OCCURRENCE, CHARACTERISTICS, AND EFFECTS ON BEER FLAVOUR
by R. B. Gilliland, B.A., B.Sc, F.R.I.C.
(Arthur Guinness Son & Co. (Dublin) Ltd., St. James’s Gate, Dublin)
Received 21st January, 1961
The genus Brettanomyces includes small slow-growing yeasts which have so far been isolated exclusively from beer or wine. Records of their occurrence and characteristics are surveyed. While they can produce a desirable and characteristic flavour in some high-gravity beers, they may cause spoilage of beers of normal gravities. Forty-four strains of Brettanomyces were isolated from naturally conditioned beers with off-flavours. These strains were classified into three species: Brett. bruxellensis, Brett. schanderlil, and an unidentified species. Each of these species was shown to be capable of producing off-flavours when added in pure culture to beer; Brett. bruxellensis was the most deleterious to flavour.
Brettanomyces spp. have long been known as important factors in the production of condition and flavour in certain high-gravity beers and in Iambic beer. They have also been mentioned as possible spoilage organisms in beer but have generally been considered as of little importance in this connection. Thus Wiles27, 28 stated that Brettanomyces had only once been reported as a spoilage organism of beer, and that although Brett. Bruxellensis was sometimes detected it was not of common occurrence and was unlikely to be a cause of spoilage. On the other hand, Shimwell21, 22 found a Brettanomyces to be the cause of a fret in beer and suggested that it might be widely encountered as a spoilage organism. It has been stated that Brettanomyces can cause a mawkish “nose” and turbidity in beers of original gravity less than 1060.3,4 Recently Brettanomyces spp. were isolated on a number of occasions from spoiled wines and interest has revived in their ecology and characteristics. In the following paper the occurrence of Brettanomyces as spoilage organisms in naturally conditioned beer is recorded and the characteristics of 44 strains isolated from beer are studied. In view of the fact that the genus Brettanomyces has very rarely been discussed in English brewing literature, a survey of references to its occurrence and properties is included.
Occurrence of Brettanomyces.—The earliest reference to Brettanomyces, or secondary yeasts, was in a patent for the use of these organisms for the preparation of English beers which was taken out by Claussen in 1903.5 In a paper to the Institute of Brewing in 1904, Claussen6 described the importance of Brettanomyces for secondary fermentation and the production of the characteristic flavour of English stock beers. Claussen did not give a detailed description of these organisms, which he included in the genus Torula. Shortly after this, Seyffert20 of the Kalinkin Brewery in St. Petersberg announced that he had isolated a “Torula” in 1889 from English beer which produced the typical “English” taste in lager beer, and which was similar in other respects to Claussen’s Brettanomyces. In 1899 J. W. Tullo, in the Chemist’s Laboratory, Arthur Guinness Son & Co. Ltd., Dublin, had already isolated two types of “secondary yeast” from Irish stout, and in an unpublished report described the characteristics of these yeasts and their importance in secondary fermentation. At this time the “secondary yeasts” were important constituents of the flora of all stock beers, and in particular of those beers, designed for the export trade, which underwent long maturation in the brewery. These export beers depended indeed on the “secondary yeasts,” not only for their characteristic flavour but also for the production of condition in bottle by means of their ability to ferment higher polysaccharides which the “primary yeast” could not ferment.
In 1908 Shionning,10 who worked with Claussen’s cultures, gave a detailed description of two species which he called “Torula A” and “Torula B.” In 1920, Kufferath12 isolated two strains from Lambic beer and investigated their characteristics. This study was continued by Kufferath & van Laer13 who established the genus and named two species.
