Alcoholic Fermentation as a Strategy to Mitigate Pesticides and Mycotoxins

In this study, alcoholic fermentation was explored to reduce the pesticides—2,4-dichlorophenoxyacetic acid (2,4-D) and procymidone—and the mycotoxin ochratoxin A (OTA), besides evaluating their effects on quality parameters of the process. Fermentation (at 20 °C for 168 h) was conducted in synthetic must (YPD and 10 g hL−1 of yeast Saccharomyces cerevisiae) to which the following were added: 2,4-D (6.73 mg L−1) and procymidone (2.24 mg L−1), simultaneously (treatment 1); and OTA (0.83 and 2.66 µg L−1) (treatments 2 and 3). The control treatment had no contaminants. All contaminated treatments showed lower ethanol production and cell viability. The highest decrease in contaminants was found after 168 h of fermentation: 22 and 65% in the cases of both pesticides while OTA decreased 52 and 58% (treatments 2—0.83 µg L−1 and 3—2.66 µg L−1). Decrease in procymidone and OTA due to metabolic processes in yeast cells is mainly related to peroxidase activity and glutathione production.


Introduction
Mycotoxins and pesticides are contaminants found in raw material and processed products, such as grapes and wine (Freire et al., 2020;Scariot et al., 2022). They have given cause for concern due to risks that result from human and animal exposure to them in their diets (Čepo et al., 2018;Gavahian et al., 2020). Fungicides have often been used in grape cultures to inhibit fungal growth and, consequently, mitigate losses caused by toxigenic fungi and other pathogens found in this culture (Carpinteiro et al., 2010;Gava et al., 2021). Procymidone is one of the most common fungicides applied to grapevines (Čuš et al., 2010a, b;Romanazzi & Feliziani, 2014;Gava et al., 2021;Shen et al., 2021).
Procymidone is an amide fungicide which has been used to control diseases in grapes, such as gray mold, that are mainly caused by Botrytis cinerea. Its occurrence in grapes has become increasingly severe and has led to decrease in productivity (Shen et al., 2021). Control of the disease has been proven by several studies that report incidence of procymidone in grapes and processed products. Sala et al. (1996) found procymidone residues in white must after pressing (2.0 mg L −1 ) and in red must after destemming and crushing (3.55 mg L −1 ). Rose et al. (2009) showed that procymidone concentration in chardonnay grapes was higher than the maximum residue limit tolerated in Australian grapes (2.0 mg kg −1 ) (FSANZ, 2006), since 4.5 mg kg −1 was reported. Čuš et al. (2010a, b) identified maximum concentrations (0.05 and 0.13 mg L −1 ) of the fungicide in wine, respectively. Besides, other studies have focused on decrease of the fungicide in stages of the vinification process, such as filtration and clarification (Doulia et al., 2016(Doulia et al., , 2017. Another problem that the vitiviniculture sector has faced is cross-contamination caused by pesticides whose origins are other agricultural crops, such as rice and soybean, which also cause negative impact on grapevines, mainly the herbicide 2,4-dichlorophenoxyacetic acid (2,4-D) (Rossouw et al., 2019). Its effect may be found several kilometers from the target, mainly when there are prevailing winds when it is applied (Felsot et al., 2010).
Frequent use of active ingredients in vineyards (Costa et al., 2019;Medina et al., 2007), too much irrigation, humid climate, and fine grape skin enable colonization by toxigenic fungal species that may produce mycotoxins under stress conditions (Csutorás et al., 2013;González-Domínguez et al., 2019;Torović et al., 2020). Due to their high toxicity and risks to health, mycotoxins have been investigated in raw material and food (Costa et al., 2019;Kochman et al., 2021;Medina et al., 2007). The most important mycotoxin in wine is ochratoxin A (OTA), which is produced by different fungal species that belong to both genera Penicillium and Aspergillus (Čepo et al., 2018;Kersh et al., 2022). Taking into consideration their carcinogenic properties (IARC, 1993), their levels in wine must be strictly controlled, mainly regarding OTA intake in the daily diet, i.e., from 10 to 13% of the intake may originate in wine (Miraglia & Brera, 2002). In the legislation, maximum OTA level in wine is 2 μg L −1 (ANVISA, 2011;EC, 2006) while weekly OTA intake was established at 0.12 μg kg −1 of body weight (EFSA, 2010).
