Volatile aroma and sensory analysis of black raspberry wines fermented by different yeast strains

Research article Received: 15 January 2014 Revised: 29 September 2014 Accepted: 2 November 2014 Published online in Wiley Online Library: 27 January 2015 (wileyonlinelibrary.com) DOI 10.1002/jib.199 Volatile aroma and sensory analysis of black raspberry wines fermented by different yeast strains Byoung-Ho Kim 1,2 and Seung K. Park 2 * Robus coreanus Miquel is a small berry fruit used for Korean black raspberry (KBR) wine-making. Twelve different yeast strains were investigated by laboratory-scale fermentation to develop a wine with a high flavour quality. Volatile aroma compounds from the wines were analysed using headspace solid phase microextraction gas chromatography mass spectrometry and sensory evaluation was performed to evaluate the flavour characteristics. The volatile aroma compounds that mostly contributed to the flavour of KBR wines were those related to fruity (esters) and floral (terpenes) aromas. Fifteen out of the 67 identified volatile compounds showed higher odour activity values than other compounds in the wines, and these compounds were considered as important contributors to the final aromas of the wine. Additionally, the KBR wine fermented by the M1 yeast strain had the highest sensory preference because of higher fruity and floral aroma characters compared with other wines. In addition to the M1 strain, the other yeast strains that produced favourable sensory characteristics included Enoferm CSM, Uvaferm VRB, ICV GRE, ICV Opale and LevureSeche Active. Of these strains, the M1 strain produced a particularly excellent black raspberry wine, and thus could be applied for further large-scale production of black raspberry wines. It is also expected that this work will expedite research on the production of high-quality black raspberry wines with beneficial physicochemical properties, functionality and good sensory characteristics. Copyright 2015 The Keywords: black raspberry wine; aroma analysis; fermentation; HS-SPME-GC-MS; odour activity value Introduction Robus coreanus Miquel is a shrub of the Rosaceae family that grows in various regions of Asia, including Korea, China, and Japan (1). Various bioactive compounds have been isolated and identified from this fruit; these compounds, include phenolic acids, triterpenosides, flavonoids and ellagitannin, and these are known to have beneficial physiological effects, such as cholesterol and glucose-lowering, anti-inflammatory, anti-allergic and anti-oxidative activities (2 6). Therefore, R. coreanus consumption in the form of juices, functional foods and alcoholic beverages has increased. Because of the commercial value and wide range of applications of R. coreanus berries, it would be beneficial to develop new functional black raspberry wines with good sensory acceptability. A recent study reported data on yeast selection, sensory evaluation and instrumental analysis of volatile and other compounds from red raspberry (R. idaeus L.) fermented with 16 yeasts strains (7), but very few reports have compared the composition and sensory properties of the active aromatic compounds of KBR wines fermented with different yeast strains. The subtle balance of aromatic compounds in alcoholic beverages is often used as an organoleptic fingerprint for specific products and brands (8). The formation of volatile compounds during alcoholic fermentation depends not only on the particular yeast species, but also on the particular strain (9). The vital and fundamental compounds contributing to wine flavour are produced during alcoholic fermentation (10). Must fermented with different Saccharomyces cerevisiae strains can develop diverse aromatic profiles (11). It is therefore important to identify the differences between volatile compounds produced by various yeast strains in order to select the best strain for production of the desired wines (12). Several different volatile compounds can be present in a given wine; however only compounds at concentrations above or around their odour threshold contribute to the unique wine aroma. The odour activity value (OAV) is a useful parameter to assess the relative importance of individual chemical components present in a wine. The OAV is generally expressed as the ratio between the concentration of the volatile compound and its odour threshold. Aromatically active compounds are volatile compounds whose concentrations in wine are above their odour threshold (OAV > 1). However, a particular compound