A history of research on yeasts 4: cytology part I, 1890–1950

Most of our knowledge of the behaviour, fine structure and composition of the cell wall, mitochondria and vacuoles of yeasts is of very recent date. By contrast, the nucleus of the yeast cell has long been the subject of voluminous and controversial literature. (Matile, Moor and Robinow, 1969147 p. 274.) Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151 Improvements in microscopy . . . . . . . . . . . . . . . . . 152 Improvements in stains . . . . . . . . . . . . . . . . . . . . . . 152 Nuclei . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157 Ascospores, cell fusion and sexual reproduction . . .163 Ballistoconidia (ballistospores) . . . . . . .169 Chlamydospores. . . . . . . . . . . . . . . . . . . . . . . . . . . .172 Dimorphism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .174 This first part of the fourth article in a series on the history of research on yeasts,1,2 describes the work on yeast cytology up to about 1950. During this early period of its development, cytology depended greatly on the use of stains. The structures of cells and their organelles, nuclei, vacuoles, mitochondria and so forth, were described, often incorrectly, before their functions were understood and before phase contrast, interference and electron microscopy came into general use. Cytological research, in general, intensified greatly during the 1870s and 1880s. Major factors for this were advances both in the optics of microscopes and in techniques of staining cells and tissues. At that time, ‘ …︁ it was not uncommon for the leading cytologists, most of whom worked in German laboratories, to publish up to seven papers a year. This may well be the first branch of biology which attracted research on …︁ a modern scale’ (105 p. 61). Perhaps it also provides an early example ofexcessive haste to publish many ill-considered papers. Many of the older publications give confused descriptions of the nature of the yeast cells' organelles, such as nuclei and vacuoles, as well as of the identity of specialized cells, such as ascospores or chlamydospores. Despite these confusions, some attempt is made here to give a coherent account ofthe gradual and uneven emergence of understanding of yeast cytology. Original drawings and photographs, published from 1890 onwards, are reproduced in this article to show how parts of the cells, as seen with the microscope, have been interpreted and, also, to indicate some of the problems of interpretation which the earlier workers faced. Table 1 lists chronologically some of the main developments in the early history of cytology in general that must have had a considerable impact on yeast cytology in particular. Table 2 summarises some of the findings of yeast cytology during the same period. Early in the nineteenth century, the availability ofmicroscope objectives corrected for chromatic and spherical aberrations, and which consequently gave clearer images than before, had been of major importance for early microbiology25. The end ofthat century saw further improvements in the resolution that could be obtained (Figure 1). In 1878, Ernst Abbe,3 having established the relation between the resolution of an objective, the refractive index of the medium between lens and object, and the angular aperture of the lens, demonstrated the improved performance of lenses of high numerical apertures4 1. Improvement of maximum resolution given by microscopes during the nineteenth century (data from Turner208) Additional advances came from improvements inthe manufacture of the glass used in lenses. Collaborating with Abbe, and aided by grants from thePrussian government, Schott5 investigated the effect of introducing different oxides into vitreous fluxes. As a result, Schott and Abbe developed newoptical glasses, such as borate glass, enabling Abbe to make the first apochromatic6 objectives, that is, objectives with the same focal length for blue, green and red light2. By 1886, the Zeiss factory had produced oil-immersion7 apochromatic objectives, designed by Abbe, with a focal length of2 mm and a numerical aperture of 1.4 (36, p. 265, 194, 195). Although dyeing with aniline has found extensive use in the textile industry, it has proved to be of only limited use for microscopical techniques, despite the histologist being much wealthier than the dyer.11 However, by 1902, Heidenhain12 could say categorically that some of these dyes had facilitated