Introduction

The aroma of a wine is influenced by the action of several different compounds on the sensory organs. These compounds are produced through metabolic pathways during the ripening and harvesting of grapes and during the fermentation and storage of wine (Rapp and Mandery, 1986; Rapp and Versini, 1995). Of these factors, it is well known that the grape varieties gives characteristic aromas. This varietal aroma allows the classification and even the labelling of wines, so it is important to know the typical aromatic pattern of a wine variety in order to ensure its quality (Marais, 1983).

Verdicchio is a white grape variety grown almost exclusively in the Marche region in central Italy. Genetic analysis has shown a very close relationship between Verdicchio, “Trebbiano di Soave” and “Trebbiano di Lugana or Turbiana” (Costacurta et al., 2003). The genetic origin of Verdicchio and therefore of Trebbiano di Soave, is however still being debated. One hypothesis, based on the documented 15th century migration of farmers from the province of Verona in northern Italy to the Marche, partially depopulated by the plague, is now the most accredited. The settlers are thought to have brought their vines with them and thus it is surmised that they originally came from the Veneto region. In a recent work (Ghidoni et al., 2010) it was reported that the three varieties are genetically identical, at least as far as the analysed part of the genome is concerned, but they could be traced to three different biotypes of the same variety, in relation to phenotypical traits related to the environment. Indeed, the length of the separation would appear to justify some olfactory differences between Trebbiano di Soave and Verdicchio, over and beyond the demonstrated genetic identity (Vantini et al., 2003). Verdicchio is not characterised by a large quantity of terpenes. Indeed, in analysis carried out in the past, a very low content of the main monoterpenes was observed. The floral terpene note could derive from the Ho-terpendiol I (about 20 μg L^-1), also present in bound forms, which could then be released over time (Nicolini et al., 2003). What was instead observed was the presence of a characterising compound, methyl salicylate, which in this variety can reach 45 μg L^-1 in free forms and more than 500 μg L^-1 in bound forms, as reported for Trebbiano di Soave and Verdicchio (Versini et al., 2005). When added to wine, it is described as providing a floral note, slightly balsamic, tending to cinnamon-chestnut honey (Bauer et al., 2008). A study involving Trebbiano di Lugana wines (genetically similar to Trebbiano di Soave and Verdicchio) highlighted that the grapes also contain the precursors of sulphured aromatic compounds, in particular 3-mercaptohexanol and 3-mercapto hexyl acetate, which could impart tropical and citrus fruit scents to the wines (Mattivi et al., 2012). During the tasting of this wine, notes of anise/fennel are sometimes perceived and even balsamic scents in aged wines. One of the objectives of this work was to understand which compounds could contribute to these descriptors. For this reason we used gas chromatography–olfactometry (GC-O), a powerful tool for the study of chemical compounds responsible for wine aroma (van Ruth, 2001).

Materials and methods

Results and discussion

The initial exploration was carried out using four different types of Verdicchio wines (Massaccio, San Sisto, Le Moie and Titulus) by analysing a series of vintages and using SPME with GC×GC-ToF-MS. Two-dimensional analysis made it possible to find many compounds characterising the different types of Verdicchio. Indeed, it is possible to notice clear separation in the PLS-DA (partial least squares discriminant analysis) plot (Figure 1). As shown in the graph, the first component can separate types of Titulus, especially old vintages, from other wines and San Sisto, Massaccio and Le Moie are separated by the second component.

Figure 1. Partial least squares discriminant analysis (PLS-DA) scatterplot of Verdicchio wines from grapes grown in different areas. Eleven vintages of Massaccio (2002–2015), eight vintages of San Sisto (2001–2015), three vintages of Titulus (2013–2015) and three vintages of Le Moie (2008–2015).

Save View full size Expand inline Collapse inline

The score scatter each point represents the vintages (each vintage with three technical replicates).

