Molecular and Physiological Analysis of Saccharomyces cerevisiae Strains Associated with Taberna, an Alcoholic Beverage from the Sap of Coyol Palm (Acrocomia aculeata Jacq. Lodd. Ex Mart.)

Abstract

In Chiapas, Mexico, traditional fermented beverages represent an important cultural resource and serve as reservoirs of native microorganisms, particularly Saccharomyces cerevisiae. In Taberna, a beverage produced from the sap of Acrocomia aculeata, morphological variation among S. cerevisiae colonies has been reported across different fermentation stages. This study aimed to determine whether colony morphological traits are related to intraspecific variability among isolates and to assess whether this variability is associated with the locality of origin and fermentation stage. Twenty Saccharomyces cerevisiae isolates collected from three fermentation stages (initial, intermediate, and final) in two localities (Benito Juárez and Tierra y Libertad) were characterized. Macromorphological and physiological traits—including thermotolerance, osmotolerance, cycloheximide tolerance, ethanol resistance, and carbohydrate fermentation—were evaluated. Genetic variability was assessed by Arbitrarily Primed-Polimerase in Chain Reaction (AP-PCR)using the microsatellites (CAG)₅, (GAC)₅, and MR, and by Restriction Fragment Length Polymorphism (RFLP) of the ITS-5.8S and NTS regions; data were analyzed using Unweighted Pair Group Method with Arithmetic mean (UPGMA). The isolates exhibited high physiological and molecular heterogeneity, primarily associated with locality. Isolates C15, K8, and X5 grew at temperatures up to 45 °C, and isolated Ñ5 tolerated ethanol concentrations up to 15%. Genetic profiles showed intraspecific polymorphism and geographic differentiation. These findings highlight substantial physiological variation among isolates, suggesting potential relevance for future applications in traditional fermentation processes.

1. Introduction

The sap extracted from various palm species of the Arecaceae family has traditionally been used in natural fermentation processes for the production of alcoholic beverages [1]. In palm sap, spontaneous fermentation involves both acidogenic and alcoholic phases, which confer distinctive sensory characteristics to each beverage [2]. In southern Mexico, particularly in the state of Chiapas, a traditional beverage known as Taberna is produced. This drink is white, sweet, and effervescent, and is obtained through the fermentation of the sap of the coyol palm (Acrocomia aculeata Jacq., Lodd. ex Mart.). Palm trees are selected based on their height (approximately 5 m), physiological maturity (5–15 years), and the specific position of the incision made on the stem. After selecting the palm, the trunk is carefully cleaned to remove spines and leaves. A longitudinal cut is then made, and a rectangular cavity known as a canoa (approximately 5–7 cm²) is carved to reach the inner tissues of the stem.

The production process of Taberna has been orally transmitted across generations. Producers from ejidos (a form of communal land tenure in which a community collectively owns and uses land for productive activities, without individual private ownership) such as Tierra y Libertad traditionally harvest palm trees during the full moon, guided by the belief that the stem is less susceptible to decomposition during the fermentation process. Additionally, they emphasize that production should occur during warm seasons, as higher temperatures prevent the sap from becoming excessively viscous and thereby reduce the risk of impaired yield. When production is adversely affected, producers incorporate approximately 5–10 mL of fermented sap obtained from other palms to decrease viscosity and reestablish the production of clarified sap with sensory characteristics acceptable to consumers [3].

The traditional production of Taberna follows a clearly defined sequence of steps (Figure 1): Preparation of the trunk: cleaning of the stem and opening of the canoe; stimulation of sap secretion: the cut is gradually deepened during the first days to stimulate sap flow; Protection of the extraction area: the canoe is covered with wood or palm leaves to prevent environmental contamination and excessive exposure to air; Periodic collection of the sap: the fresh sap is collected twice daily (morning and afternoon), ensuring prior removal of sediments or insects; Transfer to fermentation containers: the sap is poured into pre-washed 50 L plastic containers; Spontaneous fermentation: fermentation proceeds without external inoculation, allowing the combined action of lactic and acetic acid bacteria and both non-Saccharomyces and Saccharomyces yeasts. Maturation and consumption: the beverage is consumed fresh, generally 1–3 days after the onset of fermentation [3].

The collected sap is sweet, slightly turbid, and rich in various sugars, primarily sucrose, glucose, and fructose, which make it an optimal substrate for the native microbiota. During fermentation, these microorganisms convert sugars into ethanol, organic acids, and secondary metabolites that contribute to Taberna’s characteristic aroma, balanced acidity, and natural effervescence [1,2,3,4]. In fermentation systems, changes in substrate composition, together with biological and environmental factors, shape the distribution, colonization, and succession of microorganisms. These processes ultimately result in the establishment of stable microbial communities that directly influence the sensory and functional development of the product [5]. Among these microbial groups, yeasts play a central role due to their ability to synthesize compounds of industrial and commercial relevance. In this type of traditional fermentation, the beverage is typically consumed before the fermentation process is fully completed.