cytological research97, one of the earliest uses of an aniline dye as a biological stain having been by Beneke13 in 1862 (41 p. 78). Ranvier's14 treatise of 1875 on histological technique lists six aniline dyes175. The method of staining …︁ grew and grew, till to be an histologist became practically synonymous with being a dyer, with this difference, that theprofessional dyer knew what he was about, while the histologist with few exceptions did not know, nor does he to the present day (144 p. 190). A thorough knowledge of the chemical properties of different types of stains and of the circumstances under which they most readily combine with different types of substances would be of great value in cytology, but at present the chemistry of these processes is not sufficiently well defined to justify a departure from the empirical rules which will be found in most text books of microscopy (78 p. 42). The warning was justified; but the fuss about what was and was not an ‘artefact’, which was prominent throughout the first half of the twentieth century, was not. All preparations made for examination under the miscoscope are ‘artefacts’. Barely visible, usually colourless structures are first fixed in solutions which dehydrate and often shrink them and dissolve all the lipids of the cells. On the other hand, there was also some misplaced scepticism, such as that of Baker18 who, in astandard work he published in 1945, describes the‘Golgi element’ as ‘elusive’. He writes that a dense lipoidal substance is present, but that it can be seenonly in specially treated cells: ‘it is not usually possible to be certain whether the appearance is that of an artifact’19 (18 p. 13). In a personal communication, Baker described a conventional method of revealing the Golgi apparatus20 as like throwing mud at something and hoping that it sticks. Unfortunately the different new names which Schwarz has proposed for the different components of the cell, for which Strasburger and others have already suggested many superfluous names, have been accepted to some extent, even though this troublesome accumulation of names only serves to make communication between authors more difficult.27 Today, when precise details of the cell cycle become familiar to students in their first year, it is difficult to imagine the bafflement of the early microscopists. But they had little useful guidance from previous work and often embellished their accounts of what they had seen with products of, sometimes, unfettered imagination. For decades, one source of confusion and debate was what happens to the nucleus when a cell divides. Did it disappear, to be reconstructed in each new cell after division? Carl von Nägeli was among some eminent microscopists who were led bywhat they saw through the microscope ‘to speculate about what lay beyond its power of resolution’ (87 p. 334). Indeed, he supported an earlier suggestion that the nucleus disappears after its function has been discharged (93 p. 113). …︁ I have investigated the beer yeast, Saccharomyces cerevisiae …︁ With haematoxylin staining, I succeeded in identifying in each cell a single spherical nucleus. It is almost in the middle of the cell next to the large vacuole …︁ 29 But, as Matile and his colleagues have pointed out, the yeast nucleus proper ‘has no marked affinity for haematoxylin, and it seems likely that what Schmitz regarded as the nucleus was the readily stainable nucleolus’30 (147 p. 225). Only the question of the yeast nucleus remains controversial; the small size of the cells, the strong affinity of the cytoplasm for stains, together with the presence of many different substances disseminated in the cell and capable of taking up the stains, makes the differentiation of the nucleus extremely difficult.35 Marie Antoine Alexandre Guilliermond (from Le Laboratoire de Cryptogamie, Muséum National d'Histoire Naturelle, Paris: by kind permission) Let us take a culture …︁ and put it in fresh must.Immediately, we will observe in the nucleus aradical transformation. From the start, thenucleus becomes vacuolar, while the protoplasm remains homogeneous. Fig. 1 [see our Figure 4] gives an excellent example of this first transformation.45 F. A. Janssens. Photograph, kindly supplied by G. L. Hennebert, from the Archives of the Botany Unit, Faculty of Sciences, Catholic University of Louvain, Louvain-la-Neuve Drawings of cells of Saccharomyces cerevisiae (from Plates I and II of Janssens and