The volatile compounds determining this separation were whisky lactone, 5-methylfurfural, 2-furancarboxaldehyde, 2,4,5-trimethyl-1,3-dioxolane, benzofuran and 2-methylthiophene for the San Sisto samples. The first three compounds derive from the use of wood, while dioxolane is a compound correlated with ageing. Le Moie wines were characterised by a higher content of norisorenoids such as -damascenone and safranal. Titulus wines were richer in other norisoprenoids, such as TDN and also in some sesquiterpenes such as alpha calacorene and monoterpenes such as alpha-terpineol and alpha terpinolene. Wines from the Massaccio areas were the richest in benzenacetaldehyde and linalool. San Sisto and Massaccio were also the richest in 4-methyl 1-pentanol and 2-acetyl furanone. It is also possible to observe that the typical trend of TDN increases over the years in Massaccio wines (Figure 2). This compound can derive from different precursors linked to sugar molecules and is formed by hydrolysis during ageing (Versini et al., 2001; Winterhalter and Schreier, 1994; Winterhalter, 1991); pH, temperature and presence of oxygen are important parameters in the formation and degradation of TND (Silva Ferreira et al., 2004).

Figure 2. Increase of TDN (g/L) content during ageing in 11 vintages of Verdicchio wines produced in the Massaccio area, 2002–2015.

Save View full size Expand inline Collapse inline

After analysing the vertical collection of old vintages as a survey study, a horizontal collection of Verdicchio wines produced in 2016 were analysed in depth. In this case, several different tests were carried out in order to understand the volatile profile of this variety. Results from SPE and SPME analyses are shown in Tables 1 and 2. As the data refer to a single vintage, no definitive conclusions can be drawn, although the wines from Massaccio, San Sisto and Le Moie areas were again distinguishable (Figure 3, heatmap free). Le Moie wine was distinguished by a higher content of hexyl acetate, octanoic and decanoic ethyl ester and the corresponding acids, by the highest amount of Ho-terpendiol I and of 3-methyl-2,4-nonanedione. Le Moie was also the area richest in the bound form. Massaccio was higher in norisoprenoids such as TDN and its precursor riesling acetale, methyl salicylate, homofuraneol and 2-(H)-pyran-2,6(3H)-dione. Some of these compounds will be also discussed in the GC-O section. Samples from San Sisto H1 were the richest in benzyl alcohol, methionol, 2-methyltetrahydrothiophen-3-one and 4-vinylguaiacol. It was also clear that the amount of terpenes increased from San Sisto H1 to San Sisto H3. The latter was indeed the richest in terpenoids such as linalool, citronellol, limonene and alpha-terpineol, both in the free and in the bound forms (Figure 4) and in -damascenone. TDN, vitispirane and riesling acetale reached their maximum in the H2 harvest data and then decreased slightly in H3; the same trend was reported in previous studies (Coelho et al., 2007).

The most vigorous vines resulted in a wine with a higher content of C6 compounds such as 1-hexanol, trans and cis 3-hexenol, trans 2-hexenol and hexyl acetate, while C₁₃ norisoprenoids, especially TDN, were higher in the wine from less vigorous vines. This is in accordance with the literature (Kwasniewski et al., 2010), where an increase in solar radiation leads to an increase of its precursor carotenoids.

The wines from different clones differed in the content of thiols (data not reported in this paper) and also in the content of methyl salicylate. Late harvested Arkezia, obtained from grapes with noble rot and a withering stage, was characterised by the presence of marker compounds for Botrytis infection, such as benzaldehyde, acetic acid, furfural and terpinen-4-ol. (Fedrizzi et al., 2011)

However, it will be necessary to confirm these preliminary observations made for the 2016 vintage by considering other vintages.

To better understand the sensorial impact of some compounds in Verdicchio aroma, SPE extracts from the three different areas were injected into GC-O and sniffed by six expert panellists: 48 main odorants were found in the SPE extract (Table 3). What is perceived in the glass is a mixture of odour and taste molecules, which may combine to act in a suppressive or additive manner, or synergistically (Frijters and Schifferstein, 1994). In GC-O, by contrast, the odorants are delivered to the olfactory epithelia as single entities, which simplifies the recognition task. Aroma extract dilution analysis (AEDA) was used to determine the relative odour potency of compounds present in the sample extract.