at a concentration of OAV < 1 may also contribute to the overall aroma perception because of certain additive or synergistic effects between compounds with similar aromatic natures (13). To understand the chemical compounds that exhibit important sensory characteristics in wine, it is necessary to determine the effects of the volatile compounds on the sensory properties of wines. Therefore, identification of active aromatic compounds in wines * Correspondence to: S. K. Park, Department of Food Science and Biotechnology, KyungHee University, Yongin-si, 446-701, Republic of Korea. E-mail: skpark@khu.ac.kr 1 Hitejinro R & D Centre and Hitejinro Brewery Co. Ltd, Chungcheongbuk-do, Cheongwon-gun, 363-823, Republic of Korea 2 Department of Food Science and Biotechnology, KyungHee University, Yongin-si, 446-701, Republic of Korea 87 J. Inst. Brew. 2015; 121: 87 94 Copyright 2015 The

88 using sensory and instrumental data is of utmost importance. To select the optimal yeast strain for a desired wine quality, more research on the correlation between volatile compounds and sensory properties is required. The objectives of this study were to investigate the effects of different strains of S. cerevisiae on the volatile compounds of wine made from KBR using headspace-solid phase microextraction gas chromatography mass spectrometry (HS- SPME-GC-MS) and to identify the optimal yeast strain for production of high-quality black raspberry wine using instrumental and sensory analysis. Materials and methods Black raspberry pulp Ripe raspberry fruit, harvested in June 2011, was purchased from a farm (35 44 49 N, 126 70 27 E) located in Jeongeup, Korea, and was kept at 20 C in a freezer until used. Before use, frozen black raspberry fruit was thawed in an incubator (SI-1000R, JEIO TECH, Korea) at 25 C and manually ground to a fine pulp. The raspberry pulp contained seeds and pulp residues and was not filtered. The initial Brix value of the pulp, measured using a hand refractometer (ATAGO, Japan), was 8 Brix and the ph was 3.5. The pulp was mixed with sucrose to adjust the sugar concentration to 24 Brix. Yeast strains Twelve different strains of S. cerevisiae were tested 71B, V1116, M1, Windsor, Enoferm CSM, Uvaferm VRB, ICV GRE, BM 4X4, QA23, ICV D47, ICV Opale and Levure Seche Active; 11 commercial strains from Lallemand (Montreal, QC, Canada) and one yeast strain from Institut Oenologique de Champagne (Epernay, France) were obtained as active dry yeasts. The 11 yeast strains are used in the commercial production of wine and one strain (Windsor) is used for ale beer production. The strain Levure Seche Active yeast is widely used for the production of black raspberry wine in Korea. These strains were recommended by the commercial suppliers for enhancing the floral and fruity notes of wine and for their resistance to high alcohol levels. Starter culture for black raspberry wine Each strain was activated by inoculating into a 2% sucrose solution followed by incubation for 4 h at 30 C. The activated strains were then inoculated into 100 g of defrosted raspberry pulp and incubated for 3 days at 25 C for starter cultures of raspberry wine. The inoculum concentration for all strains was 200 mg (dry weight)/kg raspberry pulp. Black raspberry wine production After production of starter cultures of raspberry wines for 3 days at 25 C, 400 g of the thawed pulp and 64 g (additional mass of sucrose to adjust the sugar concentration to 24 Brix) of sucrose were added to each starter culture in 1 L Erlenmeyer flasks. Raspberry musts were fermented for 10 days at 25 C. The fermentation was considered complete when the weight reduction of the fermentation mash was constant. At the end of fermentation, fermented musts were centrifuged at 3400 g for 10 min to B. H. Kim and S. Park remove cell debris, and the resulting wines were stored at 4 C in glass bottles filled completely to avoid oxygenation. All experiments were carried out in triplicate. HS-SPME-GC-MS analysis A small portion of fused silica fibre with a polymeric phase was exposed to the headspace to extract the aroma. A 100 μm polydimethylsiloxane-divinylbenzene (PDMS-DVB) SPME fibre (Supelco, Inc., Bellefonate, USA) was used because it shows high affinity for volatile compounds from various alcoholic beverages (14,15). Five millilitres of black raspberry wine was hermetically sealed in a 10 ml vial having a silicone septum and an aluminium cap (Supelco, Bellefonate, USA). A stainless steel needle containing a PDMS-DVB fibre was inserted through the