23.1
Glycosidic Precursors to Wine Odorants

23.1.1 Introduction

Glycosides are plant secondary metabolites consisting of a non‐sugar component, called an aglycone, attached to one or more sugars (Figure 23.1.1).¹ Glycosides are ubiquitous in the plant kingdom and found in all major plant organs, including fruit, leaf, seed, flower, bark, and root [1, 2].² Most aglycones are non‐polar or semi‐polar – glycosylation increases their water solubility and lowers reactivity, which in turn facilitates the transport, accumulation and storage, and detoxification of these compounds [1–3]. Glycoside metabolism in plants involves glycoside transferases and hydrolases to catalyze the glycosylation and hydrolysis of aglycones [3, 4]. Several glycosides of non‐volatile phenolic aglycones are discussed earlier in this book, including anthocyanins and flavonols (Chapter 11). However, the focus of this chapter is on glycosides of volatile aglycones, including aliphatic alcohols (e.g., C₆ compounds), shikimate derivatives (e.g., benzyl alcohol, phenols, vanillin) and mevalonate/deoxyxylulose phosphate (DXP) derivatives (e.g., monoterpenoids, C₁₃‐norisoprenoids) (Figure 23.1.2), which can serve as precursors of odorous compounds following hydrolysis [5]. These glycosides are often referred to as bound compounds, which can be converted to free odorants during fermentation and storage. Several factors are important in determining the eventual concentrations of free aroma compounds arising from glycosidic precursors, including the amounts of precursors present in juice or must, enzymatic activity during fermentation, and pH/temperature during storage.

Figure 23.1.1 Structure of a generic glycoside, showing the aglycone (R¹) attached at the anomeric hydroxyl of D‐glucose (β‐anomer shown; hence β‐D‐glucoside). The bond between aglycone and oxygen is known as the ether bond, whereas the one between the oxygen and sugar moiety is the glycosidic bond. Different sugars (R²) can be attached to the 6‐position hydroxyl of glucose, usually through their own anomeric position as indicated (). Note for the β‐anomer of D‐glucose the hydroxyls are all in the equatorial position and the hydrogens, although not drawn, are all axial

Figure 23.1.2 Examples of the types of glycosides of volatile aglycones (monoterpenoid, C₁₃‐norisoprenoid, phenolic) identified in grapes and wines³

23.1.2 Formation of glycosidic aroma precursors in grape berries

Glycosides in wine originate from the grape berry, where they accumulate during ripening, and appear to be well correlated with the concentration of their corresponding aglycone(s). The aglycone is always attached directly to β‐d‐glucose, yielding a monosaccharide (i.e., an O‐β‐D‐glucoside, Figure 23.1.1, R² = H), but the glucose can be further substituted by other sugars (α‐l‐arabinofuranose, α‐L‐rhamnopyranose, β‐D‐xylopyranose, β‐D‐apiofuranose, β‐D‐glucose, Figure 23.1.2) to give the corresponding disaccharides [1, 6]. The mechanisms underpinning glycosylation of volatiles in grape berries have not been fully elucidated but their formation putatively involves the reaction of an alcohol with a sugar. This is an energetically unfavorable reaction, and requires activated nucleotide sugars and the presence of uridine diphosphate (UDP) glycosyltransferase enzymes (UGTs, Figure 23.1.3) [7]. The aglycone components are often biosynthesized in the grape berry, either directly in the case of monoterpenoids (Chapter 8, and polyhydroxylated variants outlined below) and higher alcohols (Chapter 6), or as a result of the breakdown of carotenoids (Chapter 8), as described in more detail below. Grapes can also produce trace concentrations of volatile phenol glycosides, but concentrations capable of releasing suprathreshold amounts of aglycones are typically observed only following contamination of grapevines by wildfire smoke (Chapter 12). The major classes of glycosidic precursors to wine aroma compounds, and factors determining their levels in grapes, are summarized in Table 23.1.1.

Figure 23.1.3 Putative grape berry glycosylation reaction showing formation and transfer of an activated glycosyl residue (e.g., UDP‐glucose) to an aglycone acceptor under the action of UGT⁴

Table 23.1.1 Classes of wine aroma compounds bound as glycoconjugates in grape berries, examples of the volatiles formed from each, and factors affecting berry glycoside concentrations

Aglycone class	Key volatiles formed	Major factor(s) determining concentration
Monoterpenoidsa	Linalool, geraniol	Cultivar: higher in Muscat‐type grapes Maturity: accumulation commences ~4 weeks after veraison and continues through ripening
C13‐Norisoprenoidsa	TDN, β‐damascenone	Cultivar (for TDN): higher in Riesling Growing conditions (for TDN): more cluster exposure to sun, lower N and water. Maturity: accumulation commences 1–3 weeks after veraison and continues through ripening
Volatile phenols	Guaiacol, 4‐methylguaiacol, vanillin, syringol	Growing conditions: exposure of grapes to smoke around veraison, or upon application of oak extracts to vines
Higher alcohols	Hexanol	Unknown

^a May also be formed by rearrangement of odorless polyols, as described below.

