Fungal Microbiota of Malbec Grapes and Fermenting Must Under Different Sanitary Conditions in the Southern Oasis of Mendoza Winemaking Region

Abstract

This study characterised the diversity of filamentous fungi and yeasts during Malbec grape fermentation in the Southern Oasis of Mendoza (Argentina) winegrowing region, under different sanitary conditions and SO₂ treatments, using morphological and ITS-RFLP-based molecular methods. Alternaria, Cladosporium and Penicillium were present in both sound and damaged grapes, while Aspergillus and Botrytis were primarily found in damaged grapes. The predominant yeast species in both sound and damaged grape must, at lower and higher maturity levels, were Aureobasidium pullulans and Hanseniaspora spp. At higher grape ripening levels species diversity increased, with Hanseniaspora vineae, Metschnikowia pulcherrima and Candida membranifaciens dominating, and others such as Pichia kudriavzevii and Issatchenkia terricola appearing. A. pullulans and M. pulcherrima were highly tolerant to SO₂. Notably, the species Meyerozyma guilliermondii, Zygoascus hellenicus and Hanseniaspora uvarum were exclusively present in damaged grape must, while Zygosaccharomyces bailii was also found in sound grape must. Hanseniaspora spp. and P. kudriavzevii predominated at mid-fermentation and persisted at the end of the process, highlighting their resistance to wine conditions and their potential to influence post-fermentative dynamics. These findings emphasise the significant influence of grape sanitary status on mycobiota composition, with important implications for fermentation behaviour and final wine quality.

1. Introduction

Argentina ranks fifth among global wine-producing countries, with a total production of 10.9 MhL in 2024, positioning it as the leading producer in the Southern Hemisphere [1]. Mendoza stands out as the principal winegrowing province of the country, including the Southern Oasis region, where the San Rafael Denomination of Origin (DO) (Res. N° C31/2007, INV, Argentina) is located. This region presents a wide range of altitudes, temperatures, and soil types, all of which contribute to the production of high-quality wines [2]. Despite its long-standing viticultural tradition, limited information is available regarding the native microbiota involved in fermentation in this area, which remains a subject of concern. Malbec is the emblematic red grape variety of Argentina and the most extensively cultivated in the country. In 2024, it accounted for 23.5% of the national vineyard area, thus establishing itself as the leading variety in terms of both in cultivation and wine production [3]. Its adaptability to local environmental conditions enables the production of wines that are widely appreciated in both domestic and international markets.

The mycobiota—yeasts and filamentous fungi—plays a key role in the biochemical transformation of the must, influencing both fermentation kinetics and the sensory profile of the final product [4,5]. The composition and dynamics of these fungal communities depend on a number of factors such as grape variety [6,7], vineyard management practices [8,9,10,11], edaphoclimatic conditions [12,13,14], vine and grape health [5] and the ripening stage of the grape [15,16]. These environmental and anthropogenic variables shape microbial community structures at different scales, thereby affecting wine composition and quality [17]. Particularly, the sanitary status of grape berries at harvest may affect the composition of the grape surface microbiota, especially with regard to the presence of spoilage microorganisms [18,19,20,21]. These microorganisms have the capacity to exert a substantial influence on the winemaking process, as their metabolic activity during fermentation may lead to the formation of off-flavours, turbidity, and other undesirable alterations that compromise wine quality [11,13].

Recent studies have demonstrated that the grape-associated mycobiota not only influences the development of aromatic compounds and the production of secondary metabolites during fermentation, but may also serve as a marker of microbial terroir, contributing to regional wine characteristics [4,5,22,23]. The concept of microbial terroir refers to the distinctive microbial community of each vineyard environment, shaped by factors such as climate, soil, cultivar, and viticultural practices, impacting the authenticity and typicity of the wines produced in different winegrowing regions worldwide [23,24,25,26,27,28]. In Argentina, however, studies characterising the native mycobiota remain scarce [29,30,31,32,33,34]. In our viticultural region of southern Mendoza, bioprospecting studies have been carried out to obtain microbial tools for specific purposes [35,36,37,38,39,40]. Although some recent studies have addressed related aspects [41] a comprehensive characterisation of the mycobiota associated with grapes of varying health statuses and their corresponding must and wine ecosystems is still lacking. The aforementioned study focused specifically on potentially spoilage-related or mycotoxigenic fungi, and thus included only samples from rotten grapes.

Despite the differences in grape mycobiota between winegrowing regions, several yeast species have been recurrently identified in grapes and fermenting musts worldwide. The most representative species on the surface of healthy grape and in must before fermentation are: Aureobasidium pullulans, Lachancea thermotolerans, Hanseniaspora uvarum, Hanseniaspora guilliermondii, Hanseniaspora opuntiae, Metschnikowia fructicola, Metschnikowia pulcherrima and Candida stellata [4,16,18,25,35,40,42,43]. As alcoholic fermentation progresses, the highly fermentative agent Saccharomyces cerevisiae becomes dominant [43], while certain ethanol-tolerant yeast species, including Candida zemplinina (now Starmerella bacillaris) and Zygosaccharomyces bailii, persist until the end of the process [11,16]. Brettanomyces is often present and active at different stages of the fermentation, introduced either from grape surfaces or from the winery environment [44]. Meanwhile, Hanseniaspora osmophila [45], Hanseniaspora vineae [21], Starmerella bacillaris, H. uvarum, Issatchenkia terricola, Issatchenkia orientalis, Saccharomycodes vini and Zygoascus hellenicus have been associated with musts from rotten and botrytised grapes [18]. Additionally, Pichia kluyveri was significantly more abundant on berries exhibiting sour rot symptoms [46]. Regarding filamentous fungi, the genera Alternaria and Cladosporium have frequently been reported in vineyard environments, while Aspergillus and Penicillium have been found in lower proportions [16,21,25,39,47,48].

In this context, the aim of this study was to characterise the mycobiota present before and during the fermentation of Malbec grapes of contrasting health status in the Southern Oasis of Mendoza winegrowing region. The work examined differences in the fungal communities (filamentous fungi and yeasts) in musts under distinct sanitary conditions, including the presence of the common oenological antimicrobial agent, SO₂. To this end, the berries were carefully sampled to obtain the must, followed by microbial isolation and enumeration on both general-purpose and selective/differential media, with subsequent identification performed using morphological and molecular methods. The influence of these communities on the physicochemical characteristics of the resulting wines was also evaluated.

2. Materials and Methods

2.1. Studied Area and Sampling

Samples of Malbec grapes (Vitis vinifera L.) were harvested during the 2016 vintage in the Southern Oasis of Mendoza winegrowing region (34.5°–36° S latitude and 70°–66.5° W longitude), which includes the San Rafael Denomination of Origin (DO). Samples were collected from four vineyards, each located in a different district of the region: Rama Caída (lat. 34.66° S, long. 68.37° W), Las Paredes (lat. 34.59° S, long. 68.41° W), Villa Atuel (lat. 34.82° S, long. 67.95° W) and Cuadro Nacional (lat. 34.56° S, long. 68.24° W). These vineyards correspond to different zones (sub-regions) of this region delimited by the Corporación Vitivinícola Argentina (COVIAR) according to the edaphoclimatic characterisation [2], designated as 1-2, 1-1, 3-1, and 3-3, respectively (Figure 1). These zones reflect specific combinations of climate and soil characteristics.

Approximately three grape bunches of sound grapes, with no visible signs of damage, were randomly sampled from every five vineyard rows, collecting a total of 3 kg of grapes. Damaged grape samples were collected considering those bunches affected by several types of rot, insects, hail and/or heavy rains (

Table 1. Characteristics of grape used for fermentation trials from different sub-regions of the Southern Oasis of Mendoza winegrowing region.

