Volatile compounds in wild strawberry and their odorants of wild strawberry wines: Effects of different stages of fermentation

ABSTRACT

Wild strawberries are grown widely in southern Shaanxi (China), but information about their sensory and chemical characteristics is scant. In this work, exploratory research was conducted to correlate the results of instrumental analyses of the aroma compounds in wild strawberry wine and their sensory perception. Headspace solid-phase micro extraction (HS-SPME) was used for the pretreatment, and gas chromatography-mass spectrometry (GC-MS) was to analyze the aroma compounds. A total of 78 volatile compounds were identified, including 25 alcohols, 25 esters, four acids, five ketones, four aldehydes, nine terpenes, five phenols, and one styrene, but only 74 could be quantified. Odor activity values (OAVs) were determined for 21 odor-active compounds. Six were identified as the characteristic aroma substances for the wild strawberries from a principal component analysis (PCA) of the composition, in particular methyl 2-methylbutyrate, ethyl 2-methylbutyrate, ethyl 3-methylbutyrate, (E)-3-hexen-1-ol, 1-octen-3-ol, and phenylacetaldehyde. Sensory evaluation by a trained panel using descriptive analysis also revealed that sweetness was the main attribute in wild strawberry and its wines, and differences in varietal flavors were also observed. Wild strawberry wine is proved to exhibit a high quality of its fruit flavor, and thus winemaking is a performable practice for wild strawberry.

KEYWORDS:

• Aroma composition
• Dynamic changes
• GC-MS
• PCA
• Wild strawberry

Introduction

Wild strawberries are smaller than garden strawberries, measuring between 15 and 23 mm. When ripe, they are almost completely white but with red “seeds”. The fruit flesh can range from soft white to orange and is fragrant with a slight pineapple flavor. Although they have smaller fruits and reduced yields compared with garden strawberries, they accumulate broader and more augmented blends of volatile compounds. Wild strawberries are typically consumed fresh and used in the food industry for the production of jams.^[1,2] Because of the large diversity and potency of aromas occurring in natural and domesticated populations, plant breeders regard wild strawberries as important contributors of novel aroma molecules. Nevertheless, although much attention has focused on the cultivation of wild strawberries, especially in Europe and the United States,^[3] the wild strawberry in southern Shaanxi of China was mostly ignored due to its delicate and typical flavor.

In addition, wild strawberries are relatively perishable and thus often cannot be served as a fresh table fruit after transportation. Winemaking is popular worldwide as a promising way for the better utilization of both garden and wild strawberries in terms of postharvest quality preservation.^[1] Aroma is the most important and distinguishing characteristic of fruit wines, as they contribute to the sensory quality of the final wine. More than 360 volatile flavor compounds have been identified in strawberries and strawberry wine,^[4–8] and the nonvolatile compounds have also been extensively studied.^[9] From these studies, the strawberry aroma has been characterized by the combination of different compounds such as alcohols, esters, organic acids, aldehydes, ketones, terpenes, and sulfur compounds, each of which contribute to the character of different cultivars.^[3] Furthermore, as fermentation proceeds, the overall flavor of the wine is influenced by an increase in volatiles and the resulting interactions between aroma compounds, which is determined by the metabolism of yeast (Saccharomyces cerevisiae) and wine storage conditions.^[10,11]

To date, however, little information has been published on the volatiles of wild strawberry wine, especially the aroma profiles for wild strawberries grown in southern Shaanxi of China. Additionally, the understanding of how volatiles change during the course of fermentation in wild strawberry wine is incomplete. Therefore, in the current work, we aim to study and discuss the changes in the volatile components of wild strawberry wine during its fermentation. We investigate the odor-active compounds of the fruit and its wines by gas chromatography-mass spectrometry (GC-MS) combined with headspace solid-phase micro extraction (HS-SPME) to finally provide valuable information on the in-depth processing of wild strawberries.

Materials and methods

Materials

The wild strawberries were collected from the area of southern Shaanxi, China. These samples were immediately stored in ice blocks and transported to the College of Enology, Northwest A&F University, where the research was conducted under laboratory conditions. The samples were evaluated within 24 h of harvest.

Micro-fermentation

Wild strawberries were selected and then crushed into mashes (initial sugar 45.85 g/L, pH 3.85, total acidity 6.02 g/L). Sulfur dioxide was added to reach a final concentration of 50 mg/L, and pectolytic enzymes (30 mg/L, Lallemand, France) were added to the crushed pulp. Micro-fermentations were carried out at 20°C in 5 L glass containers. The commercial active dry yeast X16 (Lallemand, France) was first inoculated as suggested by the manufacturer. Then, sugar was added so that the final alcohol content would be 11% (v/v). The loss weight was measured to monitor the development of the alcoholic fermentation, with the whole fermentation lasting 11 days. After fermentation, the wine was racked, and sulfur dioxide (about 60 mg/L) was added. During the fermentation, 50 mL samples were taken during the early fermentation (the second day), the middle stage of fermentation (the fifth day), the late fermentation (the eighth day), the finished fermentation (the eleventh day), and 10 days after fermentation, and were stored at −20°C until required for further analysis.

Physicochemical parameter analysis

The physicochemical parameters of wine (alcohol content, residual sugars, pH, titratable and volatile acidity, and free and total sulfur dioxide) were determined in triplicate according to official methods.^[12]

Wild strawberries and their wine aroma analysis

Sample preparation

HS-SPME was used to gather the wine volatiles following the method proposed and validated by Zhang et al.^[13] and Wu et al.^[14] First, 1 g NaCl and 10 μL of an internal standard solution (1 g of 4-methyl-2-pentanol per 1 L of ethanol) were added into the sample vial. Afterwards, the sample vial was equilibrated at 40°C for 30 min on a magnetic stirrer hot plate. The pretreated (conditioned at 270°C for 1 h) SPME fiber (50/30 μm DVB/Carboxen/PDMS, Supelco, Bellefonte, PA) was then inserted into the headspace, to extract for 30 min with continued heating and stirring. The fiber was then immediately desorbed in the GC injector for 25 min.

