Portal de Periódicos da CAPES

A history of research on yeasts 14: 1 medical yeasts part 2, Cryptococcus neoformans

2010; Wiley; Volume: 27; Issue: 11 Linguagem: Inglês

10.1002/yea.1786

1097-0061

James A. Barnett,

Plant Pathogens and Fungal Diseases

This rare, life-threatening, opportunistic, iatrogenic mycosis [cryptococcosis] has been recorded more frequently since the 1960s because of the use of aggressive immunosuppressive therapies and the sudden appearance of AIDS in the 1980s (Edouard Drouhet,2 1997) [80, p. 10]. This organism which is highly pathogenic for men and animals has evoked a great deal of interest. Many cases of disease which led to the isolation of the organism were reported and, as appears from the long list of synonyms,3 many unwarranted names have been given to it (Jacomina Lodder4 and Nelly Kreger-van Rij, 1952) [151, p. 375]. The name Cryptococcus neoformans may refer to a group of seriously pathogenic yeasts, which can infect any organ of the human body, notably causing meningoencephalitis. The names of the members of this group have varied from time to time; Table 1 lists 43 of their names and synonyms and Table 2 gives their most recent nomenclature. Of necessity, this article returns several times to consider these various names. Fungal taxonomists have a system of nomenclature, dating back to 191015, as follows: when sexual reproduction is discovered in a fungus previously thought to be asexual, the new sexual form is placed in a different genus from that of the asexual form, even though they may differ in no other way. Accordingly, when in 1975 June Kwon-Chung, at the National Institutes of Health, Bethesda, found sexual reproduction in Cr. neoformans, she named this sexually reproducing yeast Filobasidiella neoformans130. However, as this yeast appears to be asexual outside the laboratory, herein it is called Cryptococcus neoformans; besides which the yeast is generally known by this name, and the illnesses it causes are called 'cryptococcosis'—not 'filobasidiellosis'. For many years, cryptococcosis was called 'torulosis',5 since the causative organism had been named Torula6 neoformans by Joseph Weis7 in 1902230 and Torula histolytica in 1916213. In 1894 and 1895 Otto Busse (Figure 1), professor of pathology at Greifswald in Germany, described a dangerous pathogen taken from a lesion of a woman's tibia and, as he considered the organism to be Saccharomyces-like, he called the disorder 'Saccharomycosis hominis'49. Busse's yeast (Figure 2) was also reported from the same material by Busse's surgeon-colleague, Abraham Buschke47; but neither author gave it a name. Coincidentally, in the same years (1884 and 1885), Francesco Sanfelice (Figure 3), at the University of Cagliari, Italy, isolated from fermenting fruit juice a yeast194-196 which he called Saccharomyces neoformans [195, p. 241], being aware that it was comparable to Busse's pathogen [197, p. 468]. Thus, right from the start of this history, Sanfelice's finding indicated that this yeast is not an obligate human pathogen but is also to be found living elsewhere. Then, to continue the story, in 1901 Jean-Paul Vuillemin (Figure 4), at Nancy University in France, named Busse's yeast Cryptococcus hominis and renamed Sanfelice's yeast Cryptococcus neoformans [228, pp. 737 and 747, respectively], because neither yeast formed ascospores, as would be expected of a member of the genus Saccharomyces.8 In addition, Sanfelice isolated another yeast, this time from the lymphatic system of an ox (Bos taurus), naming it Saccharomyces litogenes [196, p. 524]. (Lodder and Kreger-van Rij designated S. lit[h]ogenes a synonym of Cr. neoformans in 1952 [151, p. 374]). And at about this time, around 1900, there were several clinical reports of finding yeasts resembling Cr. neoformans (e.g.35, 43, 68, 182, 227). Otto Busse (1867–1922)126 Otto Busse's drawings, published in 1895, of the cells of Filobasidiella neoformans. These cells had been isolated from various lesions in a 31 year-old woman. Figures 11 and 12 show the encapsulated cells49. Busse's excellent illustrations of these yeast cells, stained in situ in a section of part of the tibia and lung, are reproduced in No. 3 of the current history series [20, p. 369] Francesco Sanfelice. Courtesy of Marianna Viviani Jean-Paul Vuillemin (1861–1932)186. Reproduced courtesy of the Société Mycologique de France The pathogenic yeasts, which do not sporulate,9 possess generally the characteristics of the genus Torula and may be considered as belonging to this genus. However, Vuillemin has given them the generic name of Cryptococcus. This name is in general use, so we will adopt it.10 In 1934, Lodder149 and the author28, independently, reported studies in which the strain of Busse and Buschke was compared with strains from torula meningitis and found indistinguishable. This showed that, whereas one name, blastomycosis, had been used for several distinct diseases, one disease, cryptococcosis, had been described under several names … and there is now a vast literature on the subject [29, p. 1301]. Indeed, Benham's recommendation was adopted in 1952 by Lodder and Kreger-van Rij in the first major taxonomic study of all known yeasts151 and a more recently published work shows that there have been at least 49 synonyms