Leblanc108) Budding cells of Saccharomyces cerevisiae, showing vacuole and nucleus with a dark nucleolus, in typical position near the vacuole, and pale chromatin area, traversed by a spindle-fibre (Helly fixation, acid fuchsin staining; photomicrograph magnification ×4500, by C. F. Robinow) Alexandre Guilliermond's drawings, made in 1914 (see footnote 46), of meiosis in the ascus of Schizosaccharomyces octosporus show the spindle fibre and spindle pole bodies (Guilliermond's Figures 5, 6 and 7 85) Janssens and Leblanc wrote: ‘ …︁ the use of Heidenhain's staining method …︁ has always given us irreproachable preparations.’46 Unfortunately, however, Heidenhain's iron haematoxylin does not stain yeast chromatin,47 so the parts of yeast cells which these authors had called ‘nuclei’ were, in fact,nucleoli. A good example of such a nucleolus is in their Fig. 1 (shown here as a part of Figure 4). On the other hand, excellent complete nuclei, mostly unstained, with large stained nucleoli can be seen in their Fig. 51 (also reproduced here inFigure 4). For comparison with Janssens and Leblanc's drawings, a modern photomicrograph of budding cells of Saccharomyces cerevisiae showing nucleus, nucleolus and vacuole is reproduced in Figure 5. The ‘dancing’ volutin bodies in the vacuole were examined more carefully by Henneberg48 in 1916. He called them vacuolar bodies (Vakuolkörper) (99 p. 51), and described their rapid movements in the vacuole. He decided that volutin was probably not a reserve material, observing that when an Amoeba consumes yeast cells, the volutin accumulates in the Amoeba and so is not easily digested. During mitosis in most plant and animal cells, thenuclear membrane is disrupted and segregation of the chromosomes involves the formation of a bi-pyramidal spindle, built mainly of microtubules.49 These tubules drive the poles of the spindle apart and draw the chromosomes toward the poles. However, the nuclear membrane of Saccharomyces cerevisiae does not break down during mitosis and a long spindle is formed, of a single microtubule116. In 1966, Robinow and Marak published a paper181 on an intranuclear fibre (i.e. spindle) in the nucleus of Saccharomyces cerevisiae (see Figure 5), John Marak's electron micrographs having shown the fibre to be composed of microtubules. Towards the end of their paper, they point out that this wasnot the first time a fibre had been reported in yeast nuclei (181 p. 149). In 1917, Guilliermond had described50 and clearly illustrated examples of such a spindle fibre (fuseau) in fixed and stained meiotic nuclei of Schizosaccharomyces octosporus85. Three of his illustrations, reproduced here in Figure 6, show the fibre extending throughout the length of the nucleus, which is undergoing meiosis. These spindle fibres in Guilliermond's drawings can be compared to those in some of Robinow's phase contrast photomicrographs of Schizosaccharomyces japonicus, an example of which is given in Figure 7. Such intranuclear fibres are found regularly in the dividing nuclei of Schizosaccharomyces pombe and S. japonicus. Robinow's phase contrast photomicrographs of two successive stages of mitosis in Schizosaccharomyces japonicus, showing spindle fibres (×2800, from Fig. 17 of180) Differences between the nucleus of Saccharomyces cerevisiae and those of most eukaryotes account for the contradictory nature of many publications on this subject up to the nineteen fifties. The yeast nucleus is not always spherical and during life stands out from the cytoplasm far less clearly than its often larger companion, the vacuole. The nucleolus does not float in the centre of the nucleus and is either crescentic or angular and brick-shaped. It tends to remain close to the part of the nuclear envelope adjoining the vacuole. Most effective, moreover, in misleading microscopists hasbeen the lack of affinity of yeast chromatin fornuclear stains. It is the nucleolus, not the chromatin, that is deeply stained by Heidenhain's haematoxylin. Affinities for stains are reversed if fixed yeast is subjected to Feulgen style hydrolysis with M-HCl and then subjected to Giemsa's stain, atechnique Robinow used in 1942, following Piekarski169, to demonstrate the nucleoids of bacteria179. Using this procedure, the chromatin of the yeast nucleus is deeply stained, while the nucleolus is barely visible. Although the genetical rôle of chromosomes was not established until after 1900, in 1848 Wilhelm Hofmeister51 had published illustrations