Corot Noir lees (above) were collected after the completion of malolactic fermentation. The lees were freeze dried and analyzed for condensed tannin.

Rich. Full-bodied. Intense, even dark and brooding. All are wine descriptions to make a red wine producer swoon. Unfortunately, winemakers struggling with tough growing years or hybrid grapes are more likely to hear descriptors like thin, light and weak palate—hardly traits to make consumers sit up and take notice. The problem? The quality and concentration of phenolic compounds (the pigments and polymers that impact the color, texture and structure of red wines). Whether caused by cultivar or climate, the challenges of low color, poor color stability and inadequate tannins are top concerns for the production of high-quality red wines.

Color
Concerns about color generally fall into one of three categories. Consumers and critics are harder on wines with faults that are visually obvious: namely, wines with low color intensity or hues that don’t seem “typical” for a particular red varietal. Consumers are arguably less cognizant of the issues of color stability, but wine producers know that the darkest youthful blush can fade to a sickly orange in maturity if strong color complexes aren’t formed early. “Good” wine color is nothing to blink at; though wine pigments contribute little taste or smell, consumer satisfaction has been positively correlated to color in a variety of products including red wines. Darker color enhances the perception of fruit and flavor intensity, and what red wine fan doesn’t look for intensity and rich fruit?

Anthocyanins, the pigmented phenolics responsible for wine color, are located in the vacuoles of hypodermal cells, right under the grape skin. To move into the juice, anthocyanins have to first exit the vacuoles and then the cells themselves—a process made easier by physical cell damage. Subsequently, anthocyanin extraction occurs more readily in mature grapes with softened fruit tissue. During winemaking, traditional belief dictates that extraction is enhanced late in fermentation by higher alcohol content, but the bulk of evidence indicates that peak anthocyanin extraction actually occurs around the fourth day of fermentation, and slows to nothing after about the 10th day.

Moving anthocyanins into must is just the beginning of the story, however. In their monomeric (non-bonded) state, individual anthocyanins are unstable at wine pH, which pushes them to change into colorless forms. Lasting pigmentation is achieved when anthocyanins bind with other compounds to form stable complexes. While scientists have hardly scratched the surface of the complex chemical reactions that produce wine color, two phenomena have been well described: co-pigmentation and polymeric pigment formation.

Young color
Copigmentation, which is most important in the visual appearance of young red wines, occurs when anthocyanins bind non-covalently with other phenolic cofactors. This binding stabilizes the anthocyanin in a colored form, changing light absorption so that color deepens and moves from red toward blue. In V. vinifera cultivars, the cofactor to anthocyanin ratio is generally around 0.05:2. Copigment formation appears to be limited by cofactor availability, extraction rate and possible sorption onto other grape or yeast components.

Studies of copigmentation have largely focused on reactions involving malvidin-3,5-glucoside, the predominant anthocyanin in V. vinifera. In hybrid cultivars, where non-malvidin anthocyanins and diglucosides are more common, the color shift and enhancement that comes with copigmentation is less pronounced. This is one reason that hybrid red wines are described as having “non-vinifera” color. It’s also likely that hybrid red wine color is affected by the type and amount of acylated non-malvidin anthocyanins, i.e., monomers of perlargonidin, cyanidin, peonidin, delphinidin or petunidin bound to cinnamic or aliphatic acids. These acylated anthocyanins appear in colors ranging from orange to pink and blue, and they may result in “untypical” red wine color.

At the moment, there’s more empirical data about production methods that may increase copigmentation than solid understanding of the science that makes it happen. The traditional practice of cofermenting low- and high-cofactor cultivars together (such as Trebbiano and Malvasia with Sangiovese) to enhance color is well documented, but the degree of copigmentation achieved varies by cultivar, region and year. Both high alcohol content (approaching 16%) and warmer fermentation temperatures have been found to disrupt copigmentation formation. In the latter case, extraction of both anthocyanins and cofactors increase, but reaction thermodynamics limits their bonding capability.