In recent years, the study of yeast biodiversity in traditional and tropical fermentations has gained increasing attention, particularly in spontaneous fermentation systems where multiple species and strains coexist with complex dynamics [6]. Several studies have shown that these fermentations harbor a high diversity of yeasts, whose composition varies depending on the substrate, environmental conditions, and local practices, exerting a direct influence on the final product’s sensory qualities. In such a context, S. cerevisiae frequently prevails during the advanced stages of fermentation, although it coexists with a wide diversity of non-Saccharomyces yeasts during the initial phases, contributing to the ecological complexity of the system [7]. Recent research has demonstrated that even within a single spontaneous fermentation, multiple S. cerevisiae strains may coexist, exhibiting significant genetic and functional differences that impact metabolite production and organoleptic profiles. Despite these advances, yeast diversity associated with traditional fermentations in tropical regions, particularly those derived from plant saps, remains underexplored [8].

S. cerevisiae is one of the most important yeasts in both traditional and modern fermentative processes owing to its high efficiency in ethanol production and its ability to generate a broad spectrum of secondary metabolites that modulate microbial community dynamics [9]. During fermentation, microbial communities undergo ecological succession involving dispersal, interspecific interactions, fluctuations in relative abundance, and the potential emergence of novel variants [10]. Previous studies have demonstrated that S. cerevisiae strains exhibit substantial variability in their responses to thermal and osmotic stress, as well as to conditions associated with carbohydrate fermentation [11]. The spontaneous fermentation of coyol sap, characterized by its dynamic composition and rapid fermentative activity, may facilitate the emergence of isolates with adaptive and potentially novel traits.

At the molecular level, ribosomal DNA markers such as the ITS1–5.8S–ITS2 region have been widely used for yeast identification and are considered the universal DNA barcode for fungi. However, recent studies have demonstrated that, rather than being fully conserved, these markers exhibit notable intraspecific and intragenomic variation, largely due to the presence of multiple rDNA copies that may diverge within a single genome [12,13,14,15]. This variability highlights their potential not only for taxonomic identification but also as informative tools for assessing fine-scale genetic diversity within species. In this context, the non-transcribed spacer (NTS) region, which forms part of the repeated rDNA units, has gained increasing attention due to its higher evolutionary rate and its contribution to genetic heterogeneity through mechanisms such as mutation, recombination, and copy number variation [16,17,18]. Unlike conventional approaches that rely solely on ITS regions, the incorporation of more variable regions such as NTS enables a deeper resolution of intra-specific diversity. Therefore, the novelty of this study lies in the combined use of conserved ribosomal markers and highly variable rDNA regions to improve strain-level discrimination of S. cerevisiae. This approach provides a more comprehensive understanding of genetic diversity and may have important implications for the selection and exploitation of yeast strains with desirable biotechnological and fermentation-related traits. Studies such as that conducted by Perez-Brito et al. [19] demonstrate that the use of these tools enables the determination of intraspecific variability among isolates of the same species.

Therefore, the present study seeks to comprehensively characterize the phenotypic, physiological, and molecular profiles of S. cerevisiae isolates obtained from Taberna. Specifically, the study aims to: (i) evaluate their fermentative capacities and stress tolerance responses across different fermentation stages; (ii) determine whether these characteristics are associated with their localities of origin; and (iii) explore how variability among isolates may influence the fermentation process and contribute to the distinctive attributes of the beverage. Overall, this work provides new insights into the role of native S. cerevisiae strains in Taberna production and highlights their potential applications in fermentation biotechnology.

2. Materials and Methods

2.1. Biological Material

The twenty Saccharomyces cerevisiae isolates used in this study were selected from previously isolated colonies obtained from Taberna [3] in the localities of Tierra y Libertad and Benito Juárez, located in the municipalities of Jiquipilas and Villaflores, respectively, in the state of Chiapas, Mexico. The isolation procedure was carried out previously but is not described in the cited reference; therefore, this information is reported here as unpublished data. They were preserved in YPD broth (20 g/L D-glucose, 10 g/L yeast extract, 20 g/L Casein peptone) and glycerol (1:1) by ultrafreezing at −80 °C in the Biotechnology Laboratory of the Institute of Biological Sciences, Universidad Autónoma de Ciencias y Artes de Chiapas (UNICACH) [3].

