18
Taints, Off‐flavors, and Mycotoxins

18.1 Introduction

Undesirable flavors and aromas are a major reason for consumer rejection of wines and other foodstuffs. Many, if not all, flavor compounds can contribute either positively or negatively, with a given compound being indicative of inferior product quality or becoming disagreeable when the concentration exceeds norms for a particular product. ¹ Typically, these flavors are classified as either:

• Taints, when the responsible flavor compound is from external contamination, that is, the environment. Taints frequently arise from packaging contamination, but also from equipment, additives, and processing aids, air (or other gases), or microbial activity external to the product.
• Off‐flavors, when the flavor is from biochemical or non‐enzymatic changes to compounds in the original product. ² Off‐flavors are typically encountered due to inappropriate handling and storage (e.g., light, oxidation, temperature, microbial load, processing operations) and are perhaps more easily avoided than the often accidental contamination leading to taints.

This chapter will primarily focus on taint compounds arising from external (exogenous) contamination. Taint compounds are generally more potent than off‐flavors, but either defect often results in a decrease in desirable product attributes rather than the creation of a distinct character from the offending compound(s). Ridgway et al. [1] review common food “tainting” compounds and include their origins, descriptors, and thresholds, whereas Boutou and Chatonnet [2] provide data specific to wine. Off‐flavors from fermentative sulfur compounds, oxidation, and related compounds ordinarily formed in wine production are described in more detail elsewhere in this book (and in the literature, e.g., see Reference [3]), but are compiled in a cross‐referenced table at the end of this chapter (see Table 18.2), for convenience.

The majority of compounds responsible for wine taints have been identified in other foodstuffs. In spite of extensive research, however, it is not always easy to determine the source/s of taints, and compounds presented in this chapter have been categorized based on their suspected or known origins. Compounds causing aroma and flavor defects can possess a wide range of functional groups, but often contain common structural motifs known for low sensory detection thresholds (e.g., sulfur, nitrogen, or halogen atoms). Infrequent taints arising from pesticides, hydraulic oils and other hydrocarbons, paints, resins and plasticizers, metals, and brine used for tank cooling, or deliberate adulterations to manipulate sensory attributes, are not discussed here (for more information see, for example, References [3] to [6]). While this chapter primarily considers compounds that affect organoleptic qualities, some contaminants with negative health effects, for example, mycotoxins from diseased grapes, will also be discussed.

18.2 Common wine taints

18.2.1 Taints from cork or wood

“Cork taint” is the commonest and probably most well‐known taint encountered in wine and consequently the taint of interest when evaluating a wine in a restaurant. The major cause is the potent and earthy/musty/moldy‐smelling 2,4,6‐trichloroanisole (TCA, Table 18.1). In addition to having an undesirable aroma, TCA can inhibit olfactory signal transduction, thereby interfering with the perception of other odorants [14]. The requirements for formation of TCA are:

• 2,4,6‐Trichlorophenol (TCP), whose origin is described in more detail below.
• Microorganisms capable of O‐methylating TCP to produce TCA.

Although cork closures are an obvious (and eponymous) source of TCA, all wood‐containing materials are potential sources, including oak barrels (Figure 18.1) or processing aids/equipment, cardboard packaging materials, or even the winery itself (e.g., wooden structures). Related haloanisoles, such as di, tetra, and pentachloroanisoles or 2,4,6‐tribromoanisole (TBA), arising from their respective halophenols in the same manner as TCA, can also contribute to cork taint. Several non‐haloanisole compounds have been identified in corks, such as geosmin, 1‐octen‐3‐one and related 1‐octen‐3‐ol, 2‐methylisoborneol (MIB), and 2‐methoxy‐3,5‐dimethylpyrazine (MDMP); these taints are typically produced by molds and have musty or earthy aromas analogous to TCA [2,15] (Table 18.1). 3‐Isopropyl‐2‐methoxypyrazine (IPMP) and guaiacol can also contribute to cork‐derived taints, but are most often derived from other sources and are discussed elsewhere (Chapters 5 and 12). Because cork taint compounds are microbially derived, it is common for more than one of the aforementioned compounds to be present when cork taint is detected, although TCA has received by far the most attention (see Reference [15] for a comprehensive review).

