Institute of Biotechnology, Faculty of Chemical and Food Technology, Slovak University of Technology in Bratislava, Radlinského 9, 812 37 Bratislava, Slovakia.
Institute of Analytical Chemistry, Faculty of Chemical and Food Technology, Slovak University of Technology in Bratislava, Radlinského 9, 812 37 Bratislava, Slovakia.
J Chromatogr B Analyt Technol Biomed Life Sci. 2021 Mar 15;1167:122565. doi: 10.1016/j.jchromb.2021.122565. Epub 2021 Feb 2.
The present work compares VOC profiles of 37 Tokaj varietal wines produced from Furmint, Lipovina and Muškát žltý grape varieties determined by SPME-GC-MS. The identified VOCs underwent treatment using one-way ANOVA and PCA for determination of compounds capable to distinguish wines based on grapevine variety. The presence of high concentration of terpenoids was characteristic for Muškát žltý wines, while Furmint and Lipovina expressed quite high similarity, nevertheless Lipovina was characterized by higher relative concentration of 1,1,6-trimethyl-1,2-dihydronaphtalene. Enantioselective analysis of the dominant terpenoids (limonene, linalool, hotrienol and α-terpineol) was performed by heartcut-2D-GC: R form was dominant for limonene and linalool and S enantiomer for hotrienol and α-terpineol. Statistical analysis (ANOVA) confirmed R-linalool as the most significant compound responsible for the main differences between studied varietal wines. LDA identified both total linalool and R-linalool as the only significant (p ≤ 0.05) sorting keys to distinguish investigated varietal wines. Both models allow perfect distinguishing of Muškát žltý from Lipovina and Furmint (p ≤ 0.05) with posterior probability of re-identification 100% but low distinction of Lipovina and Furmint from each other. The probability of posterior re-identification of Lipovina raised from 76.9 to 84.6% and total probability from 86.5 to 89.2% when using dominant enantiomer concentration as sorting variable.
本研究通过 SPME-GC-MS 比较了 37 种由福明特（Furmint）、利波维纳（Lipovina）和麝香黄葡萄（Muškát žltý）品种酿造的托卡伊（Tokaj）葡萄酒的 VOC 图谱。通过单因素方差分析（One-way ANOVA）和主成分分析（PCA）对所鉴定的 VOC 进行处理，以确定能够根据葡萄品种区分葡萄酒的化合物。萜烯类化合物的高浓度是麝香黄葡萄酒的特征，而福明特和利波维纳表现出相当高的相似性，但利波维纳的 1,1,6-三甲基-1,2-二氢萘的相对浓度较高。通过二维 GC 的中心切割（heartcut-2D-GC）对主要萜烯（柠檬烯、芳樟醇、霍香醇和α-松油醇）进行了对映体选择性分析，结果表明柠檬烯和芳樟醇的 R 形式占主导地位，而霍香醇和α-松油醇的 S 对映体占主导地位。统计分析（ANOVA）证实 R-芳樟醇是导致研究品种葡萄酒主要差异的最显著化合物。LDA 确定总芳樟醇和 R-芳樟醇是唯一显著的（p ≤ 0.05）分类关键指标，可用于区分所研究的品种葡萄酒。两种模型都可以完美地区分麝香黄葡萄酒与利波维纳和福明特（p ≤ 0.05），后验再识别的概率为 100%，但利波维纳和福明特之间的区分度较低。当使用主要对映体浓度作为分类变量时，利波维纳的后验再识别概率从 76.9%提高到 84.6%，总概率从 86.5%提高到 89.2%。
