By Christine Pascal¹

The scientific literature is full of articles on wine oxidation mechanisms, their antioxidant composition, even their redox potential …

The appearance of oxidation or reduction signs in wines is a central issue of winemaking work: one of the major challenges for the winemaker is to assess the capacity of a wine to withstand the appearance of these signs and adapt the winemaking decision to its intended profile and shelf-life.

However, making the link between what is known in the literature and enological practice is not easy. The terms used sometimes do not have the same meaning between scientific literature and practical vocabulary, which can confuse thinking:

• Are polyphenols, ascorbic acid, sulfites antioxidants or pro-oxidants?
• What is the redox potential of a wine?
• Does saying that a wine is sensitive to oxygen mean that oxidation signs will quickly appear if exposed to oxygen or on the contrary, that it will develop a capacity to resist oxidation?
• Does the high antioxidant content in a wine mean that the wine will resist oxidation better?

In a series of short chapters, we propose to try to reconcile academic knowledge and empiricism on concepts such as redox potential, oxidation, antioxidant content or wine oxidability. We will then try to make a link with the usual enological practices and their consequences on the wine shelf-life.

Chapter 1: Signs of wine oxidation as described in practice

To begin on this topic about wine oxidation and wine oxidability, we need to recall what winemakers understand by oxidation. The signs of oxidation in wines are in practice described as:

• an evolution of the color with appearance of orange tints
• an aromatic evolution described as a loss of varietal notes (boxwood, passion fruit) and / or an appearance of so-called oxidative notes (honey, wax, chocolate, stone, dry fruit, dry grass, nuts, etc.)
• sometimes an evolution of the taste balance with increased astringency and / or loss of wine body.

These different oxidation signs can appear successively or simultaneously without the appearance of one of these signs indicates the possible appearance of another.

The moment from which a wine is considered oxidized is related to the acceptability of these characteristics in relation to the product type and the winemaker intention. For example, it is acceptable or even desirable to have notes of dry fruit on a sweet wines while it is not on a dry Sauvignon Blanc wines from certain appellations which must keep citrus or exotic fruits notes.

The concept of premature oxidation, widely used today, is the very proof of these questions of acceptability. The term “premature” is likely related to the appearance of oxidation notes earlier than expected.

In the end, the notion of acceptability of the presence of oxidative notes and the multiple nuances that they can cover make the analytical determination of the level of wine oxidation, as described by winemakers, even more difficult.

To remember:

• There are many wine oxidation signs in practice (color, aroma, taste balance).
• The moment from which a wine is considered oxidized depends on acceptability criteria linked to the winemaker intention and the wine type.

Chapter 2: The Study of Oxidation Mechanisms in the Scientific Literature

Let’s now try to summarize what the literature tells us about oxidation. We have chosen to stay at a simple level of description of current knowledge and extract the main messages.

Academic research has demonstrated over the past fifty years probable mechanisms of oxidation and molecules formed or transformed during these mechanisms. The difficulty of this research theme lies in the diversity of the molecules present in the wine. The work therefore consisted of studying oxidation mechanisms:

• of simple molecules, easily accessible in commerce or purifiable from grapes or wine (monomeric polyphenols such as caftaric acid, catechin / epicatechin, malvidin-3-O-glucoside, varietal thiols such as 3-mercaptohexanol, etc.)
• in media modeling the must or wine matrix (acidic pH of around 3-3.5, 0 or 12 % ethanol to approximate the conditions of a must or wine, possible presence of tartaric acid …)
• in the possible presence of molecules described as antioxidants (sulphites, glutathione or ascorbic acid).

The mechanisms highlighted are often hypothetical, quite rarely formally demonstrated due to the strong reactivity and the very short “shelf-life” induced of the molecules involved in these reactions. In addition, the behavior of more complex molecules such as oligomers or products of oxidation reactions are completely excluded from these mechanistic models.

It is generally accepted that the oxidation mechanism of wines begins with the oxidation of certain polyphenols, the orthodiphenols, in the presence of Fe³⁺ ions. In a mechanism comprising a good number of similarities, in musts, it is enzymes whose active site is a Cu²⁺ ion which catalyze this first step.