Sub-Region	Raw Materials	Total Soluble Solids (°Brix)	pH
Rama Caída	Sound grape	19.9	3.74
	Damaged grape	20.0	3.65
Las Paredes	Sound grape	19.6	3.55
	Damaged grape	17.5	3.66
Villa Atuel	Sound grape	24.4	3.78
	Damaged grape	24.9	3.82
Cuadro Nacional	Sound grape	22.8	4.23
	Damaged grape	25.5	4.21

).

All samples were placed in plastic bags under aseptic conditions and immediately transported to the laboratory, where they were stored at 4 °C until further processing.

2.2. Fermentation Trials

In the same plastic bags utilised for sample collection, manual crushing and destemming were performed separately for each individual sample collected under aseptic conditions. The resulting grape must was then used for fermentation trials.

Fermentations were carried out under conditions simulating traditional red winemaking, using 1 L Erlenmeyer flasks containing 800 g of Malbec red grape must, including skins, seeds, pulp, and juice to replicate full maceration. Four different treatments were analysed for each sampling site (

Table 2. Studied treatments with different grape health states and the presence or absence of the antimicrobial agent evaluated.

Treatment	Grape Must Composition
Sound	100% sound grape
Damaged	20% damaged grape, 80% sound grape
Sound-SO2	100% sound grape, with SO2 *
Damaged-SO2	20% damaged grape, 80% sound grape, with SO2 *

* 80 mg/L SO₂, as sodium metabisulphite.

): fermentation with sound grapes (100%) and fermentation with 80% sound grapes and 20% damaged grapes, with or without the addition of 80 mg/L of SO₂, as sodium metabisulphite (Na₂S₂O₅).

After must conditioning, the commercial yeast S. cerevisiae IOC 18-2007 (Institute Œnologique de Champagne, Epernay, France), characterised as a killer toxin-producing strain, was inoculated at the manufacturer’s recommended concentration to initiate fermentation and ensure its completion in the damaged grape experiments, as proposed by Barata et al. [49]. Alcoholic fermentation was conducted at 25 ± 1 °C. During the skin contact period, punch-downs were carried out once a day for two minutes. The progress of fermentation was monitored by weight loss, considering the process complete when weight loss was less than 0.5 g in two consecutive days. The resulting wines were racked, stabilised by cold treatment for 30 days at 4 °C, and bottled in 250 mL glass bottles.

2.3. Enumeration and Isolation of Filamentous Fungi and Yeasts

Both fungi and yeasts were isolated from the crushed grape berries (grape must) prior to the addition of the commercial yeast (before fermentation). Yeasts were also isolated at the middle and end of fermentation, as well as from the resulting wine, following the protocol proposed by Barata et al. [49] with minor modifications.

Viable fungi and yeasts counts were carried out by the serial dilutions method on the general medium WL (Wallerstein Laboratory) and on the Lysine agar (Oxoid), which is unable to support S. cerevisiae growth, and on two selective/differential media for the detection of spoilage species: DBDM (Dekkera/Brettanomyces Differential Medium) [50] and ZBDM (Zygosaccharomyces bailii Differential Medium) [51]. Plates were incubated at 28 °C, for 3–5 days for WL and Lysine media and for 15–20 days for DBDM and ZBDM media. A proportional and representative number of each colony type was recovered. Yeast isolates were purified by streak plating and subcultured onto YPD agar (10 g/L yeast extract, 20 g/L peptone, 20 g/L dextrose, 20 g/L agar, pH 4.5), while fungal isolates were subcultured on Czapek-Yeast extract-Agar (CYA) medium [52] and single conidial colonies were obtained. Cultures were incubated at 25 °C. Subsequently, they were grown in liquid YPD with 30% of glycerol and finally stored at −20 °C for later identification.

2.4. Fungal and Yeast Identification

Filamentous fungal isolates were identified based on the morphological criteria described by Pitt and Hocking [52].

The yeast and yeast-like isolates were identified using the PCR-RFLP (Polymerase Change Reaction-Restriction Fragment Length Polymorphisms) method. DNA was extracted following the protocol described by Querol et al. [53]. The quality and quantity of the extracted DNA were assessed by electrophoresis on a 1% agarose gel, using a λ DNA/HindIII marker (Promega, Madison, WI, USA) as a molecular size standard. The DNA was then used to amplify the ITS1-5.8S-ITS2 region using universal primers ITS1 (5′TCCGTAGGTGAACCTGCGG-3′) and ITS4 (5′-TCCTCCGCTTATTGATATGC-3′) as described by Esteve-Zarzoso et al. [54]. The resulting PCR products were subsequently digested with CfoI, HinfI, HaeIII and DdeI restriction enzymes following the supplier’s instructions. Amplified products and their restriction fragments were analysed on 1.4 and 2.2% agarose gels, respectively, in 1 × TAE buffer (Tris-acetic acid-EDTA, pH 8). Gels were stained with ethidium bromide. The amplified and restriction fragment sizes were estimated comparing with a DNA 100 bp ladder and analysed using YeastID database (http://www.yeast-id.org, accessed on 15 December 2024) to assign species.

2.5. General Wine Composition

The physicochemical parameters of the wines (acetic acid, alcohol, citric acid, density (15 °C/15 °C), fructose, glucose, glycerol, lactic acid, malic acid, pH, sucrose, tartaric acid, total acidity and total sugar) were determined using a wine analyser based on Fourier Transform Infrared Spectroscopy (FT-IR, Bruker Corporation, Billerica, MA, USA).

3. Results and Discussion

With the purpose of resembling the possible conditions in a winery when a proportion of the grape to be processed is affected by physical, meteorological or biological factors, the present study evaluated the mycobiota of must obtained from the Malbec wine grapes (Vitis vinifera L.) under different sanitary states (sound or 20% damaged grapes) and the must during fermentation, with or without the addition of sulphites, from four vineyards located in different edaphoclimatic zones (subregions) of the Southern Oasis of Mendoza winemaking region. The study also examined the impact of this mycobiota on the quality of the resulting wine. Understanding the fungal microbiota in the grape must obtained under this situation is important for winemakers to identify highly frequent yeast species that can alter the normal progress of the fermentation process or dangerous genera and species with wine spoilage potential.

3.1. Diversity of Filamentous Fungi Detected in Malbec Grape Must

Figure 2 shows the diversity and concentration of filamentous fungi (CFU/mL) present in Malbec grape must, obtained from sound and damaged (20% of the sample) grapes, with and without the addition of SO₂, in the different subregions of the Southern Oasis of Mendoza winegrowing region. In general, it was observed that both the total fungal counts and the diversity of fungal genera were higher in samples containing damaged grapes compared to those made exclusively from sound grapes, in all subregions (edaphoclimatic zones) studied. As expected, the addition of SO₂ significantly reduced both fungal abundance and diversity, regardless of grape condition. The higher total count of filamentous fungi was observed in damaged grape must from Cuadro Nacional (approximately 7.0 × 10⁴ CFU/mL). In contrast, the lowest counts were recorded in Rama Caída when SO₂ was added (around 5.5 × 10³ CFU/mL), highlighting the effectiveness of SO₂ in reducing fungal populations.