Figure 3. Hierarchical clustering analysis performed using volatile aroma compound profiles of 2016 wines produced from grapes from three areas (Le Moie, Massaccio, San Sisto) and three harvest data (San Sisto) (Ward algorithm and Euclidean distance analysis using MetaboAnalyst v. 3.0).

Save View full size Expand inline Collapse inline

The heatmap was generated using 70 compounds. The rows in the heatmap represent compounds and the columns indicate samples. The colours of the heatmap cells indicate the abundance of compounds across different samples. The colour gradient, ranging from dark blue through white to dark red, represents low, middle and high abundance of a compound.

Figure 4. Hierarchical clustering analysis performed using volatile aroma compound profiles of 2016 wines produced from grapes from three areas (Le Moie, Massaccio, San Sisto) and three harvest data (San Sisto) (Ward algorithm and Euclidean distance analysis using MetaboAnalyst v. 3.0).

Save View full size Expand inline Collapse inline

The heatmap was generated using 30 most significant bound compounds after enzymatic hydrolysis (the highest Fisher ratios). The rows in the heatmap represent compounds and the columns indicate samples. The colours of the heatmap cells indicate the abundance of compounds across different samples. The colour gradient, ranging from dark blue through white to dark red, represents low, middle and high abundance of a compound.

Aroma compounds with a higher FD value (>3) likely contributed considerably to overall Verdicchio wine aroma. In particular, we observed that the most important compounds in terms of sensorial impact were ethyl 2-methylpropanoate, with notes of apple fruit, ethyl hexanoate with a bilberry-fruity note, -damascenone with a baked apple note, furaneol and homofuraneol with a cotton candy scent, vanillin with a characteristic sweet pastry note, phenylacetic acid associated with honey-withered flower, one unknown compound with a toasted almond smell, isovaleric acid with a cheese odour, -ionone with a violet note and methional and methionol with a boiled potato odour. In some wines, a passion fruit/tropical note was also found at the retention index very close to 3-mercapto-hexanol. Moreover, in all samples a liquorice note was present, probably associated with the compound pantolactone, together with a clove/spice note given by 4-vinylguaiacol and a clear celery/vegetal scent, probably due to sotolone.

Furthermore, we focused our attention on the anise note noticeable in some vintages of Verdicchio. GC-O analysis made it possible to find the area where the six expert panellists sensed this note. In the literature, the compounds that can contribute to this note are trans and cis anethole, estragol (Zeller and Rychlik, 2007), ethyl hexanoate (San-Juan et al., 2010), R carvone (Pinar et al., 2017) and 3-methyl-2,4 nonanedione (Pons et al., 2008). Except for ethyl hexanoate, which is eluted early, all the other compounds showed an RI on a polar column between 1687 and 1800. Analysis of the pure standards led to identification of the 3-methyl-2,4 nonanedione (3-MND) compound responsible for the anise note. Other researchers have reported an anise-like note in the same area that has been tentatively identified (Martí et al., 2003; Peťka et al., 2006). This compound was found in wine by French researchers (Pons et al., 2008); it was related to the plum note in prematurely aged red wines and was also correlated with a loss of fruitiness (Pons et al., 2013). 3-MND is an intriguing compound that has recently emerged as the most potent known agonist for the human receptor OR1A1, with a submicromolar half-maximal effective concentration (Geithe et al., 2017). These results further emphasise the controversial role of this compound, whose presence at variable concentrations has previously been described as reminiscent of mint, anise, fruit kernels and prunes (Pons et al., 2008) and has been associated with prematurely aged wines, oxidised soybean oil (Kao et al., 1998) and freshly brewed green tea (Guth and Grosch, 1993). The odour description of this compound in wine model solution is strong, depending on its concentration, as reported by Pons et al. (2008). When the concentration is 100 ng L⁻¹ it is perceived as minty, while at around 1 g L⁻¹ it recalls anise, kernel and prune and at 10 g L⁻¹ the perception is of anise.

