Volatile profiles of Cabernet Gernischet wine under rain-shelter cultivation and open-field cultivation using solid-phase micro-extraction–gas chromatography/mass spectrometry

ABSTRACT

The volatile profiles of Vitis vinifera cv. Cabernet Gernischet wines in vintages 2010 and 2011 from rain-shelter cultivation and open-field cultivation were compared and detected by headspace solid phase micro-extraction with gas chromatography-mass spectrometry. The results showed that the number and concentration of volatile compounds quantified in wine were lower under rain-shelter cultivation compared with open-field cultivation. The strength of the fruity, fatty, caramel, and floral aromas of the 2010 vintage wines was lower under rain-shelter cultivation condition, and the compounds contributing to herbaceous aroma series were also lower with rain-shelter cultivation in the 2011 vintage wines. However, the strength of fruity, fatty, caramel, floral, and chemical aromas was higher with rain-shelter cultivation in vintage 2011. The cultivations in the two vintages could be clearly divided into three groups by principal component analysis. The difference between two treatments was significant in vintage 2010, whereas there was no significant difference in vintage 2011. The present work reveals the effects of rain-shelter cultivation on Cabernet Gernischet wine volatile compounds and explains that the technique is helpful in improving the wine quality in some rainy regions.

KEYWORDS:

• Rain-shelter cultivation
• Wine
• Volatile compounds
• HS-SPME
• GC-MS

Introduction

With increasing production and consumption, the Chinese wine industry has developed with unprecedented speed in recent years, and the grape planting area in China has increased rapidly. It reached 680 mha in 2013, ranking fourth in the world, according to the International Organization of Vine and Wine (OIV) statistical report on world vitiviniculture.^[1] Most of China enjoys a marked continental monsoonal climate, characterized by hot and rainy summer–autumns and cold and dry winter–springs. These climatic characteristics lead to bad grape growth, sugar accumulation, organic acid degradation, and phenolic compound formation. This seriously hinders the development of the grape and wine industry in China and the improvement of China’s wine quality.

To overcome the disadvantages caused by local climatic conditions, protected cultivation was introduced to viticulture in China in the 1980s. Protected cultivation is a fruit culture practice featuring an artificial microclimate that meets fruit growth requirements in some installations in the case of an unfavorable natural environment for fruit growth.^[2] In the south of China, rain-shelter cultivation is usually implemented in the production of table grape berries.^[3] It has been indicated that the incidence of some major diseases, such as downy mildew (Plasmopara viticola), powdery mildew (Uncinula necator), botrytis (Botrytis cinerea), rip rot (Glomerella cingulata), and sour rot (imperfect yeasts), can be effectively reduced by keeping rainwater away from leaves and fruits in rain-shelter cultivation,^[4] and the technique also reduced the use of pesticides in vineyards.^[5]

The aroma is one of the important sensory characteristics of wines and it influences the consumers’ overall acceptability.^[6–8] As reported, there were more than 800 volatile compounds found in grapes and wines, and their concentration ranged from hundreds of mg/L to the μg/L or ng/L level with different cultivars.^[9–11] There are three main groups of wine flavors that originated from varietal, fermentative, and wine ageing aromas.^[12] The concentration of these volatile compounds changes with grape variety, environmental conditions, cultural techniques, grape maturity state, and winemaking and ageing techniques.^[13–15]

It has been reported that rain-shelter cultivation effectively delayed the maturation of grape berries,^[3,16] slowed sugar accumulation by reducing photosynthetically active radiation (PAR), increased berry and cluster weight, and improved economic returns.^[17] During the past several decades, although rain-shelter cultivation has been studied for its commercial value on table grapes and other fruits,^[4,17–19] few studies have focused on wine grapes and wine. To our knowledge, there has been no published literature on the aroma profiles in the wines of Cabernet Gernischet grape (Vitis vinifera L.) cultured under a rain-shelter model.

In view of the preceding discussion, the purpose of the present study was to compare the aroma profiles of Cabernet Gernischet wine whose grapes cultured under rain-shelter and open-field conditions. From a practical standpoint, this study was carried out to provide sufficient experimental evidence for further application and diffusion of rain-shelter cultivation. Based on the experiment, the cultivation of Cabernet Gernischet grapes and development of the wine industry with regard to organic production system is expected.

Materials and methods

Plant materials and vinification

The two-year study (2010 and 2011) was conducted at the Grape Demonstration Base of the College of Enology, Northwest A&F University, Jingyang County of Shaanxi Province, China (34°40’56’’ N, 108°38’53’’ E). The mesoclimate of the vineyard during the growing season (from April to September) of the two vintages is shown in Table 1. The own-rooted vines of Cabernet Gernischet (Vitis vinifera L.) were planted in 2006 and were oriented north-south on flat terroir with sandy soil. The vines were spaced at a distance of 1.8 × 2.7 m, spur-pruned and trained to a bilateral cordon 0.8 m above the ground; the heights of the grapevines’ canopies were from 0.8 to 1.8 m above the ground, and the shoots were trained upwards, and each vine carried ca. 20 grape clusters. The vertical shoot-positioned canopies were uniformly managed. All vines were divided into two groups. Group one was cultured with rain-shelter cultivation technology. Shelters were built along with vine rows before berry coloration and were 2.2 m high, 1.7 m wide and covered with colorless and transparent polyethylene film. Group two was the control and was cultured in an open field (open-field cultivation).

Table 1. Mesoclimate of vineyard during the growing season of the 2010 and 2011 vintages.

