molecules

Article Effects of Concentration on Oral Aroma Release After Wine Consumption

Carolina Muñoz-González, María Pérez-Jiménez, Celia Criado and María Ángeles Pozo-Bayón *

Instituto de Investigación en Ciencias de la Alimentación (CIAL), Consejo Superior de Investigaciones Científicas (CSIC)-Universidad Autónoma de Madrid (UAM), Campus de Excelencia Científica, 28049 Madrid, Spain * Correspondence: [email protected]; Tel.: +34-91-0017-961; Fax: +34-91-0017-905

 Received: 25 July 2019; Accepted: 3 September 2019; Published: 6 September 2019 

Abstract: This paper evaluates, for the first time, the effects of ethanol concentration on the dynamics of oral (immediate and prolonged) aroma release after wine consumption. To do this, the intraoral aroma release of 10 panelists was monitored at two sampling points (0 and 4 min) after they rinsed their mouths with three rosé wines with different ethanol content (0.5% v/v, 5% v/v and 10% v/v) that were aromatized with six fruity (ethyl butanoate, , , , and ). Overall, the results indicated that the extent of the effects of ethanol content on the oral aroma release were influenced by the subject, the ethanolconcentration and the type of aroma compound. This effect was also different in the immediate than in the prolonged aroma release. In the first in-mouth aroma monitoring, an increase in the ethanol content provoked a higher release of the more polar and volatile esters (ethyl butanoate, ethyl pentanoate), but a lower release for the more apolar and less volatile esters (ethyl octanoate, ethyl decanoate). Regarding the prolonged oral aroma release, an increase of ethanol content in wine increased the oral aroma release of the six esters, which might also increase the fruity aroma persistence in the wines. Future works with a higher number of individuals will be needed to understand the mechanisms behind this phenomenon.

Keywords: wine; ethanol; intra-oral SPME; oral aroma release; aroma persistence

1. Introduction Ethanol, the most abundant volatile compound in wines and the second most abundant after water, is a determinant of its sensory quality [1]. The presence of ethanol has been related to different sensations, such as burning feeling or palate warming [2], physical and perceived viscosity [2] or sourness and sweetness balance [3], among others. Apart from its contribution per se, it has been suggested that ethanol concentration plays a significant role in the overall aroma perception in wines [4–7], although the mechanisms behind these effects are not completely known. However, the increasing interest of consumers for light, fruity, and low alcohol beverages has increased the necessity to deeper understand this phenomenon in order to promote new types of low alcoholic wines. The studies about the effects of ethanol concentration on aroma compounds have been addressed through sensory and/or chemical approaches. From a sensory point of view, studies performed orthonasally (when a wine is smelled) have found that an increase in ethanol content provokes a decrease of the fruity and floral notes [4–7], while an increase in wood, pepper, or chemical notes [7], which would produce an odor imbalance. In this regard, in vitro static studies have shown that the presence of ethanol might change wine polarity which would modify the distribution of volatile compounds between the gas and liquid phases according to their physico-chemical properties (for a

Molecules 2019, 24, 3253; doi:10.3390/molecules24183253 www.mdpi.com/journal/molecules Molecules 2019, 24, 3253 2 of 10 Molecules 2019, 24, x FOR PEER REVIEW 2 of 12

