Influence of Terroir on Microbial Diversity and Wine Volatilome

Abstract

In this research, the differences between two terroirs belonging to the Protected Designation of Origin (PDO) Montilla–Moriles (Spain) were analyzed. Both areas share soil and climate characteristics, grape varieties, viticultural practices, and winemaking processes. Therefore, the objective of this study was to establish differences between both areas based on the microbiome of the must, the oenological parameters, and the majority and minority volatile compounds of the wines, thus determining the identity traits that make the wines from both areas so different. The results obtained are quite revealing, since at the microbiome level qualitative differences were established between the various areas. In the quality area, the predominant species is Torulaspora delbrueckii while in the production area it is Hanseniaspora opuntiae. Regarding the volatilome, it was observed that the aromatic profile of the wines from the production area has more citrus-fruity aromas and the quality area has honey-floral aromas, thus producing unique wines from each of the areas.

1. Introduction

Alcoholic fermentation, the process of transforming grape must into wine, has been extensively studied for many years. This process is carried out by yeasts, which convert the sugars present in the must into alcohol and carbon dioxide [1,2]. The microbiota of the must is sometimes unknown, which makes spontaneous fermentation a risky and unpredictable practice. In addition to yeast, filamentous fungi and bacteria are also found in the microbiota of a must [3,4].

Non-Saccharomyces, once considered undesirable during wine fermentation, have recently gained recognition for their positive contributions to wine quality [5]. These yeasts are found in vineyards and on the surface of the grapes, and they are the first to appear during spontaneous fermentation [6]. This type of fermentation is the most traditional method for carrying out alcoholic fermentation in wineries, producing a more complex organoleptic profile. Their concentration is influenced by grape variety, geographical-climatic conditions, and viticultural practices [7]. They produce a high concentration of glycerol, extracellular enzymes involved in the decrease in ethanol content and the organoleptic profile [8,9,10,11]. Jackowski et al. [12] used Non-Saccharomyces yeasts as an alternative to reduce the alcohol content in beverage fermentation. This measure could be used to mitigate the effect that global warming has had on over-ripening, producing wines with a high alcohol content.

The aroma of fruit wine is influenced by several factors, one of the main factors being the alcoholic fermentation carried out by microorganisms [10]. Sun et al. [13] studied the enzymatic activity of Non-Saccharomyces for the production of kiwi fruit wine, where β-glucosidase activity improved the aromatic complexity by releasing volatile compounds such as higher alcohols, ethyl esters, acetates, and other volatile aromatics, thus improving the aromatic quality of the fruit wine.

In the present study, some genera of Non-Saccharomyces were isolated: Hanseniaspora spp., which is the main genus present in ripe grapes, produces enzymes and aromatic compounds that increase the diversity of flavors; Lachancea spp. is also present on the surface of the grape and produces lactic acid, increasing acidification; Pichia spp. is known for its ability to increase the composition of aromatic compounds, providing thiols, terpenes, and fruity esters; and Torulaspora spp. is a genus commonly used in oenology and is recognized for its ability to improve wine quality. The objective of this study was to distinguish the production and quality areas of the Protected Designation of Origin (P.D.O.) Montilla–Moriles (Córdoba, Spain) and assess their oenological potential through analysis of microbiota and their volatilome.

2. Materials and Methods

2.1. Isolation of Autochthonous Wine Yeasts

Wine yeasts were isolated from natural musts from two regions within the Spanish wine appellation P.D.O. Montilla–Moriles, (Córdoba, Spain): (i) it is the “Quality Zone (QZ)”, which is the Sierra de Montilla, made up of the following wineries—Lagar Los Borbones (N37°35′39.245″, W4°37′57.912″), Lagar Saavedra (N37°33′17.64″, W4°38′8.518″) and Lagar Cañada Navarro (N37°31′55.743″, W4°33′29.137″); (ii) it is the “Production Zone (PZ)” which is made up of the following—Cooperativa de Aguilar de la Frontera (Coop. Aguilar de la Fra) (N37°30′53.136″, W4°39′25.811″), Cooperativa de San Acacio (Coop. San Acacio) (N37°38′52.44″, W4°41′52.079″), and Cooperativa de la Unión (Coop. La Unión) (N37°34′47.657″, W4°38′13.818″). A total of 292 isolates were collected during the spontaneous fermentation of the must. Specifically, 174 isolates were obtained from the QZ area, while 118 isolates were collected from the PZ area (Figure 1a).

2.2. Culture Media and Enzymatic Screening Procedures

2.2.1. Medium Wallerstein Laboratory (WL)

Oxoid brand dehydrated medium was prepared and mixed at 6% (w/v) with 2% (w/v) agar and distilled water, according to the manufacturer’s recommendations. It was then sterilized in an autoclave for 15 min at 120 °C. The plates were incubated at 28 °C for two days. Colonies were initially isolated on WL nutrient medium to allow the differentiation between Non-Saccharomyces yeasts (and bacteria) based on colony morphology and on their color [14].

2.2.2. Medium Lysine

The medium lysine was used to test if the isolated yeasts were Non-Saccharomyces. It was prepared from Oxoid brand dehydrated medium, mixing 6.6% (w/v) with distilled water as recommended by the manufacturer. It was then sterilized in the autoclave for 30 min with flowing steam and dispensed into Petri dishes. The result was positive if after making two consecutive passes growth was observed.

2.2.3. Determination of Yeast Killer Activity

Killer activity was measured by the method described by Ramírez et al. [15]. Yeast was tested on pH 4.0 methylene blue plates (MB) [16] seeded with 100 μL of a 48 h grown culture of the sensitive strain. The strains being tested for killer activity were either covered solid cultures or replica plated onto the sown MB plates. Then the plates were incubated for 7 days at 21 °C.

2.2.4. Medium for Detecting β-Glucosidase Activity

β-Glucosidase activity was evaluated as reported by Lodder and Kreger., [17], on a medium containing 0.5% (w/v) arbutin, 0.1% (w/v) yeast extract, and a 1% solution of ferric (v/v) chloride and 2% agar (w/v). This medium was sterilized in an autoclave for 15 min at 120 °C. The sowing was incubated at 28 °C for two weeks. The justification for this medium is based on the hydrolysis of arbutin by β-glucosidase, which reacts with ferric chloride causing the latter to change the color of the medium from brown to black if the result is positive.