septum of the sample vial for 30 min at 60 C to extract the headspace volatiles using a Gestel SPME autosampler (MPS-3, Gerstel GmbH, Műlheim an der Ruhr, Germany). Thereafter, the volatile compounds were desorbed for 5 min into the injector of an Agilent 5975 GC/MS(Santa Clara, CA, USA) equipped with a DB-WAX fused silica capillary column (0.25 mm 30 m i.d., film thickness 0.25 μm). The oven temperature was held at 45 C for 3 min, then programmed to run from 45 to 230 C at 20 C/min, and then held at 230 C for 10 min. The detector was operated in the electron-impact mode (70 ev), and mass spectra were acquired by scanning over the mass/charge (m/z) range of 40 500 with an acquisition rate of 3.15 scans/s. Volatile compounds were identified by comparing their retention indices and matching their mass spectra with those of reference compounds in the data system of the Wiley library and NIST Mass Spectral Search Program (ChemSW Inc., NIST 98 Version Database) connected to an Agilent 5975 mass spectrometer. The results are expressed as the mean ± standard deviation of three samples. OVAs The contribution of each volatile compound to the black raspberry wine aroma was evaluated qualitatively via its associated descriptor and quantitatively via measurement of the OAV. OAVs were calculated using the equation OAV = c/t, where c is the total peak area or concentration of each compound in the wine sample and t is the odour threshold value of the compound in water/ethanol solution (16). Odour threshold values were obtained from the literature (7,10,12,17 21) and were also determined in 14% ethanol solution using a triplicate evaluation with 25 panellists. Sensory analysis Samples fermented by the 12 different yeast strains were analysed in triplicate by 25 panellists trained at Hitejinro Brewery Co. Ltd. These were experienced testers in the field of manufacturing KBR wine in Korea. Sensory evaluation was performed using the quantitative descriptive (QDA) methodology (25). A constant volume of 30 ml of each wine was evaluated in a wine tasting glass covered with a glass Petri dish at 12 C. During the analysis, the wine tasters indicated different perceived descriptors and the intensity of each attribute was rated on a scale from 0 to 5, where 0 indicates that the descriptor was not perceived and values 1 5 indicate that its intensity was very low, low, medium, high or very high, respectively. The mean scores were obtained and plotted as a polygonal graph. wileyonlinelibrary.com/journal/jib Copyright 2015 The J. Inst. Brew. 2015; 121: 87 94

Volatiles in Korean black raspberry wine Statistical analysis All experiments were repeated three times. Statistical analyses were calculated using analysis of variance (ANOVA) and significance was set at p < 0.05. Principal component analysis (PCA) was performed using the SPSS version 20 software (SPSS Inc., Chicago, IL, USA). Comparative analyses were performed using the Duncan s test for significant results. Results and discussion Analysis and identification of volatile compounds in black raspberry wines by HS-SPME-GC-MS Sixty-seven volatile compounds were identified by HS-SPME-GC- MS from black raspberry wines fermented by 12 different yeast strains (Table 1). The volatile compounds included 25 alcohols, 11 hydrocarbons, 24 esters, two aldehydes, three acids and two miscellaneous compounds. Several other unidentified compounds were also present. The most abundant volatile compounds were 3-methyl-1-butanol, ethyl tetradecanoate, 2- phenylethanol, butyl benzoate, dimethyl styrene and myrtenol. The ethyl ester group contained the following compounds: ethyl acetate, 3-methylbutyl acetate, ethyl hexanoate, dimethyl styrene, butyl benzoate, phenylethyl acetate, ethyl dodecanoate, ethyl tetradecanoate, ethyl cinnamate, ethyl palmitate, ethyl stearate, ethyl oleate, ethyl linoleate and ethyl linolenate. Ethyl esters are one of the most important groups of aromatic compounds in black raspberry wines, and their concentrations depend on the yeast strain, fermentation temperature, aeration and sugar content (7). Ethyl hexanoate (apple, fruit and sweet), ethyl benzoate (fruity, floral) and ethyl cinnamate (white flower) were the ethyl esters expected to have the greatest influence on the aromas of black raspberry wines because of their extremely low odour thresholds (0.001 0.05 mg L 1 ). Compounds were also identified from the terpene group in the black raspberry wines. When compared with terpenes present in other raspberry wines reported in the literature (25), the composition of terpenes in the KBR wine was slightly different. Shamaila et al. (25) reported that red raspberry wine cultivars contain the terpenes α-pinene, γ-terpinene, sabinene, caryophyllene, α-ionone, β-ionone and α-ionol, whereas black raspberry wines mainly contain β -pinene, limonene, α-pinene, ο-cymene, β-linalool, terpene-4-ol, α-terpineol, β-citronellol, mytenol, verbenone and geraniol. Among these compounds, terpene-4-ol (pine tree-like, flowers) had the greatest effect on the sensory characteristics owing to its odour threshold (0.005 mg L 1 ). The only fatty acid identified in our black raspberry wines was octanoic acid (fatty acid, cheese, harsh and rancid) and it was present at high concentrations in all of the wines studied. This result was consistent with previous literature (7,26). For all raspberry wines, octanoic acid was measured above the odour threshold of 5 mg/l proposed by other authors (21,27). Fatty acids such as octanoic acid have a negative effect on the sensory characteristics of wines, and this compound is associated with the odour descriptors harsh and rancid (28). Five fatty acids ethyl esters were identified in our wines, namely, ethyl palmitate, ethyl oleate, ethyl linoleate, ethyl linolenate and ethyl stearate. The fatty acids ethyl esters are produced by formation of ester bonds between ethanol produced during fermentation and palmitic acid, oleic acid, linoleic acid, linolenic acid or steric acid present in black raspberry fruit. These compounds are associated with the odour descriptor waxy (17). The most predominant alcohols obtained were 1-propanol, 2-methyl-1-propanol and 3-methyl-1-butanol. Among these compounds, 3-methyl-1-butanol was present at the highest concentration. Alcohols can have both positive and negative effects on the aroma and flavour of wine depending on their concentrations; they are considered favourable at concentrations <300 mg/l (7,29,30). Correlation between volatile compounds and OAVs Table 1 shows the odour descriptors, odour thresholds and OAVs for each compound identified by HS-SPME-GC-MS. High OAVs, which indicate that the influence of the compound on the wine s aroma is high, were exhibited by ethyl acetate, 3-methylbutyl acetate, ethyl hexanoate, ethyl decanoate, ethyl benzoate, ethyl tetradecanoate, ethyl cinnamate, β-pinene, limonene, α-pinene, ο-cymene, β-linalool, terpene-4-ol, α-terpineol and β-citronellol. These compounds were mainly esters and terpenes, and contributed to the floral and fruity aromas of the wines. The highest total OAV was found in the M1 strain ( OAV = 8.82 10 9 ), whereas the lowest total OAV was found in the V1116 strain ( OAV = 4.65 10 9 ). Strains Uvaferm VRB, ICV Opale, and Levure Seche Active also showed high OAVs. M1 and Enoferm CSM had high OAVs because they abundantly produced active aromatic compounds, such as esters and terpenes. A histogram of the total OAVs of terpenes and esters detected in the wines produced using the 12 different yeast strains is shown in Fig. 1. The order of total OAVs abundances was M1 > Enoferm CSM > ICV Opale > Levure Seche Active > Uvaferm VRB. PCA of active aromatic compounds PCA was applied to the OAVs of esters and terpenes from Table 1. The first principal component (PC1) accounted for 39.2% of the total variance, and the second principal component (PC2) accounted for an additional 24.0% of the total variance (Fig. 2). Black raspberry wines produced by M1, BM 4X4, QA23 and ICV D47, positioned in the upper right quadrant, were more related to the presence of ethyl acetate, 3-methylbutyl acetate, ethyl benzoate, ethyl hexadecanoate, α-terpineol and β-linalool. These compounds have aroma descriptors such as fruity, apple, floral, pine and lavender (7,17,18). In the upper left quadrant, the black raspberry wines produced with strains Windsor, ICV Opale and Levure Seche Active were mainly related to the presence of ethyl hexanoate, α-pinene, β-pinene, limonene and ο-cymene, which have agreeable aroma descriptors, such as apple, fresh, lemon and citrus. The black raspberry wines produced with Uvaferm VRB, ICV GRE and QA23 (lower-left quadrant) were characterized by ethyl decanoate and terpene-4-ol, which may contribute to the fruity, pine tree-like aroma of black raspberry wine. Finally, the black raspberry wines fermented with 71B and Enoferm CSM were characterized by the presence of ethyl cinnamate and β-citronellol, which are associated with the aroma descriptors flower and green lemon. 89 J. Inst. Brew. 2015; 121: 87 94 Copyright 2015 The wileyonlinelibrary.com/journal/jib

