The latest on mouse, the wine taint of our times

Mouse taint (mousiness) is a wine fault that I’d not experienced until a few years ago. Now it seems to be everywhere, but that’s probably because I have a strong interest in natural wine, and it’s the increase in people working without added sulfites that has led to its rise. And also there’s the question of awareness: once you’ve spotted a specific fault, you seem to see it more frequently.

Just how common is it? It’s hard to tell. The only survey I’m aware of is a graph from a webinar by Excell laboratories in France showing a graph of enquiries they’ve received about analysis for mouse. These rose dramatically between April 2018 and February 2019. This could reflect a rise in awareness of this fault as much as an increasing incidence. As we’ll see, it’s a complicated fault and coming up with any sort of global figure would be hard.

In January 2019, I caught up with two of the leading French experts on mouse, aka goût de souris: Dr Nicolas Richard (pictured above) who works with Inter Rhône, and Dr Bertrand Chatelet, who works for Sicarex in Beaujolais. We discussed the latest research on the topic.

Three molecules have been identified as contributing to the taint. They are part of a group of compounds called tetrahydropyridines. They are 2-ethyltetrahydropyridine (ETHP)

2-acetyltetrahydropyridine (ATHP) and 2-acetylpyrroline (APY). These compounds are made by lactic acid bacteria (LAB) and also some strains of Brettanomyces. Brettanomyces and LAB can produce mousy taints independently. It also seems that while Brettanomyces can theoretically produce APY, in practice it just produces ETHP and ATHP.

Acetaldehyde (the oxidation product of alcohol) is involved in the generation of these compounds. Acylation reactions convert L-lysine and L-ornithine into the tetrahydropyridines.

These compounds exist in an equilibrium in which they go from volatile to non-volatile depending on pH. This is a critical point. At low pH they are not volatile and so you can’t smell them. But when your saliva mixes with them, the pH of the saliva is higher than that of wine, and so after a while you suddenly detect them retronasally. The aromas in question are mouse cage, popcorn, sweet corn, rice cakes, crackers, grilled bread, salami, charcuterie and dirty socks.

But, points out Richard, the perception of mouse taint is complicated. There are variations in mouth pH, variations in sensitivity, and the wine varies with aeration. ‘The first problem comes from the mouth pH,’ he says. ‘The variability is high. The variation for each person can be as much as one pH unit from day to day. This alters the perception of the mousy flavour.’

Wine pH is between 2.8 and 4, with most white wines at about pH 3.3 and most reds 3.6 on average, but this varies widely with style. Mouth pH is close to 7 and varies between 5.76 and 7.96 among individuals. With each individual, though, their mouth pH will vary by an average of 0.91, depending on the time of the day, food consumed, and physiological state.

Aside from this, is there different sensitivity among people in their ability to perceive mouse compounds? ‘Yes,’ says Richard, ‘and there are at least three molecules. You can be anosmic for one, and very sensitive to another, and have medium sensitivity for the third. There are as many combinations as there are humans.’ So the sensitivity to mouse taint varies for each of the compounds, and the sensitivity to one is independent to the sensitivity to another.

Bertrand Chatelet, Sicarex, Beaujolais

Chatelet cites an unpublished study by University of Bordeaux researcher Sophie Tempère, who studied the ability of 23 people to detect APY, one of these compounds. She found detection thresholds that differed 1000-fold between the least and most sensitive subjects, suggesting that some simply couldn’t smell it at all. This explains the assertion that some people don’t get mouse, while others do, although this is far from clear cut because of the mass of variables involved. A figure of 30% of winemakers unable to detect mousiness has been cited widely, but its source is unknown: it appears in the Oxford Companion to Wine, but there is no corroboration of this number in the scientific literature.

There is also a temporal effect. ‘The wine will vary from minute to minute,’ says Richard. ‘You can wait for 30 minutes or one hour after opening the wine, and the mousiness will appear and reveal itself.’ If you are opening a wine that you think may have a trace of mouse, or where you’ve experienced mouse in the past, serve it cold and drink it fast, and don’t decant it!

There is currently no accepted chemical analysis for mouse taint, but there are three ways to look at it that don’t involve putting the wine in your mouth and waiting for a pH change.  

  • The first is the ‘palm and sniff’ technique, first proposed in the 1950s. Here, you put the wine on your skin and smell to see whether there is any mouse. Skin changes the pH of the wine so any mouse present should reveal itself.
  • The second is to use a paper strip that has been soaked in sodium hydroxide and dip it in the wine and then smell it. The strip raises the wine pH.
  • The final one is to change the pH of wine by pouring it into a glass containing a measure of bicarbonate of soda (5 g/l dose).

As an example, APY is detectable at 55 micrograms a litre at pH 3.2, but just 9 micrograms a litre at pH 5. A study by Inter-Rhône tested the pH level at which people are best at identifying the presence of APY. ‘The experiments showed that pH is 5 is the best level to have the most powerful discrimination and the best consensus between subjects,’ says Richard. ‘No one has any results for ATHP and ETHP,’ he adds.

As well as production by lactic acid bacteria and Brettanomyces, there is also the suggestion that the Maillard reaction might be involved in mousiness in some situations. It contributes to the flavours of bread, biscuits and tortillas by making teytrohydropridines, among other compounds. But this is an area of some uncertainty. Baked goods are an example of where we actually quite like the contribution of these compounds.

The risk factors for mouse are several, but the biggest is winemaking in the absence of added sulfites. High pH, wild ferments and potentially also the use of whole bunch are also cited by Chatelet. Richard says that because of the rise of natural winemaking, mouse has become quite a problem. He thinks that part of the reason for this might be the increased use of semi-carbonic maceration, which is very popular in Beaujolais. ‘It is a particularly high risk for mouse, because there is an interface the liquid and solid, and there is aeration,’ he says. ‘Natural winemakers do a lot of semi-carbonic, and they have a lot of mousiness, so I made a correlation. But I have no experimental evidence.’

Redox potential also makes a difference, and mouse can reappear with each exposure to air. It’s quite common to open an affected bottle and for the first half to be fine, and the second half to be mousy after it has been exposed to air. Anecdotally, winemakers working naturally often say time is critical. Sometimes they have mousy wines, even bottled ones, and they just wait until the mouse goes away.

If I was a winegrower, how could I avoid mouse? The simplest way is to use some sulfites during the fermentation, but for many natural producers this is something they want to avoid. Making your wine an uninviting habitat Brettanomyces is a good idea generally. Working at low temperature and pH is important, as is limiting dissolved oxygen.

So that’s it: the latest on this most annoying of wine faults, which is still a bit of a mystery.