GC-MS analysis

The Agilent 6890 GC was equipped with an Agilent 5975 MS and fitted with a 60 m × 0.25 mm id HP-INNOWAX capillary column with 0.25 μm film thickness (J&W Scientific, Folsom, CA) to separate and identify the volatile aroma compound.^[14] The carrier gas was helium at a flow rate of 1 mL/min, and the samples were injected by placing the SPME fiber at the GC inlet for 25 min in the split-less mode. The GC oven temperature was initially maintained at 50°C for 1 min, then was ramped to 220°C at a rate of 3°C/min, and finally was maintained at 220°C for 5 min. The mass spectrometer in the electron impact (MS/EI) energy of 70 eV was measured in an m/z range from 20 to 450. The mass spectrometer was operated in auto-tuned selective ion mode (SIM) under auto-tune conditions and the area of each peak was determined by ChemStation software (Agilent Technologies).^[13]

Qualitative analysis of volatile compounds

Volatile compounds were identified by comparing the mass spectrum data to the database (Agilent) NIST 05 library and were confirmed by comparing the retention index (RI) to those of the standards. The RI values of the unknown compounds were calculated using a modified Kovats method.^[15] Wild strawberry wine solutions with 11% (v/v) ethanol were prepared in distilled water with 3.0 g/L malic acid, and the pH was adjusted to 3.8 with NaOH. The volatile compounds were then quantified by SIM mass spectrometry using standard calibration curves. Volatile standards (supplied by Professor Duan, China Agricultural University) purchased from Aldrich (Milwaukee, Wis., U.S.A.) and Fluka (Buchs, Switzerland) were dissolved in synthetic matrices at concentrations typically found in grape wine. Volatile standards were then extracted and analyzed under the same conditions as described in the sample preparation and GC-MS analysis sections. The calibration curves for each volatile standard were created by plotting the response ratio of the target compound and the internal standard against the concentration ratio. The limits of quantification (LOQs) were set at signal-to-noise ratios of 10. Odor activity values (OAVs) were calculated by dividing the concentration by its odor thresholds from the literature.

Sensory evaluation

The sensory analysis was performed as described by Tao et al.^[16] A panel of tasters, consisting of 25 students who had been trained for 4 h per week over 2 months using the “Le Nez du Vin” aroma kit (54 aromas, Yixiangle, Hongkong), conducted the wine sensory analysis. During the strawberry wine tasting, each panelist used five or six descriptors from “Le Nez du Vin”, and they also were asked to score the intensity of each term using a 5-point scale (0 = none detected; 1 = weak; hardly recognizable note; 2 = clear but inadequate note; 3 = clear and regular note; 4 = clear and strong note; 5 = intense note). The data was used to calculate a modified frequency (MF) score that was a mixture of intensity and frequency of detection calculated with the following formula:

MF% = 5*, where F (%) is the detection frequency of an aromatic attribute expressed as percentage, and I (%) is the average intensity expressed as percentage of the maximum intensity. The descriptors with MF score greater than 1 were chosen, which seemed to effectively describe the wine.

Data analysis

All statistical analyses were performed using the SPSS statistical package version 17.0 for Windows (SPSS Inc., Chicago, IL, USA). One-way analysis of variance (ANOVA) and Duncan’s multiple range tests were applied to determine significant differences among samples using a significance level of p < 0.05. A principal component analysis (PCA) was carried out using metaboanalyst 3.0 (http://www.metaboanalyst.ca/faces/home.xhtml) with the concentrations of volatile compounds to differentiate samples.

Results and discussion

Physicochemical parameter analysis

According to the National Standards of the People’s Republic of China,^[17] wild strawberry wine belongs to the semidry wine category (residual sugar above 4.0 g/L, Table 1). The related parameters all conformed to the general enological requirements of the national standard. Compared with grape wine, this wine has a relatively high value of pH (3.71).

Table 1. Physicochemical parameters of the wild strawberry wine.

Parameters	Content
Alcohol content (%)	10.7 ± 0.3
Residual sugars (g/L)	5.96 ± 0.26
Titratable acid (g/L)	8.05 ± 0.14
Volatile acid (g/L)	0.71 ± 0.03
Free SO2 (mg/L)	12.8 ± 0.27
Total SO2 (mg/L)	76.4 ± 0.18
pH	3.71 ± 0.11

Qualitative analysis of volatile compounds in wild strawberries and their wine

Seventy-eight volatiles were identified by comparison of the mass spectra to the standard NIST05 library. The RI of the reference standards and a comparison of the RI values are reported in the literature.^[13,18] Four aroma compounds (1,9-nonanediol, 3-hydroxy-3-methylbutanoic acid, trans-(2-ethylcyclopentyl) methanol, and octanoic acid) could not be quantified due to the low signal (signal-to-noise ratios less than 10). In these 78 volatile compounds, most of them are frequently found in fruits and fruit wines. The aromatic characteristics and flavor thresholds of the related volatile compounds in berries and wine are given in the next section.

Aroma changes during wild strawberry wine fermentation

Seventy-eight volatile compounds were found in the wild strawberries and strawberry wine, including 25 alcohols, 25 esters, five ketones, four aldehydes, nine terpenes, four acids, five phenols, and one styrene. During the fermentation, the composition of aroma compounds changed over time (Fig. 1). During the early stage of fermentation, the wine showed the lowest total amount of aroma compounds (89.006 mg/L) even compared to the relatively low concentrations of the fruit before fermentation (116.74 mg/L). As fermentation progresses, the concentration of aroma compounds reached a peak at the end of fermentation (901.393 mg/L), especially for alcohols and esters (808.341 mg/L and 75.862 mg/L, respectively). The complexity of the aroma profile in wild strawberries at the end of fermentation may be caused by the higher concentrations of alcohols compared with samples at different stages of alcoholic fermentation. However, the highest concentrations of esters and ketones (aldehydes) were achieved 10 days after bottling with concentrations of 75.86–96.25 mg/L and 2.32–30.69 mg/L, respectively, which created a complex aroma profile for the wild strawberry wines. With regard to acids, although their concentrations gradually increased, they did not significantly increase the fraction of the volatile compounds during fermentation.

Figure 1. Change of volatiles through the course of fermentation; A: different kinds of volatiles in total; B: alcohols; C: other esters;D: ethyl esters; E: acids; F: aldehydes and ketones; G: volatile phenols; H: terpenes and styrene.