44 reviewchemical on properties this topic see(for [ 1a]). review Although on this very topic relevant, see [1]). these Although studies very may relevant, not give these a clear studies understanding may not 45 ofgive how a clear the wineunderstanding is actually of perceived how the when wine itis isactually consumed. perceived when it is consumed. 46 When a wine is consumed, a fraction of aroma co compoundsmpounds travel from the mouth to the olfactory 47 receptors via retronasal afterafter havinghaving undergoneundergone oraloral processingprocessing (swallowing,(swallowing, breathing,breathing, interactioninteraction 48 with saliva, adsorption withwith mucus, etc.)etc.) [[8–11].8–11]. This generates a pulse of aroma known as immediate 49 aroma perception. Moreover, another aroma fraction remains in the oral cavity when the wine is no 50 longer in the mouth, and being released over time, which is responsible for the prolongedprolonged aroma 51 perception or aromaaroma persistence.persistence. Therefore, Therefore, th thee mechanisms mechanisms involved involved in in retronasal release and 52 perception are dynamic and more complex than thosethose behind orthonasal perception. In In this this regard, 53 some retronasalretronasal sensory sensory dynamic dynamic studies studies have have found found that ethanolthat ethanol increased increased the duration the duration and intensity and 54 ofintensity floralnotes of floral [12 notes]. In another [12]. In another study, the study, effect the of effect ethanol of ethanol on the on olfactory the olfactory intensity intensity of specific of specific wine 55 volatilewine volatile phenols phenols has also has been also observed been observed [13]. However, [13]. However, to our knowledge,to our knowledge, there are there no in are vivo no studiesin vivo 56 thatstudies have that measured have measured the effects the of effects ethanol of content ethanol on content the partitioning on the partitioning and release and of aroma release compounds of aroma 57 incompounds the mouth in considering the mouth considering the dynamics the of dynamics wine consumption. of wine consumption. This could This be due, could among be due, others, among to 58 analyticalothers, to analytical problems relatedproblems to therelated high to abundance the high abundance of ethanol of compared ethanol tocompared other volatiles to other of volatiles interest. 59 Toof interest. the authors’ To the knowledge, authors’ onlyknowledge, one in vivoonlystudy one in has vivo measured study has the measured effects of ethanolthe effects concentration of ethanol 60 onconcentration the retronasal on aromathe retronasal release inaroma model release wines in [14 model] showing wines that [14] the showing presence that of ethanol the presence increased of 61 theethanol overall increasedin vivo thearoma overall release. in vivo This aroma might release. be explained This might by changes be explained in surface by changes tension in increases surface 62 thetension surface increases area of the wine surface coating area the of mouthwine coating during the consumption, mouth during as wellconsumption, as by the Marangonias well as by eff theect 63 andMarangoni the Rayleigh-B effect andénard the convection Rayleigh-Bénard described convec by Tsachakition described and collaborators by Tsachaki [15 and,16] usingcollaboratorsin vitro 64 approaches.[15,16] using However, in vitro approaches. all these studies However, have beenall th performedese studies in have model been wines performed and have in notmodel taken wines into 65 accountand have the not complexity taken into of account the wine the matrix complexity composition, of the whichwine matrix could have composition, affected the which observed could e ffhaveects. 66 Inaffected this regard, the observed Tsachaki effects. and collaboratorsIn this regard, [16 Tsacha] showedki and that collaborators the in vitro [16]aroma showed release that in the real in wines vitro 67 wasaroma closer release to thatin real of anwines aqueous was closer solution to that than of to an model aqueous wines. solution Therefore, than to to model understand wines. Therefore, the effects 68 ofto ethanolunderstand content the oneffects aroma of releaseethanol and content perception, on aroma it is importantrelease and to perception, consider the it realis important wine matrix to 69 compositionconsider the real and wine to use matrix closer composition approaches toand that to ofuse wine closer consumption. approaches to that of wine consumption. 70 With thesethese antecedents antecedents in mind,in mind, the objectivethe objectiv of thise workof this was work to study was how to ethanolstudy concentrationhow ethanol 71 aconcentrationffects the dynamic affects (immediatethe dynamic and (immediate prolonged) and oral prolonged) aroma release oral aroma in real release wines. in Toreal this wines. end, To a 72 dealcoholizedthis end, a dealcoholized rosé wine (0.5% rosév /winev ethanol (0.5% content) v/v ethanol dosed content) with increasing dosed with amounts increasing of ethanol amounts (5% and of 73 10%)ethanol was (5% used. and 10%) The threewas used. wines, The aromatized three wines, with aromatized six fruity with esters six (4 fruity mg/L), esters were (4 evaluated mg/L), were by 74 10evaluated panelists by at 10 two panelists time points at two (30 time sec “immediatepoints (30 sec aroma “immediate release” aroma and 4 minrelease” after and wine 4 expectorationminutes after 75 “prolongedwine expectoration oral release”) “prolonged following oral the release”) intra-oral following SPME procedure the intra-oral for oral SPME monitoring. procedure These for results oral 76 willmonitoring. provide These more informationresults will provide about how more aroma information release changesabout how during aroma consumption, release changes and willduring be 77 importantconsumption, to understand and will be the important role of ethanol to understand on aroma the persistence. role of ethanol on aroma persistence.

78 2. Results and Discussion 79 To evaluate the eeffectffect of ethanol content on the in vivo aroma release during wine consumption, consumption, 80 the oraloral aromaaroma release release of of 10 10 panelists panelists was was monitored monitored at two at two sampling sampling points points (Figure (Figure1) after 1) they after rinsed they 81 theirrinsed mouths their mouths with three with ros threeé wines rosé aromatized wines aromatiz withed six with fruity six esters fruity (Table esters1) presenting(Table 1) presenting di fferent 82 ethanoldifferent content ethanol (0.5% contentv/v ,(0.5% 5% v/ v/andv, 5% 10% v/v vand/v). 10% v/v).

83 Figure 1. Sampling procedure followed for oral aroma monitoring. 84 Figure 1. Sampling procedure followed for oral aroma monitoring.

85

Molecules 2019, 24, 3253 3 of 10

Table 1. Physicochemical properties of the aroma compounds employed in this study.

Aroma CAS MW a Solubility d log P b BP c ( C) Aroma Descriptor e Compounds Number (g/mol) ◦ (mg/L) 105-54-4 116.16 1.85 125.79 4900 Pineapple Isoamyl acetate 123-92-2 130.19 2.26 134.87 2000 Banana Ethyl pentanoate 539-82-2 130.19 2.34 148.37 2210 Fruity Ethyl hexanoate 123-66-0 144.22 2.83 170.05 309 Apple Ethyl octanoate 106-32-1 172.27 3.81 210.70 70 Fruity apricot Ethyl decanoate 110-38-3 200.32 4.79 247.43 16 Grape a Molecular weight. b Hydrophobic constant estimated using molecular modeling software EPI Suite (U.S. EPA 2000 2007). c Boiling point estimated using molecular modeling software EPI Suite (U.S. EPA 2000 2007). d Solubility− in water estimated using molecular software EPI Suite (U.S. EPA 2000 2007). e From Flavornet database− (http://www.flavornet.org; accessed October 2009) or PubChem (https://pubchem.ncbi.nlm.nih.gov− /).