Figure 18.1 Potential origins of taints include (a) cork wood with the outer ring of bark used to make (b) cork closures (shown in a piece of cork bark), and (c) an oak wood barrel (the discoloration around the bung is mold growth).

Source: (b) Reproduced with permission of Duc‐Truc Pham. (c) Reproduced with permission of Paul Grbin

The transfer of TCA from corks and other contaminated materials into wine has been well studied, and TCA is expected to be a good analog for other haloanisoles. In general, TCA‐containing material must be in direct contact with wine, or at least in close proximity to allow aerial transfer (TCA vapor can also taint corks in this way [16]), and there is no evidence that TCA can otherwise form in bottled wine. Similarly, TBA is known to contaminate wine and other materials such as corks, barrels, and winemaking equipment via the atmosphere [7].

Several factors will affect the amount of TCA transferred into wine from a cork. TCA concentration varies within cork and other materials (i.e., on the surface or internal), and migration down the full length of a cork in a sealed bottle is slow enough to be an improbable route of transmission. The rate of transfer will also be affected by the ease of TCA migration within and polarity of the contaminated matrix ³ [15]. Because haloanisoles are highly non‐polar (e.g., TCA has log P = 3.7 and water solubility = 0.01 g/L), uncontaminated cork and other similarly hydrophobic materials such as plastics are effective at removing TCA from contaminated wine. In contrast, more polar cork taint compounds bearing pyrazine, alcohol, and ketone functionalities would be more readily transferred into wine due to their greater solubility and lower affinity for cork [15]. TCA and geosmin are able to persist in bottled wine and it seems that other cork taint compounds would be quite stable as well.

18.2.2 Taints from other winery sources

Halophenols, the parent compounds of TCA and other haloanisoles, are widely used as wood preservatives due to their biocidal properties. As a result, 2,4,6‐trichlorophenol (TCP) and its brominated equivalent, 2,4,6‐tribromophenol (TBP, also used as a flame retardant in numerous products), are ubiquitous environmental pollutants. Alternatively, halophenols arise from the combination of phenol and halogen sources via electrophilic aromatic substitution (Chapter 11), leading to a range of substituted chloro‐ and bromophenols. For instance, chlorophenols can be formed when electrophilic chlorine (e.g., chlorine bleach, chlorine gas) comes into contact with materials containing phenols (e.g., wood, cardboard, plastics, cleaning products), with subsequent contamination of winemaking aids and additives or the wines directly. As an example, bleaching (i.e., hypochlorite washing) of corks has been reported to be a common source of TCP (and eventually TCA) [8], although this process is no longer used by major cork manufacturers. Beyond serving as haloanisole precursors, halophenols are implicated as taint compounds in their own right. They tend to have plastic/chemical/medicinal aromas but higher thresholds than haloanisoles, although potent halophenols with low ng/L thresholds do exist, such as 2,6‐dichlorophenol (2,6‐DCP) and 6‐chloro‐o‐cresol (6‐CC) [17] (Table 18.1). ⁴ Similar to TCA, the most likely route of wine contamination is direct contact with tainted materials, although airborne transmission could be possible for wines handled in a contaminated atmosphere. ⁵

18.2.3 Taints from the vineyard

In addition to appearing in tainted corks, the mold metabolites geosmin, 1‐octen‐3‐one, 1‐octen‐3‐ol and MIB can also be found at high levels in infected grapes (e.g., Botrytis cinerea, Figure 18.2) [9]. Concentrations of many of these compounds will decrease during fermentation by >90% [10]:

• Carbonyl containing mold‐derived taint compounds, such as the ketone, 1‐octen‐3‐one, will be reduced almost completely to corresponding and less potent alcohols, for example, 1‐octen‐3‐ol (Chapter 22.1).
• The tertiary alcohol, MIB, decreases to undetectable levels, possibly as a result of carbocation formation and subsequent dehydration/degradation.
• Losses of the other mold metabolites can be around 50%, likely due at least in part to volatilization during fermentation or adsorption on to yeast lees.

