Courtesy of Wynboer

Written by Rodney Hart & Neil Jolly ARC Infruitec-Nietvoorbij, Stellenbosch
Residual urea in wines following fermentation using experimental and commercial active dried wine yeasts

Urea, a by-product of arginine metabolism by the yeast Saccharomyces cerevisiae during fermentation, is the main precursor of ethyl carbamate (EC).


Urea, a by-product of arginine metabolism by the yeast Saccharomyces cerevisiae during fermentation, is the main precursor of ethyl carbamate (EC), a potential carcinogen and natural constituent of all fermented food and beverages (Lachenmeier et al., 2010). Ethyl carbamate is formed when excess urea reacts spontaneously with ethanol. This reaction is affected by time and increased temperature, thereby, putting wines that undergo long periods of maturation and storage (e.g. red wines) at a higher risk of developing elevated levels of EC.

Although EC was not initially considered a problem in wine, Canada set a regulatory maximum limit (30 µg/L) in 1986 (Waldner & Augustyn, 2005). This was followed by a lower voluntary limit (15 µg/L) in the United States of America (USA) in 1990 (Mira de Orduña et al., 2001; Uthurry et al., 2006). Kodama et al. (1994) indicated that the urea content of wine should be below 5 mg/L and 2 mg/L in order to keep the EC below the Canadian legal and USA voluntary limit, respectively.

Subsequently, the classification of EC was updated from “possibly carcinogenic to humans” to “almost certainly carcinogenic to humans” during 2007 by the International Agency for Research on Cancer (IARC), following numerous avenues of research on EC in fermented foods and beverages (European Food Safety Authority Journal, 2007). Consequently, some member countries of the European Union suggested maximum levels of EC in fermented beverages. It can be expected that more countries will set regulatory limits for the amount of EC allowed in wines as a result of public health concerns, which will have an impact on the South African export market.

Urea formation in wine is dependent on numerous factors with the yeast strain used for fermentation having a major impact (Schehl et al., 2007). The quantity of urea formed in local wines by the various commercial wine yeast strains is not generally known. This makes it difficult to evaluate new experimental yeast strains generated by the ARC Infruitec-Nietvoorbij’s (ARC Inf-Nvb) yeast development programme. Therefore, the purpose of this study was to investigate residual urea in wines following small-scale vinification by experimental wine yeast strains compared to a selection of commercial strains used by the South African wine industry.


Commercially ripe Cabernet Sauvignon, Pinotage and Chardonnay grapes originating from the ARC’s Nietvoorbij Research Farm were used for vinification trials in 2004 and 2005. The standard ARC Inf-Nvb vinification protocol was followed. After bottling the wines were kept at 8°C until analysed. Two repetitions of each treatment were conducted.

Fourteen commercial, active dried wine yeast strains (randomly coded as A, B, D, E, F, G, H, J, L, M, N, O, P and Q) were selected to be used with three cultivars based on manufacturer’s recommendations. These commercial strains were compared to three experimental, active dried yeast strains (coded as i, ii and iii). A low urea producing yeast strain Prise de Mousse (PdM) (Lallemand) and a high urea producing yeast strain UCD522 (Lallemand), as determined by Ough et al. (1990), were used as references.

Urea elution and measurement
Calibration samples and wines were subjected to urea analysis using ion-exchange chromatography (Almy & Ough, 1989). The residual urea was determined from the calibration curve.

Statistical Procedures
Data were subjected to an appropriate factorial analysis of variance (ANOVA).


Urea levels in wine can be affected by numerous factors including the amount of nitrogen fertiliser used in the vineyard, the grape cultivar, and urea secretion by yeasts, which in turn are influenced by the nitrogen requirements of yeast during fermentation (Waldner & Augustyn, 2005; Adams & van Vuuren, 2009; Siesto, 2009). Of these, the yeast strain used for alcoholic fermentation can have the biggest impact (Dahabieh et al., 2010). However, some red grape cultivars have higher natural levels of arginine (precursor for urea) than white grape cultivars (Stines et al., 2000; Sanliang, 2002), making the red wines more likely to have higher residual urea. Similarly, in this investigation, the residual urea in the Chardonnay wines was observed to be generally lower than that of the red wines (Fig. 1).

