Identification of Foodborne Yeasts

Transcription

1 Journal of Food Protection, ol. 0. No., Pages {March 1) Copyright' International Association of Milk Food and Environmental Sanitarians Identification of Foodborne Yeasts TIBOR DEAR'* and LARRY R. BEUCHAT sity of Georgia, College of Agriculture. Department of Food Science, Agricultural Experiment Station, Experiment, Georgia 0 (Received for publication July, 1) ABSTRACT Improvements in identification procedures for yeasts lag behind recent developments in taxonomy. Sophisticated genetical and biochemical methods cannot be used in routine identification of yeasts. In foods representing specific and often selective ecological niches for yeasts, usually a restricted number of species are present, and for these a simplified identification key has been devised based only on to 1 tests. Yeasts representing 1 species reported to be present in foods are included in a key, and methods of simplified identification are described. Yeasts play an important role both in production and spoilage of our foods. Recently, yeasts have obtained an ever widening application in biotechnology. The continuous and increasing importance of these microorganisms has resulted in recent developments in their taxonomy, nomenclature and identification. However, the increasing number of described species, the frequent change of names with burdening synonyms, and the more sophisticated biochemical and genetical methods involved in yeast classification and identification, render difficulties for the uninitiated to follow these developments closely. Several comprehensive treatments and excellent reviews on foodborne yeasts were published (//,,0,,,) before the two recent treatises on the taxonomy and identification of yeasts appeared (,1). Hence, it seemed practical to review and summarize current taxonomic and identification schemes for yeasts which commonly cause spoilage of foods. The review is intended to provide food technologists and food microbiologists with an overview of specific groups of yeasts while attention is focused on their identification. It is, however, beyond the scope of this paper to deal with the industrial application of yeasts and their control in foods. CURRENT TRENDS IN YEAST TAXONOMY Yeasts are traditionally characterized, classified and identified by morphological and physiological criteria (e.g., shape of cells, mode of sexual and asexual reproduction, anaerobic fermentation and aerobic "assimilation" of sugars and certain growth requirements) {0). In addition to conventional diagnostic tests, biochemical methods provide important data for characterization of yeasts. Similarities or differences in the most important macromolecules in cells (DNA, RNA, proteins and polysaccharides) can be used to elucidate not only the degree of relatedness but also to reveal evolutionary connections. Extensive investigations of 1S ribosomal RNA (rrna) have revolutionized our conception of kingdoms of living organisms (1) and radically changed the taxonomy of bacteria (,). A similar method of molecular phylogeny concerns the S rrna of eucaryotic organisms and data are accumulating for fungi (,). Though the sporadic data available on S rrna sequences of yeast do not yet allow for a comprehensive treatment of higher taxa, other methods of molecular biology have been applied for decades and provide important clues for taxonomical considerations on genus and species level (1,). It is generally accepted that any difference in the guanine plus cytosine (G C) base composition of DNA greater than 1. to.0 mol% excludes the possibility that two strains belong to the same species (). On the other hand, similar G C ratios do not necessarily provide evidence of species identity. Identity can, nevertheless, be confirmed by the sequence homology (complementarity) between DNA molecules, which is generally higher than 0% and, in most cases, is over 0% between DNA extracted from strains of the same species (,). Much research has been devoted to determining the chemical composition of cell walls, capsular polysaccharides, whole cell hydrolysates and antigenic determinants (,,,1,) that also provided valuable information for classification. Somewhat less value has yet evolved from investigations of electrophoretic patterns of enzymes, although coenzyme Q analysis shows promise as a tool for characterizing yeast genera (). Two examples will illustrate the impact of molecular 'Permanent address: Department of Microbiology, University of Hor biologticulture and Food, Somloi ut, Budapest, HIII, Hungary. cades ago the genus Saccharomyces encompassed and biochemistry on yeast classification. Two de more JOURNAL OF FOOD PROTECTION. OL. 0. MARCH 1

2 DEAK AND BEUCHAT than 0 species (). It was, however, heterogenous, i.e., the group consisted of haploid and diploid, conjugating and nonconjugating species; recognized species also had great strain variability with regard to conventional physiological tests and could be separated only with a great degree of uncertainty. Numerous efforts have been made to resolve the confused situation (,1,,0,); however, the most significant approach was the determination of DNA base composition and homology (1,,). These studies resulted in reducing some 0 Saccharomyces species to biotypes and nomenclatural synonyms of a single species Saccharomyces cerevisiae () (Table 1), as well as a transfer of several other species to the genera Torulaspora and Zygosaccharomyces while amalgamating simultaneously several of them () (Table ). More recently, the number of recognized Saccharomyces species has been reduced to seven (), with the rejection of such wellknown names as Saccharomyces carlsbergensis and Saccharomyces bayanus that have clear applications in breweries and wineries, respectively. The controversy still continues and suggestions have been made for once again establishing some formerly separate species on the very same basis of DNA relatedness (0,1). The genus Candida is a large but clearly heterogenous assemblage of asporogenous (anamorphic, imperfect) yeasts. After merging the former genus Torulopsis with Candida (), it now contains more than onethird of the yeast species, some 00 (). Studies on the composition of cell walls and whole cell hydrolysates of a number of Candida species support data that may help in resolving the heterogeneity of this enormously distended genus. According to Arx and Weijman (), the genus Candida should be restricted to those anamorphs that show ascomycetous affinity, cell walls being composed mainly of mannan and glucan, and containing no or only a very low amount of chitin. Other yeasts that should be excluded from Candida show basidiomycetous affinity in that their cells walls contain moderate or high amounts of chitin. Even among these, two groups can be distinguished on the basis of monosaccharides found in cell hydrolysates. One group contains xylose in addition to glucose and mannose, and the other group contains fucose, rhammose and/or galactose, but not xylose. The latter group shows affinity to the smuts (Ustilaginales). These findings correlate well with and are supported by other features such as the type of asexual propagation, i.e., conidiation, according to modern terminology (1). Several workers have made proposals for taxonomic revision of Candida (,) which are pending (). GROUPS OF FOODBORNE YEASTS From a taxonomic point of view, yeasts are not homogenous. During evolution, adaptation of unicellular life forms has occurred several times among different groups of fungi. Even the definition and circumscription of yeasts are not unequivocal (1). However, for practical of Saccharomyces cere TABLE 1. Species put into synonymy visiae () Species recognized by Lodder () aceti bayanus (beticus, cheriensis oviformis, pastorianus) capensis cerevisiae (ellipsoideus, intermedius, vini, willianus) chevalieri (lindneri) coreanus diastaticus globosus heterogenicus hienipiensis inusitatus italicus norbensis oleaceus oleaginosus prostoserdovii uvarum logos) Species described after 1. cordubensis 1. gaditensis 0. hispanica (carlsbergensis, TABLE. Saccharomyces species transferred to genera Zygosaccharomyces and Torulaspora. Species recognized by Lodder () Present classification (1) Saccharomyces amurcae Zygosaccharomyces cidri S. cidri Z. cidri S. bisporus Z. bisporus S. eupagycus Z. florentinus S. florentinus Z. florentinus S. microellipsodes Z. microellipsoides S. montanus Z. fermentati S. mrakii Z. mrakii S. rouxii Z. rouxii S. delbrueckii Torulaspora delbrueckii S. fermentati T. delbrueckii S. inconspicuus T. delbrueckii S. saitoanus T. delbrueckii S. rosei T. delbrueckii S. vafer T. delbrueckii S. kloeckerianus T. globosa S. pretoriensis T. pretoriensis purposes, yeasts may be defined as unicellular fungi in which vegetative reproduction occurs mostly by budding. This definition fits most foodborne yeasts. It is to be noted, however, that some yeasts form welldeveloped true hyphae and/or pseudohyphae as well as budding cells. These forms, collectively called yeastlike organisms, are borderline between filamentous and unicellular fungi. Those yeasts that form spores by any mode of sexual reproduction can be classified accordingly to the appropriate fungal division. Several species exhibit an anamorphic state which could be related to ascomycetous or basidiomycetous teleomorphic (perfect) forms of yeasts. This traditional subdivision of yeasts has become more complicated as the systematics of fungi improve. The endogenous formation of sexual spores in yeasts differs substantially from the characteristic development of ascospores in which yeast cells lack dicaryotic ascogenous JOURNAL OF FOOD PROTFCTION. OL. 0, MARCH 1