It is at this point of the mechanism that everything gets complicated because the two products of this polyphenol oxidation reaction are unstable / reactive. And they are the ones who will lead to the appearance of practical oxidation signs by creating chain reactions:

• The oxidized forms of polyphenols, quinones, unstable, have a strong reactivity for nucleophiles such as thiols, other polyphenols …
• Hydrogen peroxide, in the presence of Fe²⁺, initiates the Fenton reaction (Singleton 1987) and gives rise to very high energy radicals, the hydroxyl radicals HO°. These are known to oxidize ethanol to ethanal or tartaric acid to glyoxylic acid, but they can undoubtedly attack a large number of other compounds.

These reactions are in fact the reasons for the appearance of practical oxidation signs and it is this link that we will try to describe in the next chapter.

To remember:

• The oxidation mechanisms have been widely studied by scientists and the commonly accepted hypothesis is that polyphenols are the molecules that oxidize first.
• The products resulting from this oxidation are unstable and lead, through chain reactions, to the appearance of the oxidation signs observed in practice.

Chapter 3: Birth of molecules contributing to oxidation signs: the “double” role of polyphenols

In the previous chapter, we recalled that, according to the literature, the first molecules oxidizing in wine are polyphenols. The products of this oxidation reaction, quinones and hydrogen peroxide H₂O₂, are unstable and lead by chain reactions to the appearance of signs of oxidation observed in practice.

So let’s go back to these chain reactions and the effects they produce, always taking up the essence of what the literature teaches us.

The literature describes the formation or disappearance of molecules that could explain the oxidation manifestations described in practice from reactions of quinones, or H₂O₂ via the Fenton reaction. Without being exhaustive, let us cite 3 examples:

• The appearance of brown/yellow/orange pigments linked to condensation reactions of flavan-3-ols (tannins), particularly in the presence of ethanal or pyruvic acid (Oszmianski 1996, Fulcrand 1996, Guyot 1996) produced by Fenton reaction.
• The loss of varietal thiols, in particular 3MH, 4MMP due to their reaction with quinones. (Waterhouse and Laurie 2006, Nikolantonaki, 2010).
• The appearance of so-called oxidation aromas (methional, phenylacetaldehyde) probably due to the attack of hydroxyl radicals, produced by Fenton reaction on the corresponding alcohols (Nikolantonaki and Waterhouse 2012).

While the first step in the reaction mechanism is oxidation of polyphenols, it is followed by different types of chemical reactions, some of which (nucleophilic additions for example) are not redox reactions. The term “oxidation” mechanism is therefore simplifying to describe all of the chemical reactions that take place in wine following the oxidation of polyphenols. It follows that:

• Reasoning only on the “oxidability” of compounds (“their redox potential”) does not explain the practical oxidation signs. Reasoning like this, must be limited to mechanistic studies to understand the formation of the first quinones and the coupled oxidations (Danilewicz 2012). But then come the reactions of nucleophilic additions which depart from the concept of redox potential and which generate the evolutionary compounds identified empirically. We will come back to the use of redox potential in enology in a future chapter.
• The reaction chains allow most of the polyphenols to be involved in these reactions, whether they contain an orthodiphenol function or not. We will talk about this again in a future chapter because this implies that in practice, when reasoning about the risk of oxidation of a wine, the overall concentration of polyphenols must be considered.

Finally, in this context, using the word “antioxidant” to refer to polyphenols seems somewhat misleading for winemakers. In chemistry, an antioxidant is a molecule that slows down or prevents an oxidation reaction from occurring. Polyphenols meet the chemical definition of the term because by oxidizing first, they prevent the direct oxidation of other compounds. However, their oxidation does not stop the wine evolution described as oxidative in practice. This is why they are sometimes described as pro-oxidants because they cause the appearance of so-called oxidation signs. We therefore prefer to avoid the use of these 2 confusing terms. We simply refer to polyphenols as oxidizable molecules.