Regarding forms of detoxification of contaminants, alcoholic fermentation with Saccharomyces cerevisiae along with malolactic fermentation with lactic bacteria has stood out as a promising biological method that decreases concentrations of mycotoxins and pesticides (Boeira et al., 2021;Doulia et al., 2016Doulia et al., , 2017Scariot et al., 2022;Torović et al., 2020). Its potential may be related to a mechanism of defense against oxidative stress caused by contaminants since it may activate protection molecules and activities of enzymes, such as reduced glutathione (GSH) and peroxidase (PO) (Boeira et al., 2021;Dong et al., 2007;Halliwell & Gutteridge, 2015;Junior et al., 2021). Therefore, studies of mitigation of mycotoxins and pesticides by alcoholic fermentation with yeasts should be carried out. In this study, alcoholic fermentation was explored to reduce the pesticides-2,4-dichlorophenoxyacetic acid (2,4-D) and procymidone-and the mycotoxin ochratoxin A (OTA).

Conditions of Alcoholic Fermentation
Submerged cultivation was conducted in 500-mL Erlenmeyer flasks with 240 mL synthetic must and 10 mL inoculum. Fermentation was carried out at 26 °C for 168 h. One gram per hectoliter of the acid-lactic bacterium O. oeni was added 48 h after the beginning of alcoholic fermentation.
Decrease in contaminant concentrations and their effects on alcoholic fermentation were evaluated by adding 2,4-D, procymidone, and OTA to the flasks. Both pesticides were evaluated simultaneously, i.e., 2,4-D at 6.73 mg L −1 and procymidone at 2.24 mg L −1 (treatment 1) were added at the same time. Both concentrations were found in a study of the occurrence of contaminants in southern Brazil (unpublished data). This procedure was also used in relation to both OTA concentrations in the must: 0.84 µg L −1 (treatment 2) was the concentration found by the study of its occurrence while 2.66 µg L −1 (treatment 3) was the maximum limit established by the Brazilian legislation and the European one in the case of grapes (2.0 µg kg −1 ).
Solutions of the mycotoxin and pesticides were added to the flask and followed by solvent evaporation under nitrogen flow. Afterwards, synthetic must was added to them and homogenized for 10 min. Then, the inoculum was also added. The control treatment was carried out with no contaminant. Besides, a treatment with neither yeasts nor contaminants was conducted to evaluate the fermentation medium. Aliquots were aseptically collected every 24 h to monitor PO activity and concentrations of the following: cell, reducing sugar, GSH, ethanol, 2,4-D, procymidone, and OTA.

Analytical Determinations
Biomass concentration was quantified by a spectrophotometer (FEMTO Cirrus-80) by optical density (OD) at 600 nm. OD was converted into dry biomass with the use of the yeast standard curve by the gravimetric method (dry weight). Cells were dried at 60 °C up to constant weight (Chang et al., 2018).
Cell viability (%) was carried out by the trypan blue dye exclusion assay (0.2%). The dye (20 µL) was added to a sample aliquot (500 µL). After a minute, a drop of the solution was placed in a Neubauer chamber and viable and non-viable cells were counted in a microscope. Calculation was estimated by the relation between viable cells and non-viable ones (Strober, 2015).
Maximum growth rate (µ max ) resulted from exponential regression applied to the phase of logarithmic growth (Hamidi-Esfahani et al., 2007). Intracellular GSH was quantified by the colorimetric method determined by Owens and Belcher (1965) with the use of 5,5-dithiobis-2-nitrobenzoic acid (DTNB) evaluated by a spectrophotometer (FEMTO Cirrus-80) at 412 nm. GSH concentration (mg L −1 ) was calculated with the use of the L-GSH standard curve. PO activity was evaluated by a spectrophotometer at 470 nm in agreement with Garda-Buffon et al. (2011), who use hydrogen peroxide and guaiacol as the substrate. An enzyme unit (U) was defined as the amount of enzyme which is capable of oxidizing or hydrolyzing 1 µmol substrate per min (Feltrin et al., 2017). Molar absorptivity of guaiacol (Ɛ 470 ) de 26,600/(M cm) was considered.