3 IDENTIFICATION OF FOODBORNE YEASTS hyphae and ascocarps. Hence, this group of yeasts can be considered as a separate division of fungi [Endomycota (), Zymomycota ()]. Among the small number of yeasts having basidiomycetous affinity, some are true basidiomycetes whereas others are more similar to the smuts, which in turn may also be grouped in a separate division (). Consequently, this tripartite division can be realized also among the imperfect yeasts (Table ). TABLE. Classification of yeasts among fungi [after Moore (,), modified and completed]. Tax a Eumyceteae Chytridiomycota Zygomycota Endomycota Endomycetes Endomycetales* Taphrinomycota Ascomycota Ustomycota Ustomycetes Sporidiomycetes Sporidales b Basidiomycota Phragmobasidiomycetes Aphyllophorales c,i Formtaxa of imperfect fungi Deuteromycota Blastomycota Endoblastomycetes Brettanomycetales' i B asidioblastomycetes Sporobolomycetales b Cryptococcales c a True yeasts and yeastlike organisms (ascomycetous yeasts and their imperfect forms). b Smutlike yeasts (ustomycetous yeasts and their imperfect forms). c Basidiomycetous yeasts and their imperfect forms. d Mostly filamentous fungi and a few yeastlike forms. The classification of the true yeasts (Endomycetales) differs according to different authors (Table ); however, the main groups are more or less similar. No further details of classification will be given here. Readers are referred to the recent edition of the most widely recognized taxonomic manual "The Yeasts" (1) which includes 0 genera and 00 species. The previous edition included genera and species (). Other authors recognize different numbers of these taxa, e.g., genera and species by Barnett et al. (), and genera and about 00 species by Arx et al. (). Thus, taxonomy and classification of yeasts are in a state of flux. Some important genera of foodborne yeasts deserve further mention. Of the yeastlike fungi forming septate hyphae and budding cells, and often pseudohyphae as well, few species are of importance in food spoilage. Kregervan Rij (1) merged these species into the genus Saccharomycopsis, although most species have been ascribed to separate genera, e.g., Endomyces, Endomycopsella and Yarrowia (). The hyphae of these yeasts sometimes form arthroconidia. Members of another genus, Endomyces [Dipodascus, according to Arx ()1, also form septate hyphae which break into arthroconidia but they have no budding cells. Hence, these species are usually not considered as yeasts (1), although they look like yeasts and behave similarly. Their anamorphs, Geotrichum species, are frequently found in decaying foods and on food processing machinery. In the genus Schizosaccharomyces, no budding cells occur either, but rather fission is the characteristic way of vegetative reproduction. One species forms true hyphae breaking up splitting cells. Schizosaccharomyces TABLE. Families of the order Endomycetales according to different authors* Kregervan Rij (1) Arx et al. () Gams et al. () Novak () 1. Endomycetaceae Endomyces. Ascoideaceae Dipodascus. Spermophthoraceae Metschnikowia. Saccharomycetaceae.1. Schizosaccharomycetideae Schizosaccharomyces.. Nadsonioideae Nadsonia Hanseniaspora.... Lipomycetoideae Lipomyces Saccharomycetoideae Hansenula Pichia Saccharomycopsis Saccharomyces Torulaspora Zygosaccharomyces 1. Endomycetaceae Endomyces Dipodascus. Ascoideaceae Saccharomycopsis. Metschnikowiaceae Metschnikowia. Schizosaccharomycetaceae Schizosaccharomyces. Saccharomycodaceae Nadsonia Hanseniaspora. Saccharomycetaceae Lipomyces Hansenula Pichia Saccharomyces Torulaspora Zygosaccharomyces a Only the most important genera are listed for comparison. Dipodascaceae Dipodascus Schizosaccharomyces Endomycetaceae Endomyces Saccharomycopsis Metschnikowiaceae Metschnikowia. Saccharomycodaceae Nadsonia Hanseniaspora. Saccharomycetaceae Lipomyces Hansenula Pichia Saccharomyces Torulaspora Zygosaccharomyces Endomycetaceae Endomyces Dipodascus Ascoideaceae Saccharomycopsis Nematosporaceae Metschnikowia Schizosaccharomycetaceae Schizosaccharomyces Saccharomycodaceae Hanseniaspora Nadsowiaceae Nadsonia Lipomycetaceae Lipomyces Hansenulaceae Hansenula Pichia Saccharomycetaceae Saccharomyces Torulaspora Zygosaccharomyces JOURNAL OF FOOD PROTECTION. OL. 0. MARCH 1

4 DEAK AND BEUCHAT are, however, unrelated to Endomyces and are in fact only distantly related to true yeasts (), although they have been traditionally considered as such because of their strong fermenting ability. Apiculate (lemonshaped) cells are characteristic of the genus Hanseniaspora. Slight differences can be only found among species which are common in fruits and must, mostly in the anamorphic state, called Kloeckera. A unique needleshaped ascospore is the feature of the genus Metschnikowia. Spore formation can, however, be seen rarely. More frequently, large, lipidcontaining, thickwalled cells called chlamydospores can be found. Most of the true yeasts belong to genera which are collected in the family Saccharomycetaceae (Table ). The largest genus is Pichia which now accomodates the nitrateassimilating species that formerly belonged to Hansenula (). This large assemblage of yeasts may form single budding cells, frequently pseudohyphae and rarely even true hyphae. Some species ferment strongly, whereas others have weak or no fermentative capabilities. Spores are surrounded with a brim that gives them a hat or saturn shape. Many common foodborne yeasts belong to this genus. To alleviate the heterogeneity of the genus, several yeasts closely related or similar to some Pichia are classified in separate genera, mainly by reason of the mode of spore formation and the shape of spores. Species of Citeromyces, Issatchenkia, Williopsis and Debaryomyces are important in food mainly because of their salt or sugar tolerance. They possess little fermentative activity. Dekkera is a genus characterized by peculiar metabolic properties in consequences of which they produce acetic acid and grow very slowly. They may cause spoilage of alcoholic beverages and soft drinks, often in anamorphic forms, called Brettanomyces. Strong fermentation and mostly or exclusively single budding cells are the main features of the genera Saccharomyces, Torulaspora, Zygosaccharomyces and Kluyveromyces. The latter genus is distinguished by beanshaped spores easily liberating from asci, whereas spores in the other genera are spherical and remain in asci. The mode of sexual reproduction varies. Cells of Saccharomyces are diploid and directly transform into asci when the spores form. In Zygosaccharomyces, conjugation between independent haploid cells precedes ascus formation, whereas in Torulaspora conjugation occurs between the mother cell and bud. These features are not always easily observed and the identification of species by physiological criteria is also difficult because most of these criteria become uncertain and variable after amalgamating a number of species on ground of DNA homology. These genera contain several species of industrial importance and commonly occur in a wide variety of foods. As interesting as they are from a taxonomical standpoint, the small groups of basidiomycetous yeasts have little practical importance to the food industry. A few species of Leucosporidium, Rhodosporidium, Sporidiobolus and Filobasidiella are sometimes found in food. They are usually isolated in the anamorphic state because selfsporulating strains are rare. These haploid anamorphs have been grouped in various imperfect genera such as Candida, Sporobolomyces, Rhodotorula or Cryptococcus. Among the imperfect yeasts that do not reproduce sexually, three main groups can be distinguished according to their affinity to perfect counterparts. One group is ascomycetous and two are basidiomycetous whose taxonomic denomination is still debated (see Table ). There is sufficient evidence to distinguish between these groups (Table ); moreover, it is also possible to differentiate ascomycetous and basidiomycetous yeasts by simple tests (diazonium blue B staining and urease reaction). Unfortunately, no such test is available for the easy differentation of smutlike yeasts from other basidiomycetous yeasts. Efforts to form homogenous genera of imperfect yeasts according to their connection with perfect forms have not TABLE. Some genera of the family Saccharomycetaceae including important foodborne yeasts. Genus Citeromyces Debaryomyces Dekkera Hansenula c Issatchenkia Kluyveromyces Pichia Saccharomyces Torulaspora Williopsis Zygosaccharomyces No. of species True hypha _b Pseudohypha Spherical spores Spores with brim Fermentation Nitrate assimilation Sexual characteristics H/D, het H, hom, hg D, hom H/D, hom/het D, hom H/D, hom/het H/D, hom/het D, hom H, hom, hg D/H, hom H, hom, ig "Abbreviations: H, haploid; D, diploid; hom, homothallic; het, heterothallic; hg, heterogamous conjugation between mother cell and bud; ig, isogamous conjugation between two cells. b, negative; v, variable;. positive; ; strongly positive. ^'Recently grouped in Pichia (). JOURNAL Oh FOOD PROTECTION. OL. 0, MARCH 1

5 IDENTIFICATION OF FOODBORNE YEASTS TABLE. Some important differences between yeasts of ascomycetous or basidiomycetous affinity. Character Morphological Capsules, mucous colonies Carotenoid pigments Ballistoconidia Cell wall structure Septal spores Budding Physiological and biochemical Cell wall composition Diazonium blue B reaction Production of starchlike compound Urease activity Extracellular DNAase Type of coenzyme Q Guanine cytosine mol % Ascomycetous trait bilayered simple or none holoblastic glucan, mannan ( ),,, () <0 Basidiomycetous trait ; multilayered dolipores enteroblastic glucan, chitin or mannan, chitin (),, >0 yet fully been successful. It has been suggested that the genus Candida should retain only ascomycetous imperfect species (,). To this end, several species have been transferred to Rhodotorula and Cryptococcus. However, the large genus Candida is still heterogenous, becoming a reservoir of all imperfect forms that cannot be characterized by some peculiar feature such as the mode of vegetative reproduction or the shape of cells. The absence of pseudohyphae no longer sets apart the former Torulopsis species which were merged with Candida (). The identification of an enormous number of Candida species, among them many important foodborne yeasts, can be achieved only with great difficulty by applying physiological criteria. The imperfect genera Sporobolomyces, Rhodotorula and Cryptococcus comprise anamorphic forms of basidiomycetous affinity. Teleomorphs of some species are also known. These yeasts do not ferment carbohydrates. Most of them produce yellow, pink or red pigments and a polysaccharide capsule rendering the colony mucous. A few species occur frequently in food as soil or air contaminants. The genus Trichosporon includes yeastlike fungi which develop true hyphae and arthroconidia as well as budding cells. Unambiguous basidiomycetous features distinguish them from the ascomycetous anamorph, Geotrichum, which has similar appearance. At this point it should be noted that only one name is to be used for a species, that of the teleomorph if it does exist. Many yeasts have been described in their anamorphic state because the mode of sexual reproduction was not known. In some yeasts, the ability to form sexual spores is lost, whereas others are heterothallic and do not sporulate in the absence of the opposite mating type. Even many homothallic yeasts normally develop as anamorphs, and special conditions are required to induce spore formation. If once, however, the teleomorph is discovered for a species formerly known as anamorph, the name of the latter reduces into a synonym of the teleomorph. A greater number of anamorphic synonyms are known (Table ), in addition to the other numerous synonyms created by nomenclatural changes as consequences of taxonomic rearrangements (). SIMPLIFIED IDENTIFICATION METHODS In contrast to classification which considers data obtained by sophisticated methods of molecular biology and biochemical analysis, identification of yeasts is generally based on morphological and physiological features which can be determined by routine diagnostic tests. Traditional identification procedures rely heavily upon morphological characteristics of sexual reproduction, whereas physiological characteristics (fermentation and assimilation properties) are mainly considered for determination of species. At the other extreme, identification schemes have been elaborated by applying physiological criteria only (,). Other identification systems integrate ecological data, simple morphological observations and selected determinative tests (,1). Morphological and physiological characteristics can be performed by traditional methods (0) or by identification kits (1,). Comparison of unkown strains with descriptions of recognized species can be achieved by identification keys, tables, punch cards and computers (,1). Hence, identification of yeasts can be done by many techniques, each having its merit and proper place with specific advantages and disadvantages. For ecological surveys of foods, one is interested in the general profile of yeast flora and the dominant species whose physiological attributes determine the fate of food under given environmental conditions (0,,). Hence, considering needs of food monitoring and technology, the most satisfying identification system would be one which gives reliable discrimination of a wide range of species with less labor, material and time. Unfortunately, no such ideal method is known. JOURNAL OF FOOD PROTECTION, OL. 0, MARCH 1