As shown in Figure 2, in Rama Caída, the total fungal count in must from damaged grape samples (6.7 × 10⁴ CFU/mL) was higher than in sound samples (2.8 × 10⁴ CFU/mL). The addition of SO₂ markedly reduced the fungal population in both treatments, resulting in final counts of 5.5 × 10³ CFU/mL and 6.0 × 10³ CFU/mL for damaged and sound grapes, respectively. In Las Paredes, damaged and sound grape musts presented comparable fungal concentrations (5.2 × 10⁴ and 5.0 × 10⁴ CFU/mL, respectively), which were reduced following SO₂ addition (1.6 × 10⁴ and 2.1 × 10⁴ CFU/mL, respectively). Unexpectedly, in Villa Atuel, the SO₂ treatment of sound grape must led to a higher fungal count (2.8 × 10⁴ CFU/mL) compared to the untreated sample (7.8 × 10³ CFU/mL). In contrast, the antimicrobial agent slightly decreased fungal counts in damaged grape musts, from 2.9 × 10⁴ to 2.6 × 10⁴ CFU/mL. The increase in the sound sample was mainly due to a significantly higher presence of Penicillium, suggesting that local strains may be more tolerant to SO₂ than those in other subregions. Some filamentous fungi, including Penicillium, may respond to stress conditions such as SO₂ exposure, by activating sporulation or other growth-related mechanisms that temporarily increase CFU counts. Previous studies have reported that low levels of SO₂ may not only be ineffective at inhibiting fungal growth but could, under specific conditions, promote it, as sulphur is an essential element for microbial metabolism [55,56]. Meanwhile, the slight decrease observed in the SO₂-treated damaged must likely reflects a more typical antimicrobial effect, possibly influenced by a higher initial microbial load or differences in the physiological (sanitary) condition of the grapes. In Cuadro Nacional, fungal populations were higher in musts from damaged grapes (7.1 × 10⁴ CFU/mL) than from sound ones (5.4 × 10⁴ CFU/mL), and SO₂ addition consistently reduced the total counts to 2.6 × 10⁴ and 3.5 × 10⁴ CFU/mL, respectively. In terms of diversity, Las Paredes was the subregion with the greatest fungal diversity (six genera), followed by Villa Atuel and Cuadro Nacional (four genera each), and Rama Caída (three genera).

Regarding fungal genera, Alternaria was detected in all samples and subregions with concentrations ranging from 10³ to 10⁴ CFU/mL. This finding, which shows this genus as part of the main wine grape mycobiota is consistent with previous studies carried out in the same viticultural region [39,41], as well as with research from various winemaking regions worldwide using both culturable and non-culturable methods [7,16,47,57]. Alternaria was found to be present at higher concentrations in musts obtained from damaged grapes than from sound grapes, and its resistance to SO₂ was demonstrated (Figure 2), which is consistent with previous studies [58,59,60]. The genus Alternaria is ubiquitously distributed and includes both saprophytic and opportunistic plant-pathogenic species capable of affecting crops during cultivation as well as causing decay of plant products during harvest and storage. Under conditions of elevated disease pressure, Alternaria can behave as an opportunistic pathogen, contributing to grape berry rot in the vineyard [60].

The genus Cladosporium was also detected in all the analysed samples and subregions at approximately 10³–10⁴ CFU/mL, except in Villa Atuel, where it was only present in the damaged-SO₂ sample (Figure 2). Comparable concentrations of Cladosporium were observed in musts obtained from both damaged and sound grapes, except in Rama Caída, where a notably higher concentration was found in the damaged sample. This genus appeared to be more sensitive to SO₂ than Alternaria, as SO₂ treatment generally reduced Cladosporium counts by approximately one order of magnitude (10²–10³ CFU/mL). Cladosporium has previously been reported as a dominant genus in the fungal microbiota of healthy [7,12,39,47] and damaged grapes [21](Agregar acá tbn. la CITA Lu2025). However, it has also been found at relatively low abundances in healthy, rotten and botrytised grapes [45]. Its sensitiveness to SO₂ has also been demonstrated by Kostelnikova et al. [56].

In addition, Penicillium was detected at significant concentrations (10³–10⁴ CFU/mL) in all of the subregions studied, except for Rama Caída (Figure 2). This genus has also been reported at high concentrations in vine leaves, grape surfaces and grape must [7,39,57], as well as in withered grapes [61]. Conversely, other studies have found Penicillium only in damaged grapes at low concentrations [41,45].

Botrytis was detected only in musts from damaged grapes without SO₂ addition, with fungal counts ranging from 10³ to 10⁴ CFU/mL, except in Villa Atuel, where this genus was not detected. Notably, in Cuadro Nacional, it was present in all treatments at a high concentration (10⁴ CFU/mL). Interestingly, similar counts were observed in both sound and damaged grape must samples. It is likely that the “apparently” healthy grapes from this subregion were affected by grey or noble rot, despite showing no visible signs of damage. These results are consistent with those reported by Lleixá et al. [45], who found Botrytis predominantly in damaged grapes, especially in botrytised berries, and at much lower levels in sour rotten and healthy grapes. In contrast, other authors have reported the presence of this genus even in healthy grapes [12,57]. Furthermore, SO₂ appeared to inhibit the development of Botrytis, as fungal counts were lower in the SO₂-treated samples than in respective untreated samples, or the fungus was not detected at all, as observed in the “damaged-SO₂” treatment in Rama Caída and Las Paredes (Figure 2).

As for the Aspergillus genus, it was detected only in damaged grapes from the Las Paredes and Villa Atuel subregions, with counts of around 10³ CFU/mL. In the latter subregion, it was also detected in sound grapes but at a lower concentration (around 10² CFU/mL). This genus has previously been reported at high relative abundance in other varieties of sound grapes, such as cv. Merlot, and at low relative abundance in Cabernet Sauvignon must [7] as well as rotten grape must [21,45].

Regarding the Acremonium genus, it was detected at a lower frequency than the other genera. It was present only in sound grapes from the Las Paredes (10³ CFU/mL) and in the sample containing damaged grapes from Cuadro Nacional (10² CFU/mL). In agreement with our findings, Medina et al. [62] also reported Acremonium as less prevalent than other dominant genera in sound grapes.

Overall, Alternaria, Cladosporium, and Penicillium were present in both sound and damaged grapes, while Aspergillus and Botrytis were primarily found in damaged grapes. Consistently, Aspergillus has been reported to be more abundant in musts and during the early stages of fermentation derived from rotten or botrytised grapes, whereas Botrytis was notably concentrated in botrytised samples [45]. This pattern reflects the ability of Aspergillus and Botrytis to cause diseases in grapevines and to alter the structure and integrity of grape bunches [21,63,64]. The occurrence of these genera in damaged grapes highlights the importance of avoiding the use of infected or deteriorated fruit in winemaking, since this can modify the microbial composition of the grape must and consequently affect the quality of the final wine. In agreement with our results, these fungal genera were the most frequently reported in the literature [7,13,16,47,61,62], reinforcing their ecological relevance in the fungal community associated with grape must under varying sanitary conditions.

3.2. Diversity of Yeasts Detected in Malbec Grape Must During Fermentation

Table 3. Yeast and yeast-like species identification.

