Functional Properties (Acid and Bile Tolerance) and Antibiotic Susceptibility of Lactic Acid Bacteria Isolated from Newborn Calves for the Design of a Probiotic Product

Diarrhea in young calves in dairy farms is one of the main causes of economic losses, morbidity and mortality. The use of probiotic products as feed additives or complements is a novel alternative for the prevention of intestinal syndromes. In order to include beneficial bacteria in the design of a probiotic product, their functional and safety characteristics must be studied. The aim of this work is to evaluate the behavior of the strains in some “in vitro” gastrointestinal conditions such as acid stress and bile salts in the specific physiological concentration of young calves. The antibiotic susceptibility of a group of lactic acid bacteria from calves which were identified due to their beneficial properties was also studied. The strains, genetically identified and used for the resistance assays were: Lactobacillus johnsonii CRL1692, CRL1693, CRL1699, CRL1700, CRL1701 and CRL1706; L. amylovorus CRL1697; L. murinus CRL 1695 and CRL1705; L. mucosae CRL1696 and CRL1698; L. salivarius CRL1694 and CRL1702; and Enterococcus faecium CRL1703. The results of gut resistance assays showed that all the strains were resistant to pH 4 and to a bile salts concentration of less than 0.5%. However, some of them were sensitive to pH 2. The most pH-sensitive strains were found to be L. johnsonii and L. amylovorus, and enterococci. However, pre-treatment at low pH increased the growth rate of the L. salivarius strains. The minimal inhibitory concentration showed that the strains were sensitive to Tetracycline, Erythromycin, Chloramphenicol and Ampicillin, while most of them were resistant to Kanamycin. The results allowed the selection of the most adequate strains to be included in a probiotic product that can be utilized most successfully in young calves.


Introduction
One of the main causes of mortality in newborn calves of dairy farms is neonatal diarrhea, which causes severe economic losses [1][2][3][4]. As a novel alternative in this field, the use of probiotics as feed additives or complements is being proposed for the prevention of intestinal infections by restoring the balance of the microbiome. Probiotics are defined as "live microorganisms which when administered in adequate amounts confer a beneficial health on the host" [5]. In a previous work, fourteen Lactic Acid Bacteria (LAB) strains were selected for their beneficial properties to include some of them in the design a new probiotic product for newborn calves [6]. Later, the evaluation of the technological properties of most strains was also performed [7]. At the same time, functional properties and others related to the safe status of the strains must be determined for the final design of such product [8]. As the probiotic supplement is going to be administered orally, the bacterial constituents should be resistant to the conditions of the gastrointestinal tract (GIT). Moreover, the number of live organisms that reach the gut should be large enough to produce their beneficial effect on the host [9]. On the other hand, LAB can be reservoirs of resistance genes, which could be transferred to other bacteria and even to human pathogens [10], which is why some of their safety-related characteristics should be studied. The probiotic

Research Article
Functional Properties (Acid and Bile Tolerance) and Antibiotic Susceptibility of Lactic Acid Bacteria Isolated from Newborn Calves for the Design of a Probiotic Product the determination of minimal inhibitory concentration (MIC), the EFSA (European Food Safety Authority) Panel on Additives and Products or Substances used in Animal Feed (FEEDAP) has proposed epidemiological breakpoints to evaluate and identify phenotypic resistance. Moreover, the European Union has strongly recommended the study of the antibiotic susceptibility of bacteria included in feed additives for veterinary use and later used in the Food Chain.
The aim of this work was to evaluate some of the functional (acid and bile tolerance) and safety (antibiotic susceptibility) related properties of a LAB strains group previously selected for their beneficial characteristics to be used in the design of a probiotic product to prevent diarrhea in newborn calves.

Resistance to gastrointestinal conditions
The strains were washed twice and the pellets were re-suspended in saline solution at neutral pH 7.0 (control) or at pH 2 or pH 4 adjusted with 1N HCl (Anedra, San Fernando, Argentina) and incubated for 90 min at 37ºC (the estimated time that the liquid food remains in the abomasum during the calves' gastric digestion) [23]. After incubation, the bacterial cells were washed three times in standard saline solution (0.9% NaCl, Anedra, San Fernando, Argentina). These cultures were adjusted to 0.9-1 Optical Density at 560nm and later 2 µl corresponding to 5 x 10 8 CFU/mL) were inoculated into polystyrene microplates (Deltalab SL, Barcelona, Spain) containing MRS (Merck, Darmstadt, Germany) broth supplemented with different bovine oxbile concentrations (0.1%, 0.3%, 0.5% and 1%) (Fluka-70168, Sigma-Aldrich, St. Louise, USA) and also into MRS pH 4 (adjusted with 1N HCl). The bacterial growth was determined by changes in the Optical Density (OD) at 560nm in a microplate reader (VersaMax Tunable Microplate reader, Sunny Valley, USA) at different time periods (3, 5, 7, 12, 20 and 24 h). Final ΔOD was calculated as the increase in OD between OD 0 (OD 0 is OD at t=0) and OD 24 (OD 24 is OD at t=24). The growth curves were performed from the OD of each time period. The growth rate was calculated as: µ/h (growth rate); µ = ln2/g; generation time g = (t 2 -t 1 ) log2/ (log OD 2 -log OD 1 ).