Parameter	April	May	June	July	August	September	April to September
Vintage 2010
Average temperature (°C)	13.2	19.9	25.4	27.2	24.7	21.3	22.0
Rainfall (mm)	41.0	42.7	23.8	77.0	145.1	85.9	415.5
Sunlight hours	170	150.5	219.6	158	145.4	152.1	995.6
Vintage 2011
Average temperature (°C)	16.5	19.4	25.5	26.5	24.1	18.3	21.7
Rainfall (mm)	15.2	83.5	25.7	50.3	87.2	321.1	583.0
Sunlight hours	224.6	197.5	200.5	183.1	189.1	94.7	1089.5

Small-scale wine making was conducted in triplicate using the following protocol: control and rain-shelter cultivation fruit (20 kg for each of the three replicates per treatment) were harvested at their physiological maturity, the harvest date was August 22 for open-field cultivation and September 18 for rain-shelter cultivation in vintage 2010, and that was August 30 and September 15 for each in vintage 2011 (Table 2). The grapes were crushed, destemmed, and placed separately in stainless steel tanks immediately after transported to the lab. Fifty mg/L sulfur dioxide and 25 mg/L pectinase (LALLZYME EX, Lallemand Company, France) were added to the musts, and the contents were mixed by hand. Considering the low acid content in grape must in both years, tartaric acid was added to the must. The must acid content was 8 g/L in each treatment, according to OIV.^[20] After maceration of the musts for 24 h, 200 mg/L active dry wine yeast (Saccharomyces cerevisiae strain RC212, Lavlin, France) was added to the musts according to commercial specifications.

Table 2. Physicochemical parameters of must and wine in the winemaking phases of different treatments.

			Total sugar (g/L)^a	Total acidity (g/L)^b	pH	Ethanol (% vol)
Must after filling the tank	Vintage 2010	T1^c	179.2 ± 5.4Aa	4.7 ± 0.3Aa (8.0)^e	—	—
		T2^d	171.2 ± 4.2Aa	4.5 ± 0.2Aa (8.0)	—	—
	Vintage2011	T1	176.7 ± 5.3Aa	3.6 ± 0.2Aa (8.0)	—	—
		T2	153.7 ± 4.0Bb	3.9 ± 0.2Aa (8.0)	—	—
At the end of alcoholic fermentation	Vintage 2010	T1	1.8 ± 0.2Aa	5.6 ± 0.4Aa	3.6 ± 0.2Aa	12.6 ± 1.2Aa
		T2	1.9 ± 0.2Aa	6.4 ± 0.5Aa	3.3 ± 0.1Aa	10.05 ± 1.0Aa
	Vintage2011	T1	1.7 ± 0.4Aa	4.8 ± 0.4Bb	3.4 ± 0.1Aa	11.3 ± 0.6Aa
		T2	2.1 ± 0.3Aa	5.9 ± 0.4Aa	3.2 ± 0.1Aa	11.2 ± 0.5Aa

^aConcentration represented by glucose.

^bConcentration presented by tartaric acid.

^cT1: Rain-shelter cultivation.

^dT2: Open-air cultivation.

^eThe data outside the brackets represent the acid content in must before acidity adjusted (the acid content in grape after harvest), and the data inside the brackets represent the acid content in must after acidity adjusted (the acid content of must before fermentation).

The same letter mean are not statistically different in different treatments (p < 0.05).

Alcoholic fermentation was carried out at 22 to 25°C to dryness (reducing sugar < 4 g/L). Throughout this period, three mass homogenizations were performed per day to dissolve the cap of the wine. Temperature and density were also recorded three times per day to evaluate fermentation arrests. At the end of alcoholic fermentation, the wines were separated from pomace and added with SO₂ at 50 mg/L. After fermentation, the wine samples were bottled and stored at 4°C prior to analysis (for physicochemical parameters, see Table 2). The aromatic compounds were detected 4 months after the wines were well fermented in each year.

Physicochemical parameter analysis

The total reducing sugars, ethanol, pH, and total acidity were determined according to the National Standard of the People’s Republic of China-GB/T 15038-2006.^[21] The total reducing sugars were detected by Fehling’s solution reaction, and the sugars were represented by glucose.

Wine tasting

A wine tasting experiment was held when the wine samples were well done and stored at 4°C for 3 months. The tasting team was composed of 10 professional panelists. Each test session consisted of six wine samples coded with random numbers from 1 to 6; the order of the samples was randomized in each test session. Wine samples were stored at 4°C and presented at 20°C (ca 30 mL) for tasting. The tasting scores were defined according to the methods of Li^[22] and Tao.^[23]

Gas chromatography-mass spectrometry (GC‐MS) analysis of volatile compounds in wines

The extraction and analysis of wine volatile compounds were based on available methods developed by the Center for Viticulture and Enology, China Agricultural University.^[24–26] In this method, volatile thiols and volatile phenols were not measured but they can have a huge impact in red wines. And the method of detecting these two kinds of volatile compounds is developing in our lab. Five milliliters of wine with 10 μL of internal standard 4-methyl-2-pentanol (4M2P, 1.0018 g/L) and 1 g of NaCl added were blended in a 15-mL sample vial tightly capped with a PTFE-silicon septum containing a magnetic stirrer. Afterwards, the vial containing the sample was equilibrated at 40°C for 30 min while being stirred. Then the pretreated solid-phase micro-extraction (SPME) fiber (50/30-μm DVB/Carboxen/PDMS, Supelco, Bellefonte, PA, USA) was then inserted into the headspace, and extracted for 30 min with continued heating and agitation of the sample. After that, the fiber was immediately desorbed in the GC injector for 8 min. The analysis of volatile compounds was performed on an Agilent 6890 GC equipped with an Agilent 5975 MS fitted with a 60 m × 0.25 mm HP-INNOWAX capillary column with 0.25 μm film thickness (J&W Scientific, Folsom, CA, USA). The flow rate of helium was 1 mL/min. The oven temperature was programmed at the successive temperature and time points as follows: 50°C for 1 min, increased to 220°C at a rate of 3°C/min and held at 220°C for 5 min. The ion source temperature was 230°C and the MS transfer-line temperature was 280°C. Electron ionization mass spectrometric data from m/z 30-350 were collected. The ionization voltage was 70 eV.