2.2.5. Medium for Detecting Cellulase Activity

Cellulase activity was assessed by plating yeast on plates containing 1% (w/v) yeast extract, 2% (w/v) agar (w/v), 2% (w/v) dextrose, 2% (w/v) peptone, and 0.4% carboxymethylcellulose (CMC, Sigma, Kanagawa, Japan). Plates were left at 30 °C over five days [18]. Colonies were cleaned with distilled water before staining with Lugol’s solution. Cellulase activity was scored as positive when a clear halo appeared around the colony.

2.3. Yeast Identification

Yeast identification was carried out using MALDI TOF mass spectrometry (Matrix-Assisted Laser Desorption/Ionization Time-Of-Flight Mass Spectrometry, MALDI TOF-MS) at the Central Research Support Service of the University of Córdoba. This technique allows the identification of microorganisms by generation of a mass spectrum based on the protein profile, which is specific for each genus and species [19].

In the process of preparing the samples for analysis by MALDI-TOF-MS, the different yeast strains under study were obtained directly from the medium solids used for their short-term preservation, and they were dispensed into Eppendorf tubes in which 300 μL of Milli-Q water and 900 μL of absolute ethanol (96%) were added.

2.4. Determination of Oenological Parameters

pH, volatile and total acidity, and ethanol were analyzed according to the OIV official methods [20].

2.4.1. Analytical Determinations of Volatile Compounds

Determination Major Aroma Compounds

Chromatographic conditions were previously described by Peinado et al. [21]. Major volatile compounds were quantified in a gas chromatograph HP 6890 Series II (Agilent technologies, Santa Clara, CA, USA) equipped with a CP-WAX 57 CB (50 m in length, 0.25 mm in internal diameter, and 0.4 μm in coating thickness) capillary column and a flame ionization detector (FID). Analytical standards were used subjected to the same chromatographic conditions as the samples used to quantify the volatile compounds.

Determination Minor Aroma Compounds

Aroma extraction and determination was carried as described by Palenzuela et al. [22]. The extraction was realized by using stir bar sorptive extraction with 0.5 mm film thickness and 10 mm length (Gerstel GmbH, Mülheim an der Ruhr, Germany) covered by PDMS. Wine samples were previously diluted (1:10), and 0.1 mL of ethyl nonanoate (0.4464 mg/L) as internal standard was used. Extraction was carried out at 1500 °C over 100 min.

After that stirs bars were introduced into glass desorption tubes and these were desorbed in a GC/MS equipped with a Gerstel TDS 2 thermodesorption system. The retention time of the pure chemical compounds obtained from Merck (St. Louis, MO, USA), Sigma–Aldrich (St. Louis, MO, USA), and spectral libraries were used to identify and quantify the aroma compounds.

2.5. Calculation of Aroma Series

The odor activity values (OAV) of the volatile compounds were calculated as the ratio between the concentration of each compound and its odor perception threshold. Compounds with similar odor descriptors were categorized into aromatic series, with the value of each series determined by summing the OAVs of the volatile compounds within it. The results are provided in Supplementary Material (Table S1). In this study, eleven aroma series were identified: Chemical, Citrus, Creamy, Floral, Fruity, Green, Green fruit, Herbal, Honey, Toasty, and Waxy.

2.6. Statistical Analysis

To identity differences among the various wine zones and among wineries, all the analyzed parameters and aroma compounds were subjected to statistical analysis by using the Statgraphics Centurion (v. 16.1.11) statistical package. All the experiments were carried out in triplicate. All the quantified compounds were subjected to an ANOVA Simple, and F-tests were used for the establishment of homogenous groups (different letters indicate significant differences al 95% of confidence level) and principal component analysis (PCA).

3. Results and Discussion

3.1. Microbiological Analysis

A total of 292 isolates were obtained during the spontaneous fermentation of the must, as mentioned in the section, “Isolation of Autochthonous Wine Yeasts” in Materials and Methods.

Total isolates from the QZ zone were used to differentiate the percentage of Saccharomyces and Non-Saccharomyces yeasts between the various wineries. It was shown that there was a greater number of isolates that corresponded to the Non-Saccharomyces genus. However, in Figure 1c, which corresponds to the PZ zone, it is observed that the highest percentage of the isolates obtained belong to the Saccharomyces genus. The results obtained in Figure 1a show that in QZ there is a greater number of yeast isolates during spontaneous fermentation of the must compared to PZ. This may be due to the absence of the qualitative killer activity of the yeasts since, as shown in Figure 2a, most of the yeasts found do not present killer activity, leading to the absence of biological control. In PZ it was found that there is greater killer activity of the yeasts, which could modify the population patterns of the fermentation and its evolution [23], obtaining a lower number of cultivable yeast isolates. Killer activity is carried out by yeasts that produce substances known as “killer substances”. These substances are antimicrobial protein compounds that inhibit the growth and development of yeast strains susceptible to these toxins, but not their own producers [24]. These proteins have a heterogeneous structure in size and complexity. This explains the lower number of cultivable isolates in the ZP zone, since by having a higher killer activity it can be concluded that in this zone there is a greater number of yeasts sensitive to these toxins. Figure 1b illustrates the distribution of yeast isolates from the QZ zone, highlighting the proportions of Saccharomyces and Non-Saccharomyces yeasts in the various wineries. The data reveal a predominance of Non-Saccharomyces isolates in this zone. In contrast, Figure 1c, which represents the PZ zone, shows a markedly different pattern. In this zone, the majority of isolates belong to the Saccharomyces genus, indicating a significant change in the dynamics of the yeast population between the two zones.

