Journal of Chromatography A, 881 (2000) 557–567

www.elsevier.com / locate / chroma

Review

Monoterpenes in grape juice and wines

J.J. Mateo*, M. Jimenez
´
Department of Microbiology and Ecology, University of Valencia, Dr. Moliner 50, E-46100 -Burjasot, Valencia, Spain

Abstract

The importance of monoterpenes on varietal flavour of wines has been reviewed. These compounds were mainly found
linked to sugar moieties in the grape juice and wines, showing no olfactive characteristics. In this way, mechanisms to
liberate terpenes were studied, making a comparative study between acidic and enzymic hydrolysis of terpene glycosides.
Finally, analytical techniques developed to study these compounds, in both free or glycosidically forms, and also to
fractionate glycosidic precursors, have been discussed.  2000 Elsevier Science B.V. All rights reserved.

Keywords: Reviews; Wine; Fruit juices; Food analysis; Monoterpenes; Terpenes; Glycosides

Contents

1. Introduction ............................................................................................................................................................................ 557

1.1. Classification of grape varieties........................................................................................................................................ 558
1.2. Classification of monoterpenes......................................................................................................................................... 558
1.3. Quantitation of monoterpenes .......................................................................................................................................... 558
2. Glycoside hydrolysis ............................................................................................................................................................... 560
2.1. Acidic hydrolysis ............................................................................................................................................................ 561
2.2. Enzymic hydrolysis......................................................................................................................................................... 561
2.2.1. Grape glycosidases .............................................................................................................................................. 561
2.2.2. Yeast glycosidases............................................................................................................................................... 561
2.2.3. Exogenous (fungal) glycosidases.......................................................................................................................... 562
3. Methodology .......................................................................................................................................................................... 563
3.1. Isolation of glycosides..................................................................................................................................................... 563
3.2. Analytical determination of monoterpenes ........................................................................................................................ 564
3.3. Glycoside fractionation.................................................................................................................................................... 565
4. Conclusion ............................................................................................................................................................................. 566
References .................................................................................................................................................................................. 566

1. Introduction

It has emerged from numerous studies that the

*Corresponding author. Tel.: 134-96-398-3145; fax: 134-96- terpenoid compounds form the axis for the sensory
398-3099. expression of the wine bouquet which is typical of its
E-mail address: [email protected] (J.J. Mateo) variety and that they can therefore be used ana-

0021-9673 / 00 / $ – see front matter  2000 Elsevier Science B.V. All rights reserved.
PII: S0021-9673( 99 )01342-4
558 ´
J. J. Mateo, M. Jimenez / J. Chromatogr. A 881 (2000) 557 – 567

lytically for varietal characterization. Apart from the viticultural region. The classification, however, indi-
hitherto known compounds (terpene ethers, mono- cates those cultivars for which analysis for mono-
terpene alcohols) [1–5] numerous monoterpene com- terpenes are likely to give data which could be useful
pounds, in particular monoterpene diols, in grape in an investigation of fruit vatietal character, i.e.,
must and wine, has been identified. At present, about those listed in groups (1) and (2). In the cultivars
50 monoterpene compounds are known of which the listed under (3), monoterpenes are at such low
most important are shown in Fig. 1. The dominating concentration that these compounds could only play
monoterpene alcohols, particularly from Muscat a minimal role in determining varietal flavour. It is
varieties, are linalool, geraniol, nerol, citronellol, from among the more numerous cultivars of group
a-terpineol. In the case of koshu, an indigenous (3) that a large volume of the world’s wine is
Japanese variety, terpinene-4-ol was identified as the produced.
dominating monoterpene compound [6]. The flavour
threshold of nerol and a-terpineol is three to four 1.2. Classification of monoterpenes
times higher than that of linalool (100 mg / l). The
linalool oxides have flavour thresholds of 3000–5000 Three types of categories of monoterpenes exist in
mg / l [7]. grapes with some interrelationships between the
Terpene compounds belong to the secondary plant categories.
constituents, of which the biosynthesis begins with On the top of the complex are the free aroma
acetyl-coenzyme A (CoA) [8]. Microorganisms are compounds, commonly dominated by linalool, ge-
also able to synthesize terpene compounds [9] but raniol, and nerol, together with the pyran and furan
the formation of terpenes by Saccharomyces cere- forms of the linalool oxides. However, depending on
visiae has not yet been observed. Terpenes are not how the juice has been treated and on factors, which
changed by the yeasts metabolism during fermen- may include climate, many additional monoterpenes
tation [7]. Several authors have shown that terpenes can be found in this group, i.e. citronellol, a-ter-
play a significant role in the varietal flavour of wines pineol, ho-trienol, nerol oxide, myrcenol, the
and, in the berries, they are located in skin and ocimenols plus several other oxides, aldehydes and
linked to sugars [10–12]. hydrocarbons. In wines, several monoterpene ethyl
ethers and acetate esters have also been found among
1.1. Classification of grape varieties the free aroma compounds.
Second, there are the polyhydroxylated forms of
A number of surveys have been made of mono- the monoterpenes, or free odourless polyols. A most
terpene concentration in different grape varieties significant feature of the polyols is that, although
[11,13–16]. However, since the reported quantitative these compounds make no direct contribution to the
data were obtained by different techniques and from aroma, some of them are reactive and can break
samples of fruit from diverse areas, direct com- down with great ease to give pleasant and potent
parison on the analytical figures from different volatiles, i.e. diendiol (3,7-dimethylocta-1,5-diene-
surveys is not feasible. Nevertheless, a general 3,7-diol) can give ho-trienol and nerol oxide [5].
classification of those varieties which have been Third, there are the glycosidically conjugated
screened is possible allowing division into (1) in- forms of the monoterpenes which also make no
tensely flavoured muscats, in which total free mono- direct contribution to the aroma of the grape. Glyco-
terpene concentrations can be as high as 6 mg / l; (2) sides are, in most cases, more abundant than the
non-muscat but aromatic varieties with total mono- unglycosilated forms of individual monoterpenes and
terpene concentration of 1–4 mg / l; and (3) more polyols.
neutral varieties not dependent upon monoterpenes
for their flavour (Table 1). 1.3. Quantitation of monoterpenes
Data for many of these assigments were made
from a limited number of samples (often only one) The occurrence of terpene glycoside precursors in
and, in most cases, the fruit was taken from a single grapes was better demonstrated. Glycoside precur-
 ´
J. J. Mateo, M. Jimenez / J. Chromatogr. A 881 (2000) 557 – 567 559