6 DEAR AND BEUCHAT The traditional method, though it is the only ine acceptable for taxonomical purposes, requires considerable experience and skill in the performance and evaluation of some 0 specified tests. Computergenerated keys also require numerous (0 to 0) tests and can be applied, provided a program and all data are at hand. Readymade identification kits perform reliably only for the restricted TABLE. Some known anamorphs of foodbome yeasts. Teleomorph Anamorph Citeromyces matritensis Clavispora lusitaniae Debaryomyces hansenii Dekkera anomala D. bruxellensis D. intermedia Dipodascus geotrichum Filobasidium capsuligenum F. uniguttulatum Filobasidiella neoformans Hanseniaspora guilliermondii H. occidentalis H. osmophila H. uvarum H. valbyensis H. vinae Hyphopichia burtonii Issatchenkia orientalis Kluyveromyces marxianus var. lactis K. marxianus var. marxianus K. thermotolerans Leucosporidium scottii Metschnikowia pulcherrima M. reukaufii Pichia anomala Pichia canadensis P. capsulata P. fabianii P. fermentans P. guilliermondii P. holstii P. humboldtii P. jadinii P. membranaefaciens P. nakasei P. norvegensis P. stipitis Rhodosporidium infirmominiatum R. toruloides R. paludigenum R. dacryoideum Saccharomyces cerevisiae S. exiguus Sporidiobolus pararoseus S. salmonicolor Tremella aurantia Torulaspora delbrueckii Wickerhamiella domercqiae Yarrowia lipolytica Zygosaccharomyces rouxii Candida globosa C. lusitaniae C. famata Brettanomyces anomalus B. bruxellensis B. intermedius Geotrichum candidum Candida japonica Cryptococcus uniguttulatus C. neoformans Kloeckera apis K. javanensis K. corticis K. apiculata K. japonica K. africana Candida variabilis C. krusei C. sphaerica C. kefyr C. dattila C. scottii C. pulcherrima C. reukaufii C. pelliculosa C. melinii C. molischiana C. fabianii C. lambica C. guilliermondii C. silvicola C. ingens C. utilis C. valida C. citrea C. norvegensis C. shehatae Cryptococcus infirmominiatus Rhodotorula glutinis R. graminis R. minuta Candida robusta C. holmii Sporobolomyces shibatanus S. laurentii Cryptococcus laurentii Candida colliculosa C. domercqiae C. lipolytica C. mogii domain of some 0 species of clinical importance (). Other keys specially designed for the most important foodbome yeasts consider an even smaller number of species (1,,0). Hence, a real need has existed for devising a simplified identification scheme for a wide range of foodbome yeasts. Two approaches have been taken in our laboratory to simplify the procedure of identification. First, the number of diagnostic tests has been reduced by selecting the most efficient ones for discriminating yeasts. Secondly, the number of yeast species considered has been reduced by selecting only those yeasts which have been found in foods. For the identification of yeasts, numerous standard tests are applied to determine morphological and physiological characteristics (Table ). Responses of yeasts to diagnostic tests can be described as positive or negative. Sometimes, however, the response is not unequivocal but delayed, weak, uncertain or variable (assigned d, w, ± and v, respectively). This can be attributed to the method of investigation or it can be an inherent property of yeast (e.g., the use of a substrate depends on an inducible enzyme). Yeast species differ, of course, in responses to different tests. Obviously, the most efficient discriminating test would be one that would divide a given number of species into two equal groups (positive and negative) with no variable responses. The efficiency (e) of any one test can be calculated, according to Gower and Barnett (0) from the expression: e = (p 0.) (q 0.) r where p is the ratio of species with positive responses, q is the ratio of negative species and r denotes the ratio of variable responses. Out of all possible tests, the most efficient is the one for which the value e is smallest. This formula was used in our laboratory to select tests for developing a simplified identification key (). On the basis of data collected for species (), the efficiency of various fermentation and assimilation tests was first calculated (Table ). For example, the fermentation of glucose was reported positive for 1, negative for 1 and variable for 1 species. Hence, the efficiency of glucose fermentation was calculated as 0.0. Overall, the most efficient test proved to be maltose assimilation. Although there were only two species with variable urease reaction, the efficiency of this test is relatively low because it divides the yeasts into two rather unequal groups. The construction of the key started with maltose assimilation. This approach divided the species into two groups consisting of positive and negative yeasts, as the 1 variable species were added to each group. The next most efficient test was then selected for the division of each branch. Although sucrose assimilation appeared efficient for separating the whole group of yeasts studied, its application in the second level after maltose proved to be redundant because nearly all maltosepositive yeasts assimilated sucrose, and most maltosenegative species responded similarly to sucrose. It was observed that maltosepositive species are separated efficiently by raffinose assimilation and mal JOURNAL OF FOOD PROTFCTION. OL. 0, MARCH 1

7 IDENTIFICATION OF FOODBORNE YEASTS tosenegative species are separated efficiently by galactose assimilation (Table ). The selection of tests was continued successively until 1 main groups were formed. However, to avoid an unduly increase in the number of tests, the selection of the most efficient test was not rigorously applied. In contrast, an attempt was made to minimize the number of tests by choosing, whenever it was practicable, a test that occurred elsewhere on a different branch of the key, even if it was slightly less effective. Finally, only seven tests were used in constructing the key (Fig. 1). In this respect, the key markedly differs from other keys which TABLE. Characteristics used in the identification of yeasts. 1. Morphological a. Sexual reproduction Mode of conjugation Characteristics of spores and sporangia Formation of basidiospores b. egetative reproduction Budding, conidiogenesis c. Microscopic growth egetative cells True hyphae and pseudohyphae d. Macroscopic growth Liquid cultures Colonies on solid media. Physiological a. Fermentation of sugars b. Assimilation of carbon sources c. Assimilation of nitrogen sources d. itamin requirements e. Temperature of growth f. Growth at low water activity g. Resistance to cycloheximide h. Production of starchlike compound i. Production of acetic acid j. Splitting of urea. Biochemical, genetic a. Composition of cell wall b. Coenzyme Q pattern c. DNA base composition d. DNA sequence homology e. S rrna structure make use of 0 to 0 or more tests, e.g., those generated by computer programs (,). The other way of simplifying the key was its application for a particular group of yeasts occurring in a relatively defined habitat, i.e., in foods. Although foods represent widely diverse niches from a microecological point of view (), the occurrence in foods of many yeasts which have adapted to other specific habitats is highly improbable (). Hence, omitted from the keys are those yeasts which are exclusively found in sea water, tree exudates, nectars of cacti and other exotic niches, or which are intestinal parasites of various animals and symbiotes of insects (Table ). Most genera in Table are represented by a solitary species, often described with a single strain. Besides these, however, there are some 0 species of Candida and 0 species of Pichia found in particular habitats only, none of which has been reported from foods. These were not considered in the key either. Based on an extensive survey of literature from the 's on, a list of 1 yeast species ever reported in foods was compiled (Table 1). In the list, yeast genera are represented, of them by a single species. Nearly half of the species come from the genera Candida and Pichia ( and 1 species, respectively). Occurrence of species was registered only once, whereas 0 species can be considered commonly associated with foods, encountered mainly in spoilage. According to the type of food, the widest range of yeasts resides on fruits, vegetables and other plant materials as well as in must, wine and cider, from which about 0 yeast species were described (,,,,). Some 0 to 0 different species form the yeast flora in meat and meat products (1,1,,,), beverages (,,) and beer (,,,,). Only about 0 can be found in foods which offer a more restraining environment, such as salted, brined and fermented products as well as those containing high sugar concentration (,0,,). This range is nevertheless much broader than the number of wellknown xerotrophic (osmophilic) yeasts (e.g., Zygosaccharomyces bailii, Zygosaccharomyces rouxii, Torulaspora delbrueckii, Pichia subpelliculosa, Debaryomyces hansenii, Candida versatilis, Candida lactiscondensi). About a dozen species, usually TABLE. Efficiency of some identification tests. Responses of yeasts Negative Test Positive ariable Efficiency Glucose fermentation Maltose fermentation Maltose assimilation Sucrose assimilation Cycloheximide resistancy Cellobiose assimilation Growth at C Galactose assimilation Raffinose assimilation Nitrate assimilation Urease reaction Gluconate assimilation JOURNAL OF FOOD PROTECTION. OL. 0, MARCH 1