The isolates were selected according to the fermentation stages: initial (days 1 to 4), intermediate (days 5 to 10), and final (days 11 to 15), as well as the two localities where Taberna is produced: Benito Juarez, Municipality of Villaflores (16°25′36″ N, 93°19′14″ W), and Tierra y Libertad, Municipality of Jiquipilas (16°22′44″ N, 93°51′38″ W), both in the state of Chiapas, in southwestern Mexico (

Table 1. Origin of isolates and fermentation stage of S. cerevisiae from Taberna.

Isolates	Locality	Stage
G30, H22, L22, F31, C15	Benito Juarez, Villaflores, Chiapas	Initial
T35, Q25, M18, M24, Ñ35		Intermediate
W36, X15, X16, Y14		Final
E7, F13	Tierra y Libertad, Jiquipilas, Chiapas	Initial
J5, K8		Intermediate
X5, Ñ5		Final

). The selection of isolates for this study was based on the diversity of S. cerevisiae morphotypes present during spontaneous fermentation at the time of sampling carried out by Ambrocio-Ríos et al. [3].

2.2. Assessment of Macroscopic Morphology and Sexual Reproduction in S. cerevisiae Isolates

The conditions for the characterization of these aspects were followed by following the previously reported method and according to the established criteria [18,20,21]. Asci and ascospore formation were induced using actively growing cells obtained from 24–48 h YPD cultures (20 g D-glucose, 5 g, yeast extract, 10 g casein peptone, 20 g bacteriological agar, 1 L distilled water) of each isolate were inoculated onto McClary sporulation agar plates (1 g D-glucose, 1.8 g potassium chloride, 8.2 g sodium acetate trihydrate, 2.5 g yeast extract, 1.5 g agar, 1 L distilled water) and incubated at 27 °C for 7 days [20]. After incubation, slides were prepared in 10% ammonium hydroxide and examined under a light microscope (Carl Zeiss, MicroImagen GmbH 37081, Gottingen, Germany) to confirm the presence of asci and ascospores and to document their characteristics [20,21].

2.3. Physiological Characterization S. cerevisiae Isolates

Tolerance to high osmotic pressures, thermotolerance and ethanol tolerance of the twenty S. cerevisiae isolates were assessed following the methodology described in a previous study [20].

The ability of the S. cerevisiae isolates to grow in the presence of different concentrations of cycloheximide (Sigma-Aldrich, St. Louis, MO, USA) was evaluated in liquid medium. Tubes containing 4.5 mL of sterilized distilled water were supplemented with 0.5 mL of liquid YNB + D-glucose medium (6.7 g YNB, 5.0 g D-glucose, 100 mL distilled water) containing cycloheximide to achieve a final concentration of 0.1% and 0.01% (w/v). The YNB + D-glucose medium with cycloheximide was sterilized by microfiltration before use. Each tube was inoculated with 100 µL of an aqueous suspension of active yeast cells from a 24–48 h YPD culture, adjusted to turbidity pattern 2 of the McFarland standard [21] (bioMérieux, Marcy-l’Étoile, Lyon, France) and incubated at 29 °C [21]. Resistance assays to different factors were performed in duplicate.

Carbohydrate Fermentation Tests

Fermentation tests were conducted in 150 × 12 mm screw-cap tubes with a small inverted tube (Durham tube) to trap gas. To each tube 2 mL of Wickerham’s basal fermentation medium was added (4 g yeast extract, 7.5 g casein peptone, 1 L of distilled water) and sterilized by autoclaving. Subsequently, 1 mL of a sterile sugar solution, prepared by dissolving 6.0 g of each sugar (except raffinose, prepared with 12 g) in 100 mL of distilled water and sterilized by microfiltration, was aseptically added to each tube. The sugars tested included D-glucose, D-galactose, sucrose, maltose, lactose, raffinose, fructose, D-xylose, and D-arabinose, all analytical grade and purchased from Sigma-Aldrich, Toluca, Mexico. Each tube was inoculated with 100 µL of an aqueous suspension of active yeast cells from a 24–48 h YPD culture, incubated at 27 °C. Fermentation activity was recorded daily for two weeks [22].

2.4. Molecular Characterization

Total DNA extraction was performed using the Kick Quick DNA Fungal and Bacteria Mini Prep kit (Zymo Research, Irvine, CA, USA) according to the manufacturer’s instructions. DNA concentration and purity were assessed by measuring absorbance at 230, 260, and 280 nm using an Eppendorf BioSpectrometer Basic (Hamburg, Germany) [21].

