AWRI REPORT

Getting proactive about protein

By Jacqui McRae, Agnieszka Mierczynska-Vasilev, Markus Herderich and Dan Johnson, The Australian Research Institute, PO Box 197, Glen Osmond, South Australia 5064

Heat tests and bentonite fining are common practices across the world. Recent research at the AWRI has identified a more convenient and reliable format for the heat test, which is being adopted across winery labs. Promising results are also being seen for a number of possible alternatives to bentonite for protein stabilisation.

INTRODUCTION This prompted AWRI researchers comparing the amount of haze formed Grapes contain proteins that to revisit the heat test in some depth, following different periods of heating persist through winemaking and, looking at the effect of different heat suggested that heat test samples if not removed, can cause hazes in test conditions on the amount of haze should be heated for a minimum of white, and sparkling . Most formed, the predicted bentonite dose two hours for consistent results. winemakers use bentonite fining to and the relationship between the haze Predicted bentonite dose was remove protein and prevent haze formed in the heat test and the haze usually greater after longer heating formation, and use a heat test to formed after longer-term storage. and cooling times (e.g. after the determine the required bentonite original 24-hour test compared to a COOLING IS AS IMPORTANT AS dose. Recent research on wine five-hour test) although the cooling HEATING proteins at the AWRI has closely temperature did not change the test investigated the chemistry behind The first key finding from this work outcome, providing the cooling time the heat test, and developed new was that cooling time and temperature was at least three hours. recommendations to maximise were very important in achieving TESTING LONGER-TERM STABILITY both convenience and reliability. In consistent heat test results. This is addition, possible alternatives to because protein haze production Preventing haze formation is always a bentonite for protein stabilisation following heating is a three-stage balance between over-fining wine with continue to be investigated, with some process (Van Sluyter 2015): bentonite, which can strip colour and interesting results. • Proteins in the wines are unfolded aroma, and under-fining wine, which from their normal configuration due increases the risk of the wine becoming REVISITING THE HEAT TEST to the high temperature. hazy. Given that different combinations Heat tests are widely used to • The unfolded proteins start to of heating and cooling time resulted in determine the amount of bentonite interact with each other and with different predicted bentonite doses, this required to remove enough protein other wine components to form meant either that the lower predicted from a wine to prevent a haze from aggregates. developing. The test is performed • As the wine is cooled, the aggregates AT A GLANCE by heating a wine sample, cooling it grow larger and interact with other down, and measuring the turbidity aggregates to form a visible haze. • Proteins in wine have potential to (cloudiness) before and after heating. For the heat test to work, the wine cause haze if not removed. This The original heat test method (Pocock samples must be heated long enough might be considered desirable or et al. 1973) was designed to produce for the proteins to unfold and start undesirable, depending on wine style. the greatest amount of haze in a wine to aggregate and then be cooled sample in the shortest amount of fast enough and long enough for the • Bentonite fining is the most time and included six hours of heating aggregates to form a visible haze. common treatment used to prevent at 80°C and 16 hours of cooling at Because of this, wine samples that are protein haze, with a heat test used 4°C followed by a further two hours not cooled immediately from 80°C to to determine the bentonite dose. at room temperature. This test was ≤ 20°C (or are not cooled for a long • AWRI researchers have confirmed reliable but it involved a 24-hour enough time) will produce less haze the reliability of a shorter heat test, turnaround time and may have over- than samples that are removed from requiring only two hours of heating predicted the amount of bentonite heat and cooled for longer. The results and three hours of cooling. required to stabilise a wine. Since this suggest that the cooling period in the test was developed, many variations heat test should be a minimum of • Enzymes, pasteurisation and of heating and cooling times and three hours for consistent results. magnetic nanoparticles are all temperatures have been used by wine Heating time was also found to be showing favourable results as laboratories around Australia and important, with different amounts of potential alternatives to bentonite the world, with reports received of haze formed when wines were heated for ensuring haze-free wines. variable results. for different times. Experiments

