Article pubs.acs.org/JAFC
Piperitone Profiling in Fine Red Bordeaux Wines: Geographical Influences in the Bordeaux Region and Enantiomeric Distribution Magali Picard,*,†,‡ Sophie Tempere,†,‡ Gilles de Revel,†,‡ and Stéphanie Marchand†,‡ †
Université Bordeaux, ISVV, EA 4577 Œnologie, F-33140 Villenave d’Ornon, France INRA, ISVV, USC 1366 Œnologie, F-33140 Villenave d’Ornon, France
‡
S Supporting Information *
ABSTRACT: Piperitone was recently identified in red Bordeaux wines, and this study was designed to further explore its contribution to wine aroma. Firstly, a geographical origin effect was detected within the Bordeaux region (left versus right banks of the Gironde estuary), with significantly higher mint aroma intensities and piperitone levels in wines from the left bank (Medoc appellations). Statistical analysis of chemical and sensory data highlighted a correlation with the proportion of Cabernet Sauvignon in the wine blends. Accordingly, it was suggested that these sensory and chemical differences may be of varietal origin. Secondly, the contribution of piperitone to minty aromas in the aging bouquet was confirmed by combining both sensory and chemical data from a wide range of red Bordeaux wines. Subsequent investigation of the enantiomeric distribution of piperitone found a significantly higher proportion of the (+)-(6S) enantiomeric form in wines displaying a strong aging bouquet. KEYWORDS: piperitone, aging bouquet, red wines, Bordeaux areas, Cabernet Sauvignon
■
INTRODUCTION The pleasure of tasting a high-quality wine derives not only from hedonic feelings but also from the taster’s mental representations, built on knowledge, culture, and experience. Verdù Jover et al.1 defined the perception of quality in red wines as a multidimensional structure, where the expression of complex aging bouquet is one of the key intrinsic attributes. As the aim of enology is to enhance wine quality,2 identifying both viticultural and enological parameters that permit full characterization of the personality of a high-quality wine and give the ability to taste the “terroir” where the grapes originated is among the current challenges. It is well-known that the concept of terroir is an important factor in wine quality and style.3 Terroir is a rather complex notion, defined by a well-characterized agricultural system, including soil and climate conditions, as well as viticultural and enological practices, that results in typical wine profiles, with specific, distinctive characteristics.4−6 In a previous sensory study, the conceptualization of red Bordeaux wines by professionals highlighted that the terroir dimension was intimately associated with the definition of wine aging bouquet, as several terms related to vineyard characteristics were elicited: “soil”, “hot”, “dry climate”, “great Cabernet Sauvignon”.7 Nevertheless, characterizing the aging bouquet of red Bordeaux wines involves assessing not only a unique olfactory concept but also different aromatic wine profiles, each depending on several viticultural and enological parameters. Indeed, the Bordeaux region is naturally divided by the Gironde Estuary into left- and right-bank areas. Each of these two areas in the Bordeaux region is characterized by several terroirs, where the plurality of soil profiles (gravel, sand, clay, and limestone), climatic conditions, cultivars, and human practices produce specific wine signatures. Whereas red Bordeaux wines are mixtures of several varietal wines, it is well-known that left-bank wine blends are most © XXXX American Chemical Society
often dominated by Cabernet Sauvignon, whereas the right bank mainly features Merlot. It has been reported that the aromatic spectrum of Cabernet Sauvignon wines is highly dependent upon the wine origin and can exhibit aromas as diverse as “blackcurrant”, “licorice”, “green pepper”, “mint”, “cedar”, “spice”, and “smoke”.8,9 Merlot aromas are also linked to the wine origin but are known to more particularly express “black and red fruits” notes, which over aging evolve toward “dried prune”,“violet flower”, “vanilla”, “leather”, and “undergrowth” nuances. In addition to increasing complexity and modulating typicality, the plurality of grape varieties