Introduction

Wine quality is affected by both cultural and climatic factors, some of which are difficult to evaluate [1]. Cultural practices, as is well known, influence directly and indirectly fruit and wine quality. Cluster thinning is a viticulture tool used to correct overcropping, to improve fruit composition, and to find a balance between shoot growth and berry development [2]. The timing of cluster thinning influences yield components such as berries per cluster, berry weight, and size. For wine grapes, it is usually conducted before veraison [3, 4]. Cluster thinning has been shown to determine yield reduction, leading to an increase in sugar, color, and flavor in fruit at harvest in many wines [2, 5–8]. Some authors reported that it increased sugar and total phenolic concentration in Tempranillo and Grenache wines [7], anthocyanins and seed tannins in Carrot Noir, and potassium levels in Cabernet Sauvignon and Carignane wines [9, 10].

Syrah (Vitis vinifera L.) is an international red cultivar, considered to be one of the noble black grape varieties due to its ability to produce dark, full-bodied, and age-worthy wines. The variety may have originated in Southern Iran; nowadays, it is cultivated in all regions of the world particularly in Australia, where it is called Shiraz. Significant plantings also exist in South Africa and South America. In Sicily (South Italy), Syrah is widely cultivated, mostly in the province of Trapani and Agrigento on a total area of 4870 ha. The productivity of the vineyard is controlled by agronomic tools to avoid the decrease in wine quality.

The wine aroma profile is important as it contributes to the quality of the final product; it is due to the combined effects of several volatile compounds mainly alcohols, aldehydes, esters, acids, monoterpenes, and other minor components, such as sesquiterpenes and C13-norisoprenoids already present in the grapes or being formed during the must fermentation and maturation process of wine. Several factors, such as environment, cultural practices, ripeness and grape variety, winemaking and aging, influence the type and amount of volatile compounds. The volatile aroma compounds of a wine are, moreover, closely related to its sensory quality which is determining on the consumer’s acceptability [11].

The influence of cluster thinning on Syrah wine composition from different countries was studied by different authors who reported positive effects on the compounds related to the color and overall improvement in the wine quality [12–14].

To the best of our knowledge, no researches are present in the literature on the effects of cluster thinning on Syrah wine aroma volatile compounds. As regards Syrah wine volatiles, information is reported in literature on Australian wines from vines cultivated in cool and warm climate grape-growing regions and mainly on the key aroma compounds responsible for the peppery notes [15–19].

The aim of our research was therefore to evaluate the effect of cluster thinning on the aroma volatile compounds which are responsible for the sensory quality of Syrah wines obtained by vines cultivated under Mediterranean climate.

Materials and methods

Sampling

The research has been carried out in 2012 on a Syrah vineyard located in the province of Palermo (Agro di Monreale, Sicily, Italy) (37°53′N 13°09″E). The vines were planted in 2001 on 140 Ru rootstock. The vineyard was vertical shoot-positioned with Guyot pruning and oriented north–south. Manual cluster thinning in the early stage of veraison was applied, removing the clusters more distal from the bud (2nd or 3rd). Manual cluster thinning treatments were compared with not thinning controls. For the experimental design, randomized complete two blocks (thinned and control) with two replicates were used. Harvest date was based on the technological maturity of the untreated grapes. Sugar content, pH, and total acidity, during ripening from veraison until technological maturity (21–24°Brix), were assessed in 100 berries randomly collected from different positions within thinned and control plants. These data are reported in Table 1 together with the grape productivity and the average weight of the bunches and berries at harvest for 50 plants (homogeneous against pruning). After harvesting, the grapes were transferred to an experimental winery of the Istituto Regionale Vini e Oli (IRVOS) in Marsala (Sicily, Italy). The grapes from the two theses (thinning and control) were divided into four aliquots and then subjected to the same treatment.