Alcohols

Alcohols were the most abundant volatile group in wild strawberry wine tested, which has been reported to improve aromas in other alcoholic fermented beverages.^[19,20] Normally, alcohols arise from sugar catabolism, the deamination of amino acids, and the decarboxylation of α-keto acids, followed by reduction to the corresponding aldehydes.^[21] Among the 25 alcohols identified in this study, the aroma-active alcohols (OAV > 1, Table 2) during the alcoholic fermentation of wild strawberries were mainly 2,3-butanediol, isopentanol, 2-phenylethanol, (E)-3-hexen-1-ol, and 1-octanol. Among these, 2,3-butanediol was responsible for the buttery and creamy aroma and was present in the highest amounts in the final wild strawberry wine with a relative high OAV (1.35). Such an increase was probably due to diacetyl and acetoin being converted by yeast metabolism.^[22] Accordingly, isopentanol gave an odor of whiskey, malt, and caramel. The concentration of this compound was more than its corresponding odor threshold (30 mg/L) from the middle stage of fermentation^[23] and especially increased at the end of fermentation (215.60 mg/L), and thus contributed to the strong aroma of the wild strawberry wine. In addition, 2-phenylethanol, which had a floral (rose, muscadine) aroma, was significantly enhanced at the end of fermentation (218.24 mg/L), but sharply decreased to its odor threshold (14 mg/L) 10 days after bottling. This result agrees with a previous study that showed that fermentation releases volatile phenols in a similar fashion to acid hydrolysis, which may explain the increase of 2-phenylethanol.^[24] The latter alcohols (isopentanol, 2-phenylethanol), together with the other detected aroma-inactive (OAV < 1) alcohols, such as isobutanol, 1-propanol, and 1-butanol, were formed from branched amino acids via the Ehrich pathway during fermentation. Furthermore, they contribute to the wine bouquet with a fruity aroma at concentrations below 300 mg/L, but with a pronounced pungent smell at a total concentration greater than 400 mg/L.^[25,26] A similar trend occurred in 1-octanol during fermentation, with an initial concentration of 0.008 mg/L in the unfermented fruit reaching 2.67 mg/L at the end of fermentation (odor threshold 0.12 mg/L), which imparted an overly intense citrus and rose aroma to the bouquet. Compared with those volatile compounds that reached a peak at the end of fermentation, (E)-3-hexen-1-ol and 1-octen-3-ol (OAV > 1) decreased during alcoholic fermentation. In particular, (E)-3-hexen-1-ol plummeted close to its odor threshold of 1.0 mg/L 10 days after bottling from its high initial concentration in wild strawberries (9.18 mg/L),^[3] but still provided a green aroma to the final wild strawberry wine. Consequently, (E)-3-hexen-1-ol and 1-octen-3-ol are potential impact odorants in wild strawberries.

In general, the concentration of higher alcohols declined after 10 days of bottling storage, ranging from an undesirable concentration of 808.34 mg/L to 312.88 mg/L, which has been well known for wine aroma modifications.^[18] Such an improvement in the wine aroma profile may be attributed to esters produced by the esterification of alcohols and acids during wine storage. On the other hand, decreased concentrations of higher alcohols also alleviated the strong smell during wine storage.^[18]

Table 2. Volatiles with OAV above 1 of different stages during fermentation (n = 3).

RI	Numbers	Aroma compounds	Perception Threshold (mg/L)	Fruit (mg/L)	Early (mg/L)	Middle (mg/L)	Late (mg/L)	Finish (mg/L)	10 days after bottling (mg/L)
977	1	methyl 2-methylbutyrate	0.001^c	264.0	138.0	83.0	39.0	2.0	1.0
1830	2	phenethyl acetate	0.25^a	0.0	0.0	11.9	11.9	35.4	6.9
890	3	ethyl acetate	15^b	0.2	0.5	1.3	2.3	4.1	5.5
1064	4	ethyl 2-methylbutyrate	0.001^d	89.0	63.0	68.0	96.0	64.0	55.0
1095	5	ethyl 3-methylbutyrate	0.003^d	304.3	330.0	351.3	280.0	115.7	40.3
1232	6	ethyl hexanoate	0.014^a	0.6	0.8	3.5	3.7	9.2	4.3
1437	7	ethyl octanoate	0.005^e	29.6	2.8	12.0	11.0	158.4	25.8
2055	8	ethyl cinnamate		–	–	–	–	–	–
1220	9	isopentanol	30^a	0.3	0.5	1.2	1.4	7.2	4.2
1365	10	(E)-3-hexen-1-ol	1^b	9.1	1.0	0.0	0.0	4.1	1.5
1351	11	1-octen-3-ol	0.01^c	1.8	0.7	1.1	1.2	1.5	0.6
1557	12	1-octanol	0.12	0.1	2.8	8.6	10.0	22.3	9.6
1579	13	2,3-butanediol	120^g	0.6	0.2	0.4	0.3	2.8	1.4
1926	14	2-phenylethanol	14^a	0.3	1.3	4.3	4.2	15.6	0.0
1290	15	Octanal	0.05^b	0.9	2.2	4.1	4.1	0.1	0.7
1611	16	4-methoxy-2,5-dimethyl-3(2H)-furanone	0.016^c	1.3	27.9	33.4	22.7	36.4	15.1
1651	17	phenylacetaldehyde	–	–	–	–	–	–	–
1833	18	β-damascenone	0.00005^d	60.0	60.0	20.0	–	–	140.0
2076	19	Eugenol	0.006^a	0.5	1.2	3.8	3.0	3.8	1.0
1455	20	1-heptanol	0.2^f	0.5	0.9	1.1	1.2	1.7	0.7
1462	21	6-methyl-5-hepten-2-ol	0.16^c	0.1	1.0	1.0	1.1	1.4	0.6

^aÁlvarez M.G et al. 2011; ^bMoyano, L et al. 2002; ^cDu et al. 2011; ^dGuth 1997; ^eWu et al. 2011b; ^fCai et al. 2014; ^gJiang and Zhang 2010.