of ‘Klumpen’ (that is, lumps), which were the large mitotic chromosomes52 of a Tradescantia sp.103; and, early in the 1880s, Flemming had produced clear descriptions of animal chromosomes during nuclear division69. In 1903 there was a crucial development. After giving evidence that Mendel's laws could be explained by the behaviour of the chromosomes in meiosis, Sutton53 wrote: ‘We have seen reason …︁ to believe that there is a definite relation between chromosomes and allelomorphs …︁ ’ (205 p. 240). Yet, even as late as 1921, having been visiting T. H. Morgan54 in America, Bateson55 felt it necessary to reaffirm ‘that chromosomes are definitely associated with the transferable characters’, that is, the genes132. Because yeast chromosomes are difficult to observe, even by electron microscopy (147 p. 289), different numbers of chromosomes have been attributed to various yeasts, including Saccharomyces cerevisiae. The haploid number is currently thought to be 16, but four chromosomes, presumably considered to be the diploid number, were reported for this species in 1905 both by Swellengrebel207 and by Fuhrmann,56 who published convincing drawings of its mitotic chromosomes (Figure 8)73. Nonetheless, despite these findings, confusion continued about the identity of the nucleus and other organelles. Fuhrmann's drawings of mitosis in cells of Saccharomyces (ellipsoideus) cerevisiae73. The cells were fixed in 1% PtCl4 + glacial HAc +2% OsO4 (15:1:3)101 and stained with eosin; but cell 16 is stained with alizarin and cell 18 with methylene blue Boveri58 expressed the idea that the nucleus, andnot the protoplasm, is the really significant part of the egg in matters of heredity (139 p. 181). In order to decide whether the nucleins or the histones or the protamines are of importance for the hereditary qualities, it would be necessary to decide whether the nuclei of the eggs of one form [of animal] contain always the same base as that found in the sperm of the same species …︁ (139 p. 180). …︁ I have not felt justified in coming to any definite conclusions regarding the cytology of yeasts, based upon the preparations at my disposal, but neither do I consider the results of other cytological investigations to be sufficiently satisfactory to justify a clear interpretation of thechromosome mechanism in yeasts (224 pp. 87–88). Lindegren's illustration of mitosis in the yeast cell, published in 1946 (Fig. 5 of136) The nuclear vacuole contains the chromosomes and the nucleolus …︁ The wall of the nuclear vacuole does not break down at any time in the life cycle; it is a permanent structure (135 p. 278). Diagrams of the yeast cell 1910–1969. (A) From Wager and Peniston of 1910; 1, nucleolus; 2, peripheral layer of chromatin; 3, chromatin patch on one side of nucleolus; 4, nuclear vacuole; 5, central volutin granule in the vacuole; 6,chromatin network; 7, granules of fatty substance; 8, volutin granules; 9, glycogen granules (216 p. 76). (B) Fig. 1 of Lindegren's paper of 1952135. (C) From Matile, Moor and Robinow of 1969, that is, after the application of phase contrast and electron microscopy: ER, endoplasmic reticulum; F, filament; G, Golgi apparatus; L, lipid granule (sphaerosome); M,mitochondrion; Mt, thread-like mitochondrion; N, nucleus; Nc, centriolar plaque; Nm, nuclear membrane; Nn, nucleolus; Pi, invagination; vacuole; cell (Fig. 1 The of of the nucleus of the yeast cell, which has in a series of during the is not definitely of the central vacuole of the cell as a vacuole; including consider this structure with to the nucleus (135 p. Table summarises many of the of yeast nuclear cytology between and It seems that which can are to …︁ than those which are such is the biological of this In this one of the seems to be on the Budding and fusion of by C. in of the (A) Saccharomyces cerevisiae. was on beer at The series of four are then after and contains four two the is the same 18 budding having the of four budding is a cell with two drawings show its after and in the between the two has almost and they have show two and for In the two in an ascus Fig. was in beer and the were obtained by to of of budding as in each series showing fusion clearly Fig. 5 may have been the fusion as some form of but his that and do is probably the first of an effect of in a In was when about Schizosaccharomyces the ascus and to and involves no sexual The next year, however, that of the same yeast were after the fusion of two but he did