8 0 DEAR AND BEUCHAT TABLE. Efficiency of some secondary identification tests. Test Positive 1. Primary test: Maltose assimilation positive ( species) Raffinose assimilation Glucose fermentation 1 Cycloheximide resistance Erythritol assimilation Sucrose assimilation 1. Primary test: Maltose assimilation negative ( species) Cycloheximide resistance Galactose assimilation Glucose fermentation Cellobiose assimilation Sucrose assimilation Responses of yeasts Negative ariable Efficiency / \ / \ / \ / \ Ce N0 ure G ure d Ce ure A A A A A A A A Figure 1. Master key of identification. Assimilation tests: M, maltose; R, raffinose; G, galactose; Ce, cellobiose; N0, nitrate; d, glucose fermentation; ure, urease reaction. Figures denote the sixteen subgroups, each formed by four basic tests. found in soil, were described only in the environment of food plants. The 0 most frequent foodborne yeasts which are widespread in nearly all kinds of foods are listed in Table 1. It has to be noted that among the yeasts reported from foods there are a few species originally found in quite different habitats. An example would be the single strain of Cryptococcus skinneri isolated from grass in 1 (1), until one further strain was identified from ground meat by Dalton et al. (1) in 1. This is a reminder that keys confined to species occurring in a given habitat may be loaded with error. This defect can be, however, negligibly small when the composition of yeast flora is surveyed, as in ecological studies in which erroneous identification of a few strains would not change the ratio of dominant species or disturb the overall picture of microflora. One should always reckon with the possibility of finding a yeast species in food, the occurrence of which is just accidental. Keys of this kind concerning restricted habitats, e.g., wines, strawberries or clinical materials, have been described and applied with good success (,,,1,1,1,). TABLE. Yeast genera omitted from the key of foodborne yeasts". No. of Genus species Habitat 1. True yeasts and yeastlike organisms Ambrosiozyma Botryoascus Coccidioascus Cyniclomyces Guilliermondella Hormoascus Neurospora Pachytichospora Sporopachydermia Sterigmalosporidium Wickerhamia Wingea. Basidiomycetous yeasts Chianosphaera Fibulobasidium Holtermannia Sirobasidium Tremella. Imperfect yeasts Aciculoconidium Malassezia Mastigomyces Oosporidium Phaffia Sarcinosporon Schizoblaslosporion Sympodiomyces Ambrosia beetles Ambrosia beetles Intestinal parasite b Rabbit stomach 0 Tanning fluid Ambrosia beetles Plant pathogen Gut of a baboon, soil Cacti, tanning fluid, sea Arctic sea Gut of a rodent Bark beetles, soil Dead tree limbs b Basidiospores Basidiospores Basidiospores Basidiospores Fruit flies Animal skin c Tropical fruit Slime flux Slime flux Skin lesion Soil Sea water "Besides the genera listed, some 0 species of the genus Pichia and some 0 species of the genus Candida have also been omitted. These species have been described..from specific habitats, e.g., tunnel of bark beetles in wood, insect frass on trees, rotting cacti, arctic sea, antarctic soil, desert, etc. b No living culture known. ^'Requires specific cultivation. JOURNAL OF FOOD PROTECTION. OL. 0, MARCH 1

9 IDENTIFICATION OF FOODBORNE YEASTS 1 TABLE 1. List of foodhorne yeasts*. Species A B C Arthroascus javanensis Brettanomyces abstinens B. claussenif B. custersianus B. custersii B. lambicus B. naardenevis Bullera albcf B. armeniaca B. crocea B. tsugae Candida acuta C. agretis C. albicans C. apicola 0 C. bacarum C. beechii C. blankii C. boidinif C. buffonii C. butyri C. cacaoi C. cantarellif C. castellii C. catenulata 0 C. diddensiae 0 C. diversa c C. etchellsii 0 C. ethanolica C. fennica C. foliorum C. fragariorum C. friedrichii c C. fructus C. fusiformata C. glabrata c C. glaebosa C. graminis C. gropengiesseri C. hellenica C. humilis C. incommunis C. inconspicua c C. insectamans C. intermedial C. ishiwadae C. lactiscondensi c C. magnoliae c C. maltosa C. maritima C. membranaefaciens C. mesenterica 0 C. milleri C. mogii C. montana C. multisgemmis C. musae C. norvegica c C. oleophila C. parapsilosis c Types of food b D E F G H I K JOURNAL OF FOOD PROTFCTION, OL. 0. MARCH 1

10 DEAK AND BEUCHAT Species C. pinus C. puslula C. rugosa c C. sake c C. salmanticensis C. saniamariae C. silvae C. silvatica C. silvicultrix C. solani c C. sorboxylosa c C. spandovensis C. steatolyticcf C. st el lata" C. tenuif C. tropicalis 0 C. vanderwaltii C. vartiovaaraf C. veronae C. versatilis c C. vinaria C. vini c C. wickerhamii C. zeylanoldes c Citeromyces matritensis c Clavispora lusitaniae c Cryptococcus albidus Q C. curvatus C. dimennae C. flavus C. gastricus C. humicolus C. hungaricus C. kuetzingii C. Iaurentii c C. luteolus C. macerans C. skinneri C. terreus Debaryomyces castellii D. hansenif D. marama D. polymorphus Dekkera anomala c D. bruxellensis 0 D. intermedia c Eeniella nana Endomyces fibuliger c Endomycopsella vini Filobasidiella neoformans Filobasidium capsuligenunf F. uniguttulatum Geotrichum candidunf C. capitatunf G. fermentans G. fragrans Hanseniaspora guilliermondi^ H. occidentalis H. osmophila H. uvarum c Types of food ' JOURNAL OF FOOD PROTECTION. OL. 0, MARCH 1

11 IDENTIFICATION OF FOODBORNE YEASTS Types of food b Species H. valbyensis c H. vineae c Hyp lopichia burtonii'' lssaichenkia orientalis~ I. terricola c Kluyveromyces delphensis K. marxianus 1 K. thermotolerans c he ucosporidium scottif Lipomyces starkeyi Lodderomyces elongisporus c Melschnikowia pulcherrima c M reukaufi'f Nadsonia elongata Pachysolen tannophilus Pichia angustcf P. anomala e P. canadensis P. capsulata P. carsoni'f P. delftensis P. etchellsif P. fabianii P. farinosa c P. fermentans c P. fluxuum P. guilliermondii Q P. haplophila P. holslii P. humboldtii P. jadinii c P. kluyverf P. media P. membranaefaciens^ P. minuta P. nakasef P. norvegensis P. ohmeri 0 P. onychis Q P. pastoris P. pijperi 0 P rhodanensis P silvicola P stipitis P subpelliculosa c P loletana Rhodosporidium diobovatum R infirmominiatum c Rhodolorula acheniorum R R R R R R aurantiaca c glutinis c graminis ingeniosa minuta 0 mucilaginosa 0 Saccharomyces S. S. S. S. dairensis 0 exiguus c kluyveri 0 telluris cerevisiae c JOURNAL OF FOOD PROTECTION, OL. 0. MARCH 1

12 DEAK AND BEUCHAT Species S. unisporus c Saccharomycodes ludwigii c Saccharomycopsis malanga Schizosaccharomyces japonicus c S. malidevorans c S. octosporus c S. pombe c Schwanniomyces occidentalism Sporidiobolus pararoseus' Sporobolomyces salmonicolor' S. alborubescens S. holsaticus S. puniceus S. roseus Q Stephanoascus ciferrii Sterigmatomyces nectairi Torulaspora delbrueckii c T. globosa Trichosporon culaneuirf T. brassicae T. pullulans* Trigonopsis variabilis Wickerhamiella domercqiae' c Williopsis beijerinckii W. californica c W. mrakii W. saturnus Yarrowia lipolytica 0 Zygosaccharomyces bailii c Z. bisporus c Z. cidri Z. fermenlati Z. florenlinus Z. microellipsoides c Z. mrakii Z. rouxif Types of food ''References:,,,,1',,,,,,,0,,,, (out of some 00 articles surveyed, only selected comprehensive treatments and reviews are listed here). 'Types of food: A, fruits, vegetables, raw plant materials; B, fermented, salted or brined foods; C, concentrates, syrups, jams, honey; D, beverages, soft drinks, juices; E, must, wine, cider; F, beer, ale; G, meat, poultry, meat products; H, milk, cheese, other dairy products; I, other various food products (e.g., pastries, bread, confectioneries, salads, etc.); K, food plant environment, air, soil, sewage. c Most frequently occurring yeasts (). IDENTIFICATION KEY The simplified identification method so far described is illustrated in the master key shown in Fig. 1. This key divides foodborne yeasts into sixteen subgroups by the use of seven basic tests. Of course, four tests are enough for separating each of the sixteen branches. The remaining three tests can be used to complete the identification of each strain. To this end, some additional tests are also necessary, together with a microscopic investigation. The subgroups contain to 1 species (in the average 1). It is possible to arrive directly at species identity by the seven basic tests in the smallest subgroups only. In the other cases, 1 to tests are needed, and in the three largest subgroups, or additional tests must be done to complete species identification. These tests are selected respectively for each group, and can be performed after the results of the basic tests have been read. This is a peculiarity of the scheme that it is partitioned into two stages. The most frequently used additional tests are the assimilation of xylose, erythritol, mannitol, melibiose and trehalose. In a few cases other tests are used (e.g., growth at C, resistance to cycloheximide and fermentation of certain sugars). In addition to characteristics of morphology viewed by microscopy, macroscopic properties of growth are also considered which can be ascertained without further tests (e.g., color of colony and formation of a pellicle). JOURNAL OF FOOD PROTECTION. OL. 0, MARCH 1