in wine blends confers specific aromatic and chemical fingerprints.10,11 Whereas several aromatic notes characterizing red wine aging bouquet typicality could have been associated with specific molecular markers,12 the chemical deciphering of the soughtafter freshness expressed by old red Bordeaux wines remained under investigation. In a recent study, a multiple-step approach, combining sensory evaluations of red Bordeaux wines and aromatic reconstitutions of fractions extracted from red Bordeaux wines, identified piperitone as a molecular marker for the specific minty aroma in wine aging bouquet.13 As piperitone is a monoterpene ketone, its varietal origin was hypothesized on the basis of previous findings concerning carvone, another monoterpene ketone,14−16 and monoterpenes, such as 1,4- and 1,8-cineole,8,17 linalool, and α-terpineol.18 It has also been established that terpene levels vary drastically between wines made from Muscat and non-Muscat varieties, so these compounds are considered relevant molecular markers for Muscat wine typicality.15,19,20 Furthermore, piperitone has one Received: June 23, 2016 Revised: September 19, 2016 Accepted: September 25, 2016
A
DOI: 10.1021/acs.jafc.6b02835 J. Agric. Food Chem. XXXX, XXX, XXX−XXX
Article
Journal of Agricultural and Food Chemistry
None of the wines included in the two tasting sessions presented premature aging character, and they all were tasted within 30 min after the bottles were opened. Wine-Professional Tasting Panels. First Tasting. As previously described,7 a panel of 13 wine professionals (panel 1: 3 women and 10 men, age range 31−70 years, mean age 42 years) was selected. Second Tasting. For this second session, the sensory experiments were divided into three sequences. The first tasting session, in 2015, was conducted by the same panel (panel 1), whereas the panel for the second and third tastings in 2016 had changed slightly (panel 2: 11 wine professionals, 4 women and 7 men, age range 26−70 years, mean age 40 years). All of the panelists consumed wine regularly, worked in the Bordeaux area, and had considerable experience in tasting Bordeaux wines. As both tasting sessions focused on their experience and their own mental representation of wine aging bouquet, the panelists were informed about the context of the study but not given prior common training. Sensory Analyses. Sensory Conditions. Sensory analyses were performed as described by Martin and de Revel.24 Samples containing about 20 mL of wine were evaluated in a dedicated room in individual booths, using covered, black, ISO glasses (NF EN ISO 8589:2007). The presentation order was randomized among the panelists in a Latin square arrangement. For all tasting sessions, wines were presented in a small range of vintages to avoid any marked aging effect (i.e., the same wine series encompassed only vintages belonging to a 5 year period). Experimental Design. (a) First Tasting. The methodology of this first session has already been presented in detail.7 Wine professionals were invited to generate a free vocabulary correlating with the perceived typicality of the aging bouquet. The wines were separated by vintage into five sets of six. Panelists initially performed an ortho- and retronasal typicality evaluation task,25 based on their own concept of aging bouquet. More precisely, they replied to the following question: “Imagine that you have to explain the aging bouquet of red Bordeaux wines to someone. For each wine tasted: do you think this wine is a good or poor example of aging bouquet?”. Aging bouquet typicality was measured by the panel on a 10 cm unstructured scale, ranging from “‘poor example”’ on the left to “‘good example”’ on the right. The distance between the left end and the score-mark made by each panelist was measured and converted into a figure between 0 and 10, corresponding to “the typicality score”. Then, using a maximum of five aromatic descriptors, each panelist described the aromas of wines that they had previously identified as the most typical of aging bouquet. (b) Second Tasting. The typicality task and sensory profiling were performed according to our previously validated methodology.7,12 For sensory profiling, the intensity