Table 1 Physical and chemical analyses for Syrah grapes samples at harvest
Full size table

The grapes were destemmed and the musts fermented in contact with the skins, the musts were sulfited (0.05 g L−1) and added of dry yeast NDA21 (0.2 g L−1). Alcoholic fermentation lasted 9 days at 22–26 °C; malo-lactic fermentation was carried out with the addition of biomass (0.005 kg L−1 dregs). At the end of malolactic fermentation, the wines were racked, and sulfur dioxide was added (up to 25 mg L−1 of free sulfur dioxide). Clarification and cold treatments for tartaric stabilization were made. After decanting, the wines were bottled and stored. A total of 16 sample wines were made: eight samples for theses (control and thinned); each sample was analyzed in duplicate. Chemical analyses were carried out on the musts immediately before fermentation, on the wines after fermentation and after a period of 17 months. The bottled wines were stored in a conditioned room at 12–15° C and 65–70 % relative humidity. At this time, the wines were analyzed for aroma volatile compounds.

Chemicals analysis

Physical–chemical parameters: °Brix; nitrogen; pH value; titratable acidity; volatile acidity; alcohol content; total and monomer anthocyanins; total flavonoids and polyphenols; lactic and malic acid, were determined according to the EEC official method [20]. The color intensity and color hue were determined by spectrophotometry.

Volatile extraction and analysis

For the isolation and concentration of volatiles, the headspace solid-phase microextraction technique (HS-SPME) was used. All extractions were carried out using a DVB/CAR/PDMS fiber, of 50-/30-μm film thickness (Supelco, Bellefonte, PA, USA). Volatile compounds were identified and quantified by gas chromatography coupled on line with mass spectrometry (GC–MS); a capillary column, CP-Wax 52 CB, 60 m, 0.25 mm i.d. was used. All the analyses were carried out following the method previously reported [11, 21–23]. Each peak quantified was required to have a minimum signal-to-noise ratio (S/N) of 5. Quantitative results were obtained using the method of standard additions. Standard solutions were added to multiple aliquots of each sample wine. The sample alone was also analyzed. The quantification was based on a calibration curve generated by plotting the detector response versus the amount spiked of each standard. Each sample measurement was repeated twice. The standards used were purchased from Sigma-Aldrich s.r.l. (Milan, Italy) at the highest purity available. To quantify compounds whose standards were not available, the calibration curve of a compound of the same class of substances with the most similar peak area was used.

Statistical analysis

Statgraphics plus software, 5.1 version, was used to perform statistical analysis of the data. One-way ANOVA was performed on the data, to investigate the effects of cluster thinning on aroma volatiles and physical–chemical parameters. Principal component analysis (PCA) was performed on the amount of all the volatile compounds in the analyzed wine samples to investigate the differences between thinning wines and controls.

Results

Table 1 reports the productivity data for the two theses. Cluster thinning treatments decreased the grape production of about 27 % when compared to the control; otherwise, the average weight of the bunches and berry increased.

Tables 2, 3, and 4 report the results of chemical analyses carried out on the musts, and on the wines at the end of fermentation and after 17 months, respectively. In the musts (Table 2), the thinning treatment increased the sugar content and pH, while decreased the total acidity and the amount of the yeast assimilable nitrogen. At the end of fermentation (Table 3), the control samples showed a higher amount of total acidity than the thinned ones, whereas the thinned samples had a higher alcohol, total extract polyphenol, anthocyanin and flavonoid content, and pH values. The color intensity and the color hue were higher in the wine samples from thinned plants. After aging (Table 4), the total amount of flavonoids, polyphenols, anthocyanins, monomer anthocyanins, and dTAT (anthocyanin/tannin adducts) was higher in the thinned samples than in the control. The color intensity was higher in the thinned sample wines than in the control, and opposite behavior resulted for the color hue.

Table 2 Physicochemical parameters of the analyzed Syrah must samples before yeast inoculation
Full size table
Table 3 Physicochemical parameters of the analyzed Syrah wine samples at the end of fermentation
Full size table
Table 4 Chemical parameters of Syrah wine samples after an aged period of seventeen months
Full size table