Arbitrary primers PCR amplification (AP-PCR) was carried out using three 15 bp microsatellites: (GAC)₅, (CAG)₅, and MR (5′-GAC GGT GGC GGT TCT-3′), with annealing temperatures of 50 °C, 60 °C, and 42 °C, respectively [12]. PCR reactions were performed in a 25 µL total volume containing 12.5 µL of DreamTaq Green PCR Master Mix (Thermo Scientific, Waltham, MA, USA), 1 µL (10 mM) of the microsatellite primer (Integrated DNA Technologies, Coralville, IA, USA), 10 ng of DNA, and nuclease-free H₂O to complete the reaction volume [11]. DNA amplification was conducted in a T-100 Thermal Cycler (Bio-Rad, Hercules, CA, USA), with the following program: initial denaturation at 95 °C for 5 min; 40 cycles of 40 s at 95 °C (denaturation), 60 s at the primer-specific annealing temperature for each primer, and 60 s at 72 °C (extension); followed by a final extension of 5 min at 72 °C [19]. Amplified DNA fragments were separated by electrophoresis on 1.5% (w/v) agarose gels (Sigma, St. Louis, MO, USA) in 1X Tris-Borate-EDTA (TBE) buffer (Promega, Madison, WI, USA) at 100 V for 40 min and stained with ethidium bromide. A 15,000 bp molecular weight marker (1 kb plus, Invitrogen™ Carlsbad, CA, USA) was included on each gel as a reference. DNA band patterns were visualized using a GelMax Imager transilluminator (UVP, Upland, CA, USA), and images were acquired using VisionWorks software (Version 2019, UVP, Upland, CA, USA) [19,20].

The ITS 5.8S-region was amplified with the universal primers ITS1 Fw (5′-TCC GTA GGT GAA CCT GCG G-3′) and ITS4 Rev (5′-TCC TCC GCT TAT TGA TAT GC-3′) (Integrated DNA Technologies, Coralville, IA, USA). PCR reactions were performed in a 25 µL reaction volume containing genomic DNA (10 ng), 12.5 µL of DreamTaq Green PCR Master Mix (Thermo Scientific, Vilnius, Lithuania), 1 µL (10 mM) of each primer, and nuclease-free H₂O to complete the final volume. DNA amplification was performed in a T-100 Thermal Cycler using the following program: initial denaturation at 94 °C for 1 min, followed by 30 cycles of 1 min at 94 °C, 2 min at 58 °C, and 60 s at 72 °C, with a final extension of 5 min at 72 °C [19,20].

The complete non-transcribe spacer (NTS) was amplified with the pair primers JV51ET Fw (5′-TGA ACG CCT CTA AGY CAG AAT-3′) and JV52ET Rev (5′-TTA TAC TTA GAC ATG CAT GGC-3′) (Integrated DNA Technologies, Coralville, IA, USA) [13]. PCR reactions were carried out in a 25 µL volume containing genomic DNA (10 ng), 12.5 µL of DreamTaq Green PCR Master Mix, 1 µL of each primer (10 mM), and nuclease-free H₂O to complete the final volume. The PCR program was as follows: initial denaturation at 94 °C for 5 min, followed by 30 cycles of 30 s at 94 °C, 60 s at 57 °C, and 3 min at 72 °C, with a final extension step of 5 min at 72 °C [19,20].

In both cases, the PCR products were digested with the restriction enzymes AluI (Thermo Scientific, Vilnius, Lithuania), HaeIII, TaqI, and MspI (Promega, Madison, WI, USA). Each 25 µL digestion reaction contained 10 µL of the amplified PCR product, 12.3 µL of nuclease-free H₂O, 2 µL of 10X buffer (Promega, Madison, WI, USA), 0.2 µL of bovine serum albumin (Promega, Madison, WI, USA), and 0.5 µL of the corresponding restriction enzyme. Digestions were performed according to the manufacturer’s recommended conditions. Digestion products were analyzed by electrophoresis on 2% agarose gels (Sigma Aldrich, Toluca, Mexico) in 1X buffer (Promega, Madison, WI, USA). Gels were stained with ethidium bromide (0.4 µg/mL). DNA digestion band patterns were visualized using a GelMax Imager transilluminator (UVP, Upland, CA, USA), and images were captured with VisionWorks software (Version 2019, UVP, Upland, CA, USA) [19,20].

2.5. Statistical Analysis

For the statistical analysis, morphological traits, including colony characteristics such as shape, color, surface, elevation, sheen, margin, and texture, as well as the presence of asci and ascospores, were considered. Additionally, results from physiological tests, including thermotolerance, osmotolerance, and the fermentation of specific carbohydrates, were included. All data were used to generate a Euclidean distance matrix using the “dist” in Rstudio. Qualitative data were numerically coded prior to analysis. Subsequently, a hierarchical clustering was performed on the distance matrix using the “hclust” function. The resulting clustering dendrogram was displayed using the “as.dendrogram” function to allow more efficient manipulation and show the fermentation stages and locations of study [23,24].