The residual urea in Pinotage wines did not exceed 5 mg/L and will comply with Canadian legislation (Fig. 2a). In contrast to the findings of Ough et al. (1990), the roles of the high and low urea producer reference yeasts were reversed, with the PdM produced wines having a higher residual urea (2.39 mg/L) compared to that of the UCD552 produced wine (1.53 mg/L). Potentially, only wines produced by the commercial yeast strains B, D, F and N; and two experimental yeast strains i and iii will not comply with the voluntary USA EC limit if stored incorrectly. However, if these wines were to be stored at low temperatures (<20°C) (Waldner & Augustyn, 2005), no problems regarding the formation of unacceptably high EC levels are foreseen.

Cabernet Sauvignon
All the Cabernet Sauvignon wines complied with the Canadian regulation with the highest wine showing a residual urea of 2.66 mg/L (Fig. 2b). As was observed in the Pinotage wines, PdM produced wines had a higher residual urea (2.52 mg/L) compared to that of the UCD552 produced wine (1.40 mg/L). Only one commercial yeast, strain E and the experimental yeast strains i and ii produced wine which may lead to EC concentrations not complying with the voluntary USA limit.

Residual urea in the Chardonnay wines showed the same trends as the research done by Ough et al. (1990), with the UCD552 produced wine having a higher residual urea (1.78 mg/L) compared to the PdM produced wine (1.32 mg/L) (Fig. 2c). The commercial yeast, strain A and the experimental yeast, strain i were the only yeast strains evaluated that produced wines with more residual urea (>2 mg/L) than the high urea producer reference yeast strain UCD522. No problems regarding the formation of unacceptable EC levels are foreseen for the remaining wines.


Differences in residual urea observed in wine produced by different yeast strains support the findings of others with regard to the role of the yeast strain in urea formation. This study also confirmed the role of the grape cultivar and it cannot be assumed that a particular yeast strain will behave the same in red and white fermentations. The choice of yeast strain used for red and white grape vinification with regard to urea production is therefore of cardinal importance. It is important that a low urea producing yeast strain is used, especially if the wines are destined to be exported to Canada or the USA. The yeast manufacturer can be approached for assistance in the choice of yeast strain. None of the three experimental yeast strains evaluated in this study produced wines with low enough urea concentrations not to be at risk of exceeding the voluntary USA EC limit. However, all the experimental yeast strains evaluated have urea production within the range of local and imported commercial yeast strains currently available in South Africa.


Urea, a by-product of arginine metabolism by the yeast Saccharomyces cerevisiae during fermentation, is the main precursor of ethyl carbamate (EC), a potential carcinogen. Although EC is not considered a problem in young wines, urea can be measured in a young wine to give an indication of potential EC levels in wine after ageing or incorrect storage. This study, conducted over the 2004 and 2005 vintages with wines produced by experimental and commercial yeast strains, showed that residual urea in Chardonnay wines were generally lower than that observed in Pinotage and Cabernet Sauvignon wines, respectively. The roles of the high and low urea producer reference yeasts were reversed in red wines, with the Prise de Mousse (PdM) wines having more residual urea compared to that of the UCD552 produced wine. Furthermore, residual urea in all wines was <5 mg/L and complies with Canadian legislation. Some wines may not comply with the voluntary, but stricter, USA EC limit if the wines are to be stored under conditions allowing EC to form. None of the three experimental yeast strains evaluated in this trial produced wines within the voluntary USA EC limit. However, the experimental yeast strains all had urea production within the range of existing local and imported commercial yeast strains available in South Africa.


The ARC Infruitec-Nietvoorbij for infrastructure and financial support, and Winetech for financial support. Thanks to Frikkie Calitz and Mardé Booyse for the statistical analyses of the data.


