Simultaneous assay of mousy off-flavor markers

The compounds described as being responsible for mousy off-flavor in wine are 2-acetylpyrroline (APY), 2-acetyltetrahydropyridine (ATHP) and 2-ethyltetrahydropyridine (ETHP). However, assay of these compounds is difficult due to their physical-chemical characteristics. The aim of this study was to develop a simple and effective method for simultaneously determining the concentration of these three N-heterocyclic compounds in wine. This new method is a decisive tool for better defining this spoilage phenomenon in wine and guarding against it.


Introduction
Mousy off-flavor is a wine fault of microbial origin.It is described as particularly unpleasant, reminiscent of urine or a mouse cage, but also and most often grilled foods such as popcorn, crackers and bread crust, rice, or sausage skin 1 2 3 .While this type of spoilage may have been misnamed in olfactory terms, it appears more clearly on the palate.The N-heterocyclic compounds involved are not very volatile at wine pH 4 .The pH in the mouth is higher than in wine, which explains why this fault is perceived more strongly on the palate via retro-olfaction.Described by Peynaud and Domercq 1 as early as 1956, this fault was thought to be linked to the presence of Brettanomyces-type yeasts, according to Australian professionals 2 4 .However, before the 2010s, mousy off-flavor was an extremely rare fault.It has become more frequent in recent years, associated with an increase in wine pH, as well as with winemaking practices that tend to significantly reduce the use of sulfur dioxide (SO 2 ).Until now, quantitative data for mousy off-flavor markers has been limited due to the analytical challenges associated with the nature of these compounds and their low concentrations, in the order of µg.L -1 .Concentrations previously reported in wine range from 0.7 to 106 µg.L -1 for ATHP 5 ; for APY, a concentration of 7.8 µg.L -1 has been reported by Snowdon et al. 2 , while ETHP concentrations found in wine can exceed 150 µg.L -1 4 .There are currently no detection thresholds for these compounds (though they would be very useful), as measurements have been carried out using different sensory protocols and in different matrices 3 , making comparison difficult.With the aim of filling this gap and understanding the parameters influencing the appearance of mousy off-flavor, this study has developed a simple and effective method for simultaneous assay in wine of the three N-heterocyclic compounds involved.Hence, a technique has been developed using Stir Bar Sorptive Extraction (SBSE, Twister®) followed by gas chromatography-mass spectrometry (GC-MS) analysis.The SBSE technique is already used for multi-fault assay of wine 6 .
Lastly, wines suspected of taint with mousy off-flavor may contain APY or ATHP, or both of these compounds, but sometimes neither (presence below the limit of detection).As for ETHP, it is present in almost all wines produced without SO 2 or with limited sulfite addition.Further analyses will be needed to confirm these initial results with more data and a more diverse range of samples.

Conclusion
A relatively simple method, accessible to laboratories with the necessary equipment, has been developed to determine the concentrations of the N-heterocyclic compounds responsible for mousy off-flavor.It consists in the use of SBSE or Twister® extraction followed by gas chromatography with mass spectrometry detection.This method makes it possible to check tasting results, to confirm the presence of the olfactory fault or, on the contrary, eliminate doubts when perception of the fault is difficult on tasting.In addition, the mechanism for generation of mousy off-flavor has yet to be fully elucidated.Several hypotheses point to very high microbial diversity 8 or a modification of the wine's redox balance.This new method is expected to contribute to a better understanding of this all-too-frequent fault. To 10 mL of the sample at pH 10.5, 2 g of NaCl is added.Once the NaCl has completely dissolved, the internal standard (IBMP-d 3 ) is added to the vial, along with a PDMS Twister® (Gerstel, 20 mm × 1 mm; length × phase thickness).After extraction with stirring at 600 rpm for 1 hour, the Twister® is removed from the sample and desorbed into the GC-MS for analysis (Figure 1).In this method, the imine and enamine forms of ATHP are detected, unlike the enamine forms of APY and ETHP.Either the equilibrium is very strongly shifted towards the imine forms, or the extractability of the enamine forms by PDMS is low due to their chemical characteristics.The ATHP concentration is thus determined by summing the results for A-3,4,5,6-THP and A-1,4,5,6-THP.The limits of detection (LOD) and quantification (LOQ) of APY, ATHP and ETHP for white, rosé and red wines are given in Table 1.

Analysis of wine samples
The optimized method was applied to the analysis of commercial wine samples from different origins and produced using a variety of winemaking practices.This study included analysis of 6 wines produced with "conventional" use of SO 2 and 68 wines produced without addition of SO 2 .Classification into the two categories, conventional or limited sulfite addition, was based on indications concerning the use of SO 2 on the label or from the producer.It has been demonstrated that almost all wines produced without SO 2 or with limited sulfite addition contain ETHP, whereas this compound is not quantifiable in wines where sulfites have been added.On the other hand, APY and ATHP were detected only in wines produced without SO 2 or with limited sulfite addition.In all, 11 wines contained APY and 39 wines contained ATHP (Figure 2).Concentrations were always below 7.7 µg.L -1 for APY, 54.5 µg.L -1 for ATHP and ranged from 14.0 µg.L -1 to 120.5 µg.L -1 for ETHP.The results highlight several points.First, the conventional use of SO 2 could limit the appearance of the heterocyclic compounds responsible for mousy off-flavor.Secondly, the presence of APY is generally associated with the presence of the other two heterocyclic compounds.

FIGURE 2 .
FIGURE 2. Distribution of the mousy off-flavor in the 74 wines analyzed (adapted from Kiyomichi et al. 7 ).
a LOD: limit of detection, LOQ: limit of quantification.