In 1940, Custers7 published a monograph on Brettanomyces which reported the first systematic investigation of the genus. He collected 17 strains including, as well as those of Claussen and of Kufferath & van Laer, a strain isolated by Shimwell, and strains newly isolated from English ale and stout, and from Iambic beer. Custers considered the genus to be almost exclusively of brewery origin, although he identified as a Brettanomyces a strain isolated by Krumbholz & Tauschanoff11 from a French wine fermentation, and called by them Mycotorula intermedia. Etchells & Bell9 isolated many strains from commercially brined cucumber fermentations which they classified in two newly described species of Brettanomyces; these species were, however, transferred to the genus Torulopsis by Lodder & Kreger-van Rij.14 Schanderl,17 and Schanderl & Draczynski,18 described the isolation of Brettanomyces from bottled wines and they also pointed out that the organism isolated by Osterwalder16 from apple wine and described by him as Monilia vini was probably a species of Brettanomyces. Barret, Bidan & André2 found Brettanomyces in wines from Arbois. Galzy & Rioux10 found them in the “fleur” pellicles of wine from the Midi region of France. Domercq8 and Peynaud & Domercq16 described the characteristics of Brettanomyces isolated from wines and musts of the Girond area and named two new species. Verona26 found Brettanomyces in Italian wine, and it has also been isolated from Brazilian wines.32 Van der Walt & van Kirken23, 24 found Brettanomyces to be the most prevalent genus in South African wines showing turbidity, and in an important series of papers they described studies on the characteristics of the strains isolated and made recommendations for rationalizing the nomenclature of some species. Very recently the occurrence of Brett. bruxellensis in pilsener beer has been recorded by Windisch30 and in “weissbier” by Zaake.31 Zaake also found Brett. clausenii in pilsener, in export beer and in Belgian beer.31
Properties of Brettanomyces.—The cells of Brettanomyces are small, oval or round, frequently with a pointed end (ogive). In some strains elongated cells and branched chains (“trees”) are frequently seen. On suitable media most strains show a tendency to form pseudomycelium. In malt extract Brettanomyces spp. grow slowly but give a high final attenuation1 and a characteristic aroma. Under aerobic conditions they produce considerable acid; this may be the reason why malt-agar cultures have a short life. In agar-streak cultures they give yellow or brown raised growth. The genus has always been considered anascosporogenous, but van der Walt & van Kirken28 have recently reported sporulation in a number of species which would necessitate the placing of the genus in another family. The fermentative abilities of the various strains in the species as described by Lodder & Kreger-van Rij14 and by van der Walt & van Kirken24 are given in Table I.
One of the most interesting characteristics of the metabolism of Brettanomyces is that young aerobic cultures exhibit a negative Pasteur effect. That is, they have a very much stronger fermentative ability under aerobic conditions than under anaerobic conditions, in contrast to other yeasts.7 Another important characteristic of the genus is that secondary products of fermentation accumulate to a much greater extent than is the case with Saccharomyces. Ethyl acetate, glycerol, acetic acid, succinic acid and 2,3-butane-diol have been estimated by Peynaud & Domercq16 in the completed fermentations with Brettanomyces. These authors consider that ethyl acetate and acetamide are the two main organoleptic products of Brettanomyces fermentations, but that some strains may also produce a butyric smell.
Brettanomyces spp. appear to be homogeneous in their requirements for biotin and thiamin as growth factors.24 Some species also require one or more of the growth factors pyridoxin, nicotinic acid or inositol. No species was found to require pantothenate, whereas this requirement is common amongst Saccharomyces. All strains of Brettanomyces tested could grow in media containing 1000 p.p.m. of actidione, whereas 1 p.p.m. is sufficient to prevent the growth of Saccharomyces cerevisiae. It is interesting that the strain of the organism named Brett. sphaericus by Etchells & Bell9 but transferred to the genus Torulopsis by Lodder & Kreger-van Rij14 could not withstand 4 p.p.m. of actidione. Brettanomyces share actidione resistance with some strains of Pichia, Candida, Lipomyces, etc. A wort-agar medium containing 10 p.p.m. actidione was found most useful for isolating Brettanomyces from samples containing many Saccharomyces.