Fig. 1. Main monoterpene compounds in grape juice and wines. Numbers correspond to Table 2.
560 ´
J. J. Mateo, M. Jimenez / J. Chromatogr. A 881 (2000) 557 – 567

Table 1
Classification of some grape varieties based on monoterpene content
(1) (2) (3)
Muscat varieties Non-muscat aromatic Neutral varieties
varieties
Canada Muscat Traminer Aryan
Gewurztraminer Huxel Bacchus
Muscat of Alexandria Kerner Bobal
Muscat de Frontignan Morio-Muskat Cabernet-Sauvignon
Muscato Bianco del Piemonte ¨
Muller-Thurgau Carignan
Muscat Hamburg Riesling Cencibel
Muscat Ottonel Schurebe Chardonnay
Moscato italiano Schonburger Chasselas
Siegerebe Chenin Blanc
Sylvaner Cinsault
Wurzer Clairette
Dattier de Bevrouth
Doradillo
Forta
Merlot
Nobling
Rkaziteli
¨
Rulander
Sauvignon Blanc
Semillon
Shiraz
Sultana
Terret
Trebbiano
Verdelho
Viognier

sors are numerous and fairly abundant as, in flavour- flavourant aglycons (i.e., with the same state of
ant grapes, they are evaluated between 6.5 and 28 oxidation than linalool) the most abundant are
mg / l juice. Most grape varieties contain free and apiosylglycosides (up to 50% according to grape
bound glycoside terpenes but concentrations are variety) followed by rutinosides (6–13%) and finally
higher in flavourant cultivars. In general, bound glucosides (4–9%). A more accurate analysis indi-
glycosides forms are more abundant than the free cates that all glycosides are not present in all
ones [11,16]. cultivars and that their proportions also differ accord-
The sugar moieties were identified to rutinose ing to grapes [23].
(6-O-a-L-rhamnopyranosyl-b-D-glucopyranoside), 6- The glycoside flavour potential from grapes re-
O-a-L-arabinofuranosyl-b-D-glucopyranosides, 6-O- main unfortunately quite stable during winemaking
b-D-apiofuranosyl-b-D-glucopyranosides or b-D- and in young wines as well.
glucopyranosides [17–19]. The aglycon part is often
formed with terpenols, but linalool oxides, terpene
diols and triols can also been found. However, other 2. Glycoside hydrolysis
flavour precursors can occur such as linear or cyclic
alcohols, e.g. hexanol, phenylethanol, benzyl alcohol, Terpene glycosides can be hydrolysed by acids
C 13 norisoprenoids, phenolic acids and probably [24,25] or enzymes [10,11,17]; under the former
volatile phenols such as vanillin [20–22]. conditions, rearrangements of the monoterpenol may
If we consider only glycosides with the most occur, whereas under the latter conditions the
 ´
J. J. Mateo, M. Jimenez / J. Chromatogr. A 881 (2000) 557 – 567 561

changes in the natural monoterpenol distribution are as 3-oxo-a-ionol and 3-hydroxy-b-damascenona