Figure 2b shows the qualitative activity of the β-glucosidase enzyme observed in the different isolates in various musts. It is noteworthy that PZ yeasts exhibit higher activity levels compared to QZ yeasts. β-glucosidase is a significant and well-studied enzyme due to its main role in the hydrolysis of substrates with glycosidic bonds, thus releasing glucose and aglycones [25]. This enzyme is responsible for the cleavage of glycosidic bonds such as 1,4-β and 1,6-α bonds, promoting the formation of free terpenes, phenylpropenes, and aliphatic esters. In fruit-based juices and wines, numerous aroma precursors exist in the form of glycosides. In grape-derived wines, the predominant aromatic glycosides are terpenes [26]. Under the catalytic action of β-glucosidase, these glycosides release volatile aromatic compounds, improving the sensory properties of the wine [27]. The aromatic compounds that are related to this activity are geraniol, nerol, citronellol, linalool, and α-terpineol [28]. Consequently, it can be concluded that PZ yeasts have a high potential to improve the aromas associated with compounds derived from glycosides; this explains, for example, that the concentration of citronellol in the production zone is higher compared to the quality zone. Also, in the minor compounds we found geraniol and nerol, however, between the zones, Figure 2c shows the results of the cellulolytic activity of the different isolates in the various musts. Cellulolytic activity allows the degradation of polysaccharides in the cell walls, thus increasing the fruity and sweet aroma of the wines due to the release of aromatic precursors and also allows better clarification and filtration of the must [29,30]. The yeasts from the production area have a lower activity than those from the quality area. This is related to the results obtained in the aromatic series section, where the wines that present a sweeter and fruitier aroma are those where the cellulitic activity is higher. Figure 2d shows the different species of yeasts that were isolated in each of the wineries and cooperatives. During the first phase of spontaneous fermentation, different genera of Non-Saccharomyces yeasts were isolated, such as Candida spp., Hanseniaspora spp., Pichia spp., and Torulaspora spp. These results agree with the research carried out by Stefanini and Cavalieri, [31]. Within the same areas there is a diversity in terms of the number of isolated species; this is due to the existence of different factors that influence the yeast communities associated with grapes, such as soil and climate factors and vineyard management [32,33]. The Non-Saccharomyces yeast species found in this study were Hanseniaspora opuntiae, Lachancea thermotolerans, Rhodotorula mucilaginosa, Torulaspora delbrueckii, Kluyveromyces marxianus, Hanseniaspora uvarum, Pichia kudriavzevii, Candida glabrata, and Pichia terricola, distributed among the cooperatives and wineries as shown in Figure 2d.

3.2. Oenological Parameters

In

Table 1. Oenological parameters of the PZ and QZ zones. Values with different superscript letters in the same row are a significantly different test (p < 0.05).

Oenological Parameters	Lagar Los Borbones	Lagar Saavedra	Lagar Cañada Navarro	Coop. Aguilar de la Fra	Coop. San Acacio	Coop. La Unión
pH	4.07 ± 0.06 e	3.71 ±0.10 c	3.24 ± 0.06 a	3.61 ± 0.10 b	3.81 ± 0.11 d	3.71 ± 0.07 c
Volatile acidity (g L−1)	0.71 ± 0.10 b	0.49 ± 0.10 a	0.81 ±0.09 c	0.91 ± 0.10 d	2.20± 0.11 f	1.22 ± 0.10 e
Total acidity (g L−1)	4.51 ± 0.08 a	5.92 ± 0.08 c	8.62 ± 0.07 f	6.40 ± 0.11 d	7.31 ± 0.11 e	5.84 ± 0.10 b
Ethanol (% v/v)	12.6 ± 0.2 a	14.6 ± 0.2 c	12.6 ± 0.1 a	13.6 ± 0.4 b	15.00 ± 0.5 d	13.4 ± 0.2 b

the values shown represent the averages of triplicate samples (data are mean ± standard deviation (SD)). Values with different superscript letters in the same row are a significantly different test (p < 0.05). The ethanol content, volatile acidity, total acidity, and pH have values in accordance with the wines produced in the terroirs; however, significant differences are observed between the different wineries. Regarding total acidity, only Coop. La Unión and Coop. San Acacio present higher volatile acidity values than the rest of the wines. This may be due to a low level of contamination of the must with harmful microorganisms from the grapes or to the presence of Non-Saccharomyces yeasts typical of the terroir. With respect to ethanol content, only two wineries showed a higher ethanol content; this may be related to the sugar content of the grape itself.

3.3. Major Aroma Compounds

Table 2. Major aroma compounds of the PZ and QZ zones. Values shown represent averages of triplicate samples (data are mean ± SD). Values with different superscript letters in the same row are a significantly different test (p < 0.05).

Compounds (mg/L)	Lagar Los Borbones	Lagar Saavedra	Lagar Cañada Navarro	Coop Aguilar de la Fra.	Coop San Acacio	Coop La Unión
Alcohols
Methanol	69.2 ± 1.9 a	104.2 ± 7.3 c	98.6 ± 7.5 bc	75.0 ± 7.0 a	137.2 ± 6.2 d	87.9 ± 6.3 b
1-Propanol	32.2 ± 1.4 ab	35.5 ± 1.3 b	29.4 ± 1.4 a	39.9 ± 3.1 c	62.2 ± 1.4 d	33.1 ± 2.3 b
Isobutanol	144.8 ± 3.3 e	136.5 ± 5.4 d	126.6 ± 1.8 c	115.5 ± 3.8 b	44.6 ± 0.7 a	41.8 ± 1.7 a
Isoamyl alcohols	367.5 ± 8.0 c	444.0 ± 17.4 e	416.1 ± 6.6 d	345.8 ± 11.6 b	207.2 ± 4.2 a	338.4 ± 9.0 b
2-Phenylethanol	142.0 ± 12.3 c	248.0 ± 18.7 d	131.7 ± 10.5 c	68.5 ± 6.9 b	43.5 ± 3.2 a	39.4 ± 2.3 a
Carbonyl Compounds
Acetaldehyde	101.6 ± 8.4 b	33.9 ± 2.0 a	189.6 ± 16.0 d	143.4 ± 5.4 c	195.2 ± 9.4 f	105.2 ± 7.3 d
Acetoin	85.4 ± 5.9 c	18.0 ± 1.3 a	77.4 ± 4.0 bc	71.4 ± 2.1 b	145.1 ± 14.7 d	21.1 ± 1.1 a
Esters
Ethyl lactate	51.1 ± 2.0 b	10.1 ± 0.2 a	13.1 ± 0.9 a	277.7 ± 26.6 d	105.9 ± 6.1 c	34.5 ± 1.5 b
Diethyl succinate	11.8 ± 1.1 c	8.17 ± 0.60 b	4.70 ± 0.30 a	13.4 ± 1.3 d	8.60 ± 0.80 a	5.84 ± 0.44 a
Ethyl acetate	52.0 ± 1.8 d	19.1 ± 0.7 a	86.1 ± 1.7 f	69.8 ± 2.3 e	31.8 ± 0.4 c	24.9 ± 1.3 b
Polyols
2,3-Butanediol (levo)	546.8 ± 22.5 b	438.8 ± 29.2 a	683.4 ± 38.4 c	921.5 ± 0.6 e	1326.6 ± 74.7 f	779.7 ± 51.7 d
2,3-Butanediol (meso)	288.7 ± 25.2 b	228.0 ± 11.7 a	225.1 ± 12.7 a	309.3 ± 2.6 b	353.0 ± 21.4 c	254.0 ± 15.8 a
Glycerol (g/L)	11.5 ± 1.2 c	15.5 ± 1.1 d	10.8 ± 0.9 bc	9.50 ± 1.40 ab	9.20 ± 0.81 ab	8.50 ± 0.60 a