Species	Restriction Fragments (bp)				Amplified Product (bp) e
	CfoI a	HaeIII b	HinfI c	DdeI d
Aureobasidium pullulans	190 + 180 + 100	450 + 150	290 + 180 + 130	ND	600
Candida membranifaciens	290 + 290 + 100	400 + 120 + 80	340 + 340	ND	600
Hanseniaspora uvarum	350 + 330 + 120	800	380 + 180 + 150	300 + 180 + 100 + 50	800
Hanseniaspora vineae	280 + 180 + 150 + 90	680 + 90	400 + 370	460 + 220 + 80	800
Issatchenkia terricola	130 + 100 + 90 + 85 + 45	300 + 120	250 + 100 + 100	ND	450
Metschnikowia pulcherrima	205 + 100 + 95	280 + 100	200 + 190	ND	400
Meyerozyma guilliermondii	320 + 270 + 50 + 10	420 + 120 + 50 + 20 + 5	300 + 300 + 10	ND	600
Pichia kudriavzevii	210 + 180 + 70 + 50 + 6	400 + 100 + 40	230 + 160 + 140	ND	500
Saccharomyces cerevisiae	380 + 330 + 150	350 + 250 + 150 + 125	380 + 370 + 120	ND	850
Zygoascus hellenicus	330 + 330	630	350 + 170 + 130	ND	650
Zygosaccharomyces bailii	320 + 280 + 100 + 100	700 + 90	350 + 230 + 160 + 60	ND	800

^a–d restriction enzymes used; ^e 5.8S-ITS-amplified product size. ND: not determined. Light grey numbers indicate the bands that could not be visualised in the agarose gels and are registered in YeastID database.

presents the identification of the yeast species isolated from grape must and during fermentation, based on the PCR-RFLP technique targeting the ITS1-5.8S-ITS2 region of the rRNA gene. A total of 11 species were identified using this molecular approach, with most classifications achieved by comparing the CfoI, HaeIII and HinfI restriction profiles with those available in the YeastID database, complemented by information from the literature [36,54]. Furthermore, it was necessary to compare the DdeI restriction profile as well, in order to distinguish H. uvarum from H. guilliermondii [65,66]. It is worth noting that some isolates did not survive long enough to be identified using this technique. This was the case for certain isolates that were presumptively identified as belonging to the Hanseniaspora genus based on their micro- and macromorphological characteristics on the WL agar medium. These yeasts are therefore presented in this work as Hanseniaspora spp. Likewise, other distinct morphotype found only in Las Paredes subregion did not survive for molecular identification. Apart from these morphotypes, the resulting species were: A. pullulans (a yeast-like fungus), Candida membranifaciens, H. uvarum, H. vineae, I. terricola, M. pulcherrima, Meyerozyma guilliermondii, Pichia kudriavzevii, S. cerevisiae, Z. hellenicus and Z. bailii.

3.2.1. Diversity of Yeasts Detected in Malbec Grape Must Before Fermentation

Figure 3 shows the diversity and concentration of yeasts and yeast-like microorganisms (CFU/mL) present in Malbec grape must, obtained from both sound and partially damaged grapes (20% of the sample), with and without the addition of SO₂, prior to fermentation in the different subregions (edaphoclimatic zones) of the Southern Oasis of Mendoza winegrowing region. In all the subregions studied, total yeast counts were found to be higher in must from grapes with 20% damage than in must from sound grapes, although the difference never exceeded one order of magnitude. A previous study reported similar total yeast counts in healthy and 30% rotten grape must samples (6.1 and 6.7 log CFU/mL, respectively) [18]. As expected, the addition of the antimicrobial agent (SO₂) reduced total yeast counts in both sound and damaged grape samples, with the exception of the Las Paredes subregion, where the total yeast count in the 20% damaged grape must added with 80 mg/L SO₂ was higher than in the respective untreated sample. When the different subregions were compared, the highest total yeast concentration was found in the Cuadro Nacional subregion, at around 1.5–2.0 × 10⁶ CFU/mL in damaged grape samples. In contrast, the lowest total yeast concentrations were recorded in the Rama Caída and Las Paredes subregions, at approximately 5.5 × 10³ and 2.0 × 10⁴ CFU/mL, respectively, in SO₂-treated sound grape musts. These results can be related to differences in grape ripeness and total soluble solids (TSS) content of the grapes across the vineyards studied, as shown in Table 1. In the Cuadro Nacional subregion, the TSS content of sound and damaged grapes was 22.8 and 25.5 °Bx, respectively, whereas in the Rama Caída and Las Paredes subregions, the TSS content of sound grapes was 19.9 and 19.6 °Bx and of damaged grapes was 20.0 and 17.5 °Bx, respectively. These findings suggest a positive relationship between grape maturity and total yeast load.

Analysing the microbial concentration by subregion, Figure 3a shows that the total yeast count in sound grapes from Rama Caída was 8.4 × 10⁴ CFU/mL, while an increase up to 1.9 × 10⁵ CFU/mL was observed in the must containing damaged grapes. Decreases of more than one order of magnitude (5.5 × 10³ and 9.0 × 10³ CFU/mL) in yeast counts were observed in treatments with the antimicrobial agent applied to musts from sound and damaged grapes, respectively, confirming the effective inhibitory action of SO₂ on certain yeast species and the overall microbial load. Regarding Las Paredes (Figure 3b), yeast counts were in the order of 10⁴ CFU/mL in musts from both sound and damaged grapes (3.6 × 10⁴ and 6.9 × 10⁴ CFU/mL, respectively). While the total count decreased in the SO₂-treated sound grape must (2.2 × 10⁴ CFU/mL), it unexpectedly increased significantly in the SO₂-treated damaged grape must treatment (1.1 × 10⁵ CFU/mL). These results suggest that local strains of H. uvarum and other Hanseniaspora spp. may be resistant to the antimicrobial treatment and possibly even activated under stress conditions such as SO₂ exposure. Although Hanseniaspora is generally considered as highly sensitive to SO₂, our findings are consistent with several studies showing that tolerance within H. uvarum is strain-dependent and influenced by environmental factors. For instance, H. uvarum strains isolated from grape must and berries have been shown to persist during fermentation despite the addition of 30 mg/L of SO₂ [67,68], and even 40 mg/L of free SO₂ [69], indicating the presence of sulphite-tolerant genotypes. Furthermore, Mancic et al. [70] reported that strains recovered from wild, non-vinous fruits could tolerate 50 mg/L of total SO₂, suggesting that sulphite resistance may also be associated with the ecological origin of the strains or their preadaptation to oxidative environments. The total yeast counts in musts from both sound and damaged grapes in the Villa Atuel subregion were similar (1.8 × 10⁵ and 2.7 × 10⁵ CFU/mL, respectively) (Figure 3c), whereas in Cuadro Nacional, the total yeast counts in musts from sound and 20% damaged grapes were 3.2 × 10⁵ and 2.2 × 10⁶ CFU/mL, respectively (Figure 3d). Interestingly, the incorporation of 20% damaged grapes into the Malbec must in Cuadro Nacional led to an increase in total yeast counts by nearly one order of magnitude, substantially higher than the differences observed in the other subregions. These findings are consistent with the highest TSS content and pH (25.5 °Bx and pH 4.21; Table 1), which likely favoured the microbial development and proliferation at harvest time. This fact could be explained by the health status of the damaged grapes from Cuadro Nacional, which showed visible signs of grey rot and were therefore selected as “damaged” in this study. Grape damage results in high nutrient availability [44], which in turn supports higher microbial cell counts and greater species diversity compared to sound grapes [18]. All the species identified in our study increased in number, particularly S. cerevisiae. The relatively high presence of S. cerevisiae in sound grape samples, besides its expected abundance in damaged grape must, could be explained by contamination from injured berries hidden in apparently sound bunches, as previously observed by Barata et al. [18]. As in the Rama Caída subregion, the addition of SO₂ reduced the total yeast counts in both treatments compared to the respective untreated musts in Villa Atuel and Cuadro Nacional subregions (Figure 3c,d). Overall, with the exception of certain Hanseniaspora species, A. pullulans, M. pulcherrima and S. cerevisiae were among the most SO₂-resistant species.