B. H. Kim and S. Park Table 1. Odour activity values of volatile compounds Peak no. RI cal Compounds Odour descriptors Odour threshold (mg L 1 ) 71B V1116 M1 Windsor Enoferm CSM Uvaferm VRB OAVs ( 10 6 ) ICV GRE BM 4X4 QA23 ICV D47 ICV Opale Levure Seche Active 1 0.005 d,e 604 Ethyl acetate Solvent, fruity a 7.5 a 0.7 ± 0.09 1.2 ± 0.05 2.9 ± 0.22 0.6 ± 0.03 2.0 ± 0.19 1.5 ± 0.07 0.8 ± 0.01 1.5 ± 0.07 1.2 ± 0.06 1.4 ± 0.08 0.9 ± 0.02 1.2 ± 0.02 2 896 1-Propanol Alcohol b 750 b < 0.1 <0.1 <0.1 <0.1 3 1015 2-Methyl-1-propanol Alcohol, 65 b 0.1 ± 0.01 0.2 ± 0.01 0.1 ± 0.01 0.1 ± 0.01 < 0.1 0.1 ± 0.01 0.1 ± 0.01 0.2 ± 0.01 < 0.1 < 0.1 < 0.1 < 0.1 banana, solvent b 4 1045 3-Methylbutyl acetate Banana, apple, estery b 0.03 b 47.7 ± 0.67 2.6 ± 0.03 36.9 ± 3.00 34.8 ± 1.33 59.1 ± 4.70 43.5 ± 2.67 12.3 ± 0.67 25.3 ± 1.00 23.0 ± 1.00 14.7 ± 1.00 7.2 ± 0.30 17.5 ± 1.00 5 1099 β-pinene Fresh c 0.033 c 6.6 ± 0.01 8.0 ± 0.30 6 1114 1-Butanol Alcohol, fusel b 150 b < 0.1 1.9 ± 0.02 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 7 1174 Limonene Fruity, lemon d 0.015 d 25.8 ± 44.70 96.8 ± 6.00 18.5 ± 1.33 86.2 ± 4.70 109.4 ± 8.00 94.1 ± 4.70 8 1192 Dimethyl octatetraene Unknown 9 1223 α-pinene Fresh c 0.006 c 24.2 ± 41.70 48.3 ± 1.67 10 1241 3-Methyl-1-butanol Malty, fusel a 65 a 5.4 ± 0.05 < 0.1 4.6 ± 0.13 5.0 ± 0.01 5.7 ± 0.25 5.1 ± 0.06 5.0 ± 0.04 4.7 ± 0.02 4.8 ± 0.03 4.3 ± 0.03 4.8 ± 0.05 4.8 ± 0.04 11 1277 Ethyl hexanoate Apple, fruity, 0.014 b 241.1 ± 3.60 280.3 ± 17.00 361.2 ± 22.00 161.5 ± 30.00 321.6 ± 15.00 243.5 ± 16.00 125.1 ± 7.10 84.4 ± 5.70 61.7 ± 2.10 47.5 ± 0.70 66.7 ± 3.60 75.7 ± 1.42 sweetish b 12 1282 γ-terpinene Sweet, pine-like e 0.2 e 5.1 ± 0.50 5.6 ± 0.10 2.1 ± 0.10 13 1316 Terpinolene Plastic, petroleum Unknown 14 1331 ο-cymene Weak citrus c 0.011 c 304.2 ± 17.00 80.6 ± 0.10 292.9 ± 6.40 203.6 ± 4.60 194.2 ± 1.80 381.1 ± 15.00 393 ± 7.30 338.4 ± 20.00 327.4 ± 17.00 301.7 ± 11.00 296.3 ± 21.00 149.8 ± 9.10 15 1353 α-terpinene Herbaceous e 1.4 e 0.4 ± 0.04 0.4 ± 0.03 0.4 ± 0.03 0.4 ± 0.03 0.6 ± 0.01 0.5 ± 0.01 0.5 ± 0.02 2.6 ± 0.20 1.7 ± 0.08 1.8 ± 0.02 0.6 ± 0.01 16 1382 Tridecane Unknown 17 1398 Pinanol Unknown 18 1474 1-Hexanol Vegetable, grass b 8 b 0.1 ± 0.03 10.3 ± 0.84 0.1 ± 0.01 0.1 ± 0.01 < 0.1 0.01 < 0.1 < 0.1 0.01 < 0.1 < 0.1 < 0.1 19 1502 Tetradecane Unknown 20 1513 3-Ethoxy-1-propanol Unknown 21 1579 Dimethyl styrene Spicy-type flavour Unknown 22 1599 Acetic acid Vinegar f 600 f < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 23 1609 1-Heptanol 0.425 c 3.8 ± 0.10 4.4 ± 0.40 2.5 ± 0.21 4.1 ± 0.14 0.9 ± 0.05 1.0 ± 0.07 1.2 ± 0.09 1.0 ± 0.05 1.3 ± 0.05 1.1 ± 0.02 1.8 ± 0.09 1.0 ± 0.05 24 1650 Octanol Coconut, walnut c,h 0.2 c,h 7.2 ± 0.35 1.2 ± 0.05 5.3 ± 0.10 8.1 ± 0.30 4.5 ± 0.40 5.7 ± 0.25 8.7 ± 0.35 6.7 ± 0.30 7.5 ± 0.20 6.3 ± 0.10 7.7 ± 0.50 7.2 ± 0.10 25 1662 Ethyl octenoate Unknown 26 1671 Decenal Orange, green Unknown 27 1678 Benzaldehyde Fragrant, sweet, 0.35 c 2.4 ± 0.14 2.5 ± 0.03 aromatic c 28 1699 Butyl benzoate Unknown 29 1714 β-linalool Flower, lavender b 0.025 b 901.8 ± 15.00 1550 ± 12.00 832.5 ± 13.00 808.6 ± 38.00 856.9 ± 40.00 805.3 ± 22.00 771.8 ± 14.00 670 ± 46.00 680.9 ± 12.00 676.7 ± 3.20 891.8 ± 12.00 821 ± 22.00 30 1772 Terpinene-4-ol Pine tree-like, 3702 ± 76.00 909.9 ± 24.00 4064 ± 36.00 3501 ± 84.00 3771 ± 28.00 3865 ± 38.00 3672 ± 82.00 3486 ± 30.00 3475 ± 46.00 3553 ± 54.00 3300 ± 86.00 3907 ± 80.00 flowers d,e 31 1776 Verbenol Unknown 32 1801 Ethyl decanoate Fruity, apple b 0.2 b 96.9 ± 1.15 58.1 ± 3.75 33 1833 Ethyl benzoate Floral, fruity c 0.053 c 999.1 ± 4.60 142 ± 2.10 872.7 ± 69.00 741.4 ± 31.00 746.3 ± 20.00 691.7 ± 34.00 418.1 ± 30.00 377 ± 22.00 464.6 ± 12.00 390.4 ± 4.70 642.5 ± 32.00 634.9 ± 24.00 34 1843 Diethyl succinate Faint pleasant a 200 a 0.1 ± 0.001 0.4 ± 0.01 0.1 ± 0.01 0.1 ± 0.01 0.1 ± 0.01 0.1 ± 0.01 0.1 ± 0.01 < 0.1 0.1 ± 0.01 < 0.1 0.1 ± 0.01 0.1 ± 0.01 35 1865 α-terpineol Pine, lily of the 0.25 b,d 283.8 ± 2.80 48.9 ± 1.68 289.7 ± 3.10 306.4 ± 9.50 290.1 ± 1.00 294.8 ± 6.30 294.9 ± 8.90 280.8 ± 4.80 293.1 ± 4.50 288.9 ± 3.20 309.1 ± 6.50 297.1 ± 11.00 valley b,d 36 1871 Verbenone Spicy, camphoraceous Unknown 37 1926 β-citronellol Green lemon b 0.1 b 114.8 ± 1.00 621.7 ± 5.10 133.3 ± 2.40 154.3 ± 4.10 64.7 ± 0.80 160.3 ± 1.40 83.9 ± 4.80 135.4 ± 9.00 136.3 ±.3.00 125.5 ± 1.50 178.3 ± 5.30 155.9 ± 11.00 38 1950 Myrtenol Woody, minty g Unknown 39 1955 Geraniol Floral, rose-like h 0.13 h 6.7 ± 0.09 7.1 ± 0.62 6.9 ± 0.69 5.8 ± 0.30 6.6 ± 0.08 7.2 ± 0.08 5.5 ± 0.08 4.2 ± 0.23 5.1 ± 0.38 3.2 ± 0.23 6.7 ± 0.15 5.0 ± 0.15 40 1966 Phenylethyl acetate Apple, rose, 0.2 b 4.9 ± 0.68 75.5 ± 3.65 2.2 ± 0.20 1.3 ± 0.05 5.1 ± 0.60 3.0 ± 0.30 2.0 ± 0.05 2.6 ± 0.20 3.0 ± 0.25 2.6 ± 0.05 2.8 ± 0.25 2.0 ± 0.10 sweet, flowery b 41 1983 Carenol Unknown 42 1989 Ethyl dodecanoate Fruity, floral f 0.5 f 2.9 ± 0.26 32.2 ± 0.22 6.3 ± 0.42 4.3 ± 0.42 4.6 ± 0.42 8.7 ± 0.46 1.9 ± 0.08 2.8 ± 0.04 1.8 ± 0.10 4.5 ± 0.06 20.6 ± 1.22 20.7 ± 0.50 43 1993 ρ-cymen-8-ol Unknown 44 2008 Carveol Unknown 90 wileyonlinelibrary.com/journal/jib Copyright 2015 The J. Inst. Brew. 2015; 121: 87 94