In order to understand the contribution of this compound to the Verdicchio aroma, 3-MND was added to a “neutral” Verdicchio (3-MND not detectable). The tasting was carried out by ten trained people (eight males, two females), who determined that the fruity note was masked only by the addition of 500 ng L⁻¹ of 3-MND, as the anise note predominated in this wine. The samples preferred by the panellists were those with a 3-MND concentration of between 25 and 50 ng L⁻¹. No oxidative note was detected in any sample. In Verdicchio wines 3-MND instead seems capable of imparting a positive and characteristic note.

Unfortunately, a well-perceived smell on the nose (olfactometry) is frequently not detected by chemical detectors (FID or MS). This is the case of some sulphur compounds and some carbonyl compounds, such as 3-MND. In such cases, it is necessary to use specific tests for the quantification of trace compounds. In order to quantify these compounds in our samples we used a derivatisation method with headspace solid phase microextraction using GC- tandem mass spectrometry. The derivatisation reagent PFBHA reacts with the carbonyl group formed oximes, which showed relatively specific mass spectra and high sensitivity. This method has been extensively employed to identify and quantify aroma components in several matrices, including wine (Flamini et al., 2005) and beverages (López-Vázquez et al., 2012; Saison et al., 2009). Our method was modified from that of Moreira et al. (2013) and Saison et al. (2009), where the determination of carbonyl compounds in beer was reported. Table 1 shows the content of 3-MND in our samples.

As Verdicchio is not an aromatic grape, one of the special characteristics of this variety that has been shown in previous studies (Versini et al., 2005) is the presence of a large amount of methyl salicylate. Methyl salicylate was found in V. riparia grapes (Schreier and Paroschy, 1980), in V. vinifera sp. (Cabaroglu et al., 1997) and in the Frontenac interspecific hybrid (Mansfield et al., 2011). The sensory impact of this compound in wine is not altogether clear. In the literature it is described as having an odour of wintergreen, mint and a fresh green character (Mansfield et al., 2011). We found that in the free forms the concentration was in the range 25–40 g L⁻¹, while in the bound forms, after hydrolysis, a higher concentration of up to 0.5 mg L⁻¹ was found (Table 2). As the olfactory threshold of this compound is 50–100 g L⁻¹, it is possible that in some Verdicchio wines the balsamic note of methyl salicylate contributes to the fresh/anise scent. It is well known that wine odours are complex mixtures of many volatile chemicals, present in different concentrations. These chemicals can interact synergistically or additionally in the mixtures according to unpredictable rules and are the basis of the overall odour sensation (Brattoli et al., 2013). A recent study (Parker et al., 2017) proposed the sensory role of monoterpene glycosides during tasting through retronasal perception of odorant aglycones released in-mouth. For this reason, studies are underway to quantify the precursors of methyl salicylate in wines, hence the total quantity, not only that deriving from hydrolysis of glycosylated compounds. In the literature it has been suggested that it can be bound as methyl salicylate 2-O--D-glucoside or as methyl salicylate 2-O- -D-xylopyranosyl(1–6)–D-glucopyranoside (primeveroside), although nine different glycosides have been reported in plants to date (Mao et al., 2014).

Conclusion

This work confirms that it is possible to obtain wines with very different characteristics from this variety of grapes, producing premium wines with a distinctive pattern of volatiles, reproducible across several vintages and variable depending on the different location of the vineyard and winemaking techniques. Young wines are characterised by fruity, thiolic notes, while wines aged for longer are distinguished by their norisoprenoid content and by balsamic, anise and liquorice notes, which can be attributed to methyl salicylate released by precursors and the presence of 3-MND and pantolactone. With a new derivatisation and GC-MS-MS method it was possible to quantify 3-MND in the range10–50 ng L⁻¹. With the use of GC-O, 48 sensorially important compounds were found for this wine. Analysis of the compounds after hydrolysis confirmed that a high amount of methyl salicylate characterised this variety. Methyl salicylate has a balsamic note from wintergreen oil that is often perceived in Verdicchio tasting. In-depth studies are underway to understand which precursors are present in grapes and whether they are in any way dependent on agro-climatic variables.