To remember:

• The chain reactions following the polyphenols oxidation in wine are not all oxidation reactions but give rise to what is described in practice as manifestations of “oxidation”.
• Using the word “antioxidant” to designate polyphenols seems somewhat misleading to winemakers because their oxidation does not prevent the appearance of changes described as oxidative in practice. On the contrary, it is the origin.
• We therefore prefer to designate polyphenols as oxidizable molecules.

Chapter 4 The Concept of Antioxidant

We have previously shown that the use of the term antioxidant to designate polyphenols is misleading in enology because their oxidation does not prevent the appearance of evolution signs described in practice. On the contrary, it is at the origin.

It is interesting here to focus on the exogenous antioxidants used in enology.

The SO₂

Let’s start with the best known, the most used and perhaps the most decried: SO₂. It has the most comprehensive action to limit the mechanisms described above:

• It limits the formation of quinones enzymatically by denaturing oxidases (polyphenoloxidase and laccase to a lesser extent) when it is present in sufficient doses
• When quinones are formed, it can react with these quinones to neutralize them but also with hydrogen peroxide, thus limiting Fenton reaction.

Thanks to its action on both products of the polyphenol oxidation reaction, SO₂ is the most effective antioxidant to fight against the wine oxidation.

Ascorbic acid

On the contrary, the beneficial effect of ascorbic acid on the wine preservation against oxidation is more questionable. This compound has an ability to oxidize in place of polyphenols (Danilewicz 2012). However the product of oxidation is unstable and leads to the formation of brown pigments, just as troublesome as those related to the reaction of flavan-3-ols (link previous chapter). In addition, hydrogen peroxide is also formed when it oxidizes, opening the way for the Fenton reaction (Barril et al. 2009, Bradshaw et al. 2011). Ascorbic acid therefore limits the formation of quinones but leads to the formation of yellow pigments and do not limit Fenton reaction. Its behaviour is therefore relatively close to the polyphenols one. This is why the combined addition of SO₂ and ascorbic acid is recommended in practice.

Glutathione

Glutathione has the ability to react with quinones to neutralize them (this is how the Grape Reaction Product is formed- Cheynier 1989). It could also oxidize directly. Its oxidation product has not been shown in wines, except in its sulphonated form probably produced by reaction with SO₂ (Arapitsas 2016). In practice, this shows that the addition of 2 antioxidants does not necessarily lead to double protection against oxidation! Furthermore, compared to SO₂, it has no action on the enzymes forming the quinones and its behavior towards the Fenton reaction products is not documented to our knowledge. However, the literature does not yet provide data indicating that glutathione leads to the formation of compounds that can be linked to practical manifestations of oxidation (yellow/orange coloration, decrease in varietal aromas, appearance of oxidation aromas). Its retarding effect on the appearance of oxidation marks has been widely described in the literature (Nikolantonaki 2018, Kritzinger 2013, Lavigne and Dubourdieu, 2002; Roussis et al., 2007; Ugliano et al., 2011). It would constitute an alternative to the antioxidant action of SO₂, even if its mechanism of action is less well documented. From a practical point of view, glutathione is present in grapes at varying levels, it is also produced during alcoholic fermentation or extracted from lees. However, its analysis is not easy. The addition of pure glutathione is not currently allowed in enology by the OIV. It can be brought by adding enological products said to be rich in glutathione. Knowledge of their glutathione richness and the form of glutathione they contain (reduced or oxidized) would be a plus (Kritzinger 2013).

All the compounds designated as “antioxidants” present in wine or used as such in enology do not necessarily have the same impact on the appearance of practical oxidation signs (color evolution, varietal aromas decrease, evolution aromas appearance). In the next chapter, we will ask ourselves whether measuring the “antioxidant” content provides information on the potential wine shelf-life, or even its resistance to oxidation.

