A history of research on yeasts 8: taxonomy

Taxonomy is divisible into three parts: first, there is the sorting of individuals into likes and unlikes; this is pigeon-holing or classification. Next, there is the labelling or naming of the groups sorted (nomenclature); the third part is the comparison of the unknown with the known and the identification, where possible, of the unknown with the previously recognized and named specimen, group, or population (Cowan, 1970 63 p. 146). … the only truly scientific foundation of classification is to be found in an appreciation of the available facts from the phylogenetic point of view (Kluyver and van Niel, 1936 128 p. 369). Taxonomy is built upon the basic fields of morphology, physiology, ecology and genetics. Like other scientific disciplines it is a synthesis of many kinds of knowledge, theory, and method, applied in this case to the particular field of classification. Its potentialities and its limitations are largely those of the basic fields whose raw material it utilizes (175 p. 3). Criteria for classifying yeasts have included: (a) the sizes and shapes of their cells; (b) the structures of the cell walls; (c) the modes of vegetative reproduction; (d) whether or not they reproduce sexually and, if so, their mode of sexual reproduction; and (e) their ability to utilize various exogenous compounds. In addition, from the 1970s, characteristics of their DNA and RNA have become increasingly popular criteria. As new laboratory techniques have been introduced and more kinds of yeasts have been discovered, their names and classification have changed. Moreover, further changes have come about as criteria for assessing evolutionary affinities have developed. Another factor in producing instability of nomenclature has been ignorance of earlier publications, so that some yeasts have been described more than once and given different names. The taxonomically correct name for any organism is generally its earliest name to have been published which accords with the requirements of the International Code of Botanical Nomenclature,5 to which yeast taxonomists profess to adhere. For example, according to the Code, a Latin description must be given for any taxon published on or after 1 January 1935, otherwise the name is nomen nudum [naked name] but may be 'clothed' and, hence, validated later by publishing a Latin description. Names published on or after 1 January 1958 must indicate the 'type'. ('Type' is here used as the Greek word , in the sense of 'the original pattern'). Dead, dried type material of the original organism must be deposited in a publicly accessible herbarium. Freeze-dried living cultures of yeasts may also now be recognized as types. The type of a genus is a species and the type of a species is a specimen or culture. This article differs from those previously published in this series on the history of yeast research, which were chiefly concerned with experimental work and with the evidence for the validity of the conclusions from that work. Although they undertake experiments, yeast taxonomists are largely concerned with the way they give order to the various kinds of yeast and by what criteria they should group them—into classes, orders, families, genera and species. Such decisions, generally more a matter of scholarly endeavour than of experimental science, are necessarily arbitrary, often depending on the whims of the taxonomists. The description given by Kützing is in fact so indefinite, and the yeast-like nature of the organisms described by him so doubtful, that it is not permissible to use this name to refer to a yeast genus.15 Jacomina Lodder in 1956. Photo courtesy of F. W. Lodder Nelly Kreger-van Rij. © Delft Microbiology Archives; http://www.beijerinck.bt.tudelft.nl The oily plasma liquid concentrates around the nucleoli; slight granulations appear at its surface and are soon replaced by a true membrane; the rounded cell is formed in this way; during this time, the membrane of the mother-cell becomes very thin, very transparent. When the cells thus formed, to the number of one, two or three in each elongate cell have attained the size of cells which float freely at the surface of the liquid, the very stretched membrane of the mother cell ruptures, and the round cells are freed, carrying with them a part of the mother cell's membrane destined to disappear little by little …22 Emil Christian Hansen's27 (Figure 3) development of techniques for obtaining pure yeast cultures in the early 1880s100 (described in the third article of the present series23) made yeast taxonomy a practical proposition and, by the end of the nineteenth century, reports of about 200 species had been published, and the identity of about 90 of these is known today (Table 1). Guilliermond's splendid, pioneering book on the yeasts, published in 191296, names a further five species. Ascospore-forming yeasts, such as Saccharomyces species, were isolated notably from industrial fermentations, whereas many of the non-ascospore producing yeasts, such as Candida albicans and Malassezia furfur, were found in clinical practice, often the putative causes of mycotic diseases. Emil Christian Hansen, about 1875. Reproduced by courtesy of Carlsberg, Copenhagen Between 1893 and 1911, a number of new ascospore-forming yeast genera were described (Table 2). Paul Lindner's description of Schizosaccharomyces pombe in 1893158 was the first report of a yeast which reproduced vegetatively by fission and not by budding. Yeasts with different kinds of ascus were found and each kind of yeast tended to have a characteristic number of ascospores in each ascus; the yeasts also formed various kinds of bud and filament. And in 1894, Martinus Beijerinck described another Schizosaccharomyces