Data management and analysis were performed using ChemStation Software (Agilent Technologies, Inc.). The retention indices of reference standards and mass spectra matching in the standard NIST 08 library were used to identify the volatile compounds. A comparison of retention indices with those reported in the literature was processed when standards were unavailable. For quantification, a synthetic model wine solution was prepared in distilled water containing 14% vol ethanol (high-performance liquid chromatography [HPLC] quality); and 5 g/L tartaric acid, and the pH was adjusted to 3.8 with 5 M NaOH solution. 4-Methyl-2-pentanol (10 μL added before detection) was employed as the internal standard. The standard solution was diluted into eight levels successively, and volatile compounds of each level were extracted and analyzed using the same method as that used for the wine samples. Volatile compounds without chemical standards were estimated with equations of the compounds that had the same functional group and/or similar numbers of C atoms.

Reagents and standards

NaOH, NaCl and tartaric acid were purchased from Beijing Chemical Works. Ethanol was purchased from Sigma-Aldrich; 4-Methyl-2-pentanol (98.0%, Sigma-Aldrich, Milwaukee, WI) was used as internal standard. The manufacturers and purity information of the chemical standards used for identification and quantification of volatile compounds in this experiment is listed in Table 3, and the information of curves of the standards is listed in the additional materials.

Table 3. The parameters of volatile compounds identified in the Cabernet Gernischet wines.

Compounds	RI^a	ID^b	Purity	Manufacturers	Threshold (μg/L)	Reference of threshold^c	Aroma descriptor	Reference of aroma descriptor^c	Aroma series^d	Reference of aroma seriesd^c
Alcohols
1-Hexanol	1353	A	0.97	Sigma–Aldrich	1100	[1]	Herbaceous, grass,woody	[1,2]	3	[1,2]
(E)-3-Hexen-1-ol	1365	A	0.98	Sigma–Aldrich	1000	[3]	Herbaceous, green	[3]	3	[3]
(Z)-3-Hexen-1-ol	1387	A	0.96	Sigma–Aldrich	1000	[1]	Herbaceous, green,bitter, fatty	[1,3]	3,8	[1,3]
(E)-2-Hexen-1-ol	1409	A	0.95	Sigma–Aldrich	15,000	[3]	Herbaceous, green	[3]	3	[3]
1-Propanol	1057	B			306,000	[1]	Alcohol, ripe fruit	[1]	1,7	[1]
1-Butanol	1158	A	0.99	Sigma–Aldrich	150,000	[1]	Medicinal, phenolic	[1-3]	7	[1-3]
Isobutanol	1111	A	0.99	Sigma–Aldrich	75,000	[1]	Alcohol, solvent,green, bitter	[1]	3,7	[1]
1-Pentanol	1256	A	0.99	Sigma–Aldrich	64,000	[6]	Balsamic, bitter almond	[6]	7,8	[6]
Isopentanol	1220	A	0.99	Sigma–Aldrich	60,000	[1]	Solvent, sweet,alcohol, nail polish	[1,7]	7,5,8	[1,7]
3-Methyl-2-butenol	1322	B
3-Methyl-3-buten-1-ol	1256	B			600	[8]
3-Methyl-1-pentanol	1328	B			500	[8]	Pungent, solvent,green	[2]	3,7	[2]
4-Methyl-1-pentanol	1316	A	0.97	Sigma–Aldrich	50,000	[9]	Almond, toasted	[6,9]	4,5,9	[6,9]
1-Heptanol	1455	A	0.99	Sigma–Aldrich	200	[8]	Oily	[6]	8	[6]
1-Octanol	1562	A	0.99	Sigma–Aldrich	800	[8]	Jasmine, lemon	[10,11]	2	[10,11]
2-Ethyl-1-hexanol	1493	A	0.99	Sigma–Aldrich	8,000	[8]
2-Nonanol	1517	A	0.99	Sigma–Aldrich
1-Decanol	1765	A	0.99	Sigma–Aldrich	400	[9]
1-Dodecanol	1974	A	0.98	Sigma–Aldrich
Benzyl Alcohol	1892	A	0.98	Sigma–Aldrich	200,000	[1]	Roasted, sweet	[1,6]	1,5,9	[1,6]
2-Phenylethanol	1928	A	0.99	Sigma–Aldrich	14,000	[9]	Roses, honey	[7]	2	[7]
laevo-2,3-Butanediol	1542	A	0.99	Sigma–Aldrich	150,000	[6]	Fruity, sweet, butter	[6]	1,5,8	[6]
(S)-3-Ethyl-4-methylpentanol	1509	B
3-(Methylthio)-1-propanol	1726	B
1,9-Nonanediol	1414	B
(3-Methyl-1,5-pentanediol)	1060	B
Esters
Isoamyl acetate	1122	A	0.95	Sigma–Aldrich	160	[1]	Banana	[7]	1	[7]
Phenethyl acetate	1830	A	0.99	Sigma–Aldrich	1800	[1]	Fruity, rose	[1,2]	1,2,5	[1,2]
Ethyl acetate	890	A	0.99	Sigma–Aldrich	7500	[4]	Fruity, sweet	[4]	1	[4]
Ethyl hexoate	1232	A	0.99	Sigma–Aldrich	80	[1]	Banana, green apple	[7]	1	[7]
Ethyl heptanoate	1334	B
Ethyl 2-hexenoate	1346	B		Sigma–Aldrich
Ethyl lactate	1350	A	0.98	Sigma–Aldrich	154,636	[9]	Fruity, buttery	[1]	1,8	[1]
Ethyl octanoate	1437	A	0.99	Sigma–Aldrich	580	[1]	Sweet, floral, fruity,banana, pear	[2]	1,2	[2]
Ethyl laurate	1576	B
Ethyl dodecanoate	1848	A	0.98	Sigma–Aldrich	1500	[12]	Flowery,fruity	[12]
Isoamyl hexanoate	1227	B
Methyl salicylate	1792	A	0.99	Sigma–Aldrich
Ethyl decanoate	1639	A	0.99	Sigma–Aldrich
γ-Butyrolactone	1645	B
Diethyl succinate	1682	A	0.99	Sigma–Aldrich
Ethyl 9-decenoate	1693	B
n-Hexyl isobutyrate	1378	B
Ethyl benzenepropanoate	1898	B
Ethyl octadecanoate	2183	B
Ethyl hydroxycinnamate	1522	B		Sigma–Aldrich
Ethyl eicosanoate	2649	B		Sigma–Aldrich
Acids				Sigma–Aldrich
Propanoic acid	1551	A	0.99	Sigma–Aldrich	8100	[9]	Pungent, rancid, soy	[6]	8	[6]
Hexanoic acid	1860	A	0.99	Sigma–Aldrich	420	[9]	Cheese, fatty	[7]	8	[7]
Octanoic acid	2075	A.	0.99	Sigma–Aldrich	500	[9]	Rancid, cheese, fatty acid	[7]	8	[7]
n-Decanoic acid	2292	A	0.98	Sigma–Aldrich	1000	[9]	Fatty, rancid	[7]	8	[5]
Acetic acid	1463	A	0.99	Sigma–Aldrich
2-Methyl-propanoic acid	1590	B
Carbonyl compounds
Benzaldehyde	1534	A	0.99	Sigma–Aldrich	2000	[1]	Roasted, almond	[3]	9	[3]
Terpenes
Linalool	1547	A	0.97	Sigma–Aldrich	25	[9]	Flowery, muscat	[7]	2,5	[7]
Geraniol	1793	B		Sigma–Aldrich	20	[14]	Citric, geranium	[7,14]	2	[7,15]
C13-norisoprenoid
β-Damascenone	1833	B		Sigma–Aldrich	0.14	[13]	Sweet, exotic	[6,7]	1,2,5	[6,7]
Others
4-Methyl-1-pentene	556	B
Butylated hydroxytoluene	1925	B

