance to enteric pathogens, aid in lactose digestion, anti-colon cancer effect, Development, v. 9, n. 8, e266984958, 2020 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v9i8.4958 4 small bowel bacterial overgrowth, and immune system modulation are some of the beneficial actions of probiotic bacteria (Lee and Salminen, 2009).
Certain criteria need to be met by a bacterium to qualify as a probiotic: it must be harmless for ingestion (safe), able to colonize the gut epithelium, and especially be able to resist harsh conditions found in the gastrointestinal tract (gastric acidity, bile salts, pepsin, pancreatin, and other enzymes) (Lee and Salminem, 2009). Therefore, knowledge of these characteristics is important for evaluating possible positive and negative effects of probiotic consumption.
In the past, the GIT was considered the main potential source of probiotic bacteria (Espinoza & Navarro, 2010). However, the scientific community has focused attention on fermented foods and recognized the autochthonous lactic microbiota of different foods with probiotic potential. Thus, the search for new probiotic strains derived from dairy (Nuryshev, Stoyanova & Netrusov, 2016;Haghshenas, et al. 2017) and non-dairy (Saito et al. 2014;Mortezaei et al. 2020) is justified by the possibility of detecting strains with benefits to human health and with good technological performance.
Coalho cheese is a semi-hard cheese, typically produced and widely consumed in the Northeast of Brazil by coagulation of milk with rennet or other appropriate coagulating enzymes, complemented or not by the action of selected lactic acid bacteria (LAB) (Brasil, 2001). Several studies have isolated autochthonous LAB in this cheese (Santos et al. 2015;Medeiros et al. 2016;Bruno et al. 2017) searching for strains with technological and/or probiotic features, as most probiotic bacteria belong to LAB group. The main representative LAB are Carnobacterium, Enterococcus, Lactobacillus, Lactococcus, Leuconostoc, Oenococcus, Pediococcus, Streptococcus, Tetragenococcus, Vagococcus, and Weissella (Jay, Loessner & Golden, 2005).
Thus, the aims of our research were to characterize the probiotic potential of 24 autochthonous LAB strains isolated from Coalho cheese produced in Pernambuco state, Brazil.

Materials and Methods
This research was based on a laboratory study. Lactic acid bacteria (LAB) strains isolated from artisanal Coalho cheese were studied in the research. The results obtained were qualitatives and quantitatives with statistical analysis (Pereira et al.,2018). Research, Society and Development, v. 9, n. 8, e266984958, 2020 (CC BY 4. glycerol. Prior to assays, the strains were transferred to MRS broth and cultivated at 37ºC/24h at least three times in aerobic conditions.

Resistance to simulated gastrointestinal tract
A sequence of tests simulating gastrointestinal conditions were used according to Burns et al. (2011) with modifications to verify the strains' tolerance to the GIT. LAB were inoculated into saliva solution (CaCl2 0.22 g L -1 , NaCl 16.2 g L -1 , KCl 2.2 g L -1 and NaHCO3 1.2 g L -1 ) containing bovine pepsin 0.3% w/v (Sigma-Aldrich). The pH was quickly lowered to 2.0 immediately after mixing, with HCl 5N and incubated in a water bath for a period of 90 min at 37°C. After simulated saliva-gastric digestion, 1 mL of the sample was centrifuged (4.000 g, 5 min, 5°C), the supernatant was removed, and the pellet was washed twice with buffered phosphate saline (PBS) solution (pH 7.4) and resuspended to the original volume in 1% (w/v) bovine bile (Sigma-Aldrich) at pH 8.0. The cell suspension was incubated in a water bath for 10 min/37°C. Next, it was centrifuged and the cells were washed as described above and resuspended to the original volume in 0.3% (w/v) bovine bile and 0.1% (w/v) pancreatin (Sigma-Aldrich) at pH 8.0. The cell suspension was then again incubated in a water bath at 37°C/90min. Cell viability was monitored before the beginning of the first test and at the end of each stage by plating on MRS agar at 37°C/48h. The test was performed in triplicate. in diameter were made in the agar layer of plates containing pathogens, and 180µL of the CFS from each strain were placed in a well. Plates were incubated overnight at 37ºC and the diameters of the inhibition halos were measured.

Autoaggregation and coaggregation with pathogens strains
Autoaggregation and coaggregation assays were performed according to Kos et al. (2003), but modified as follows: cell suspension of each LAB (approximately 10 8 CFU.mL -1 ) was prepared to determine autoaggregation and coaggregation. Tubes containing only the LAB and tubes with LAB and pathogenic bacteria (S. aureus ATCC 25923 and E. coli ATCC 35218) were vigorously mixed (10 seconds) and then left at room temperature (20ºC) for 5 hours. An aliquot of the cell suspension was picked up each time and absorbance (OD600)