2.1. Effects of Ethanol Concentration and Individual Differences on the Immediate Oral Aroma Release Immediate aroma release refers to the aroma compounds trapped in a SPME fiber immediately after the panelists rinsed their mouths with the wine samples and expectorated (Figure1). Data obtained for the 10 volunteers in each of the wines were submitted to two-way ANOVA analyses considering individuals and ethanol content as factors and their interactions. Results (Table2) showed that the effect of subjects on aroma release was highly significant (p < 0.0001) for all the assayed compounds (Table2), which indicated a high interindividual variability on the immediate oral release of esters. The variability on the aroma released observed among individuals equally trained in the same consumption procedure and using the optimized SPME conditions is not surprising and it could be related to differences in oral physiological factors, such as, salivary flow rate and composition or oral cavity volume, among others, as it has been previously described in different works [8–11].

Table 2. Results of the two-way ANOVA performed to compare the statistical significance of the ethanol content on the immediate intra-oral aroma release (n = 10). Each sample was analyzed in triplicate.

Ethyl Isoamyl Ethyl Ethyl Ethyl Ethyl Butyrate Acetate Pentanoate Hexanoate Octanoate Decanoate Mean 4,596,854.41 7,107,442.85 6,254,852.57 8,892,546.68 26,049,220.47 12,679,314.42 SEM a 355,864.23 509,112.21 453,805.55 744,257.30 1,704,957.45 850,164.95 R2 b 0.94 0.94 0.91 0.87 0.95 0.93 F c 25.27 24.43 17.228 11.47 30.10 21.29 Pr > F d <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 F 56.14 59.49 40.55 31.37 78.78 58.32 Subject Pr > F <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 F 4.04 1.06 3.73 0.57 6.75 4.29 Ethanol (%) Pr > F 0.024 0.356 0.031 0.567 0.003 0.019 Subject*Ethanol F 12.30 9.57 7.17 3.31 8.54 4.65 (%) Pr > F <0.0001 <0.0001 <0.0001 0.000 <0.0001 <0.0001 a Standard error of mean. b Coefficient of determination. c F statistic. d Probability values.

Regarding the effect of ethanol content on the immediate oral aroma release, it seemed that this effect was less important than the subject effect and aroma compound and ethanol concentration-dependent (Table2). To better visualize these results, data were plotted, and shown in Figure2. To do so, data were normalized considering the values obtained for the 0.5% v/v as 0% and calculating the ratio for the other two wines. As can be seen, the oral release of two compounds (isoamyl acetate andethyl hexanoate) was not significantly affected by variations in the ethanol content. Moreover, certain compounds (ethyl butyrate and ethyl pentanoate) showed a significantly higher immediate oral release after the consumption of the wines with high ethanol content (5% v/v and 10% v/v) compared to 0.5% v/v, while others (ethyl octanoate and ethyl decanoate) showed the opposite behavior. The differences Molecules 2019, 24, x FOR PEER REVIEW 5 of 12

108 Regarding the effect of ethanol content on the immediate oral aroma release, it seemed that this 109 effect was less important than the subject effect and aroma compound and ethanol concentration- 110 dependent (Table 2). To better visualize these results, data were plotted, and shown in Figure 2. To 111 do so, data were normalized considering the values obtained for the 0.5% v/v as 0% and calculating 112 the ratio for the other two wines. As can be seen, the oral release of two compounds (isoamyl acetate 113 andethyl hexanoate) was not significantly affected by variations in the ethanol content. Moreover, 114 certain compounds (ethyl butyrate and ethyl pentanoate) showed a significantly higher immediate 115 oralMolecules release2019, 24after, 3253 the consumption of the wines with high ethanol content (5% v/v and 10%4 ofv/v 10) 116 compared to 0.5% v/v, while others (ethyl octanoate and ethyl decanoate) showed the opposite 117 behavior. The differences observed among compounds are not unexpected, and they could be related 118 toobserved the physicochemical among compounds properties are notof the unexpected, aroma compounds, and they such could as be hydrophobicity related to the physicochemical(log P) or boiling 119 pointproperties (BP), ofas theit has aroma been compounds, previously suchsuggested as hydrophobicity by different authors (log P) or[5,15,17–19]. boiling point Interestingly, (BP), as it hasthe 120 lessbeen polar previously and volatile suggested compounds by different of authorsthis study [5, 15(ethyl,17–19 octanoate]. Interestingly, and ethyl the lessdecanoate) polar and were volatile less 121 releasedcompounds in the of thiswines study with (ethyl high et octanoatehanol content. and ethyl This decanoate) could be due were to lessthe fact released that these in the compounds wines with 122 couldhigh ethanol be more content. soluble This in the could wine be duematrix to theas etha factnol that concentration these compounds increases. could beTherefore, more soluble the higher in the 123 winethe ethanol matrix content, as ethanol the concentration higher the retention increases. of Therefore,these compounds the higher in the the wine ethanol matrix content, and thethe higherlower 124 the retentionimmediate of oral these release. compounds On the in contrary, the wine the matrix more and polar the lowerand volatile the immediate compounds oral (ethyl release. butyrate On the 125 andcontrary, ethyl thepentanoate) more polar were and more volatile released compounds in the wines (ethyl with butyrate high ethanol and ethyl content. pentanoate) These compounds were more 126 wouldreleased have in the had wines less withaffinity high for ethanol the wine content. matrix These as ethanol compounds concentration would have increased, had less and affinity therefore, for the 127 winethey could matrix have as ethanol been more concentration intra-orally increased, released. and However, therefore, the they immediate could have release been of more the intra-orallycompound 128 isoamylreleased. acetate However, (showing the immediate a similar release log P ofand the BP compound values than isoamyl those acetate of ethyl (showing pentanoate) a similar was log not P 129 significantlyand BP values affected than those by the of ethylethanol pentanoate) content. Interestingly, was not significantly this is the aff onlyected compound by the ethanol with content. a non- 130 linearInterestingly, structure this of is thetheonly compounds compound assayed, with a non-linearwhich means structure that ofnot the only compounds the phsysicochemical assayed, which 131 characteristics,means that not but only the the structure phsysicochemical of the compound characteristics, could be but important the structure to explain of the compoundoral aroma couldrelease. be important to explain oral aroma release.