minimal. [31]. These compounds are totally glycosilated in the
grape and, opposite with terpenes, they are found in
2.1. Acidic hydrolysis the same quantities in all the grape varieties, aro-
matics or neutral, and they are capable of awarding
Experiments on both whole juice and monoterpene certain typicity to the wine flavour because they have
glycosides isolated from juice [26,27] have demon- lower threshold values than terpenes and they con-
strated that significantly different patterns of volatile tribute characteristic aromatic features [32].
monoterpenes are produced when each is hydrolysed
at different pH values. Furthermore, there appears to 2.2.1. Grape glycosidases
be a pH-dependent interrelationship between several The potential use of enzymic systems of grapes to
of the grape monoterpenes. Thus e.g. the isomerics liberate terpenes from grape juices or wines has been
ocimenols appear to be formed hydrolytically in always the subject of different works regarding the
juice at pH 1 at the expense of linalool, nerol and enzymic hydrolysis of terpene glycosides.
geraniol, the last three compounds being pH 3 Grapes have an enzyme with b-glucosidase activi-
products [24]. ty (Table 2) [31,33–37] but only low a-rhamnosid-
For acidic hydrolysis, samples, with no free ase, a-arabinosidase or b-apiosidase activities have
volatile compound, were dissolved in a buffer at been detected [31]. On the other hand, grape b-
acidic pH and solutions were heated and resulting glucosidase was not quite stable and showed low
free volatile compounds extracted and analysed by activity at grape juice or wines pH values [38].
GC [24,25,27–29]. Similarly, certain b-glucosidasae activity has been
Acidic hydrolysis of terpene glycosides can observed in grape leaves [35–37].
provoke a molecular rearrangement of the mono- In general, b-glucosidases with a vegetal origin
terpenols, which were transformed in other com- show a low activity on monoglucosides of terpenes
pounds. Nevertheless, this way to liberate terpenes with a tertiary alcohol group (linalool, a-terpineol),
simulate the reactions which takes place during and they are only capable to hydrolyse mono-
ageing of wines and the different terpenic alcohols glucosides of terpenes with a primary alcohol group
were produced in similar quantitative rations. (geraniol, nerol, citronellol) [34,36,39].
Recently, some evidence has been obtained on the
2.2. Enzymic hydrolysis presence of an endoglycosidase in grape skins able to
split the heterosidic linkage of disacharide glycosides
To enrich wine flavour by release of free aromatic releasing disaccharide and aglycon; the enzyme was
compounds from natural glycoside precursors, par- quite tolerant to glucose inhibition [40].
ticularly pathways are required. Enzymic hydrolysis
was most interesting because it produces a more 2.2.2. Yeast glycosidases
‘‘natural’’ flavour in the wine [10,11,29]. Enzymic Regarding yeast b-glucosidases, yeasts of the
hydrolysis of glycosides is carried out with various Hansenula species isolated from fermenting must
enzymes which act sequentially according to two were reported to have an inducible b-glucosidase
steps: firstly, a-L-rhamnosidase, a-L-arabinosidase or activity, but this enzyme was inhibited by glucose
b-D-apiosidase make the cleavage of the terminal [41]. Among other yeast strains, b-glucosidases from
sugar and rhamnose, arabinose or apiose and the Candida molischiana [42] and C. wickerhamii [43]
corresponding b-D-glucosides are released; sub- also possess activities towards various b-glucosides
sequently liberation of monoterpenol takes place and they were little influenced by the nature of
after action of a b-D-glucosidase [30]. aglycon [44].
Enzymic hydrolysis of glycoside extracts from Finally, the situation regarding Saccharomyces
Muscat, Riesling, Semillon, Chardonnay, Sauvignon cerevisiae is more complex because this yeast is
and Sirah varieties have provoked the liberation not capable of modifying the terpenic profile of the wine;
only of terpenes, but also C 13 norisoprenoids, such so, it can produce citronellol from geraniol and nerol,
562 ´
J. J. Mateo, M. Jimenez / J. Chromatogr. A 881 (2000) 557 – 567

Table 2
Results obtained by hydrolysis of terpene glycosides treated with different enzymatic preparations. Data are normalized to 100 in untreated
wines
No. Terpenes Grape skin S. cerevisiae Exogenous glycosidases
glucosidase
Klerzyme 200 Pectinol C Rohapect C Hemicellulase
Sweet wine Dry wine
1 trans-Furan linalool oxide 108.2 122.9 247.3 174.3 943.1 109.7 129.7
2 cis-Furan linalool oxide 124.3 296.8 149.2 133.8 456.1 87.0 117.6
3 Linalool 101.2 204.6 113.6 111.0 187.8 107.5 108.8
4 Ho trienol 93.3 113.2 130.8 87.1
5 Neral
6 a-Terpineol 230.0 150.8 190.8 113.8 141.7
7 Geranial
8 trans-Pyran linalool oxide 124.9 109.3 109.1 113.1 104.2 104.4
9 cis-Pyran linalool oxide 104.5 88.9 136.3 111.4 98.0
10 Citronellol 177.6 176.6 551.9 239.8 131.8 146.5 335.4
11 Nerol 593.0 847.6 424.5 720.0 1173.2 190.6 642.0
12 Geraniol 381.9 833.4 491.8 633.2 806.2 107.0 359.1
13 Diol I 104.2 91.4 151.9 112.7 152.3 113.8 98.0
14 Endiol 133.0
15 Diol II 101.4 12.5 174.7 114.0 124.3 100.1 97.6
16 Hydroxy-cityronellol 152.6 159.0 178.4 552.9
17 8-Hydroxydihydrolinalool 275.9 230.8 152.9 397.6 105.9 93.2
18 Hydroxynerol 141.4
19 trans-8-Hydroxylinalool 133.4 466.0 1343.0 431.5
20 Hydroxygeraniol 140.2 475.0
21 cis-8-Hydroxylinalool 271.8 688.9 4083.0 990.0 916.9
22 Geranic acid 362.6 522.8 410.2
23 Triol