of the minty aromatic descriptor was evaluated, using a 10 cm unstructured scale, ranging from ‘“odor not perceived”’ on the left to “‘very intense”’ on the right. The distance between the left end and the score-mark made by each panelist was measured and converted into a score between 0 and 10, indicating the sensory intensity of the mint aromatic attribute. Quantitation of Piperitone in Red Bordeaux Wines. Instrumentation. Piperitone was analyzed by solid phase microextraction coupled to gas chromatography−mass spectrometry (SPME-GC-MS), as previously reported.13 Chromatographic analyses were carried out using an HP7890N GC system coupled to an HP 5975c quadrupole mass spectrometer, equipped with a Gerstel MPS2 autosampler (Gerstel GmbH, Mulheim, Germany) and a BP21 capillary column (50 m × 0.32 mm; 0.25 μm film thickness, SGE, Courtaboeuf, France). The carrier gas was helium N55 with a flow rate of 1 mL/min. The temperature program was 40 °C for 1 min, increasing to 220 °C at 3 °C/min (final isotherm for 20 min). The MS source temperature was set at 220 °C. Samples were spiked with 20 μL of D-camphor as an internal standard (10 mg/L in alcoholic solution). Identification was carried out in electron ionization (70 eV) and SIM mode. Piperitone was detected by its three main spectrum ions at m/z 82, 95, 110, 124, 137, and 152 amu, and the internal standard Dcamphor was identified by ions at m/z 81, 95, and 108 amu. On the basis of the best measured signal-to-noise ratio, the ions chosen for
asymmetrical carbon atom, indicating the possible presence of two enantiomers. (−)-(6R)-Piperitone has mainly been identified in Sitka spruce and some specific Mentha species, whereas (+)-(6S)-piperitone is present in most mint species and eucalyptus leaves.21−23 However, neither the enantiomeric piperitone profile, related to aging bouquet typicality, nor the influence of the dominant grape variety in wine blends has previously been studied in the Bordeaux region. This study, using an extensive range of red wines from different vintages and various appellationsMedoc appellations for left bank and Pomerol and Saint Emilion appellations for right bankwas designed to (i) investigate whether the composition of wine blends (predominance of Cabernet Sauvignon or Merlot) affected piperitone levels and the perception of mint aroma and (ii) further explore the contribution of piperitone to the aromatic profile of the red Bordeaux wine aging bouquet and, more precisely, the impact of its enantiomeric distribution.
■
MATERIALS AND METHODS
Chemicals. Dichloromethane (99.9%) was supplied by VWR Chemicals (Fontenay-sous-Bois, France), and sodium sulfate (99%) was obtained from Merck (Fontenay-sous-Bois, France). Reference Compounds. Standard compounds were obtained from commercial sources as follows: piperitone (i.e., (6R,6S)-3methyl-6-(1-methylethyl)-2-cyclohexen-1-one; CAS Registry No. 8981-6, purity = 92%) was supplied by Chemos (Regenstauf, Germany), the enantiomer (−)-(6R)-piperitone (CAS Registry No. 4573-50-6, specific rotation angle α20 D = −51.06°, purity = 98%) by Santa Cruz Biotechnology (Santa Cruz, CA), and D-camphor (purity = 97%) by Merck. Wines. First Tasting Session. As presented in our previous paper,7 28 commercial red wines from various Bordeaux appellations (ListracMedoc, Margaux, Medoc, Pauillac, Pessac-Leognan, Pomerol, SaintEmilion, Saint-Estephe, and Saint-Julien; Table S1 in the Supporting Information) were subjected to ortho- and retronasal sensory characterization. All of these wines, tasted in 2013, were oak barrelaged and produced between 1994 and 2005. Second Tasting Session. A second set of 51 commercial red wines from 4 Bordeaux appellations (Margaux, Medoc, Pauillac, and Pomerol), 6 different vineyards, and various vintages from 1994 and 2005 was subjected to both sensory and chemical analyses. Two appellation series (one Margaux, MX-1, and one Pomerol, P-1) also included younger vintages (2007, 2008, 2011, 2012, and 2013) (Table 1). All of these wines were oak barrel-matured before bottling. Two wine series (P-1 and MX-1; Table 1) were both tasted and analyzed in 2015, whereas the last four series (P-2, MX-2, PC, and MC; Table 1) were studied in 2016.