and Supplementary Material (S2) show the results obtained for the volatility analysis for each of the wineries or cooperatives belonging to each of the areas already described. As shown, in the QZ there is a significant difference compared to the PZ in some volatile compounds in the higher alcohols. Isoamyl alcohol is produced from the decarboxylation of α-keto acids that arise from the degradation of sugars or the amino acids leucine and isoleucine. Its perception threshold is 300 mg/L [34]. This product is one of the most influential volatile compounds in the organoleptic profile, providing aromas of roses and honey [35]. There is a significant difference between both areas, although only the fermented musts from the San Acacio Coop. did not exceed the perception threshold. The highest values were found in the QZ. The volatile compound 2-phenylethanol or β-phenylethanol has a perception threshold of 140 mg/L [34], which gives the wine aromatic notes of honey and roses [36]. There is a significant difference between both areas, since only the wineries in the QZ pass the perception threshold, positively influencing the aroma of the wine. However, isobutanol and propanol, despite a significant difference between both areas, do not pass the perception thresholds and do not influence the aroma.

Methanol is not a by-product of fermentation carried out by yeasts. This compound is produced naturally from the enzymatic hydrolysis carried out by endogenous pectinases of the methoxyl groups of grape pectins [37]. In the study conducted by Peinado et al. [21], it was suggested that the concentration of methanol in wines should always be low, because methanol is potentially toxic; however, none of our samples passed the perception threshold [37].

Among the carbonyl compounds, acetaldehyde is formed during the first phase of alcoholic fermentation from pyruvic acid. Yeast produces this intermediate compound in the conversion of glucose into ethanol; the enzymes responsible are pyruvate decarboxylase or/and alcohol dehydrogenase I [38]. At low concentrations, acetaldehyde has a pleasant and fruity aroma, while at high levels, its smell becomes irritating and spicy [39]. The total acetaldehyde concentration is between 10 and 200 mg/L, with a sensory perception threshold of around 100 to 125 mg/L for free acetaldehyde. The results obtained show that all samples exceed the perception threshold except for Lagar Saavedra, positively influencing the sensory profile of the wines. However, there is a significant difference between both areas, with the highest values being found in the PZ.

With respect to the ethyl esters, ethyl acetate shows a significant difference between both zones, with the highest values for QZ. However, all samples showed an ethyl acetate concentration of less than 80 mg/L, which improves the taste of the wine and enhances the fruity smell [40]. Although there is a significant difference between the concentration of ethyl lactate between both zones, it has an olfactory threshold of 150 mg/L, and no sample passed the threshold, except for Coop. Aguilar de Fra; therefore only this one presents a pineapple, varnish, balsamic aroma.

Regarding the polyols, glycerol is a viscous polyalcohol with a certain sweet taste that contributes to the body of fermented beverages [41]. It has also been shown to improve flavor intensity [42], palatability and sweetness sensation [43], suppress musty flavor, and influence aroma volatility [44]. The results obtained from this compound showed a significant difference between the zones, since only those from the QZ exceeded the perception threshold, 1000 mg/L.

3.4. Minor Aroma Compounds

Table 3. Minor aroma compounds of the PZ and QZ zones. Values shown represent averages of triplicate samples (data are mean ± SD). Values with different superscript letters in the same row are a significantly different test (p < 0.05).