the intensity of this transformation depends on the by changing the composition of the medium includ-
yeast strain used [45]. Other authors propose a more ing inductive compounds, as well as in filamentous
complex scheme: geraniol was transformed by these fungi [48,49].
yeasts into geranyl acetate, citronellyl acetate and
citronellol, while nerol was transformed into neryl 2.2.3. Exogenous ( fungal) glycosidases
acetate; in addition, geraniol was transformed into Taking into account that enzymic systems of
linalool and nerol was cyclized to a-terpineol at must grapes are not suitable to hydrolyse terpene glyco-
pH [46]. sides in grape juice or wine and that more studies are
Few data are available regarding glycosidase needed regarding the ability of S. cerevisiae to
activities of enological yeast strains and the tech- produce all the enzymes which take part in this
nological properties of the enzymes. Last published process, several exogenous enzymes, mainly with a
results [47] have detected b-glucosidase activity in fungal origin, have been developed to liberate ter-
different Saccahromyces strains on the basis of its penes in wines.
hydrolytic activity on para-nitrophenyl-b-D-gluco- The most suitable enzymic preparations to be used
side (pNPG) and terpene glucosides of Muscat juice during winemaking process are those which possess
(Table 2); it is quite glucose independent but is all b-D-glucopyranosidase, a-L-arabinofuranosidase,
inhibited by ethanol. These results could open new a-L-rhamnopyranosidase and b-D-apiofuranosidase
pathways regarding other glycosidase activities in S. activities (Table 2) [39]. These enzymes are present
cerevisiae; a-rhamnosidase, a-arabinosidase or b- only in low quantities in the majority of commercial
apiosidase activities could be induced in wine yeast fungal enzymic preparations, mainly regarding b-D-
 ´
J. J. Mateo, M. Jimenez / J. Chromatogr. A 881 (2000) 557 – 567 563

apiosidase activity [50]. Determination of free vola- solvent extraction followed by silica gel chromatog-
tile compound terpenols, norisoprenoids and volatile raphy and preparative thin layer chromatography.
phenols indicate that concentration in enzyme treated Subsequently, Banthorpe and Mann [61] applied the
wines highly increase, not only in aromatic varieties, same technique to petals of Tanacetum vulgare.
but also in neutral ones, from 265 to 2000% (Table Similarly other authors [62–64] extract plant materi-
2) [31]. al directly with organic solvents, after clean-up of
The ability of purified enzymes, synthetic sub- the extracts on silica gel, conventional chromato-
strates and suitable analytical methods are needed for graphic procedures were used to isolate b-D-gluco-
the rigourous progress in the field of enzymic sides and derivatives of several terpenoids. Unfor-
hydrolysis of terpenyl glycosides [11,19,30,51–54]. tunately, these techniques were either inappropriate
Enzymes have been isolated from fungal enzymic or found to be unsuitable for the isolation of
preparations, vegetal extracts or synthetic culture monoterpene glycosides from grape juice and wines.
media inoculated with fungal cultures (mainly Asper- In the case of grape juice, the presence in aqueous
gillus niger), and have been purified by different solution of large amounts of glucose and fructose,
chromatographic techniques (gel filtration, ion-ex- together with other free sugars, makes the problem
change chromatography, affinity chromatography, of small amounts of glycosides extremely difficult.
chromatofocusing) [31,49,55–57]. On the other hand, the isolation from wines in which
The method used to actually enrich wine in most of the sugars have been removed by fermen-
volatile compounds is only effective when dry wines tation is also complicated by the presence of glycerol
are used, because b-glucosidase in fungal enzymic and the isomeric 2,3-butanodiols. Several techniques
preparations is hardly inhibited by glucose; so, the were tried unsuccessfully, including solvent extrac-
hydrolysis of terpene glycosides is not completed in tion and subsequent TLC, gel filtration of both juice
sweet wines obtained from Muscat de Frontignan and wines and ion-exchange chromatography of juice
grapes (Table 2) [31]. on a column of cation exchanger in the calcium form
When enzymic preparations have been used to [65] and on an anion exchanger in the bisulphite
improve the aromatic characteristics of the wines, form [66,67].
undesirable odours have been sometimes detected,
even if liberation of terpenes has been produced, 3.1. Isolation of glycosides
showing high concentrations in vinyl-phenols (4-
vinyl-phenol, 4-vinyl-guayacol) [50]. The isolation of glycosides from wines and juices
On the other hand, fungal enzyme preparations by selective retention of the compounds on a solid-
generate oxidation artefacts during the hydrolysis of phase adsorbent is a commonly used technique. By
glycosides [58]; this fact implies that, perhaps, the washing the adsorbent with water following the
review of all the information regarding the use of adsorption step, free sugars and other polar con-
these enzymic preparations will be necessary. stituents can be removed while the less polar glyco-
sides are retained. Elution with an organic solvent
then give the glycosidic fraction.
3. Methodology Different methodologies have been proposed to
extract glycoside precursors from grape juices and
Glycosides of monoterpenes have been recognised wines.
as plant constituents and a variety of techniques have Williams et al. [51] used glass column chromatog-
been used to isolate these hydrophylic compounds. raphy containing C 18 reversed-phase adsorbent to
Croteau and Martinkus [59], after lyophilizing aque- extract glycosides from juice or dealcoholised wine.
ous leaf extracts, used TLC, ion-exchange chroma- After washing with water and eluting free com-
tography and gel permeation chromatography to pounds with 20% aqueous acetic acid, precursors
purify neomethyl-b-D-glucoside. Francis and Allcock were eluted in two fractions with 30% aqueous acetic
[60] isolated b-D-glucosides of geraniol, nerol and acid and methanol. A modification of the methodolo-
citronellol from aqueous extracts of rose petals by gy has been proposed by using 1 g solid-phase
564 ´
J. J. Mateo, M. Jimenez / J. Chromatogr. A 881 (2000) 557 – 567