Regarding species richness, Villa Atuel had the highest number of yeast species or morphotypes (ten), followed by Cuadro Nacional (six), Las Paredes (five) and Rama Caída (four) (Figure 3). In Las Paredes and Cuadro Nacional, the species number increased in the trial with 20% damaged grapes (five and six species, respectively) compared to sound grapes (two and four species, respectively). In Las Paredes, Hanseniaspora spp., H. uvarum and Z. hellenicus were only detected in damaged must, whereas in Cuadro Nacional, H. vineae and Z. bailii, were recovered in damaged must, although Z. bailii was also present in sound must with SO₂. By contrast, the number of species in samples containing damaged grapes was the same as in samples containing sound grapes, both in Rama Caída (three species) and in Villa Atuel (nine species). Nevertheless, M. guilliermondii was only recovered in damaged grape must in Rama Caída, while H. uvarum was only found in damaged grape must in Villa Atuel. Regardless, higher yeast counts were observed in must containing damaged grapes than sound grapes in all subregions studied for most yeast species detected under both sanitary conditions. These findings are consistent with previous studies. Barata et al. [18] isolated twelve ascomycetous yeast species in must containing 30% rotten grapes, from which four species (Candida apicola, Issatchenkia occidentalis, Z. hellenicus and Z. bailii) were only found in must containing damaged grapes. In addition, the species that were detected in both must samples were isolated in higher cell concentrations in 30% damaged grape samples. These findings align with those of Hall et al. [46], who reported that although overall fungal diversity remained relatively stable in healthy and sour rot-affected grapes, the abundance of certain species increased in symptomatic berries, associated with physical damage to the berry and the presence of Drosophila spp., which may facilitate fungal proliferation. These observations underscore that microbial shifts associated with grape deterioration are primarily driven by changes in relative abundance and opportunistic growth under favourable conditions, rather than the introduction of new species.

As demonstrated in the preceding results, it can be determined that, regardless of the subregion studied, the diversity of yeast mycobiota in the grape must depends significantly on the sanitary condition and degree of ripeness of the grapes, as can be seen in

Table 4. Yeast and yeast-like species isolated from grape must of different health conditions, degrees of maturity, and in the presence or absence of SO₂ in the Southern Oasis of Mendoza (Argentina) winegrowing region.

Species	Sound Grape	Damaged Grape (20%)	SO2 Tolerance (80 mg/L)	Degree of Grape Maturity
				19–20 °Bx a	23–25 °Bx b
A. pullulans	+	+	+	+	+
C. membranifaciens	+	+	+	−	+
Hanseniaspora spp.	+	+	+	+	+
H. uvarum	−	+	+	+	+
H. vineae	+	+	−	−	+
I. terricola	+	−	−	−	+
M. pulcherrima	+	+	+	−	+
M. guilliermondii	−	+	−	+	−
P. kudriavzevii	+	+	+	−	+
S. cerevisiae	+	+	+	−	+
Z. hellenicus	−	+	+	+	−
Z. bailii	+	+	+	+	+

+ or – indicates the presence or absence of the species under condition evaluated in the Southern Oasis of Mendoza winegrowing region. ^a Total soluble solids content range in the Rama Caída and Las Paredes subregions; ^b Total soluble solids content range in the Villa Atuel and Cuadro Nacional subregions.

. It summarises the yeast morphotypes and species present in grape must with different health conditions and maturity levels, along with their tolerance to the antimicrobial agent SO₂, in the Southern Oasis of Mendoza (Argentina) winegrowing region.

As regards particular species, A. pullulans was present in the grape must before fermentation in higher numbers than the other species, with counts ranging from 5.5 × 10³ to 1.2 × 10⁵ CFU/mL, in all analysed subregions and treatments, except for the must obtained from 20% damaged grapes and treated with SO₂ in Rama Caída (Figure 3, Table 4). Although this species could not be specifically detected in this sample, its tolerance to SO₂ was demonstrated by its recovery in all other samples that underwent the same treatment. Furthermore, it was shown to be the most resistant species identified. This demonstrates its predominance on the grape surface and in fresh must, especially in less mature samples (Rama Caída and Las Paredes, see Table 1), as well as its resistance to antimicrobials commonly used in oenology. Previous studies have reported that various A. pullulans strains can resist the addition of 29 mg/L of SO₂ to Chardonnay grape must [71], as well as 120 mg/L of SO₂ to Malbec grape must [72]. A. pullulans has been reported as the most abundant species on the grape surface, in fresh grape juice and during the earliest stages of fermentation in various winegrowing regions around the world. It has been frequently isolated from healthy grapes or their must in the same winemaking region as the present study (Southern Oasis of Mendoza) in Argentina [38,40,73], as well as from fresh and withered red grapes in Italy [10,11,47,48], in musts of Pinot Noir, Shiraz and Chardonnay from healthy grapes in Australia [25,71], and in sound grapes from the Cape South Coast wine region in South Africa [57]. Furthermore, A. pullulans has also been detected in botrytised grapes [45] and rotten grapes [18,41]. Our results are consistent with those of Barata et al. [18], who identified A. pullulans as the predominant species isolated from grapes at different developmental stages, including immature, mature, damaged, and undamaged berries. Although A. pullulans is typically present in similar numbers in both sound and damaged grapes, its relative abundance tends to decrease significantly as ripening advances, due to the proliferation of oxidative and fermentative ascomycetous species, as previously observed by Barata et al. [44].

The group of yeasts referred to as Hanseniaspora spp., which could not be recovered to be identified at species level, was also predominant in grape must before fermentation. This morphotype was found in all subregions and in almost all treatments, primarily in must obtained from damaged grapes, with counts ranging from 6.0 × 10³ to 2.2 × 10⁵ CFU/mL (Figure 3). Particularly, H. uvarum was detected in grape must from vineyards located in Las Paredes (Figure 3b) and Villa Atuel (Figure 3c) but only in treatments involving damaged grapes (with and without SO₂). Meanwhile, H. vineae was detected in grape must from damaged grapes in the Villa Atuel (Figure 3c) and Cuadro Nacional (Figure 3d) subregions, with similar counts (around 10⁴ CFU/mL), but also detected in sound grapes in Villa Atuel, probably linked to its advanced state of maturity (24.4 °Bx; Table 1). Moreover, unlike H. uvarum, which can withstand the presence of SO₂, H. vineae was found to be sensitive to this antimicrobial compound, as this species was only present in very low quantities or not present at all in treatments involving it (Figure 3 and Table 4).

Previous studies have reported the presence of H. uvarum in damaged grapes, rotten [41], sour rotten and botrytised grapes [18,44,45,74]. Nevertheless, this species is widely known as predominant in healthy grapes and during the first stages of fermentation in several winegrowing regions of the world [7,10,12,25,40,43,47,75,76]. Moreover, its presence and relative abundance have been reported to increase with grape ripeness [18,77], although this was not clearly evident in our results. The Hanseniaspora genus is the main yeast group isolated from grapes and musts. While the species H. uvarum is widespread in both sound and damaged grapes and their musts, other Hanseniaspora species such as H. guilliermondii, H. occidentalis and H. osmophila are predominantly found in grapes affected by sour rot [45,49], while H. opuntiae is mainly associated with Botrytis-infected grapes, although it has also been detected in healthy grapes [74]. Additionally, H. vineae has been identified as one of the most prevalent species in sour rot-affected grapes [21]. Based on our results showing a higher proportion of Hanseniaspora spp. in damaged grapes from all subregions of the Southern Oasis of Mendoza, we hypothesise that aforementioned Hanseniaspora species may have comprised the group of unidentified isolates (Hanseniaspora spp.), given that they have been demonstrated to be prevalent in damaged grapes.