Compounds	Lagar Los Borbones	Lagar Saavedra	Lagar Cañada Navarro	Coop Aguilar de la Fra.	Coop San Acacio	Coop La Unión
Alcohols
Octanol	132.9 ± 7.7 c	56.2 ± 3.2 b	137.7 ± 11.7 c	207.2 ± 8.9 d	22.9 ± 1.5 a	149.6 ± 6.1 c
Hexanol	1099.2 ± 38.2 a	1310.9 ± 60.7 b	1488.1 ± 65.0 bc	2057.0 ± 60.7 b	1741.3 ±140.9 b	2663.8 ± 204.6 c
Decanol	3.10 ± 0.10 a	5.20 ± 0.20 b	4.30 ± 0.4 ab	9.10 ± 0.80 d	7.50 ± 0.30 c	10.8 ± 1.7 e
2-Ethyl-1-hexanol	82.1 ± 3.7 a	95.5 ± 3.2 d	117.2 ± 8.7 e	57.0 ± 2.2 b	48.2 ± 2.9 c	61.3 ± 2.1 b
Guaiacol	23.1 ± 1.1 c	24.4 ± 1.0 e	12.9 ± 1.2 c	4.10 ± 0.30 a	11.7 ± 0.8 d	6.70 ± 0.40 b
4-Vinylphenol	55.4 ± 3.4 b	84.0 ± 4.1 e	35.4 ± 2.5 c	9.20 ± 0.60 a	26.2 ± 1.7 d	7.20 ± 0.40 a
4-Ethylguaiacol	13.3 ± 0.7 bc	1.10 ± 0.10 ab	5.00 ± 3.80 c	0.0 ± 0.0 a	3.30 ± 0.30 d	0.0 ± 0.0 a
Dodecanol	11.9 ± 1.0 a	8.80 ± 0.30 bc	10.1 ± 0.7 d	7.90 ± 0.50 ab	7.60 ± 0.50 e	9.70 ± 0.50 cd
2-Furanmethanol	19.2 ± 0.9 b	18.9 ± 1.7 c	8.10 ± 0.60 b	2.70 ± 0.10 a	6.90 ± 0.40 c	4.10 ± 0.40 a
Terpenoids
E-Nerolidol	2.60 ± 0.10 a	2.70 ± 0.10 a	2.70 ± 0.30 a	2.60 ± 0.10 a	2.70 ± 0.20 a	2.70 ± 0.10 a
B-Citronellol	15.3 ± 0.6 b	12.4 ± 0.8 a	27.8 ± 2.1 b	45.5 ± 3.1 c	25.8 ± 1.7 a	28.5 ± 1.3 b
Z-Nerolidol	0.50 ± 0.01 a	0.20 ± 0.01 b	0.50 ± 0.01 c	0.10 ± 0.01 a	0.0 ± 0.0 c	0.0 ± 0.0 a
Limonene	27.0 ± 2.9 b	25.9 ± 1.9 c	25.1 ± 1.1 c	17.1 ± 1.5 a	21.5 ± 1.4 c	20.9 ± 1.0 b
Trans-Methyldihydrojasmonate	2.80 ± 0.30 a	3.00 ± 0.10 d	2.40 ± 0.30 b	1.90 ± 0.20 a	1.80 ± 0.20 c	2.90 ± 0.10 d
B-Farnesene	7.70 ± 0.10 f	4.50 ± 0.40 b	3.00 ± 0.10 a	5.60 ± 0.30 d	6.50 ± 0.40 e	4.50 ± 0.40 c
E-Citral	0.20 ± 0.01 b	0.90 ± 0.10 c	0.50 ± 0.01 b	0.50 ± 0.01 b	0.50 ± 0.01 a	1.00 ± 0.10 d
Z-Citral	4.80 ± 0.70 c	2.40 ± 0.10 b	2.50 ± 0.20 bc	0.80 ± 0.01 a	2.80 ± 0.20 d	2.70 ± 0.20 bc
Farnesol	7.60 ± 0.10 d	0.0 ± 0.0 a	0.0 ± 0.0 a	5.60 ± 0.30 a	0.0 ± 0.0 a	0.0 ± 0.0 c
B-Damascenone	0.50 ± 0.01 b	0.30 ± 0.01 a	0.40 ± 0.01 a	3.00 ± 0.20 d	2.40 ± 0.10 a	2.80 ± 0.20 c
B-Ionone	0.10 ± 0.01 c	0.20 ± 0.01 d	0.10 ± 0.01 b	0.10 ± 0.01 a	0.10 ± 0.01 b	0.20 ± 0.01 d
Aldehydes
(E,E)-2,4-Nonadienal	0.20 ± 0.01 bc	0.20 ± 0.01 bc	0.10 ± 0.01 b	0.30 ± 0.01 d	0.10 ± 0.01 a	0.20 ± 0.01 c
E-2-Octenal	1.80 ± 0.10 f	1.10 ± 0.01 c	0.90 ± 0.01 b	0.0 ± 0.0 a	1.40 ± 0.10 d	1.50 ± 0.10 e
2,4-Decadienal,(E,E)	0.30 ± 0.10 c	0.20 ± 0.10 b	0.10 ± 0.10 a	0.20 ± 0.01 b	0.20 ± 0.01 ab	0.20 ± 0.10 b
E-2-Nonenal	0.40 ± 0.01 a	0.90 ± 0.01 d	0.50 ± 0.01 b	0.70 ± 0.10 c	1.00 ± 0.10 e	0.90 ± 0.10 de
Heptanal	0.80 ± 0.01 a	0.80 ± 0.01 a	0.90 ± 0.10 a	0.80 ± 0.01 a	0.80 ± 0.01 a	1.30 ± 0.10 b
Phenylacetaldehyde	1.80 ± 0.01 a	2.10 ± 0.10 a	2.10 ± 0.10 a	3.00 ± 0.10 a	1.80 ± 0.10 a	2.80 ± 0.10 a
Hexanal	4.62 ± 0.32 a	5.80 ± 0.40 b	6.10 ± 0.20 bc	5.60 ± 0.20 b	5.70 ± 0.60 b	6.70 ± 0.50 c
Octanal	0.70 ± 0.10 d	0.90 ± 0.10 bc	0.80 ± 0.10 ab	1.00 ± 0.10 c	1.40 ± 0.10 a	1.30 ± 0.10 d
Nonanal	1.82 ± 0.10 d	1.90 ± 0.10 b	1.00 ± 0.10 a	4.20 ± 0.20 c	4.60 ± 0.10 b	5.30 ± 0.30 e
Decanal	2.40 ± 0.10 c	2.30 ± 0.20 b	1.30 ± 0.10 a	8.40 ± 0.60 c	8.10 ± 0.40 b	10.00 ± 0.60 d
Benzaldehyde	4.90 ± 0.40 d	4.90 ± 0.20 cd	4.30 ± 0.40 bc	2.20 ± 0.10 a	4.30 ± 0.30 b	4.00 ± 0.50 b
4-Methylbenzaldehyde	1.40 ± 0.20 d	1.10 ± 0.90 b	1.00 ± 0.10 b	0.80 ± 0.10 a	1.00 ± 0.10 b	1.30 ± 0.10 c
3-Methylbenzaldehyde	0.80 ± 0.10 c	0.30 ± 0.01 a	0.50 ± 0.10 b	0.50 ± 0.10 b	0.80 ± 0.01 d	0.80 ± 0.10 c
Hexylcinnamaldehyde	1.01 ± 0.01 ab	1.01 ± 0.01 ab	0.90 ± 0.01 a	1.01 ± 0.02 bc	1.10 ± 0.10 d	1.00 ± 0.02 c
Ketones
Acetophenone	0.10 ± 0.01 a	2.20 ± 0.10 bc	0.20 ± 0.01 a	2.20 ± 0.10 b	2.40 ± 0.40 bc	2.70 ± 0.50 c
3-Methylacetophenone	0.10 ± 0.01 a	0.70 ± 0.10 c	0.70 ± 0.10 d	0.40 ± 0.30 b	0.10 ± 0.00 a	0.80 ± 0.01
3-Heptanone	1.35 ± 0.10 ab	1.40 ± 0.10 bc	1.20 ± 0.13 e	1.10 ± 0.10 a	1.70 ± 0.10 de	1.50 ± 0.10 cd
2-Decanone	2.80 ± 0.01 b	2.80 ± 0.01 a	2.80 ± 0.01 a	3.30 ± 0.10 d	2.90 ± 0.01 a	3.10 ± 0.01 c
3-Octen-2-one	1.30 ± 0.10 e	3.10 ± 0.20 d	1.90 ± 0.10 b	3.00 ± 0.60 d	3.20 ± 0.10 a	2.40 ± 0.10 c
Z-Geranylacetone	1.50 ± 0.01 b	1.50 ± 0.01 a	1.50 ± 0.10 a	2.6 ± 0.20 bc	2.40 ± 0.10 a	2.7 ± 0.20 c
Benzophenone	6.70 ± 0.53 b	3.30 ± 0.10 c	5.20 ± 0.40 d	0.90 ± 0.10 a	1.20 ± 0.10 e	1.30 ± 0.01 b
E-Geranilacetone	2.40 ± 0.20 a	0.70 ± 0.10 a	2.00 ± 0.20 c	4.40 ± 0.20 e	0.80 ± 0.10 d	1.60 ± 0.10 b
Esters
Ethyl butanoate	25.2 ± 1.0 a	59.0 ± 2.8 c	55.7 ± 2.7 c	55.7 ± 5.0 c	36.5 ± 3.2 b	31.5 ± 1.8 b
γ-Nonalactone	10.1 ± 0.9 a	14.7 ± 0.9 b	20.1 ± 1.0 c	20.4 ± 1.2 c	23.7 ± 1.6 c	85.5 ± 5.6 d
Ethyl 3-methylbutanoate	1.20 ± 0.10 c	2.90 ± 0.01 d	9.00 ± 0.30 e	0.90 ± 0.01 b	0.60 ± 0.01 a	0.50 ± 0.01 a
Ethyl isobutanoate	36.5 ± 1.7 d	167.5 ± 2.0 f	150.3 ± 3.0 e	28.5 ± 0.4 c	14.4 ± 0.4 a	18.0 ± 1.0 b
Butyl acetate	6.40 ± 0.32 c	2.30 ± 0.10 a	2.90 ± 0.10 b	14.6 ± 0.9 d	5.10 ± 0.10 c	14.6 ± 0.5 d
Octyl acetate	1.60 ± 0.10 d	0.70 ± 0.10 b	0.90 ± 0.10 c	4.0 ± 2.00 e	0.60 ± 0.10 ab	0.50 ± 0.01 a
Hexyl hexanoate	3.80 ± 0.30 d	3.50 ± 0.10 b	0.0 ± 0.0 a	3.50 ± 0.20 b	3.50 ± 0.10 c	3.50 ± 0.20 b
Ethyl benzoate	0.20 ± 0.01 c	0.10 ± 0.01 a	0.30 ± 0.01 d	0.40 ± 0.01 e	0.40 ± 0.01 f	0.0 ± 0.0 b
Isoamyl acetate	132.3 ± 7.8 a	166.7 ± 8.6 a	904.0 ± 53.0 d	688.9 ± 25.4 c	332.7 ± 20.1 b	689.2 ± 29.1 c
2-Phenylethyl butanoate	1.10 ± 0.10 b	6.80 ± 0.40 e	2.00 ± 0.20 c	2.70 ± 0.10 d	0.40 ± 0.01 a	1.10 ± 0.01 b
Ethyl hexanoate	1.00 ± 0.10 a	43.1 ± 1.9 b	43.3 ± 1.7 b	58.6 ± 4.1 c	59.0 ± 4.0 c	58.1 ± 4.9 c
Ethyl octanoate	22.0 ± 1.7 a	20.3 ± 0.8 a	19.4 ± 0.2 a	80.5 ± 3.1 d	38.0 ± 4.0 b	55.0 ± 3.3 c
cis-3-Hexenylbutyrate	4.00 ± 0.01 ab	4.40 ± 0.30 c	4.20 ± 0.30 bc	3.80 ± 0.20 a	4.10 ± 0.20 bc	3.90 ± 0.10 ab
2-Methylbutanoate	0.70 ± 0.01 a	4.50 ± 0.30 b	5.80 ± 0.30 c	0.60 ± 0.01 a	0.10 ± 0.01 a	0.70 ± 0.10 a
Hexyl acetate	1.20 ± 0.10 a	0.60 ± 0.01 a	1.80 ± 0.10 a	0.90 ± 1.50 c	2.50 ± 0.20 a	8.30 ± 0.30 b
Ethyl decanoate	19.1 ± 0.8 a	31.0 ± 1.0 c	23.5 ± 0.5 b	51.4 ± 4.0 d	55.8 ± 2.7 e	57.4 ± 0.6 e
Ethyl heptanoate	0.0 ± 0.0 a	0.10 ± 0.01 c	0.10 ± 0.01 b	0.10 ± 0.01 d	0.10 ± 0.01 e	0.20 ± 0.01 f
Z-3-Hexenol acetate	1.30 ± 0.10 b	0.0 ± 0.0 a	5.40 ± 0.30 c	0.0 ± 0.0 a	10.3 ± 0.6 d	3.40 ± 0.30 b
Phenylethyl hexanoate	0.20 ± 0.01 b	0.20 ± 0.01 c	0.20 ± 0.01 b	0.20 ± 0.01 c	0.20 ± 0.01 b	0.01 ± 0.01 a
Ethyl benzoate	0.20 ± 0.01 b	0.10 ± 0.01 b	0.30 ± 0.01 d	0.40 ± 0.01 e	0.0 ± 0.0 a	0.10 ± 0.01 c
Ethyl dodecanoate	19.1 ± 0.7 b	31.1 ± 1.0 d	23.5 ± 0.4 c	9.70 ± 0.50 a	20.3 ± 0.9 b	57.4 ± 0.6 e
Phenethyl benzoate	1.20 ± 0.10 d	2.10 ± 0.01 ab	2.20 ± 0.01 c	3.40 ± 0.20 e	3.20 ± 0.10 a	3.30 ± 0.10 de
Ethyl undecanoate	0.50 ± 0.01 c	0.60 ± 0.01 d	0.60 ± 0.01 bc	0.50 ± 0.01 bc	0.60 ± 0.01 b	0.0 ± 0.0 a
B-Phenylethyl acetate	26.5 ± 1.7 a	3.30 ± 0.20 b	4.20 ± 0.20 b	0.0 ± 0.0 a	0.20 ± 0.01 c	35.0 ± 3.0 d
Ethyl hexadecanoate	16.4 ± 0.9 e	10.6 ± 0.6 a	17.7 ± 0.8 b	33.0 ± 2.3 c	87.1 ± 4.1 b	69.0 ± 3.0 d
2-Phenylethylacetate	1154.2 ± 88.9 a	669.0 ± 35.0 b	1619.2 ± 107.9 d	650.4 ± 6.9 b	385.4 ± 14.4 c	1274.0 ± 99.0 c
Geranol acetate	1.30 ± 0.01 c	1.30 ± 0.10 a	2.90 ± 0.01 c	2.40 ± 0.10 b	2.90 ± 0.01 a	4.80 ± 0.10 d
Butyrolactone (mg/L)	21.0 ± 16.0 b	21.0 ± 16.0 b	20.0 ± 14.0 b	8.60 ± 0.70 a	8.80 ± 0.60 a	9.10 ± 0.40 a
2-Phenylbenzo acetate	2.30 ± 0.01 a	2.50 ± 0.01 d	2.30 ± 0.01 b	2.10 ± 0.01 a	2.01 ± 0.01 c	2.01 ± 0.01 a
Ethyl tetradecanoate	8.70 ± 0.10 ab	8.10 ± 0.40 a	10.0 ± 0.9 c	9.30 ± 0.70 bc	8.20 ± 0.30 ab	11.2 ± 1.0 d
Furanoic
5-HMF	16,521.0 ± 14 a	1140.1 ± 38.5 b	1106.0 ± 13.6 b	468.0 ± 14.0 a	466.4 ± 28.1 c	484.2 ± 35.6 a
Furfural	1707.7 ± 12.1 a	1536.7 ± 96.0 d	1013.6 ± 66.7 c	436.2 ± 38 b	735.4 ± 44.0 e	561.6 ± 46.5 a
5-Methylfurfural	645.0 ± 2.6 d	226.2 ± 15.2 b	377.4 ± 20.6 c	13.5 ± 0.7 a	26.0 ± 2.7 e	21.0 ± 2.0 d
Pentylfuran	3.30 ± 0.20 c	7.10 ± 0.20 d	7.70 ± 0.40 e	0.0 ± 0.0 ab	0.0 ± 0.0 ab	0.0 ± 0.0 a

