Characterization of natural haze protein in sauvignon white wine

Abstract

The natural precipitate of a white Sauvignon wine was dissolved, lyophilized and analyzed by molecular exclusion FPLC and/or electrophoresis (SDS–PAGE, native and isoelectrofocusing) using three types of stains (Coomassie blue, silver stain and PAS stain). The protein spots were digested and analyzed by MALDI-TOF/TOF MS. Direct analysis of the precipitates by SDS–PAGE gave no bands with Coomassie blue stain and five bands with silver stain. It was not possible to identify these bands by MALDI-TOF/TOF MS. However, applying molecular exclusion FPLC prior to SDS–PAGE improved sensitivity and enabled three bands to be identified. These were the previously described thaumatin-like protein (VVTL1) and two proteins that to our knowledge have never been described as components of protein haze: β-(1-3)-glucanase and ripening-related protein grip22 precursor.

Introduction

One of the main factors affecting the stability of white wine during storage is the instability of proteins (Bayly and Berg, 1967, Hsu and Heatherbell, 1987a). Though wine proteins are usually found at very low concentrations, some of them may precipitate due to lack of stability (Bayly and Berg, 1967, Waters et al., 1991). The appearance of haze in a bottle of wine can seriously harm the image of the wine and affect its acceptance by the consumer.

The wine industry systematically tries to stabilize white wines and prevent protein precipitation. Currently the best way to eliminate the risk of protein precipitation is bentonite fining (Hsu and Heatherbell, 1987b, Pocock and Waters, 2006). Alternative methods such as ultrafiltration (Hsu, Heatherbell, Flores, & Watson, 1987) and zirconium oxide treatment (Salazar, Achaerandio, Labbe, Guell, & Lopez, 2006) can also be effective but their use is still experimental.

Bentonite interacts electrostatically with proteins and this produces flocculation (Hsu and Heatherbell, 1987b, Hsu et al., 1987). However, protein instability does not correlate well with total protein concentration because individual proteins behave differently (Bayly and Berg, 1967, Hsu and Heatherbell, 1987a). Other factors may also have a role in protein precipitation (Siebert, 1999, Waters et al., 1995). It has recently been suggested, for example, that sulfate may be a key factor (Pocock, Alexander, Hayasaka, Jones, & Waters, 2007).

The literature contains several tests for determining the stability of wine proteins and there are even some commercial tests on the market. However, these tests sometimes provide different results, which creates uncertainty for the winemaker (Dubourdieu et al., 1988, Sarmento et al., 2000, Toland et al., 1996). As there is no clear criterion for selecting the correct dose of bentonite to stabilize a wine, winemakers usually treat wine with high doses.

However, an excess of bentonite can have negative effects on wine aroma (Lubbers, Charpentier, & Feuillat, 1996) and mouthfeel (Guillou, Aleixandre, Garcia, & Lizama, 1998). Bentonite also affects the foaming properties of sparkling wines (Martinez-Rodriguez and Polo, 2003, Vanrell et al., 2007). For these reasons, the chemical nature of protein precipitates is still a subject of great interest in oenology.

Several techniques have been used to study the protein fraction of white wines: gel electrophoresis, isoelectric focusing (Dawes, Boyes, Keene, & Heatherbell, 1994), capillary electrophoresis (Dizy & Bisson, 1999), HPLC (Sung-Won, 2004), affinity chromatography, FPLC (Canals, Arola, & Zamora, 1998) and chromatofocusing on FPLC (Dawes et al., 1994). More recently, N-terminal sequencing (Ferreira et al., 2000, Waters et al., 1998, Waters et al., 1996) and mass spectrometry (Hayasaka et al., 2001) methods have been used.

Few of these techniques have been used to study unstable proteins, however. Bayly and Berg (1967) reported that the proteins with the lowest pI were the most unstable against the heat test. These data were confirmed by Hsu and Heatherbell (1987b), who reported that unstable proteins have a low isoelectric point (4.1–5.8) and a low relative molecular mass (13–30 kDa). More recently Waters et al. used the back-addition technique (Waters et al., 1991, Waters et al., 1996, Waters et al., 1998) to identify pathogenesis-related proteins (thaumatin-like proteins and chitinases) as causes of haze.

All the above studies were conducted using heat forced precipitates (Hsu & Heatherbell, 1987b) or back-addition techniques (Waters et al., 1991). To our knowledge only Koch and Sajak (1959) and Sarmento et al. (2000) have analyzed the naturally occurring precipitate. The aim of the present study was to characterize the proteins in a natural precipitation of a white Sauvignon wine.