^aRetention indices on HP-Innowax column.

^bIdentification of the compounds, ‘A’ means identified by mass spectrum and RI agree with standards, “B” means tentatively identified by mass spectrum agrees with the mass spectral database and RI agrees with literature.

^c[1] (Peinado et al., 2004)^[65], [2] (Peinado et al., 2006)^[31], [3] (Franco, Peinado, Medina, et al., 2004)^[66], [4] (Guth, H., et al., 1997)^[40], [5] (Noble et al., 1987)^[29], [6] (Sánchez-Palomo et al., 2010)^[32], [7] (Noguerol-Pato et al., 2012)^[34], [8] (Tao & Zhang, 2010)^[67], [9] (Ferreira et al., 2000)^[68], [10] (La Guerche, Dauphin, Pons, et al., 2006)^[69], [11] (Jiang & Zhang, 2010)^[12], [12] (Li, H., et al., 2006)^[70], [13] (Pineau, Barbe, Van Leeuwen, et al., 2007)^[71], [14] (Gómez-Míguez et al., 2007)^[13].

^d1: fruity, 2: floral, 3: herbaceous (or vegetal, fresh), 4: nutty, 5: caramel (or sweet), 6: earthy, 7: chemical, 8: fatty, 9: roasted.

Odor activity values (OAVs) and aroma series

The OAVs of each volatile compound were calculated to estimate the sensory contribution of these odorants to the general flavor of wine. The OAVs were calculated by dividing the concentration of each compound in the sample by its odor threshold value of the compound in water/ethanol solution.^[27,28] To predict the overall aroma profile of the wine in the whole size from the data obtained from GC-MS analysis, nine aroma series were grouped based on similar odor descriptors of the aroma active compounds. The aroma series were modelled from the wine aroma wheel^[29] and literature (Table 3: fruity:1, floral: 2, herbaceous (or vegetal, fresh): 3, nutty: 4, caramel (or sweet): 5, earthy: 6, chemical: 7, fatty: 8, and roasted: 9). Due to the high complexity of olfactive perceptions, some volatile compounds were included in two or more aroma series according to some authors.^[30] In this study, except for the compounds without thresholds, the total intensities for each aroma series were calculated as the sum of the OAVs (∑OAV) of each compound assigned to this series list in Table 3 according to some studies.^[24,31–34]

Statistical analysis

Statistical analysis was performed using the SPSS 18.0 software application (Chicago, IL, USA). Duncan analysis was used to assess statistically significant differences in the content of various volatile compounds between wines of the treatments with three replications of the same sample. Significant differences were calculated at 0.05 levels. Principal component analysis (PCA) was also used in the experiment. Origin 8.1 (Systat Software Inc., Richmond, CA, USA) and Excel 2010 (Microsoft Corp., Redmond, WA, USA) were used to draft the graph.

Results and discussion

Physicochemical parameters

To avoid the damage caused by diseases, when the berries were harvested, they may not have attained the optimal maturity. The oenological parameters (Table 2) show that good ripeness and uniformity were achieved in the grapes. However, it should be noted that, to overcome the diseases of grape berries in vintage 2011, the treatment under open-field cultivation was harvested earlier than the other (rain-shelter cultivation). Except for the total sugar content of the open-field cultivation in vintage 2011, there were no significant differences in the total reducing sugars in the must, total acidity before fermentation and alcohol content of the wines for all treatments (Table 2).