Genetic variability among the isolates was assessed by analyzing polymorphisms identified in AP-PCR banding profiles of the (CAG)₅, (GAC₅) and MR microsatellites, as well as in restriction enzyme digestion patterns of the NTS region. The resulting DNA polymorphism data were compiled into a binary matrix, which was subsequently subjected to Principal Component Analysis (PCA) to elucidate patterns of genetic differentiation [25].

Cos2 values were used to evaluate the quality and contribution of the variables to the principal components. The analysis was performed using the “Factoextra” package in RStudio (Version 2026.01.1+403). Point colors were assigned based on the sample’s location, and concentration ellipses were added to highlight groupings within the data.

The data obtained from the physiological and molecular characterization were analyzed using UPGMA clustering based on the Jaccard similarity index [24].

3. Results

3.1. Assessment of Macroscopic Morphology and Formation of Asci and Ascospores

The twenty S. cerevisiae isolates displayed phenotypic variability in colony morphology, enabling their classification into eight distinct morphotypes (Figure 2a). Of these, three morphotypes were identified among isolates from Tierra y Libertad, whereas five were observed in isolates from the Benito Juárez locality. All S. cerevisiae isolates exhibited the ability to sporulate, forming asci and ascospores when grown on McClary medium (Table S1, Supplementary Materials).

3.2. Physiological Characterization

All Saccharomyces cerevisiae isolates exhibited resistance to elevated temperatures, growing at temperatures up to 40 °C. Notably, isolates C15 from Benito Juarez and K8 and X5 from Tierra y Libertad also grew at 45 °C (Table S1). In osmotic tolerance assays, all isolates tolerated 50% glucose, but only C15, T35, M24, W36, X15, X16, and Y14 (from Benito Juarez), and E7, J5, and X5 (from Tierra y Libertad) grew at 60% glucose. In a medium containing 10% NaCl and 5% glucose, weak growth was observed in four isolates. None of the isolates tolerated cycloheximide at either 0.1% and 0.01% concentrations (Table S1, Supplementary Materials).

All S. cerevisiae isolates exhibited growth at an ethanol concentration of 10%. However, tolerance decreased as ethanol levels increased. At 12%, only isolates L22, W36, X15, T35, M18, M24, H22, Y14, and Ñ35 from Benito Juarez, as well as all six isolates from Tierra y Libertad, exhibited growth. At 13%, growth was observed in isolates T35 and X15 from Benito Juarez, and in all isolates from Tierra y Libertad except K8. Finally, isolate Ñ5 was the only strain capable of tolerating an ethanol concentration of 15% (Table S2, Supplementary Materials).

In the fermentation assays, all isolates demonstrated the ability to ferment D-glucose, D-galactose, sucrose, and fructose, confirming their capacity to metabolize the main sugars present in coyol palm sap. Fermentation of maltose was observed in most isolates; however, two isolates—C15 and T35 from Benito Juarez—were unable to utilize this disaccharide. Raffinose fermentation exhibited notable heterogeneity: while several isolates fermented it completely, others showed partial metabolism, and three isolates (F31, X15, and J5) displayed limited or no activity, suggesting variation in α-galactosidase expression. None of the isolates were able to ferment lactose, D-arabinose, or D-xylose, consistent with the typical metabolic profile of S. cerevisiae (Table S3, Supplementary Materials).

When the results were grouped into general fermentative profiles (Table S3, Supplementary Materials), five distinct patterns were identified in isolates from Benito Juarez and three in those from Tierra y Libertad, indicating that sugar-fermentation capacity is structured by locality rather than uniform across isolates. This variability suggests that certain isolates may contribute differentially to the progression and outcome of the fermentation process, particularly in shaping the production of ethanol, organic acids, and secondary metabolites that define the sensory attributes of Taberna.

Based on colony-level morphological characterization, distinctive traits were identified that allowed the isolates to be classified into eight different groups; of these, five corresponded to the locality of Benito Juarez and three to Tierra y Libertad (Figure 2a).

According to the morphological and physiological characterization, hierarchical clustering analysis using the UPGMA method, applying a cutoff distance of 0.15, allowed the discrimination of four well-defined clusters. Isolates C15 and T35 exhibited the greatest distance between them and were positioned in independent clades, indicating pronounced morphophysiological divergence. Additionally, a third cluster was predominantly composed of isolates from Tierra y Libertad, suggesting a locality-associated structuring pattern (Figure 2b). Overall, these findings indicate the presence of phenotypic variability potentially influenced by local environmental conditions.

3.3. Molecular Characterization

The AP-PCR molecular analysis revealed high intraspecific variability among the S. cerevisiae isolates. Each microsatellite marker generated a distinct distribution of isolates according to their genetic profiles, showing a clear association with the locality of origin and no apparent relationship with the fermentation stage. Using the (GAC)₅ microsatellite, nine genetic profiles were identified (Figure 3).