not any cytological the fusion or its he seems to have cells or stains, on the of and that he had not what happens to the In 1900, gave evidence of the fusion of nuclei at as staining was with in yeasts, but did not that the was In with we have nuclear after before formation of the as Janssens and Leblanc they in must this to The Saccharomyces we have show no of T. in (from drawings of formation by on for at Fig. The cells were fixed with acid + PtCl4 + and stained with stain and in cell with two cells formation of ascospores Guilliermond's drawings of in Schizosaccharomyces stained with Heidenhain's haematoxylin 1, of Guilliermond's for his The yeast of Saccharomyces a marked to which is not only from the large number of but also by the fusion of by nuclear In contrast, considered that can be any that and of one of Guilliermond's drawings, reproduced here in Figure after Guilliermond's 1905 was with of Saccharomyces cerevisiae, various other names were reported for about of fusion of ascospores when they were about to For one only so that one must for a long time in a to Guilliermond's published in of of ascospores of Saccharomyces cerevisiae (from Guilliermond's Fig. 6 In suggested that and illustrated p. it is to draw to whether it does or does not after previous of two cells, whether this or or whether is …︁ there remains no that in species may …︁ the Saccharomyces Saccharomyces the idea of ‘ …︁ no ascus to most yeast and in in Saccharomyces cerevisiae and S. of single had produced haploid which Such produced few or no The work was published in and did not become until when it was by the who colleagues who in a which is not well p. the in they In the there only a haploid of very limited size …︁ after which the cells to This can in a large number of diploid which at to with larger cells …︁ there is to indicate …︁ that the may form p. Three of for one on a microscope (Fig. 1 were made by F. C. glass in pp. by of the of showing cell in Saccharomyces cerevisiae and the products of those of the (A) two ascospores I and cells and cells and are to the cells have 7 the two ascospores have Fig. (B) the ascus two of which the as a dark in the base of the ascus in each I evidence drawings and some of which are shown in Figure The in published however, did not and in having diploid not while with a haploid a fusion of or to be a p. it is not to in ‘ …︁ the in yeast cytology in this of being in a p. Indeed, the present understanding of yeast cytology has from the use of and interference as microscopy the from ascospores to be given to the in size and of ascospores in of the most (Figure is that of most of the cells of this budding yeast about the ascospores may in cells and ascus with ascospores of was for 7 on at The by with an used interference contrast with a A part of the cell on typical or when are discharged into the by of a Ballistoconidia of by and in The yeast was in beer for a was stained and with a 2 mm Zeiss apochromatic drawings of the formation and of these are shown in Figure but their of these cells was not as clear as that in the much earlier paper of and in a of the of and his colleagues described the …︁ developed from cells at the of of yeast each at its a which at is discharged into the p. on species of and in described with such yeasts, a the discharged on the they pp. Three all these he the and of under a microscope and illustrated the with excellent drawings, shown in Figure drawings show a the is to its and has suggested that in the the and of the formation and of of drawings by A. H. and of of on a single by was at about stages in the of a and the first after its development, a became in at the end of which time a would to at its of a from its to about it on its for and was then before a of at the point of of the and in the size to about that of the 6 and The and were together for a or of about stages in the of the and of the Fig. are cells, of the yeasts, and C. In and published a of the yeast that of which they described which seems to us to the (Figure This had been proposed by de in for cells with However, were called as by in who describes and their (Figure He that his well with the idea that the is a which has an of two of the of published by and in The yeast was for in an of 1% 1% the cells were stained with Fig. of of in (Fig. 7 of of Unfortunately, many either to and stages in the from cells to …︁ or use the p. In a paper on

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.