extraction C 18 cartridges; hydrophylic compounds from must or wine, be they acidic [16,24], with the
were eluted with water, free terpenes with dichloro- well-known drawback of terpene rearrangements, or
methane and glycosides with methanol [26]. This enzymic [74,75] are all based on the analysis of
method has been improved in latter years, but it has monoterpenols released [76]. Since the monoter-
a disadvantage because separation is different de- penols must be first isolated from the hydrolysed
pending on the commercial origin of the cartridges sample, such methods require at least two steps
[68]. preceding the final analysis by GC (Fig. 2). Further-
A second approach to the problem has been more, they elicit no information about the glycosidic
proposed by Montpellier researchers by using Am- conjugation of each monoterpenol, allowing only
berlite XAD-2 resin because it possesses an excellent gross characterization of the bound aroma fraction.
capacity for adsorption of free terpenols from grape In order to achieve a method suitable for the direct
juice [11]. This resin had been previously used to and individual analysis of momoterpenyl glycosides,
isolate naringin and limonin from grape juices [69]. both GC and high-performance liquid chromatog-
Once wine has been eluted through resin, free raphy (HPLC) have been investigated.
compounds were eluted with pentane and glycosides HPLC is a priori a well suited technique as the
with ethyl acetate. Amberlite XAD-2 resin displays compounds of interest are not volatile. Liquid chro-
extraction capacities similar to those of coated matography of monoterpenyl glycosides has already
octadecyl silica. Furthermore, it has the advantage of been considered for extraction and sample clean-up
being sold in large particle sizes; it is thus possible to purposes, using adsorption on silica gel [60,64] Also,
use it in a wide preparative column at atmospheric reversed-phase chromatography on C 18 bonded silica
pressure. A modification of this method has been [51,76] and apolar resin copolymer [11] as well as
suggested and free compounds were eluted with combined gel permeation and hydrophobic inter-
pentane:dichloromethane to improve extraction [70]. action chromatography [77] have been investigated
Nevertheless, even with extensive washing of the for the same purpose. HPLC has also been used for
adsorbent beds both Amberlite XAD-2 and Amber- the chromatography of heavier glycosides, e.g. with
lite XAD-16 (the Amberlite XAD-16 resin has a diterpenyl [78,79] or triterpenyl [80] aglycones.
surface area and capacity greater than that of Amber- Among all the liquid chromatography methods
lite XAD-2) retained free glucose. Both of these studied, the best involved a reversed-phase HPLC
adsorbents had the disadvantage of retaining free system although good results were also obtained by
glucose in addition to the adsorbed glycosides [71]. using another apolar packing, a polystyrene–di-
Reversed-phase silica gel has been found to be a vinylbenzene copolymer [81].
particularly suitable adsorbent for the isolation of GC is recognised as a highly efficient and resolu-
glycosidic terpenes [72]. The commercial availability tive separation technique but it can only be applied
of this adsorbent in uniform, prepacked cartridges to volatile derivatives of monoterpenyl glycosides.
was an additional advantage in development of the Comparing both techniques, similar results are ob-
isolation step. tained by HPLC techniques and by GC of trimethyl-
More recently, a method has been proposed to silyl derivatives of the monoterpenyl glycosides [52].
extract glycosilated flavour precursors from grape So, when severe chromatographic conditions are not
juices by microwaves [73]. This method was com- considered as restrictive, GC turns out to be a good
pared to the method using a column of Amberlite alternative to HPLC; however, as glycosides are not
XAD-2 resin. Although a further purification of the volatile, an additional derivatisation step is required.
extracts was required, this method provided the The use of GC–MS for determination of synthetic
advantages of rapidity, ease of extraction and possi- glycosides can be a way to establish satisfactory
bility of extract berries without deseeding or crush- conditions for the separation and identification of
ing. monoterpenyl glycosides, to facilitate detection of
new glycosides through identification of characteris-
3.2. Analytical determination of monoterpenes tic fragmentation patterns and to determine indi-
vidual glycosides and their aglycones [53]. Trifluoro-
Investigations of the precursor compounds isolated acetylation derivatization proved to be more suitable
 ´
J. J. Mateo, M. Jimenez / J. Chromatogr. A 881 (2000) 557 – 567 565