91 Volatiles in Korean black raspberry wine 45 2015 Benzylalcohol Faint aromatic i 200 i < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 46 2035 Ethyloctyl succinate Unknown 47 2043 2-Phenylethanol Rose, sweetish i 14 i 16.3 ± 0.32 0.3 ± 0.02 14.2 ± 0.29 11.3 ± 0.23 11.8 ± 0.08 12.3 ± 0.37 10.2 ± 0.10 7.7 ± 0.15 12.3 ± 0.34 9.7 ± 0.28 11.3 ± 0.30 10.2 ± 0.30 48 2059 Dodecanol 1 i 1.1 ± 0.02 1.1 ± 0.05 1.3 ± 0.10 1.2 ± 0.07 1.2 ± 0.11 1.1 ± 0.07 0.9 ± 0.02 0.8 ± 0.05 0.8 ± 0.01 0.8 ± 0.04 1.3 ± 0.08 1.2 ± 0.04 49 2103 4-Phenyl-2-butanol Unknown 50 2113 Perillol Sweet, herbaceous e Unknown 51 2132 Isopropyl myristate Unknown 52 2135 Nerolidol Rose, apple, 0.25 c 1.7 ± 0.04 1.1 ± 0.04 2.2 ± 0.12 2.6 ± 0.16 2.6 ± 0.16 1.9 ± 0.04 1.7 ± 0.04 1.1 ± 0.04 1.8 ± 0.12 1.5 ± 0.16 1.2 ± 0.04 2.6 ± 0.12 green, citrus c 53 2141 Ethyl tetradecanoate Mild waxy c 4 c 1.5 ± 0.03 0.1 ± 0.01 3.4 ± 0.01 3.9 ± 0.19 2.0 ± 0.07 2.6 ± 0.17 1.6 ± 0.11 2.4 ± 0.15 2.0 ± 0.10 2.6 ± 0.09 5.4 ± 0.27 7.3 ± 0.32 54 2150 Octanoic acid Rancid, harsh, fatty 0.5 i 25.8 ± 1.1 14.1 ± 1.02 34.4 ± 1.1 48.9 ± 3.5 32.4 ± 1.34 28.6 ± 0.46 30.2 ± 1.64 29.7 ± 0.44 33.2 ± 1.20 35.8 ± 0.68 34.6 ± 1.38 33.7 ± 2.90 acids vegetable oil i 55 2177 ρ-cymen-7-ol Unknown 56 2196 Ethyl cinnamate White flowers i 0.0011 i 1009 ± 9.10 845.5 ± 73.00 1636 ± 82.00 1455 ± 9.10 2273 ± 73.00 1590 ± 73.00 1591 ± 64.00 1800 ± 18.00 1446 ± 82.00 1546 ± 91.00 1636 ± 38.00 1909 ± 18.00 57 2208 Ethyl pentadecenoate Unknown 58 2222 Oleic acid Faint fatty odour Unknown 59 2272 Ethyl palmitate Mild waxy c 2 c 36.5 ± 1.87 33.5 ± 2.91 54 ± 4.45 66.7 ± 1.56 35.4 ± 1.1 29.6 ± 0.66 31.3 ± 0.29 33.9 ± 0.74 35.4 ± 0.08 45.3 ± 1.86 51.9 ± 2.14 62.9 ± 2.10 60 2283 Ethyl-2-aminobenzoate Unknown 61 2287 Ethyl-9-hexadecenoate Unknown 62 2300 Ethyl-11-hexadecenoate Unknown 63 2385 Ethyl stearate Waxy c 3 c 1.7 ± 0.11 0.8 ± 0.05 1.2 ± 0.12 2.8 ± 0.23 1.0 ± 0.07 1.1 ± 0.01 0.9 ± 0.06 0.6 ± 0.03 1.2 ± 0.08 1.0 ± 0.04 1.0 ± 0.02 2.4 ± 0.08 64 2396 Ethyl oleate Unknown 65 2421 Ethyl linoleic acid Unknown 66 2455 Ethyl linolenic acid Unknown 67 2512 Isobutylidene phthalide Unknown OAVs ( 10 9 ) 7.72 ± 0.13 4.65 ± 0.33 8.82 ± 0.33 7.68 ± 0.23 8.78 ± 0.20 8.19 ± 0.21 7.46 ± 0.22 7.30 ± 0.16 7.02 ± 0.18 7.15 ± 0.19 8.46 ± 0.22 8.23 ± 0.19 Bold indicates that correlation is significant at p < 0.05 level. RIcal, linear retention index on the DB-WAX capillary column; n. d., not detected. Data are presented as the mean ± standard deviation (n = 3). Odour activity values (OAVs) were calculated using the equation OAV = c/t, where c is the peak area ( 10 6 ) of each compound in the black raspberry wines and t is the odour threshold value (in mg L 1 ) of the compound in water/ethanol solution (16). a Duarte et al. (7). b Vilanova et al. (13). c Pino et al. (17). d Noguerol-Pato et al. (18). e Choi (22). f Saberi et al. (10). g Nykanen et al. (23). h Marais (24). i Liang et al. (20). j Vilanova et al. (13). J. Inst. Brew. 2015; 121: 87 94 Copyright 2015 The wileyonlinelibrary.com/journal/jib

B. H. Kim and S. Park Figure 1. Total odour activity values (OAVs) histogram of terpenes and esters detected in 12 different yeast strains. 1, 71B; 2, V1116; 3, M1; 4, Windsor; 5, Enoferm CSM; 6, Uvaferm VRB; 7, ICV GRE; 8, BM 4X4; 9, QA23; 10, ICV D47; 11, ICV Opale; 12, Levure Seche Active. Figure 2. Principal component analysis (PCA) of active aroma compounds obtained by HS-SPME-GC-MS. 1, 71B; 2, V1116; 3, M1; 4, Windsor; 5, Enoferm CSM; 6, Uvaferm VRB; 7, ICV GRE; 8, BM 4X4; 9, QA23; 10, ICV D47; 11, ICV Opale 12, Levure Seche Active. 92 Sensory evaluation for preference Sensory evaluation of the 12 KBR wines was performed by 25 trained panellists on the basis of a quantitative descriptive analysis. ANOVA was used to differentiate the black raspberry wines by evaluation scores. The wine fermented by M1, as shown in Fig. 3, reached the highest value of fruity and floral aroma. Wines made with Uvaferm VRB, Levure Seche Active, Enoferm CSM and ICV Opale obtained high assessment scores because of the high OAVs of terpenes and esters. In contrast, wines made with V1116 and BM 4X4 obtained the lowest scores for overall acceptability. The order of total acceptability in terms of aromatic preference was M1 (4.9) > Ivaferm VRB (4.8) > Enoferm CSM (4.4)> Levure Seche Active (4.3) > ICV Opale (4.0) (Fig. 3). wileyonlinelibrary.com/journal/jib Copyright 2015 The J. Inst. Brew. 2015; 121: 87 94