Cluster analysis using the UPGMA method, based on the molecular characterization of the isolates with microsatellite markers, revealed high variability, with a Jaccard similarity coefficient of 0.28. Four main clusters were identified; most isolates from Tierra y Libertad grouped within a single cluster, with the exception of K8 and E7. In contrast, isolates from Benito Juarez exhibited greater intraspecific variability (Figure 4).

The PCA based on the molecular marker (GAC)₅ shows genetic variability among the isolates, allowing partial differentiation of the localities but not of the fermentation stages. The PCA generated with the (CAG)₅ microsatellite reveals a clear genetic structure by locality, with isolates from Tierra y Libertad forming a more differentiated lineage, while those from Benito Juarez cluster consistently. In contrast, the PCA based on the MR marker shows genetic differentiation between localities, with no evidence of genetic structure associated with the fermentation stage (Figure 5).

According to the RFLP analysis of the ITS-5.8S region, no intraspecific variability was detected among the isolates, confirming that this region is highly conserved. However, in the RFLP analysis of the NTS region, the profiles generated with the enzymes AluI, HaeIII, and MspI were identical across all isolates, with no variation detected. In contrast, the analysis with the enzyme TaqI revealed three distinct clusters (Figure 6).

The dendrogram reveals three major patterns in the genetic structure of the isolates at a cutoff distance of 0.10. Isolate E7 (TL, initial stage) is clearly segregated, forming an independent lineage and exhibiting the highest genetic distance within the dataset, which suggests pronounced divergence. The remaining isolates are distributed into two well-supported clusters that correspond to their locality of origin, indicating a defined population structure and genetic differentiation associated with geographic provenance (Figure 7).

Considering the set of molecular markers employed (the different microsatellites and the NTS region analyzed using the TaqI enzyme), observed intraspecific variability among the isolates, with 18 distinct genetic profiles identified (

Table 2. Band profiles obtained by AP-PCR using the microsatellites (GAG)₅, (GAC)₅ and MR, and RFLP with the restriction enzymes Taql of the NTS regions, of all S. cerevisiae isolates obtained at different fermentation stages of Taberna made in the localities of Benito Juarez and Tierra y Libertad, Chiapas.

S. cerevisiae Isolates	Microsatellites			NTS	G. P.
	(CAG)5	(GAC)5	MR	TaqI
BENITO JUÁREZ
Initial stage (1–4 d)
G30	A	A	A	A	A
H22	A	A	A	A	A
L22	B	A	A	A	B
F31	C	B	A	A	C
C15	D	C	A	A	D
Intermediate stage (5–10 d)
T35	B	D	B	A	E
Q25	E	E	C	A	F
M18	F	F	D	A	G
M24	F	E	E	A	H
Ñ35	G	E	D	A	I
Final stage (11–15 d)
W36	H	E	D	A	J
X15	I	G	F	A	K
X16	B	E	F	A	L
Y14	G	H	F	B	M
TIERRA Y LIBERTAD
Initial stage (1–4 d)
E7	B	H	D	C	N
F13	A	H	G	C	O
Intermediate stage (5–10 d)
J5	J	I	H	C	P
K8	J	I	H	C	Q
Final stage (11–15 d)
X5	A	H	H	C	R
Ñ5	A	H	H	C	R

Identification pattern types are represented by an identical letter code in a column. G. P.: General Profile. The general profile was determined by combining the results of the different typing approaches, which were assigning a different letter to each pattern type.

).

The UPGMA analysis, based on the integration of physiological and molecular characteristics, revealed the formation of four main clusters at a genetic distance of 0.25. Isolates from Tierra y Libertad were predominantly grouped within a single cluster (distance of 0.20), suggesting high phenotypic and genotypic similarity and a strong within-locality association. However, isolate E7 exhibited relatively greater divergence within this group. In contrast, isolates C15, L22, and T35 were positioned in independent clades, reflecting greater heterogeneity and intraspecific diversity among the Benito Juarez isolates. Overall, the clustering pattern indicates a partial structuring associated with locality of origin, evidencing a clear differentiation between Tierra y Libertad and Benito Juárez, as well as higher genetic variability within the latter (Figure 8).

4. Discussion

Our investigation revealed substantial physiological and genetic variability among the S. cerevisiae isolates obtained at different stages of the Taberna fermentation process in two localities of Chiapas, Mexico. Each isolate exhibited distinct responses to various stress factors, reflecting a mosaic of local adaptations. A total of three macromorphological profiles were recorded in Tierra y Libertad and five in Benito Juarez. The observed morphological diversity is consistent with previous studies, indicating that colony characteristics largely depend on the environmental origin of the yeasts. At a biogeographical scale, S. cerevisiae has been shown to develop typical strains with recurrent genetic, physiological, and phenotypic attributes within specific environments, where microbial interactions, competition, and environmental conditions act as key evolutionary forces [26,27,28].