13 IDENTIFICATION OF FOODBORNE YEASTS TABLE 1. Twenty most common foodborne yeasts (,,0). a Species Occurrence Candida glabrata C. sake C. tropicalis C. versatilis Cryptococcus albidus Debaryomyces hansenii Issatchenkia orientalis Kluyveromyces marxianus Pichia anomala P. fermentans P. guilliermondii P. membranaefaciens Rhodotorula glutinis R. mueilaginosa Saccharomyces cerevisiae S. exiguus Sporobolomyces roseus Torulaspora delbrueckii Zygosaccharomyces bailii Z. rouxii a See footnote in Table 1. A, D, A, D A, C, A, C, A, C, A, C, A, C A, B C, D, E, G, I E, F, G, I, K C, D, E, F, G, C, E, F, G, I E, G, H, I, K C, D, E, F, G, C, D, E, F, G, D, E, F, H, I B, D, E, F, G, C, D, E, F, G, C, D, E, F, G, C, D, E, F, G, D, E, G, H, I, D, E, F, G, H, C, D, E, F, G, D, E, F, G, H, E, F, G, H, K C, D, E, F, G, D, E, F, I C, D, E, F, H, H, I H, H, K H, H, H, K I, I, I H I I I I, K I I K K I, K Overall, to 1 tests are used in the simplified identification scheme, hence eliminating at least 0% of the otherwise necessary diagnostic tests. The simplified method is economic, saving both labor and materials. Even time can be spared if it is considered that only one petri dish and three test tubes are needed for one strain, hence many isolates can be examined simultaneously. By applying the most frequently used secondstage tests at the start, identification can be done in a single step, arriving at species identity in most cases. Further tests are necessary only for the differentiation of a few very similar species. For the onestep method, two petri dishes and three test tubes are used to examine one strain (Fig. ). Individual identification keys for each of the 1 subgroups formed by the master key are arranged in the familiar dichotomous fashion in Table 1. Physiological and morphological characters are listed (left column) for which results can be adequate (positive, middle column) or inadequate (negative, rightside column). In either case, the answer may be a number referring to another entry in the key, or it may be the name of a species with which the unknown strain can be identified with high probability. Ce Te ure d N0 Figure. Arrangement of tests for the onestep way of identification. Abbreviations as shown in Figure I: X, xylose; Mt, mannitol; Er, erythritol; Te, trehalose; Me, melibiose; D, glucose. The accuracy of identification depends mainly on the certainty of results of the seven basic tests. An erroneous reading of a result would lead to a subgroup within which the final identification is also mistaken or impossible. It should be noted that the simplified scheme is based on selected yeast species and even correct results differing from those in the keys are likely to occur sometimes. Hence, species identification should not be based only on those features included in the key. The amount of data available generally exceeds that considered in the key. To confirm identification, all data are to be compared with the characters of the likely identical species. Standard methods of identification are described in detail by Lodder () and Kregervan Rij (1) and can also be found in several other sources (,,,,,0). No effort is made to recapitulate these procedures and media here. Two comments will be made on methodology concerning the use of the simplified key and the identification of yeasts isolated from foods. As set forth before, not all standard procedures are to be done for the application of the simplified key. On the contrary, only a few of them are necessary. Great reliance is given to the sugar assimilation tests on which the master key is based. Great care should be exercised in executing and evaluating these tests. For routine purposes the following regime of work has been found practical: 1. Preparation of media and glassware, accurate labelling. For sugar assimilation tests, yeastnitrogen agar base, dispensed in 1ml quantities into test tubes, is melted and tempered at C. Potatoglucose agar is poured on sterile microscopic slides in 0. to 0.ml quantities and allowed to solidify for investigating filamentous growth.. Inoculation from agar slope culture. Potatoagar slides and urea broths are inoculated directly from young agar cultures. The slides should be inoculated lightly using a straight wire, while a loopful of cells are to be suspended into 0. ml of rapid urea broth (Difco). Results of the latter test can be recorded in to h of incubation at C.. Preparation of cell suspension. A cell suspension of visible turbidity (about cells/ml) is prepared in sterile water. Liquid nitrate assimilation medium is inoculated with one loopful of this suspension. Before inoculating media for sugar assimilation and fermentation tests, the suspension is supplemented by to drops of 0.% yeast extract solution.. Nitrate assimilation. The auxanographic nitrate assimilation test can be evaluated easier than tests involving a liquid medium; it requires, however, one additional petri dish per strain. A recent modification of nitrate assimilation on agar slopes () was found reliable and practical.. Glucose fermentation. The method involves observation of gas formation JOURNAL OF FOOD PROTECTION. OL. 0, MARCH 1

14 DEAK AND BEUCHAT in % glucose broth in Durham tubes. Broth is inoculated with 0.1 ml of cell suspension. Sugar assimilation. Auxanograms are prepared by pouring melted nitrogen agar base media inoculated with 0. ml of cell suspension into petri dishes. The inoculum should be wellmixed with the medium. After solidification, small amounts of selected carbon compounds are placed on the agar surface; up to five different subtrates can be tested on one plate. Incubation. All inoculated media (except those for the urease test) are incubated at to C. Readings are made on the nd, th and th days. Slow and latent reactions that would be positive within or wk are considered negative for the purposes of the simplified identification key. TABLE 1. Key of simplified identification of foodborne yeasts. Character" Positive Reaction Negative Master key 1. Maltose. Raffinose. Glucose ferm.. Cellobiose. Nitrate. Nitrate. Urease. Galactose. Galactose. Cellobiose. Urease 1. Glucose ferm. 1. Glucose ferm. 1. Cellobiose 1. Urease Subgroup 1 Subgroup Subgroup Subgroup Subgroup Subgroup 1 Subgroup 1 Subgroup 1 Subgroup Subgroup Subgroup Subgroup 1 1 Subgroup Subgroup 1 1 Subgroup 1 Subgroup 1 Subgroup I: Maltose raffinose glucose ferm. 1. Nitrate. Galactose. Acetate produced. True hyphae. Erythritol. Erythritol. Growth at C. Growth without vitamins. True hyphae. Galactose. Erythritol 1. Spores formed 1. Inositol 1. Erythritol 1. Pellicle 1. Pseudohyphae 1. Inulin 1. Galactose 1. Pseudohyphae 0. Xylose 1. Melibiose. Growth at C. Pellicle. Inulin. Pseudohyphae cellobiose, species Dek. anomala Pi. anomala Pi. subpelliculosa Pi. jadinii 1 Hypho. burtonii steatolytica 1 1 friedrichii membranaefaciens Pi. guilliermondii De. castelli Schw. occidentalis maritima Br. custersii versatilis Wi. beijerinckii Pi. fabianii 1 E'ces fibuliger 1 fennica hellenica 1 1 De. polymorphus silvicultrix Pi. ohmeri intermedia salmanticensis Klu. marxianus Pi. onychis Subgroup : Maltose raffinose glucose ferm. cellobiose, 1 species 1. Urease. True hyphae Schi. japonicus. Nitrate Sci. pombe JOURNAL OF FOOD PROTFCTION. OL. 0, MARCH 1

15 IDENTIFICATION OF FOODBORNE YEASTS Character 0. Galactose. Psuedohyphae. Galactose. Conjugating cells. Xylose. Melibiose. Cells > (Jim. Growth at C Subgroup : Maltose raffinose 1. Red colonies. Erythritol. True hyphae. Melibiose. Xylose. Starch production. True hyphae. Erythritol. Arthroconidia. Pseudohyphae. Melibiose 1. Mucous colonies Positive versatilis mogii Zy. cidri Zy. florentinus Sa. cerevisiae Sa. kluyveri glucose ferm. nitrate,1 species fusiformata Rh. acheniorum Rsp. infirmominiatum Sp. puniceus Tr. pullulans Lsp. scottii Cr. albidus Rh. ingeniosa Reaction Negative Ci. matritensis E'cop. vini Klu. thermotolerans Tsp. delbrueckii Cr. Rh. Sp. 1 macerans glutinis roseus bacarum fragariorum Subgroup : Maltose raffinose 1. Urease. Arthroconidia. Melibiose. Starch production. Red colonies. Inositol. True hyphae. Erythritol. Cells > xm. Red colonies. Starch production 1. Growth at C 1. Xylose 1. Sphaerical cells 1. Pseudohyphae 1. Growth at C 1. Galactose 1. Hyphae 1. Galactose 0. Cellobiose 1. Pseudohyphae. Erythritol. Pellicle. Mucous colonies. Erythritol. Spores >. Pellicle glucose ferm. nitrate, species Tr. cutaneum Cr. hungaricus Cr. humicolus Cr. luteolus 1 Cr. hungaricus 1 Sp. alborubescens Cr. curvatus 1 F'ella neoformans 1 Steph. ciferrii ' ces fibuliger blankii Pi. carsonii Lipo. starkeyi De. marama De. hansenii 1 Cr. flavus Bu. crocea Bu. alba Cr. laurentii 1 1 Bu. armeniaca Spori. pararoseus Rh. mucilaginosa 1 Bu. armeniaca F'ium uniguttulatum 1 0 E'cop. vini glaebosa De. hansenii glaebosa Subgroup : Maltose raffinose 1. Red colonies. True hyphae. Erythritol. Starch production. Mucous colonies. Pseudohyphae. Galactose nitrate urease, species Sp. Cr. Cr. holsaticus graminis albidus buffonii albidus Rh. Cr. 1 Bu. aurantiaca terreus tsugae JOURNAL OF FOOD PROTECTION. OL. 0, MARCH 1