Reducing sugars were determined by the spectrophotometric method of 3,5-dinitrosalicylic acid (DNS) (Miller, 1959), which uses a glucose standard curve.
Ethanol was quantified in agreement with Klarić et al. (2015). A gas chromatograph (CG-2010 Shimadzu) equipped with a split/splitless and flame ionization detector (FID) and a Crossbond column (diphenyl dimethyl polysiloxane), 30 m in length, 0.25 mm in inner diameter, and 0.23 μm in thickness (75% phenyl-methylpolysiloxane film), was used. Manual injection volume was 1 µL of fermented medium, initial temperature was 230 °C, column pressure was 2.33 psi, and nitrogen carrier gas flow was 0.5 mL min −1 . Average velocity of nitrogen gas was 10 cm s −1 . Oven temperature was kept at 26 °C for 7 min and then increased to 50 °C at 1 °C min −1 up to 200 °C (at 15 °C min −1 ) for 4 min. FID temperature was 250 °C while nitrogen carrier gas flow was 25 mL min −1 . Ethanol was determined by the standard curve of ethanol diluted in ultrapure water and results were expressed as percentages.

Extraction and Quantification of Pesticides in the Fermented Medium
The herbicide 2,4-D and the fungicide procymidone were extracted from fermented media by the QuEChERS method described by Payá et al. (2007), with modifications, i.e., citrate salts sesquihydrate and dihydrate were eliminated. Extraction was carried out by MeCN and the extract was cleaned by both salts magnesium sulfate (MgSO 4 ) and sodium chloride (NaCl). The resulting extract was used for identifying and quantifying the pesticides by LC-DAD and GC-MS.
To separate the fungicide procymidone, the RTX-5MS column (30 m × 0.25 mm ID × 0.25 μm) was used in a Shimadzu gas chromatograph QP2010 Plus (Kyoto, Japan) equipped with an autosampler (AOC-20i) coupled to a mass spectrometer with a quadrupole mass filter (GC-MS). Injection volume was 1 µL, as described by Barbosa et al. (2020).

Extraction and Quantification of OTA
OTA was extracted from the fermented medium by the QuEChERS method described by Fernandes et al. (2013), with the use of acidified MeCN (1% acetic acid) and partition with addition of a mix of salts: MgSO 4 , NaCl, sodium citrate tribasic dihydrate, and sodium citrate dibasic sesquihydrate. Resulting dry extract was resuspended in the mobile phase (60% MeCN, 40% Milli-Q water acidified with 1% acetic acid), followed by chromatographic injection (20 µL).

Reduction in Contaminant Concentrations
Reduction of contaminants in alcoholic fermentation was evaluated by Eq. (1), where Q0 i is the amount of residue of contaminants at the beginning of the process and Q i is the amount of residue of contaminants in the processed product (Doulia et al., 2017;Hou et al., 2020).

Redox-Specific Degradation and Velocity of Degradation in Relation to GSH and PO of Contaminants in Alcoholic Fermentation
Estimate of the redox-specific degradation (RSD) of contaminants in fermented medium was based on GSH concentration and PO activity in yeast cells by Eq. (2), where PDE (%) is the percentage of degradation of the contaminant determined in the fermented medium, GSH is GSH concentration (mg mL −1 ), PO is peroxidase activity (U mL −1 ), and t is sample collection time in the fermentation process (h).
Estimate of velocity of degradation in relation to GSH and PO (VDE GSH*PO ) was based on the degraded mass of the contaminant, GSH concentration, PO activity, and fermentation time (t), expressed as Eq. 3 (Garda-Buffon et al., 2011).