Hydrophobicity
The cell hydrophobicity was determined according to Vinderola and Reinheimer (2003). Cultures of the strains were harvested in the stationary phase by centrifugation (12000g, 5min, 5ºC), washed twice in 50mM of K2HPO4 (pH 6.5) buffer and resuspended in the same buffer. The cell suspension was adjusted to a 560nm absorbance value of approximately 1.0 (OD560 1.0) with the buffer and 3 ml of the bacterial suspensions were put in contact with 0.6ml of n-hexadecane (Merck Schuchardt, Germany) and vortexed for 120s.
After separating the two phases, the aqueous phase was carefully removed and the absorbance was measured. The decrease in the absorbance of the aqueous phase was taken as a measure of the cell surface hydrophobicity (H %), which was calculated with the formula H %=[(A0-A)/A0]*100, where A0 and A are the absorbance before and after extraction with nhexadecane, respectively. Assays were performed in triplicate. Research, Society and Development, v. 9, n. 8, e266984958, 2020 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v9i8.4958

ß-galactosidase activity
ß -galactosidase activity (ß -gal) in whole cells was determined according to the method of Miller (1972) modified by Vinderola and Reinheimer (2003). Overnight strain cultures were harvested in the stationary phase by centrifugation (12000g, 5min, 5ºC), washed twice in 60 mM Na2HPO4.7H2O/40 mM NaH2PO4 buffer (pH 7.0), inoculated (1% v/v) in MRS-lac broth and incubated at 37ºC for 24h. After the incubation period, cells were harvested and washed twice as previously described, and 560nm absorbance was adjusted to [(A420*1.75*A560 b )/(15min*1ml*A560 a )], where A560 a was the absorbance just before assay and A560 b was the absorbance value of the reaction mixture.

Bile salt deconjugation ability
The strain activity to produce bile salt hydrolase (BSH) was determined according to Zago et al. (2011). Bile salt plates were prepared by adding 0.5% (w/v) of sodium salts The inoculum of each strain in MRS without supplementation was included as a negative control.

Antibiotic resistance
The disc diffusion method was used to evaluate the antibiotic susceptibility of LAB.
Tests were done according to the criteria of the National Committee of Clinical Laboratory Standards (NCCLS, 1997) with modifications. Cells were grown in MRS broth at 37ºC for 18h to obtain a density of approximately 10 7 cells/ml. The cell suspension was inoculated in MRS agar plates with the aid of a sterile "swab". Antibiotic discs were dispensed on to the media and incubated at anaerobic conditions at 37ºC for 24h. Seven discs (LABORCLIN ® ) of antibiotics were tested: ampicillin (AM 10µg), erythromycin (E 15µg), vancomycin (VA 30µg), chloramphenicol (C 30µg), tetracycline (TE 30µg), streptomycin (S 300µg), and norfloxacin (NOR 10µg). Inhibition-zone diameters were measured after incubation and susceptibility is expressed in terms of resistance (R), moderate susceptibility (MS), and susceptibility (S), based on CLSI M100-S15 guidelines for the standard strains (NCCLS).
Each experiment was performed in triplicate.

Identification of LAB strains
The identification of LAB strains was determined at the Stab Vida (Caparica, Portugal).
Total DNA of isolates was obtained from cell culture stored on FTA TM indicating Micro card.
The identity of isolates was analysed by PCR amplification of the 16S rDNA gene and DNA Sanger sequencing (sense and antisense) of the amplified product. The identity of isolates was checked by nucleotide BLAST of the NCBI database (www.ncbi.nlm.nhi.gov/blast)

Resistance to simulated gastrointestinal tract
The evaluated LAB strains showed high resistance to the combined stress at the various steps of the simulated gastrointestinal tract. All strains survived simulated gastric juice, exposure to pH 2.5 and simulated duodenal juice (Table 1). However, only 22 LAB remained viable after the passage to simulated GIT conditions. Six strains (15, 16, 37, 126, 143 and 174) were able to overcome all barriers with a reduction until one log cycle ( Figure   1). Research, Society and Development, v. 9, n. 8, e266984958, 2020 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v9i8.4958 Several authors have reported the GIT resistance of strains isolated from different sources. Saito et al. (2014) concluded that lactobacilli strains isolated from the fermentation of cocoa survived the barrier of simulated GIT and Lactobacillus fermentum showed higher resistance than Lactobacillus plantarum to the simulated gastrointestinal digestion at pH 3, losing only approximately one log order of cell viability. Nejati and Oelschlaeger (2016) also observed that Lactococcus lactis isolated from traditional Iranian dairy products had better survival after exposure to simulated gastrointestinal tract stresses in comparison to the control Lactobacillus rhamnosus GG probiotic. In our study, major decreases in the viability of strain cells (around 4 log orders) were registered after the incubation with bile, with cell counts at the end of the simulated GIT conditions between 10 3 and 10 5 CFU.mL -1 . Han, Kong, Chen, Sun and Zhang (2017) reported poor bile resistance for L. acidophilus. All LAB strains in their study grew on the plates containing 0.3% bile salts. The strains with shorter lag time (less than 4 h) grew on plates with 1% bile salts. L. pentosaceus specifically survived in 2% bile salts. log CFU.mL -1
Salmonella typhi ATCC 6539 and S. aureus ATCC 25923S were not inhibited by the evaluated LAB.
Development of mechanisms to survive in competition with other microorganisms in the GIT complex is a desirable feature in probiotic strains. Cabral et al. (2016) also observed antagonistic activity of the lactic acid bacteria isolated from artisanal and industrial Coalho cheese samples against Escherichia coli, Klebsiella pneumoniae and Staphylococcus aureus. Bruno et al. (2017) reported inhibition zones in plates with Listeria monocytogenes, Escherichia coli, Staphylococcus aureus and Salmonella sp. In their study, seven LAB isolated from Coalho cheese showed inhibition halos around the well. All diameters of the produced halos were under 10 mm.