Fig. 2. GC–MS chromatogram of terpenes obtained by enzymic hydrolysis of glycosides. Peak numbers correspond to Table 2.

for the qualitative and quantitative analysis of mono- solvent and subjected to another purification or
terpenyl diglycosides but trimethylsylilation, as well hydrolysis.
as HPLC methods, provided complementary results The critical points of this method are: (a) the
[52,81]. incomplete elimination of volatile components dur-
ing the evaporation of wine alcohol under vacuum;
3.3. Glycoside fractionation (b) the partial recuperation of the precursors with an
acidic solvent, with the opportunity of hydrolytic
Williams et al. [51], laid stress on the existence, in adulterations and chemical transformations during
the aromatic varieties of grapes, of two classes of the solvent evaporation.
precursors of C 10 and C 13 components with different Besides, Strauss et al. [82] fractionated by doplet
polarity, separable by chromatography on C 18 car- counter-current chromatography (DCCC) the C 10 ,
tridges. The method described from these authors C 13 and aromatic ring precursors in must previous
about the separation of two glycosides classes about isolation on C 18 RP column, and Bitteur et al. [52]
the absorption of dealcoholized wine on C 18 car- described an analytical HPLC method with C 18
tridges of adeguated size to the wine volume used, column to fractionate aroma precursors, using ace-
the elimination of hydrophylic components with tonitrile and water as solvents.
water, the elution of more polar precursors with 30% Following developments of chromatographic tech-
acetic acid and the elution of monohydroxilated niques in totally liquid phase [83] led to the frac-
terpenic alcohols with methanol. The components in tionation of the precursors in leaves of Renan
both fractions were recuperated by evaporation of Riesling by using multilayer coil counter-current
566 ´
J. J. Mateo, M. Jimenez / J. Chromatogr. A 881 (2000) 557 – 567