species, the eight-ascospored Sz. octosporus, and held that this yeast should not be considered as a Saccharomycete39. Anastasia Grigor'evna Konokotina. Photo courtesy of V. I. Golubev Zygosaccharomyces has characteristic sexual copulation of the cells. Saccharomycodes was distinguished from Saccharomyces species as (a) it has lemon-shaped cells, bipolar budding with a wide isthmus and a transverse septum between daughter- and mother-cell and (b) it forms one to four spherical ascospores which fuse pair-wise within the ascus104. Saccharomycopsis: Holger Schiönning28 (Figure 5) gave this yeast a new genus, as it forms both buds and septate hyphae (see Table 2). He found that the ascospore wall has two layers, the outer of which (the exosporium) detaches from the spore during germination217. Pichia was based on Hansen's previously described non-sugar-fermenting Saccharomyces membranaefaciens101, which has hemispherical or irregularly angular ascospores. Hansen's Pichia membranaefaciens became P. membranifaciens (140 p. 319). Willia anomala was Hansen's new name for Saccharomyces anomalus, which he, himself, had described in 1891103 (see Table 2) and which Beijerinck had also named Saccharomyces acetaethylicus in 189238. Finding that the genus Willia had, unknown to Hansen, already been used by Müller in 1889 for another kind of fungus, Paul and Hans Sydow29 renamed the genus Hansenula in 1919 (234 p. 44). In its turn, nearly all the Hansenula species were moved to the genus Pichia in 1984138. Nematospora: Vittorio Peglion30 isolated from hazel nuts (Corylus avellana) this budding yeast, Nematospora coryli197 with its elongate and tapering ascospores, one end of which forms a fine thread (see Table 2), the Greek word nema () meaning a thread. Monospora, as the following chronology shows, provides a fairly typical story of the vagaries and confusions in the naming of yeasts. 1884 Elias Metschnikoff gave the name Monospora bicuspidata to a yeast which lives in the body cavity of the branchiopod crustacean Daphnia magna176. 1899 Because the name Monospora had already been used for some algae of the Florideae, Feodor Kamenski31 renamed Metschnikoff's yeast Metschnikowia bicuspidata116, 117. 1913 Genkel substituted Metschnikowiella for Metschnikowia83 because the latter name had been used for the genus of a sponge. 1920 Unaware of the renaming of Monospora, David Keilin drew attention to the previous use of Monospora for an angiosperm by Hochstetter (151 p. 292) as well as for an alga. Keilin accordingly renamed the genus Monosporella and drew attention to further confusion: The species Monospora, discovered by Metschnikoff, was described by him under the name M. bicuspidata, and under this name it is referred to in his various publications, nevertheless all the authors I have consulted (Zopf, Hansen, Dangeard, Guilliermond, Lafar, Saccardo)32 wrongly name the species M. cuspidata Metschnikoff. I do not know who changed the specific name, but incline to the view that the error may have arisen through a misprint or misquotation, none of the authors mentioned having apparently referred to Metschnikoff's original papers (118 p. 83). 1952 In their major work on yeast taxonomy, Lodder and Kreger-van Rij accepted Keilin's generic name Monosporella169. 1962 Nicolau van Uden33 accepted Kamenski's Metschnikowia because, as a sponge is an animal, use of the same name for a yeast is permissible251. As van Uden commented subsequently180, none of the Metschnikowia species was isolated in pure culture by their authors, their descriptions being based on morphological observations of the yeasts in the hosts on which they were parasitic. This was so until van Uden, himself, obtained pure cultures of M. krissii and M. zobellii from marine sources253. Holger Ludvig Schiönning. Reproduced by courtesy of Carlsberg, Copenhagen In 1907 Lindner isolated a filamentous, fermentative ascospore-forming yeast causing 'chalky bread'34162. Studies of this species, which Lindner called Endomyces fibuliger, changed many mycologists' perceptions of the evolutionary relationships of yeasts. Lindner's E. fibuliger formed abundant branched septate hyphae (Table 2) as well as budding, typically yeast-like, cells. Asci, containing two or four hat-shaped spores, were often formed at anastomoses between hyphal cells. Since Saccharomyces species did not form hyphae, Lindner revived Reess's genus, Endomyces35, considering it to be a link between Willia (later called Hansenula or Pichia) and some ascosporogenous moulds. In 1924, Albert Klöcker36 transferred E. fibuliger to Saccharomycopsis, of which genus he gave a valid description (121 pp. 298–301). However, the genus Saccharomycopsis generally lapsed into desuetude until Kreger-van Rij restored it in all its glory in 1984131. One sees that all attempts up till now with a view to the transformation of moulds into true yeasts have failed and that they seem to constitute an independent group of fungi.37 … I look at the term species, as one arbitrarily given for the sake of convenience to a set of individuals closely resembling each other, and that it does not essentially differ from the term variety, which is given to less distinct and more fluctuating forms. The term variety, again, in comparison with more individual differences, is also applied arbitrarily, and for mere convenience sake (68 p. 52). … the placement of different species in synonomy because adequate techniques were not used, is unfortunate (264 p. 320). However, these are not abstract species, which are usually laid down according to various combinations of any inherited signs not connected with the conditions of life of their possessors. Instead of them with the help of the same morphological methods which were used