shows all the minor compounds. Seventy-eight volatile compounds were determined, with the chemical families with the highest content being esters, aldehydes, and alcohols. As a result, of the esterification between alcohols and acids during must fermentation, esters contribute, together with alcohols, to the sweet and fruity aroma of wines [45]. Other compounds that contribute to the sweet aroma descriptor are the compounds included in the group of furanoic compounds, which are furfural and 5-methylfurfural [46], both of which have a higher concentration in wines from QZ than in PZ, as reflected in Figure 3. However, they do not pass the perception threshold, so they have no influence. Another compound is pentylfuran, which has a fruity, green odor descriptor with vegetal nuances, and its presence in wines from the quality zone can be explained by the high temperatures reached during the maturation period as explained by this author [47]. However, this only passes the perception threshold in the Lagar Cañada Navarro winery and in the Lagar Saavedra winery. According to other authors [48,49], another important group are the terpene compounds that have an important role in influencing the aroma of white wines, providing floral aromas. Within this group we have the b-citronellol, a compound that passes the threshold in all wines belonging to PZ; however, in the QZ, it is only present in Lagar Cañada Navarro; this compound gives the wines pink and green smells. Among the different families of esters, acetate esters are the most representative. The levels of acetate esters in all the wines obtained have representative values; however, only ethyl butanoate and ethyl isobutanoate exceeded the perception threshold in all wines. In the analyzed wines, we found isoamyl acetate, ethyl butanoate, ethyl 3-methylbutanoate, ethyl hexanoate, and ethyl octanoate, which passed the perception threshold, giving the wine a sweet aroma; this is supported by the study carried out by Martín-García et al. [45] and by Meng et al. [50]. 2-Phenylethanol exceeded the threshold in all wines except in the Lagar Saavedra winery and in the Coop. Aguilar de la Fra; this contributes notably to the aroma of the wines, giving the wine floral and fruity notes as shown by Lopez de Lerna et al. [51].