23.1.2.1 Carotenoid‐derived precursors

Of the various aglycone classes, formation of C₁₃‐norisoprenoid precursors deserves special discussion due to their genesis from multiple precursors and intermediates. As mentioned in Chapter 8, the key odorous C₁₃‐norisoprenoids are derived from precursors linked to carotenoid degradation [8, 9]. Carotenoids are widely distributed, naturally occurring pigments present in mature grape berries at total concentrations around 0.4–2.5 mg/kg, primarily as lutein and β‐carotene. Carotenoids have key roles in photosynthesis (including in green berries) and concentrations in unripe grapes are often 3‐fold higher than at maturity [10, 11]. Carotenoid degradation will commence 1 week pre‐veraison, and C₁₃‐norisoprenoid precursors will begin to accumulate about 1–2 weeks after veraison [12]. While oxidative degradation could potentially occur spontaneously, several carotenoid cleavage dioxygenases (CCD) have been identified in grapes [13], and at least one (encoded by VvCCD1) is demonstrated to produce C₁₃‐norisoprenoids from carotenoids [12]. Exemplary pathways for formation of two odor‐active C₁₃‐norisoprenoids from two different carotenoids involve:

• Direct formation in grapes from oxidative cleavage of carotenoids, as appears to happen with β‐ionone (Chapter 8).
• Formation of non‐odorous C₁₃‐norisoprenoid intermediates, which can undergo acid‐catalyzed rearrangements during storage; for example, allenic triol can serve as a precursor to β‐damascenone during acidic storage (Figure 23.1.4a). These intermediates often contain multiple OH groups (i.e., they are polyols) and may exist in grapes as glycosides. This pathway appears to occur for TDN, vitispirane, actinidols, and some β‐damascenone precursors, and will require an initial hydrolysis step to release the glycoside, as described further below.

Figure 23.1.4 Acid‐catalyzed rearrangements of non‐odorous precursors leading to formation of (a) β‐damascenone from carotenoid‐derived allenic triol (earlier polyol intermediates in this rearrangement sequence may also be accumulated as glycosidic precursors in grapes, where glucose is conjugated to one of the hydroxyl groups) and (b) (–)‐cis‐rose oxide from citronellyl diol, which can form from geranyl diol after enzymatic reduction ([H]) of a double bond

23.1.2.2 Polyhydroxylated monoterpenoids

Free and glycosidically bound monoterpenoid diols (i.e., diendiols related to linalool, geraniol, nerol, and citronellol) have been identified in grape juices [14–18]. Odorless polyhydroxylated variants are produced enzymatically, and glycosylated as outlined above. The pool of bound monoterpenoids fluctuates but generally increases with grape berry ripening, and exceeds the concentration of free monoterpenoids in mature berries. The main exception to this relates to a linalyl diendiol ((E)‐3,7‐dimethylocta‐1,5‐dien‐3,7‐diol), which was found to increase rapidly during ripening and eventually exceed the concentration of all other free monoterpenoids [19]. As with aroma precursors, such as C₁₃‐norisoprenoids described above, acid‐catalyzed hydrolysis of glycosides and rearrangements lead to a range of wine aroma compounds, such as hotrienol, cis‐rose oxide (and linalool or nerol oxides [14]), and wine lactone (Chapters 8 and 25). An example showing the formation of cis‐rose oxide from citronellyl diol (which can arise from geranyl diol) [20] is shown in Figure 23.1.4b.

Like other grape‐derived aroma precursors such as varietal thiol conjugates (Chapter 23.2), glycosides are distributed between the pulp (including juice) and skin of grape berries to varying extents depending on the aglycone, with a slight predominance in favor of skin [15, 21, 22]. From the analysis of volatiles released during hydrolysis experiments, it can be inferred that juice and must concentrations of different glycosides are in the order of tens to thousands of μg/L [23–25].

23.1.3 Glycosidic aroma precursors – extraction

Because glycosidic aroma precursors are proportionally higher in the skins, considerable research has explored strategies to increase glycoside extraction in wines that are pressed prior to fermentation, that is, most white wines. Most commonly, this is achieved through skin contact, often in conjunction with enzyme preparations:

• Pre‐fermentation skin contact facilitates extraction of glycosidic precursors. Macerating crushed grapes for 4–12 hours at 10 °C [26] or 6–23 hours at 15–18 °C prior to pressing and fermenting [27–31], led to significant increases (up to 2‐fold in total for a compound class) for a range of volatile compounds arising from glycosides, including monoterpenoids, benzenoids, and C₁₃‐norisoprenoids.
• The use of commercial pectolytic enzymes (i.e., pectinases, also used for juice clarification, Chapter 21) can aid in pre‐fermentative extraction of aroma precursors from skins of white and red grape varieties [32]. Treatment of grapes [33] and macerating musts [34] with pectinases enhanced glycoside concentrations in a similar manner to skin contact alone.
• Conversely, the addition of pectinase during clarification of juice has also been found to significantly lower the concentration of glycosides [35]. This may be because many pectinase preparations can possess glycosidase side‐activity (Table 23.1.2), which hydrolyses glycosides and releases bound aroma compounds, as discussed below. This aspect of pectinase use will affect wine aroma release from glycosides rather than extraction of the glycosides from cells.

Table 23.1.2 Enzymatic activities (nkat/mg)⁶ of some commercial pectinase/hemicellulase preparations that have been used in winemaking. Data from Reference [38]

Enzyme preparation	β‐Glucosidase	α‐Arabinosidase	α‐Rhamnosidase	β‐Apiosidase
AR 2000	5.7	14.7	0.3	1.08
Cellulase A	6.1	0.6	0.07	–
Hemicellulase	7.1	7.0	0.9	–
Novoferm 12	8.4	0.5	0.05	0.15
Pectinase 263	7.2	1.4	0.3	0.2
Pectinol D5S	0.5	0.7	–	–
Pectinol VR	0.2	0.1	–	–
Pektolase 3PA	1.5	3.8	0.04	0.3
Rohament CW	3.3	0.7	0.4	–
Ultrazym 100	0.5	0.1	–	0.03

23.1.4 Hydrolysis of glycosidic aroma precursors – mechanisms

Being non‐volatile, glycosidic aroma precursors need to be hydrolyzed to release the volatile aglycone. Hydrolysis can be acid‐ or enzyme‐catalyzed [6, 36], and the resulting products will differ depending on the mode of hydrolysis [37] and structure of glycosides and aglycones [1].

