Potential Probiotic Yeasts Sourced from Natural Environmental and Spontaneous Processed Foods

In the last decades, there has been a growing interest from consumers in their food choices. Organic, natural, less processed, functional, and pre-probiotic products were preferred. Although, Saccharomyces cerevisiae var. boulardii is the most well-characterized probiotic yeast available on the market, improvement in probiotic function using other yeast species is an attractive future direction. In the present study, un-anthropized natural environments and spontaneous processed foods were exploited for wild yeast isolation with the goal of amplifying the knowledge of probiotic aptitudes of different yeast species. For this purpose, 179 yeast species were isolated, identified as belonging to twelve different genera, and characterized for the most important probiotic features. Findings showed interesting probiotic characteristics for some yeast strains belonging to Lachancea thermotolerans, Metschnikowia ziziphicola, Saccharomyces cerevisiae, and Torulaspora delbrueckii species, although these probiotic aptitudes were strictly strain-dependent. These yeast strains could be proposed for different probiotic applications, such as a valid alternative to, or in combination with, the probiotic yeast S. cerevisiae var. boulardii.


Introduction
Historically, nature represented the primary source of most pro-technological microorganisms, used today for industrial applications in pharmaceuticals, foods, beverages, and the agrochemical industry.
Although the search for new strains with unexplored properties continues, today's investigations in this field are going through a revolution [1]. The vast number and variety of bioactive molecules isolated from microbial natural products has greatly contributed to the improvement of human well-being and health during the past century. Studies of fungal communities from specific environments have indicated non-anthropized natural environments or food processing matrices as a natural source of microbial isolation [2].
Spontaneous fermentation, and other traditional methods of processing and preserving food and beverages using unselected microorganisms, is frequently practiced around the world [3]. Until the 1960s, this practice was necessary due to the lack of knowledge and the scarce availability of well-characterized commercial starter strains to be applied in industrial processes. Today, after some forty years, all industrial fermentation in the food field is managed using starters, which have reached their maximum application. Today, a new trend can be observed: a return to the past, with increasing production of "artisanal" foods, and the promotion of high quality ingredients (some local or sustainable) intended as natural products. This spontaneous processed food represents a or 100 mL (wine, beer and sugarcane juice) of each sample was collected using a sterile bag. All the samples were maintained at 4 • C until arrival in the laboratory for processing. NE samples were then aseptically maintained overnight on a rotary shaker at 150 rpm at 4 • C, to facilitate microbial release, while sourdough and cheese samples were subjected to stomacher homogenization (25 • C for 5 min) adding 100 mL of sterile physiological solution. Ten-fold dilutions were made for all samples and spread onto Wallerstein Laboratories (WL) agar (Oxoid, Hampshire, UK) with 0.02% biphenyl to prevent mold diffusion, Rose Bengal medium (Oxoid, Hampshire, UK) with chloramphenicol to inhibit bacteria development, and Lysine Agar medium (Oxoid, Hampshire, UK)) for non-Saccharomyces yeasts selection. The plates were incubated at 25 • C for 5 days. Representative yeast colonies were selected on the basis of their micro-and macro-morphological differences, from the highest diluted plates of each matrix and the numbers of isolations made in relationship to the relative abundance of each plate. Pure isolates were maintained at 4 • C on YPD medium (yeast extract 1%, peptone 2%, dextrose 2%, and agar 2%) for subsequent analyses, and in YPD broth supplemented with 80% (w/v) glycerol for long-term storage at −80 • C.