chromatography (MLCCC). In this case, the pre- [3] A. Rapp, H. Mandery, M. Gunter, ¨ ¨
in: L. Nykanen, P.
Lethonen (Eds.), Flavour Research of Alcoholic Beverages
cursors were previously isolated on Amberlite XAD-
— Instrumental and Sensory Analysis, Kauppakirjapino,
2 resin column. This technique has been used to Helsinki, 1984, p. 255.
isolate two novel terpenoid glucose esters from [4] A. Rapp, H. Mandery, Experientia 42 (1986) 857.
Riesling wines [84]. [5] P.J. Williams, C.R. Strauss, B. Wilson, J. Agric. Food Chem.
A separation system for volatile precursors using 28 (1980) 766.
[6] J. Shimizu, M. Watanabe, Agric. Biol. Chem. 45 (1981)
commercial C 18 RP columns and inert solvents has 2797.
been recently performed. The glycosides of mono, di [7] A. Rapp, in: H.F. Linskens, J.F. Jackson (Eds.), Wine
and trihydroxilated terpene and norisoprenoid al- Analysis, Springer-Verlag, Berlin, 1988, p. 28.
cohols as well the related shikimate pathway ones [8] P. Manitto, Byosynthesis of Natural Products, Ellis Horwood,
Chichester, 1980.
have been isolated on C 18 RP cartridges and frac- [9] R. Hock, J. Benda, P. Schreier, Z. Lebensm. Unters. Forsch.
tionated in classes having different polarity with 179 (1984) 450.
eluents at increasing percentages of methanol. The [10] R. Cordonnier, C. Bayonove, C.R. Acad. Sci. 278 (1974)
benzyl alcohol glycosides appear the most polar, 3387.
[11] Z. Gunata, C. Bayonove, R. Baumes, R. Cordonnier,
while those of terpene monohydroxilated alcohols
J.Chromatogr. 331 (1985) 83.
and of geranic acid the lest polar. The terpene diols, [12] B. Wilson, C.R. Strauss, P.J. Williams, Am. J. Enol. Vitic. 37
linalool furanoid and pyranoid oxides as well (1986) 107.
norisoprenoid precursors show intermediate polarity [13] A. Terrier, J.N. Boidron, P. Ribereau-Gayon, C.R. Acad. Sci.
and place themselves in well fixed fractions accord- 275 (1972) 941.
[14] A. Rapp, W. Knipser, H. Hastrich, L. Engel, in: A.D. Webb
ing their polarity [28].
(Ed.), Grape and Wine Centennial Symposium Proceedings,
University of California, 1982, p. 304.
[15] R. Di Stefano, Vini Italia 23 (1981) 29.
[16] E. Dimitradis, P.J. Williams, Am. J. Enol. Vitic. 35 (1984)
4. Conclusion 66.
[17] P.J. Williams, C.R. Strauss, B. Wilson, P.A. Massy-Westropp,
The presence of terpenes, in their different forms, Phytochemistry 21 (1982) 2013.
in grape juices and wines represents an enormous [18] Z. Gunata, C. Bayonove, R. Baumes, R. Cordonnier, J. Sci.
Food Agric. 36 (1985) 857.
potential in a way to increase the varietal characteris- [19] S. Voirin, R. Baumes, C. Bayonove, O. M’Baibaroua, C.
tics of the wines, contributing the final product with Tapiero, Carbohydr. Res. 207 (1990) 39.
higher fruit-like characteristics. Actually, researchers [20] M.S. Allen, M.J. Lacey, W.V. Brown, R.L.N. Harris, in: P.
have sufficient information and tools to study the Ribereau-Gayon, A. Lonvaud (Eds.), Actualities
presence of terpenes and their evolution in grape Oenologiques 89, Dunod, 1989, p. 25.
[21] C. Bayonove, R. Cordonnier, P. Dubois, C.R. Acad. Sci. 281
juices and wines, but it is not yet possible to translate (1975) 75.
all of the acquired knowledge to the wineries be- [22] W. Schwab, P. Schreier, Phytochemistry 25 (1990) 164.
cause an efficient methodology to improve the [23] C. Bayonove, Y.Z. Gunata, J.C. Sapis, R.L. Baumes, I.
terpene content of all the wines present in the market Dugelay, C. Grassin, Rev. Oenol. 64 (1992) 15.
has not been found. So, much work remains to be [24] P.J. Williams, C.R. Strauss, B. Wilson, R.A. Massy-Wes-
tropp, J. Agric. Food Chem. 30 (1982) 1219.
done to obtain a methodology, probably of enzymic [25] R. Di Stefano, Vigne Vin 9 (1982) 45.
nature, which allows that wine consumer sense of [26] R. Di Stefano, Bull. OIV 721–722 (1991) 219.
smell can, at last, appreciate the whole organoleptic [27] P.J. Williams, C.R. Strauss, B. Wilson, Am. J. Enol. Vitic. 32
richness of the product that he has on his hand or in (1981) 230.
his mouth. [28] J.J. Mateo, N. Gentilini, T. Huerta, M. Jimenez, R. Di
Stefano, J. Chromatogr. A 778 (1997) 219.
[29] R. Di Stefano, personal communication.
[30] Y.Z. Gunata, S. Bitteur, J.M. Brillouet, C.L. Bayonove, R.E.
Cordonnier, Carbohydr. Res. 134 (1988) 139.
References [31] Y.Z. Gunata, I. Dugelay, J.C. Sapis, R. Baumes, C.
Bayonove, J. Int. Sci. Vigne Vin 24 (1990) 133.
[1] A. Rapp, W. Knipser, Vitis 18 (1979) 229. [32] A. Razungles, C.L. Bayonove, R.E. Cordonnier, R.L.
[2] A. Rapp, H. Mandery, H. Ullemeyer, Vitis 23 (1984) 84. Baumes, Vitis 26 (1987) 183.
 ´
J. J. Mateo, M. Jimenez / J. Chromatogr. A 881 (2000) 557 – 567 567