Funding

This study was supported by Fazi Battaglia winery.

References

• Bauer K., Garbe D. and Surburg H., 2008. Common Fragrance and Flavor Materials: Preparation, Properties and Uses. John Wiley & Sons: Chichester, UK.
• Brattoli M., Cisternino E., Dambruoso, P.R., de Gennaro G., Giungato P., Mazzone A., … Tutino M., 2013. Gas Chromatography Analysis with Olfactometric Detection (GC-O) as a Useful Methodology for Chemical Characterization of Odorous Compounds. Sensors (Basel, Switzerland), 13(12), 16759–16800. doi:10.3390/s131216759
• Cabaroglu T., Canbas A., Baumes R., Bayonove C., Lepoutre J.P. and Günata Z., 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. doi:10.1111/j.1365–2621.1997.tb15434.x
• Coelho E., Rocha S.M., Barros A.S., Delgadillo I. and Coimbra M.A., 2007. Screening of variety- and pre-fermentation-related volatile compounds during ripening of white grapes to define their evolution profile. Analytica Chimica Acta, 597(2), 257–264. doi:10.1016/j.aca.2007.07.010
• Costacurta A., Calò A. and Crespan M., 2003. The varietal identification and characterisation work of the ‘Istituto Sperimentale per la Viticoltura’ in the past fifteen years. Acta Horticulturae, 603, 261–273.
• Fedrizzi B., Tosi E., Simonato B., Finato F., Cipriani M., Caramia G. and Zapparoli G., 2011. Changes in wine aroma composition according to botrytized berry percentage: A preliminary study on Amarone wine. Food Technology and Biotechnology, 49(4), 529–535.
• Flamini R., Vedova A.D., Panighel A., Perchiazzi N. and Ongarato S., 2005. Monitoring of the principal carbonyl compounds involved in malolactic fermentation of wine by solid-phase microextraction and positive ion chemical ionization GC/MS analysis. Journal of Mass Spectrometry, 40(12), 1558–1564. doi:10.1002/jms.928
• Frijters J.E.R. and Schifferstein H.N.J., 1994. Perceptual interactions in mixtures containing bitter tasting substances. Physiology & Behavior, 56(6), 1243–1249. doi:10.1016/0031–9384(94)90372–7
• Geithe C., Noe F., Kreissl J. and Krautwurst D., 2017. The Broadly Tuned Odorant Receptor OR1A1 is Highly Selective for 3-Methyl-2,4-nonanedione, a Key Food Odorant in Aged Wines, Tea and Other Foods. Chemical Senses, 42(3), 181–193. doi:10.1093/chemse/bjw117
• Ghidoni F., Emanuelli F., Moreira Maia F., Imazio S., Scienza A. and Grando M.S., 2010. Variazioni del genotipo molecolare in Verdicchio, Trebbiano di Soave e Trebbiano di Lugana. ITALUS HORTUS, 17(3 suppl.), 373–380.
• Guth H. and Grosch W., 1993. Identification of potent odourants in static headspace samples of green and black tea powders on the basis of aroma extract dilution analysis (AEDA). Flavour and Fragrance Journal, 8(4), 173–178. doi:10.1002/ffj.2730080402
• Kao J.-W., Wu X., Hammond E.G. and White P.J., 1998. The impact of furanoid fatty acids and 3-methylnonane-2,4-dione on the flavor of oxidized soybean oil. JAOCS, Journal of the American Oil Chemists’ Society, 75(7), 831–835. doi:10.1007/s11746–998–0233–2
• Kwasniewski M.T., Vanden Heuvel J.E., Pan, B.S. and Sacks G.L., 2010. Timing of Cluster Light Environment Manipulation during Grape Development Affects C13 Norisoprenoid and Carotenoid Concentrations in Riesling. Journal of Agricultural and Food Chemistry, 58(11), 6841–6849. doi:10.1021/jf904555p