Esters

Considering their relatively low odor threshold and diversity, esters are positive contributors to wine flavor, especially in young wine.^[27] Twenty-five esters were identified and quantified by GC-MS, and 18 were ethyl esters. The production of ethyl esters was first catalyzed enzymatically during fermentation and then by intracellular ethanolysis of acetyl-CoA in yeast metabolism.^[28] As the fermentation continued, ethyl esters increased significantly until 10 days after bottling (94.43 mg/L) and occupied 86.0–98.2% of the total esters at all fermentation stages (4.17–94.43 mg/L, Table 3). Of these, ethyl acetate was the most abundant ester in wild strawberry wine, with a sweet, fruity aromatic profile at concentrations lower than 80 mg/L;^[29] however, high levels of ethyl acetate (about 150 mg/L) can impart off-flavors, such as varnish or nail polish notes, which are caused by contamination with bacteria. Other aroma-active ethyl esters, including ethyl 2-methylbutyrate, ethyl 3-methylbutyrate, ethyl octanoate, ethyl hexanoate, and ethyl cinnamate, which have been characterized as typical berry and pineapple smells, also increased the complexity of the wild strawberry wine. In contrast, the concentrations of the other esters at the end of fermentation (9.55 mg/L) decreased greatly after storage for 10 days (1.82 mg/L). Besides the major ester of phenethyl acetate, the other remaining esters were methyl 2-methylbutyrate, butyl formate, hexyl acetate, methyl 2-hydroxy-3-methyl lcharacteristic of apples, pears, oranges, and other tropical fruits. Unfortunately, most of the above-mentioned esters did not reach perception thresholds in our study, but have been observed in strawberries from different regions or in different cultivars.^[3] The lactone γ-decalactone once was reported to be an intense and important strawberry aroma, but was lower than its organoleptic sensory threshold (0.088 mg/L) in the wild strawberries tested.^[30,31] As the OAVs of methyl 2-methylbutyrate, ethyl 2-methylbutyrate, ethyl 3-methylbutyrate, ethyl hexanoate, ethyl octanoate, and ethyl cinnamate were above 1, they were identified as the varietal flavor for the wild strawberries and for the wines produced by yeast metabolism, which was partly confirmed by previous studies.^{[4,6,8,9,32,33]}

Table 3. The aroma compounds identified and their concentrations in the wild strawberry wine during fermentation (mg/L).

