deeper reds are often more desirable and so attract a higher price

Red, red, red wine: A touch too much anthocyanin? 
[August 15, 2011]

Adultery and red wine

Various approaches to statistical analysis of spectroscopic data can reveal whether red wine has been adulterated with anthocyanins to artificially improve, "correct", its red colour.

E. Ferrari, Lorenzo Tassi, Giorgia Foca and Alessandro Ulrici of the Università di Modena e Reggio Emilia, working with M. Vignali of Vinicola San Nazaro in Italy point out that the red of red wines are commonly and legitimately "corrected" by blending with Rossissimo, a wine rich in anthocyanins. The process is a natural part of the Italian oenological industry, they say. Anthocyanins, ironically named from the Greek meaning "blue flower" belong to the class of molecules called flavonoids. They can, of course, be blue, purple and red, depending entirely on the pH in which the molecules find themselves. Anthocyanins occur in all tissues of higher plants, including leaves, stems, roots, flowers, and fruits, so giving rise to a range of colours in all kinds of vegetation. The content of anthocyanins in red wine, is between 240 and 350 milligrams per litre and given that red wine is usually at acidic pH these indicator compounds are red in the bottle and glass.

Knowing that deeper reds are often more desirable and so attract a higher price there are some parts of the wine market, particularly in Asia, where anthocyanins extracted from black rice and other sources are added to improve wine colour. "This practice does not produce negative effects on health," the Italian researchers say, "however, in many countries, it is considered as a food adulteration." For those markets where artificially boosting the reds is either illegal or simply frowned upon, there is a requirement for a simple and quick approach to testing red wine for such adulteration.

Efficient discrimination

The team thus set out to develop an efficient method for discriminating between red wines containing anthocyanins from natural grapevine sources or that has been adulterated with anthocyanins from black rice. Key to success would be the development of a reliable spectroscopic technique that needs minimal sample preparation but provides rapid results with few false positives.

The team prepared two series of samples from five original wines, to which they added different amounts of Rossissimo or a solution of black rice anthocyanins. The samples were prepared so that desired Colour Index values were reached. The researchers then analysed the "wines" using Fourier-transform near-infrared (FT-NIR) and proton nuclear magnetic resonance spectroscopies.

The spectroscopic data was then subjected to multivariate classification. "Initially, partial least-squares discriminant analysis (PLS-DA) was used as classification method," the team says, "then also variable selection/classification methods were applied, i.e. iPLS-DA and WILMA-D."

Blind classification

The researchers found that blind classification using variable selection of the NIR spectra gave them a classification efficiency of only about 70 percent with their test set samples. They suggest that this was an artefact of the matrix effect combined with the lack of sensitivity of NIR for minor components of the samples. However, by contrast they were able to get an efficiency validation of a much more satisfactory level of greater than 95 percent when they analysed the NMR data. They focused specifically on the spectral region corresponding to aromatic groups (i.e. chemical shifts lying between delta 6.5 and 9.5 ppm) and used wavelet-based variables selection.

Of course, NMR is a lot more expensive and slower than FT-NIR, so the pros and cons of each must be weighed up practical analyses. However, the team points out that two-dimensional correlation analysis between the FT-NIR and proton NMR data highlighted the fact that the more sensitive and intelligible NMR spectra could be used interpret overlapping NIR bands, which could be used to improve the analytical algorithms used to determine the anthocynanin origin in the samples. The legitimacy and ethics of food and drink adulteration aside, one wonders whether given the alleged health benefits of flavonoids associated with their purported antioxidant properties, that the adulterers in boosting the apparent quality of their reds might inadvertently be improving the health of their customers.