23.1.4.1 Acid‐catalyzed hydrolysis

Glycosides undergo slow acid hydrolysis during wine storage [3, 36, 38] and hydrolytic studies have been conducted using mild (e.g., wine‐like pH, 50 °C, several weeks) or harsh (e.g., pH 1, 100 °C, 1 hour) acidic conditions. The former is more suited to examination of compositional changes during wine aging while the latter “forcing conditions” are more useful for evaluating the maximum pool of odorants that could be released from grape glycosidic precursors. Acid hydrolysis can result in cleavage of either the ether or glycosidic bonds (see Figure 23.1.1), depending on the structure of the aglycone, and released volatile compounds will often undergo further acid‐catalyzed rearrangements (Figures 23.1.4 and 23.1.5); higher temperatures and lower pH will lead to greater release and rearrangement of aglycones [37, 39, 40]. Glycosylation stabilizes an aglycone, such that free geraniol rearranges about an order of magnitude faster than geranyl glucoside to form related odorants (e.g., linalool). Glycosylation may also change the relative amounts of products formed following hydrolysis; for example, 50% more β‐damascenone is produced relative to its odorless counterpart 3‐hydroxy‐β‐damascone from a glycosidic precursor, as opposed to from the corresponding precursor aglycone [1, 39].

Figure 23.1.5 Acid‐catalyzed glycoside hydrolysis of (a) an activated alcohol (i.e., allylic‐stabilized carbocation intermediate) such as geraniol (ether bond cleavage) and subsequent reaction of liberated carbocation to form new compounds including linalool and α‐terpineol, and (b) unactivated alcohol citronellol (glycosidic bond cleavage) to produce the volatile aglycone, which is similar to the mechanism of enzymatic hydrolysis [4]

23.1.4.2 Enzyme‐catalyzed hydrolysis

Aglycone liberation by enzymes (glycosidases) can be single step or sequential [1, 3, 6, 36]. Simple glucosides can be cleaved by the action of β‐glucosidase (single step) whereas disaccharides incorporating a sugar other than glucose often require specific enzymes (exoglycosidases, e.g., α‐arabinosidase, etc., see Table 23.1.2) to hydrolyze the intersugar linkage, cleaving the terminal sugar prior to the glucose‐aglycone bond being hydrolyzed (sequential). There are, however, diglycosidases (endo‐β‐glucosidases) that can cleave the disaccharide directly in a single step. Grapes contain endogenous glycosidases, with β‐glucosidase being the most dominant, but these are essentially inactive at juice pH and inhibited by glucose. Given the abundance of glycosides in grapes and their relatively slow release during fermentation, alternative sources of glycosidases have been evaluated for their potential to liberate aglycones and improve wine aroma. Yeast and bacteria have glycosidase activity during fermentation (Section 23.1.5 and [40]), but commercial preparations typically involve exogenous glycosidases from Aspergillus niger, which have greater pH, temperature, glucose, and ethanol tolerance. Ideally, an enzyme preparation will have a broad range of activities to sequentially hydrolyze different disaccharide classes. However, commercial preparations show considerable variation in their specificity towards different glycosides (see Table 23.1.2) [3, 36, 38], which means that different preparations can lead to different sensory profiles. Glycosides can also be released enzymatically by oral microflora [41] (e.g., volatile phenol glycosides, Chapter 12), which can result in a more intense “aftertaste” following consumption of the wine.

The glycosidase side activity of pectinase preparations have been frequently shown to increase free volatile concentrations (Table 23.1.3). However, this glycosidase activity is not necessarily desirable [3, 38, 42]. In the case of red winemaking, β‐glucosidase activity towards anthocyanin glucosides should be minimal, due to the important stabilizing influence of glucosylation on anthocyanins (Chapter 16). Similarly, enzyme preparations with cinnamate esterase activity are avoided due to their hydrolytic effect on hydroxycinnamoyl tartrate esters (Chapters 13 and 23.3); release of the free hydroxycinnamic acids provides a source of precursors for “Brett” off‐flavor formation (Chapters 12 and 23.3). The presence of other esterases is also cause for concern, considering the significance of volatile esters to wine aroma (Chapter 7). As an example, a wine treated with AR 2000 for 15 and 30 days revealed a dramatic decrease in several important “fruity” acetate esters. There was also a several‐fold increase in diethyl succinate and diethyl malate, which may not be so consequential but is indicative of the catalytic role of esterases in achieving equilibrium concentrations of esters, through concurrent esterification and hydrolysis reactions (Chapter 7). An alternative enzyme with glycosidase activity, Novarom G, showed no such effect on esters [42].

Table 23.1.3 Effect of exogenous enzyme treatment on the concentrations (μg/L) of classes of aroma compounds in wines of selected grape varieties

Variety	Monoterpenes		Norisoprenoids		Benzenoidsa		Study
	Control	Treated	Control	Treated	Control	Treated
Muscat	4384	7718	9	542	–b	–	[6]
	1520	2010	–	–	24 839	28 213	[43]
Riesling	2418	3119	NDc	407	–	–	[6]
Sauvignon Blanc	53	198	ND	141	–	–	[6]
	17	26	–	–	19 410	22 570	[44]
Shiraz	145	1138	ND	847	–	–	[6]
Traminer	72	336	–	–	15 691	25 606	[44]
Palomino	43	55	–	–	17 686	20 288	[44]
Chardonnay	45	79	–	–	24 438	30 053	[43]
	7	9	–	–	10 574	10 218	[45]
Emir	14	25	48	74	1226	1874	[29]
Airén	56	93	–	–	26 128	32 045	[43]
	ND	ND	–	–	12 217	13 640	[45]
Macabeo	52	85	–	–	30 264	37 665	[43]
	8	9	–	–	8687	9498	[45]

^a Concentrations predominantly result from 2‐phenylethanol in the tens of mg/L range.