Reproductive modes also contribute to this variability. All isolates were able to form asci and ascospores, confirming their potential for sexual reproduction. Sporulation represents a specialized form of division that generates haploid gametes through meiosis when cells are deprived of nitrogen and fermentable carbon sources. During this process, the plasma membrane and cell wall give rise to the ascus that encloses the resulting spores. Although S. cerevisiae typically grows as a diploid in rich media and reproduces asexually by budding [29,30,31], sporulation constitutes a strategy for survival, dispersal, and genetic diversification that promotes adaptation to changing environments. Recombination and reassortment during meiosis contribute to the elimination of deleterious alleles, increase genetic variability, and enhance population fitness [16,28].

In the physiological characterization, all isolates tolerated ethanol concentrations of 10% and temperatures up to 40 °C, with a reduction in growth at 12% ethanol and 42 °C. However, isolates C15 (Benito Juarez) and K8 and X5 (Tierra y Libertad) tolerated temperatures up to 45 °C; the latter two also grew at 14% ethanol. This behavior aligns with previous reports for S. cerevisiae and other mesophilic yeasts, whose ability to withstand temperatures up to 42 °C depends on the expression of heat shock proteins (HSPs). Sustained exposure to higher temperatures inactivates RNA polymerase II and limits growth [32]. Similarly, previous studies have documented native strains capable of tolerating 15% ethanol and 40 °C [33], suggesting that the Taberna isolates possess comparable traits. Ethanol and heat resistance are associated with changes in gene expression [33], including the induction of HSPs involved in both thermal and ethanol stress responses. Moreover, extracellular vesicles enriched in HSPs—particularly the Ssa2 subunit of the HSP70 family—appear to play a relevant role in thermotolerance [34].

Regarding osmotic stress, all isolates grew in 50% glucose, whereas growth at 60% varied. Under conditions of 10% NaCl and 5% glucose, only three isolates from Tierra y Libertad and one from Benito Juárez were able to grow. These results reflect the physiological challenge posed by hyperosmolarity in traditional fermentations, where yeasts must counteract loss of turgor pressure and activate signaling pathways to restore osmotic balance [33,34,35,36,37]. The HOG pathway, along with the cell wall integrity (CWI) and MAPK pathways, regulates glycerol accumulation, water influx, and recovery of cell volume—mechanisms essential for survival in hyperosmotic environments [38].

In the cycloheximide resistance assays, none of the isolates showed growth, indicating generalized sensitivity to this antibiotic. Cycloheximide inhibits eukaryotic protein synthesis by blocking translational elongation at the ribosomal level, thereby impairing cellular growth and metabolic activity [36]. This uniform sensitivity suggests a conserved translational machinery among the Taberna isolates and the apparent absence of adaptive resistance mechanisms, such as mutations in ribosomal proteins, alterations in membrane sterol composition, or the overexpression of multidrug transporters regulated by PDR1/PDR3 [39]. From a physiological perspective, sensitivity to cycloheximide reflects the integrity of essential cellular processes and can be indicative of limited exposure to selective pressures that drive antifungal resistance. Recent studies have also demonstrated that translation inhibition by cycloheximide can interfere with key stress-response pathways, including protein homeostasis and adaptive responses to environmental stress, reinforcing its value as a probe for cellular functionality [40].

From an applied standpoint, this phenotype is particularly relevant in fermentation and microbial ecology. Cycloheximide-containing media are widely used as selective tools to suppress Saccharomyces cerevisiae and allow the detection of resistant non-Saccharomyces yeasts or contaminants in mixed fermentations [41]. Therefore, the consistent sensitivity observed in these isolates may facilitate their differentiation from resistant yeasts and support their potential use in controlled fermentation systems, where predictable behavior and reduced risk of resistant subpopulation emergence are desirable traits.

The differences observed in carbohydrate fermentation suggest a possible relationship between the metabolic variability of the strains and their tolerance to ethanol stress. Sugar metabolism begins with transport and phosphorylation, which constitute the first steps of glycolysis. These processes primarily depend on hexose transporters (Hxt) and key enzymes such as glucokinase and hexokinase [42]. Variations in the activity or expression of these components may influence the efficiency of sugar metabolism, which in turn could affect the ability of cells to adapt to stress conditions generated by ethanol accumulation during fermentation. The presence of atypical profiles could be associated with the practice of backslopping, common in spontaneous fermentations, which favors the emergence of mutations and the coexistence of diverse strains. Isolates C15 and T35 were unable to ferment maltose, possibly due to the absence or inactivity of genes in the MAL locus, which encodes maltose permease and maltase [43,44]. Raffinose was also not fermented by several isolates, particularly those from the initial stage. This behavior may be explained by the absence or low activity of α-galactosidase encoded by MEL1, required to release galactose and enable fermentation of the remaining monosaccharides [45].