Adams, C. & van Vuuren, H.J.J., 2009.  Effect of timing of diammonium phosphate addition to fermenting grape must on the production of ethyl carbamate in wine.  Am. J. Enol. Vitic. 61, 125 – 129.

Almy, J. & Ough, C.S., 1989. An urea analysis for wines. J. Agric. Food Chem. 37, 968.

Dahabieh, M.S., Husnik, J.I. & van Vuuren H.J.J., 2010. Functional enhancement of Sake yeast strains to minimize the production of ethyl carbamate in Sake wine. J. Appl. Microbiol. 109:3, 963 – 973.

European Food Safety Authority (EFSA), 2007. Opinion of the Scientific Panel on Contaminants in the Food chain on a request from the European Commission on ethyl carbamate and hydrocyanic acid in food and beverages, The EFSA J. 551, 1 – 44. [WWW document]. URL online: [Accessed on 24 February 2011].

Kodama, S., Suzuki, T., Fujinawa, S., De La Teja, P. & Yotsuzuka, F., 1994. Urea contribution to ethyl carbamate formation in commercial wines during storage. Am. J. Enol. Vitic. 45, 17 – 24.

Lachenmeier, D.W.,  Lima, M.C.P., Nóbrega, I.C.C., Pereira, J.A.P., Kerr-Corrêa, F., Kanteres, F. & Rehm, J., 2010. Cancer risk assessment of ethyl carbamate in alcoholic beverages from Brazil with special consideration to the spirits cachaça and tiquira. BMC Cancer. 10, 266.

Mira de Orduña, R., Patchett, M.L., Liu, S. & Pilone, G.J., 2001. Growth and arginine metabolism of the wine lactic acid bacteria Lactobacillus buchneri and Oenococcus oeni at different pH values and arginine concentrations. Appl. Environ. Microbiol. 67, 1657 – 1662.

Ough, C.S., Stevens, D., Sendovski, T., Huang, Z. & An, D., 1990. Factors contributing to urea formation in commercially fermented wines. Am. J. Enol. Vitic. 41, 68 – 73.

Sanliang, G., Hakim, A., Du, G., Fugelsang, K., Anthony, B., Trombella, B. & Hodson, G., 2002. Effect of Arginine Addition to Must on Wine Urea, Ethyl Carbamate, and Ethyl Carbamate Potential in Cabernet Sauvignon, Syrah, Chardonnay, and Sauvignon blanc Grapes Viticulture and Enology Research Center California State University. [WWW document]. URL online: [Accessed on 19 April 2011].

Schehl, B., Senn, T., Lachenmeier, D.W., Rodicio, R. & Heinisch, J.J., 2007. Contribution of the fermenting yeast strain to ethyl carbamate generation in stone fruit spirits. Appl. Microbiol. Biotechnol. 74, 843 – 850.

Siesto, G., 2009. Traceability of wine Saccharomyces cerevisiae strains by molecular and biochemical parameters. 14th Workshop on the Developments in the Italian PhD Research on Food Science Technology and Biotechnology – University of Sassari Oristano, September 16 – 18, 2009 [WWW document]. URL online: [Accessed on 07 February 2011].

Stines, A.P., Grubb, J., Gockowiak, H., Hensche, P.A., Høj, P.B. & van Heeswijck, R., 2000. Proline and arginine accumulation in developing berries of Vitis vinifera L. in Australian vineyards: Influence of vine cultivar, berry maturity and tissue type. Aust. J. Grape Wine Res. 6, 150 – 58.

Uthurry, C.A., Suárez Lepe, J.A., Lombardero, J. & García del Hierro, J.R., 2006. Ethyl carbamate production by selected yeast strains and lactic acid bacteria in red wine. Food Chem. 94, 262 – 270.

Waldner, M., & Augustyn, O., 2005. Ethyl carbamate in South African wine. Wynboer. [WWW document]. URL online: [Accessed on 07 February 2011].