[33] R. Di Stefano, Riv. Vitic. Enol. 2 (1989) 11. [58] M.A. Sefton, P.J. Williams, J. Agric. Food Chem. 39 (1991)
[34] A.P. Aryan, B. Wilson, C.R. Strauss, P.J. Williams, Am. J. 1994.
Enol. Vitic. 38 (1987) 182. [59] R. Croteau, C. Martinkus, Plant Physiol. 64 (1979) 169.
[35] C. Biron, C. Cordonnier, O. Glory, Z. Gunata, J.C. Sapis, [60] M.J.O. Francis, C. Allock, Phytochemistry 8 (1969) 1339.
Conn. Vigne Vin 22 (1988) 125. [61] D.V. Banthorpe, J. Mann, Phytochemistry 11 (1972) 2589.
[36] Y.Z. Gunata, C.L. Bayonove, C. Tapiero, R.E. Cordonnier, J. [62] I. Sakata, T. Mitsui, Agric. Biol. Chem. 39 (1975) 1329.
Agric. Food Chem. 38 (1990) 1232. [63] F. Bohlmann, M. Grenz, Phytochemistry 16 (1977) 1057.
[37] R. Di Stefano, E. Garcia, Riv. Vitic. Enol. 48 (1995) 53. [64] R. Tschesche, F. Ciper, E. Breitmaier, Chem. Ver. 110 (1977)
[38] M. Lecas, Z.Y. Gunata, J.C. Sapis, C.L. Bayonove, Phtoch- 3111.
emistry 30 (1991) 454. [65] S.J. Angyal, G.S. Bethell, R.J. Beveridge, Carbohydr. Res.
[39] R.E. Cordonnier, Y.Z. Gunata, R.L. Baumes, C.L. Bayonove, 73 (1979) 9.
Conn. Vigne Vin 23 (1989) 7. [66] B. Lindberg, K.N. Slessor, Carbohydr. Res. 5 (1967) 286.
[40] Z. Gunata, C. Blondeel, M.J. Vallier, J.P. Lepoutre, J.C. [67] L. Larsson, O. Samuelson, Acta Chem. Scand. 19 (1965)
Sapis, N. Watanabe, J. Agric. Food Chem. 46 (1988) 2748. 1357.
[41] M. Grossman, A. Rapp, W. Rieth, Dtsch. Lebensm. Rdsch. [68] R. Di Stefano, personal communication.
83 (1987) 7. [69] R.L. Johnson, B.V. Chandler, J. Sci. Food Agric. 33 (1982)
[42] P. Gonde, R. Ratomahenina, A. Arnaud, P. Galzy, Can. J. 287.
Biochem. Cell. Biol. 63 (1985) 1160. [70] R. Di Stefano, Vitic. Enol. Sci. 44 (1989) 158.
[43] M. Leclerc, P. Gonde, A. Arnaud, R. Ratomahenina, P. [71] P.J. Williams, W. Cynkar, I.L. Francis, J.D. Gray, P.G. Iland,
Galzy, J. Gen. Appl. Microbiol. 330 (1984) 509. B.G. Coombe, J. Agric. Food Chem. 43 (1995) 121.
[44] Z.Y. Gunata, C.L. Bayonove, R.E. Cordonnier, A. Arnaud, P. [72] P.J. Williams, in: T.E. Acree, R. Teranishi (Eds.), Flavor
Galzy, J. Sci. Food Agric. 50 (1990) 499. Science Sensible Principles and Techniques, American
[45] I. Dugelay, Z. Gunata, J.C. Sapis, R. Baumes, C. Bayonove, Chemical Society, Washington, DC, 1993, p. 287.
J. Int. Sci. Vigne Vin 26 (1992) 177. [73] S. Bureau, A. Razungles, R. Baumes, C. Bayonove, J. Food
[46] R. Di Stefano, G. Maggiorotto, S. Gianotti, Riv. Vitic. Enol. Sci. 61 (1996) 557.
42 (1992) 43. ¨
[74] A. Rapp, M. Gunter, H. Ullemeyer, Z. Lebensm. Unters.
[47] J.J. Mateo, R. Di Stefano, Food Microbiol. 14 (1997) 583. Forsch. 180 (1985) 109.
[48] O. Shoseyov, B.A. Bravdo, R. Ikan, I. Chet, Phytochemistry [75] B. Wilson, C.R. Strauss, P.J. Williams, J. Agric. Food Chem.
27 (1988) 1973. 32 (1984) 919.
[49] I. Dupin, Z. Gunata, J.C. Sapis, C. Bayonove, J. Agric. Food [76] K.H. Engel, R. Tressl, J. Agric. Food Chem. 31 (1983) 998.
Chem. 40 (1992) 1886. [77] R. Croteau, S. El-Hindawi, H. El-Bialy, Anal. Chem. 137
[50] Z. Gunata, Rev. Oenol. 74 (1994) 22. (1984) 389.
[51] P.J. Williams, C.R. Strauss, B. Wilson, P.A. Massy-Westropp, [78] M.S. Ahmed, R.H. Dobberstein, J. Chromatogr. 245 (1982)
J. Chromatogr. 235 (1982) 471. 373.
[52] S. Bitteur, Z. Gunata, J.M. Brillouet, C. Bayonove, R. [79] H.C. Makapugay, N.P.D. Nanayakkara, A.D. Kinghorn, J.
Cordonnier, J. Sci. Food Agric. 47 (1989) 341. Chromatogr. 283 (1984) 390.
[53] S.G. Voirin, R.L. Baumes, Z.Y. Gunata, S.M. Bitteur, C. [80] F. Orsini, L. Verotta, J. Chromatogr. 349 (1985) 69.
Bayonove, J. Chromatogr. 590 (1992) 313. [81] S. Bitteur, R. Roset, J. Chromatogr. 394 (1987) 279.
[54] G.K. Skouroumounis, R.A. Massy-Westropp, M.A. Sefton, [82] C.R. Strauss, P.R. Gooley, B. Wilson, P.J. Williams, J. Agric.
P.J. Williams, J. Agric. Food Chem. 43 (1995) 974. Food Chem. 35 (1987) 519.
[55] Z. Gunata, J.M. Brillouet, S. Voirin, R. Baumes, R. Cordon- [83] G.K. Skouroumounis, P. Winterhalter, J. Agric. Food Chem.
nier, J. Agric. Food Chem. 38 (1990) 772. 42 (1994) 1066.
[56] G. Spagna, F. Andreani, E. Salatelli, D. Romagnoli, P.G. [84] B. Bonnlander, B. Baderschnider, M. Messerer, P. Winterhal-
Pifferi, Proc. Biochem. 33 (1998) 57. ter, J. Agric. Food Chem. 46 (1998) 1474.
[57] Z. Gunata, I. Dugelay, M.J. Vallier, J.C. Sapis, C. Bayonove,
Enz. Microb. Technol. 21 (1997) 39.