At the molecular level, the use of microsatellites (GAC)₅, (CAG)₅, and MR revealed high polymorphism among the isolates. According to several authors, such microsatellites denote substantial genetic diversity within a population, representing diverse subpopulations that may reflect distinct evolutionary histories and microenvironmental adaptations. Microsatellites capture fine-scale differences that are not detected with conserved markers such as the ITS-5.8S region, which in our study was confirmed to be highly conserved [7,46,47,48,49]. Principal component analysis revealed clustering associated with locality of origin, but not with fermentation stage, supporting the notion that microenvironmental factors exert strong selective pressure [50]. In the NTS region, analysis with TaqI showed that genetic profiles were associated with locality. Although NTS regions have an indirect relationship with physiological and phenotypic characteristics, they contain replication origins, silencing promoters, Sir-binding sites, and sequences involved in recombination. Variations in the NTS region can alter silencing and recombination recruitment, leading to changes in rDNA copy number, ribosomal gene expression, and protein synthesis rates [46,47,48,51]. Recent studies have demonstrated that NTS regions accumulate numerous variants that may persist at different frequencies within the same isolate, reflecting a complex dynamic of ribosomal evolution [52,53].

The integration of molecular characterization with the observed phenotypic traits provides a more comprehensive understanding of the mechanisms underlying strain variability. Recent studies have shown that phenotypic diversity in Saccharomyces cerevisiae is strongly influenced by multiple layers of genetic variation, including single-nucleotide polymorphisms, structural variants, and copy number variations, all of which can directly affect physiological processes such as stress tolerance and metabolite production [54]. In this context, the correspondence observed between molecular profiles and physiological characteristics in this study suggests that the detected genetic differences may be associated with functional adaptations to specific fermentation conditions. Furthermore, integrative approaches combining genomic, phenotypic, and metabolic data have demonstrated that linking these levels of information enables the identification of genotype–phenotype relationships and improves the prediction of strain performance under relevant environmental stresses [55]. Recent research also highlights that intraspecific genetic diversity in S. cerevisiae is reflected in significant differences in stress tolerance and metabolic outputs, reinforcing the importance of connecting molecular data with physiological traits for a more robust mechanistic interpretation [15].

Altogether, the physiological and molecular characterization demonstrates that the S. cerevisiae isolates associated with Taberna production exhibit considerable diversity, both in their adaptive capabilities and in their genetic profiles. This variability may directly influence the fermentation process, modulating fermentation rates, ethanol yield, metabolite production, and, consequently, the organoleptic profile of the beverage. Previous studies on palm sap fermentations, tequila, and agave-based beverages have shown that differences among Saccharomyces strains affect microbial succession, process stability, and the sensory profiles of the final product [1,5,18,22,23,56,57,58]. Our results suggest that certain native strains may act as key contributors to the stability and quality of the fermentation, helping to shape the distinctive character of traditional beverages from Chiapas.

5. Limitations

This study presents some limitations that should be considered when interpreting the results. First, the number of isolates obtained from certain localities, such as Tierra y Libertad, was limited and reflects the diversity and abundance of S. cerevisiae morphotypes present at the time of sampling. Therefore, the observed diversity may not fully represent the total microbial variability of the region.

Second, the experimental approach was primarily qualitative, focusing on the characterization and comparison of physiological and molecular profiles rather than on quantitative measurements of strain performance. As a result, the findings provide comparative insights but do not allow for precise quantification of differences among strains. Finally, the results are specific to the spontaneous fermentation of coyol sap under the environmental and cultural conditions of the studied region. Therefore, extrapolation of these findings to other fermentation systems or geographical contexts should be made with caution. Despite these limitations, this study offers valuable insights into the diversity and potential biotechnological significance of native S. cerevisiae strains associated with traditional fermentations.

6. Conclusions

The physiological and molecular analyses conducted in this study demonstrate that native S. cerevisiae isolates associated with Taberna fermentation exhibit substantial intraspecific diversity. Differences in tolerance to temperature, ethanol, and osmotic stress, together with variability in carbohydrate fermentation, highlight the metabolic plasticity of these isolates. At the molecular level, the high polymorphism detected through microsatellites and the variation observed in the NTS region underscore the genetic complexity of these populations.

This diversity suggests that certain native strains may significantly influence the stability, efficiency, and sensory characteristics of Taberna fermentation. Understanding these patterns provides a foundation for preserving traditional practices, developing autochthonous starter cultures, and exploring the biotechnological potential of these yeasts. Future studies linking genetic variation with functional and sensory traits will help to further elucidate the role of these strains in traditional fermented beverages.