Figure 18.2 Sources of taints from the vineyard, shown in images of (a) Botrytis on a grape berry, (b) powdery mildew on a grape cluster, (c) MALB on a grapevine leaf, (d) MALB floating in a red grape must, and (e) bushfire smoke drifting in the vicinity of a vineyard. Photographs courtesy of Tijana Petrovic (a), Bruce Bordelon (b and d), Erik Glemser (c), and Tony Mills (e)

Overall, geosmin and 1‐octen‐3‐ol seem to be the most stable of these malodorous compounds, and the most likely to cause a taint in wine made from diseased grapes. Several other earthy‐smelling compounds arising from rotting grapes (e.g., fenchone, fenchol, 2‐octen‐1‐ol, heptan‐1‐ol) can potentially contribute to taint problems if present at high enough concentrations after fermentation ([2, 9, 18]). Apart from eliminating rotten grape bunches prior to fermentation, wine amelioration techniques are limited to fining with activated charcoal or other typical fining agents and these compounds appear to be challenging to remove selectively – although thermal treatments could potentially be used for geosmin remediation [19] (Chapter 26.2).

Contamination of grapes in the vineyard by insects can lead to taints, most notably multicolored Asian ladybeetle (MALB, Figure 18.2) taint caused by extraction of 3‐isopropyl‐2‐methoxypyrazine (IPMP). Insect‐derived taints contributing methoxypyrazines were covered earlier (Chapter 5); other insects such as millipedes and earwigs and their feces [20] can conceivably taint grapes and wines, although there is a lack of information on the extent of such contamination and the compounds responsible.

Smoke taint occurs in the vineyard as a result of contamination of grapes by bushfire (wildfire) smoke (Figure 18.2), thereby conferring undesirable smoke, ash, and medicinal characters to wine. This outcome is a result of volatile phenols present in smoke being transferred to the grape berry and subsequently glycosylated. This is covered in more detail in Chapter 12.

Eucalyptus aromas can arise in red wine from exposure of grapes in the vineyard to 1,8‐cineole emanating from nearby Eucalyptus trees (Chapter 8). Although this illustrates the potential for exogenous volatiles in the vineyard to contaminate grapes and wine (anecdotal information exists about grapes being tainted by nearby food processing plants, for example), it is unclear if 1,8‐cineole contamination results in increased consumer rejection. A limited number of consumer studies using red wines spiked with different amounts of 1,8‐cineole indicate that the concentrations encountered in eucalyptus‐affected wines are not objectionable to most consumers (see Reference [21] and references therein for more information).

Mycotoxins are another class of compounds arising in the vineyard due to fungal diseases [9]; their primary concern is not due to any adverse effect on flavor but because of concerns for human health. Ochratoxin A (OTA), produced by Aspergillus spp., is one such metabolite implicated as a human carcinogen and associated with liver and kidney toxicity [22]. In the European Union (EU), wine consumption is estimated to be the second greatest dietary source of OTA after cereals. Conditions that promote Aspergillus growth (warmer, more humid) typically have higher OTA concentrations. OTA from wine can be limited through physical (e.g., removal of moldy grape bunches, lighter pressing, or filtration – the most effective treatment), chemical (e.g., fining agents such as bentonite, chitosan, and charcoal, or oak products – currently the most studied remedy), or microbiological (e.g., yeast or bacteria, through adsorption rather than metabolism) processes [19]. However, a range of studies have shown that average concentrations are well below the designated EU limit (2 μg/L) and the need for OTA remediation is expected to be rare.

Table 18.1 Indicative odor descriptors, detection thresholds, and odor activity values of the main compounds that impart a taint to wine [2,7–13] ^a

Compound b	Structure	Odor descriptor	Threshold (ng/L)	OAV (max.)
2,4,6‐Trichloroanisole (TCA)		Mold, earth	3	12
2,4,6‐Tribromoanisole (TBA)		Mold, earth	8	5
2,6‐Dichlorophenol (2,6‐DCP)		Plastic, medicinal	32	6
6‐Chloro‐o‐cresol (6‐CC)		Disinfectant	70	7
(–)‐Geosmin		Earth	50	6
1‐Octen‐3‐one		Mushroom	70	6
1‐Octen‐3‐ol		Mushroom	40 000	5
(–)‐2‐Methylisoborneol (MIB)		Earth	55	3 c
2‐Methoxy‐3,5‐dimethylpyrazine (MDMP)		Fungal, earth	2	2 (7 d )

^a Descriptor and threshold data refer to different matrices, including water, model wine, and red or white wine.