Long-term color
The second type of stable bound anthocyanin compounds is called polymeric pigments, but the moniker is deceiving, as not all “polymeric pigments” are polymers, or pigmented. The name was originally applied to compounds that were observed to resist sulfite bleaching, and it has become a sort of blanket term for any complex with anthocyanins bound covalently with other wine compounds. Polymeric pigment formation shifts color toward brick red and provides the primary source of color in wines upwards of two years old.

To date, dozens of polymeric pigment types have been identified. Common adducts (or compounds formed from the direct addition of two other compounds) include anthocyanin-acetaldehyde; portosin, which contributes a blue color; vitisins, which are orange; and pinotin, which enhances red color. A variety of tannin-anthocyanin and anthocyanin-tannin compounds also may form. It is likely that tannin concentration early in fermentation impacts polymeric pigment (but not copigment) formation.

As with copigmentation, much of the science of polymeric pigment formation is still unclear. Concentrations are known to correlate positively with ethanol, but beyond that scientists can only speculate that any processing methods that increase the availability of oxygen and tannins required for hypothesized formation reactions will increase polymeric pigment content. These include micro-oxidation at the middle or end of alcoholic fermentation, tannin additions early in fermentation, enzyme additions and saignée. Hybrid cultivars, which have naturally low condensed tannin concentrations, will necessarily be limited in their potential for pigment formation.

Wine color chemistry can be investigated two ways: by delineating individual chemical reactions (like anthocyanin-acetaldehyde binding) and tracking them throughout the winemaking process; or holistically, by examining all the pigments in a finished wine and working backward to determine processing impacts on formation. In the latter case, a number of studies have yielded conflicting results about the impact of such methods as extended maceration and cold soak (which seem to have no impact), and freezing and thermovinification (which may have some) on color enhancement. In short, given the variety of possible pigment complexes (many of which are still unknown), it’s likely that the mechanisms and compounds that provide “good” and lasting red wine color vary by cultivar, region and processing method.

Tannins
If color chemistry isn’t confusing enough, tannin reactions and analysis are even more fun. Like “polymeric pigment,” the term “tannin” is somewhat confusing. The broadest definition of tannins includes any plant polyphenolic capable of cross-linking collagen fibers in animal hides, a categorization that is purely functional, and not altogether helpful from a winemaking standpoint. It does point to one important fact: Tannins are capable of precipitating proteins, providing the astringency and mouthfeel associated with high-quality red wines.

From a winemaker’s point of view, there are two types of tannin relevant to winemaking: condensed and hydrolysable. Condensed tannins are ogliomers or polymers (short or long chains) of flavan-3-ols, a family of small phenolic compounds including (+) catechin, (-) epicatechin, gallocatechin and gallocatechin gallate, the four most commonly found in tannin chains. These tannins originate in skins and seeds—and while seeds have a higher tannin concentration, they are less extractable, so skin tannin predominates in most wines. Concentration of condensed tannins in the grape varies by cultivar, site and viticultural practices.

Condensed tannins take many forms; variables include the chain length (called degree of polymerization), the compounds that initiate and elongate the tannin chains and the frequency and linkage type of any branches. At present, scientists are unable to isolate individual condensed tannin compounds for identification and quantification. Instead, condensed tannins must be measured indirectly by chemically breaking apart the chains and looking at the pieces—a bit like trying to figure out what a jigsaw puzzle looks like by examining each puzzle piece individually. For this reason, tannin content in grapes and wine is generally reported as overall concentration and mean degree of polymerization (mDP), since it’s impossible to quantify individual tannin compounds.

The second type, called hydrolysable tannins, is wood-derived and historically found only in barrel-aged wines. Grape and wood-derived tannin types aren’t interchangeable; condensed tannins bind with proteins at a rate of about 20:1 in reactions independent of temperature and ethanol content, and have the potential to bind with anthocyanins in color-stabilizing reactions. In contrast, hydrolysable tannins bind with proteins at a rate of around 40:1 in reactions that increase with temperature and decrease as ethanol content increases, and are not known to participate in polymeric pigment reactions. In short, condensed tannins should aid in color stabilization and provide a higher perception of astringency, while hydrolyzed tannins will be less likely to stabilize color but may promote fuller mid-palate and a softer mouthfeel due to lower, slower protein interaction.