16 DEAK AND BEUCHAT Character Subgroup : Maltose raffinose 1. Galactose. Cellobiose. Mucous colonies. Erythritol. Xylose. Acetate production. Trehalose. Acetate production. Erythritol. Growth at C 1 1. Sucrose 1. Glucose ferm. 1. Glucose ferm. 1. Xylose 1. Pseudohyphae nitrate urease, Positive 1 species Pi. holstii Pi. anomala Pi. silvicola Dek. intermedia Br. lambicus Pi. angusta ishiwadae 1 1 vartiovaarai Reaction b Negative 1 1 Pi. Pi. Dek. Wi. versatilis etchellsii versatilis capsulata incommunis canadensis bruxellensis californica Subgroup : Maltose raffinose 1. Urease. Arthroconidia. Cellobiose. Glucose ferm.. Growth at C. Erythritol. Acetate production. Mannitol. Erythritol. Glucose ferm.. Growth at C 1. Cells irregular 1. Rhamnose 1. Maltose ferm. 1. Cellobiose 1. Pseudohyphae 1. Glucose ferm. 1. Growth at C 1. Cycloheximide resistance nitrate galactose, 1 species Tr. cutaneum F'ium capsuligenum F'ella neoformans Cr. humicolus Br. naardensis 1 diddensiae veronae Pi. stipitis Tr. brassicae Cr. gastricus Dek. intermedia 1 Pi. media 1 1 butyri tenuis Pi. carsonii 1 0. Hyphae 1. Rhamnose. Pellicle. Lactose. Reddish colonies. Glucose ferm.. Xylose. Growth at C. Spores. Maltose ferm. 0. Conjugating cells 1. Cells > \i.m. Pseudohyphae. Pellicle tropicalis Cla. lusilaniae oleophila Pi. stipitis Me. pulcherrima hod. elongisporus albicans Zy. rouxii Sa. cerevisiae Pi. carsonii maltosa Pi. etchellsii sake Me. reukaufii 0 sake parapsilosis 1 Tsp. delbrueckii mullisgemmis catenulala Subgroup : Maltose raffinose 1. Urease. Bipolar budding. Growth at C. True hyphae. Cellobiose. Glucose ferm.. Sucrose ferm. nitrate galactose, 1 species Schi. octosporus Hsp. vineae E'ces fibuliger Hsp. osmophila mesenterica S' cop sis malanga JOURNAL OF FOOD PROTFCTION. OL. 0, MARCH 1

17 IDENTIFICATION OF FOODBORNE YEASTS Reaction b Character". Acetate production. Cellobiose. Growth at C. Glucose ferm. 1. Sorbose 1. Mannitol 1. Conjugating cells 1. Xylose 1. Acetate production 1. Cells > jlm Positive Dek. bruxellensis 1 Cla. lusitaniae Pi. toletana Zy. rouxii musae Dek. bruxellensis Sa. cerevisiae Negative E'cop. vini 1 1 insectamans Pi. rhodanensis solani Tsp. delbrueckii Subgroup : Maltose galactose 1. Red colonies. Nitrate. Raffinose. Nitrate. Raffinose cellobiose urease, species Rh. graminis Rh. mucilaginosa Cr. terreus Cr. dimennae Rh. minuta Cr. skinneri Subgroup : Maltose galactose 1. Arthroconidia. Nitrate. Xylose. Pseudohyphae. Trehalose. Erythritol. Cylindrical cells. Xylose. Cylindrical cells cellobiose urease, species Geo. fermentans Pachy. tannophilus Br. claussenii Pi. farinosa Br. naardensis Br. Klu. wickerhamii abstinens cacaoi gropengiesseri marxianus Subgroup : Maltose galactose 1. Arthroconidia. Xylose. Nitrate. Erythritol. Acetate production. Conjugating cells cellobiose glucose ferm., 1 species Geo. candidum magnoliae Pi. farinosa Eeniella nana Geo. 1 fragrans. Melibiose. Trehalose. Cycloheximide resistance. Xylose. Cycloheximide resistance 1. Xylose 1. Acetate resistance 1. Cells > xm 1. Cells > xm 1. Xylose 1. Cycloheximide resistance 1. Cycloheximide resistance 1. Raffinose 0. Growth at C 1. Growth with lysine. Trehalose ferm. Zy. cidri Zy. mrakii Klu. marxianus 1 Zy. bailii Sa. cerevisiae 1 Klu. marxianus 1 0 Sa. exiguus Sa. unisporus humilis 1 Tsp. delbrueckii Zy. florentinus Zy. microellipsoides 1 Zy. rouxii Zy. bisporus 1 1 spandovensis Tsp. delbrueckii 1 milleri Sa. dairensis Subgroup 1: Maltose galactose 1. Arthroconidia. Mannitol. Xylose. Urease. Nitrate. Raffinose cellobiose glucose ferm., 1 species Geo. candidum Spori. salmonicolor Rh. mucilaginosa Geo. capitatum Geo. fragrans 1 Rh. minuta JOURNAL OF FOOD PROTFCTION. OL. 0, MARCH 1

18 0 DEAR AND BEUCHAT Reaction" Character Positive Negative. Nitrate. Xylose. Cells > i.m. Cells triangular. Erythritol Pi. Tri. Pi. vanderwaltii humboldtii variabilis haplophila Wic. 1 domercqiae 1. Mannitol 1. Trehalose 1 zeylanoides vinaria rugosa Subgroup 1: Maltose 1. Nitrate. Pseudohyphae. Trehalose. Cells < i.m. Raffinose. Bipolar budding. Cells > i.m. Growth at C. Growth at 0 C. Sucrose ferm.. Raffinose 1. Mannitol 1. Xylose 1. Trehalose 1. Trehalose galactose glucose ferm. cellobiose,1 species Wi. mrakii Pi. minuta Wi. salurnus S' codes ludwigii Hsp. guilliermondii Hsp. occidentalis Hsp. uvarum agrestis 1 1 beechii santamariae Wi. Hsp. 1 Pi. 1 Pi. californica norvegica valbyensis norvegensis pijperi montana Subgroup 1: Maltose galactose glucose ferm. 1. Urease. True hyphae. Bipolar budding. Conjugating cells. Actidione resistance. Trehalose. Growth with ethylamine. Acetate resistance. Cells > i.m. Growth with lysine. Mannitol 1. Trehalose 1. Erythritol 1. Xylose 1. Cells > i.m 1. Nitrate 1. Erthritol 1. Cells > i.m 1. Raffinose 0. Trehalose 1. Xylose. Cells > i.m. Pseudohyphae. Cells > i.m. Raffinose. Nitrate. Xylose. Pellicle. Growth without vitamins 0. Spores hatshaped 1. Growth at C. Cells < i.m cellobiose, species Schi. malidevorans E'cop. vini Nad. elongata Zy. florentinus Zy. bailii Tsp. delbrueckii 1 1 cantarellii fructus Sa. cerevisiae 1 boidinii Sa. cerevisiae apicola 1 fructus Sa. cerevisiae Br. custersianus Sa. cerevisiae lactiscondensi Pi. fermentans Iss. orientalis Pi. kluyveri castellii Tsp. globosa Zy. rouxii Zy. bisporus Klu. delphensis Pi. pastor is 1 magnoliae 1 diversa glabrala stellata 1 0 Iss. terricola Pi. nakasei Sa. telluris Subgroup 1: Maltose galactose glucose ferm. urease, species JOURNAL OF FOOD PROTECTION. OL. 0, MARCH 1