^b Common names are widely used in the literature since the systematic names are more complex; for example, geosmin is (4S,4aS,8aR)‐4,8a‐dimethyl‐1,2,3,4,5,6,7,8‐octahydronaphthalen‐4a‐ol.

^c For red grape must. Although often found in rotten grapes or musts, MIB is not detectable after fermentation, either due to metabolism by yeast, binding to lees, or other mechanisms.

^d For a wine stored in a stainless steel tank in contact with slightly toasted oak chips. Temperatures encountered with traditional barrel‐making readily eliminate MDMP if present in the wood.

18.3 Off‐flavors in wine

Off‐flavor compounds arise from chemical or microbiological transformations of wine components, and thus can often be identified even in sound wines, but at sufficiently low concentrations so as not to be sensorially detectable or decrease consumer acceptance. One example is indole, which is only found at suprathreshold concentrations as a result of suboptimal (i.e., stuck or sluggish) fermentation (Chapter 5). More common wine off‐flavors include acetic acid and ethyl acetate associated with volatile acidity, sulfur compounds such as H₂S related to reductive characters, nitrogenous compounds imparting vegetal or mousy flavors, carbonyl compounds from oxidation, and ethylphenols indicative of spoilage by Brettanomyces (Dekkera) yeast. These compounds are described in other chapters, as summarized in Table 18.2. As a caveat, these compounds are ordinary constituents of wine, and they may not be considered off‐flavors in certain wine styles or by particular groups of consumers or wine experts (e.g., see References [23] and [24]). For example, the ethylphenols responsible for “Brett” aromas can be at concentrations well in excess of threshold in certain very expensive French red wines, where they may considered part of the wine’s style rather than a fault. Similarly, oxidized aromas are entirely appropriate in Sherry wines, as are high levels of volatile acidity in botrytized wines such as Sauternes.

A final defect worthy of discussion is the off‐odor reminiscent of crushed geranium leaves (“geranium taint”), attributable to 2‐ethoxyhexa‐3,5‐diene [25]. This unsaturated ethyl ether, having a threshold of 100 ng/L [26], arises from reduction and subsequent rearrangement (cf. isoprenoid rearrangements, Chapter 8; also see nucleophiles/electrophiles, Chapter 10) of the preservative sorbic acid (i.e., (2E,4E)‐2,4‐hexadienoic acid, Figure 18.3) by lactic acid bacteria (Chapter 22.5). Sorbic acid or its salts may be used by winemakers to inhibit yeast growth in sweet wine, but adequate levels of SO₂ and sterile conditions at bottling are necessary to prevent bacterial growth and possible development of a geranium off‐odor. It follows that sorbic acid should not be added to a wine that will subsequently undergo malolactic fermentation or be subjected to any other bacterial activity (even inadvertently).

Figure 18.3 Formation of geranium off‐odor through microbial reduction (indicated by [H]) of sorbic acid to the corresponding alcohol and acid‐catalyzed rearrangement to the malodorous ether, 2‐ethoxy‐3,5‐hexadiene. This may occur directly by addition of ethanol as a nucleophile (Nu:) to the secondary carbocation intermediate arising from protonated sorbic alcohol, or indirectly by addition of ethanol to C2 of protonated 3,5‐hexadien‐2‐ol formed when Nu: is water. Addition of ethanol to the primary carbocation intermediate leads to another ether, 1‐ethoxy‐2,4‐hexadiene (i.e., the ethyl ether of sorbic alcohol) – this ether and the two diene alcohols may partly contribute to the off‐odor [25]

Table 18.2 Cross‐references to chapters that describe compounds causing some of the more common off‐flavors in wine

Off‐flavor	Odor descriptor	Compounds	Chapter
Volatile acidity	Vinegar Nail polish remover	Acetic acid Ethyl acetate	3  6
Mousy	Mousy, cardboard, cracker	Cyclic imines	5
Atypical aging Foxy	Acacia, mothball	ortho‐Aminoacetophenone	5
Oxidized	Rotting apple, green, earthy	Various aldehydes, sotolon	9
Reduced	Rotten egg, putrid, onion, vegetal	Various sulfur compounds	10
Brett	Leather, horse stable, spice	4‐Ethylphenol 4‐Ethylguaiacol	12