RI^a	Quantitative ion (m/z)^b	Aroma compounds	CAS	Fruit		Early		Middle		Late		Finish		10 days after bottling
				Mean	SD	Mean	SD	Mean	SD	Mean	SD	Mean	SD	Mean	SD
977	57	methyl 2-methylbutyrate	868-57-5	0.264a	0.015	0.138b	0.078	0.083bc	0.01	0.039c	0.005	0.002c	0	0.001c	0
1172	56	butyl formate	592-84-7	0.007a	0	0.001b	0.001	0.004c	0	0.008d	0	–	–	0.004c	0
1272	43	hexyl acetate	142-92-7	0.003a	0	0.004a	0	0.340b	0.001	0.367c	0.017	0.661d	0.004	0.071e	0.005
1402	73	methyl 2-hydroxy-3-methylbutanoate	17417-00-4	0.005a	0	0.013b	0	0.013b	0	0.013b	0	0.022c	0	0.010d	–
1416	88	methyl 3-hydroxy-2-methylpropionate	80657-57-4	0.014a	0.001	0.004b	0.001	0.005b	0.001	0.003b	0	0.011c	0.001	0.005b	–
1724	85	γ-Decalactone	706-14-9	0.003a	0	0.006b	0	0.008c	0	0.005b	0	0.012d	0.001	0.005b	0.001
1830	104	phenethyl acetate	103-45-7	0.001a	0	0.006a	0	2.964b	0.052	2.966b	0.063	8.838c	0.193	1.727d	0.066
		∑other esters		0.2963a	0.016	0.172b	0.077	3.415c	0.01	3.401c	0.022	9.546d	0.006	1.822e	0.005
890	43	ethyl acetate	141-78-6	2.534a	0.156	7.526b	0.145	18.847c	0.558	34.061d	0.876	62.014e	1.874	81.865f	2.857
1064	102	ethyl 2-methylbutyrate	7452-79-1	0.089a	0.004	0.063a	0.05	0.068a	0.055	0.096 a	0.002	0.064a	0	0.055a	0.001
1095	88	ethyl 3-methylbutyrate	108-64-5	0.913ab	0.107	0.990ab	0.133	1.054a	0.031	0.840b	0.036	0.347c	0.015	0.121d	0.001
1232	88	ethyl hexanoate	123-66-0	0.009a	0	0.011a	0.001	0.049b	0.001	0.052c	0.003	0.129d	0.001	0.060e	0.001
1255	83	ethyl tiglate	5837-78-5	0.009a	0	0.005b	0	0.006c	0	0.003d	0	–	–	–	–
1346	97	(E)-ethyl-hex-2-enoate	27829-72-7	0.002a	0	0.002a	0	0.005b	0	0.004bc	0	0.003c	0	0.001d	0
1416	73	ethyl 2-hydroxy-3-methyl butanoate	2441-06-7	0.012a	0.001	0.015b	0.001	0.014b	0	0.011a	0	0.028c	0	0.012a	0
1602	73	ethyl 3-hydroxy-2-methyl butanoate	18267-36-2	0.013a	0.001	0.204b	0.002	0.178c	0.021	0.146d	0.001	0.200b	0.001	0.100e	0.006
1437	88	ethyl octanoate	106-32-1	0.148a	0.016	0.014b	0.001	0.060c	0.003	0.055c	0.002	0.792d	0.01	0.129a	0.004
1639	88	ethyl decanoate	110-38-3	0.171a	0.01	0.003b	0.001	0.019b	0.003	0.014b	0.002	0.122c	0.011	0.018b	0.001
1678	105	ethyl benzoate	93-89-0	0.002a	0	0.021ab	0	0.073bc	0.001	0.089c	0.001	0.230d	0.033	0.143e	0.04
1694	55	ethyl trans-2-decenoate	7367-88-6	0.250a	0.021	0.001b	0	0.003b	0.001	0.002b	0	0.069c	0.001	–	–
1736	85	ethyl 3-oxoheptanoate	7737-62-4	0.005a	0	0.063b	0.007	0.120c	0.001	0.088d	0.013	0.122c	0.001	0.041e	0
1816	120	ethyl 2-hydroxy-benzoate	118-61-6	0.002a	0	0.002b	0	0.012c	0	0.015d	0	0.041e	0.001	0.016d	0
1848	88	ethyl laurate	106-33-2	0.002a	0	0.001a	0	0.013b	0.005	0.011b	0	0.018c	0	0.003a	0
1898	104	ethyl 3-phenylpropionate	2021-28-5	0.001a	0	0.006a	0	0.746b	0.002	0.764b	0.001	1.937c	0.001	0.282d	0.001
2055	177	ethyl cinnamate	103-36-6	0.014a	0	0.020a	0.004	0.110a	0.003	0.132a	0.005	0.428a	0.037	11.162b	0.63
1350	45	ethyl lactate	97-64-3	–	–	0.032a	0	0.033a	0	0.032a	0.001	0.071b	0.001	0.424c	0.007
		∑ethyl esters		4.176a	0.265	8.978b	0.32	21.409c	0.576	36.416d	0.935	66.316e	1.524	94.431f	3.603
1057	31	1-propanol	71-23-8	2.320a	0.222	0.222b	0.091	0.408b	0.22	0.988 c	0.123	6.111d	0.122	3.030e	0.156
1111	43	isobutanol	78-83-1	0.804a	0.021	0.734a	0.072	2.933a	1.401	17.417b	1.381	14.057c	0.976	13.892c	0.092
1220	70	isopentanol	123-51-3	9.584a	0.154	14.631a	3.54	36.059b	4.468	40.691b	0.586	215.597c	7.821	127.297d	6.889
1255	42	1-pentanol	71-41-0	0.110a	0.004	0.120a	0.008	0.118a	0	0.056b	0.029	0.006c	0	0.002c	0
1319	57	(Z)-2-heptanol	543-49-7	0.047a	0.012	0.016b	0.001	0.017b	0	0.020bc	0.001	0.031c	0.001	0.012b	0
1355	56	1-hexanol	111-27-3	–	–	4.100a	0.276	4.602b	0.071	4.511b	0.081	6.443c	0.127	2.667d	0.067
1365	41	(E)-3-hexen-1-ol	928-97-2	9.128a	0.019	1.036bc	1.42	0.030b	0.001	0.031b	0.002	4.051d	0.145	1.546c	0.105
1387	82	(Z)-3-hexen-1-ol	33467-74-2	0.011a	0.001	0.147b	0.004	0.122c	0.018	0.094d	0	0.187e	0	0.008a	0
1392	59	3-octanol	589-98-0	0.025a	0.001	0.016b	0.001	0.033c	0.001	0.049d	0.002	0.072e	0	0.027a	0.001
1351	57	1-octen-3-ol	3391-86-4	0.018a	0.001	0.007b	0	0.011c	0.001	0.012c	0	0.015d	0	0.006b	0.001
1455	70	1-heptanol	111-70-6	0.109a	0.002	0.172b	0.005	0.216c	0.009	0.235c	0.004	0.343d	0.02	0.132a	0.004
1462	95	6-methyl-5-hepten-2-ol	1569-60-4	0.009a	0.001	0.163b	0.001	0.167b	0.007	0.171b	0.001	0.224c	0.007	0.101d	0.015
1490	57	2-ethyl-1-hexanol	104-76-7	0.033a	0.003	0.005b	0	0.010c	0.001	0.013c	0	0.016d	0.001	0.006b	0.001
1502	81	cis-4-hepten-1-ol	6191-71-5	0.121a	0.014	0.015b	0	0.010b	0	0.020b	0	0.012b	0	0.007b	0
1517	45	2-nonanol	628-99-9	–	–	–	–	0.002	0	0.003	0.001	0.005	0	0.002	0
1557	56	1-octanol	111-87-5	0.008a	0.001	0.338b	0.018	1.032c	0.005	1.199c	0.012	2.671d	0.001	1.154c	0.182
1579	45	2,3-butanediol	123513-85-9	72.6a	22.344	20.140b	6.746	52.051c	14.912	40.415bc	10.185	338.919d	4.942	162.426e	4.478
1615	67	(Z)-5-octen-1-ol	64275-73-6	0.003a	0	0.077b	0.002	0.134c	0.004	0.125d	0.001	0.178e	0	0.071f	0.002
1662	56	1-nonanol	143-08-8	0.025a	0.006	0.020a	0	0.064b	0.004	0.072b	0.002	0.213c	0.008	0.085d	0.004
1765	70	1-decanol	112-30-1	0.005a	0.001	0.001b	0	0.006a	0	0.011c	0	0.062d	0	0.024e	0.002
1890	79	benzyl alcohol	100-51-6	0.039ab	0.002	0.277c	0.006	0.049ad	0.008	0.036b	0.002	0.052d	0.002	0.087e	0.003
1926	91	2-phenylethanol	60-12-8	4.023a	0.322	17.916b	0.502	60.430c	0.774	58.925c	4.922	218.237d	11.278	0.052a	0.005
2046	91	3-phenylpropanol	122-97-4	0.014a	0.001	0.208b	0.021	0.533c	0.008	0.390d	0.063	0.840e	0.004	0.243b	0.002
		∑alcohols		99.036a	22.816	60.359b	1.58	159.038c	20.19	165.483c	13.196	808.341a	0.322	312.878d	11.486
1146	83	mesityl oxide	141-79-7	0.007a	0.001	0.002ab	0.001	0.008a	0.004	0.006ab	0.004	0.003ab	0	0.001b	0
1187	57	diisobutyl ketone	108-83-8	0.076a	0.012	0.082ab	0.019	0.099ab	0.012	0.107b	0.001	0.034c	0.001	0.152d	0.013
1290	43	octanal	124-13-0	0.047a	0.001	0.109b	0.016	0.207c	0.009	0.207c	0.027	0.005d	0	0.037ad	0.001
1298	45	acetoin	513-86-0	11.820a	0.383	15.941b	0.381	12.982c	0.423	0.354d	0.025	0.667d	0.009	29.815e	0.649
1325	43	6-methyl-5-hepten-2-one	110-93-0	0.007a	0.001	0.012b	0.001	0.014c	0.001	0.014c	0	0.019d	0.001	0.009e	0.001
1394	57	1-nononal	124-19-6	0.014ab	0.001	0.007c	0.001	0.013a	0.001	0.015b	0.001	0.003d	0	0.001e	0
1534	77	benzaldehyde	100-52-7	0.003a	0	0.146b	0.003	0.394c	0.01	0.388c	0.006	0.968d	0.062	0.410c	0.001
1611	43	4-methoxy-2,5-dimethyl-3(2H)-furanone (DMMF)	4077-47-8	0.020a	0	0.447b	0.005	0.534c	0.055	0.363d	0.041	0.582c	0.012	0.241e	0.005
1651	91	phenylacetaldehyde	122-78-1	0.063a	0.002	0.016b	0.001	0.034c	0.001	0.029d	0.001	0.046e	0.004	0.022f	0.001
		∑aldehydes and ketones		12.058a	0.395	16.762b	0.334	14.286c	0.473	1.483d	0.045	2.326d	0.066	30.687e	0.659
1164	93	β-myrcene	127-91-3	0.008a	0	0.008a	0.004	0.020b	0.001	0.019b	0.001	0.019b	0	0.008a	0
1193	68	limonene	5138-86-3	–	–	0.006a	0	0.009b	0.001	0.001c	0	0.004d	0	–	–
1240	93	(Z)-β-ocimene	3338-55-4	0.007a	0	0.010b	0	0.020c	0.001	0.001d	0	0.050e	0.002	0.019c	0
1290	93	terpinolene	586-62-9	–	–	0.001a	0	0.002a	0.001	0.002a	0.001	0.005b	0	0.002a	0
1547	71	linalool	78-70-6	0.003a	0	0.007b	0.001	0.018cd	0.002	0.021c	0	0.036e	0.001	0.016d	0.002
1703	59	α-terpineol	10482-56-1	0.007a	0.001	0.003b	0.001	0.006a	0	0.007a	0	0.017c	0.001	0.008a	0.001
1770	69	citronellol	106-22-9	0.019a	0.001	–	–	0.001b	0	0.001b	0	0.016a	0.002	0.012c	0.003
1833	69	β-damascenone	23696-85-7	0.003a	0	0.003a	0	0.001b	0	–	–	0.010c	0.001	0.007d	–
1864	43	geranylacetone	689-67-8	0.002a	0	0.003b	0	0.008c	0	0.013d	0	0.018e	0.001	0.008c	0
		∑terpenes		0.049a	0.001	0.041a	0.007	0.085b	0.002	0.065c	0	0.175d	0.004	0.080b	0.006
1590	43	isobutyric acid	79-31-2	0.618a	0.023	1.803b	0.012	3.369c	0.262	3.403c	0.2	12.680d	0.915	7.518e	0.165
1668	74	2-methylhexanoic acid	4536-23-6	0.147a	0.008	0.428b	0.01	0.512bc	0.042	0.475bc	0.034	0.990d	0.066	0.542c	0.003
		∑acids		0.765a	0.03	2.232b	0.002	3.880c	0.304	3.878c	0.235	13.670d	0.848	8.059e	0.168
1746	138	1,2-dimethoxybenzene	91-16-7	0.005a	0	0.009b	0.001	0.021c	0	0.019c	0	0.031d	0.002	0.012e	0
2020	94	phenol	108-95-2	0.031a	0.002	0.015b	0.001	0.032a	0	0.029a	0.003	0.065c	0.002	0.115d	0.007
2027	164	isoeugenol	97-54-1	0.015a	0.001	0.016a	0.002	0.052b	0.001	0.048b	0.004	0.074c	0.001	0.030d	0.001
2049	137	4-ethyl guaiacol	2785-89-9	0.011a	0.002	0.009a	0.001	0.043b	0.001	0.051b	0.005	0.105c	0.008	0.144d	0.005
2076	164	eugenol	97-53-0	0.003a	0.001	0.007b	0.001	0.023c	0.001	0.018d	0.002	0.023c	0	0.006b	0
		∑phenols		0.068a	0.002	0.063a	0.006	0.187b	0.004	0.179b	0.015	0.320c	0.009	0.315c	0.003
1263	104	styrene	100-42-5	0.291a	0.008	0.403b	0.01	0.563c	0.038	0.527c	0.006	0.710d	0.005	0.422b	0.011
		∑total volatiles		116.74a	23.47	89.006b	1.464	202.852c	21.589	211.422c	14.398	901.393d	2.624	448.692e	15.919

