|
One of the most popular red wines to come out of Australia is the Shiraz, produced from the Shiraz grape that is grown in most regions of the country. The flavour and aroma depend on the locality of the vineyard. Warmer regions give plum, blackberry and chocolate notes while the cooler areas tend more towards the distinctive black pepper flavour associated with this wine. Maturing in American oak, rather than French oak, helps to confer a spicy, vanilla-type characteristic. The compounds behind the pepper aroma and flavour have proved elusive so far. Scientists from the Australian Wine Research Institute have attempted to identify them by gas chromatography with olfactory (GC-O) and mass spectrometric (GC/MS) detection with a spectacular lack of success, finding no candidates. One of the reasons, according to the Institute's Alan Pollnitz, is the incredible complexity of the volatile fraction of grapes and wines. Several hundred compounds have to be sifted through, many of them at trace levels. So, Pollnitz and some of his colleagues decided to try a different approach, again based on sensory and GC/MS but backed up by chemometric analysis of the data. Guided by experts, they collected potentially peppery grapes from 12 vineyards in Victoria and South Australia over two different years. As a first step, grapes from each set were homogenised and subjected to sensory analysis by 10 experts who rated the "black pepperiness" on a scale of 0-10. They observed that the pepper attribute was independent of other characteristics such as grassy, green apple, tobacco, compost and red berry but correlated with the spicy attribute. There was also a strong correlation between the intensity of the black pepper flavour on the palate and the intensity of the aroma. This preliminary test allowed the researchers to concentrate on the more peppery grapes in the ongoing studies. In the next step, the volatiles in the grapes were analysed by static headspace GC/MS in a broad, non-targeted manner. Blended samples were held in a vial and the volatile compounds were cryofocussed in a cooled inlet system held at -50°C prior to injection. This temperature control protected thermally labile compounds against decomposition and aided the detection of trace compounds. About 13,000 mass spectra were obtained for each sample. They were processed by multivariate analysis, constructing composite mass spectra by summing the integral m/z values in each spectrum. Using a broad range of m/z (from 35-220), an excellent correlation was obtained between the predicted scale of the pepperiness and that rated by the sensory panel. This confirmed that GC/MS could predict the degree of spicy/black pepper attribute of a grape without knowing anything about the compounds involved. Of course, most chemists will find this "non-chemical" approach unsatisfactory, so the team decided to dig a little deeper. A partial least-squares analysis revealed that five ions in the range m/z 50-100 strongly influenced the prediction, explaining 99% of the variations in pepper. They comprised four ions at m/z 51, 60, 74 and 92 giving strong positive influences an ion at m/z 85 giving a strong negative influence. A reexamination of the GC/MS data using extracted ion chromatograms, looking for spectra containing the four positively influencing ions but missing m/z 85, provided only one candidate compound. It was identified from its Kovats retention index and mass spectrum as the sesquiterpene alpha-ylangene. This is a surprising marker compound for the peppery aroma and flavour in Shiraz grapes, because it does not have peppery/spicy attributes and, in fact, is not a strong aroma compound at all. Nevertheless, it was the most abundant sesquiterpene out of 28 identified by Pollnitz and his team. It probably reflects other factors associated with the peppery note, such as cool climate, location and grape cultivar. Alternatively, it might be converted to the peppery compound(s) during biosynthesis. As confirmation of the effectiveness of alpha-ylangene as a marker, its concentration was measured by GC/MS in the Shiraz grapes. There was a good correlation for high peppery grapes but a much poorer correlation for less peppery samples, although it gave the best correlation of all the identified sesquiterpenes. The team predicted that the concentration-pepperiness relationship is non-linear. Although present in grapes, alpha-ylangene was absent from peppery wines, so cannot be used as a wine marker. The compound is probably retained in the chapeau de marc (the solid material, including grape skins, that rises to the surface) during winemaking, so is effectively removed. Having identified a marker for the black pepper attribute in Shiraz grapes and peppery wines, the team are now attempting to identify the actual compounds responsible in ongoing studies. Related Links:
The views represented in this article are solely those of the author and do not necessarily represent those of John Wiley and Sons, Ltd. |
|
|
|
|
|