Starting at véraison, tannin length (mDP) increases and perception of bitterness decreases as grapes mature, but extractability decreases, making optimal harvest time a balance between tannin quality and quantity. Extraction during maceration and fermentation is another balancing act, as skin tannins are extracted rapidly during early alcoholic fermentation but eventually plateau, while seed tannin is extracted slowly and consistently throughout. Since a higher proportion of skin-derived tannins has been found to enhance perception of red wine quality, winemakers must optimize both quantity and proportion of skin and seed tannins extracted. Additional studies suggest that warmer fermentation and saignée favor overall extraction, and that the latter results in higher average mDP. Both punch down and pump over enhance tannin extraction, but the degree and type seem to vary by cultivar and method.

Tannin woes
Low tannin concentration and poor tannin quality are common complaints of red wine producers, and at least one study suggests that this emphasis on polyphenolics is well founded, since as much as 70% of total variance among red wines as been attributed to tannin astringency. Unfortunately, recent research suggests that both of these parameters are more complicated than traditionally thought, and means of both assessing and influencing tannin quantity and quality are anything but straightforward.

While it would seem that final wine tannin concentration equals whatever percentage of tannin can be extracted from the grape, a number of obstacles prevent a direct translation from grape to wine. A recent survey comparing wine tannin to tannin per berry in a range of V. vinifera and hybrids grown in the Finger Lakes growing region of New York found no correlation between the two. Hybrid grapes usually have lower tannin concentrations than vinifera grapes, but variation among wines is much greater than that found among berries. To understand potential wine tannin concentration, it’s helpful to think in terms of tannin extractability, defined as [(tannin in wine)/(tannin in grape)] X 100. A comparison of hybrid and vinifera winegrape cultivars has shown that tannin extractability is generally lower in hybrid cultivars, regardless of the tannin concentration found in the fruit.

This disparity between grape tannin content and tannin extractability means that tannins in hybrids are either harder to extract from the fruit, or that some component in the must is acting as a sponge to remove tannins post-release. Grape polysaccharides and proteins are known to bind a small amount of tannin during fermentation, but recent work at Cornell (Springer and Sacks, 2014) demonstrated that hybrids have higher binding capacity due to higher concentrations of certain proteins. These proteins have tentatively been identified as factors in disease resistance in grapes, which may mean that hybrid cultivars with greater disease resistance also have the highest tannin-binding capacity. If true, this poses a challenge for producers hoping for full-bodied hybrid red wines.

On the surface, the solution to low-tannin reds is simple: The broad availability of exogenous tannin additives means that winemakers can adjust tannins at will. Exogenous tannins can be roughly divided into three categories: Grape-derived condensed tannins, plant (but non-grape) derived condensed or hydrolysable tannins, and any mixture of the two. In addition to grape material, exogenous tannins are commonly extracted from oak, acacia or from the South American Quebracho tree. Producing high-purity tannin extracts is very difficult, however, so exogenous tannin products range from 10%-45% tannin, with the remainder consisting of inseparable non-tannin phenols, drying and solubility agents and non-phenolic plant material. The obvious practical implications of these formulations are a) It’s nearly impossible to know how much active tannin is available in a given tannin product (and concentration may vary from batch to batch in the same product), and b) products may have unknown activity or sensory characteristics due to non-tannin content.

In addition to the general problem of variable product content, tannin additions are especially problematic in hybrid wines, where high tannin-binding capacity may render additions moot. Manufacturer recommendations developed for V. vinifera cultivars are likely to be insufficient to impact sensory characteristics of high-binding hybrid cultivars. Ongoing research at the Cornell Enology Extension lab suggests final tannin concentration is significantly impacted by cultivar and timing of tannin additions, and addition rates must be much higher than manufacturer recommendations. The sensory impact of additions as high as three times the recommended rate is also under investigation, as off-odors or flavors may outweigh the potential benefits to overall wine sensory quality.

What is quality?
In addition to overall low tannin concentration, red wine is often threatened by sensory attributes deemed to be signs of “poor quality” tannins. Though critics often toss out terms like “hard,” “soft” or “green” tannins, sensory science has found it almost impossible to define these terms in meaningful ways. A few sensory characteristics can be tentatively assigned causes; tannins with mDP < 5 are thought to enhance bitterness, for example, while mDP > 5 provides astringency. “Hard” tannins may be the result of high tannin concentration and a greater percentage of seed tannin in the mix, but “soft” tannins are largely in the mouth of the beholder (that is, no one can agree on what that means, so tying it to a chemical phenomenon is impossible).