^aRetention indices were calculated on HP-INNOWAX column. ^bQuantitative ion for peak area evaluation of volatile compounds. Results are the means of three independent experiments ± standard deviation. Values with different letters in the same row are significantly different according to the Duncan test (p < 0.05).

Acids

Organic acids are not only the predominant precursors of esters but also the major compounds that confer fruity, fatty, and rancid notes to wine.^[34] Four acids, isobutyric acid, 2-hydroxy-2-methyl butanoic acid, 2-methylhexanoic acid, and octanoic acid, were identified in the present study. Nevertheless, only isobutyric acid and 2-methylhexanoic acid were quantified because 2-hydroxy-2-methylbutanoic acid and octanoic acid could only be detected without quantification or analyzed qualitatively, respectively, as they were only present in trace amounts at the mid-fermentation stage. Although the concentration of acids increased gradually until the fermentation process completed with a peak concentration of 13.67 mg/L in this study, isobutyric acid was the only odor-active compound that contributed a cream and cheese aroma (OAV > 1). In addition, 50% of the total acid concentration was depleted rapidly after 10 days of bottle aging to an appropriate concentration (8.059 mg/L) for a pleasant and fresh aroma. These changes are similar to those in higher alcohols, two of which were used for synthesizing the corresponding esters during wine storage to further ameliorate the pungent smell of the wine after fermentation.