It is interesting to note that H. uvarum was found in musts obtained from grapes of different maturity levels (approximately 20 °Bx in Las Paredes and 25 °Bx in Villa Atuel), whereas H. vineae was only detected in musts from more mature grapes (23–25 °Bx) (Table 1 and Table 4). The proportion of each microorganism in the microbial consortium of wine grapes varies depending on the stage of ripening and availability of nutrient. The presence of H. vineae in sound and damaged grapes, as previously reported [21,41,77], and its occurrence in sound grapes with a high sugar content suggest that visually intact berries may have microfissures and softens, which increases nutrient availability and explains the appearance of oxidative or weakly fermentative ascomycetous populations, such as Hanseniaspora spp., as harvest time approaches [44].

Similarly, M. pulcherrima was found in the grape must prior to fermentation in Villa Atuel (Figure 3c) and Cuadro Nacional (Figure 3d) across all treatments. In both subregions, its counts were higher in treatments with damaged grapes than in those with sound grapes, although the values remained the same order of magnitude (10⁴ CFU/mL). This species has previously been found in Botrytis-infected grapes in Greek vineyards [74]. Nevertheless, it has also been identified as one of the predominant species in healthy grapes at harvest time and in must during initial days of fermentation [4,6,11,35,38,40,75,78]. In our study, M. pulcherrima was only detected in subregions where the grapes exhibited a higher level of maturity, with a total soluble solids content ranging from 23–25 °Bx (Table 1 and Table 4). This observation is consistent with the findings of Barata et al. [44], who reported that Metschnikowia species are typically associated with environments rich in nutrients. Such environments involve either damaged or sound grapes, especially at an advanced stage of ripeness. This could explain the possible presence of this weakly fermentative species in both sound and 20% damaged grape musts in the two subregions where the grape had reached higher maturity levels.

The spoilage yeast Z. bailii was found in significant counts of around 10³–10⁴ CFU/mL in Rama Caída, Villa Atuel and Cuadro Nacional (Figure 3a, Figure 3c and Figure 3d, respectively). In Rama Caída, it was present in grape must made from sound grapes with a count of around 10³ CFU/mL. In Villa Atuel and Cuadro Nacional, it was more prevalent in treatments including damaged grapes (around 10⁴ CFU/mL) than sound grapes. This species exhibited tolerance to the SO₂ treatment in both subregions. Zygosaccharomyces spp. have been identified in musts with high sugar content (225–246 g/L) from rotten and botrytised grape samples by Lleixá et al. [45], and in low abundance from sound grapes by Perpetuini et al. [11]. In both studies, the genus appeared as a non-dominant but significant component of the fungal community under specific conditions. Z. bailii, in particular, has been consistently isolated from grapes affected by noble rot, sour rot, and Botrytis in various viticultural regions worldwide [41,49,74]. The genus Zygosaccharomyces includes yeasts with spoilage potential, and Z. bailii is considered a serious threat to wine quality and preservation. According to Malfeito-Ferreira and Silva [79], it belongs to the sensu stricto group of spoilage yeasts, being associated with visible sediment formation, cloudiness in dry wines, and refermentation in sweet wines, often accompanied by the production of off-flavours such as acetic acid and acetaldehyde [80].

Remarkably, S. cerevisiae was found in the grape must before fermentation in the two subregions with the highest grape maturity levels (23–25°Bx): Villa Atuel and Cuadro Nacional. In Villa Atuel, it was only detected at very low levels (10² CFU/mL), exclusively in the musts that had not been treated with SO₂ (Figure 3c), and in the ZBDM selective/differential medium. Although S. cerevisiae is rarely isolated from visually unblemished berries, previous studies have reported its presence, albeit at very low frequencies, on grapes and at the start of alcoholic fermentation [11,18,81,82]. Recent research suggests that it may act as an endophyte closely associated with grape berry tissues [83]. Conversely, in Cuadro Nacional, S. cerevisiae was present at significantly higher levels (10⁵ and 10⁶ CFU/mL) in all four treatments (Figure 3d). The detection of this species in sound grapes, prior to the inoculation of the commercial S. cerevisiae (IOC 18-2007) strain used to initiate and complete fermentation, is likely related to the ripeness stage and the overall condition of the grapes. Nevertheless, the particularly high counts of S. cerevisiae in musts containing damaged grapes can be attributed to clear signs of rot (Botrytis-infected grapes) in the raw material used for this treatment. When grape skin is visibly damaged, the availability of high sugar concentration on the berry surface promotes the growth of ascomycetes with strong fermentative activity, such as S. cerevisiae, as previously reported [44].

M. guilliermondii and Z. hellenicus were exclusively found in the must containing 20% damaged grapes in the Rama Caída (Figure 3a) and Las Paredes (Figure 3b) subregions, respectively. The former species was sensitive to SO₂, whereas the latter species showed resistance. Both species were detected at around 10³–10⁴ CFU/mL in the DBDM selective medium. M. guilliermondii has previously been reported in grapes affected by sour rot, as noted by Barata et al. [49], and it was also detected in the selective DBDM medium. This demonstrates the importance of using selective media to detect and isolate less frequently found species. More recently, the presence of this species was confirmed using next-generation sequencing (NGS) to profile fungal communities in Pinot Noir musts from sound grapes of various vineyards in southern Australia [25]. Likewise, Z. hellenicus has only been isolated from sour rotten grapes or their resulting juice samples [18,44,49]. However, it was previously detected in sound Tempranillo grapes in Spain in an overripe state, but not at the mature or unripe stages [77]. It is interesting to note that the presence of these species, which have previously been associated with rotten grapes, were confirmed in musts containing damaged grapes, particularly in subregions with lower maturity level (Table 4). This suggests that their occurrence is more closely related to the health status of the grapes rather than their ripeness stage.

In the Villa Atuel subregion (Figure 3c), I. terricola, P. kudriavzevii (I. orientalis) and C. membranifaciens were found in musts from sound grapes at counts in the order of 10³–10⁴ CFU/mL and, in the case of the two latter species, they were also detected in musts with damaged grapes with counts in the same order of magnitude (from 10³ to 10⁴ CFU/mL). Their presence in must derived from riper grapes aligns with findings by Barata et al. [18], who suggested that these populations of grape berry yeasts, classified as copiotrophic oxidative or weakly fermentative ascomycetes, tend to appear or increase in abundance during ripening or near harvest, primarily in response to changes in nutrient availability [44]. Candida spp. and Pichia spp. are commonly reported as spoilage yeasts associated with Botrytis-affected grapes, that are involved in the formation of pellicles on wine surface and the production of off-flavour compounds such as oxidised odours and volatile phenols [80]. Various Candida species, with the exception of C. zemplinina, which is normally found in healthy grapes [18,38,49], have been found in grapes affected by sour rot. Certain strains of C. membranifaciens, however, have been reported to be associated with viticultural environments and to have oenological potential for their use as starters in mixed fermentations with S. cerevisiae [34]. Regarding P. kudriavzevii (I. orientalis) and I. terricola, these species have only been found in grapes affected by sour rot [18,21,49], although I. terricola has also been detected by NGS in healthy grapes (Vitis vinifera L.) cv. Montepulciano from vineyards managed organically or conventionally in Italy [11].