Effects of Contaminants on Alcoholic Fermentation
Biochemical and Kinetic Characterization of Alcoholic Fermentation: Cell Concentration and Viability, GSH, PO, Reducing Sugar, Ethanol, and µ max . It is fundamental to follow cell concentration, GSH production, PO activity, reducing sugar concentration, ethanol production, and cell viability to evaluate effects of mycotoxins and pesticides on alcoholic fermentation. Therefore, these parameters were monitored daily in the 168-h fermentation (Table 1). Regarding cell concentration, exponential growth was found up to 72 h of fermentation in all culture media (Tables 1 and 2). After 72 h, cell concentrations decreased differently as the result of treatments. The highest decrease was found after 168 h in treatments exposed to OTA. Both treatments 2 and 3 (exposed to OTA) led to decrease in cell concentrations that ranged from 28 to 40% after 168 h, by comparison with the other treatments (Table 1). Both 2,4-D and procymidone exhibited no significant difference from the control treatment at the end of alcoholic fermentation, after 168 h (Table 1). Maximum growth rate (µ max ) also decreased in contaminated treatments, i.e., 9 and 12.5% in treatments exposed to OTA and 1.0% in the treatment exposed to pesticides, by comparison with the control treatment (Table 2). Besides, 24 h after fermentation, there was decrease in cell viability in all contaminated treatments by comparison with the control treatment (Table 1) while 168 h after fermentation, there was decrease of 15% in cell viability in treatment 1 and 4 and 7% in treatments 2 and 3, respectively, by comparison with the control treatment. Simultaneous addition of pesticides to culture media decreased cell viability significantly.
The mechanism of action of pesticides in cells is distinct and not very clear, depending on the organism (Owsiak et al., 2021). Herbicides, including 2,4-D, have been known for affecting biological systems negatively (Ritcharoon et al., 2020). Since negative effects affect gene expression, they trigger responses to stress and lead to interruption of cell cycle control, of immune responses, and of DNA repair (Bharadwaj et al., 2005).
Toxicity of 2,4-D towards yeast cells is mainly due to the activity of the non-dissociated form (Cabral et al., 2003). It suggests that the lipid bilayer in plasma membranes is one of the biological targets of the herbicide; it may be either due to direct interaction between this highly lipophilic form and membrane lipids, thus affecting spatial organization of membranes (Heipieper et al., 1994), or due to lipid peroxidation as the consequence of its activity as a pro-oxidant agent (Teixeira et al., 2004). On the other hand, fungicides aim at disruption of integrity of 1 3 cell membranes and cell walls of fungi. Toxic effects of fungicides may not only result in instability in cell walls, changes in osmolarity, and production of reactive oxygen species (ROS) (Hayes et al., 2014), but also induce oxidative stress as the result of accumulation of free radicals in cells (Grosicka-Maciąg, 2011). Different activities of both pesticide classes may have influenced high decrease in viability found in treatment 1. Teixeira et al. (2004) showed that exposure of S. cerevisiae cells to 2,4-D at 0.45 and 0.65 mM induces a period of growth latency in which the cell population loses viability, followed by resumption of exponential growth of the adapted population. It shows that toxic compounds in alcoholic fermentation influence concentration and/or cell viability in yeasts and may cause cell death induced by chemical stress of contaminants. Table 1 Biochemical characterization of cultures with Saccharomyces cerevisiae in synthetic must (YPD) with and without 2,4-D, procymidone, and OTA Treatment 1: 6.73 mg L −1 2,4-D + 2.24 mg L −1 procymidone. Treatment 2: 0.83 µg L −1 OTA. Treatment 3: 2.66 µg L −1 OTA. All experiments exhibited coefficient of variations below 5%. Different lowercase letters in a column show significant differences among days. Different uppercase letters on lines show significant differences among treatments (p < 0.05)
Increase in PO activity and in GSH concentration was different in every treatment. The control treatment exhibited the highest enzyme activity after 72 h of fermentation; it was 15% higher than the treatment exposed to pesticides and 18% lower than treatments exposed to OTA. However, the highest concentration of GSH was found after 96 h; it was 93% higher than the treatment exposed to pesticides and 53 and 13% higher than treatments 2 and 3 exposed to OTA, respectively. Production of these molecules results from metabolic pathways that are typical of alcoholic fermentation (Boeira et al., 2021;Scariot et al., 2022) but, in treatment 1, the opposite of the control treatment was observed. In 72 h of fermentation, there was increase in GSH production, which was 7% higher that the control treatment and 9 and 18.5% lower that treatments 2 and 3. The highest PO activity was found after 96 h, i.e., 20% higher that the control treatment and 18% lower than treatments 2 and 3.