Effects of rain-shelter cultivation on volatile compounds and their OAVs

In this study, a total of 57 volatiles were detected in all samples, in vintage 2010, 53 and 51 free volatile compounds were identified in rain-shelter cultivation (T1) and open-field cultivation (T2), respectively, whereas the number was 37 for both in vintage 2011. There were 44 and 45 free volatile compounds quantified, respectively, in rain-shelter cultivation (T1) and open-field cultivation (T2) in vintage 2010; there were 27 and 28, respectively, in vintage 2011. In the two vintages, no significant difference in the total concentration of volatile compounds between the two treatments of wines was found. Table 3 lists all the volatile compounds identified and their aroma descriptors and thresholds. The quantitative results and OAVs of these compounds are shown in Table 4.

Table 4. The concentration of volatile compounds and their odor activity values (OAVs) at the end of alcoholic fermentation (AF) in two vintages.

	Vintage 2010 Contents (μg/L)		Vintage 2011 Contents (μg/L)		Vintage 2010 odor activity values^c		Vintage 2011 odor activity values
Compounds	T1^a	T2^b	T1	T2	T1	T2	T1	T2
Alcohols
1-Hexanol	1601.8Aa	1439.7Aa	9858.5Aa	10457.9Aa	1.5	1.3	9.0	9.5
(E)-3-Hexen-1-ol	42.8Aa	36.4Aa	tr	tr	0.0	0.0	–	–
(Z)-3-Hexen-1-ol	281.3Aa	250.9Aa	tr	tr	0.3	0.3	–	–
(E)-2-Hexen-1-ol	19.1	nd	nd	nd	0.0	–	–	–
1-Propanol	5048.9Aa	4234.0Bb	tr	tr	0.0	0.0	–	–
1-Butanol	415.2Ab	485.9Aa	20831.5Aa	19841.5Aa	0.2	0.2	0.5	0.7
Isobutanol	32137.7Ab	40123.1Aa	12430.0Aa	16335.6Aa	0.4	0.5	0.2	0.2
1-Pentanol	tr	tr	6649.6Aa	2578.1Bb	—	—	0.1	0.0
Isoamyl alcohol	24353.8Aa	27075.7Aa	77193.0Aa	104872.7Aa	0.4	0.5	1.3	1.7
3-Methyl-3-buten-l-ol	11.78Ab	15.90Aa	nd	nd	0.0	0.0	—	—
3-Methyl-1-pentanol	48.0Aa	58.7Aa	1072.7Aa	1356.5Aa	0.1	0.1	2.1	2.7
4-Methyl-1-pentanol	28.0Aa	38.7Aa	599.9	nd	0.0	0.0	0.0	—
1-Heptanol	59.6Aa	42.9Bb	nd	nd	0.3	0.2	—	—
1-Octanol	18.3Aa	10.6Bb	117.0Aa	120.3Aa	0.0	0.0	0.1	0.2
2-Ethyl-1-hexanol	nd	7.3	nd	nd	—	0.0	—	—
2-Nonanol	72.9	tr	nd	nd	—	—	—	—
1-Decanol	11.1Aa	11.8Aa	nd	nd	0.0	0.0	—	—
1-Dodecanol	nd	1.7	nd	nd	—	—	—	—
Benzyl alcohol	475.5Aa	374.7Aa	24024.4Aa	11589.9Aa	0.0	0.0	0.1	0.1
2-Phenylethanol	17956.3Aa	22128.4Aa	6135.6Ab	12471.2Aa	1.3	1.6	0.4	0.9
laevo-2,3-Butanediol	343.0Aa	410.2Aa	nd	nd	0.0	0.0	—	—
(S)-3-Ethyl-4-methylpentanol	193.4Aa	86.7Bb	nd	nd	—	—	—	—
3-(Methylthio)-1-propanol	16.1Aa	19.1Aa	13.3Aa	13.3Aa	—	—	—	—
1,9-Nonanediol	2.6	nd	nd	nd	—	—	—	—
(3-Methyl-1,5-pentanediol)	1.0	nd	nd	nd	—	—	—	—
Total alcohols (mg/L)	107.1Ab	123.4Aa	215.3Ab	264.7Aa
Esters
Isoamyl acetate	tr	tr	334.4Aa	281.4Ab	—	—	2.1	1.8
Phenethyl acetate	tr	30.7	5526.1Aa	2225.1Bb	—	0.0	3.1	1.2
Ethyl acetate	56218.64Aa	53056.0Ab	20098.0Aa	23688.5Aa	7.5	7.1	2.7	3.2
Ethyl hexoate	718.93Aa	638.2Aa	639.7Aa	714.4Aa	9.0	8.0	8.8	9.3
Ethyl enanthate	81.9Aa	71.0Aa	tr	tr	0.0	0.0	—	0.0
Ethyl 2-hexenoate	29.9Aa	33.3Aa	nd	nd	—	—	—	—
Ethyl lactate	1582.0Bb	4798.4Aa	tr	4842.2	—	—	—	—
Ethyl octanoate	2112.1Aa	1294.3Bb	tr	tr	3.6	2.2	—	—
Ethyl laurate	tr	tr	241.3Aa	186.3Bb	—	—	—	—
Isoamyl hexanoate	nd	nd	110.9	tr	—	—	—	—
Methyl salicylate	28.4	nd	tr	tr	—	—	—	—
Ethyl decanoate	211.5Aa	115.4Aa	51.2Aa	22.3Bb	—	—	—	—
γ-butyrolactone	44.09Aa	53.9Aa	nd	nd	—	—	—	—
Diethyl butanedioate	154.5Bb	232.1Aa	tr	tr	—	—	—	—
Ethyl 9-decenoate	1089.6Aa	548.7Ab	8596.1Aa	3238.4Ab	—	—	—	—
n-Hexyl isobutyrate	5.6Aa	14.3Aa	nd	nd	—	—	—	—
Ethyl benzenepropanoate	12.8Aa	22.3Aa	113.9Bb	205.3Aa	—	—	—	—
Ethyl octadecanoate	16.7	nd	nd	nd	—	—	—	—
Ethyl hydroxycinnamate	14.4	nd	nd	nd	—	—	—	—
Ethyl eicosanoate	nd	24.7	nd	nd	—	—	—	—
Total esters (mg/L)	62.3Aa	60.9Ab	59.5Aa	52.9Ab
Acids
Propanoic acid	nd	4234.0	nd	nd	—	0.5	—	—
Hexanoic acid	1464.1Aa	1480.8Aa	1744.9Aa	1238.3Aa	3.5	3.5	4.2	2.9
Octanoic acid	1464.2Aa	1442.4Aa	1134.5Aa	780.5Aa	2.9	2.9	2.3	1.6
N-Decanoic acid	tr	32.7	238.9Aa	155.6Aa	—	0.0	0.2	0.2
Acetic acid	35385.5Aa	31412.0Aa	17526.0Aa	22524.6Aa	—	—	—	—
2-methyl-Propanoic acid	99.1Aa	116.2Aa	tr	tr	—	—	—	—
Total acids (mg/L)	38.4Aa	38.7Aa	20.6Aa	24.7Aa
Carbonyl compounds
Benzaldehyde	tr	tr	335.7Aa	495.7Aa	—	—	0.2	0.2
Total carbonyl compounds	tr	tr	335.7Aa	495.7Aa
Terpenes
Linalool	tr	1.8	3.3Bb	5.2Aa	—	0.1	0.1	0.2
Geraniol	7.1Aa	6.3Aa	63.9Aa	63.1Aa	0.4	0.3	3.2	3.2
Total terpenes	7.1Bb	8.0Aa	67.2Bb	68.3Aa
C13-norisoprenoid
β-Damascenone	5.5Bb	7.9Aa	detected	detected	39.4	56.4	—	—
Total C13-norisoprenoid	5.5Bb	7.9Aa	detected	detected
Others
4-Methyl-1-pentene	nd	46.3	nd	nd	—	—	—	—
Butylated hydroxytoluene	90.9Bb	237.3Aa	nd	2.9	—	—	—	—
Total others	90.9Bb	283.6Aa	nd	2.9

