New approaches for understanding the formation of mouthfeel properties in wines and grapes stars

Abstract

It is of great interest for the wine sector to understand the factors that increase the acceptability of the product. The identification of these factors will provide the industry with useful tools to increase the acceptance of our wines and thus add value to them. The degree of acceptability and consumer satisfaction is directly related to the perception of quality. Perceived quality is a multidimensional variable, comprising both intrinsic (organoleptic properties) and extrinsic (label design, packaging...) factors of the product, whose interaction with the consumer and its characteristics (for example: level of expertise with wine or culture) give rise to the formation of perceived quality. Intrinsic properties, especially flavour, play a very important role in quality formation, especially in repurchase situations. Flavour is the result of the integration of aromatic stimuli, taste and sensations mediated by the trigeminal nerve (tactile sensations in the mouth). In the last decade there has been a significant increase in the interest for understanding flavor formation, which implies understanding how this multisensory integration occurs at both the perceptual and cognitive levels. Among the senses involved in flavour (smell, taste, tactile sensations), the formation of tactile sensations is generally the least understood, which is especially important in the case of wine. The perception of tactile properties occurs mainly in the oral cavity, where mechanical and chemical processes take place. To date, despite advances in instrumental chemical analysis, the taste properties (acid, sweet, bitter, salty and umami) and mouthfeel sensations of food in general and wine in particular, can only be determined by sensory strategies. These techniques make it possible to obtain descriptions of the products and identify the differences that exist among them. This information is of great interest to the industry, as it helps to increase the understanding of the factors that are involved in consumer's preference and the perceived quality of the product. This knowledge can be used in the development of new products, as well as in the development of quality control strategies, among others. However, it is difficult to obtain a complete sensory profile in terms mouthfeel properties of wine, especially for red wines. This difficulty lies in the fact that the compounds that originate these properties are mostly unknown. This fact generates a lack of an adequate standardised vocabulary that objectively describes the dimensions of wine mouthfeel, which definitely hinders the study and therefore the understanding of the mechanisms that originate these sensations. En este contexto, el objetivo principal de esta Tesis Doctoral ha sido aumentar el conocimiento sobre la formación de las propiedades táctiles en boca y los compuestos y parámetros químicos que conducen y modulan esta percepción en vinos y uvas. Para lograr tal objetivo, se trabajó bajo la premisa de que la comprensión de las sensaciones táctiles en boca inducidas por los vinos debe abordarse combinando estrategias tanto sensoriales como químicas. In this context, the main objective of this Doctoral Thesis has been to increase knowledge about the formation of mouthfeel properties and the chemical compounds and parameters that lead and modulate this perception in wines and grapes. To achieve this goal, we worked under the premise that the understanding of mouthfeel properties induced by wines must be addressed by combining both sensory and chemical strategies. Thus, the first section was devoted to develop a sensory-directed chemical strategy to obtain odourless fractions with consistent taste and mouthfeel properties. The fractions from three wines were sensory characterised using different techniques: sorting task, repertory grid, triangulation, and Rate-All-That-Apply (RATA) with wine experts. The most surprising result was the sensory properties of the anthocyanic fraction (containing anthocyanins and derivatives), which was especially dry, bitter and persistent as the original wine. In addition, this approximation allowed to generate a list of 18 terms related to mouthfeel properties and was successfully used in the description of wines and fractions by RATA analysis. Next, this senso-chemical methodology was applied in the study of the "green character" of red wines. At the beginning of this study the "green character" was an ill-defined term. Winemakers declared that it was a default linked to mouthfeel sensations and taste of certain red wines. They noted that it was a sensation that they were observing recurrently in their wines associated with climate change and that was generating a depreciation of their products. In order to identify the compounds or groups of compounds responsible for this sensation, first of all, the study of the definition of this ill-defined descriptor was addressed. Thus, the "green character" resulted to be a multidimensional term associated with dimensions of aroma, taste and mouthfeel sensations, which was effectively negatively correlated with wine preference. Then, following the strategy of section I, different odourless fractions of wines with high and low green character were obtained. One of the most outstanding results was that the anthocyanic fraction of one of the greenest wines was especially "adherent". This fraction was suggested as responsible for the green character in that wine. These results, in line with those obtained in section I, suggested that anthocyanins and their derivatives could be involved in the formation and modulation of certain mouthfeel properties of some red wines. To deepen in the study of the sensory properties of the anthocyanic fraction, a large-scale study was carried out considering 42 red wines. Two different fractions were obtained for each wine containing tannins or anthocyanins, respectively, and they were chemically and sensory characterized in terms of taste and mouthfeel sensations. The results confirmed the sensory activity of the anthocyanic fractions studied, which varied significantly in the terms: bitterness, dryness, granulosity and body among the fractions derived from the wines studied. The study of the mouthfeel properties of red wines requires considering tannins (this work focuses mainly on condensed tannins, polymerised flavanols also known as proanthocyanidins). The literature is consistent with respect to the sensory activity of these compounds, stating that they are mainly responsible for wine astringency (i.e., dryness). The modulation of this sensation during the ripening of the grape and the ageing of the wine is widely recognised. However, the correlation between the changes that occur in the chemical composition during grape ripening and wine ageing and the sensory changes remain relatively unknown. In order to increase knowledge in this field, the effect of grape maturity and oxygen supply on both the taste and mouthfeel properties of grapes and red wines was analysed. As far as the ripeness of the grape is concerned, it represents an important factor that determines the composition of the grape and consequently the sensory properties of the wines made with these berries. The study consisted of the elaboration of a total of 21 wines (7 wines in triplicate) with grapes of the