In treatments with OTA, PO activity in treatment 2 was higher after 168 h by comparison with both treatment 1 (27%) and the control treatment (52%). However, increase in GSH concentration agreed with the one of the contaminant after 48 h in treatment 3 (2.66 μg L −1 ), the highest concentration in the shortest fermentation time and after 72 h in treatment 2 (0.84 μg L −1 ). They were 33 and 90% higher than the control treatment and 16 and 9% higher than treatment 1, respectively.
These differences are strongly related to contaminants added to the media (Viegas et al., 2005). Both treatments exposed to OTA (treatments 2 and 3) exhibited the highest PO activities after 168 h (11.47 U mL −1 ), by comparison with the other treatments (Table 1). It is due to the direct correlation among the mycotoxin, its toxicity, and the potential of yeast cells to produce specific enzymes that act on the maintenance of metabolic activity of microorganisms (Boeira et al., 2021). In contaminated treatments, increase in GSH production took place before the highest PO activity was found. This metabolic alteration may show that cells, after 48 h of culture, develop mechanisms of biodegradation of compounds that are oxidative to yeasts and may be associated with degradation of toxic compounds, a metabolic pathway induced by the contaminants throughout the culture (Garda-Buffon & Badiale-Furlong, 2010).
Contaminants lead to generation of ROS and convert GSH, which is the most abundant antioxidant molecule in the intracellular medium, into oxidized glutathione (GSSG), thus, decreasing toxicity in the medium (Lu, 2013). Antioxidant activity of GSH is mostly carried out by reactions catalyzed by GSH peroxidase (GPx), which reduce hydrogen peroxide and lipid peroxide as GSH is oxidized to GSSG. Regeneration of GSH from GSSG takes place through glutathione reductase. Thus, to balance redox reactions, cells may induce glutathione reductase activity in order to increase GSH and the relation GSH/GSSG (Bitani et al., 2022).
This study is extremely important since data on activities of protection molecules and enzyme systems, such as GSH and PO, were related to different toxic compounds. Thus, it shows that both pesticides and the mycotoxin respond differently to production of molecules GSH and PO related to the fermentation period connected with molecule conversion to keep redox balance of cells.
The end of alcoholic fermentation was confirmed by low levels of reducing sugars, between 1.7 and 2.3 mg mL −1 after 168 h (Table 1). Briz-Cid et al. (2018) evaluated the influence of four treatments with fungicides (metrafenone, boscalid + kresoxim-methyl, fenhexamid, and mepanipyrim) on Tempranillo wine. The authors found that fermentation kinetics is influenced not only by grape composition but also by fungicides. Grapes treated with a mix of boscalid (200 mg mL −1 ) and kresoxim-methyl (100 mg mL −1 ) exhibited delay at the beginning of alcoholic fermentation but ended together with the others. Therefore, high concentrations of contaminants (mycotoxins and pesticides) in grapes may result in slow fermentation or paralyze it and, consequently, harm ethanol production (Kłosowski et al., 2010). Table 2 Maximum growth rate (µ max ) of cultures conducted in synthetic must (YPD) with and without 2,4-D, procymidone, and OTA Treatment 1: 6.73 mg L −1 2,4-D + 2.24 mg L −1 procymidone. Treatment 2: 0.83 µg L −1 OTA. Treatment 3: 2.66 µg L −1 OTA

Treatments µ max (h −1 ) Exponential phase interval (h) Reduction of µ max (%)
Control treatment 0.0573 0-72 -Treatment 1 0.0567 0-72 1 Treatment 2 0.0552 0-72 9 Treatment 3 0.0501 0-72 12.5 Ethanol concentration is also related to GSH production (Table 1) (Margalef-Català et al., 2017;Wen et al., 2005), which may be classified into three phases throughout alcoholic fermentation. In the first phase, glucose levels decrease gradually while concentrations of ethanol and glutathione increase. In the second phase, ethanol is used as the carbon source for cell growth and glutathione synthesis. In the third phase, both glucose and ethanol are consumed and cells stop multiplying (Wen et al., 2005). Table 1 shows that, between 48 and 72 h of fermentation, increase in ethanol levels may be related to increase in GSH production in