• López-Vázquez C., Orriols I., Perelló M.-C. and De R., 2012. Determination of aldehydes as pentafluorobenzyl derivatives in grape pomace distillates by HS-SPME-GC/MS. Food Chemistry, 130(4), 1127–1133. doi:10.1016/j.foodchem.2011.07.140
• Mansfield A.K., Schirle-Keller J.-P. and Reineccius G.A., 2011. Identification of Odor-Impact Compounds in Red Table Wines Produced from Frontenac Grapes. American Journal of Enology and Viticulture, ajev.2011.10067. doi:10.5344/ajev.2011.10067
• Mao P., Liu Z., Xie M., Jiang R., Liu W., Wang X., Meng S. and She G., 2014. Naturally Occurring Methyl Salicylate Glycosides. Mini Reviews in Medicinal Chemistry, 14(1), 56–63.
• Marais J., 1983. Terpenes in the Aroma of Grapes and Wines: A Review. South African Journal of Enology and Viticulture, 4(2), 49–58. doi:10.21548/4–2-2370
• Martí M.P., Mestres M., Sala C., Busto O. and Guasch J., 2003. Solid-Phase Microextraction and Gas Chromatography Olfactometry Analysis of Successively Diluted Samples. A New Approach of the Aroma Extract Dilution Analysis Applied to the Characterization of Wine Aroma. Journal of Agricultural and Food Chemistry, 51(27), 7861–7865. doi:10.1021/jf0345604
• Mattivi F., Fedrizzi B., Zenato A., Tiefenthaler P., Tempesta S., Perenzoni D., … and Vrhovsek U., 2012. Development of reliable analytical tools for evaluating the influence of reductive winemaking on the quality of Lugana wines. Analytica Chimica Acta, 732, 194–202. doi:10.1016/j.aca.2011.11.051
• Moreira N., Meireles S., Brandão T. and De P., 2013. Optimization of the HS-SPME-GC-IT/MS method using a central composite design for volatile carbonyl compounds determination in beers. Talanta, 117, 523–531. doi:10.1016/j.talanta.2013.09.027
• Nicolini G., Versini G., Moser S., Carlin S. and Malossini U., 2003. Aroma characteristics of wines from the cultivar ‘Manzoni bianco’ (‘White Riesling’ x ’Pinot blanc’). Mitteilungen Klosterneuburg Rebe Und Wein, Obstbau Und Früchteverwertung, 53(7–8), 251–257.
• Parker M., Black C.A., Barker A., Pearson W., Hayasaka Y. and Francis I.L., 2017. The contribution of wine-derived monoterpene glycosides to retronasal odour during tasting. Food Chemistry, 232, 413–424. doi:10.1016/j.foodchem.2017.03.163
• Peťka J., Ferreira V., González-Viñas M.A. and Cacho J., 2006. Sensory and chemical characterization of the aroma of a white wine made with Devín grapes. Journal of Agricultural and Food Chemistry, 54(3), 909–915. doi:10.1021/jf0518397Büttner A., Lopez Pinar A., Rauhut D. and Ruehl E., 2017. Effects of bunch rot (Botrytis cinerea) and powdery mildew (Erysiphe necator) fungal diseases on wine aroma. Frontiers in Chemistry, 5(MAR). doi:10.3389/fchem.2017.00020
• Pons A., Lavigne V., Darriet P. and Dubourdieu D., 2013. Role of 3-methyl-2,4-nonanedione in the flavor of aged red wines. Journal of Agricultural and Food Chemistry, 61(30), 7373–7380. doi:10.1021/jf400348h
• Pons A., Lavigne V., Eric F., Darriet P. and Dubourdieu D., 2008. Identification of volatile compounds responsible for prune aroma in prematurely aged red wines. Journal of Agricultural and Food Chemistry, 56(13), 5285–5290. doi:10.1021/jf073513z