earlier there have been revealed completely real groups of organisms distinguished among themselves within the bounds of each genus by the conditions of life, the habitat occupied by them and the generality of the specific adaptations to these conditions.40 Often … the most commonly used system has proven inadequate for delineating natural species (204 p. 188). … many of the yeast species described in the literature represent an amalgam of distantly related organisms, while other species have been separated on trivial grounds from closely related taxa. The situation is particularly acute among species placed in the imperfect genera (204 p. 162). … in future … such terms as 'related' and 'relationship', with their phylogenetic connotations, [should] be replaced by less pretentious ones that imply no more than degrees of resemblance ('similar', 'similarity')250. If two organisms are related, they must retain in their genomes base sequences that are descendent from a common ancestral base sequence; closely related organisms will have retained a greater proportion of base sequences in common than organisms that have widely diverged … members of the Enterobacteriaceae … considered by experienced taxonomists to constitute well-defined species usually shared at least 70–80% similar DNA sequences … (204 p. 162). Lynferd Wickerham. Photo courtesy of C. P. Kurtzman Vladimir Ilyiich Kudryavtsev. Photo courtesy of M. R. Ushakov As the study of yeast nucleic acids progresses, so the interpretation of precise ancestral interrelationships alters. In any case, the dividing line between one species and another must necessarily be arbitrary and must depend on a study of their present-day characteristics, including their nucleic acids. A large-scale study of 26S D1/D2 ribosomal sequences of 500 species145, has shown that there are fewer than 1% nucleotide substitutions in strains considered to be of the same species. This '<1%' rule for species has been widely accepted, although it is not always applicable. The D1/D2 domain of Clavispora lusitaniae, for example, is polymorphic, with more than 6% variation between two mating type strains150. Hence, identification of a species should encompass thorough studies of the whole organism, physiology, structure and so forth, as well as variations in DNA sequences. To conclude, some confusions can be avoided by using 'species' to mean simply the lowest principal rank in the nomenclatural hierarchy, consisting of a generic name and a specific epithet, decided on by a competent taxonomist. Thus Saccharomyces cerevisiae is a species; so is Saccharomyces carlsbergensis, although most yeast taxonomists no longer accept the latter species. Species are groups of actually or potentially interbreeding natural populations, which are reproductively isolated from other such groups173. Hybrids between Saccharomyces species obtained by David Yarrow273. Reproduced by courtesy of D. Yarrow, by permission of The Netherlands Academy of Sciences The gradual finding that many yeasts thought to be asexual can in fact reproduce sexually has caused nomenclatural confusions, which must be unintelligible to those not familiar with the oddities of fungal taxonomy. Fungal taxonomists accept that a given yeast or other fungus may exist both in a state (often called the 'perfect' state or stage or 'teleomorph') which produces sexual 'spores' such as ascospores, and in an asexual ('imperfect') state, or 'anamorph', without sexual reproduction (107 pp. 437–438). Astonishingly for other biologists, these two states are classified in different genera, so that, for example, the asexual state of Kluyveromyces lactis is Candida sphaerica32. Sensibly, G. L. Hennebert and L. K. Weresub proposed that the two states could be combined nomenclaturally, as the 'holomorph'110, which could keep the same name as the teleomorph. Few hyphae, prostrate, breaking up into shorter or longer pieces. Conidia, arising by budding from the hyphae or on top of each other, are small and hyaline.45 It was only partly because Berkhout's dissertation was not widely known (it was cited by Langeron and his colleagues) that there was a subsequent proliferation of generic names proposed for asexual filamentous yeasts, such as Blastodendrion192, Myceloblastanon193, Geotrichoides, Mycocandida, Mycotoruloides153, Mycokluyveria60. And as the genera proliferated, so did the new combinations and seemingly endless lists of synonyms. Writing from the Faculté de Médecine of Paris in 1932, Langeron and Rodolfo Talice46 published a paper on classifying those fungi which characteristically formed both filaments and yeast-like cells. This paper was largely a microscopical study of the different sorts of cell produced by each kind of yeast: blastoconidia, chlamydospores, the mode of budding and the greatly varied appearance of filamentous growths. The authors included 27 figures of clear drawings which illustrate the many formations of filaments. Analysis of these characteristics led Langeron and Talice to classify their organisms into eight genera, which today would be put into only four genera (Table 3) and most of them would be called Candida albicans. This would inevitably lead to confusion and justified irritation among the increasing number of workers in various fields who use or encounter yeasts of this group. We have therefore maintained Candida and Torulopsis as separate genera (252 p. 897). The 'food yeast', Candida utilis, provides an example of nomenclatural confusion. In 1926 Wilhelm Henneberg47 found this yeast in several German yeast factories, having been cultivated (without a systematic name) during World War I for food and fodder. He named the yeast Torula utilis (109 pp. 56–59). However, unknown to Henneberg, as the name Torula