References

1. 1. Ridgway, K., Lalljie, S.P.D., Smith, R.M. (2010) Analysis of food taints and off‐flavours: a review. Food Additives and Contaminants: Part A: Chemistry, Analysis, Control, Exposure & Risk Assessment , 27 (2), 146–168.
2. 2. Boutou, S. and Chatonnet, P. (2007) Rapid headspace solid‐phase microextraction/gas chromatographic/mass spectrometric assay for the quantitative determination of some of the main odorants causing off‐flavours in wine. Journal of Chromatography A , 1141 (1), 19.
3. 3. Hudelson, J. (2011) Wine faults: causes, effects, cures , The Wine Appreciation Guild, San Francisco, CA.
4. 4. Jackson, R.S. (2008) Postfermentation treatments and related topics, in Wine science: principles and applications , Academic Press, San Diego, CA, pp. 418–519.
5. 5. Cowey, G., Coulter, A., Holdstock, M. (2009) Don't get contaminated this vintage! AWRI Technical Review , 178, 22–27.
6. 6. Ribéreau‐Gayon, P., Glories, Y., Maujean, A., Dubourdieu, D. (2006) Chemical nature, origins and consequences of the main organoleptic defects, in Handbook of enology: the chemistry of wine stabilization and treatments , 2nd edn (eds Ribéreau‐Gayon, P., Glories, Y., Maujean, A., Dubourdieu, D.), John Wiley & Sons, Ltd, Chichester, UK, pp. 233–284.
7. 7. Chatonnet, P., Bonnet, S., Boutou, S., Labadie, M.D. (2004) Identification and responsibility of 2,4,6‐tribromoanisole in musty, corked odors in wine. Journal of Agricultural and Food Chemistry , 52 (5), 1255–1262.
8. 8. Pollnitz, A.P., Pardon, K.H., Liacopoulos, D., et al. (1996) The analysis of 2,4,6‐trichloroanisole and other chloroanisoles in tainted wines and corks. Australian Journal of Grape and Wine Research , 2 (3), 184–190.
9. 9. Steel, C.C., Blackman, J.W., Schmidtke, L.M. (2013) Grapevine bunch rots: impacts on wine composition, quality, and potential procedures for the removal of wine faults. Journal of Agricultural and Food Chemistry , 61 (22), 5189–5206.
10. 10. La Guerche, S., Dauphin, B., Pons, M., et al. (2006) Characterization of some mushroom and earthy off‐odors microbially induced by the development of rot on grapes. Journal of Agricultural and Food Chemistry , 54 (24), 9193–9200.
11. 11. Darriet, P., Pons, M., Lamy, S., Dubourdieu, D. (2000) Identification and quantification of geosmin, an earthy odorant contaminating wines. Journal of Agricultural and Food Chemistry , 48 (10), 4835–4838.
12. 12. Pons, M., Dauphin, B., La Guerche, S., et al. (2011) Identification of impact odorants contributing to fresh mushroom off‐flavor in wines: incidence of their reactivity with nitrogen compounds on the decrease of the olfactory defect. Journal of Agricultural and Food Chemistry , 59 (7), 3264–3272.
13. 13. Bowen, A.J. and Reynolds, A.G. (2012) Odor potency of aroma compounds in Riesling and Vidal blanc table wines and icewines by gas chromatography–olfactometry–mass spectrometry. Journal of Agricultural and Food Chemistry , 60 (11), 2874–2883.
14. 14. Takeuchi, H., Kato, H., Kurahashi, T. (2013) 2,4,6‐Trichloroanisole is a potent suppressor of olfactory signal transduction. Proceedings of the National Academy of Sciences , 110 (40), 16235–16240.
15. 15. Sefton, M.A. and Simpson, R.F. (2005) Compounds causing cork taint and the factors affecting their transfer from natural cork closures to wine – a review. Australian Journal of Grape and Wine Research , 11 (2), 226–240.
16. 16. Barker, D.A., Capone, D.L., Pollnitz, A.P., et al. (2001) Absorption of 2,4,6‐trichloroanisole by wine corks via the vapour phase in an enclosed environment. Australian Journal of Grape and Wine Research , 7 (1), 40–46.
17. 17. Capone, D.L., Van Leeuwen, K.A., Pardon, K.H., et al. (2010) Identification and analysis of 2‐chloro‐6‐methylphenol, 2,6‐dichlorophenol and indole: causes of taints and off‐flavours in wines. Australian Journal of Grape and Wine Research , 16 (1), 210–217.
18. 18. Sadoughi, N., Schmidtke, L.M., Antalick, G., et al. (2015) Gas chromatography–mass spectrometry method optimized using response surface modeling for the quantitation of fungal off‐flavors in grapes and wine. Journal of Agricultural and Food Chemistry , 63 (11), 2877–2885.
19. 19. La Guerche, S. (2004) Recherches sur les déviations organoleptiques des moûts et des vins associées au développement de pourritures sur les raisins: étude particulière de la géosmine. PhD Thesis, Universitéde Bordeaux II, France.
20. 20. Kehrli, P., Karp, J., Burdet, J.P., et al. (2012) Impact of processed earwigs and their faeces on the aroma and taste of “Chasselas” and “Pinot Noir” wines. Vitis , 51 (2), 87–93.
21. 21. Capone, D.L., Jeffery, D.W., Sefton, M.A. (2012) Vineyard and fermentation studies to elucidate the origin of 1,8‐cineole in Australian red wine. Journal of Agricultural and Food Chemistry , 60 (9), 2281–2287.
22. 22. Quintela, S., Villarán, M.C., López de Armentia, I., Elejalde, E. (2013) Ochratoxin A removal in wine: a review. Food Control , 30 (2), 439–445.
23. 23. Goode, J. (2005) Brettanomyces, in The science of wine: from vine to glass , University of California Press, Los Angeles, CA, pp. 136–143.
24. 24. Jackson, R.S. (2009) Olfactory sensations, in Wine tasting: a professional handbook , 2nd edn, Academic Press, San Diego, CA, pp. 55–128.
25. 25. Crowell, E.A. and Guymon, J.F. (1975) Wine constituents arising from sorbic acid addition, and identification of 2‐ethoxyhexa‐3,5‐diene as source of geranium‐like off‐odor. American Journal of Enology and Viticulture , 26 (2), 97–102.
26. 26. Wurdig, G. (1977) Technologie du traitement a lʼacide sorbique. Bulletin de lʼOIV , 50, 547–558.