Statistical analyses
Growth determinations were performed in triplicate. Data express the mean ±SD. The experimental results were used to create the Main Effects Plot using the Minitab 16 Statistical Software.

Antibiotic susceptibility
The strains selected to determine the antibiotic susceptibility were those expressing beneficial properties: surface characteristics (hydrophobicity degree and autoaggregation patterns), antagonistic activity (production of hydrogen peroxide, organic acids, or bacteriocins) and functional properties (resistance to acid and bile salts). In all cases, only one of each of the species of Lactobacillus under study was included in these assays.

Minimum inhibitory concentration (MIC):
The Minimum Inhibitory Concentration (MIC) was determined by the microdilution method in solid media in LSM agar (LAB Susceptibility test medium) as the culture medium [25]. The bacteria assayed at this stage were: L. salivarius CRL1694, L. amylovorus CRL1697, L. johnsonii CRL1693, L.mucosae CRL1696 and L. murinus CRL1695. The following antibiotics and concentration ranges were assayed: Vancomycin Antibiotics were stored at -20°C until the preparation of stock solutions, re-suspending antibiotics in distilled water or methanol (Cicarrelli, San Lorenzo, Argentina) for those not soluble in water (Chloramphenicol, Erythromycin, Rifampicin), and dilutions were performed in twofold series according to CLSI specifications [27]. Agar plates were prepared with 1 mL of each of the antibiotic solutions and 9 mL of LSM agar melted and cooled to 45±5°C. The bacterial inoculum was adjusted to 0.16 to 0.2 OD at 625nm, corresponding to approximately 3x10 8 CFU/mL and to McFarland standard 1, following the procedures proposed by ISO 10932/IDF 233 described in Shao et al. [28]. In order to determine the effect of different bacterial concentrations (10 5 CFU/mL, 10 6 CFU/mL and 10 7 CFU/mL) in the susceptibility assays, serial dilutions in saline solution were prepared and added to the plates. All the plates were inoculated with 2 µL of bacterial suspensions and incubated for 48 h in microaerophilic conditions. MIC was calculated from the lowest antibiotic concentration that caused inhibition of the microorganism. The interpretation of the results was performed was performed on the basis of the EFSA document [15], according to the results obtained with the highest microorganism concentration. The L plantarum ATCC14917 strain was also used as quality control. All the antibiotic assays were performed in triplicate.