was already used for some moulds200, Augusto Berlese48 had, in 1895, substituted the generic name Torulopsis for Torula43 and, accordingly, Lodder had adopted the name Torulopsis utilis in 1934 (166 p. 144). Now, in their book of 1952, Lodder and Kreger-van Rij separated filamentous from non-filamentous genera and somewhat extended the meaning of 'filamentous'. Hitherto, for a species to be classified in the genus Candida, it was necessary for it to form pseudohyphae or true hyphae with blastoconidia.49 With this new system, Candida included yeasts which produce only simple pseudohyphae, that is groups of cells, each remaining to its thus a so the species was renamed Candida utilis p. 1926 Torula Saccharomyces 1934 Torulopsis Hansenula 1952 Candida Candida asexual state of Hansenula Pichia Table some genera of particular which were published at various up to the early part of the The following the of some other genera which were thought to be and In isolated from such as and an yeast to which he gave the name to indicate its and not as a He the characteristic and of to the of this kind of yeast, as at the same time, of the in The story that the name was to the of is as to the of the p. In and used as a generic name for two new species they isolated from and and commented that they had characteristics in common with van and van ascospore in hence, became the asexual state of a species to a new genus, that is D. A in the classification of budding asexual yeasts was the in of the genus by Albert and van of which they to those of filamentous authors strains of or yeasts which formed of cultures and by that the from the of cells were to which they called and were to their by small or and van of this by the is reproduced as of article part in this that this was also characteristic of the the group of which and and van that, had affinities to in the sense of of ancestral In described two new species, for which he the genus yeasts from having cells and spherical to (Figure drawings of of and species. state of state of as yeasts had been known pure culture methods had been yeasts had, the been to a number of genera, such as and In and who was on yeasts that were with placed those to in a new genus, This genus included thus the species which had isolated from the of in pp. and In described a or yeast and called it Saccharomyces Since ascospores were not found in this species, in moved it to his new genus However, as this name was already in the name for these yeasts in but it to in and retained Reess's specific epithet, species of were found later to be asexual states of with K. the of isolated the yeast, from in This yeast is both and cells, the of each depending on and other This is another genus with only one species, named after those who isolated it from and described by in This yeast has bipolar budding, each bud connected with its mother-cell by a wide thus from those called Torula species, which was at that (Figure of bipolar budding, with wide between and their with the of courtesy of Guilliermond's for yeasts, which in The criteria he used for identification included the appearance of vegetative cells, the or of ascospores and their number and and also the ability to (Table It is to be that the of recognized and species, by the authors in will not with Moreover, it be that in several very can be However, the can only be as in the present state of yeast because this will lead to a more precise of the species in © Delft Microbiology © Delft Microbiology In the first of these with the ascosporogenous yeasts is well the of yeasts to the group, the However, that the techniques available to for whether a yeast can form ascospores that a yeast is to do The methods are (a) p. (b) an containing only about as or (c) the surface of or The that a particular is can only be as of in the ability to the is If the is only it is always that under conditions no could be and for this the or not of the particular be as a in such A only in and A in all from Lodder p. 1934 with permission from and Lodder used another of which had because was and gave the same is to and (Table which was by its with apparently that is or fungi As in many the authors combined into a species yeasts that had been considered as distinct species and moved some species between For example, became a of D. Zygosaccharomyces species were considered from those of so the genus was and all its members were into the genus was to yeasts, such as E. Pichia which could utilize the type species was E. which had and has to Saccharomycopsis Lodder and Kreger-van Rij published a classification for all yeasts known at the and, until the of taxonomy, subsequent were based on major and publications, not from the with Lodder and Kreger-van were Taxonomy of which with the genus Hansenula largely on studies of from and book on the of all ascospore-forming yeasts, published in in authors their yeasts well and were both were to the that they were a phylogenetic classification and both introduced a of as than had been used described a new set of which are used in to this and all these were later produced and this is of their previously strains from genera of yeasts on on of with which he his for Table lists these and also those used by Kudryavtsev. In Lodder published the of a who included most of the yeast taxonomists of that time, more than three as many as those for the first classifying them in genera and species. by of the and by use was made of to on which yeast was being as of for and to of them for The genera were classified into four (a) (b) (c) and (d) other yeasts (Table In most the genera were distinguished an was that and Torulopsis were by a the ability of the to use as the of for Such in the ability to use various for or with methods which had