19 IDENTIFICATION OF FOODBORNE YEASTS 1 Character" 1. Red colonies. Nitrate. Buds on short neck. Raffinose. Nitrate. Raffinose. Raffinose. Sorbose Positive Ste. nectairi Rh. mucilaginosa acuta Cr. kuetzingii pustula Reaction Negative Spori. Rh. minuta 1 Ya. lipolytica foliorum salmonkolor Subgroup 1: Maltose galactose glucose ferm. urease,1 species 1. Hyphae. Arthroconidia. Trehalose Art. javanensis. Erythritol pinus. Pseudohyphae. Mannitol. Cellobiose. Xylose. Pellicle. Spores hatshaped 1 Pi. zeylanoides norvegica delftensis. Cells > u,m vini 1. Cellobiose 1. Pellicle 1. Xylose 1. Growth without vitamins Pi. Pi. norvegensis membranaefaciens sorboxylosa ethanolica E'cop vini si/vatica 1 montana 1 Pi. fluxuum silvatica inconspicua "Names of the following substrates stand for the assimilation of them: cellobiose, erythritol, galactose, inositol, inulin, lactose, maltose, mannitol, melibiose, nitrate, raffinose, rhamnose, sorbose, sucrose, trehalose and xylose; ferm, fermentation of sugars (glucose, sucrose, maltose); urease, hydrolysis of urea. b Names and abbreviations of genera; Art, Arthroascus; Br, Brettanomyces; Bu, Bullera; Ca, Candida; Ci, Citeromyces; Cla, Clavispora; Cr, Cryptococcus; De, Debaryomyces; Dek, Dekkera; E'ees, Endomyces; E'cop, Endomycopsella; F'ella, FilobasidieUa; F'ium, Filobasidium; Geo, Geotrichum; Hsp, Hanseniaspora; Hypho, Hyphopkhia; Iss, Issatchenkia; Klu, Kluyveromyces; Lipo, Lipomyces; Lod, Lodderomyces; Lsp, Leucosporidium; Me, Metschnikowia; Nad, Nadsonia; Pachy, Pachysolen; Pi, Pichia; Rh, Rhodotorula; Rsp, Rhodosporidium; Sa, Saccharomyces; S'codes, Saccharomycodes; S'copsis, Saccharomycopsis; Schi, Schizosaccharomyces; Schw, Schwanniomyces; Sp, Sporobolomyces; Spori, Sporidiobolus; Ste, Sterigmatomyces; Steph, Stephanoascus; Tri, Trigonopsis; Tr, Trichosporon; Tsp, Torulaspora; Wi, Williopsis; Wic, Wickerhamiella; Ya, Yarrowia; and Zy, Zygosaccharomyces.. Microscopic investigations. Morphological data can be of great value in identification. Although microscopy can be tedious work, it is worth applying several times in the course of identification. Data on microscopic morphology can be collected from young agar cultures made at the time of purification of strains and from liquid cultures (e.g., glucose fermentation broth) as well as from the examination of slide cultures. Care should be taken to distinguish true and pseudohyphae. No specific test is necessary for demonstration of sexual spores when applying the simplified key. Spores can sometimes be observed during microscopic investigations, and this is then a valuable piece of information in identification. Formation of conjugation tubes, clumping of cells are often hints of sexual formation of spores. In mycological investigations of foods, yeast strains are often isolated from selective media formulated to suppress bacterial growth and/or to retard spreading of molds [e.g., dichloranrose bengal chloramphenical agar (), oxytetracyclineglucoseyeast extract agar (), acidified glucoseyeast extract agar or other general purpose media ()]. Identification should never be initiated with colonies picked directly from such selective media. Moreover, not even the purity of a colony grown on a selective medium should be taken for granted. Purification of strains must be made by repeated streaking on nonselective, general purpose media, and a wellisolated colony with similar appearance to other colonies which has been subcultivated on solid media for to h should serve as the starting culture of inoculation for identification. CONCLUSIONS Taxonomy of yeasts is being reformed by modern methods of molecular biology and biochemistry. These sophisticated methods cannot be applied in routine identification procedures; moreover, in the ecological studies of foods when a large number of strains is concerned, even traditional identification procedures requiring some JOURNAL OF FOOD PROTFCTN. OL. 0, MARCH 1

20 DEAK AND BEUCHAT 0 or more tests per strain prove impractical. Simplified identification keys can be constructed by selecting the most effective identification tests and restricting for the domain of foodborne yeasts. While keys of this kind may result in misidentification in some cases, it allows food microbiologists to more easily cope with the formidable task of identification. Characterizing yeast species will facilitate a deeper insight in the role of yeasts in the microbial associations of food. This will lead to maintaining sound good manufacturing practices, and improve processing technology and product quality. While following current trends of taxonomy, it is necessary to find practical identification methods. Simplified identification schemes save labor, materials and time. Effectiveness could be improved even more by applying more rapid methods, thus enabling one to read data within a few hours. To this end, rapid enzymatic tests are promising; however, further research is needed to compare the performance of these methods with traditional methods of identification. ACKNOWLEDGMENT This review was completed by T. D. while being supported by a Senior Fulbright Research Award. REFERENCES 1. Arx, J. A. von. 1. Propagation in the yeasts and yeastlike fungi, pp.. In B. Kendrick (ed.) The whole fungus, vol.. National Museums of Canada for the Kananaskis Foundation, Ottawa, Canada.. Arx, J. A. von, and A. C. M. Weijman. 1. Conidiation and carbohydrate composition in some Candida and Torulopsis species. Antonie van Leeuwenhoek I. Microbiol. Serol. :.. Arx, J. A. von, L. Rodriques de Miranda, M. Th. Smith, and D. Yarrow. 1. The genera of yeasts and yeastlike fungi. Studies in Mycology 1:1. Centraalbureau voor Schimmelcultures, Baarn, The Netherlands.. Ballou, C. E., D. N. Lipke, and W. C. Raschke. 1. Structure and immunochemistry of the cell wall mannans from Saccharomyces chevalieri, Saccharomyces italicus, Saccharomyces diastaticus and Saccharomyces carlsbergensis. J. Bacteriol. :1.. Barnett, J. A., R. W. Payne, and D. Yarrow. 1. A guide to identifying and classifying yeasts. Cambridge University Press, Cambridge, England.. Barnett, J. A., R. W. Payne, and D. Yarrow. 1. Yeasts: characteristics and identification. Cambridge University Press, Cambridge, England.. Barr, M. E. 1. The ascomycete connection. Mycologia :1 1.. Beech, F. W., and R. R. Davenport.. The role of yeasts in cidermaking, pp. 1. In A. H. Rose and J. S. Harrison (eds.) The yeasts, vol.. Academic Press, New York.. Beech, F. W., R. R. Davenport, R. W. Goswell, and J. K. Burnett. 1. Two simplified schemes for identifying yeast cultures. pp. 1. In B. M. Gibbs and D. A. Shapton (eds.) Identification methods for microbiologists, part B. Academic Press, London.. Berry, J. M. 1. Yeast problems in the food and beverage industry. pp.. In M. E. Rhodes (ed.) Food mycology. Hall Company, Boston.. Beuchat, L. R. (ed.). 1. Food and beverage mycology. AI Publishing Company, Westport, CT. 1. Bicknell, J. N., and H. C. Douglas.. Nucleic acid homologies among species of Saccharomyces. J. Bacteriol. 1: Bowman, P. I., and D. G. Ahearn. 1. Evaluation of the Uni YeastTek kit for the identification of medically important yeasts. J. Clin. Microbiol. :. 1. Bowman, P. I., and D. G. Ahearn. 1. Evaluation of commercial systems for the identification of clinical yeast isolates. J. Clin. Microbiol. :. 1. Buhagiar, R. W. M., and J. A. Barnett.. The yeasts of strawberries. J. Appl. Bacteriol. :. 1. Campbell, J. 1. Computer identification of yeasts of the genus Saccharomyces. J. Gen. Microbiol. :. 1. CarmoSousa, L. Do.. Distriubution of yeasts in nature, pp.. In A. H. Rose and J. S. Harrison (eds.) The yeasts, vol. I, Academic Press, New York. 1. Cooper, B. H., J. B. Johnson, and E. S. Thaxton. 1. Clinical evaluation of the UniYeastTek system for rapid presumptive identification of medically important yeasts. J. Clin. Microbiol. :. 1. Dalton, H. E., R. G. Board, and R. R. Davenport. 1. The yeasts of British fresh sausage and minced beef. Antonie van Leeuwenhoek J, Microbiol. Serol. 0:. 0. Davenport, R. R. 1. Integrated taxonomic and ecological studies of yeasts and yeastlike organisms. In E. K. Novak, T. Deak, T. Torok and J. Zsolt (ed.) Proceedings of the th Internat. Spec. Symp. on Yeasts. Keszthely, Hungary. 1. Davenport, R. R. 1. Experimental ecology and identification of microorganisms, pp.. In Identification methods for microbiologists, nd ed. Society of Applied Bacteriologists Technical Series No. 1. Academic Press, New York.. Deak, T. 1. The ecology of microorganisms on fruits and vegetables. pp.. In P. Zalewsky (ed.) Food as an ecological environment for pathogenic and index microorganisms, vol. 1. Academy of Agriculture, Szczecin, Poland.. Deak, T. 1. A simplified scheme for the identification of yeasts. In A. D. King, J. I. Pitt, L. R. Beuchat and J. E. L. Corry (eds.) Methods for the mycological examination of foods. Plenum, New York.. Fox, G. E., E. Stackebrandt, R. B. Hespell, J. Gibson, J. Maniloff, T. A. Dyer, R. S. Wolfe, W. E. Balch, R. S. Tanner, L. I. Magram, L. B. Zaplen, R. Blakemore, R. Gubta, L. Bonen, B. J. Lewis, D. A. Stahl, K. R. Leuhrsen, K. N. Chen, and C. R. Woese.. The phylogeny of prokaryotes. Science 0:.. Fuchs, P. C, and T. Dolan. 1. Performance of yeast identification systems. Am. J. Clin. Pathol. :.. Fukazawa, Y., T. Shinoda, A. Nishikawa, and T. Nakase.. Synonymy of Saccharomyces cerevisiae Hansen 1 and Saccharomyces uvarum Beijerinck 1: significance of cell wall antigens in yeast classifications. Int. J. Syst. Bacteriol. 0:.. Gams, W., H. A. van der PlaatsNiterink, R. A. Samson, and J. A. Stalpers.. CBS course of mycology, nd ed. Centraalbureau voor Schimmelcultures, Baarn, The Netherlands.. Golubev, W. I.. Novye kombinatsii drozhzhevykh gribov roda Cryptococcus. Mikol. Fitopatol. 1:.. Gorin, P. A. J., and J. F. T. Spencer.. Proton magnetic resonance spectroscopy an aid in identification and chemotaxonomy of yeasts. Adv. Appl. Microbiol. 1:. 0. Gower,.1, C, and.1. A. Barnett.. Selecting tests in diagnostic keys with unknown responses. Nature :1. 1. HesselSchmalfusz, I., M. Th. Smith, and R. Hadlok. 1. Helen und Fleisch: orkommen, Systematik und Differenzierung. Fleischwirtschaft :.. Hsieh, D. Y., and J. M. Jay. 1. Characterization and identification of yeasts from fresh and spoiled ground beef. Int. J. Food Microbiol. 1:1.. International Commission on Microbiological Specifications for Foods.. Microbial ecology of foods, vol. II. Food commodities. Academic Press, New York.. Jay, J. M. 1. Yeasts and molds in meats, poultry and seafoods. pp.. In L. R. Beuchat (ed.) Food and beverage mycology. AI Publishing Company, Westport, CT.. Jay, J. M. 1. Occurrence and significance of specific yeasts JOURNAL OF FOOD PROTFCTION. OL. 0, MARCH 1