132 Figure 2. Percentage of immediate oral aroma release obtained for the volunteers (n = 10) after rinsing 133 Figuretheir mouths 2. Percentage with the of winesimmediate (considering oral aroma the release values obta obtainedined for for the the volunteers W 0.5% as (n 0% = 10) and after comparing rinsing 134 theirthis value mouths with with the the amount wines determined(considering for the the valu restes ofobtained wines). for Di thefferent W 0.5% letters as across0% and the comparing different 135 thiscompounds value with denote the statisticalamount determined differences afterfor the the rest application of wines). of theDifferent Tukey lette testrs for across means the comparison different 136 compounds(p < 0.05). denote statistical differences after the application of the Tukey test for means comparison 137 (p < 0.05). Finally, the interaction between subject and ethanol content was significant for all the aroma 138 compoundsFinally, assayedthe interaction (Table2 between). This means subject that and the ethanol e ffect content of ethanol was on significant the immediate for all oralthe aroma 139 compoundsrelease is individual assayed dependent,(Table 2). This and therefore,means that not the all effect the panelists of ethanol would on bethe a ffimmediateected in the oral same aroma way 140 releaseby changes is individual in ethanol dependent, concentration. and Figure therefore,3 shows not anall example the panelists of the would release be values affected obtained in the for same the 141 waycompound by changes ethyl decanoatein ethanol byconcentration. the 10 panelists. Figure As can2 shows be seen, an example the variation of the in release the ethanol values content obtained only 142 forsignificantly the compound affected ethyl the oraldecanoate aroma releaseby the of10 ethyl panelist decanoates. As can for threebe seen, of the the ten variation subjects in (S2, the S3 ethanol and S4). 143 contentMoreover, only the significantly effects of ethanol affected content the onoral oral aroma aroma release release of showed ethyl decanoate opposite behaviorsfor three dependingof the ten 144 subjectson the subject. (S2, S3 While and S4). the presenceMoreover, of ethanolthe effects significantly of ethanol reduced content the on release oral aroma of ethyl release decanoate showed for 145 oppositeS2 and S4, behaviors it increased depending the release on th fore subject. S3. The While different the presence effects of of ethanol ethanol concentration-depending significantly reduced the on the subject are interesting, and it should be studied in future works considering a high number of individuals. Molecules 2019, 24, x FOR PEER REVIEW 6 of 12

146 release of ethyl decanoate for S2 and S4, it increased the release for S3. The different effects of ethanol 147 concentration-depending on the subject are interesting, and it should be studied in future works

148 Moleculesconsidering2019, 24 ,a 3253 high number of individuals. 5 of 10

149 150 FigureFigure 3. 3.Immediate Immediate oral oral ethyl ethyl decanoate decanoate release release (absolute(absolute peakpeak areas)areas) obtained for the volunteers volunteers (n 151 (n== 10)10) after after rinsing their mouthsmouths withwith thethe wines.wines. DiDifferentfferent lettersletters across across the the di differentfferent compounds compounds 152 denotedenote statistical statistical diff erencesdifferences (p value (p

154 2.2.To Effects globally of Ethanol understand and Individual the effects Diffe of ethanolrences on during the Prolonge wine consumptiond Oral Aroma itRelease is important to consider not only the aroma released immediately after wine intake, but also the oral aroma release that lingers 155 To globally understand the effects of ethanol during wine consumption it is important to in the mouth once the wine is no longer in the oral cavity. Therefore, after monitoring the immediate 156 consider not only the aroma released immediately after wine intake, but also the oral aroma release aroma release, the panelists were asked to keep their mouths closed and 2 min later, a second oral 157 that lingers in the mouth once the wine is no longer in the oral cavity. Therefore, after monitoring the monitoring was carried out in the same conditions as described above (Figure1). With these data, a 158 immediate aroma release, the panelists were asked to keep their mouths closed and 2 minutes later, two-way ANOVA analysis was performed in order to understand the effect of individual differences 159 a second oral monitoring was carried out in the same conditions as described above (Figure 1). With and ethanol content on the prolonged oral aroma release. Results are shown in Table3. Similar to 160 these data, a two-way ANOVA analysis was performed in order to understand the effect of what happened in the immediate oral aroma release, a high variability among individuals and in the 161 individual differences and ethanol content on the prolonged oral aroma release. Results are shown interactions between individuals and ethanol content was observed for all the aroma compounds 162 in Table 3. Similar to what happened in the immediate oral aroma release, a high variability among assayed (Table3). 163 individuals and in the interactions between individuals and ethanol content was observed for all the 164 aromaTable compounds 3. Results of assayed the two-way (Table ANOVA 3). performed to compare the statistical significance of the ethanol content on the prolonged intra-oral aroma release (n = 10). Each sample was analyzed in triplicate.