Overall, the predominant species in both sound and damaged grape musts, regardless maturity levels, were A. pullulans and Hanseniaspora spp., while in musts from riper grapes, species diversity increased, with H. vineae, M. pulcherrima and C. membranifaciens becoming dominant, and additional species such as P. kudriavzevii and I. terricola also appearing. Notably, the species M. guilliermondii, Z. hellenicus and H. uvarum were exclusively detected in damaged grape musts. The spoilage yeast Z. bailii was mainly associated with damaged grapes, although it was also detected in sound grapes, and the strongly fermentative yeast S. cerevisiae was particularly abundant in musts from highly mature grapes affected by grey rot. These findings underscore that significant skin damage, combined with high sugar concentrations on the berry surface, promotes the proliferation of ascomycetous yeasts with greater fermentative capacity such as Pichia spp., M. guilliermondii, and Z. hellenicus, as well as spoilage yeasts like Z. bailii, which may negatively impact wine quality.

3.2.2. Diversity of Non-Saccharomyces Yeasts Detected in Malbec Grape Must at the Middle and End of Fermentation

Figure 4 shows the diversity and concentration of non-Saccharomyces yeasts (CFU/mL) present in Malbec grape musts at mid-fermentation, obtained from both sound and damaged grapes, with and without SO₂ addition, from the different subregions of the Southern Oasis of Mendoza winegrowing region evaluated. The highest total yeast concentration was recorded in the Las Paredes subregion, reaching approximately 1.2 × 10⁷ CFU/mL in musts from damaged grapes (Figure 4b). In contrast, the lowest total yeast concentration was found in the Villa Atuel subregion, at around 2.0 × 10⁴ CFU/mL in SO₂-treated musts from damaged grapes (Figure 4c). In addition to the dominance of the S. cerevisiae after inoculation of the commercial fermentative strain, the non-Saccharomyces yeasts that were detected at mid-fermentation were: Hanseniaspora spp., H. uvarum, H. vineae, P. kudriavzevii and Z. bailii. As observed in the grape must prior to fermentation, Villa Atuel exhibited the greatest yeast diversity, with four species present (H. uvarum, H. vineae, P. kudriavzevii and Z. bailii), followed by Las Paredes with three species (Hanseniaspora spp., H. uvarum and P. kudriavzevii), and Rama Caída, with only one (Hanseniaspora spp.) (Figure 4a). P. kudriavzevii was detected in Las Paredes during mid-fermentation, despite not being found beforehand (Figure 4b). In the Cuadro Nacional subregion, it was not possible to process the samples at this stage of the fermentation due to experimental limitations.

In both Rama Caída (Figure 4a) and Las Paredes (Figure 4b), Hanseniaspora spp. was either the only or the predominant non-Saccharomyces morphotype capable of surviving mid-fermentation in all treatments, respectively, with counts ranging from approximately 2.0 × 10⁶ to 1.2 × 10⁷ CFU/mL. In Las Paredes, additional but less abundant species were detected. H. uvarum was found exclusively in damaged grape treatments, as before fermentation, with counts around 10⁶ CFU/mL. A third species, P. kudriavzevii, was also detected in this subregion in mid-fermentation, with low counts (10²–10³ CFU/mL) detected in all treatments, although not visible in the bar diagram in Figure 4b. The persistence of these species regardless of SO₂ treatment suggests resistance to this commonly used antimicrobial agent in oenology. In Villa Atuel (Figure 4c), H. uvarum and P. kudriavzevii were detected in all treatments during mid-fermentation, but at lower concentrations (10³–10⁵ CFU/mL) compared to Las Paredes, unlike before fermentation, where they were only detected in treatments involving damaged grapes. Likewise, H. vineae was only detected in this subregion with high counts in both sound and damaged grapes (10⁴–10⁵ CFU/mL). As observed before fermentation, this species appeared to be sensitive to SO₂. Similarly, Z. bailii was detected only in the damaged grape treatment, with a concentration of approximately 8.0 × 10³ CFU/mL.

Species belonging to Hanseniaspora genus, mainly H. uvarum, are characterised by low ethanol tolerance (4–6% v/v) and by their markedly decrease at the mid-stages of fermentation [6,82]. However, some species have proved to tolerate higher ethanol concentrations (around 7.5–8.5% v/v). In fact, H. osmophila was reported to replace to H. uvarum as the predominant yeast during the mid-fermentation stage for both sour rot and Botrytis-affected damaged grapes [45]. H. vineae was able to coexist with S. cerevisiae until relatively advanced stages of fermentation of Isabella (Vitis labrusca L.) grape must [76]. The presence of P. kudriavzevii in mid-fermentation can be attributed to its well-known ethanol tolerance and occasional dominance in fermentation environments [75,78]. Our results are in concordance with those of Barata et al. [49], who reported that wine spoilage species Z. bailii was only recovered during fermentations with sour rotten grape, reaching 5.0 log CFU/mL and 2.5 log CFU/mL at the end of fermentation.

At the final stage of fermentation, Hanseniaspora spp. were the only yeasts detected in the sound-SO₂ treatment (5.0 × 10³ CFU/mL) from the Las Paredes subregion. The survival of the yeasts, despite the presence of both SO₂ and ethanol, highlight their tolerance to these stress conditions. However, their decline towards the end of fermentation was demonstrated, as previously reported [45,82]. Additionally, P. kudriavzevii was recovered in bottled wine from the same subregion, with counts of 10⁴ and 10³ CFU/mL in the damaged and damaged-SO₂ treatments, respectively. This observation underscores its ability to persist under wine conditions. If not controlled, this could potentially contribute to post-fermentation dynamics and wine sensory deviations.

It is worth noting that, despite using a specific selective medium to detect and isolate the Dekkera/Brettanomyces bruxellensis species, this yeast was not found in must samples, fermenting must or wine in the studied region, suggesting that the grape surface is not a source of this yeast. Therefore, if this yeast and its associated flavour defect are present in the winery, contamination and deterioration would be caused by specific environments within the winery itself, rather than by the raw material, which could be controlled by maintaining proper hygiene in the winery and implementing other relevant practices.

3.3. Physicochemical Composition of Malbec Wines

Table 5. Physicochemical parameters at bottling of Malbec wines obtained from the different treatments.

Parameters	Rama Caída				Las Paredes				Villa Atuel				Cuadro Nacional
	S	D	S-SO2	D-SO2	S	D	S-SO2	D-SO2	S	D	S-SO2	D-SO2	S	D	S-SO2	D-SO2
Acetic Acid (g/L)	0.60	0.54	0.61	0.61	0.57	0.62	0.62	0.63	0.76	0.58	0.62	0.71	0.75	0.72	0.70	0.80
Alcohol %	10.1	10.3	10.0	10.6	10.3	11.0	10.0	10.0	13.9	13.0	13.4	12.5	12.6	12.4	12.8	12.2
Citric Acid (g/L)	0.35	0.20	0.43	0.24	0.09	0.23	0.23	0.04	0.58	0.00	0.00	0.01	0.70	0.53	0.11	0.52
Fructose (g/L)	0.0	0.0	0.1	0.1	0.0	0.0	0.0	0.0	0.1	0.0	0.3	0.1	0.0	0.5	0.1	0.0
Glucose (g/L)	1.3	2.4	1.5	1.6	1.7	1.4	1.6	2.1	1.4	1.6	1.5	0.9	0.4	1.5	0.4	1.2
Glycerol (g/L)	9.3	9.1	9.2	9.5	8.6	9.5	8.3	8.5	12.4	11.8	12.1	12.6	12	11.7	12.6	13.4
Lactic Acid (g/L)	0.55	0.43	0.81	0.91	0.50	0.50	0.57	0.49	0.63	1.12	1.41	1.08	0.31	0.26	0.90	0.26
Malic Acid (g/L)	2.7	3.3	2.3	2.4	1.9	1.9	2.1	2.3	1.1	2.5	1.6	2.8	1.6	1.5	1.8	1.7
pH	3.52	3.70	3.65	3.73	3.39	3.43	3.40	3.37	3.42	4.12	3.97	4.20	3.49	3.43	3.64	3.50
Sucrose (g/L)	0.3	0.3	0.1	0.1	0.1	0.4	0.6	0.2	0.4	0.7	0.0	0.0	0.4	0.2	0.0	0.2
Tartaric Acid (g/L)	1.43	1.19	1.29	1.27	2.00	2.03	1.77	1.92	2.37	0.92	0.99	0.74	2.30	2.54	1.36	2.36
Total acidity (g/L)	5.9	5.8	5.8	5.3	5.7	5.8	6.0	5.9	6.2	5.0	4.8	5.0	6.2	6.0	5.7	6.2
Total sugar (g/L)	1.6	3.3	2.1	2.4	1.7	1.7	2.1	2.4	1.3	3.6	2.2	2.4	0.0	1.6	0.3	0.8