Within the aldehydes, those compounds that pass the threshold were decanal, (E,E)-2,4-nonadienal, 2,4-decadienal, (E,E)-, E-2-nonenal. The latter is present in all wines except in the Lagar Cañada Navarro winery and in the Lagar los Borbones, both belonging to the QZ. Nonanal is a compound that only passes the threshold in PZ wines, obtaining wines with a greater floral smell [50]. Other compounds important are limonene and butyrolactone, characterized by a citrus note with an odor reminiscent of orange [52,53]; both compounds are present in all wines.

3.5. Aromatic Series, Multivariate and Principal Component Analysis

The aroma of a wine is conditioned not only by its volatilome but also by the interactions between the different compounds that compose it, giving a positive or negative interaction, as confirmed in the study by Hein, Ebeler, and Heymann [54]. The analytical fingerprint of the volatilomes can be obtained by constructing aromatic series, whose values are determined as described in the Materials and Methods Section. Several authors [22,55] previously used this clustering approach, which offers the advantage of reducing the number of variables to consider when analyzing the differences between wineries or oenological treatments.

The OAV provides us with information on the influence that a compound has on the aroma of the wine. In this way, volatile compounds with an OAV greater than one indicates a potential contribution to the aroma of the wine. In the sections on major and minor aromas, those that have a greater influence on the series of our study have been described.

The values of the aromatic series obtained are shown in

Table 4. Aromatic series of different wines. Values shown represent averages of triplicate samples (data are mean ± SD). Values with different superscript letters in the same row are significantly a different test (p < 0.05).