Table 5 reports all the volatile constituents identified in the samples together with their LRI, quantitative data, perception threshold, and relative odors. The used analytical method allowed the identification and quantification of sixty compounds in the volatile fraction of Syrah wines. As can be seen from the table, they belonged to different classes of substances here listed in decreasing order of predominance by amount: alcohols, acids, esters, aldehydes, terpenes, and sesquiterpenes. In the analyzed wine samples, both control and thinned, numerous esters were identified; ethyl esters from C4 to C16 prevailed. Ethyl hexanoate (pineapple, fruity), ethyl octanoate (fruity, apple skin), and ethyl decanoate (wax, soapy, fruity) are the most represented with an amount higher than their odor threshold. The majority of the identified esters were present in quantities statistically higher in thinned samples than in the control ones. Alcohols are quantitatively the largest group of volatile compounds in Syrah wine. Linear and branched aliphatic from C4 to C8, sulfur and aromatic alcohols, such as benzyl alcohol and β-phenyl ethyl alcohol, were identified. Isoamyl alcohol (floral–fruity) and β-phenyl ethyl alcohol (floral) prevailed both in control and thinned samples; hexanol (herbaceous and vegetal) followed.

Table 5 Volatile compounds quantified in the Syrah wine samples
Full size table

Among fatty acids, hexanoic and octanoic were notable for their concentration. The content in acids was lowest in thinned samples. As regards the aldehydes, two furan compounds were identified, namely furfural and 5-methyl furfural (almond and caramel aroma notes). Furfural and its derivatives were present at levels lower than their perception threshold. Furfural was more abundant in thinned samples.

As shown in Table 5, seven monoterpenes and nine sesquiterpenes were identified. The total amount of terpenes evidenced statistical significant differences among the two theses. The total amount of monoterpenes and sesquiterpenes was higher in thinned wines than in the control: monoterpenes two times and sesquiterpenes three times higher. As regards, sesquiterpenes (E)-nerolidol, α-muurolene, germacrene D and γ-gurjunene were the most represented. All of these sesquiterpenes showed the highest amount in the thinned samples.

PCA was, thus, performed on the wine volatiles (Fig. 1). PCA resulted in a two-dimensional solution accounting for 88.9 % of the total variance in the data, of which 83.9 % was accounted by PC1 and 5 % by PC2. Therefore, almost all the variations in the data are explained by this two-dimensional summary. Wines from the two theses resulted well separated with thinned samples located on the positive side of the PC1 and control ones situated on the negative side. The volatile compounds which mostly influenced PC1 were diethyl succinate, ethyl octanoate, isoamyl acetate, and ethyl decanoate responsible for fruity and floral notes, whereas PC2 is mostly influenced by phenyl ethyl acetate (flowery, wine), (Z)-3-hexen-1-ol, δ-cadinene, and valencene responsible for flowery and wine, herbaceous, and woody notes (Table 6).

Fig. 1
figure 1

Principal component analysis (PCA). Projection of the analyzed Syrah wine samples in the space formed by the PC1 and PC2. a Control samples, b Thinned samples

Full size image
Table 6 Loading values of the variables mostly influencing PC1 and PC2 calculated on the volatile data
Full size table

Discussion

The results underline that the cluster thinning treatment reduces the vine productivity as expected. Otherwise, the average weight of the bunch and the berry was increased. This behavior was in agreement with different authors who demonstrated that the cluster thinning reduced in Syrah the yield per vine and crop load, but increased cluster and berry weights [12]. In our samples, cluster thinning improved the technological (sugar/acid ratio), phenolic (amount of anthocyanins and tannins), and probably aromatic maturity (greatest aroma potential) of the grapes. It increased the amount of sugar and decreased the acidity and thus increased the sugar/acid ratio. This behavior was in agreement with Gil et al. [14] who showed that thinning produced an earlier harvest and improved the grape quality in the Syrah variety.

The results highlighted that the cluster thinning impinges on the quality of Syrah wine, since the differences observed in composition among the samples. As regards the content of compounds responsible for wine color, a higher content resulted in the thinned samples; a higher color intensity and color hue were observed, too. This behavior was in agreement with different authors who demonstrated that cluster thinning results in an increased Syrah grape quality especially in compounds related to wine color [13, 14]. This could be due to the increase in the volume/surface ratio of the berries which allowed an improvement in the extraction of compounds from the skins [24]. Flavonoid, polyphenol, and mainly anthocyanin content decreased in the wine samples during the aging at the same manner in both theses. The reduction in the concentration of anthocyanins is due to various pathways involving their conversion to new more stable pigments such as anthocyanin/tannin adducts [25]. At the end of aging, the dTAT amount resulted higher in the thinned samples, leading to a more stable color.