Ethyl Isoamyl Ethyl Ethyl Ethyl Ethyl Butyrate Acetate Pentanoate Hexanoate Octanoate Decanoate Mean 133,838.96 321,509.22 199,024.90 633,104.58 5,761,616.25 6,940,717.29 SEM a 31,033.08 43,941.31 36,020.31 105,826.43 519,018.38 515,250.95 R2 b 0.83 0.96 0.86 0.90 0.91 0.91 F c 9.02 45.07 11.32 21.13 19.41 18.74 Pr > F d <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 F 19.80 121.52 31.67 63.78 55.75 48.91 Subject Pr > F <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 Ethanol F 8.91 12.64 7.44 2.56 3.76 7.55 (%) Pr > F 0.000 <0.0001 0.001 0.087 0.030 0.001 Subject*Ethanol F 3.62 10.38 1.66 1.98 2.82 4.54 (%) Pr > F 0.000 <0.0001 0.079 0.028 0.002 <0.0001 a Standard error of mean. b Coefficient of determination. c F statistic. d Probability values.

Regarding the effects of ethanol content on the prolonged oral aroma release, it can be observed that the presence of ethanol significantly increased the oral release for five of the six esters assayed. Moreover, the release of ethyl hexanoate was also increased by the ethanol content, but the effect did not reach the significance level (p-value = 0.087) (Table3). Once again, to better visualize these results, data

were plotted, and shown in Figure4. Here, data were normalized considering the values obtained for the 0.5% v/v as 0% and calculating the ratio for the other two wines. As can be observed, all the esters Molecules 2019, 24, x FOR PEER REVIEW 8 of 12

Regarding the effects of ethanol content on the prolonged oral aroma release, it can be observed that the presence of ethanol significantly increased the oral release for five of the six esters assayed. Moreover, the release of ethyl hexanoate was also increased by the ethanol content, but the effect did not reach the significance level (p-value = 0.087) (Table 3). Once again, to better visualize these results, data were plotted, and shown in Figure 4. Here, data were normalized considering the values Moleculesobtained2019 for, 24 ,the 3253 0.5% v/v as 0% and calculating the ratio for the other two wines. As can be observed,6 of 10 all the esters were more released at higher ethanol content. Moreover, the more apolar compounds were(ethyl more octanoate released and at higherethyl decanoate) ethanol content. were also Moreover, significantly the more more apolar released compounds in the 5% (ethyl v/v than octanoate in 0.5% andv/v ethyl. Interestingly, decanoate) these were results also significantly showed the more same released trend as in the the previous 5% v/v than studies in 0.5% andv /dynamicv. Interestingly, sensory thesestudies results performed showed on the this same topic trend which as the showed previous an studiesenhancement and dynamic of aroma sensory release studies in the performed presence of onethanol this topic [14,20], which and showed an increase an enhancement in the duration of aroma of release fruity/floral in the presencenotes as ofthe ethanol ethanol [14 concentration,20], and an increaseincreases in the[12]. duration of fruity/floral notes as the ethanol concentration increases [12].

FigureFigure 4. 4.Percentage Percentage of of prolonged prolonged oral oral aroma aroma release release obtained obtained for for the the volunteers volunteers (n (n= 10)= 10) after after rinsing rinsing theirtheir mouths mouths with with the the wines wines (considering (considering the the values values obtained obtained for for the the W0.5% W0.5% as as 0% 0% and and comparing comparing thisthis value value with with the the amount amount determined determined for for the the rest rest of of wines). wines). Di Differentfferent letters letters across across the the di differentfferent compoundscompounds denote denote statistical statistical di differencesfferences after after the the application application of of the the Tukey Tukey test test for for means means comparison comparison (p(p< 0.05).< 0.05). Although additional studies are required in order to elucidate the mechanisms of action of ethanol Although additional studies are required in order to elucidate the mechanisms of action of content on the persistence of aroma compounds after wine intake, the higher prolonged oral aroma ethanol content on the persistence of aroma compounds after wine intake, the higher prolonged oral release observed in the presence of ethanol (from 5% v/v onwards) could be due to different facts. aroma release observed in the presence of ethanol (from 5% v/v onwards) could be due to different On the one hand, ethanol might induce changes in surface tension affecting the distribution of the facts. On the one hand, ethanol might induce changes in surface tension affecting the distribution of wine in the mouth and pharynx during consumption, allowing the sample to better spread out and the wine in the mouth and pharynx during consumption, allowing the sample to better spread out favoring the formation of a larger surface in the oral cavity for volatile release [20]. On the other hand, and favoring the formation of a larger surface in the oral cavity for volatile release [20]. On the other in the presence of ethanol, esters could be more solubilized in the layer of wine that remains in the hand, in the presence of ethanol, esters could be more solubilized in the layer of wine that remains in mouth after wine rinsing and, therefore, these compounds could be released more slowly over time. the mouth after wine rinsing and, therefore, these compounds could be released more slowly over Finally, the mixture of wine lingering in the oral cavity could be submitted to the Marangoni and time. Finally, the mixture of wine lingering in the oral cavity could be submitted to the Marangoni Rayleigh convection phenomenon described in vitro by Tsachaki and co-authors [15]. In this case, the and Rayleigh convection phenomenon described in vitro by Tsachaki and co-authors [15]. In this case, presence of ethanol would help aroma compounds be released more easily from the layer of wine to the presence of ethanol would help aroma compounds be released more easily from the layer of wine the oral headspace. Additionally, other mechanisms related to the interactions of ethanol with salivary to the oral headspace. Additionally, other mechanisms related to the interactions of ethanol with proteins or enzymes could not be discarded. In this regard, it could be possible that the presence salivary proteins or enzymes could not be discarded. In this regard, it could be possible that the of ethanol could have inhibited certain salivary enzymes (e.g., esterases), more prone to metabolize presence of ethanol could have inhibited certain salivary enzymes (e.g., esterases), more prone to long chain esters [21], like ethyl octanoate and ethyl decanoate. Additionally, the presence of ethanol metabolize long chain esters [21], like ethyl octanoate and ethyl decanoate. Additionally, the presence (in higher concentrations than those of the aroma compounds) could have occupied the binding sites of ethanol (in higher concentrations than those of the aroma compounds) could have occupied the of the salivary proteins modifying the interactions between aroma compounds and oral proteins as binding sites of the salivary proteins modifying the interactions between aroma compounds and oral suggested previously in the presence of sugar [22]. Therefore, additional studies are required in order proteins as suggested previously in the presence of sugar [22]. Therefore, additional studies are to corroborate these hypotheses. required in order to corroborate these hypotheses. 3. Material and Methods 3. Material and Methods 3.1. Wine Samples 3.1. Wine Samples A dealcoholized rosé wine (Matarromera, Pesquera de Duero, Spain) from the Tempranillo grape variety with an ethanol concentration of 0.5% v/v (control wine) was selected for this study. To this wine, different amounts of ethanol content (Panreac Quimica S.A., Barcelona, Spain) were added to obtain two additional wines with 5% v/v and 10% v/v of ethanol content. To reinforce the aroma profile of the wines, six food-grade esters (Sigma-Aldrich, Steinheim, Germany) were employed in this study. They present different physicochemical properties and are associated with fruity notes in wines (Table1). Independent stock solutions of the compounds were Molecules 2019, 24, 3253 7 of 10 prepared in food-grade ethanol (Panreac Quimica S.A., Barcelona, Spain). From here, each aroma compound was added to the wines immediately before each session to obtain a final concentration of 4 mg/L. It was previously determined by HS-SPME-GC/MS [23] that the amount of these compounds present in the original wine was negligible compared to the amount of aroma added (Table S1).