Aldehydes and ketones

Aldehydes and ketones are additional types of main volatile compounds with fruity aromas that varied at different stages of the wine fermentation. According to our study, five ketones (mesityl oxide, diisobutyl ketone, acetoin, 6-methyl-5-hepten-2-one, and 4-methoxy-2,5-dimethyl-3(2H)-furanone (DMMF)) and four aldehydes (octanal, nonanal, benzaldehyde, and phenylacetaldehyde) were detected and analyzed. Among them, the majority has been observed in earlier studies with only four aroma-active volatiles in this study: octanal, DMMF, 6-methyl-5-hepten-2-one, and phenylacetaldehyde.^[9,31] DMMF is noteworthy as an impact compound in strawberries with a very low perception threshold (0.016 mg/L) and was first reported in wild strawberries by Pyssalo et al.^[35] DMMF is generally considered to have a “cotton candy” or “caramel” descriptor, conveying sweetness to the aroma of the wild strawberry wine.^[3] Interestingly, the concentration of DMMF was 10 times higher after 10 days of bottle storage than in wild strawberries before fermentation (0.020 mg/L). Such aglycones are dependent on the fruit development stage and are especially prominent in a free form during fruit ripening.^[36] Unlike the aforementioned active aroma compounds that were at low concentrations in all stages, a large proportion of the carbonyl compounds were acetoin and benzaldehyde, but the typical cheese and butter aromas of acetoin were still imperceptible due to their quite high odor threshold (150 mg/L). Specifically, the concentrations of aldehydes and ketones did not have a consistent trend during the enological process. For example, the amount of acetoin increased first from 11.820 mg/L in the fruit slurry to 15.941 mg/L in the wine during the early fermentation and then declined; benzaldehyde had a similar trend to acetoin, although its peak concentration was on the eleventh day, followed by a decrease during bottle storage. This decrease in benzaldehyde might be caused by oxygen from the press and filter process that may have oxidized benzaldehyde to benzoic acid.

Volatile phenols

In the fruit and related wine samples, phenols were only present in minor quantities. Two different kinds of phenols, vinylphenols and ethylphenols, were affected by Brettanomyces/Dekkera yeasts in wine and were responsible for chemical, medicinal smells and animal, smoky odors at low levels, respectively.^[37,38] In total, five phenols were detected in this study, including 4-ethyl guaiacol, eugenol, 1,2-dimethoxybenzene, phenol, and isoeugenol, and barely contributed (0.36–0.92‰) to the overall content of volatile compounds without deterioration in wine organoleptic quality. When compared with the wild strawberries prior to fermentation, the amount of phenols after bottle-aging significantly increased although to relatively low concentrations (from 0.07 to 0.32 mg/L). Of these phenols, only eugenol exceeded the perception threshold (0.005 mg/L),^[39] which contributes a dianthus and musk smell to the wild strawberry wine.

Terpenes

Isoprenoid monoterpenes, formed from the precursor mevalonate and acetyl-CoA from the plant, are associated with varietal aromas in fruits and in finished wines.^[40] Nine terpenes, which have floral and fruity aromas, were detected in small amounts (0.041–0.175 mg/L) at different stages of fermentation, including linalool, limonene, citronellol, β-terpineol, geranylacetone, and β-damascenone. Of these, notwithstanding the relatively low concentrations of β-damascenone (0–0.01 mg/L) and other terpenes, their concentrations remained near their sensory thresholds of 0.05 μg/L, imparting honey and apple aromas to the overall bouquet.^[41]

During fermentation, the concentration of terpenes increased as the maceration progressed and declined after separation of the skin residue from wine liquid, ending at a concentration 62% higher than in wild strawberries. Nevertheless, the yeast did not predominantly produce terpenes, but synthesized sterol using terpenes as an intermediate.^[42] Additionally, these compounds increased at the end of fermentation due to lagged releases of aglycones by the β-glucosidase from the yeast or hydrolysis by acids of the bound form in agreement with earlier studies.^[9,43] Another volatile compound, styrene, with a resin and flowery flavor, was also detected and quantified with significant differences among the samples at different fermentation stages. As the fermentation of wild strawberry wine progressed, the concentration of styrene increased first and then reached the highest concentration (0.71 mg/L) when the fermentation finished, while the concentration declined by 40.6% after 10 days of bottle storage.

Principal component analysis

To compare the relative differences in volatile aroma compounds of wild strawberries and their wine during fermentation, a PCA analysis was performed according to 21 selected volatiles with OAV above 1 (Table 4). As the data set shows, 76.3% of the total variance was explained by the first two principal components (PC1 and PC2), while 49.7% and 26.5% of the variance was accounted for by PC1 and PC2, respectively (Fig. 2). The wild strawberry and its wines were separated using all the variances between the two. PC1 positively correlated with volatile compounds, phenethyl acetate, 6-methyl-5-hepten-2-ol, 1-octanol, ethyl hexanoate, and DMMF, but negatively correlated with some esters and alcohols, which were the main contributions of the wild strawberry group itself. This result showed that volatile components such as methyl 2-methylbutyrate, ethyl 2-methylbutyrate, ethyl 3-methylbutyrate, (E)-3-hexen-1-ol, 1-octen-3-ol, and phenylacetaldehyde, were significantly different between the raw materials and their wines in agreement with the previous results. Nevertheless, ethyl 3-methylbutyrate and 1-heptanol are located on the negative part of PC2, which means they were both present in wild strawberries and finished wines stored for 10 days. PC2, which was formed by two groups of volatile compounds including ethyl hexanoate, ethyl acetate, ethyl cinnamate, ethyl octanoate, isopentanol, 2,3-butanediol, and β-damascenone on the positive axis, and 2-phenylethanol, octanal, DMMF, and eugenol on the negative axis, was strongly associated with the states of wine fermentation, in particular 10 days after bottling and the middle stage of fermentation, respectively. Although the finished wines were loaded in the same quadrants as wines after storage, their distinct characteristics were reflected by the volatile compounds, including phenethyl acetate, 1-octanol, DMMF, and 6-methyl-5-hepten-2-ol, according to their PC1 scores on the positive part of the axis. Similarly, the middle and late stages of fermentation showed the same volatiles that strongly related to a positive PC1, and further contributions by 2-phenylethanol, DMMF, and eugenol increased the score on the negative side of PC2 when compared with the remaining groups. Finally, the group of wines at the early stage was scattered in the third quadrant with low PC1 scores, indicating that it was less characterized by the PC1 volatiles, but the volatiles with negative PC2 scores may help differentiate this pattern. Hence, according to PCA analysis, the fermentation stages could be differentiated according to their corresponding key odor compounds.

Table 4. Concentrations of volatiles with OAV above 1 and its perception thresholds of different stages during fermentation (n = 3).

