Senior Lecturer in Wine Science, Ron Potter Centre, CSU-Riverina
Trace analysis has achieved great importance in health and environmental studies in the monitoring of pollutants, pesticides and drugs. This emphasis on what should not be around us can lead us to forget that some trace components are very important aromas, conveying a vital awareness of our environment. Amongst these are desirable food and beverage flavours. Our sensory awareness of them relies upon the remarkable sensitivity and selectivity of our own sense of smell which frequently allows detection of components at parts per billion, and sometimes even parts per trillion level.
An interest in the grape-derived aroma of wine led us to the methoxypyrazines, particularly 2-methoxy-3-isobutylpyrazine (Figure 1). This compound is responsible for the very characteristic 'green, herbaceous, or vegetative' aromas of Sauvignon blanc and Cabernet Sauvignon wines. Amongst aroma components, this is close to being the most potent. We have an ability to detect incredibly small concentrations of this compound. We have shown that it contributes to wine aroma even at 1 ng/L, a remarkable 2 parts per trillion.
The low concentration of this methoxypyrazine has challenged several research groups. In 1975, a French group first tentatively identified it in Cabernet Sauvignon grapes, and in 1982 a South African and German group tentatively identified it in Sauvignon blanc grapes. The techniques of neither research team could identify how much of the component might be present, so no application of their findings was possible.
Success in definitive identification and quantification came with our development of a gas chromatography/mass spectrometry (GC/MS) method (Figure 2). The use of mass spectrometry was essential to provide not only outstanding sensitivity, but also the powerful selectivity required to identify traces of methoxypyrazines amongst a complex matrix of other components. Mass spectrometry also allowed the use of synthetic methoxypyrazines, modified by substitution of deuterium for hydrogen at the methoxyl group, as internal standards for accurate quantitation. Although virtually identical to the natural compounds in every other respect, making them ideal internal standards, they are clearly distinguishable by mass spectrometry. Positive ion chemical ionisation was used to improve identification; selected ion monitoring to improve detection levels (Figure 3); and an isolation procedure that makes use of the basicity of the components to minimise solvent evaporation to 1 mL or less to avoid loss of the traces of these volatile components.
This provides an analytical detection theshold of 0.15 ng/L, less than one part per trillion, and a level well below the sensory detection threshold of about 2 ng/L. To provide an indication of this level of detection consider that, compared to the circumference around the earth, one part per trillion is the thickness of a human hair. The wine quantity required for this (60-240 mL) leaves sufficient from a wine bottle for relaxed contemplation of the wine aroma at the end of the day.
A little of this methoxypyrazine has a pronounced effect. Its concentration in Sauvignon blanc wines is typically 5-30 ng/L. Below 5-10 ng/L, the aroma is subdued; at 15-20 ng/L it provides an aroma that is distinctive, characteristic of the grape variety, and frequently balanced with other flavour components in the wine; at only 30 ng/L it begins to be rank and overpowering, giving a pungent aroma intensity that is particularly characteristic of some New Zealand Sauvignon blanc wines in recent years. Too little of this compound leads to an undistinguished wine, but too much gives one that is unbalanced.
This methoxypyrazine may not be the only one that is important. We have also found lesser quantities of two other methoxypyrazines, the isopropyl methoxypyrazine, and the sec-butyl methoxypyrazine. The isopropyl methoxypyrazine has a more earthy and vegetative aroma, and sometimes occurs at concentrations above the levels that we have found it to be detectable in wine (Figure 4). These low concentrations may play a role in modifying the wine aroma.
Quantification of methoxypyrazines might be purely of academic interest were it not that the growing conditions of the grape vine (soil, climate, variety, pruning, and grape maturity at harvesting) influence the berry methoxypyrazine concentration in dramatic ways.
Firstly, methoxypyrazines are much more important for some grape varieties than others. Their levels are particularly high in Sauvignon blanc, Cabernet Sauvignon, and Semillon. Methoxypyrazines contribute a wine aroma that is distinctive and important in distinguishing the wine of these grape varieties from that of other varieties.
Secondly, berry quantities at harvesting can have less than 2% of that present 6-8 weeks earlier; there is very rapid loss of methoxypyrazines as grape ripening occurs. The harvesting date dramatically influences the aroma intensity of these components in the wine, and a small amount of grapes harvested early may be blended with those harvested later to increase the methoxypyrazine aroma intensity if this is desired.
Thirdly, cool climatic conditions generate much higher methoxypyrazine levels at comparable stages of berry ripening. Cool areas, particularly those in maritime locations that moderate the diurnal temperature change (such as New Zealand, Margaret River in Western Australia and Mornington Peninsula in Victoria) can develop particularly intense methoxypyrazine aroma. This contributes to methoxypyrazine levels in New Zealand wines being generally greater than in Australian wines.
Different pruning methods have produced fruit with 8-fold concentration differences, and it seems likely that the degree of exposure of fruit to light, which in turn depends on the vine leaf canopy shape and density, will influence levels further.
At a practical level, this influence of growing conditions on berry methoxypyrazine levels challenges the production of grapes with the right flavour quality, neither too little nor too much. At the same time, it is a stimulus for research investigation, and allows our findings to be fed back into the grapegrowing industry. We have developed a technique for research into grape methoxypyrazines, and we have found that grapegrowing conditions are of paramount importance to their development. In front lies the challenge of investigating how and why this happens.