3.2. Intra-Oral SPME Sampling

3.2.1. Subjects Ten young, healthy volunteers (four male and six female) between 18–36 years old participated in this study. The sampling procedures were explained in detail to the subjects who also provided written consent to participate. The Bioethical Committee of the Spanish National Council of Research (CSIC) approved this study.

3.2.2. Intra-Oral SPME Procedure for Aroma Monitoring To study how ethanol affects the oral processing of typical wine esters, subjects rinsed their mouths with three aromatized wines containing different ethanol content (0.5% v/v, 5% v/v and 10% v/v), and then, the aroma released into the oral cavity after wine expectoration was monitored following the intra-oral SPME procedure previously developed and validated by Esteban-Fernandez and collaborators [24]. The representation of the analytical procedure is schematized in Figure1. Briefly, subjects were instructed to introduce the aromatized wine (15 mL) into their oral cavity, kept it for 30 s and then spit it off. During rinsing, special care was taken to keep the lips closed, not to swallow and not to open the velum-tongue border prior to expectoration. At 30 s after expectoration, a DVB/CAR/PDMS (Divinylbenzene/Carboxen/Polydimethylsiloxane 50/30 µm thickness, 2 cm length) coated SPME fiber (Supelco, Bellefonte, PA, USA) with a home-made adaptor, was introduced into the oral cavity of the volunteer. After 2 min of oral aroma extraction, in which swallowing was not allowed, the fiber was removed from the oral cavity and immediately placed into the split/splitless injector (splitless mode) of the Gas Chromatograph (GC) Agilent 6890N coupled to a quadrupole Mass Detector (MS) Agilent 5973 (Santa Clara, CA, USA). This measurement is referred to as an immediate oral release. Four minutes after wine expectoration, a second fiber was introduced in the oral cavity following the exact protocol described previously (Figure1). This second measurement is referred to as a prolonged oral aroma release. After each injection, the fiber was cleaned for 10 min to avoid any memory effect. Analyses were performed three times with each wine type (0.5% v/v, 5% v/v and 10% v/v) by the ten volunteers (3 wine 10 volunteers 3 repetitions = 90 injections). An inter-fiber repeatability study was previously × × carried out. This allowed the selection of the most similar fibers (RSD values <10% for the extraction of the same aroma compound) to complete the study. Moreover, the use of a short sampling time also known as “true headspace” [25], avoids any problems derived from the use of SPME, such as the changes in the distribution of the molecules between the liquid and gas phases at equilibrium, due to adsorption phenomena onto the fiber itself, or to problems of saturation, due to an excess of ethanol absorbed onto the fiber compared to the compounds of interest [25]. In our case, the equilibrium is not reached, the fiber is not saturated, and therefore, the detected composition can be considered as representative of the real intraoral-headspace composition.