Volatiles in Korean black raspberry wine Acknowledgements Lallemand Inc. is thanked for providing the 11 yeast strains and for technical advice. Soo-Yong Lee, the director of HiteJinro R&D, who allowed this study and made publication possible is thanked. There are no conflicts of interest to declare. Figure 3. Quantitative descriptive of analysis wines fermented by different yeast strains. Conclusions This is the first study to compare the association between sensory characteristics and volatile compound composition of wines produced from KBR using 12 different yeast strains. Therefore, this study provides important insights into the compounds contributing to wine aromas and presents a standard method for future comprehensive analyses. The KBR is a great resource for wine production because it contains an abundance of functional ingredients. In order to produce new black raspberry wines with good acceptability, selection of the optimal yeast strain is essential because particular yeast strains may contribute to the characteristic flavour of a wine and can also produce different amounts of aromatic compounds during fermentation. Studies on strain-specific aromatic profiles have become effective tools for the selection of a starter strain that will optimize wine fermentation. In this study, instrument-based analysis and sensory evaluation of KBR wines fermented by different yeast strains was performed. Both methods yielded similar results in terms of acceptability and preference. In general, instrument-based analysis revealed that the compounds that contributed the most to the wine flavour were associated with fruity (esters) and floral (terpenes) descriptors. The main active aroma compounds present, such as terpenes and esters, were significantly influenced by the yeast strains used. The selected yeast strains Enoferm CSM, Uvaferm VRB, ICV GRE, ICV Opale, Levure Seche Active and especially M1 produced excellent black raspberry wine and could be applied for further large-scale production of black raspberry wine. It is hoped that this work will expedite research on the production of high-quality black raspberry wines with beneficial physicochemical properties, functionality and good sensory characteristics. References 1. Ku, C.S. and Mun, S.P. (2008) Antioxidant activities of ethanol extracts from seeds in fresh Bokbunja (Rubus coreaus Miq.) and wine processing waste, Bioresource Technol. 99, 4503 4509. 2. Luther, M., Parry, J., Moore, J., Meng, J., Zhang, Y., Cheng, Z., and Yu, L. (2007) Inhibitory effect of Chardonnay and black raspberry seed extracts on lipid oxidation in fish oil and their radical scavenging and antimicrobial properties, Food Chem. 104, 1065 1073. 3. Tulio, A.Z., Reese, R.N., Wyzgoski, F.J., Rinaldi, P.L., Fu, R., Scheerens, J. C., and Miller, A.R. (2008) Cyanidin 3-rutinoside and cyaniding 3- xylosylrutinoside as primary phenolic antioxidants in black raspberry, J. Agric. Food Chem. 56, 1880 1888. 4. Yang, H.M., Oh, S.M., Li, S.S., Shin, H.K., Oh, Y.S., and Kim, J.K. (2008) Antiinflammatory activities of Rubus coreanus depend on the degree of fruit ripening, Phytother. Res. 22, 102 107. 5. Shin, T.Y. (2009) Effect of ripe fruits of Rubus coreanus on anaphylactic allergic reaction and production of cytokines, Nat. Prod. Sci. 15, 139 143. 6. Prior, R.L., Wilkes, S., Rogers, T. Khanal, R.C., Wu, X., Hager, T.J., Hager, A., and Howard, L. (2010) Dietary black raspberry anthocyanins do not alter development of obesity in mice fed an obesogenic highfat diet, J. Agric. Food Chem. 58, 3977 3983. 7. Duarte, W.F., Dias, D.R., Oliveira, J.M. Vilanova, M., Teixeira, J.A., Almeida e Silvae, J.B., and Schwan, R.F. (2010) Raspberry (Rubus idaeus L.) wine: Yeast selection, sensory evaluation and instrumental analysis of volatile and other compounds, Food Res. Int. 43, 2303 2314. 8. Meilgaard, M.C. (1975) Flavor chemistry of beer, Tech. Q. Master Brew. Assoc. Am. 12, 151 168. 