Moristel variety. These grapes were harvested at different points of maturity (each point separated by one or two weeks) in two plots of very different characteristics. The results of the study showed that the ripeness of the grape generated significant sensory effects on the astringency (i.e., dryness) of the wine and on the fruity aromas of the wines ('black fruit', 'red fruit' and 'raisins'). An important result found in this study is related to the fact that grapes harvested early, even just one week in advance of the optimal point (defined as the one in which maximum values of aromas of red and black fruits and a moderate astringency in wines are obtained) have given rise to wines with higher levels of oxidation aldehydes (acetaldehyde, metional, phenylacetaldehyde and isoaldehydes). These oxidation notes are related to lower levels of certain polyphenols capable of reacting with the aforementioned aldehydes, called in previous works as ARPs (aldehyde-reactive polyphenols). This fact is also supported by negative correlations found between the aldehydes involved and various parameters related to polyphenols (IPT, concentration of tannins or pigments). Further, the effect of oxidative and reductive ageing (key in the production of long-living wines) on the chemical composition and sensations in the mouth (mouthfeel and taste) has been studied. Therefore, the evolution of a young red wine and its phenolic fractions has been evaluated in the absence and presence of oxygen. Oxygen consumption was measured using a non-invasive method based on luminescence, in order to increase knowledge about the oxygen consumption of the different compounds in wine. In parallel, replicates of the same samples were kept in anoxia (not oxygenated). After exposure to oxygen, all samples were kept inside an anoxic chamber (in the absence of oxygen). Both chemical and sensory analyses were carried out at two different times, 6 weeks after oxygenation and 24 weeks later. The chemical analysis of the samples showed an important modulating effect of the non-tannic phenolic fractions (flavanols, flavonols, anthocyanins and derived pigments) on the tannic activity (measured as the variation of enthalpy of interaction between proanthocyanidins and a hydrophobic surface) manifested by the tannic fraction of the wine. Thus, the activity of tannins can be reduced by simple addition of non-tannic phenolic compounds. As for the study of oxygen consumption, the results showed that the presence or absence of manganese can make differences in the oxygen consumption rate of fractions with the same phenolic composition. From the study of the effect of oxygen on the chemical parameters analysed, it should be noted that while oxidative ageing induces an increase in the parameter of tannic activity, anoxic ageing, both in oxygenated and non-oxygenated samples, induces changes in the structure of tannins increasing the % of prodelphinidins. Regarding sensory changes induced by oxygen, the most significant changes related to mouthfeel properties were observed in the fractions containing proanthocyanidins, in samples evaluated after 6 weeks of oxygen consumption. These sensory differences between non-oxygenated and oxygenated samples disappear after both samples are aged under anoxic conditions. The study of the mouthfeel properties of wines was extended to the study of the phenolic fractions of grapes. This study is of great importance since the phenolic compounds present in wines have their origin in the grape. The phenolic compounds of the grape are extracted mainly from the skins and seeds during the maceration and fermentation processes. In this context, a total of 31 grape extracts were obtained and subsequently sensory characterized (through non-verbal and verbal strategies, the latter developed in this thesis) and further submitted to targeted chemical analyses (i.e., quantification of known compounds). The results showed significant sensory differences among the 31 polyphenolic grape extracts. Sensory variables were predicted from chemical parameters by PLS regression. The activity and concentration of the tannins together with their average degree of polymerisation resulted to be good predictors of the sensation of dryness. The concentration of the polymeric pigments capable of precipitating with ovalbumin seems to be involved in the "adherent" perception and the flavonols in the "bitter" taste. These results increase knowledge about the properties of grape and suggests this methodology to infer grape quality. Finally, untargeted chromatographic methods were developed and applied with the aim of identifying molecular markers that generating different taste and mouthfeel properties. This methodology overcomes the main limitations of classical directed instrumental techniques since they allow us to consider both unknown metabolites as well as those occurring at low concentrations and that, without a doubt, can play a determining role in the formation of taste and mouthfeel properties. For this study, a total of 42 wines were sensory characterised and their metabolomic profile was obtained by non-directed analysis using UPLC-HRMS-QTOF instrumental technique. The results of this study allowed to obtain very satisfactory PLS regression models predicting sensory variables from chemical parameters. Among the most interesting markers found are those derived from sulfonated flavanols, which according to the PLS models obtained, are involved in reducing the mouthfeel sensation of wine dryness. Similarly, the results suggest that amino acids and peptides are involved in modulating the dryness and oily attributes. In addition, it was possible to confirm the sensory role of anthocyanins and their derivatives in the perception of taste and mouthfeel of red wines. These results establish the basis for formulating different hypotheses related to the sensory activity of different compounds of red wines and whose involvement in taste and mouthfeel will be confirmed with reconstitution studies. The results of this Doctoral Thesis show how different strategies that combine chemical and sensory techniques have proven to be effective in the study of different oenological concepts ("green character", maturity or oxidative aging), and have allowed to expand the knowledge about the properties of taste and mouthfeel generated by wines. The development of a broad sensory vocabulary related to wine mouthfeel in section 1 has allowed to deepen the study of the compounds and / or factors causing the different sensations perceived through the development of mathematical models. In addition, the discriminant capacity of mouthfeel terms has been demonstrated in wine and in polyphenolic fractions obtained from grapes and wines. While the strategies developed during this work, which employ chemical-sensory methods along with both directed and undirected instrumental techniques, have shown satisfactory results, they still have limitations and the development of new analytical tools is a key factor for success in understanding properties of taste and mouthfeel sensations. The determination of the molecular structure of the compounds involved in taste and mouthfeel sensations is a great challenge that is fundamental to be able to fully understand their role. In this context, techniques such as voltammetry and spectrofluorometry could increase the variety of sensory attributes satisfactorily modelled.