^b –, not reported.

^c ND, not detected.

23.1.4.3 Comparison of hydrolysis by enzyme and acid

As outlined above, differences are noted depending on the nature of the glycosides and mode of hydrolysis. To summarize:

• Acid hydrolysis may result in cleavage at either the ether or glycosidic bonds, while enzymatic hydrolysis results in cleavage at the glycosidic group (see Figure 23.1.5).
• Acid hydrolysis often results in additional acid‐catalyzed rearrangements and, by comparison, enzymatic hydrolysis usually results in minimal changes to the aglycone.
• Acid hydrolysis is less sensitive to the sugar group, whereas in enzymatic hydrolysis disaccharide glycoside precursors are usually hydrolyzed more slowly because of the need for multiple selective enzymes.

During winemaking and storage both acid and enzymatic hydrolysis processes will occur, concurrently at some points. Nonetheless, much of the knowledge of aroma precursors has come from studies selectively employing either acid or enzyme hydrolysis of extracts, which are useful approaches for separately assessing the effects of these factors on the release of glycosides. The study of Loscos et al. [37] reinforces earlier research and serves as a useful example for comparing natural and accelerated hydrolysis techniques. Using precursor extracts from different grape varieties, it was shown that alcoholic fermentation tended to release the lowest amount of volatile compounds, enzymatic hydrolysis was very efficient yet the most different to other treatments, and harsh acid hydrolysis (with intermediate release of volatiles) better represented the aroma potential of grapes due to mimicking the acid‐catalyzed reactions and rearrangements that occur in wine over time. Generalizing the results:

• Enzymatic hydrolysis was effective at increasing free volatile benzenoid and monoterpenoids. However, the increase in benzenoids was in part due to vinylphenol formation due to cinnamate esterase side activity, as mentioned above and in Chapter 23.3.
• Harsh acid hydrolysis contributed most to C₁₃‐norisoprenoids and monoterpenoids.
• Alcoholic fermentation resulted in relatively high formation of benzenoids (especially 2‐phenylethanol through amino acid metabolism rather than glycoside release) and lactones (Table 23.1.4).

Table 23.1.4 Aroma compounds released (sum of relative peak areas) from precursor extracts of a range of grape varieties by hydrolysis under different conditions. Data from Reference [37]

Hydrolytic Treament	Compounds Releaseda	Controlb	Verdejo	Tempranillo	Chardonnay	Cab Sauv	Merlot	Muscat	Grenache
Fermentation (synthetic must, 200 g/L glucose, Stellevin NT 116, 54 days)	Monoterpenoids	27	35	32	54	31	47	266	36
	Norisoprenoidsc	1	13	7	13	11	9	9	7
	Benzenoidsd	13	269	282	189	231	283	184	624
Acid (pH 2.5 citrate buffer, 100 °C, 1h)	Monoterpenoids	2	22	24	81	30	66	1325	95
	Norisoprenoidsc	0.1	187	174	276	192	214	160	312
	Benzenoidsd	1	214	248	250	265	306	130	771
Enzyme (AR 2000, pH 5 citrate/phosphate buffer, 40 °C, 16 h)	Monoterpenoids	1	96	218	137	105	97	963	162
	Norisoprenoidsc	0.1	26	19	44	21	13	21	30
	Benzenoidsd	30	7022	9941	5736	10373	5242	2720	11543

^a Total amount determined from the sum of relative areas for a given class of compounds.

^b Treatment performed without addition of precursor extract.

^c 3‐Oxo‐α‐ionol not included.

^d Selected compounds including vinylphenols, vanillin derivatives and benzene derivatives.

Overall, the concentrations of volatiles released by harsh acid hydrolysis were better correlated with volatile concentrations arising through alcoholic fermentation than with enzymatic hydrolysis. However, harsh hydrolytic treatment may lead to unrealistic outcomes, as highlighted with studies on the formation of β‐damascenone from precursors that would produce negligible amounts of β‐damascenone under normal storage conditions [46]. Importantly from a winemaking perspective, due to the low level of hydrolysis, a sizeable pool of precursors remains unaffected by alcoholic fermentation that can then play a role during aging. Additionally, these glycosides can affect aftertaste through in‐mouth release by oral microflora (Chapter 12).

23.1.5 Hydrolysis of glycosides under fermentation and aging conditions

Most yeast and lactic acid bacteria (LAB) strains express glycosidases, and further hydrolysis or rearrangements of aglycones can occur under the acidic conditions of wine storage. The glycoside pool thus provides winemakers with tools to manipulate wine aroma and flavor because relative enzyme activity and substrate specificity will vary among strains and fermentation conditions [47–49]. In fact, many yeast suppliers will advertise yeast strains with glycosidase activity appropriate for a particular wine style.

23.1.5.1 Microbial strain and lees effects

The ability of yeast strains to release monoterpenes from glycosides has been demonstrated by supplementing a model juice media with glycoside precursor fraction containing no free monoterpenes (Figure 23.1.6a). The resulting model wines had 2–5‐fold higher concentrations of monoterpenes than the control. However, interpreting the effects of either aging or microbial action on glycosides is often complicated by additional changes that may occur to odorants during fermentation or storage. In fermentations of real juices, monoterpenes may appear to decrease, quite substantially in some cases (e.g., geraniol, Figure 23.1.6b). This is likely because their release from glycosides can be less than the losses of free monoterpenes present in juice, as a result of lees binding, volatilization, yeast metabolism, or chemical reactions.