21 IDENTIFICATION OF FOODBORNE YEASTS and molds in meats, poultry and seafoods, pp.. In M. E. Rhodes (ed.) Food mycology. G. K. Hall & Co., Boston.. Johannsen, E., J. G. Niemand, L. Eagle, and G. Bredenham. 1. Yeast flora of nonradurized and radurized minced beef a taxonomic study. Int. J. Food Microbiol. 1:1.. King, A. D., Ir., A. D. Hocking, and J. I. Pitt. 1. Dichloranrose bengal medium for enumeration and isolation of molds from foods. Appl. Environ. Microbiol. :.. Kiss, I., T. Deak, and K. Incze. 1. Microbial associations and interactions in food. Akademiai Kiado, Budapest.. Kleyn, J., and J. Hough.. The microbiology of brewing. Annu. Rev. Microbiol. :0. 0. KockovaKratochvilova, A. 1. Taxometric study of the genus Saccharomyces (Meyen/Reess) rd part: small species. Biologicke Prace,, Slovak Acad. Sci., Bratislava, Czechoslovakia. 1. Kregervan Rij, N. J. W. (ed.). 1. The yeasts, a taxonomic study, rd cd. Elsevier, Amsterdam.. Kunkee, R. E., and M. A. Amerine.. Yeasts in winemaking. pp. 1. In A. H. Rose and J. S. Harrison (eds.) The yeasts, vol.. Academic Press, New York.. Kurtzman, C. P. 1. Synonomy of the yeast genera Hansenula and Pichia demonstrated through comparisons of deoxyribonucleic acid relatedness. Antonie van Leeuwenhoek J. Microbiol. Serol. 0:01.. Kurtzman, C. P., L. J. Wickerham, and C. W. Hesseltine.. Yeasts from wheat and flour. Mycologia :.. Land, G. A., B. A. Harrison, K. L. Hulme, B. H. Cooper, and J. C. Byrd. 1. Evaluation of the new API 0C strip for yeast identification against a conventional method. J. Clin. Microbiol. :.. Last, F. T., and D. Price.. Yeasts associated with living plants and their environs, pp.. In A. H. Rose and J. S. Harrison (eds.) The yeasts, vol. 1. Academic Press, New York.. Leistner, L., and Z. Bern.. orkommen und Bedeutung von Hefen bei Pokelfleischwaren. Fleischwirtschaft 0:01.. Lodder, J. (ed.).. The yeasts, a taxonomic study, nd ed. NorthHolland Publishing Company, Amsterdam.. Lund, A. 1. Studies on the ecology of yeasts, pp. 1. Munksgaard, Copenhagen. 0. Martini, A.., and C. P. Kurtzman. 1. Deoxyribonucleic acid relatedness among species of the genus Saccharomyces sensu stricto. Int. J. Syst. Bacteriol. :0. 1. Meyer, S. A., and M. J. Phaff. 1. DNA base composition and DNADNA homology studies as tools in yeast systematics. pp.. In A. KockovaKratochvilova and E. Minarik (eds.) Yeasts as models in science and technics. Publishing House of the Slovak Acad. Sci., Bratislava, Czechoslovakia.. Meyer, S. A., D. G. Aheam, and D. Yarrow. 1. Genus. Candida Berkhout. pp.. In N. J. W. Kregervan Rij, (ed.) The yeasts, a taxonomic study. Elsevier, Amsterdam.. Moore, R. T. 1. Ustomycota, a new division of higher fungi. Antonie van Leeuwenhoek J. Microbiol. Serol. :.. Moore, R. T.. Taxonomic proposals for the classification of marine yeasts and other yeastlike fungi including the smuts. Bot. Marina :1.. Mossel, D. A. A.. Physiological and metabolic attributes of microbial groups associated with foods. J. Appl. Bacteriol. :1.. Mossel, D. A. A., A. M. C. KleynenSemmeling, H. M. incentine, H. Beerens, and M. Catsaras.. Oxytetracyclineglucoseyeast extract agar for selective enumeration of molds and yeasts in foods and clinical material. I. Appl. Bacteriol. :.. Mundt, J. O. 1. Fungi in the spoilage of vegetables, pp In L. R. Beuchat (ed.) Food and beverage mycology. AI Publishing Company, Westport, CT.. Novak, E. K.. Selecta Fungorum. Selected chapters of mycology. University of Horticulture, Budapest (in Hungarian).. Peppier, H. J.. Food yeasts, pp. 1. In A. H. Rose and J. S. Harrison (eds.) The yeasts, vol.. Academic Press, New York. 0. Peppier, H. J.. Yeasts, pp.. In M. P. Defigueiredo and D. F. Splittstoesser (eds.) Food microbiology: public health and spoilage aspects. AI Publishing Company, Westport, CT. 1. Phaff, H. J.. Structure and biosynthesis of the yeast cell envelope. pp. 1. In A. H. Rose and J. S. Harrison (eds.) The yeasts, vol.. Academic Press, New York.. Phaff, H. J., and W. T. Starmer.. Specificity of natural habitats for yeasts and yeastlike organisms, pp.. In F. A. Skinner, S. M. Passmore and R. R. Davenport (eds.) Biology and activities of yeasts. Academic Press, London.. Pincus, D. H., I. F. Salkin, I. L. Levy, N. J. Hurd, S. J. Taylor, and M. E. Kemna. 1. Comparative evaluation of potassium nitrate assimilation tests. Abstracts of the th Annual Meeting of the American Society of Microbiology, F, p. 1.. Pitt, J. I., and A. D. Hocking. 1. Fungi and food spoilage. Academic Press, Sydney, Australia.. Price, C. W., G. B. Fuson, and H. J. Phaff. 1. Genome comparison in yeast systematics: delimitation of species within the genera Schwanniomyces, Saccharomyces, Debaryomyces and Pichia. Microbiol. Rev. :1.. Rainbow, C.. Brewer's yeasts, pp. 1. In A. H. Rose and J. S. Harrison (eds.) The yeasts, vol.. Academic Press, New York.. Rhodes, M. E. (ed.). 1. Food mycology. G. K. Hall & Co., Boston.. Richards, M. 1. Brewery spoilage microorganisms. Brewers Dig. ():,.. Rocken, W. 1. Fremdehefennachweis in der obergorigen Brauerei mit dem Pantothenat agar. Monatsch. Brauwiss. ():. 0. Rodriques de Miranda, L. 1. Yeasts, pp.. In R. A. Samson, E. S. Hoekstra and C. van Oorschot (eds.) Introduction of foodborne fungi, nd ed. Centraalbureau voor Schimmelcultures, Baarn, The Netherlands. 1. Rosini, G. F., F. Federici, A. E. aughan, and A. Martini. 1. Systematics of the species of the genus Saccharomyces associated with the fermentation industry. Eur. J. Appl. Microbiol. Biotechnol. 1:.. Sand, F. E. M., and A. M. van Grinsven. 1. Investigation of yeast strains isolated from Scandinavian soft drinks. Brauwissenschaft :.. Schleifer, K. M., and E. Stackebrandt. 1. Molecular systematics of prokaryotes. Annu. Rev. Microbiol. :.. Skinner, F. A., S. M. Passmore, and R. R. Davenport.. Biology and activity of yeasts. Academic Press, London.. Spencer, J. F. T., and P. A. J. Gorin. 1. Systematics of the genera Saccharomyces, Schizosaccharomyces, Endomycopsis, Kluyveromyces, Schwanniomyces and Brettanomyces: proton magnetic resonance spectra of the mannans and mannosecontaining polysaccharides as an aid in classification. Antonie van Leeuwenhoek J. Microbiol. Serol. :1.. Tsuchiya, T., Y. Fukazawa, M. Taguchi, T. Nakase, and T. Shinoda. 1. Serological aspects of yeast classification. Mycopathol. Mycol. Appl. :1.. Walker, H. W., and J. C. Ayres.. Yeasts as spoilage organisms. pp.. In A. H. Rose and J. S. Harrison (eds.) The yeasts, vol.. Academic Press, New York.. Walker, W. F. 1. S ribosomal RNA sequences from Ascomycetes evolutionary implications. Syst. Appl. Microbiol. :.. Walker, W. F., and W. F. Doolittle. 1. S rrna sequences from eight basidiomycetes and fungi imperfecti. Nucl. Acid Res. :0. 0. Walt, J. P. van der, and D. Yarrow. 1. Methods for the isolation, maintenance, classification and identification of yeasts, pp.. In N. J. W. KregerRij (ed.) The yeasts, a taxonomic study. Elsevier, Amsterdam. 1. Woese, C. R., and G. E. Fox. 1. Phylogenetic structure of the prokaryotic domain: the primary kingdoms. Proc. Natl. Acad. Sci. USA :000.. Yamada, Y., M. Arimoto, and K. Kondo. 1. Coenzyme Q sys JOURNAL OF FOOD PROTECTION, OL. 0. MARCH 1