S: Sound grapes, D: sound grapes added with 20% damaged grapes, S-SO₂: sound grapes supplemented with 80 mg/L of SO₂, D-SO₂: sound grapes added with 20% damaged grapes and supplemented with 80 mg/L of SO₂.

shows the physicochemical analysis of the Malbec wines obtained from the four treatments (sound grape must and 20% damaged grape must, with or without SO₂) applied to the grape must from the four subregions (zones) studied in the Southern Oasis of Mendoza.

Ethanol content varied by subregion and correlated with the soluble solids content of the original grape must. Wines from Rama Caída and Las Paredes exhibited lower ethanol levels (10.0–11.0%), while those from Villa Atuel and Cuadro Nacional showed higher values (12.2–13.9%). A similar trend was observed for glycerol, with lower concentrations in the Rama Caída y Las Paredes subregions (8.3–9.5 g/L) and higher levels in Villa Atuel and Cuadro Nacional (11.7–13.4 g/L).

Acetic acid concentrations also followed this geographic pattern. Wines from Rama Caída and Las Paredes ranged from 0.54 to 0.60 g/L, while those from Villa Atuel and Cuadro Nacional reached higher levels (0.58–0.76 g/L and 0.70–0.80 g/L, respectively). Interestingly, within each subregion, wines made from sound and 20% damaged grapes showed similar acetic acid contents, and in some cases, wines from damaged grapes exhibited even lower values. These findings align with those of Lleixá et al. [45], who reported comparable acetic acid levels in wines from healthy and botrytised grapes, and lower values in wines from rotten grapes. However, in our study, the highest acetic acid concentrations (approaching or exceeding 0.70 g/L) were associated with treatments involving musts with greater yeast load and species diversity. At this threshold, acetic acid can impart undesirable vinegar-like aromas and sour flavours, potentially masking fruit and floral notes and degrading overall wine quality [84].

Regarding total acidity, all wines showed values between 5.3 and 6.2 g/L, with corresponding tartaric acid levels of 1.5–2.5 g/L and pH between 3.4 and 3.6. An exception was observed in wines from the Villa Atuel subregion, specifically in treatments involving damaged, sound–SO₂, and damaged–SO₂ grapes, where total acidity was lower (4.8–5.0 g/L), consistent with reduced tartaric acid content (<0.1 g/L) and higher pH values (~4.0).

Total residual sugar levels remained below 3.6 g/L in all samples. Nevertheless, across all subregions, wines made from 20% damaged grapes generally showed higher residual sugar contents (typically >2.2 g/L), except for the damaged treatment in Las Paredes. This pattern may be linked to the greater yeast diversity and concentration in damaged grapes, which could interfere with sugar metabolism. The presence of non-Saccharomyces yeasts, including potential spoilage organisms, is known to affect both the fermentation performance and metabolic activity of S. cerevisiae, potentially slowing sugar consumption [85].

As a collateral result, this study partially contributes as a first approach to the study of microbial terroir of the Southern Oasis of Mendoza, since we observed particular species in the viticultural subregions studied that may serve as biological markers. In particular, I. terricola, P. kudriavzevii (I. orientalis), and C. membranifaciens were exclusively associated with sound grapes of the Villa Atuel subregion, suggesting they could be distinctive of this area and contribute in a unique manner to the wines produced in this viticultural region. P. kudriavzevii was also found in mid-fermentation and in bottled wine of Las Paredes, demonstrating its ability to tolerate winemaking conditions; thus, this species could also be considered a biological marker of this subregion. Therefore, these results suggest that grape and wine yeast communities may exhibit a geographical character at local scale, a significant aspect of the microbial terroir concept, as observed by Chalvantzi et al. [23], with implications for regional wine features and contributing to the typicity and distinctiveness of wines from different winegrowing regions. Taken together, this work contributes to the understanding of microbial terroir by identifying yeast species and communities that may serve as microbial terroir markers of specific subregions, helping to define the microbial signature that underpins the regional identity and sensory uniqueness of Malbec wines from the Southern Oasis of Mendoza.

4. Conclusions

Understanding the fungal communities associated with grapes in contrasting health conditions, as well as their evolution during fermentation, is crucial for winemakers aiming to manage microbial dynamics and ensure wine quality. This study provided a comprehensive characterisation of the mycobiota present before and during the fermentation of Malbec grapes with distinct sanitary statuses in the Southern Oasis of the Mendoza winegrowing region, simulating realistic winery conditions in which grapes affected by physical, meteorological, or biological factors are commonly processed. The role of SO₂, a widely used oenological antimicrobial agent, was also considered.

Our findings indicate that the composition of the mycobiota was primarily influenced by the sanitary status and maturity of the grapes, rather than their geographical origin within the Southern Oasis of Mendoza winegrowing region. However, to confirm this observation, further studies involving a broader range of subregions and a correlation with the specific edaphoclimatic conditions of each zone are necessary. Notable differences in fungal communities, including both filamentous fungi and yeasts, were observed across sanitary conditions. Malbec musts containing 20% damaged grapes exhibited higher fungal loads and greater species diversity, particularly of SO₂-resistant genera such as Botrytis and Aspergillus, as well as potentially spoilage-associated yeast species including Z. bailii, P. kudriavzevii (I. orientalis), H. uvarum, M. guilliermondii, and Z. hellenicus, compared to musts made exclusively from healthy grapes. This suggests that such yeasts could serve as microbial indicators of grape damage and highlights the importance of using grapes in good health to minimise the risk of introducing spoilage microorganisms into the winemaking process, an aspect with direct implications for fermentation dynamics and final wine quality. In this context, evaluating the spoilage potential of the main species associated with damaged grapes at the species and strain level becomes crucial.

This study also provides an initial insight into the concept of microbial terroir in the Southern Oasis of the Mendoza winegrowing region. The presence of distinct fungal populations across subregions supports the microbial aspect of terroir. Certain yeast species such as I. terricola, P. kudriavzevii (I. orientalis), and C. membranifaciens, found in sound grapes, were associated with a specific subregion (Villa Atuel), suggesting their potential as biological markers of microbial terroir. Identifying region-specific fungal communities may help define the microbial signature that reinforces the regional identity and sensory uniqueness of Malbec wines. However, this remains a preliminary approach, and multi-year studies are required to evaluate the temporal stability of these communities and their influence on wine quality. To further characterise the microbial terroir and identify potential biological markers associated with specific locations, we are conducting high-throughput sequencing and metagenomic analyses of the mycobiota in the Southern Oasis of Mendoza winegrowing region.