Figure 23.1.6 Effect of yeast strain on monoterpene concentrations in (a) a model juice system supplemented with a grape glycoside fraction, showing that yeast glycosidase activity during fermentation can result in an increase in free monoterpenes (data from Reference [50]), and (b) fermentations of real juices, highlighting an apparent decrease in monoterpenes due to losses of free monoterpenes present in the juice (data from Reference [51])

Apart from hydrolyzing glycosides to varying amounts during model fermentations (e.g., up to 10‐fold variation for release of some monoterpenoids), some yeast strains produce higher concentrations of vinylphenols (attributed to esterase activity on hydroxycinnamic acids,6 Chapter 23.3) or lactones from precursor fractions [50, 52, 53]. Not surprisingly, the sensory effects of different strains were easily noted in the model wines but are less obvious when fermenting real Muscat or Riesling grape juices [51, 54]. Nonetheless, the composition of free and bound monoterpenoids and benzenoids differs between yeast strains and the initial juice [54], as shown for Riesling in Table 23.1.5. These data reinforce the notion that winemakers have a level of control on wine composition and aroma through the choice of commercial yeast strain (or by conducting uninoculated fermentation), and also indicate that substantial quantities of residual glycosides remain in the new wine that could be released during aging (Chapter 25).

Table 23.1.5 Concentrations (μg/L) of free and bound aroma compounds in Riesling juice and wines made with different yeast strains. Data from Reference [54]

Compound	Juice		PDM		D47		Fermiblanc		VL1		Nativea
	Bound	Free	Bound	Free	Bound	Free	Bound	Free	Bound	Free	Bound	Free
Linalool	70	32	55	23	59	40	42	38	63	43	48	34
Nerol	20	11	16	<1	14	1	5	2	7	1	7	<1
Geraniol	42	22	28	2	29	2	20	2	18	3	18	4
α‐Tepineol	114	8	91	32	84	35	62	24	80	37	60	50
Benzyl alcohol	16	8	14	5	14	6	11	7	13	9	11	1
2‐Phenylethanol	160	33	149	88	135	101	89	122	130	107	98	139
Total Monoterpenoids	246	73	190	57	186	78	129	66	168	200	133	88
Total Benzenoids	176	41	163	93	149	107	100	129	143	116	109	140

^aUninoculated fermentation.

Similar to the effects of yeast strain, lactic acid bacterial (LAB) strains used for malolactic fermentation (MLF) can also affect glycoside and aglycone concentrations in wine. However, declines in glycosides are not necessarily mirrored with increases in the respective concentrations of free volatiles, likely as a result of other transformations or binding of aglycones to polysaccharides produced by LAB [55]. A study of LAB strains conducting MLF in model wine (pH 3.4) containing Muscat precursor extract showed a decrease in monotepenoid glycosides compared to the control (Figure 23.1.7a), and a related increase in the corresponding aglycones or their rearrangement products (Figure 23.1.7b) [56]. Notably, there was significant glycoside hydrolysis due to MLF, and variable release according to the nature of the aglycone; up to 7% linalool glycoside was hydrolyzed in contrast to 34–38% of α‐terpineol, nerol, and geraniol precursors. However, hydrolysis of glycosides usually exceeds the amount of volatiles formed [57]. Glycosidase activity decreases at low pH and several strains show a significant decrease in precursor hydrolysis (around 20–70%) for model wine systems [56]. Similar to yeast, model glycoside systems treated with different LAB strains can be differentiated sensorially, but these changes are less apparent with real wines [58].

Figure 23.1.7 Effects of LAB strain in a model wine system supplemented with a grape glycoside fraction relative to uninoculated control post‐MLF, showing (a) a decrease in monoterpene glycoside concentrations and (b) an increase in free monoterpenoids.

Data from Reference [56]

Storage on yeast lees can either increase or decrease the concentration of compounds derived from grape precursors, depending on the grape variety (which would affect precursor pool size and pH) and yeast strain (which would affect activity of glycosidases and other enzymes). Maintaining wines on lees for 20 days after alcoholic fermentation showed an increase (around 1.5–3‐fold) in some isoprenoids and lactones in the case of Airen wine, but a decrease (around 1.5–2‐fold) in Macabeo wine, except for β‐ionone, which increased slightly [59]. Some variation was also seen with wine made from sterile Macabeo juice supplemented with grape precursor extract and stored on lees from different yeast strains for 3 and 9 months. A range of isoprenoids (e.g., linalool, α‐terpineol, Riesling acetal, vitispiranes, TDN) increased by up to 2‐fold whereas other compounds (some isoprenoids along with benzenoids and lactones) decreased by up to 3‐fold [60]. Overall, storage time seemed to be more important than yeast strain, but lees were determined to be taking an active role in the observed changes (i.e., changes were beyond what could be explained by acid hydrolysis alone).

23.1.5.2 Aging effects

Glycoside hydrolysis occurs prior to and during fermentation, but a latent pool of aroma compounds in the form of grape‐derived glycosides still remains in finished wine that can continue to transform under abiotic conditions (Chapter 25) [61]. Most of these reactions are acid‐catalyzed, and thus will occur faster at lower pH and higher temperature. However, the concentration and compositional distribution of free volatiles that can be released under real conditions is generally less than that released from the total bound pool during accelerated aging trials (as described above, e.g., 45 °C for several weeks) [39]. For example, many glycosylated β‐damascenone precursors can be converted to free β‐damascenone at high temperatures such as those encountered during cooking,7 but release of β‐damascenone is negligible under typical wine storage conditions [46]. Presumably, this indicates that hydrolysis of several β‐damascenone precursors involves high activation energies, although specific data on this topic are unavailable. Furthermore, glycoside hydrolysis during aging occurs concurrently with degradation reactions involving odorant aglycones (Chapter 25). For many key odorants (e.g., geraniol, linalool), the rate of release from glycosides is slower than the degradation reactions, and an increase in free volatiles would only be observed in cases where the pool of bound compounds is much higher in concentration than that of the free volatiles. However, other compounds (particularly TDN and cis‐rose oxide, Chapter 8) appear to be very stable once formed and can therefore accumulate during storage.