DNA Extraction
About 180 pure yeast cultures were selected and used for the yeast's DNA extraction, according to the method reported by Stringini et al. [29]. First, the isolates were pre-cultured on YPD agar for 3 days at 25 • C. Then, the cells were transferred to screwcap tubes containing glass beads and reaction buffer (Trizma 0.1 M, pH 8.0, EDTA (Ethane diyldinitrilo tetraacetic acid) 50 mM, SDS (Sodium dodecyl sulfate 1%). The tubes were vortexed, boiled for 10 min, and placed on ice to allow cell wall disruption. Next, 20 µL of Tris-HCl 1 M (pH 8.0), 15 µL of EDTA 0.5 M (pH 8.0), 50 µL of SDS 10%, and 200 µL of potassium acetate 5 M were added, and the tubes were incubated on ice for 30 min. After centrifugation, the supernatant containing the DNA was transferred to a new tube containing ice-cold isopropanol, incubated on ice for 5 min, centrifuged, and the pellet resuspended in ice-cold ethanol 70%. After centrifugation, the DNA was resuspended in a Tris-EDTA buffer and left at 45 • C for 15 min. The DNA obtained was stored at −20 • C until processing.

Yeast Species Identification
The ITS1-5.8S rRNA-ITS2 region was amplified by PCR (Polymerase Chain Reaction) using primer pair ITS1 (5 -TCCGTAGGTGAACCTCGCG-3 ) and ITS4 (5 -TCCTCCGCTTTATTGATATGC-3 ), as described by White and co-workers [30]. The amplification was performed in a reaction mix containing 0.5 µM of each primer, 10 µM of each dNTP, 1.5 mM of MgCl, 1 vol. of PCR buffer 10×, 1U of DreamTaq DNA polymerase (Fermentans, Thermo Fisher Scientific Inc., Waltham, MA, USA) and 5 µL of extracted DNA solutions in a final volume of 100 µL. The PCR was performed in a Biorad Thermal Cycler (Bio Rad, Hercules, CA, USA), using an initial denaturation at 95 • C for 5 min, followed by 35 cycles of 94 • C for 1 min, annealing at 55 • C for 2 min, elongation at 72 • C for 2 min, and a final extension at 72 • C for 10 min. PCR products were separated on 1.5% agarose gel stained with 0.5 mg/mL of ethidium bromide in 0.5× TBE (Tris-borate-EDTA) and visualized under UV. A Generuler 100 bp DNA ladder (Fermentans, Thermo Fisher Scientific Inc., MA, USA) was used to compare the size of the bands obtained. The sequencing approach was used to obtain sequences of the representative yeasts, which were compared with those already present in the data library using the BLAST program [31] and the GenBank database (http://www.ncbi.nlm.nih.gov/BLAST) [32] for yeast species identification.

Probiotic Characterization
2.4.1. Ability to Grow at 37 • C The isolates were first tested for their ability to grow at internal body temperature. The strains were pre-cultured on YPD broth for 24 h at 25 • C. They were then transferred to fresh media with an inocula of 10 6 cells/mL. Changes in optical density were monitored after 48 h of incubation at 37 • C.
The commercial probiotic S. cerevisiae var. boulardii (CODEX, Zambon Italia S.r.l., Bresso, Italy) was used as the positive control strain. The trial was conducted in duplicate.

Effect of Low pH and Bile on Yeast's Growth and Viability
All the isolates were evaluated for their ability to grow in the presence of bile and low pH. The strains were pre-cultured on YPD broth for 24 h at 25 • C. They were then used to set up the trials following the protocol described by van der Aa Kühle et al. [33], with some modifications. Yeast Nitrogen Base (YNB, Biolife, Milan, Italy) was acidified with HCl 2 N to reach pH 2.5, and added to 0.3% (w/v) bile salts (Merck KGaA, Darmstadt, Germany). It was used as the growth medium. An inoculum of about 10 6 cells/mL of each pre-culture was made in duplicate, and the changes in cell density after 48 and 120 h of incubation at 37 • C were monitored by counting the cells using a Thoma-Zeiss counting chamber, as suggested by Casagrande-Pierantoni et al. [34]. Viability of the cells was estimated using methylene blue staining. YNB without bile salts, not acidified, inoculated with yeasts, and incubated at 37 • C was used as the control. The probiotic S. cerevisiae var. boulardii (CODEX, Zambon Italia S.r.l., Bresso, Italy) was used as the positive control strain.

