Elsevier

Food Quality and Preference

Volume 17, Issues 1–2, January–March 2006, Pages 96-107
Food Quality and Preference

Prediction of perceived astringency induced by phenolic compounds II: Criteria for panel selection and preliminary application on wine samples

https://doi.org/10.1016/j.foodqual.2005.04.009Get rights and content

Abstract

In the following work, subject saliva characteristics affecting panel astringency evaluation in phenolic mixtures were studied. Sixty subjects were selected on the basis of their salivary flow, haze developing capacity and protein concentration. Subjects rated the perceived astringency of tannic acid (TA), commercial procyanidin (PA) and grape seed extract (GSE) solutions with concentration values ranging from 0.42 to 1.4 g/L. Astringency intensity perception proved to be inversely related to saliva flow rate and haze developing capacity. No significant correlations were found between saliva protein concentration and intensity of astringency perception. A panel selected on the basis of subject similarity for flow rate and haze developing capacity rated the astringency intensity of set sample training solutions of TA, PA and GSE with concentrations ranging from 0.39 to 4.48 g/L. The reactivity of the same astringent solutions with mucin was measured in an in vitro assay and expressed in terms of Nephelometric Turbidity Units (NTU). Three predictive models described by a linear regression of astringency intensity vs. NTU were found. The possibility of a practical application of the proposed assay for optimization of wine production was evaluated on 18 experimental wines. A linear correlation was found between the intensity astringency ratings of wine samples and the in vitro assay response.

Introduction

The sensation referred to as astringency is described as drying, roughing or puckering of the oral cavity (Lee & Lawless, 1991). There is evidence to indicate that it is a tactile sensation (Braslin, Gilmore, Beauchamp, & Green, 1993) generated by the reduction of lubrication in the oral cavity (Prinz & Lucas, 2000). Astringency can be elicited by various substances such as multivalent metallic cations (particularly aluminium salts), dehydrating agents (ethanol), mineral and organic acid and plant phenolic compounds. Polyphenol compounds in plant-derived food and beverage have been shown to have important physiological properties and may be responsible for both beneficial and detrimental effects on human health (Chung et al., 1998, Singh et al., 2003). Polyphenol anti-oxidant activity probably accounts for their role in preventing diseases related to oxidative stress. Conversely, polyphenols have been reported to be responsible for anti-nutritional effects including inhibition of digestive enzymes, formation of relatively less digestible complexes with dietary proteins, depressed growth in rats, altered food consumption. Polyphenolic compounds form complexes with salivary proteins. Human saliva contains prolin rich proteins (PRP) that have been demonstrated to have a high affinity with polyphenols (Haslam, 1998). These protein are present as three main classes: acidic, basic and glycosilated (Bennic, 1982). Their role in inhibiting or promoting the sensation of astringency induced by phenol compounds has still not completely clarified. The formation of insoluble aggregates phenols-protein has lead to the theory that astringency is, at least initially, due to delubrication via removal of the slippery coating on oral surfaces (Haslam, 1998, Kallithraka et al., 1998). The binding of salivary protein to polyphenols may serve to sequester or inactivate polyphenols and thus protect the alimentary tract from their deleterious effect on nutritional uptake (Bacon and Rhodes, 2000, Horne et al., 2002). Animals on a high polyphenol diet may increase secretion of PRP salivary proteins (Kim, House, & Miller, 2004) which can modulate their response to astringency (Gleddining, 1992).

Physiological factors related to saliva flow and characteristics (Fisher et al., 1994, Horne et al., 2002, Siebert and Chassy, 2003) as well as individual sensitivity to oral sensations (Pikering, Simunkova, & DiBattsta, 2004) can modulate the intensity of astringency induced by phenolic compounds. The mouth-feel perception of astringency in food products containing phenols is the result of a highly complex process. It depends on the presence of individual food components which can exert synergic or antagonistic effects on astringent compounds affecting the overall intensity of perceived sensation (Riou et al., 2001, Sowalsky and Noble, 1998).

The intensity of perceived astringency is an important determinant of consumer response to red wine. Astringency descriptors account for more than a half of the total terms in the mouth-feel wheel proposed for describing the characteristics of red wine (Gawell, Oberholster, & Francis, 2000). Polymeric flavan-3-ols, which occur either as galloylated species, conjugated with anthocyanins or in free form, are believed to be largely responsible for red wine astringency to which they contribute in a different manner depending on their chemical structure and composition (Vidal et al., 2004).