Resistance of beneficial LAB strains to gastrointestinal conditions (acid and bile salts)
The resistance of beneficial LAB strains to bile salts after pretreatment at pH 4 and pH 2 was determined in all the strains used. Growth curves were plotted to compare the behavior of each strain, and also those from the same species in a combined figure. The standard deviation was not included in the graphs, except for L. salivarius CRL1702 in order to allow a better interpretation of the results. The resistance of the two L. salivarius (CRL1694 and CRL1702) and L. mucosae (CRL1696 and CRL1698) strains to the acid conditions and bile salts was similar for each species (Figures 1a,1b). However, in the case of L. murinus (CRL 1695 and CRL1705), the two strains showed a different behavior: CRL 1705 grew at pH 4 and 1% bile salts, but both strains proved to be more resistant after pretreatment at pH 2. In the L. salivarius strains, acid pretreatment improved their growth at all bile concentrations used (with the exception of 1% where the strains failed to grow) and also at pH 4. The two L. mucosae strains grew in all conditions assayed when compared with controls; however, their growth was lower than all the other species ( Figure 1c). The growth curves of Enterococcus and L. amylovorus strains are shown in Figures 2a,1b. The six L. johnsonii strains exhibited a different behavior, with two different patterns or profiles. The strains CRL 1692, CRL1693 and CRL1700 pretreated at pH 4 showed a higher growth, in contrast with CRL 1966, CRL1701 and CRL 1706, which were stimulated when the acid pretreatment was at pH 2, as indicated in Figure 3.
All the strains assayed increased the lag phase length in bile salts after the pretreatments at low pH, either 2 or 4.
When all the strains were pretreated at pH 2, 4 or 6.5 and later transferred to evaluate their growth at pH 4, Lactobacillus johnsonii and L. murinus CRL1695 strains or enterococcus did not grow, in contrast to both L. salivarius and L. mucosae strains, and L. murinus CRL1705, which grew in acid conditions, as indicated in Table 1. The strains that were able to grow at higher bile salts concentrations were L. mucosae CRL1696 and CRL1698, L. amylovorus CRL 1697, E. faecium CRL 1703 and L. murinus CRL 1705; the other strains grew at a bile concentration of 0.5% or less. Most of the strains treated in acid conditions (pH 4) and later transferred to MRS-0.1%, 0.3% and 0.5% bile did not modify their growth, but in some cases their growth rates were lower (L. murinus CRL 1695 L. mucosae CRL1696 and CRL1698), as shown in Table 2. In addition, the growing and growth rate of L. murinus CRL 1695, L. salivarius, L. amylovorus CRL1697 and enterococcus strains pretreated at pH 2 were similar to those observed after pH 4 pretreatment. The growth rates of L. salivarius and L. murinus increased in MRS-bile salts pretreated at pH 2.
The application of the main effect analysis to the growth and growth rate data of all LAB strains used, analyzing strain, bile salt concentrations, acid pretreatment and growth at different pH values, resulted in the plot in Figure 4. L. salivarius CRL1694 and CRL1702 proved to be the strains with higher growth in all conditions studied. In contrast, L. mucosae CRL1696 and CRL1698, L. johnsonii CRL1699 and L. johnsonii CRL1701 showed a lower growth. Pretreatment in acid conditions (pH 2-pH 4) did not affect growth at pH 4 or in bile salts of all the strains. The highest bile concentration exerted a higher negative effect on the growth of all the strains.
With respect to the growth rate, the same pattern was obtained for all the strains (data not shown).

Antimicrobial susceptibility testing
In this paper, the evaluation of the most suitable media to determine the antibiotic susceptibility or resistance of beneficial LAB isolated from calves' feces was carried out. The disk diffusion assay showed that Gentamicin, Ciprofloxacin and Cephalexin were the antibiotics showing different patterns according to the culture media, because different inhibitions zone diameters were obtained in MRS, LAPTg and LSM. Also, Rifampicin, Erythromycin and Ampicillin Sulbactam showed different diameters of the inhibition zone. Minomycin was the antibiotic that caused the highest inhibition in all strains assayed. There were no inhibition halos for Sulfamethoxazole+Trimethoprim and Nalidixic Acid, as shown in Figure 5. Considering the results for each strain, L. mucosae CRL1696 grew poorly in LMS medium, L. amylovorus CRL1697 was the most sensitive, and L. johnsonii CRL1693 and L. amylovorus were inhibited by Vancomycin. These results are summarized in Table 3.
The MIC values obtained in solid media were compared with those suggested by EFSA [15], as shown in Table 4. All the strains were sensitive to Erythromycin, Ampicillin and Chloramphenicol but were resistant to Kanamycin (except for L. mucosae CRL1696). The MIC values of all the strains were slightly higher for Oxytetracycline than for Tetracycline, except for L. johnsonii CRL1693l. The MIC assays using different concentrations of the inoculum performed with the macro dilution agar technique indicated differences of one dilution, mainly in Kanamycin and Oxytetracycline.
When comparing the two methods, the resistance to Vancomycin of those strains carrying intrinsic resistance to this antibiotic showed similar patterns. No inhibition zones were observed in the disk assay and MIC values indicated that the cut-off points were higher than those suggested by EFSA [15].