Antimicrobial Activity
The antimicrobial activity of the about 180 strains was assessed by the double-layer agar technique, as described by Perricone and co-workers [35]. Six microbial species potentially pathogenic for humans were used as sensitive strains: Candida albicans, Escherichia coli, Listeria monocytogenes, Staphylococcus aureus, and Salmonella enterica. Plate Count Broth (tryptone 5.0 g/L; yeast extract 2.5 g/L; glucose 1.0 g/L) was used to allow the bacteria's growth twice at 30 • C for 24 h, while YPD broth was used for C. albicans under the same conditions. The potential probiotics were pre-cultured on YPD broth for 24 h at 25 • C, and 100 µL of the pre-culture (7 log CFU/mL) were distributed onto the surface of YPD agar plates and incubated at 30 • C for 24 h. A second soft layer of nutrient agar (beef extract 3 g/L; peptone 5 g/L; agar 15 g/L) was distributed onto the surface of YPD agar, and the potential pathogen strains were streaked on the surface of the soft layer and incubated at 37 • C for 24 h. The probiotic S. cerevisiae var. boulardii was used as the positive control strain. Plates without potential probiotics were prepared as negative controls. The antimicrobial activity of the yeasts tested was evaluated as the presence of an area of inhibition of pathogen growth.

Antioxidant Activity
The about 180 isolates were tested for their ability to scavenge the DPPH (1,1-Diphenyl-2-Picrylhydrazyl) radical following the method described by Chen et al. [36]. All the strains were pre-cultured onto YPD broth for 24 h at 25 • C and the probiotic S. cerevisiae var. boulardii was used as positive control strain. In short, 800 µL of fresh cell solution and 1 mL of DPPH solution (0.2 mM in methanol) were mixed and left at 25 • C for 30 min. The samples were centrifuged at 2000 g-force for 2 min and the scavenged DPPH was monitored by measuring the decrease in absorbance (A) at 517 nm. The trial was conducted in duplicate and the blank sample was prepared using de-ionized water. The scavenging ability of each strain was defined by solving the following equation: [1−A 517(sample) /A 517(blank) ] × 100%.

Isolation and Identification of the Isolates
The primary isolation campaign was carried out in natural environments (wood, soil, fruit, plants, sandstone pits, cellars, and dairy) and spontaneously processed foods (sourdoughs, cheeses, wine, beer, and sugarcane juice), and 179 yeast strains were isolated and identified through the sequencing of ITS1-5.8S and rRNA-ITS2 regions. As expected, NE samples showed greater variability of genera and, specifically, a high number of different Metschnikowia species, such as pulcherrima, ziziphicola, fructicola, and reukaufi. S. cerevisiae were isolated in a single case (oak moss). This supported the evidence that S. cerevisiae is vanishingly rare on fruit, even in vineyards where fruiting plants are at very high artificial densities and in which the associated winemaking would be expected to increase the overall abundance of yeast in the location [37]. Alternatively, an abundance of S. cerevisiae strains was found in SPF samples where the natural matrices were fermented. In these cases, S. cerevisiae obviously dominated in wine samples together with S. bayanus, and in sourdough samples together with Kazachstania unispora. Cheese matrices represented the most abundant source of Debaryomyces hansenii [38].

Probiotic Aptitudes
After identifying the 179 yeast species, the evaluation of potential probiotic characteristics was performed. For this purpose, a series of tests, growth at 37 • C (Table 2), the ability to survive at low pH, high bile concentrations of the antioxidant property (Table 3), and antagonistic behavior against human pathogens (Table 4), were carried out. Table 2. Evaluation of the ability of each isolate to grow at 37 • C. The CODEX, Zambon Italia S.r.l., Bresso, Italy, strain was used as the positive control. The "+"expresses the order of magnitude increase in optical density, the "±" indicates faint growth, and the "−" represents no growth.