Many sediments of old beers—particularly those which had abnormal flavours—were examined microscopically, and in a proportion of these the typical cells of Brettanomyces were observed. The sediments of the beers containing Brettanomyces were plated out on wort agar containing 10 p.p.m. of actidione and incubated at 25° C. for 7 days. Some of the resultant colonies were examined microscopically and, if it was confirmed that they were like Brettanomyces in appearance, then four colonies from each plate were seeded into sterile wort. In this way four strains were isolated from each of 11 bottles. These 44 strains were first tested to confirm that they were Brettanomyces by microscopical appearance, acid production, lack of ability to sporulate, and aroma production in malt extract. They were then further classified by seeding them into McCartney bottles each containing 10 ml. of a yeast growth medium with glucose, galactose, lactose, sucrose, maltose or raffinose as sole carbon source. These bottles were provided with Durham tubes so that gas production could be observed. The screw caps were left slightly loose so that pressure should not build up in the bottles. Where fermentation took place gas was generally observed in the Durham tubes in 3-5 days. As an additional and more reliable method of determining whether fermentation had taken place, the specific gravities of the solutions were determined at seeding and after 7 and 14 days by a falling drop method.29
The 44 cultures could be divided into three groups:—
Group 1.—21 strains fermented glucose, sucrose and maltose only. These strains agreed with the published descriptions of Brett. bruxettensis Kufferath et van Laer and were similar to authentic strains of this species. Brett, bruxellensis has been isolated from Lambic beer,12 from Cork porter,21 from French grape must11 and from Brazilian wine fermentations.32
Group 2.—11 strains fermented glucose and galactose only. These strains agreed with published descriptions of Brett, schanderlii Peynaud & Domercq and were similar to an authentic strain of this species. Brett. schanderlii has been found in French16 and South African24 wine fermentations. It has not previously been reported as occurring in beer.
Group 3.—12 strains fermented glucose, galactose, sucrose and lactose. This combination of fermentative abilities does not correspond with any established species. These strains are most like Brett. anomalus but differ from this species in that they ferment sucrose rapidly. A further study is being made of these strains and, if the differences are sufficient to establish the strains in this group as a separate species, this will be reported in a further communication. For the present these strains will be described as Brettanomyces sp. 1.
It seems strange that so many strains unable to ferment maltose (Groups 2 and 3) should be found in beer, but Custers isolated Brett, anomalus, which also is unable to ferment maltose, from English beer. It was because of this peculiar circumstance that Custers gave this species the name “anomalus.”
The distribution of the individual strains from the 11 bottles is given in Table II, from which it will be seen that from 5 of the bottles only one species of Brettanomyces was recovered; in all the other bottles two or more species were found. The infection therefore regularly occurs as a mixture of species of Brettanomyces.
Tasting experiments were carried out with 36 of the pure cultures of Brettanomyces isolated from beer. In these tests the culture of Brettanomyces was added to beer at bottling at the rate of about 5 cells of Brettanomyces per 100 cells of culture yeast present in the beer. The beers were naturally conditioned at 18° C. and, after 2-5 weeks, they were tasted against a control of the same beer which had been bottled at the same time but which contained no added infection. The results are summarized in Table III for each of the three species. The taste of the infected beers became more objectionable as the beers became older but, for simplicity, the results of the tastings at different periods have been combined. It is clear that Brett. bruxellensis has much the worst effect on flavour of beer, although the other groups have an appreciable deleterious effect. The flavours most complained of in the infected beers were “harsh”, “strong after-bitter,” “mawkish” and “old beer flavour.”
The genus Brettanomyces consists of five well characterized species which have sufficient common distinguishing attributes to justify their retention as a separate genus. They have been found as spoilage organisms in wine, but their occurrence in present-day beer has been little commented on and there are very few published reports of spoilage of beer by Brettanomyces. One reason for this is that they grow very slowly; thus, in pasteurized beers they would not have time to produce a significant change of flavor before the time of pasteurization, and even in naturally-conditioned beers the effect on flavour would not appear for some considerable time after bottling unless the infection was very heavy. A second reason for the rarity of reports of the occurrence of Brettanomyces spp. in beer is that they are difficult to isolate from mixed culture unless a special medium is used and the incubation time is prolonged, so that they can readily escape detection unless a particular search is made. It is recommended that such a search should be made for Brettanomyces in all cases where the source of off-flavours in beers is being investigated.