treatments exposed to contaminants, by comparison with the control treatment, an effect that was not observed in the control treatment, when GSH only increased after 96 h of fermentation. Dong et al. (2007) stated that cell stress-in this case, caused by contaminants-influences increase in energy consumption by yeasts and leads to changes in metabolism and accumulation of some protection molecules, such as GSH. Although some studies show initial inhibition caused by contaminants, yeasts outperform them and resume fermentation (Briz-Cid et al., 2018;Scariot et al., 2022). This fact is confirmed by sugar consumption and ethanol production, markers that are fundamental to control fermentation (Hu et al., 2022;Samphao et al., 2018;Zhang et al., 2022). Benucci and Esti (2021) conducted their study from the sequential inoculation of T. delbrueckii and S. cerevisiae compared to S. cerevisiae in a white grape must to assess arginase uptake. The authors observed that the rate of sugar consumption was significantly higher (38%) when the alcoholic fermentation was carried out by the two yeasts in relation to S. cerevisiae alone, and also verified that there was a delay of 5 days in the conclusion of the alcoholic fermentation with only S. cerevisiae. The authors highlighted high similarity in arginase activity mainly related to the presence of non-Saccharomyces yeast.
In this sense, Abedi et al. (2022) evaluated the binding capacity and stability of OTA (10 µg kg −1 ) by single and co-culture of Lactobacillus acidophilus + Lactobacillus rhamnosus, including viable cells, inactivated by heat and ultrasound. The authors showed that the co-culture of these two lactic acid bacteria was more effective than a single one to remove OTA in bakery, and the greatest binding capacity and stability of OTA was the one treated with ultrasound (8.418 and 7.829 µg kg −1 ). In our study, we also demonstrated advantages in relation to sequential inoculation of non-Saccharomyces and Saccharomyces yeasts, as well as the addition of lactic acid bacteria. The union of both microorganisms, even in the presence of contaminants, made the alcoholic fermentation to be completed.
The importance of malolactic fermentation in winemaking was investigated by du Toit et al. (2010), as it de-acidifies the wine, contributes to microbial stability, and helps in the aromatic profile of the wine through the production of metabolites, as well as in the reduction of contaminants. Other methods that can be used concomitantly, such as plasmaactivated water (Sarangapani et al., 2020) and ultrasound (Abedi et al., 2022), are also used to reduce pesticides and mycotoxins in grapes and other food marts. However, for the investigated contaminants, the alcoholic fermentation was efficient in the reduction of 22 to 65% of the levels, characterizing the potential of the process.
After 24 h of fermentation, decrease in contaminants in culture media is simultaneous to increase in PO activity (Tables 1 and 3), the period in which the yeast activates metabolic pathways to decrease toxicity in the culture medium, thus, preserving cells in the fermentation process (Garda-Buffon & Badiale-Furlong, 2010). Therefore, the highest RSD and VDE GSH*PO were found after 24 h of alcoholic fermentation in the cases of 2,4-D, procymidone, and OTA in all treatments (Table 3). The highest percentage of procymidone degradation was 61.18%deg/ GSH *U PO *h at the velocity of 1.320 µg/GSH*U*h (Table 3) and, consequently, the highest reduction (50%) (Fig. 1), which confirms that compound oxidoreduction may be related to the potential that yeast cells have to produce specific molecules (GSH and PO) to aim at maintaining metabolic activity of the yeast (Lash, 2005).
These data are relevant for the possibility of observing and describing the yeast activity as biological process to the contaminant reduction, as well as to strengthen the wine sector, highlighting the consumption of fermented products when compared to in natura. The union of Saccharomyces and non-Saccharomyces yeast strains applied in this study and the industry is also an advantage, as wineries could adopt this practice in the winemaking process categorizing the process as a biological decontamination agent and safe products for consumption.