References

1. 1. Winterhalter, P. and Skouroumounis, G. (1997) Glycoconjugated aroma compounds: occurrence, role and biotechnological transformation, in Advances in biochemical engineering/biotechnology: biotechnology of aroma compounds (ed. Berger, R.G.), Springer‐Verlag, Berlin, pp. 73–105.
2. 2. Hjelmeland, A.K. and Ebeler, S.E. (2015) Glycosidically bound volatile aroma compounds in grapes and wine: a review. American Journal of Enology and Viticulture, 66 (1), 1–11.
3. 3. Sarry, J.‐E. and Günata, Z. (2004) Plant and microbial glycoside hydrolases: volatile release from glycosidic aroma precursors. Food Chemistry, 87 (4), 509–521.
4. 4. Sinnott, M.L. (1990) Catalytic mechanism of enzymic glycosyl transfer. Chemical Reviews, 90 (7), 1171–1202.
5. 5. Baumes, R. (2009) Wine aroma precursors, in Wine chemistry and biochemistry (eds Moreno‐Arribas, M.V. and Polo, M.C.), Springer, New York, pp. 251–274.
6. 6. Gunata, Z., Dugelay, I., Sapis, J.C., et al. (eds) (1993) Role of enzymes in the use of the flavour potential from grape glycosides in winemaking in Progress in flavour precursor studies: analysis, generation, biotechnology, September 30–October 2, 1992, Wurzburg, Germany, Allured Publishing Corporation, Carol Stream, IL.
7. 7. Bowles, D. and Lim, E.‐K. (2010) Glycosyltransferases of small molecules: their roles in plant biology, in Encyclopedia of life sciences (ELS), John Wiley & Sons, Ltd, Chichester, UK, pp. 1–10.
8. 8. Baumes, R., Wirth, J., Bureau, S., et al. (2002) Biogeneration of C₁₃‐norisoprenoid compounds: experiments supportive for an apo‐carotenoid pathway in grapevines. Analytica Chimica Acta, 458 (1), 3–14.
9. 9. Mendes‐Pinto, M.M. (2009) Carotenoid breakdown products the‐norisoprenoids‐in wine aroma. Archives of Biochemistry and Biophysics, 483 (2), 236–245.
10. 10. Razungles, A.J., Baumes, R.L., Dufour, C., et al. (1998) Effect of sun exposure on carotenoids and C‐13‐norisoprenoid glycosides in Syrah berries (Vitis vinifera L.). Sciences Des Aliments, 18 (4), 361–373.
11. 11. Oliveira, C., Barbosa, A., Ferreira, A.C.S., et al. (2006) Carotenoid profile in grapes related to aromatic compounds in wines from Douro region. Journal of Food Science, 71 (1), S1–S7.
12. 12. Mathieu, S., Terrier, N., Procureur, J., et al. (2005) A carotenoid cleavage dioxygenase from Vitis vinifera L.: functional characterization and expression during grape berry development in relation to C₁₃‐norisoprenoid accumulation. Journal of Experimental Botany, 56 (420), 2721–2731.
13. 13. Young, P., Lashbrooke, J., Alexandersson, E., et al. (2012) The genes and enzymes of the carotenoid metabolic pathway in Vitis vinifera L. BMC Genomics, 13 (1), 243.
14. 14. Williams, P.J., Strauss, C.R., Wilson, B. (1980) Hydroxylated linalool derivatives as precursors of volatile monoterpenes of muscat grapes. Journal of Agricultural and Food Chemistry, 28 (4), 766–771.
15. 15. Wilson, B., Strauss, C.R., Williams, P.J. (1986) The distribution of free and glycosidically‐bound monoterpenes among skin, juice, and pulp fractions of some white grape varieties. American Journal of Enology and Viticulture, 37 (2), 107–111.
16. 16. Strauss, C.R., Wilson, B., Williams, P.J. (1988) Novel monoterpene diols and diol glycosides in Vitis vinifera grapes. Journal of Agricultural and Food Chemistry, 36 (3), 569–573.
17. 17. Luan, F., Hampel, D., Mosandl, A., Wüst, M. (2004) Enantioselective analysis of free and glycosidically bound monoterpene polyols in Vitis vinifera L. cvs. Morio Muscat and Muscat Ottonel:  evidence for an oxidative monoterpene metabolism in grapes. Journal of Agricultural and Food Chemistry, 52 (7), 2036–2041.
18. 18. Luan, F., Mosandl, A., Münch, A., Wüst, M. (2005) Metabolism of geraniol in grape berry mesocarp of Vitis vinifera L. cv. Scheurebe: demonstration of stereoselective reduction, E/Z‐isomerization, oxidation and glycosylation. Phytochemistry, 66 (3), 295–303.
19. 19. Wilson, B., Strauss, C.R., Williams, P.J. (1984) Changes in free and glycosidically bound monoterpenes in developing muscat grapes. Journal of Agricultural and Food Chemistry, 32 (4), 919–924.
20. 20. Koslitz, S., Renaud, L., Kohler, M., Wust, M. (2008) Stereoselective formation of the varietal aroma compound rose oxide during alcoholic fermentation. Journal of Agricultural and Food Chemistry, 56 (4), 1371–1375.
21. 21. Gunata, Y.Z., Bayonove, C.L., Baumes, R.L., Cordonnier, R.E. (1985) The aroma of grapes. Localisation and evolution of free and bound fractions of some grape aroma components c.v. Muscat during first development and maturation. Journal of the Science of Food and Agriculture, 36 (9), 857–862.
22. 22. Gomez, E., Martinez, A., Laencina, J. (1994) Localization of free and bound aromatic compounds among skin, juice and pulp fractions of some grape varieties. Vitis, 33, 1–4.
23. 23. Gunata, Y.Z., Bayonove, C.L., Baumes, R.L., Cordonnier, R.E. (1985) The aroma of grapes I. Extraction and determination of free and glycosidically bound fractions of some grape aroma components. Journal of Chromatography A, 331, 83–90.
24. 24. Schneider, R., Razungles, A., Augier, C., Baumes, R. (2001) Monoterpenic and norisoprenoidic glycoconjugates of Vitis vinifera L. cv. Melon B. as precursors of odorants in Muscadet wines. Journal of Chromatography A, 936 (1–2), 145–157.
25. 25. Sefton, M.A., Francis, I.L., Williams, P.J. (1993) The volatile composition of Chardonnay juices: a study by flavor precursor analysis. American Journal of Enology and Viticulture, 44 (4), 359–370.
26. 26. Rodrıguez‐Bencomo, J.J., Méndez‐Siverio, J.J., Pérez‐Trujillo, J.P., Cacho, J. (2008) Effect of skin contact on bound aroma and free volatiles of Listán blanco wine. Food Chemistry, 110 (1), 214–225.
27. 27. Cabaroglu, T., Canbas, A., Baumes, R., et al. (1997) Aroma composition of a white wine of Vitis vinifera L. cv. Emir as affected by skin contact. Journal of Food Science, 62 (4), 680–683.
28. 28. Cabaroglu, T. and Canbas, A. (2002) The effect of skin contact on the aromatic composition of the white wine of Vitis vinifera L. cv. Muscat of Alexandria grown in Southern Anatolia. Acta Alimentaria, 31 (1), 45–55.
29. 29. Cabaroglu, T., Selli, S., Canbas, A., et al. (2003) Wine flavor enhancement through the use of exogenous fungal glycosidases. Enzyme and Microbial Technology, 33 (5), 581–587.
30. 30. Selli, S., Canbas, A., Cabaroglu, T., et al. (2006) Aroma components of cv. Muscat of Bornova wines and influence of skin contact treatment. Food Chemistry, 94 (3), 319–326.
31. 31. Sánchez‐Palomo, E., Pérez‐Coello, M.S., Díaz‐Maroto, M.C., et al. (2006) Contribution of free and glycosidically‐bound volatile compounds to the aroma of muscat “a petit grains” wines and effect of skin contact. Food Chemistry, 95 (2), 279–289.