Discussion
Over the last few decades probiotics have been used in the gut, skin, respiratory or urogenital mucosa of both humans and animals [29][30][31]. One of the main objectives of these feed additives  or complements is the restoration of the indigenous microbiome of different tracts, supported by the host-specificity evidenced some years ago [32][33][34]. Some authors suggest that the autochthonous microbiota of the intestine could be better adapted to gastrointestinal conditions, and also that the bacterial resistance profiles of some of the microorganisms could be influenced by the region of the tract [13].
A basic characteristic to be evaluated when working in the design of a probiotic adjunct or complement to a specific host is its resistance to the conditions where it will be applied, the so-called functional properties [11]. On the other hand, viable bacteria in high numbers should be able to exert their beneficial effect on the target organ or mucosae. The microorganisms should cross over several biological barriers through the GIT, including gastric acid, enzymes, secretions and bile salts [12,13]. The digestive system of newborn calves is different from that of adult; in fact, calves behave as monogastrics up to the development of rumen in older animals.
The capability of the bacteria to survive the passage through the GIT is variable and strain dependent [35]. Lactic Acid Bacteria have several mechanisms that confer resistance in acidic conditions [36]. According to our results, each strain showed a different resistance profile and different sensitivity to acid in the presence of bile. However, those strains that produce higher concentrations of lactic acid [6], were the most sensible to these conditions. Similar results were obtained by other authors, where resistance to gastrointestinal conditions varied with the microorganism [37,38].
Lactic Acid Bacteria can also induce stress tolerance responses [39]. Our results suggest that the growth of microorganisms previously treated in acid conditions was higher in bile. All the strains should increase the lag phase length as observed in the growth curves. This behavior could suggest an adaptation of the strain; Burns et al. [35], evaluated the pre-adaptation and cross-resistance mechanisms to gastric conditions submitting the strains to sub-lethal acid stress. The resistance patterns observed in this group of strains could support a better adaptability of certain bacteria that should be further studied. The beneficial properties evaluated in a previous work [6], indicated that there is no relation between hydrophobicity or auto aggregation and resistance profiles.
The antibiotic susceptibility of LABs should be determined to prevent the incorporation of multi-resistance strains into the Food Chain or feed additives for animals, as stated before [40]. The standardization of the method for the assessment of antibiotic susceptibility and its interpretation was determined by different research groups [16,19,21,25,28,41,42], in order to generate a data base to establish and compare susceptibility profiles in different hosts and areas. Moreover, the selection of the technique is essential to determine the resistance criteria, as suggested by Mayrhofer et al. [21], who compared the procedures of CLSI and ISO/ IDF Standard guides. The disk diffusion method was applied for LAB against antibiotics of different groups. The LSM media was compared with two frequent media use for lactobacilli, which are nutritionally demanding bacteria. The diffusion of the antibiotic can be modified by the composition or pH of the media and is of major importance to obtain reproducible results [24]. In the case of LABs, different media were used [42][43][44]. In this work, the LSM media proposed by Klare et al. [25], and studied by others authors was included [16,42], supporting the growth of most of the strains, with the exception of L. mucosae strains. Higher inhibitions zones were observed in LSM, which could indicate a better diffusion compared with LAPTg media.  The selection of antibiotics for MIC assays was performed according to EFSA 2012 and Klare et al. [16,25]. Microorganism concentration, incubation time and atmospheric conditions must be defined. Egervärn et al. [42], observed that high inoculum concentrations and longer incubation periods in the microdilution test increase MIC values. Our results showed that some resistances profiles were affected by the inoculum of the bacteria and supported the importance of standardization of the inoculum to compare results. MIC values were compared with others authors, Kanamicyn resistance were also observed in lactobacillus isolated from pigs, human GUT and food [45][46][47]. The intrinsic resistance of heterofermentative lactobacilli to Vanomicyn, which was phenotypically determined in the strains, does not present the risk of gene transfer [14,18,40].

Conclusions
According to our results, the behavior of LAB in acid conditions and bile salts cannot be predicted based on bacterial species and should be evaluated in each of the strains. Although some of the strains showed similar resistant patterns, since all the strains were be affected by the acid pretreatment, each strain was affected in a The growth rate was determined as described in materials and methods.
different way, and in some cases the strain became more resistant. On the other hand, all the strains proved to be resistant to 0.5% bile salts after acid pretreatment.
The antibiotic resistance profile of the strains indicated that the culture media affected the inhibition zone in the disk diffusion technique for most of the antibiotics assayed. Also, almost all the strains grew in LSM media, proposed by the ISO/IDF Standard guidelines. The phenotypic studies of microbial resistance of the strains according to the EFSA breakpoints indicated they are sensitive to the antibiotics evaluated, except for Kanamicyn. Some of these strains are being included in the design of a probiotic product for calves to prevent diarrhea (Table 5).    Table 4: Antibiotic susceptibility of lactic acid bacteria isolated from newborn calves calves.

Strains
Bacterial Inoculum

Antibiotics
Antibiotics proposed by EFSA with microbial breakpoints Others Antibiotics Table 5: Beneficial, functional and safety properties of LAB selected for the design of a probiotic product for calves.