• Rapp A. and Mandery H., 1986. Wine aroma. Experientia, 42(8), 873–884. doi:10.1007/BF01941764
• Rapp A. and Versini G., 1995. Influence of nitrogen compounds in grapes on aroma compounds of wines. In G. Charalambous (Ed.), Developments in Food Science, pp. 1659–1694. doi:10.1016/S0167–4501(06)80257–8
• Saison D., De Schutter D.P., Delvaux F. and Delvaux F.R., 2009. Determination of carbonyl compounds in beer by derivatisation and headspace solid-phase microextraction in combination with gas chromatography and mass spectrometry. Journal of Chromatography A, 1216(26), 5061–5068. doi:10.1016/j.chroma.2009.04.077
• San-Juan F., Pet’ka J., Cacho J., Ferreira V. and Escudero A., 2010. Producing headspace extracts for the gas chromatography-olfactometric evaluation of wine aroma. Food Chemistry, 123(1), 188–195. doi:10.1016/j.foodchem.2010.03.129
• Schreier P. and Paroschy J.H., 1980. Volatile components of wild grapes, Vitis riparia, M. Canadian Institute of Food Science and Technology Journal, 13(3), 118–121. doi:10.1016/S0315–5463(80)73349–7
• Silva Ferreira A.C. and Guedes de Pinho P., 2004. Nor-isoprenoids profile during port wine ageing—influence of some technological parameters. Analytica Chimica Acta, 513(1), 169–176. doi:10.1016/j.aca.2003.12.027
• van Ruth S.M., 2001. Methods for gas chromatography-olfactometry: A review. Biomolecular Engineering, 17(4–5), 121–128. doi:10.1016/S1389–0344(01)00070–3
• Vantini F., Tacconi G., Gastaldelli M., Govoni C., Tosi E., Malacrinò P., Bassi R. and Cattivelli L., 2003. Biodiversity of grapevines (Vitis vinifera L.) grown in the province of Verona. Vitis, 42(1), 35–38.
• Versini G., Carlin S., Dalla Serra A., Nicolini,G. and Rapp A. 2001. Formation of 1,1,6-Trimethyl-1,2-Dihydronaphthalene and Other Norisoprenoids in Wine: Considerations on the Kinetics. In ACS Symposium Series: Vol. 802. Carotenoid-Derived Aroma Compounds (Vol. 802, pp. 285–299.
• Versini G., Moser S. and Carlin S., 2005. Methyl salicylate as a remarkable almost bound compound in some renowned Italian varietal wines. In Vino Analytica Scientia: analytical chemistry for wine, brandy and spirits, Montpellier, France: 224.
• Winterhalter P. and Schreier P., 1994. C13-norisoprenoid glycosides in plant tissues: An overview on their occurrence, composition and role as flavour precursors. Flavour and Fragrance Journal, 9(6), 281–287.
• Winterhalter P., 1991. 1,1,6-Trimethyl-1,2-dihydronaphthalene (TDN) formation in wine. 1. Studies on the hydrolysis of 2,6,10,10-tetramethyl-1-oxaspiro[4.5]dec-6-ene-2,8-diol rationalizing the origin of TDN and related C13 norisoprenoids in Riesling wine. Journal of Agricultural and Food Chemistry, 39(10), 1825–1829. doi:10.1021/jf00010a027
• Xia J., Sinelnikov, I.V., Han B. and Wishart D.S., 2015. MetaboAnalyst 3.0—making metabolomics more meaningful. Nucleic Acids Research, 43(W1), W251–W257. doi:10.1093/nar/gkv380
• Zeller A. and Rychlik M., 2007. Impact of estragole and other odorants on the flavour of anise and tarragon. Flavour and Fragrance Journal, 22(2), 105–113. doi:10.1002/ffj.1765