RI	Numbers	Aroma compounds	Quantitative standards	Perception Threshold (mg/L)	Fruit(mg/L)	Early(mg/L)	Middle (mg/L)	Late(mg/L)	Finish (mg/L)	10 days after bottling (mg/L)
977	1	methyl 2-methylbutyrate	methyl 2-methylbutyrate	0.001^c	0.264	0.138	0.083	0.039	0.002	0.001
1830	2	phenethyl acetate	phenethyl acetate	0.25^a	0.001	0.006	2.964	2.966	8.838	1.727
890	3	ethyl acetate	ethyl acetate	15^b	2.534	7.526	18.847	34.061	62.014	81.865
1064	4	ethyl 2-methylbutyrate	ethyl 2-methylbutyrate	0.001^d	0.089	0.063	0.068	0.096	0.064	0.055
1095	5	ethyl 3-methylbutyrate	ethyl 2-methylbutyrate	0.003^d	0.913	0.99	1.054	0.84	0.347	0.121
1232	6	ethyl hexanoate	ethyl hexanoate	0.014^a	0.009	0.011	0.049	0.052	0.129	0.06
1437	7	ethyl octanoate	ethyl octanoate	0.005^e	0.148	0.014	0.06	0.055	0.792	0.129
2055	8	ethyl cinnamate	–		0.014	0.02	0.11	0.132	0.428	11.162
1220	9	isopentanol	1-butanol	30^a	9.584	14.631	36.059	40.691	215.597	127.297
1365	10	(E)-3-hexen-1-ol	(E)-3-hexen-1-ol	1^b	9.128	1.036	0.03	0.031	4.051	1.546
1351	11	1-octen-3-ol	1-octen-3-ol	0.01^c	0.018	0.007	0.011	0.012	0.015	0.006
1557	12	1-octanol	1-octanol	0.12	0.008	0.338	1.032	1.199	2.671	1.154
1579	13	2,3-butanediol	2,3-butanediol	120^g	72.6	20.14	52.051	40.415	338.919	162.426
1926	14	2-phenylethanol	2-phenylethanol	14^a	4.023	17.916	60.43	58.925	218.237	0.052
1290	15	Octanal	Octanal	0.05^b	0.047	0.109	0.207	0.207	0.005	0.037
1611	16	4-methoxy-2,5-dimethyl-3(2H)-furanone	–	0.016^c	0.02	0.447	0.534	0.363	0.582	0.241
1651	17	phenylacetaldehyde	phenylacetaldehyde	–	0.063	0.016	0.034	0.029	0.046	0.022
1833	18	β-damascenone	–	0.00005^d	0.003	0.003	0.001	–	0.010c	0.007
2076	19	Eugenol	Eugenol	0.006^a	0.003	0.007	0.023	0.018	0.023	0.006
1455	20	1-heptanol	1-heptanol	0.2^f	0.109	0.172	0.216	0.235	0.343	0.132
1462	21	6-methyl-5-hepten-2-ol	6-methyl-5-hepten-2-ol	0.16^c	0.009	0.163	0.167	0.171	0.224	0.101

^aGonzalez Alvarez et al. 2011; ^bMoyano et al. 2002; ^cDu et al.^{[3]; d}Guth^{[39]; e}Wu et al. 2011b; ^fCai et al. 2014; ^gJiang and Zhang 2010.

The data represents the average value.

Figure 2. Principal component analysis (PCA) of volatile compounds with OAV above 1 in wines by GC/MS. These volatiles were numbered in sequence according to Table 3. The data represents the average value.

Sensory analysis of wild strawberry wine

Sensory analysis differs from the OAV approach in two ways. First, volatiles perceived by the organoleptic descriptors are examined in the food and wine matrix. In addition, the components are not separated before sensory analysis when evaluating the predominant aromatic and structural character of the wine. For example, both volatile and nonvolatile components are assessed simultaneously in sensory analysis, whereas only volatiles are assessed by headspace techniques in OAV. During the process of sensory analysis, 19 aroma terms were used with MF more than 1, with the MF of these terms shown in Fig. 3. The sensory profiles of the samples confirmed a marked difference between the fruit and wine. The wines at the end of fermentation showed significant differences from the wild strawberry fruit according to their fruity and floral notes, such as coconut and honeysuckle instead of aromas of orange, cherry, vanilla, and plum. On the contrary, five of the 19 sensory attributes (yoghurt, honey, strawberry, apple, and cream) were described by both samples; therefore these smells may be characteristic varietal flavors of wild strawberries.^[6] In general, sensory analysis highlighted the characteristics of wild strawberry and its wines with different aroma descriptors.

Figure 3. Descriptive analysis of aroma profile in wild strawberry fruit and wine. Wild strawberry fruit – triangle markers and wines – square markers, evaluated by the tasting panel.

Conclusion

In conclusion, among the flavor compounds of the wild strawberry, volatile compounds are considered important with OAV above 1, influencing the aroma of the studied wine greater than the other compounds. In this study, the concentrations of six volatiles (methyl 2-methylbutyrate, ethyl 2-methylbutyrate, ethyl 3-methylbutyrate, (E)-3-hexen-1-ol, 1-octen-3-ol, and phenylacetaldehyde) were higher than their odor thresholds and helped differentiate wild strawberry and its wines, indicating that they may be the typical flavor substances of the wild strawberry. Compared to the fruit, wild strawberry wine had six more compounds (ethyl hexanoate, ethyl acetate, ethyl cinnamate, isopentanol, 2,3-butanediol, and β-damascenone) as characteristic aroma compounds. By taste evaluation, the wild strawberry wine was shown to be one feasible way to preserve wild strawberries and their sensory characteristics and to avoid spoiling in storage or transportation.

Declaration of interest

The authors declare no conflict of interest.

Acknowledgment

All authors have agreed to submit this manuscript to the International Journal of Food Properties.

Funding

This work received financial support from the National Natural Science Foundation of China (31571812, 31501463), the Agriculture Research System of China (CARS-30-jg-03), and Fundamental Research Funds of the Central Universities (2014YB045).

Additional information

Funding

This work received financial support from the National Natural Science Foundation of China (31571812, 31501463), the Agriculture Research System of China (CARS-30-jg-03), and Fundamental Research Funds of the Central Universities (2014YB045).