The possibility of estimating the strength of astringency induced by different phenolic compounds could definitely help to optimize processing conditions in respect to this important driver of wine acceptability. The reactivity of polyphenols with proteins has often been used for this purpose (Bacon and Rhodes, 1998, Carvalho et al., 2004, Edelmann and Lendl, 2002, Glories, 1978, Sarni-Manchado et al., 1999). In a previous work (Monteleone, Condelli, Dinnella, & Bertuccioli, 2004) astringency induced by commercial polyphenolic extracts proved to be linearly related to the sample’s capability of developing turbidity by reacting with mucin in an in vitro assay. On the base of this linear regression a predictive model of perceived astringency was proposed.

The main objective of the present paper is to improve the predictive capacity of the model by reducing the effect of individual physiological differences on astringency ratings. The effect of mean subject salivary characteristics, in terms of flow rate, haze developing capacity and protein concentration, on the intensity of perceived sensation was investigated using tannic acid and grape seed extracts as astringent compounds. Furthermore the suitability of the in vitro assay for predicting astringency in red wine was investigated.

The experimental plan consisted of three stages:

  • 1.

    The first stage was aimed to study the effect of subject’s saliva characteristics (flow rate, haze developing capacity and protein concentration) on mean astringency ratings from phenolic mixtures in order to define subject selection criteria.

  • 2.

    The second stage was aimed to validate an astringency predictive model based on the relationship between perceived astringency and turbidity development in the mucin in vitro assay of selected phenolic compounds.

  • 3.

    The third stage was aimed to study the relationship between the in vitro assay response and the astringency experienced by the panel when tasting the same wine samples.

Section snippets

Subjects

Sixty subjects, 20 males and 40 females, aged from 22 to 27 took part in the experiment.

On the basis of flow rate and haze forming capacity values, 23 subjects were selected from the recruited panel to participate in the sensory tests aimed to build up and validate the prediction model of astringency induced by phenols.

All the subjects were students at University of Basilicata. Subjects were informed about the nature of the solutions used in the study and they gave their consent to take part in

Factors affecting mean astringency ratings from phenolic compounds

In the following paragraphs the effect of phenolic compound, solution concentration and subjects’ saliva characteristics on mean astringency ratings are discussed on the basis of experimental results from the evaluation of “subject set” samples.

Conclusions

Astringency perception in food and beverage containing phenols is a highly complex process affected by several factors: the chemical composition of astringent compounds, individual physiological factors and interaction between astringent compound and individual food product components. The identification of predictors of the strength of the sensation induced by phenolic compounds could make an important contribution to the optimization of food processing conditions. It is extremely important to

Acknowledgements

The research was funded by the Italian Ministry of Education, University and Research (MIUR) “Composti fenolici, colore, aroma e astringenza delle uve e dei vini rossi di Basilicata”—Progetto Rientro Cervelli—2002/2005.

References (43)

  • N.J. Baxter et al.

    Multiple interactions between polyphenols and salivary protein. Repeat results in complexation and precipitation

    Biochemistry

    (1997)
  • A. Bennic

    Salivary proline rich proteins

    Molecular and Cellular Biochemistry

    (1982)
  • P.A.S. Braslin et al.

    Psychophysical evidence that oral astringency is a tactile sensation

    Chemical Senses

    (1993)
  • A.J. Charlton et al.

    Polyphenol/peptide binding and precipitation

    Journal of Agricultural and Food Chemistry

    (2002)
  • K.T. Chung et al.

    Tannins and human health: a review

    Critical Reviews in Food Science and Nutrition

    (1998)
  • V. de Freitas et al.

    Structural features of procyanidin with salivary proteins

    Journal of Agricultural and Food Chemistry

    (2001)
  • R. Di Stefano et al.

    La determinazione dei polifenoli totali nei mosti e nei vini

    Vigne e Vini

    (1989)
  • A. Edelmann et al.

    Towards the optical tongue: flow-through sensing of tannin–protein interactions based on FTIR spectroscopy

    Journal of the American Chemical Society

    (2002)
  • U. Fisher et al.

    Physiological factors contributing to the variability of sensory assessments: relationship between salivary flow rate and temporal perception of gustatory stimuli

    Food Quality and Preference

    (1994)
  • Folin-Ciocalteau Index, (1992). Off. J. Eur. Communitie...
  • R. Gawell et al.

    A mouth-feel wheel: therminology for communicating the mouthfeel characteristics of red wine

    Australian Journal of Grape Wine Research

    (2000)
  • Cited by (0)

    View full text