Yeast Strains
Growth at 37 • C Yeast Strains Growth at 37 • C

Brettanomyces genus
Kluyveromyces genus        Table 4. Antimicrobial activity of the yeast strains isolated from NE and SPF. Five human pathogenic bacteria were chosen for the test. The "+"indicates the ability of the yeast to inhibit bacterial growth in a double-layer agar test, the "±" indicates faint inhibition, and the "−" indicates no inhibition action by yeast.

Yeast Strains Human Pathogenic Bacteria
C. albicans E. coli L. monocytogenes S. aureus S. enterica Brettanomyces genus      The 179 yeast strains isolated and identified were evaluated for their ability to grow at 37 • C. Table 2 shows that 130 out of the 179 strains were able to survive in a condition similar to that of natural internal body temperature. In particular, a greater capacity to grow at 37 • C was observed for the strains belonging to the Brettanomyces, Candida, Debaryomyces, and Saccharomyces genus. Alternatively, only one of the seven strains identified as Kazachstania was able to grow at this temperature, and variable results were observed within the Torulaspora genus.

Antioxidant Activity, Low pH, and Bile Effects
The antioxidant capacity of the isolated strains was measured by the DPPH method in the final medium extracts. Table 3 reports the antioxidant activities of the tested samples evaluated in comparison to CODEX strain, and used as the positive control. Results showed lower values of D. hansenii, B. bruxellensis, H. uvarum, and K. unispora in comparison with the positive control. Many strains belonging to the T. delbrueckii species (isolated from NE and SPF matrices) revealed higher antioxidant activity, together with L. thermotolerans, L. waltii, Candida spp. and M. ziziphicola isolated from bark or bark moss, and P. fermentans and K. marzianus coming from unpasteurized malt. This result strongly supports the importance of wood as un-anthropized natural habitat to isolate bioactive new strains [39].
The 179 strains were also evaluated for their ability to survive in chemical conditions similar to the conditions found in the gastrointestinal tract (Table 3). All strains belonging to the Brattanomyces, Candida, Debaryomyces, Kazachstania, Saccharomyces, and Torulaspora genera were able to survive at pH 2.5 with 0.3% bile salts for 48 hours. However, in comparison with CODEX, used as the positive control, the screened strains showed great variability in the percent viability. Indeed, Candida MMF1_1128 and MMI1_1129, Debaryomyces BAT2_1170, BB4_1204, LAIF1_1167, MM1_1193 and MMS1_1196, and Saccharomyces 4PV exhibited higher persistence than CODEX (42.7%). After 120 h of incubation at the same conditions, a general trend in viability reduction was observed for all the strains tested. The only S. cerevisiae 4PV maintained the 90.9% viability after 120 h of exposure to these stressful conditions.

Antimicrobial Activity of the Isolates
The antimicrobial activity of the 179 isolates was evaluated via a double-layer agar test and is reported in Table 4. The results indicated that all the strains belonging to Kazachstania, Kluyveromyces, Lachancea, Saccharomyces, and Torulaspora exhibited antimicrobial activity against C. albicans, E. coli, S. aureus, and S. enterica bacteria, comparable with the results shown by the positive control, CODEX. On the contrary, most strains of Candida and Debaryomyces seemed to be unable to counteract the growth of pathogens, with the exception of Candida 7, 28, B9, B10, and B9 that exhibited the same antimicrobial activity of the control. Similarly, few strains of Debaryomyces showed results comparable to the control. As expected, L. monocytogenes showed the greatest resistance to the antimicrobial activity of yeasts [40].