3.2.3. GC/MS Analyses Volatile compounds were separated on a DB-Wax polar capillary column (60 m 0.25 mm i.d. × × 0.50 um film thickness) from Agilent (J&W Scientific, Folsom, USA). Helium was the carrier gas at a 1 flow rate of 1 mL min− . The oven temperature was initially held at 40 ◦C for 2 min, then increased at 8 ◦C each minute to 240 ◦C and held for 15 min. For the MS system (Agilent 5973N), the temperature of the transfer line, quadrupole and ion source were 270, 150 and 230 ◦C respectively. Electron impact mass spectra were recorded at 70 eV Molecules 2019, 24, 3253 8 of 10 ionization voltages, and the ionization current was 10 µA. The acquisitions were performed in a scan (from 35 to 350 amu) and SIM modes. The identification of the six target compounds was based on the comparison of retentions times and mass spectra. The mass spectra were compared with those from NIST 2.0 database and with those from reference compounds analysed in the same conditions. Since no internal standard was used during the oral SPME extractions, absolute peak areas (APAs) were obtained. The use of APAs to express aroma release was sufficient for this type of analyses as the aim of the work was to compare the extent of aroma release between wine samples.

3.2.4. Statistical Analyses Two-way ANOVA and the Tukey test were used to determine significant differences in oral release parameters of the six aroma compounds considering subjects and wine type (ethanol concentration) as factors. The significance level was p < 0.05 throughout the study. The XL-Stat 2017.01 program was used for data processing (Addinsoft, Paris, France).

4. Conclusions The present results showed that the effects of ethanol content on oral aroma release after wine consumption are dependent on the subjects, the ethanol concentration and their interactions and the nature of the aroma compound assayed. The effects of ethanol content on oral aroma release were different immediately after wine consumption than four minutes later. In the first in-mouth aroma monitoring, an increase in the ethanol content provoked a higher release of the more polar and volatile esters (ethyl butanoate, ethyl pentanoate), but a lower release for the more apolar and less volatile esters (ethyl octanoate, ethyl decanoate). Regarding the prolonged oral aroma release, an increase of ethanol content in wine, increased the oral aroma release of the six esters. This effect was greater when the concentration of alcohol rose from 0.5% to 10% v/v, and it could be related to an increase of the fruity aroma persistence in the wines. Moreover, important interindividual differences have been noted in this work, which could be related to variation in oral physiology parameters among subjects. Interestingly, high interaction between subject and ethanol was also found, meaning that in each subject, retronasal aroma release is differently affected by ethanol content. Future studies with a high number of individuals are needed in order to validate these results and to investigate the mechanisms behind the observed effects. Nonetheless, it is important to state that this research has shown the effect of ethanol on aroma release within the oral cavity but no information on the travel of aroma compounds to the receptors in the nose has been provided, which could be also of interest for new studies. Overall, these findings could help winemakers produce new low alcohol wines considering the effect of ethanol on aroma release.

Supplementary Materials: The following are available online, Table S1: Endogenous concentration of the target aroma compounds in the control wine (0.5% v/v). Author Contributions: M.Á.P.-B. conceived the experiment. M.P.-J. and C.C. conducted the analyses. C.M.-G. performed the data analysis. C.M.-G. and M.Á.P.-B. wrote the manuscript. All authors reviewed the manuscript. Funding: This research was funded by the Spanish MINECO through the AGL201678936-R (AEI/FEDER, UE) Project. Acknowledgments: C.M.-G. and M.P.-J.thank the JdC and FPI programs (MINECO) for her respective postdoctoral and PhD research contracts (MINECO). The authors thank the volunteers who participated in this study and Francisco Javier Barroso for his technical assistance. Conflicts of Interest: The authors declare no conflict of interest. Molecules 2019, 24, 3253 9 of 10