Aromatic Series	Coop. Aguilar de la Fra.	Coop. La Union	Coop. San Acacio	Lagar Saavedra	Lagar Los Borbones	Lagar Cañada Navarro
Fruity	54.9 ± 0.3 d	44.8 ± 0.8 c	29.0 ± 2.0 b	28.7 ± 0.7 b	26.1 ± 0.8 a	47.0 ± 2.0 c
Green fruit	4.60 ± 0.30 b	6.70 ± 0.40 b	5.60± 0.70 c	4.40 ± 0.10 b	1.70 ± 0.10 a	6.80 ± 0.20 d
Green	7.90 ± 0.30 ab	9.40 ± 0.10 d	12.2 ± 0.7 e	7.40 ± 0.30 a	8.50 ± 0.10 c	8.10 ± 0.10 bc
Creamy	5.10 ± 0.20 b	5.00 ± 1.01 b	5.20 ± 0.30 b	3.50 ± 0.10 a	3.60 ± 0.10 a	4.50 ± 0.10 ab
Citrus	17.0 ± 0.2 d	18.8 ± 0.6 e	16.0 ± 1.0 c	9.90 ± 0.40 a	11.1 ± 0.4 b	8.90 ± 0.20 a
Chemistry	16.7 ± 0.5 b	11.7 ± 0.2 a	11.2 ± 0.3 a	16.1 ± 0.5 b	17.6 ± 0.5 c	20.6 ± 0.1 d
Honey	0.60 ± 0.30 bc	0.80 ± 0.01 d	0.50 ± 0.01 a	0.60 ± 0.01 c	0.54 ± 0.01 bc	0.50 ± 0.10 b
Waxy	23.10 ± 0.5 e	19.4 ± 0.8 d	14.0 ± 1.0 c	6.10 ± 0.20 ab	6.5 ± 0.40 b	5.10 ± 0.20 a
Herbal	2.60 ± 0.20 b	3.10 ± 0.10 c	3.20 ± 0.10 c	2.90 ± 0.10 c	1.90 ± 0.20 a	2.10 ± 0.10 a
Floral	14.5 ± 0.7 b	14.2 ± 0.3 b	10.3 ± 0.5 a	30.0 ± 2.1 d	22.0 ± 1.0 c	23.1 ± 1.0 c
Toasty	0.10 ± 0.01 a	0.10 ± 0.01 b	0.20 ± 0.10 c	0.35 ± 0.01 e	0.50 ± 0.01 f	0.20 ± 0.01 d

. The fruity series stands out in all the wineries, with Coop. Aguilar de la Fra. and Lagar Cañada Navarro being the wineries with the highest values. Likewise, the Citrus, Chemical, Waxy, and Floral series reach values higher than 10.

A star plot with eleven rays was obtained by using multivariate analysis of the aromatic series. These were standardized, so the maximum length of the rays was the same for all the series, unity being the medium value (Figure 3). In Figure 3a, it can be observed that the Lagar Cañada Navaro and Lagar Borbones wineries have very similar olfactory tendencies. The predominant aromatic series are Fruity, Green, Green fruit, Herbal, obtaining wines with fruity aromas, while in the Lagar Saavedra winery, the predominant aroma series is sweet (Honey). In Figure 3b, all the wines present the same aromatic series Citrus, Creamy, and Floral, however in the Coop. Aguilar de la Fra. the first two series are more accentuated. Through multivariate analysis of the aromatic series, a star graph with eleven rays was obtained, which were standardized, so that the maximum length of the rays was the same for all the series, with one being the average value (Figure 3). In Figure 3a, it can be observed that the Lagar Cañada Navaro and Lagar Borbones wineries present very similar olfactory tendencies. The predominant aromatic series are Fruity, Green, Green Fruit, Herbal, obtaining wines with fruity aromas, while in the Lagar Saavedra winery, the predominant aromatic series is sweet (Honey). In Figure 3b, all the wines present the same Citrus, Creamy, and Floral aromatic series, however in the Coop. Aguilar de la Fra. The first two series are more accentuated. This type of study was carried out by other authors such as Dumitriu et al., [55] and Moyano et al., [56].

With the aim of reducing the number of variables, a principal component analysis (PCA) was carried out where components are a lineal combination of the input variables, and the interpretation of the components obtained must be done by the analyst. Here, the aromatic input was selected as input variable; the aromatic and the first three components obtained explain 83.4% of the observed variability (Figure 4a,b).

The first component clearly differentiates the wineries of the quality zone from those of the production zone. The second component differentiates among the wineries of the production zone, and the third the wineries of the quality zone. The aromatic series with the greatest influence in the first component are the Citrus, Floral, and Toasty. While in component two, the series with the greatest influence are the Chemical, Fruity, and Green series. Regarding component three, the series with the greatest influence are the Herbal and Honey series. This makes us see that the wines from the cooperatives have a clear olfactory tendency towards citrus, wavy and floral smells, while those wines belonging to the quality zone have sweeter smells, such as honey, fruity, and herbaceous.

After performing the principal component analysis (PCA) between component one and component two, a clustering can be observed between the wineries in the quality zone and the cooperatives in the production zone, since in component one the series with the greatest influence are the Citrus, Floral, and Creamy series. While in component two, the values with the greatest influence are the Chemical, Fruity, and Green series. Regarding component three, the series with the greatest influence are the Herbal, Green fruit, and Honey series. This makes us observe that the wines from the cooperatives have a clear olfactory tendency towards citrus and floral smells, while those wines belonging to the quality zone have sweeter smells, such as honey, fruity, and herbaceous. This is related to the different enzymatic activities, as indicated in Section 3.1.

4. Conclusions

The study revealed notable distinctions between the quality and production zones in terms of microbial composition and enzymatic activity, which significantly influenced the wine’s aromatic profile. The production zone exhibited a greater diversity of non-conventional yeasts and higher beta-glucosidase enzyme activity, resulting in wines with enhanced glycoside-derived aromatic compounds. Conversely, while the quality zone displayed a lower variety of non-conventional yeasts, it demonstrated superior cellulase enzyme activity, yielding wines with more pronounced fruity and clarified aromas. Analysis of the volatilome further emphasized these differences, with production zone wines characterized by citric and floral notes, while quality zone wines presented a more fruity and sweet aromatic profile. These findings underscore the complex interplay between terroir, microbial ecology, and enzymatic processes in shaping the sensory attributes of wines from different viticultural areas.