Autoaggregation and coaggregation
The evaluated LAB showed autoaggregation percentage ranging from 27.27% (strain 99) to 96.43% (strain 155) ( Table 1). The tested strains showed the best coaggregation result with S. aureus, which reached levels of 58.33% of coaggregated bacteria (strain 42), and for E. coli with a coaggregation result of 47.83% with strain 46 (Table 1). Research, Society and Development, v. 9, n. 8, e266984958, 2020 (CC BY 4. Our results are similar to results reported by several studies with high percentage of autoaggregation and moderate coaggregation. Janković, Frece, Abram and Gobin (2012) investigated the aggregation and coaggregation ability of three potential probiotic strains of Lactobacillus plantarum and observed that the autoaggregation rate after 24 hours of broth cultivation was at least 80%, and coaggregation with E. coli, S. Typhimurium and L.

Hydrophobicity
The evaluated LAB presented hydrophobicity levels ranging from 5.13% (strain 155) to 57.61% (strain 15) ( Table 2). Nejati and Oelschlaeger (2016) found similar results and reported cell hydrophobicity of Lactococcus lactis strains isolated from Traditional Iranian Dairy Products ranging from 5.13 to 83.68 %. In both studies, evaluation of hydrophobicity of the 8 isolated strains revealed considerable differences between the strains.

ß-galactosidase activity
A total of 20 tested strains were able to produce the ß-galactosidase enzyme. The ßgalactosidase values found ranged from 2 to 604 Miller units (strain 46) (Table 2)  On the other hand, the ß-galactosidase activity values observed for the evaluated strains were lower than those reported by Son et al. (2017), where all the LAB strains showed higher ß-galactosidase enzyme activity when grown in MRS broth containing 1% lactose than when grown in broth without 1% lactose, thereby indicating that the addition of 1% lactose could increase b-galactosidase enzymatic activity. L. plantarum showed the highest bgalactosidase enzyme activity in the study by Son et al. (3320.99 Miller units).

Bile salt deconjugation ability
All evaluated microorganisms were able to grow in the presence of sodium taurodeoxycholate, taurocholic acid and glycocholic acid, but only 17 strains were able to show full resistance to sodium glycodeoxycholate ( Table 3). The strains 46, 60, 106 and 128 showed white opaque colonies after growth in MRS-TDCA, evidencing the ability to deconjugate sodium taurodeoxycholate. Strains 106 and 128 hydrolyzed sodium glycodeoxycholate detected by halo formation around colonies after growth in MRS-GDCA. Zago et al. (2011) reported similar results in their study. All strains evaluated for these authors demonstrated the ability to hydrolyze the sodium glycodeoxycholate and sodium taurodeoxycholate.

Antibiotic resistance
All tested strains were susceptible (S) to five antibiotics: ampicillin (AM 10 µg), erythromycin (E 15 µg), chloramphenicol (C 30 µg), tetracycline (TE 30 µg) and streptomycin (S 300 µg). Only two strains (37 and 98) were resistant (R) to vancomycin (VA 75%, coaggregation with pathogens above 40%, ß-galactosidase production over 50 Miller units, BSH production, and sensitivity to antibiotics were considered for selection. Seven LAB showed four or more of these criteria and were identified as Lactococcus lactis subsp. lactis (strain 15 and 16); Enterococcus durans (strain 37); and Enterococcus faecium (strains 106, 126, 143 and 174). There is controversy in the use of Enterococcus strains in food production, and therefore strains 15 and 16 were selected for the development of a functional product.

Conclusions
The evaluated LAB showed promising characteristics to be used as probiotic strains.
The set of tests used in this study enabled selecting seven strains with desirable characteristics for any probiotic strain and as potential candidates for the formulation of new probiotic foods.
Two were selected for further trials involving the possible use of this microorganism into a functional product: strain Lactococcus lactis subsp. lactis 15 showed good resistance to gastric barriers; and strain Lactococcus lactis subsp. lactis 16 was able to deconjugate bile salts, presented high percentages of autoaggregation and coaggregation, and ß-galactosidase production. Studies involving the technological potential of all strains are being performed.
The strains shall be subjected to in vivo tests to compare the results.