Notes

1 An analogy can be drawn between a “good” versus “bad” flavor compound and a “drug” – the dose makes the poison.

2 A distinction is not necessarily made between taints and off-flavors as the cause of an undesired aroma or flavor is not always clear.

3 A common quality control for TCA (and other taint compounds) in corks is a “soak test” – a sample of corks are soaked in aqueous ethanol or neutral wine and the extract is analysed sensorially or instrumentally. However, there are limitations to this process. Often the number of corks screened from a batch is relatively small (e.g., 5 corks) so does not account for cork variability. Additionally, taint compounds may not be uniformly distributed in the cork – a short soak test will not account for migration from within the cork matrix over many months of storage, leading to an underestimation of the potential to taint a wine. Conversely, more polar compounds such as MDMP can be more easily extracted during soaking trials, meaning cork screening may overestimate their potential importance. Thus, soaking trial conditions do not fully emulate what occurs in bottled wine, but rather give an indication of the danger presented by TCA (or similarly hydrophobic compounds) following short-term storage.

4 Often a number of chlorophenols may be identified in suspect wines, but their high thresholds relative to the concentrations found make it difficult to ascribe a primary role (and they are not included in Table 18.1). Based on their potency, compounds such as 2,6-DCP and 6-CC may be the principal cause of chlorophenol taints, but they are not necessarily detected in every analysis of tainted wines.

5 Potable water supplies and distribution systems can be a source of halophenol (and haloanisole) contamination. This route of transmission seems less likely in a winery setting amongst all the other potential sources, in comparison to a brewery, for example, because the amount of water incorporated during winemaking is comparatively low.