Table 1. Characteristics of the 51 Red Bordeaux Wines Studied in the Second Tasting Session
Bordeaux area
appellation
code series
range of vintages
number of wines
percentage of Cabernet Sauvignon in the blendsa
right bank
Pomerol Pomerol
P-1 P-2
1996−2012 1994−2005
7 12
5−15 none
left bank
Margaux Margaux Medoc Pauillac
MX-1 MX-2 MC PC
1995−2013 1995−2005 1994−2005 1994−2005
7 5 12 8
48−85 32−55 40−58 60−85
a
Given by wineries. B
DOI: 10.1021/acs.jafc.6b02835 J. Agric. Food Chem. XXXX, XXX, XXX−XXX
Article
Journal of Agricultural and Food Chemistry Table 2. Validation Parameters of the GC-MS Quantitation Method of Piperitone in Red Wines repeatability (%, n = 10)
recovery (%)
linearity range (ng/L)
correlation coefficient
LOD (ng/L)
LOQ (ng/L)
500 ng/L
2000 ng/L
reproductibility (%, n = 10)
500 ng/L
2000 ng/L
50−10000
0.999
25
80
5.3
2.2
5.1
96
98
Figure 1. Factorial correspondence analysis (FCA) plot, based on the frequency citation method, presenting the projection of the first series of 28 red Bordeaux wines and the 7 main aromatic descriptors of the wine aging bouquet (total variability = 55.3%). LB and RB correspond to “left bank” and “right bank” wines, respectively. quantitation were m/z 82 and 108 amu for piperitone and the internal standard, respectively. Each wine was analyzed in duplicate. Method Validation. Linearity was validated between 50 and 10000 ng/L by a series of 10 additions of piperitone standard in Cabernet Sauvignon red wine. The repeatability of the method was assessed by analyzing 10 replicates of a commercial red Bordeaux wine supplemented with piperitone standard at 500 and 2000 ng/L. The reproducibility of the method was determined by analyzing 10 replicates of the same spiked wine at 500 ng/L over a period of 1 month. With regard to the recovery study, a red Bordeaux wine was spiked with piperitone at the same levels as used for repeatability. All validation parameters are presented in Table 2. Enantioselective Analysis of Piperitone. Wine Sample Preparation. Two hundred milliliter samples of each wine were extracted three times at room temperature (20 °C) using 20, 10, and 10 mL of dichloromethane, with magnetic stirring (700 rpm), for 5 min each. The organic phases were separated in a funnel, collected, dried over sodium sulfate, and concentrated under nitrogen flow (approximately 100 mL/min) to obtain approximately 0.1 mL of wine extract. GC-MS Analysis of Wine Extracts. The enantioselective GC-MS analysis of piperitone was carried out on an HP 6890N (Agilent Technologies, Palo Alto, CA, USA) GC system coupled to an HP 5973 quadrupole mass spectrometer, equipped with a Gerstel MPS2 autosampler. An enantioselective column (Chiraldex Gamma-TA column; 50 m × 0.25 mm; 0.12 μm film thickness, Astec, Whippany, NJ, USA), with 2,6-di-O-pentyl-3-trifluoroacetyl derivative of γ-cyclodextrin as enantioselective stationary phase was used. A 2 μL sample of organic extract
was injected in split mode (split ratio, 30:1) using a 2 mm nondeactivated direct linear transfer, and injector temperature was maintained at 250 °C. Oven temperature was raised from 40 °C (1 min) to 120 °C at a rate of 2 °C/min (held for 10 min) and then ramped at a rate of 10 °C/min to 180 °C. The carrier gas was helium N55 (Linde, Toulouse, France) with a constant flow of 1 mL/min. Piperitone enantiomers were detected in electron ionization (70 eV) and SIM mode, selecting the following ions at the corresponding m/z 82, 95, 110, 124, 137, and 152 amu. Linear retention indices (LRI) for both piperitone enantiomers were obtained by simultaneous injection of samples and a series of alkanes (C8−C20, Sigma-Aldrich, St Quentin Fallavier, France), according to the analytical procedure described by Kovats26 and Van der dool and Kratz.27 Concerning the current (6R)/(6S) ratio, it was calculated assuming that LOD and LOQ values for both enantiomers are equivalent and close to the (6R,6S) reference compound ones. Statistical Analysis. Data from the First Tasting Session. Considering only the seven most representative aromatic descriptors of wine aging bouquet,7 quantitative analysis was based on attribute citation frequency. More precisely, the percentage of the panel (PF) who quoted each aromatic attribute was calculated using the formula PF = 100 × NP/NT (NP = number of panelists who cited the word; NT = total number of panelists). The data obtained were compiled in a wine × aromatic descriptor matrix, and factorial correspondence analysis (FCA) was then applied to the citation frequencies, based on the correlation matrix. Data from the Second Tasting Session. Typicality scores for the 51 red wines were statistically analyzed using the following published procedure.7,12 After the consensus among panelists had been validated C
DOI: 10.1021/acs.jafc.6b02835 J. Agric. Food Chem. XXXX, XXX, XXX−XXX
Article
Journal of Agricultural and Food Chemistry
Figure 2. Distribution of (a) mean aroma intensity and (b) piperitone concentrations in the second series of 51 wines studied, according to their geographical origin, from either the left or right bank. Asterisks, ∗∗ and ∗∗∗, indicate significant p values of