To remember:

• Among the exogenous antioxidants used in enology, SO₂ is the one with the most complete mechanism of action: it can prevent the formation of quinones enzymatically and neutralize the oxidation products of polyphenols, thus reducing chain reactions that follow.
• Ascorbic acid behaves similar to polyphenols with regard to oxidation. Its oxidation causes the formation of brown pigments without stopping the chain reactions.
• Glutathione reduces chain reactions by neutralizing quinones. It therefore appears to be a potential alternative to SO₂, even if its mechanism of action still needs to be documented.

Chapter 5: Does the “antioxidant” content tell us about the wine shelf-life?

We have reviewed the action of endogenous or exogenous “antioxidants” (added to wines) such as polyphenols, SO₂, ascorbic acid and glutathione to finally conclude that the use of the term antioxidant to describe their action is ambiguous for the winemaker.

The antioxidant content can be evaluated by different methods consisting globally in assessing the ability of a wine to trap radicals, to react with different types of oxidants (DPPH test, ABTS, FRAP. and even analysis of Folin Ciocalteu), or to be oxidized on electrodes of a given nature (voltametry). The antioxidant content of a wine is strongly correlated with the concentration of polyphenols (Büyüktuncel et al. 2014 ; Romanet et al. 2019 ; de Beer et al. 2004, 2006 ; Kilmartin 2001).

Other molecules found in wine such as glutathione, cysteine or methanethiol show an antioxidant capacity (Romanet et al. 2019) at a similar level as polyphenols such as catechin or caffeic acid. However, their concentration in wine is in the order of ten mg/L compared to several hundred mg/L for polyphenols. It is therefore logically polyphenols that contribute mainly to the concentration of antioxidants.

Several studies have shown that the concentration of antioxidants decreases during the oxidation of wines (Rodrigues et al. 2007, Ugliano 2013 and 2019). Overall, older wines contain fewer antioxidants than young wines (Rodrigues et al. 2007) but the variability of vintages and winemaking processes (including polyphenol extraction stages) does not allow determining wine age based on its antioxidant concentration (Romanet et al. 2019). Finally, unsurprisingly, the low antioxidant content of a wine has been correlated with the presence in sensory profiles of notes associated with an oxidative evolution of wines (Romanet et al. 2019).

What these results show is that the content of oxidable molecules in a wine, to use the vocabulary we have chosen to emphasize, decreases over time, meaning with the wine oxidation along winemaking and aging. The lower their concentration, the greater the risk of oxidation notes.

Do measurement of “antioxidant” content allow us to assess the potential wine shelf-life, or even its resistance to oxidation?

It is tempting to think that the more antioxidants a wine contains, the more resistant it is to oxidation. But to our knowledge the link between this content of oxidizable molecules and the time or amount of oxygen consumed before oxidation notes appear has not been established.

By checking the data from the scientific literature, it seems that the antioxidant content is rather the sum of the polyphenol content of grapes (modulated by all stages of winemaking, aging, storage) and endogenous or added antioxidants during the process, present in smaller quantities than polyphenols.

The methods for measuring antioxidants without separating the different classes of molecules that protect against oxidation (SO₂, glutathione) or contribute to the appearance of practical oxidation signs (polyphenols), do not provide reliable information on the ability of a wine to age! To make a connection with winemaking practices, if the intrinsic content of polyphenols (majority contributors to antioxidant content) was an image of a wine’s ability to withstand the oxidative notes appearance , winemakers would use the Folin index or even the Total Polyphenols Index for a long time for this purpose.

We know that the profession is always looking for a test to predict resistance to oxidation. Reasoning by the absurd, one could even say that the wines made with Tannat grape variety from the Southwest (so dear to my heart!), rich in polyphenols, would systematically have a higher aging capacity than Burgundian Pinots Noirs, less concentrated in polyphenols

In the next chapters, we will explain when during the wine-making process and for what purpose the content of oxidizable molecules can be used to guide winemaking choices. We will also look at the methods of measuring these contents.

To remember: The antioxidant content of wines, regardless of the measurement method used, is strongly influenced by the polyphenol content. It therefore does not provide direct information on the resistance of a wine to oxidation or its ability to age.

¹Vinventions, Enology Team, 7 Avenue Yves Cazeaux, 30230 Rodilhan.