Acknowledgements.—The author wishes to express his thanks to Mr. J. P. Lacey for his technical assistance, to Dr. A. K. Mills for his interest and encouragement, and to the Directors of Arthur Guinness Son & Co. (Dublin) Ltd. for their permission to publish this work.
1. Andrews. J., & Gilliland, R. B., this Journal. 1052, 189.
2. Barret, A., Bidan. P.. & Andre, L., C. R. Acad. Agric, 1055, 41, 420.
3. “Brettanomyces,” Brewers J.. 1052, 88, 200.
4. “Brcttanomyccs,” Brewers J., 1959, 95, 41.
5. Claussen, N. H., British Patent No. 28,184, dated 22.12.1003; this Journal, 1905, 204.
6. Claussen, N. H., this Journal, 1904, 308.
7. Custcrs, M. Th. J., Onderzoekingen over het Gislgeslacht Brettanomyces. Thesis, Delft,
8. Domcrcq, S., Étude el Classification des Levures de Vxn de la Gironde. Thesis, Bordeaux,
9. Etchclls, J. L., & Bell, T. A., Farlowia, 1950, 4 (i). 87.
10. Galzy. P., & Rioux, J. A., Progr. agric. vitic, 1055, 144, 365.
11. Krumbholz, G., & Tauschanoff, W., Zentralbl. f. Bakt. II, 1933, 88, 366.
12. Kufferath, H., C. R. Soc. biol. Belg., 1920, 185.
13. Kuffcrath. H., & van Laer, M. H., Bull. Soc. chim. Belg., 1021, 30, 270.
14. Loddcr, J., & Kregcr-van Rij, N. J. W.. The Yeasts. Amsterdam: North Holland Publishing
15. Ostcrwalder, A., Zenlralbl. f. Bakt. II, 1912, 33, 257.
16. Peynaud, E., & Domcrcq, S., Arch. Mihrobiol., 1950, 24, 266.
17. Schanderl, H., Die Mibrobiologie des Weines. Stuttgart: Verlagsbuchhandlung Eugen
18. Schanderl, H.. & Draczynski, M., Wein und Rebe, 1952, 20, 462.
19. Schionning, H., C. R. Lab. Carlsberg, 1907-1909,7, 138.
20. Seyffert. H., Woch. Brau., 1904. 21, 519.
21. Shimwell, J. L., this Journal, 1938, 563.
22. Shimwell, J. L., Amer. Brewer, May, 1947, 21, 56; this Journal, 1947, 267.
23. van der Walt, J. P., & van Kirken, A. E., Antonie van Leeuwenhoek, 1958, 24, 239.
24. van der Walt, J. P., & van Kirken, A. E., Antonie van Leeuwenhoek, 1959, 25, 145.
25. van der Walt, J. P., & van Kirken, A. E., Antonie van Leeuwenhoek, 1960, 26, 292.
26. Verona, O., Ann. Fac. Agr. Pisa, 1051,12, 123; A Hi. Accad. Hal. vile, vino, 1951, 3, 203.
27. Wiles, A. E., this Journal, 1053, 265.
28. Wiles, A. E., Brewers’ Guardian, May, 1960, 81.
29. Williams, D. O., & Stringer, W. J., Amer. Brewer, May, 1955, 39.
30. Windisch, S., Monats. Brau., I960, 13, 63.
31. Zaake, S., Die Brauerei, 1960, 842.
32. Zardetto de Toledo, O.. Teixeira, C. G., & Verona, O., Annali di Microbiologia ed
Enzimologia, 1959, 9, 22.