^aT1: Rain-shelter cultivation.

^bT2: Open-air cultivation.

Different letters means significant differences according to Duncan test in different treatments (p < 0.05).

nd not detected; tr the compound was detected but too little to quantitative.

^cThe concentrations of the volatile compounds were calculated by the curves of standards solutions that were listed in additional materials. And the odor activity values were calculated by the ratio of concentration and threshold listed in Table 3.

Alcohols

Alcohols are released as secondary products of yeast metabolism. They can be produced by yeast through either the anabolic pathway from glucose or the catabolic pathway from corresponding amino acids.^[31] They are the largest group of free volatile compounds in Cabernet Gernischet wines and account for 51.5–77.1% of the volatile content (Table 4). This is consistent with the literature.^{[11,28,35,36]} In the wine of the two vintages, the total content of alcohols was less in rain-shelter cultivation, and the difference was insignificant (p > 0.05). There were 23 and 22 alcohols identified in rain-shelter cultivation (T1) and open-field cultivation (T2) wines, respectively, in vintage 2010, whereas there were 14 and 13, respectively, in vintage 2011. The main compounds accounting for the difference were 1-propanol, 1-butanol, isobutanol, 3-methyl-3-buten-l-ol, 1-heptanol, 1-octanol, 2-phenylethanol, and (S)-3-ethyl-4-methylpentanol, in which the content of 1-propanol, 1-heptanol, (S)-3-ethyl-4-methylpentanol, and 1-octanol was observably higher in rain-shelter cultivation (T1) in vintage 2010, whereas the content of 1-butanol, isobutanol and 3-methyl-3-buten-l-ol was significantly higher in open-field cultivation (T2). For the 2011 vintage wines, the concentration of 1-pentanol and 2-phenylethanol was obviously higher under rain-shelter cultivation (T1) and open-field cultivation (T2), respectively.

Esters

Esters in wine, the primary source of fruity aromas,^[37] are the second major volatile constituent. Most esters are secondary metabolites produced by yeast during alcoholic fermentation.^[38] According to the pathway the esters produced, esters can be divided into two main groups. Acetate esters came from the reaction of acetyl-CoA with higher alcohols formed by the degradation of amino acids or carbohydrates.^[37,38] Fatty acid esters, another important group of esters in wine, are produced during yeast fermentation through ethanolysis of acyl-CoA, which is formed during fatty acid synthesis or degradation.^[37,38] In this study, in vintage 2010, 20 and 17 esters were detected, in rain-shelter cultivation (T1) and open-field cultivation (T2), respectively, whereas both were 17 in vintage 2011. Of all esters detected, there were 15 and 14 quantified in vintage 2010 in rain-shelter cultivation (T1) and open-field cultivation (T2), respectively, and both were 9 in vintage 2011. In the two treatments of the two vintages, esters accounted for 15.4-30.0% of the volatile compounds (Table 4). In vintage 2010, the content of ethyl acetate, ethyl 9-decenoate and ethyl octanoate was significantly higher in the wine from rain-shelter cultivation (T1), whereas the content of ethyl lactate and diethyl butanedioate was higher in wine from open-field cultivation (T2), and the difference was remarkable. In vintage 2011, the content of phenethyl acetate, ethyl decanoate, ethyl ester, and ethyl laurate was highest in the wine from rain-shelter cultivation (T1), and the content of ethyl 9-decenoate and isoamyl acetate was significantly higher in the wine from rain-shelter cultivation (T1). The content of ethyl benzenepropanoate was higher in the wine from open-field cultivation (T2), and the difference was most remarkable. In the two vintages, the total content of esters was higher in rain-shelter cultivation (T1), but the difference was insignificant.