References

1. Ickes, C.M.; Cadwallader, K.R. Effects of Ethanol on Flavor Perception in Alcoholic Beverages. Chemosens. Percept. 2017, 10, 119–134. [CrossRef] 2. Gawel, R.; Van Sluyter, S.; Waters, E.J. The effects of ethanol and glycerol on the body and other sensory characteristics of Riesling wines. Aust. J. Grape Wine Res. 2007, 13, 38–45. [CrossRef] 3. Zamora, M.C.; Goldner, M.C.; Galmarini, M.V. Sourness? Sweetness Interactions in Different Media: White Wine, Ethanol and Water. J. Sens. Stud. 2006, 21, 601–611. [CrossRef] 4. Escudero, A.; Campo, E.; Fariña, L.; Cacho, J.; Ferreira, V. Analytical Characterization of the Aroma of Five Premium Red Wines. Insights into the Role of Odor Families and the Concept of Fruitiness of Wines. J. Agric. Food Chem. 2007, 55, 4501–4510. [CrossRef] 5. Goldner, M.C.; Zamora, M.C.; Lira, P.D.L.; Gianninoto, H.; Bandoni, A. Effect of ethanol level in the perception of aroma attributes and the detection of volatile compounds in red wine. J. Sens. Stud. 2009, 24, 243–257. [CrossRef] 6. Guth, H. Quantitation and Sensory Studies of Character Impact Odorants of Different White Wine Varieties. J. Agric. Food Chem. 1997, 45, 3027–3032. [CrossRef] 7. King, E.S.; Dunn, R.L.; Heymann, H. The influence of alcohol on the sensory perception of red wines. Food Qual. Prefer. 2013, 28, 235–243. [CrossRef] 8. Doyennette, M.; Déléris, I.; Feron, G.; Guichard, E.; Souchon, I.; Trelea, I. Main individual and product characteristics influencing in-mouth flavour release during eating masticated food products with different textures: Mechanistic modelling and experimental validation. J. Theor. Biol. 2014, 340, 209–221. [CrossRef] 9. Gierczynski, I.; Guichard, E.; Laboure, H. Aroma perception in dairy products: the roles of texture, aroma release and consumer physiology. A review. Flavour Fragr. J. 2011, 26, 141–152. [CrossRef] 10. Muñoz-González, C.; Martín-Álvarez, P.J.; Moreno-Arribas, M.V.; Pozo-Bayón, M. Ángeles Impact of the Nonvolatile Wine Matrix Composition on the In Vivo Aroma Release from Wines. J. Agric. Food Chem. 2013, 62, 66–73. [CrossRef] 11. Salles, C.; Chagnon, M.-C.; Feron, G.; Guichard, E.; Labouré, H.; Morzel, M.; Semon, E.; Tarrega, A.; Yven, C. In-Mouth Mechanisms Leading to Flavor Release and Perception. Crit. Rev. Food Sci. Nutr. 2010, 51, 67–90. [CrossRef] 12. Baker, A.K.; Ross, C.F. Wine finish in red wine: The effect of ethanol and tannin concentration. Food Qual. Prefer. 2014, 38, 65–74. [CrossRef] 13. Petrozziello, M.; Asproudi, A.; Guaita, M.; Borsa, D.; Motta, S.; Panero, L.; Bosso, A. Influence of the matrix composition on the volatility and sensory perception of 4-ethylphenol and 4-ethylguaiacol in model wine solutions. Food Chem. 2014, 149, 197–202. [CrossRef] 14. Muñoz-González, C.; Rodríguez-Bencomo, J.J.; Moreno-Arribas, M.V.; Pozo-Bayón, M.Á. Feasibility and application of a retronasal aroma-trapping device to study in vivo aroma release during the consumption of model wine-derived beverages. Food Sci. Nutr. 2014, 2, 361–370. [CrossRef] 15. Tsachaki, M.; Linforth, R.S.T.; Taylor, A.J. Dynamic Headspace Analysis of the Release of Volatile Organic Compounds from Ethanolic Systems by Direct APCI-MS. J. Agric. Food Chem. 2005, 53, 8328–8333. [CrossRef] 16. Tsachaki, M.; Linforth, R.S.T.; Taylor, A.J. Aroma Release from Wines under Dynamic Conditions. J. Agric. Food Chem. 2009, 57, 6976–6981. [CrossRef] 17. Aznar, M.; Tsachaki, M.; Linforth, R.S.; Ferreira, V.; Taylor, A.J. Headspace analysis of volatile organic compounds from ethanolic systems by direct APCI-MS. Int. J. Mass Spectrom. 2004, 239, 17–25. [CrossRef] 18. Boelrijk, A.; Basten, W.; Burgering, M.; Gruppen, H.; Voragen, A.; Smit, G. The effect of co-solvent on the release of key flavours in alcoholic beverages; comparing in vivo with artificial mouth-MS nose measurements, Flavour research at the dawn of the twenty-first century. In Proceedings of the 10th Weurman Flavour Research Symposium, Beaune, France, 25–28 June 2002; pp. 204–207. 19. Tsachaki, M.; Gady, A.-L.; Kalopesas, M.; Linforth, R.S.T.; Atheès, V.; Marin, M.; Taylor, A.J. Effect of Ethanol, Temperature, and Gas Flow Rate on Volatile Release from Aqueous Solutions under Dynamic Headspace Dilution Conditions. J. Agric. Food Chem. 2008, 56, 5308–5315. [CrossRef] 20. Clark, R.; Linforth, R.; Bealin-Kelly, F.; Hort, J. Effects of ethanol, carbonation and hop acids on volatile delivery in a model beer system. J. Inst. Brew. 2011, 117, 74–81. [CrossRef] Molecules 2019, 24, 3253 10 of 10

21. Buettner, A. Influence of Human Salivary Enzymes on Odorant Concentration Changes Occurring in Vivo. 1. Esters and Thiols. J. Agric. Food Chem. 2002, 50, 3283–3289. [CrossRef] 22. Friel, E.N.; Taylor, A.J. Effect of Salivary Components on Volatile Partitioning from Solutions. J. Agric. Food Chem. 2001, 49, 3898–3905. [CrossRef] 23. Rodríguez-Bencomo, J.J.; Muñoz-González, C.; Andújar-Ortiz, I.; Martín-Álvarez, P.J.; Moreno-Arribas, M.V.; Pozo-Bayón, M.Á. Assessment of the effect of the non-volatile wine matrix on the volatility of typical wine aroma compounds by headspace solid phase microextraction/gas chromatography analysis. J. Sci. Food Agric. 2011, 91, 2484–2494. [CrossRef] 24. Esteban-Fernández, A.; Rocha-Alcubilla, N.; Muñoz-González, C.; Moreno-Arribas, M.V.; Pozo-Bayon, M.A. Intra-oral adsorption and release of aroma compounds following in-mouth wine exposure. Food Chem. 2016, 205, 280–288. [CrossRef] 25. Robinson, A.L.; Ebeler, S.E.; Heymann, H.; Boss, P.K.; Solomon, P.S.; Trengove, R.D. Interactions between Wine Volatile Compounds and Grape and Wine Matrix Components Influence Aroma Compound Headspace Partitioning. J. Agric. Food Chem. 2009, 57, 10313–10322. [CrossRef]

Sample Availability: Samples of the compounds are not available from the authors.

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).