32. 32. Ugliano, M. (2009) Enzymes in winemaking, in Wine chemistry and biochemistry (eds Moreno‐Arribas, M.V. and Polo, M.C.), Springer, New York, pp. 103–126.
33. 33. Itu, N.L., Rapeanu, G., Hopulele, T. (2011) Effect of maceration enzymes addition on the aromatic white winemaking. Annals of the University “Dunarea de Jos” of Galati – Fascicle VI: Food Technology, 35 (1), 77–91.
34. 34. Gil, J.V. and Valles, S. (2001) Effect of macerating enzymes on red wine aroma at laboratory scale: exogenous addition or expression by transgenic wine yeasts. Journal of Agricultural and Food Chemistry, 49 (11), 5515–5523.
35. 35. Moio, L., Ugliano, M., Gambuti, A., et al. (2004) Influence of clarification treatment on concentrations of selected free varietal aroma compounds and glycoconjugates in Falanghina (Vitis vinifera L.) must and wine. American Journal of Enology and Viticulture, 55 (1), 7–12.
36. 36. Pogorzelski, E. and Wilkowska, A. (2007) Flavour enhancement through the enzymatic hydrolysis of glycosidic aroma precursors in juices and wine beverages: a review. Flavour and Fragrance Journal, 22 (4), 251–254.
37. 37. Loscos, N., Hernandez‐Orte, P., Cacho, J., Ferreira, V. (2009) Comparison of the suitability of different hydrolytic strategies to predict aroma potential of different grape varieties. Journal of Agricultural and Food Chemistry, 57 (6), 2468–2480.
38. 38. Gunata, Z. (2003) Flavor enhancement in fruit juices and derived beverages by exogenous glycosidases and consequences of the use of enzyme preparations, in Handbook of food enzymology (eds Whitaker, J.R., Voragen, A.G.J., Wong, D.W.S.), Marcel Dekker Inc., New York.
39. 39. Skouroumounis, G.K. and Sefton, M.A. (2000) Acid‐catalyzed hydrolysis of alcohols and their β‐D‐glucopyranosides. Journal of Agricultural and Food Chemistry, 48 (6), 2033–2039.
40. 40. Maicas, S. and Mateo, J.J. (2005) Hydrolysis of terpenyl glycosides in grape juice and other fruit juices: a review. Applied Microbiology and Biotechnology, 67 (3), 322–335.
41. 41. Hemingway, K.M., Alston, M.J., Chappell, C.G., Taylor, A.J. (1999) Carbohydrate‐flavour conjugates in wine. Carbohydrate Polymers, 38 (3), 283–286.
42. 42. Tamborra, P., Martino, N., Esti, M. (2004) Laboratory tests on glycosidase preparations in wine. Analytica Chimica Acta, 513 (1), 299–303.
43. 43. Castro Vázquez, L., Pérez‐Coello, M.S., Cabezudo, M.D. (2002) Effects of enzyme treatment and skin extraction on varietal volatiles in Spanish wines made from Chardonnay, Muscat, Airén, and Macabeo grapes. Analytica Chimica Acta, 458 (1), 39–44.
44. 44. Valcarcel, M.C. and Palacios, V. (2008) Influence of “Novarom G” pectinase β‐glycosidase enzyme on the wine aroma of four white varieties. Food Science and Technology International, 14 (5), 95–102.
45. 45. Sánchez‐Palomo, E., Hidalgo, M.C.D.a.‐M., González‐Viñas, M.Á., Pérez‐Coello, M.S. (2005) Aroma enhancement in wines from different grape varieties using exogenous glycosidases. Food Chemistry, 92 (4), 627–635.
46. 46. Sefton, M.A., Skouroumounis, G.K., Elsey, G.M., Taylor, D.K. (2011) Occurrence, sensory impact, formation, and fate of damascenone in grapes, wines, and other foods and beverages. Journal of Agricultural and Food Chemistry, 59 (18), 9717–9746.
47. 47. McMahon, H., Zoecklein, B.W., Fugelsang, K., Jasinski, Y. (1999) Quantification of glycosidase activities in selected yeasts and lactic acid bacteria. Journal of Industrial Microbiology and Biotechnology, 23 (3), 198–203.
48. 48. Mansfield, A.K., Zoecklein, B.W., Whiton, R.S. (2002) Quantification of glycosidase activity in selected strains of Brettanomyces bruxellensis and Oenococcus oeni. American Journal of Enology and Viticulture, 53 (4), 303–307.
49. 49. Grimaldi, A., Bartowsky, E., Jiranek, V. (2005) A survey of glycosidase activities of commercial wine strains of Oenococcus oeni. International Journal of Food Microbiology, 105 (2), 233–244.
50. 50. Ugliano, M., Bartowsky, E.J., McCarthy, J., et al. (2006) Hydrolysis and transformation of grape glycosidically bound volatile compounds during fermentation with three Saccharomyces yeast strains. Journal of Agricultural and Food Chemistry, 54 (17), 6322–6331.
51. 51. Delcroix, A., Günata, Z., Sapis, J.‐C., et al. (1994) Glycosidase activities of three enological yeast strains during winemaking: effect on the terpenol content of Muscat wine. American Journal of Enology and Viticulture, 45 (3), 291–296.
52. 52. Fernandez‐Gonzalez, M., Di Stefano, R., Briones, A. (2003) Hydrolysis and transformation of terpene glycosides from muscat must by different yeast species. Food Microbiology, 20 (1), 35–41.
53. 53. Loscos, N., Hernandez‐Orte, P., Cacho, J., Ferreira, V. (2007) Release and formation of varietal aroma compounds during alcoholic fermentation from nonfloral grape odorless flavor precursors fractions. Journal of Agricultural and Food Chemistry, 55 (16), 6674–6684.
54. 54. Zoecklein, B.W., Marcy, J.E., Williams, J.M., Jasinski, Y. (1997) Effect of native yeasts and selected strains of Saccharomyces cerevisiae on glycosyl glucose, potential volatile terpenes, and selected aglycones of white Riesling (Vitis vinifera L.) wines. Journal of Food Composition and Analysis, 10 (1), 55–65.
55. 55. Boido, E., Lloret, A., Medina, K., et al. (2002) Effect of β‐glycosidase activity of Oenococcus oeni on the glycosylated flavor precursors of Tannat wine during malolactic fermentation. Journal of Agricultural and Food Chemistry, 50 (8), 2344–2349.
56. 56. Ugliano, M., Genovese, A., Moio, L. (2003) Hydrolysis of wine aroma precursors during malolactic fermentation with four commercial starter cultures of Oenococcus oeni. Journal of Agricultural and Food Chemistry, 51 (17), 5073–5078.
57. 57. Ugliano, M. and Moio, L. (2006) The influence of malolactic fermentation and Oenococcus oeni strain on glycosidic aroma precursors and related volatile compounds of red wine. Journal of the Science of Food and Agriculture, 86 (14), 2468–2476.
58. 58. Hernandez‐Orte, P., Cersosimo, M., Loscos, N., et al. (2009) Aroma development from non‐floral grape precursors by wine lactic acid bacteria. Food Research International, 42 (7), 773–781.
59. 59. Bueno, J., Peinado, R., Medina, M., Moreno, J. (2006) Effect of a short contact time with lees on volatile composition of Airen and Macabeo wines. Biotechnology Letters, 28 (13), 1007–1011.
60. 60. Loscos, N., Hernandez‐Orte, P., Cacho, J., Ferreira, V. (2009) Fate of grape flavor precursors during storage on yeast lees. Journal of Agricultural and Food Chemistry, 57 (12), 5468–5479.
61. 61. Gunata, Y.Z., Bayonove, C.L., Baumes, R.L., Cordonnier, R.E. (1986) Stability of free and bound fractions of some aroma components of grapes cv. Muscat during the wine processing: preliminary results. American Journal of Enology and Viticulture, 37 (2), 112–114.