General matrix effects are also at work, as ethanol, acid and sugar content all impact a taster’s perception of tannin astringency and bitterness, and aroma and visual qualities are known to impact gustatory effects in a variety of ways. Most confusing is the fact that each person’s perception of wine “texture” is individual and impacted greatly by variations in the quantity and quality of saliva. All told, developing a predictable standardized lexicon of tannin sensory effects is difficult because wines and wine tasters vary so widely, and without a standard lexicon, tying individual sensory attributes to chemical or physical phenomena is well nigh impossible.

At the end of the day, chemistry and empirical observation leaves us with a few truths and a lot of questions:
1. Increased skin contact time will increase tannin concentration but not anthocyanin concentration.
2. Color stability is dictated by copigmentation and “polymeric pigment” formation.
3. Interactions governing hybrid red wine color are largely unknown, due to variety and type of anthocyanin concentration.
4. Perception of tannin quality is complicated by matrix and cross-modal sensory interactions.
5. Many color and tannin effects are cultivar dependent.
6. Optimal production practices may be specific to site, season and even winery.

In short, phenolic chemistry is very complex, making characterization of wine pigments and tannins a major challenge and linking sensory impacts to individual compounds (or winery processes) equally difficult. Until enologists have all the answers, a good palate and a willingness to experiment will remain important tools in quality red wine production. WE

Anna Katharine Mansfield is an assistant professor of enology at Cornell’s NYSAES in Geneva, N.Y. Her research focuses on hybrid wine phenolics and fermentation nutrition

References
Boulton, R. 2001. “The Copigmentation of Anthocyanins and Its Role in the Color of Red Wine: A Critical Review.” American Journal of Enology and Viticulture, 52(2):67-87.

Dooley, L.M.; R.T. Threlfall; J.F. Meullenet, and L.R. Howard. 2012. “Compositional and Sensory Impacts From Blending Red Wine Varietals.” American Journal of Enology and Viticulture, 63(2):241-250.

Harbertson, J. 2010. “The Role and Use of Tannins During Winemaking.” Presentation at Viticulture 2010, Rochester, N.Y., February 2010.

He, F.; N. Liang; L. Mu; Q. Pan; J. Wang; M. Reeves, and C. Duan. 2012. “Anthocyanins and Their Variation in Red Wines, I. Monomeric Anthocyanins and Their Color Expression.” Molecules, 17:1571-1601.

Hernández-Hierro, J.M.; N. Quijada-Morín; L. Martínez-Lapuente; Z. Guadalupe; B. Ayestarán; J.C. Rivas-Gonzalo, and M.T. Escribano-Bailón. 2014. “Relationship Between Skin Cell Wall Composition and Anthocyanin Extractability of Vitis vinifera L. cv. Tempranillo at Different Grape Ripeness Degree.” Food Chemistry, 146:41-47.

Kennedy, J.A. 2008. “Grape and Wine Phenolics: Observations and Recent Findings.” Ciencia e Investigación Agraria, 32(5)107-120.

Manns, D.C.; C.T. Coquard Lenerz, and A.K. Mansfield. 2013. “Impact of Processing Parameters on the Phenolic Profile of Wines Produced From Hybrid Red Grapes Maréchal Foch, Corot Noir and Marquette.” Journal of Food Science, 78:C696−C702.

Sacci, K.L.; L.F. Bisson, and D.O. Adams. 2005. “A Review of Winemaking Techniques on Phenolic Extraction in Red Wines.” American Journal of Enology and Viticulture, 56(3):197-206.

Springer, L.F., and G.L. Sacks. 2014. “Protein-Precipitable Tannin in Wines from Vitis vinifera and Interspecific Hybrid Grapes (Vitis ssp.): Differences in Concentration, Extractability, and Cell Wall Binding.” Journal of Agricultural and Food Chemistry, 62(30):7515-7523.