As regards the aroma compounds in Syrah wines, limited information is present in the literature, and none of these deals with the effect of cluster thinning [18, 19]. Volatile aroma compounds are closely related to wine sensory characteristics and contribute to the tipicity and quality of wine [26]. The flavor of Syrah is commonly described as spicy, dark fruit, and berry like with different styles produced depending on the region of origin, winemaking, and viticultural decisions. In fact our volatile data are not much comparable with those reported by Mayr et al. [18] who analyzed Shiraz wines produced in warm and cool Australian regions. The ethyl esters of fatty acids are one of the most important groups of aroma compounds present in wine; mainly they are produced enzymatically during yeast fermentation and from ethanolisis of acyl-CoA [26]. These compounds have a positive effect on wine quality being responsible for their fruity sensory properties. Particularly ethyl-2-methylbutanoate and ethyl-3-methylbutanoate are responsible for strawberry-like aroma which are desirable notes for Syrah wines. The amount of ethyl esters is higher in thinned sample wines than in the control ones probably due to higher precursor availability in the grapes [27]. As regards the alcohols of interest is the highest amount of β-phenylethyl alcohol in the thinned samples. β-Phenylethyl alcohol is responsible for floral notes, and its content is closely connected with the content of phenylalanine in musts [28]. In the samples, the 3-methylthio-1-propanol was also identified. It originates from the metabolism of methionine; its aromatic contribution, of cauliflower and potatoes, is considered detrimental to wine quality [29]. It showed similar values in all the samples but always inferior to its olfactory threshold.

Thinned sample wines were also characterized by a higher amount of furfural and 5-methyl furfural responsible for almond and caramel aroma. It is well known that furfural is generated during the roasting of the barrels from hemicellulose of wood, but also from carbohydrates present in the wine during the aging period. These differences could be due to the higher content of sugar in musts from the thinned plants [26].

Among the primary aromas, the monoterpenes and sesquiterpenes hydrocarbons were the main contributors to the aroma of wine. Terpenes have a pleasant aroma and a very low olfactory threshold and are therefore perceived during wine tasting even in low concentrations. They are mainly derived from the grape, synthesized during maturation, and qualitatively and quantitatively influenced by cultivar, soil, climate, and viticulture practices. These compounds are not greatly modified during the fermentation processes, so that their presence in wine is directly related to their presence in must and therefore in the grapes [30]. In thinned Syrah wines, all the monoterpenes and sesquiterpenes identified showed the highest amount. As an example, the amount of (E)-nerolidol was about five times higher in thinned samples than in the control ones; (E)-nerolidol is an acyclic sesquiterpene, and it derives from farnesyl diphosphate and is responsible for floral notes.

The wine samples were also marked by a high amount of germacrene D, α-muurolene, and γ-gurjunene, responsible for spicy and woody notes. α-Muurolene belongs to the muurolene bicyclic sesquiterpenes, γ-gurjunene is a guaiene sesquiterpene which contains fused 5- and 7-membered rings, while germacrene D is considered a precursor of many sesquiterpene hydrocarbons such as cadinene, eudesmane (selinane), oppositane, axane, isodaucane, and bourbonene and is one of the Vitis vinifera L. sesquiterpenoids reported to occur in grapes, both red and white varieties [23].

Among sesquiterpenes, rotundone was not identified; it had been considered responsible for the peppery character of Syrah wines [16]. Rotundone concentration in musts and wines may depend on several factors such as cultivar, region, and mesoclimate; in particular, low temperatures may result in increased rotundone accumulation [17], and in effect, it was not been identified in Syrah samples from Mediterranean area [23].

Conclusion

Cluster thinning is a useful practice to reduce the excessive productivity which can characterize the Syrah cultivar. Our research, in addition to the clearly demonstrated effect of advancing grape maturity and improving the phenolic content of grapes and therefore of wine, evidenced a positive effect on the aroma compounds. The grapes of thinned plants tend to get rich in varietal aromas and probably in organic precursors from which originate the fermentation aromas. Cluster thinning, although involving a considerable reduction in yield, which is a real economic disadvantage, led to the improved quality of Syrah wines at least in the Mediterranean climate where the research was carried out. In fact, the effect of cluster thinning depends not only on the variety but also on the climatic conditions, and therefore, its use cannot be generalized.