Acids

Six volatile organic acids were detected in this study, of which 5 and 6 were found in rain-shelter cultivation (T1) and open-field cultivation (T2) in vintage 2010, respectively, and 5 was in vintage 2011. The kind of aromas were produced by yeast and bacteria during fatty acid metabolism.^[38] They were described by various fatty, pungent, rancid, fruity, or cheesy odors (Table 3). No difference was found in the content of acids between the wine of the two treatments in both vintages, and the total content was slightly higher in open-field cultivation (T2).

Terpenes and others

Terpene and norisoprenoids are two classes of important volatile compounds in wine. They originate from grape berries and are transferred to the wine during fermentation. This group of compounds belongs to the secondary plant constituents and is produced from glucose by acetyl-coenzyme A during grape ripening.^[27,39] Terpene compounds have a low olfactory threshold and are generally associated with floral and citric aromas;^[40] yeast metabolism has little effect on them during fermentation, so they are good indicators for the variety and quality of grapes.^[41] Two terpenes were totally detected in the two vintage wine samples, and both were quantified in rain-shelter cultivation (T1) and open-field cultivation (T2). The content of β-damascenone and linalool was higher in open-field cultivation (T2), and the difference was more significant. Three further compounds were detected in this study—benzaldehyde, 4-methyl-1 -pentene and butylated hydroxytoluene^[42,43]—and the content of all were less than their threshold.

Volatile thiols and volatile phenols

Volatile thiols, may play a major role in the aroma of many fresh plants, vegetables, fruit, or processed products (roasted coffee, wine, etc.).^[44–46] However, the fact that these substances are present in these complex natural products in trace amounts makes it difficult to identify them and measure their concentrations. Volatile phenols also play an important role in the wine aroma, and they contribute to the overall aroma.^[47] The volatile thiols and volatile phenols were not detected by the method used in the current article, while the new method to detect them was building in our lab.

Influence of rain-shelter treatments on aroma profiles

The total OAVs (∑OAV) for each series were calculated by adding the OAVs for each individual compound assigned to this series as listed in Table 3 (Fig. 1a and 1b). The ∑OAV of series 4 (nutty) and 6 (earthy) were not shown in Fig. 1 because of their low contribution to the overall wine aroma (∑OAV < 1). The application of aroma series makes it possible to correlate quantitative information obtained by chemical analysis and sensory characters of the samples. This has been used by many authors in recent years.^[24,31–34]

Figure 1. The total OAVs (∑OAV) of aroma series and tasting scores about volatile compounds in wines. The total OAVs (∑OAV) of aroma series in wines in 2010 (A) and 2011(B); the tasting scores about volatile compounds in wines of 2010(C) and 2011(D). T1: Rain-shelter cultivation; T2: Open-air cultivation.

The fruity series had the highest intensity in all treatments. The major contributors to this series were four esters (ethyl hexanoate, ethyl octanoate, isoamyl acetate, and ethyl acetate) and one norisoprenoid (β-damascenone), all of which had an OAV greater than 1 (Table 4). The influence of rain-shelter treatment on this aroma series differed between the two vintages. In vintage 2010, wine under rain-shelter cultivation (T1) led to a significant decrease (20%) in ∑OAV for the fruity aroma series, whereas it had a slight increase (6%) in vintage 2011.

In this study, caramel and floral series were two other major aroma series. For the caramel series, the ∑OAV was dominated by four compounds (isopentanol, phenethyl acetate, linalool, and β-damascenone), and the floral series was dominated by seven compounds (1-octanol, 2-phenylethanol, phenethyl acetate, ethyl octanoate, linalool, β-damascenone, and geraniol). Similar to the fruity aroma series, the caramel and floral series were reduced by rain-shelter cultivation (T1) in vintage 2010 and increased in vintage 2011.

The fatty aroma series mainly consisted of four organic acids (propanoic acid, n-decanoic acid, hexanoic acid, and octanoic acid) and three alcohols (isopentanol, 1-heptanol, and (Z)-3-hexen-1-ol). The influence of rain-shelter treatment on this aroma series differed between the two vintages. In vintage 2010, wine under rain-shelter cultivation (T1) decreased in ∑OAV for the fatty aroma series, and the difference was not significant, whereas it increased in vintage 2011.

Herbaceous and chemical aroma series are usually considered as negative factors to wine aroma. They could enhance the complexity of wine aroma at low concentrations, whereas at high concentrations they destroy the fruity note. In this study, four alcohols (1-hexanol, (Z)-3-hexen-1-ol, isobutanol, and 3-methyl-1-pentanol) contributed to the herbaceous series, and four alcohols (1-butanol, isobutanol, isopentanol, and 3-methyl-1-pentanol) contributed to the chemical series. The herbaceous aroma series was decreased by rain-shelter cultivation (T1); the results were similar in both vintages, and the difference was insignificant. Whereas the chemical series was increased by rain-shelter cultivation (T1) in vintage 2011, the difference was not remarkable. From the results above, it is clear that rain-shelter cultivation (T1) reduced ∑OAVs for the volatile compounds in all categories in vintage 2010, whereas the influence was opposite except for herbaceous categories in vintage 2011. The result of vintage 2011 is consistent with previous studies in other quality indexes by other researchers.^[3,48] A wine tasting experiment was held to compare the total behavior of wine under different cultivations in different vintages (Fig. 1c and 1d). From Fig. 1, wine from rain-shelter cultivation in vintage 2010 has higher scores of harmony, elegance and refinement than wine of open-field cultivation, whereas in vintage 2011, there was little difference between the two wines.