Discussion
Nowadays, there is great interest in the design of functional foods that contain probiotic microbial strains responsible for health benefits in the host. Indeed, several studies suggest the important role of probiotic microorganisms as promoters of human health because they are involved in the modulation of immune response and in the prevention of diseases such as inflammatory bowel, gastrointestinal, and atopic disorders and allergies [9,41,42]. Moreover, a microorganism defined as potentially probiotic and one that exerts beneficial effects on human health should possess the ability to tolerate acid pH and bile salts (to survive in the gastrointestinal tract), adhere to and/or persist in the mucosal and epithelial surfaces for immune-modulation, and to exercise competitiveness and antimicrobial activity against human pathogens [9,11,43]. Most of the probiotic microorganisms belong to the genera of lactic acid producing bacteria, but some yeast strains that exist in dairy and fermented products are classified as probiotics [25]. In addition, probiotic features are strain specific and accurate screening is required for selection of truly probiotic yeasts. Different from bacteria, yeasts are microorganisms much easier to handle both in the laboratory and on the industrial scale, and their management is less expensive. Moreover, traditional fermented dairy and not-dairy foods, such as fermented vegetables, craft beer, various natural cheeses and yogurts, are useful original resources for finding novel probiotics or processed foods to which probiotics could be added to give them safety and beneficial properties [44][45][46].
Generally, natural environments represent specific and peculiar ecological niches for a great number of yeasts that survive by responding to stress conditions, such as different pH values, high osmotic pressure, and salinity, to resist the action of antibiotics and to produce active compounds [47,48].
A general protocol for yeast selection was proposed by Pulvirenti and co-workers [49], and this could be applied to the selection of yeasts with functional and probiotic aptitudes, even if studies that describe the health-promoting properties of yeast remain limited [50,51]. To reinforce the possible use of yeasts as probiotics, in the present study, the main probiotic features of wild yeast strains isolated from natural environments and spontaneous processed foods were evaluated. Regarding to the ability of the yeasts to survive at close to human body temperature (37 • C), most of the yeast strains isolated possessed this ability independent of the isolation matrix. Kumura and co-workers [22], testing probiotic applications of yeasts isolated from cheese, confirmed that species belonging to Kluyveromyces, Yarrowia, Debaryomyces, Saccharomyces, and Candida were able to grow at 37 • C. Regarding antioxidant activity, the strains belonging to the Kazachstania, Pichia, Saccharomyces, and Torulaspora genus showed results comparable to or greater than the control, while variable activity was observed within the strains of the other genera tested. Previous work by [36] described the excellent antioxidant activity of Pichia strains. The ability of the yeast to survive in chemical conditions similar to the gastrointestinal tract were widely diffused in Brattanomyces, Candida, Debaryomyces, Kazachstania, Saccharomyces, and Torulaspora strains, showing results comparable both to each other and the control strain. In this regard Zivkovic et al. [52] described T. delbrueckii as the most resistant strain in the gastric juice simulated conditions, highlighting a poor survival rate of S. cerevisiae. It was interesting also to note a wide diffusion of antibacterial activity of the Kazachstania, Kluyveromyces, Lachancea, Saccharomyces, and Torulaspora strains. Although no adhesion tests on intestinal cell lines have been performed, this seemed not to be a prerequisite for the potential probiotic yeasts to exhibit inhibitory action against pathogenic bacteria [33].

Conclusions
In conclusion, our in vitro study demonstrated that wild yeasts from natural environment and spontaneous processed foods could represent a valid source of potential probiotic yeasts. In particular, the results highlighted the probiotic aptitudes of 13 yeasts isolated in moss on oak (L. thermotolerans), beech tree bark (M. ziziphicola), wine (S. cerevisiae), sugar cane juice (T. delbrueckii), papaya leaves (T. delbrueckii), wineries (T. delbrueckii), and grapes (T. delbrueckii).
Based on their features, these yeasts could be proposed, for probiotic applications, as a valid alternative to the widely available probiotic yeast S. cerevisiae var. boulardii. Further investigation is needed to clearly define the yeasts, their safety, their health-promoting efficacy, and the dosage, following the WHO criteria and EFSA recommendations.
Author Contributions: A.A., L.C., E.Z., M.C. and F.C. participated in the design and discussion of the research. A.A., L.C. and E.M. carried out the entire experimental part of the work. All authors contributed to the draft of the manuscript and read and approved the final manuscript.
Funding: This research was financially supported by MICROVERDIBIO_2017_0542_ Cariverona.

Conflicts of Interest:
The authors have no conflicts of interest to disclose.