V34N2 WINE & JOURNAL AUTUMN 2019 www.winetitles.com.au 33 AWRI REPORT

doses were under-fining the wine or the Adding AGP enzymes to juice just the particles, when an external magnet higher doses were over-fining. It was prior to flash pasteurisation and then is applied. Testing of wines following a therefore important to conduct longer- fermentation can produce heat-stable trial of the MNP treatment found that term storage trials to test which version wine without needing to add bentonite haze-forming proteins were removed, of the heat test best predicted the (Marangon 2012). AGP enzymes have even from wines with very high protein minimum dose of bentonite required for been approved by FSANZ for use in content, while other components in long-term stability. and are also legal to be the wines, such as phenolics and metal Selected wines were fined at used in wines destined for major export content, were unaffected. In addition, bentonite doses predicted by two markets. Cost analyses indicate that this since the adsorption of proteins heat tests, one with six hours’ heating protein removal treatment costs less per onto the nanoparticles is a reversible and 18 hours’ cooling and the other batch of wine than bentonite. Several process, there is potential for the with two hours’ heating and three companies are now developing AGP particles to be regenerated and reused, hours’ cooling. The wines were then enzymes for commercial application. saving on waste. stored for 12 months at 17°C and 28°C. Pasteurisation Wines fined using both versions of WHERE NEXT WITH PROTEIN the test were clear and bright after While AGP enzymes are not RESEARCH? 12 months of storage, suggesting yet commercially available, juice that the shorter (and therefore more flash pasteurisation alone can Recent developments in convenient) version of the text could be still substantially reduce protein understanding haze formation have recommended to industry. concentration and the amount of enabled better predictions of haze This shorter version (two hours’ bentonite needed to heat stabilise and development of new strategies heating at 80°C, three hours’ cooling at white wines. Heating juice changes the to prevent it. Now that the work 20°C) has now been promoted widely structures of the proteins and makes on the heat test has concluded, them more likely to stick together to to industry via technical publications protein research efforts at the AWRI form aggregates, much like the effect and a well-attended webinar. It will will continue to explore and refine of the heating step in the heat test. be included in the next issue of the alternatives to bentonite, to provide This means that after fermentation, Australian wine industry’s key analysis winemakers with sustainable, text book and will be discussed at the the aggregated proteins are more likely to drop out of solution along with the economical and efficient protein upcoming Australian Wine Industry stabilisation options. Technical Conference in July 2019. yeast lees. Heating juice in a controlled REFERENCES BENTONITE ALTERNATIVES environment like a flash pasteuriser or Marangon, M.; Van Sluyter, S.C.; Robinson, While bentonite is an effective heat exchanger for one to two minutes has been shown not to have adverse E.M.C.; Muhlack, R.A.; Holt, H.E.; Haynes, P.A.; way to remove haze-forming proteins Godden, P.W.; Smith, P.A. and Waters, E.J. from wine, it has some drawbacks. effects on the sensory properties of (2012) Degradation of haze proteins It is not selective in its action, as it wine (Marangon 2012). In that trial, by Aspergillopepsin I and II during juice flash removes all proteins, not just those that the concentration of protein in a pasteurization. Food Chem. 135(3):1157-11652. contribute to a haze. It also increases wine was reduced Mierczynska-Vasilev, A; Boyer, P.; Vasilev, K. and the time wines spend in tank, can by half after juice was heated for one Smith, P.A. (2017) A novel technology for the rapid, lead to loss of volume and quality and minute, substantially reducing the selective, magnetic removal of pathogenesis-related proteins from wines. Food Chem. 232:508-514. creates waste disposal challenges. These concentration of bentonite required Mierczynska-Vasilev, A.; Mierczynski, P.; issues and associated costs have led to stabilise the wine. In smaller-scale Maniukiewicz, W.; Visalakshan, R.M.; Vasilev, trials, heating juice for two minutes researchers around the world, including K. and Smith, P.A. (2019) Magnetic separation at the AWRI, to conduct research on removed almost all the protein in a technology: Functional group efficiency in the possible alternatives. Semillon wine and a Sauvignon Blanc removal of haze-forming proteins from wines. Enzymes wine. Further research is under way Food Chem. 275:154-160. to investigate the impact of two McRae, J.M.; Barricklow, V.; Pocock, K. and The most promising bentonite minutes of heating juice on the sensory Smith, P.A. (2018a) Predicting protein haze alternative to date is flash pasteurisation properties of wine. formation in white wines. Aust. J. Grape Wine Res. of grape juice in the presence of DOI: 10.1111/ajgw.12354. Magnetic nanoparticles aspergillopepsin (AGP) enzymes. In McRae, J.M.; Schulkin, A.; Dambergs, R.G. general, the structures of wine haze- Plasma polymer coated magnetic and Smith, P.A. (2018b) Effect of white wine forming proteins make them inherently nanoparticles (MNPs) – nanometre- composition on protein haze potential. Aust. J. resistant to enzymes, but once they scale particles that can be moved Grape Wine Res. DOI: 10.1111/ajgw.12346. are heated to 75°C, these structures around using external magnets – are Pocock, K.F. and Rankine, B.C. (1973) Heat test break down, leaving them vulnerable. the basis of another novel method for for detecting protein instability in wine. Aust. Wine Brew. Spirit Rev. 91:42-43. Such high temperatures also tend to removing haze proteins from wines break down most enzymes; however, (Mierczynska-Vasilev et al. 2017, Van Sluyter, S.C.; McRae, J.M.; Falconer, R.J., Smith, P.A., Bacic, A., Waters, E.J., Marangon, unusually, AGP enzymes are active 2019). Proteins are adsorbed onto the M. 2015. Wine Protein Haze: Mechanisms of at the same temperatures that make magnetic particles and can then be Formation and Advances in Prevention. J. Agric. haze-forming proteins vulnerable. removed from the wines, along with Food Chem. 63(16): 4020-4030. WVJ

34 www.winetitles.com.au WINE & VITICULTURE JOURNAL AUTUMN 2019 V34N2