Access features, news and views from the latest issue and from our chemistry archives.

January/February 2020

Modelling molecular assembly in wine

From time to time in these columns I have referred to molecular assembly processes in red wine. The assembly process is proposed to result from interactions between protein and polyphenolic compounds, or tannins, as polyphenolics are commonly called. The role of polysaccharides in influencing this tannin–protein interaction is still the subject of conjecture.

It is clear that there is an interaction between tannins and proteins: think what happens when milk is added to tea. In red wine production, protein is sometimes added in the final preparation of a wine prior to bottling to ‘soften the astringency’. The general argument is that, as the amount of tannin would be in far excess of the amount of protein added, then no protein should remain. If there is residual protein bound to tannin in the wine, then this could become an issue for allergen labelling.

The issue of residual protein in wine after fining has been investigated by various researchers. The research group at the University of Porto proposed in a series of articles from 2004 to 2006 that an assembly of soluble complexes resulted from the interaction of wine tannins and proteins, with additional interactions involving polysaccharides potentially giving a three-component soluble complex; that is, residual protein in the wine is possible. Rolland et al. (J. Agric. Food Chem. 2008, vol. 56, pp. 349–54) expressed a contrary view when no protein residue could be detected by ELISA assay in wines that had been previously fined with casein, ovalbumin or peanut, the last being used for research purposes only.

An alternative perspective was described by the oenology research group in Montpellier. After fining with gelatin or plant proteins and applying a sodium dodecyl sulfate treatment followed by a specific analysis for the polyphenolic content, a higher amount of these compounds was detected than the amount found without sodium dodecyl sulfate treatment (Am. J. Enol. Vitic. 2001, vol. 52, pp. 140–5; 2003, vol. 54, pp. 105–11). The increase was interpreted as the sodium dodecyl sulfate treatment releasing the protein from the bound tannin–protein complex, thereby releasing more tannin. It was also suggested that the ELISA technique used by Rolland et al. may not have been able to detect tannin–bound protein.

Until recently, evidence for soluble tannin–protein complexes was qualitative, perhaps even speculative. Seeking more rigorous evidence, the Montpellier group used a model system consisting of a polyphenolic powder extracted from Syrah and dissolved in aqueous ethanol. This solution was then fined with two hydrolysed wheat glutens and gelatin (Food Chem. 2019, vol. 279, pp. 272–8). After fining and centrifugation, the viscosity of the supernatant was measured. A small, but measurable, increase in viscosity was observed for the fined samples compared to the model wine. As any unbound fining agent did not affect the viscosity, it was argued that the increase was due to the presence of soluble tannin–protein aggregates.

Not content with the viscosity result alone, the authors used a labelled protein in the fining experiments. Residual radioactivity was found in the supernatant after fining, clearly indicating the presence of residual protein in the supernatant. This, in combination with the increased viscosity and a higher proportion of amino acids specific to the fining agent in the supernatant after fining, was clear evidence for the presence of tannin–protein aggregates.

Nanoparticle tracking analysis (NTA) is finding more applications in wine research. Recently researchers from the University of Adelaide and the Australian Wine Research Institute used NTA as well as dynamic light scattering (DLS) to examine what they called ‘the characterisation of the polydispersity of aggregates resulting from tannin–polysaccharide interactions’ (Molecules 2019, vol. 24(11), p. 2100). Using a wine-like solution, the authors tracked the interaction between grape seed tannin with either mannoprotein or arabinogalactan. The observed shift in particle size seen by NTA was consistent with aggregation, with further confirmation of this aggregation being obtained by DLS.

The two polysaccharides behaved differently in their interaction with the seed tannin. Mannoprotein formed ‘large, highly scattering aggregates’ while only a weak interaction was found with arabinogalactan. The authors suggest that their results support the proposal that polysaccharides bind with tannin and thus limit access to the tannin by protein (see Anal. Chim. Acta 2004, vol. 513, pp. 135–40 for a detailed description of the optional binding models). My issue is whether it is the protein component of the mannoprotein that is the basis for the observed aggregation. I do agree with the authors that more competitive experiments with different proteins and other wine polysaccharides need to be performed to refine our understanding of the aggregation process.

The Adelaide researchers found that a 3% variation in the ethanol concentration of the model resulted in a marked change in aggregation: higher aggregation occurred at the lower ethanol concentration. This, as the authors point out, may be significant in sensory terms as the extent of aggregation versus ethanol concentration may impact on the taster’s perception of astringency and bitterness.

The outcomes of the NTA experiments open the way for additional experimentation to enhance our knowledge of sensory perception. It is essential that experiments are performed on wines of known chemical composition. Perhaps in place of fining proteins, experiments using human saliva could be performed and the aggregation results correlated with descriptive sensory analysis.

Geoffrey R. Scollary FRACI CChem ( has been associated with the wine industry in production, teaching and research for the last 40 years. He now continues his wine research and writing at the University of Melbourne and the National Wine and Grape Industry Centre at Charles Sturt University.

Book and software reviews

To offer your services as a book or software reviewer for Chemistry in Australia, please contact Damien Blackwell at