PCA

To reveal the influence of rain-shelter cultivation on aromatic constituents, PCA was performed. Figure 2 shows the loading plot on the plane defined by the first and second principal components and the corresponding projection of samples. The first principal component (PC1) describes 70.7% of the variability in the data, and the second (PC2) describes 19.2%. PC1 is correlated highly with all of the aromatic compounds except for V29 (ethyl hexoate), V32 (ethyl lactate) and V47 (hexanoic acid), whereas PC2 is correlated highly with V4 ((E)-2-hexen-1-ol), V15 (2-ethyl-1-hexanol), V16 (2-nonanol), V18 (1-dodecanol), V22 ((S)-3-ethyl-4-methylpentanol), V24 (1,9-nonanediol), V25 (3-methyl-1,5-Pentanediol), V29 (ethyl hexoate), V32 (ethyl lactate), V36 (methyl salicylate), V37 (ethyl decanoate), V41 (n-hexyl isobutyrate), V43 (ethyl octadecanoate), V44 (ethyl hydroxycinnamate), V45 (ethyl eicosanoate), V46 (propanoic acid), V55 (β-damascenone), V56 (4-methyl-1-pentene), and V57 (butylated hydroxytoluene). The cultivations in the two vintages could be clearly divided into three groups according to their relationships between scores and their aromatic constituents. The first group was composed of rain-shelter cultivation in vintage 2010 (T1-10) with the higher positive value of PC1 and PC2, owing to its higher values of V4 ((E)-2-hexen-1-ol), V16 (2-nonanol), V24 (1, 9-nonanediol), V25 ((3-methyl-1, 5-Pentanediol), and V29 (ethyl hexoate).

Figure 2. Distribution of wine samples and loadings of volatile compounds in the first two PCs of volatile compounds. T1-10 Rain-shelter cultivation wine in vintage 2010; T2-10: Open-air cultivation wine in vintage 2010. T1-11: Rain-shelter cultivation wine in vintage 2011. T2-11: Open-air cultivation wine in vintage 2011. V1-V57: The volatile compounds in Table 4 from 1-hexanol to butylated hydroxytoluene in order.

The second group was open-field cultivation in vintage 2010 (T2-10) with higher positive and negative values of PC1 and PC2, respectively. It could be characterized by V18 (1-dodecanol), V23 3-(methylthio)-1-propanol), V32 (ethyl lactate), V41 (n-hexyl isobutyrate), V45 (ethyl eicosanoate), V46 (propanoic acid), V55 (β-damascenone), and V57 (butylated hydroxytoluene). The third group was composed of rain-shelter cultivation and open-field cultivation in vintage 2011 (T1-11 and T2-11) with the higher negative value of PC1. There was little difference between the two treatments.

Conclusions

The volatile components in wine were influenced by a lot of aspects, such as the variety of grape,^[49–51] the position of berry in the same variety,^[49,52,53] the sugar concentration of berry and wine,^[54–56] the effect of fungicides,^[57–61] the yeast stains used,^[60,62] and the agronomic and technological practices during grape growing and winemaking.^[3,50] In this study, the influence of rain-shelter cultivation on volatile compounds of Cabernet Gernischet wines was investigated and compared with those of open-field cultivation. The results showed that the number and concentration of volatile compounds quantified in wines decreased under rain-shelter cultivation compared with that of open-air cultivation. The fruity, fatty, caramel, and floral aromas also decreased in rain-shelter cultivation in vintage 2010, whereas the compounds contributing to fruity and fatty, caramel and floral aroma series increased by rain-shelter cultivation in vintage 2011. In the two vintages, these differences were closely related to climate alterations^[3] in the vineyards (Table 1). In vintage 2011, the average temperature in the vine stage was lower, and the rainfall was higher than that of vintage 2010, which was bad for grape growth in open-field cultivation.^[3] Lower temperatures and heavier rainfall caused serious diseases among the grape berries. The bad quality and lower maturity of grape berries thus influenced the wine quality.^[63] To overcome this, the application of rain-shelter cultivation in wine grapes is urgently needed. And the content of volatile thiols and volatile phenols were not detected in this study, while the method for them was building. In some researches,^[57–61] the application of fungicides increased the strength of fruity aroma due to several ethyl esters, and reduced the varietal aroma of wines. But the problem of fungicide residual is very serious. The application of rain-shelter can keep rainwater away from leaves and fruits, and decreased the diseases of fruit. Thus the use of fungicides can be avoided, and makes the cultivation of grape Cabernet Gernischet and wine industry in an organic production system.

Nomenclature

SPME-GC/MS Solid-phase micro-extraction gas chromatography mass spectrometry

GC-MS Gas chromatography mass spectrometry

OAVs Odor activity values

OIV International Organization of Vine and Wine

T1 Rain-shelter cultivation

T2 Open-field cultivation

Acknowledgments

The authors would like to thank the Center for Viticulture and Enology, China Agricultural University for technical assistance in the completion of the GC/MS experiments. The authors also like to thank Bao Jiang in Weinan Testing and Inspection and Research Center of Agricultural Products and Food, Weinan Vocational and Technical College, and Jiang-Fei Meng in the college of Enology, Northwest A&F University, for their critical review of the manuscript. The American Journal Experts helped with the editing of the English language throughout the article.

Funding

This study was funded by a grant from the National Technology System for Grape Industry (CARS-30-zp-9).

Additional information

Funding

This study was funded by a grant from the National Technology System for Grape Industry (CARS-30-zp-9).