Another important fact that guides this study is that we demonstrated the real percentage of contaminant reduction, without the influence of oenological inputs used industrially. In addition, the 2,4-D herbicide was studied in alcoholic fermentation, highlighting the importance of this study, as it is a pesticide not recommended for viticulture, but which comes from the cross contamination of agricultural crops  close to the vineyards. To date, our study is the first to evaluate the effect of the 2,4-D herbicide on process quality parameters. In this sense, studies with this focus should be investigated. The highest decrease in OTA and pesticides took place at the end of alcoholic fermentation (168 h) (Fig. 1). Besides being related to GSH and PO activity, decrease may also be related to the adsorptive capacity of S. cerevisiae (Meca et al., 2010). The adsorption process on the yeast wall is related to mannoproteins which are capable of bounding to pesticides and mycotoxins (Čuš et al., 2010b;Freire et al., 2020;Meca et al., 2010). Some studies suggest that the adsorption process is easily observed when mannoproteins are released in the first week after the end of alcoholic fermentation, mainly when yeast lees are homogenized in wine at the end of the fermentation process (battonage, the stirring technique), which ends up becoming a measurer of the potential yeasts have to mitigate contaminants (Čuš et al., 2010b;González-Rodríguez et al., 2009).
Efficacy of detoxification of different toxic compounds, such as pesticides and mycotoxins by yeast activity, depends on several factors, such as the type of strain, and concentration and incubation time of compounds, throughout the process (Yousefi et al., 2021). High doses of toxic compounds exert negative influence on both S. cerevisiae growth and alcoholic fermentation (Li et al., 2012;Scariot et al., 2022).
To better understand the relation between decrease in contaminants in alcoholic fermentation, the PCA was carried out (Fig. 2) to focus on two regions. The first component (PC 1) explains 49.5% of total variance. Therefore, about 50% of information found in the five variables of the database may be encompassed by this component. The second component (PC 2) explains 33.5% of total variance. Thus, 83% of data variance is explained by only two components.
The region delimited by a circle accounts for variables VDE GSH*PO and RSD of contaminants. The Pearson correlation shows positive and significant relation (R = 0.66, p = 3.88 × 10 −4 ) between both variables. It means that, after 24 h of fermentation, S. cerevisiae activates metabolic pathways to decrease toxicity in the culture medium, which is mainly related to increase in VDE GSH*PO and RSD of contaminants from this period of fermentation (Table 3). It highlights the fact that alcoholic fermentation mitigates contamination caused by pesticides and mycotoxins in treatments.
The region represented by a square accounts for variables GSH and PO. Positive and significant correlation (R = 0.78, p = 5.99 × 10 −6 ) was also found between both variables. It shows that treatments 2 and 3 exposed to OTA exhibited high activation of those molecules (Table 1). It may be explained by OTA toxicity, which makes yeast cells produce defense systems to keep their metabolic activity.
Positive and significant correlation between reduction of contaminant and PO activity (R = 0.65, p = 0.0005), asterisked in the graph, should also be highlighted. It shows that, the lower the compound toxicity, the more it decreases, i.e., procymidone, the compound that has the lowest toxicity in this study, exhibited the highest decrease (65%), by comparison with the other contaminants ( Fig. 1), mainly related to PO activity. On the other hand, 2,4-D, which was classified into group 2B as likely to be carcinogenic to humans (IARC, 2018), exhibited the lowest decrease (22%). It is the only contaminant that is not related to the variables shown by the PCA. This fact leads to the need for further studies to investigate other relations between enzyme pathways and 2,4-D degradation. Its mitigation by decreasing adsorption though cell walls of yeasts-unlike the other contaminants-may take place but other metabolic processes in their cells should be investigated so as to mitigate human exposure to these contaminants which are often found in grapes and wine.

Conclusion
Both pesticides and the mycotoxin influenced biochemical characteristics in alcoholic fermentation. Contamination affected cell viability (the lowest ranged from 4 to 18%), which may be related to the energy spent by the yeast to activate metabolic pathways in detoxification. Decrease in OTA, 2,4-D, and procymidone took place after 24 h of fermentation and the highest index was found at the end of the alcoholic fermentation (168 h). Decrease in procymidone and OTA due to metabolic processes in yeast cells is mainly related to PO activity and GSH production.