9. Patel, S., and Shibamoto, T. (2002) Effect of different strains of Saccharomyces cerevisiae on production of volatiles in Napa Gamay wine and Petir Sirah wine, J. Agric. Food Chem. 50, 5649 5653. 10. Saberi, S., Cliff, M.A., and van Vuuren, H.J.J. (2012) Impact of mixed S. cerevisiae strains on the production of volatiles and estimated sensory profiles of Chardonnay wines, Food Res. Int. 48, 725 735. 11. Suzzi, G., Arfelli, G., Schirone, M., Corsetti, A., Perpetuini, G., and Tofalo, R. (2012) Effect of grape indigenous Saccharomyces cerevisiae strains on Montepulciano d Abruzzo red wine quality, Food Res. Int. 46, 22 29. 12. Torrens, J., Urpi, P., Riu-Aumatell, M., Vichi, S., Lopez-Tamames, E., and Buxaderas, S. (2008) Different commercial yeast strains affecting the volatile and sensory profile of cava base wine, Int. J. Food Microbiol. 124, 48 57. 13. Vilanova, M., Campo, E., Escudero, A., Grana, M., Masa, A., and Cacho, J. (2012) Volatile composition and sensory properties of Vitis vinifera red cultivars from North West Spain: Correlation between sensory and instrumental analysis, Anal. Chim. Acta 720, 104 111. 14. Barros, E.P., Moreira, N., Pereira, G.C.E., Leite, S.G., Moraes Rezende, C., and Guedes de Pinho, P. (2012) Development and validation of automatic HS-SPME with a gas chromatography ion trap/mass spectrometry method for analysis of volatile in wines, Talanta 101, 177 186. 15. Mendes, B., Goncalves, J., and Cmara, J.S., (2012) Effectiveness of highthroughput miniaturized sorbent-and solid phase microextraction techniques combined with gas chromatography mass spectrometry analysis for a rapid screening of volatile and semi-volatile composition of wines A comparative study, Talanta 88, 79 94. 16. Hellin, P., Manso, A., Flores, P., and Fenoll, J. (2010) Evolution of aroma and phenolic compounds during ripening of superior seedless grapes, J. Agric. Food Chem. 58, 6334 6340. 17. Pino, J.A., and Queris, O. (2011) Analysis of volatile compounds of mango wine, Food Chem. 125, 1141 1146. 18. Noguerol-Pato, R., Gonzalez-Alvarez, M., Gonzalez-Barreiro, C., Cancho-Grande, B. and Simal-Gandra, J. (2012) Aroma profile of Garnacha Tintorera-based sweet wines by chromatographic and sensorial analyses, Food Chem. 134, 2313 2325. 93 J. Inst. Brew. 2015; 121: 87 94 Copyright 2015 The wileyonlinelibrary.com/journal/jib

19. Cho, Y.J., Chun, S.S., Kwon, H.J., Kim, J.H., Yoon, S.J., and Lee, K.H. (2005) Comparison of physiological activities between hot-water and ethanol extracts of Bokbunja (Rubus coreanum F.), J. Kor. Soc. Food. Sci. Nutr. 34, 790 796. 20. Liang, H.Y., Chen, J.Y., Reeves, M., and Han, B.Z. (2013) Aromatic and sensorial profile of young Cabernet Sauvignon wines fermented by different Chinese autochthonous Saccharomyces cerevisiae strains, Food Res. Int. 51, 855 865. 21. Heyman H., and Noble A. (1987) Descriptive analysis of commercial Cabernet sauvignon wines from California, Am. J. Enol. Vitic. 38, 41 44. 22. Choi, H.-S. (2002) Headspace analysis of Robus coreanus berry by solid-phase microextraction and its sniffing test by gas chromatography olfactometry, Food Sci Biotechnol. 4, 355 360. 23. Nykanen L., and Suomalainen H. (1983) Aroma of Beer, Wine and Distilled Alcoholic Beverages (Handbook of Aroma Research), Springer, New York. 24. Marais, J. (1983) Terpenes in the aroma of grapes and wines. A review, S. Afr. J. Enol. Vitic. 4, 49 58 B. H. Kim and S. Park 25. Shamaila, M., Daubeny, B.S., and Anderson, A. (1993) Sensory, chemical and gas chromatographic evaluation of five raspberry cultivars, Food Res. Int. 26, 443 449. 26. Hernandez-Orte, P., Cersosimo, M., Loscos, N., Cacho, J., Garcia-Moruno, E., and Ferreira, V. (2008) The development of varietal aroma from nonfloral grapes by yeasts of different genera. Food Chem. 37, 1064 1077. 27. Lilly, M., Lambrechts, M.G., and Pretorius, I.S. (2000) Effect of increases yeast alcohol acetyltransferase activity on flavor profile of wine and distillates, Am. Soc. Microbiol. 66, 744 753. 28. Liberatore, M.T., Pati, S., Del Nobile, M.A., and La Notte, E. (2010) Aroma quality improvement of Chardonnay white wine by fermentation and ageing in barrique on lees, Food Res. Int. 43, 996 1002. 29. Ribereau-Gayon, P., Glories, Y., Maujean, A., and Dubourdieu, D. (2006) The Chemistry of Wine and Stabilization and Treatments, pp.51 61, John Wiley, Chichester. 30. Swiegers, J.H., Bartowsky, E.J., Henschke, P.A., and Pretorius, I.S. (2005) Yeast and bacterial modulation of wine aroma and flavour, Aust. J. Grape. Wine Res. 11, 139 173. 94 wileyonlinelibrary.com/journal/jib Copyright 2015 The J. Inst. Brew. 2015; 121: 87 94