Notes

1 That is, an aglycone linked by its functional group to a sugar (usually an alcohol, giving an O-glycoside) through a glycosidic bond involving the anomeric hydroxyl. The position of the anomeric hydroxyl can be axial (i.e., makes a large (90°) angle relative to a plane passing closest to the majority of ring atoms) or equatorial (makes a small angle, as seen in Figure 23.1.1); in many cases the β-anomer has the hydroxyl equatorial and the α-anomer has it axial. However, the naming of α- and β-anomers, which is beyond the scope of this chapter, does not simply derive from the anomeric hydroxyl being axial or equatorial in the cyclic form of a sugar.

2 A complete account of all economically important plant-derived glycosides is well outside the scope of this book, but such a list would include salicin from willow bark, which is a glucoside of salicylic acid and a close relative of aspirin; indican, the glucoside of indigo dye; and naringin, a key bitter component of grapefruit juice.

3  A rhamnosyl‐glucoside is commonly called a rutinoside and a glucosyl‐glucoside can be termed a gentiobioside (or a sophoroside if the linkage is 2‐O‐ rather than 6‐O‐).

4  Conjugation can occur with either retention of configuration at the anomeric position of the sugar (as shown here) or inversion.

5  The SI unit for catalytic activity in mol/s is designated a katal (abbreviated kat); nkat represents nanokatals, which are expressed per mg of enzyme preparation.