for this species [21, p. 381]. It was Benham who, in 1935, had provided good evidence that a single species was responsible for cryptococcosis28. Another advance, which was important for developing effective physiological and genetic studies of Cr. neoformans, was the devising in 1949 of a chemically defined liquid medium, containing glucose, ammonium and inorganic salts and thiamine, which was suitable for growing this yeast199. Given the11 importance of the capsule in cryptococcal disease, tremendous effort has been applied in recent years to understanding its biology (Indrani Bose et al., 2003 [39, p. 655]). In 1976, Mercedes Edwards and colleagues published a thorough electron micrographic study of the cells of Cryptococcus neoformans, illustrating the capsule, cell wall, plasma membrane, nucleus, nucleolus, nuclear membrane, vacuoles, endoplasmic reticulum, mitochondria and ribosomes85. The micrographs showed that cells of different ages differ in the thickness of the cell walls (Figure 5). Cryptococcus neoformans: electron micrograph of two cells of different ages, published in 1967 by Mercedes Edwards and colleagues [85, p. 772]. The older cell (right) has a thicker cell wall and a thicker capsule than the cell on the left of the photograph. ca, capsule; cw, cell wall; er, endoplasmic reticulum; m, mitochondria; n, nucleus; pm, plasma membrane. Bar corresponds to 1 µm. Reproduced with permission from the American Society for Microbiology © 1967 Cells enveloped by capsules of polysaccharide have long been known to be characteristic of yeasts of the genus Cryptococcus [21, p. 451; 151, p. 371] and thought to be important for the virulence of Cr. neoformans59, 138 in addition to its ability, unlike other species of Cryptococcus, to grow at 37 °C (see e.g.21). Busse's illustrations in his paper of 1895 (Figure 2) show cells with thick cell walls and, in 1895 and 1896, Ferdinand Curtis12 (Figure 6) explicitly described these capsules and published clear drawings of them (Figure 7). His yeast, which he called 'Megelococcus myxoides', was isolated from a patient with meningitis70, 72. Consistent with Curtis's observations, in 1917 Harry Swift13 and Lionel Bull14 published an account of a capsulated yeast that caused cerebro-spinal meningitis215 and in 1930 Arthur Henrici15 published an excellent drawing of a section through meninges16 infected by Cr. neoformans ('Torula histolytica') showing 'wide capsular spaces surrounding' the yeast cells, the capsular material 'evidently secreted by the yeast and not formed by the tissues' [116, p. 228] (Figure 8). At that time, Benham not only described the capsules and included drawings of them (Figure 9) in her important and influential paper of 193528, she also used the capsules (removed from the cells by treating with HCl) to produce agglutinins in rabbits, thus beginning the antigenic characterization of this species. Ferdinand Curtis (1858–1937). Courtesy of Boualem Sendid Drawing of cells of Cryptococcus neoformans by Ferdinand Curtis in 189672, showing large encapsulated cells in IV.1 to IV.6. Curtis's notes: 'I culture on agar after 48 h at 37 °C; II culture in a solution of acid peptone and sugar; III culture 1 1/2 months in a solution of acid peptone; IV forms with thick capsules, IV.1 cell with chromatin granules, c refringent cell, IV.2 cell filled with chromatin granules, IV.3 and IV.6 cells where chromatin granules surround a clear central space, IV.5 cell contains only two small chromatin granules; a cell wall, b mass formed by small chromatin granules disseminated in the cytoplasm, c refractile particle—food reserve, d clear protoplasm, f developing bud, g capsule, h homogeneous unstained space' Drawing, published by Arthur Henrici in 1930, of a section through the meninges (membranes enveloping the central nervous system) infected with Cryptococcus neoformans, the cells of which vary in size and are surrounded by thick capsules [116, p. 228] Rhoda Benham's illustrations of Cryptococcus neoformans, published in 1934 (A–C) [27, p. 387] and 1935 (D–F) [28, p. 264]. (A) A colony; (B) budding cells (those on shaded background in Indian ink mount); (C) thick-walled cells surrounded by a capsule; (D) as (B); (E) also as (B), but thick-walled cells with capsule too; (F) cells with capsules in tissue from an inoculated rat. (A–C) From Archives of Dermatology 30 (1934), p. 387, copyright © 1934 American Medical Association. All rights reserved In the course of a study of the growth requirements of different yeasts, it was unexpectedly found that cultures of Torulopsis rotundata [now Cryptococcus albidus] gave a steel-blue colour reaction with iodine198. This finding led these authors to study capsules of 25 yeasts, including Cr. neoformans, and they found the capsules were formed when the yeasts were in media below pH 5, which favoured growth. The polysaccharide of the capsules (which was liberated into the suspending medium), detected by the blue colour it gave with iodine, included both amylose20 and a pentosan8, 157. The authors suggested that the starch reaction with iodine could be used as an aid in identifying the causative agent of cryptococcosis ('European blastomycosis'21)156. … that the major antigens on the surface of encapsulated Torula cells are polysaccharides … [169, p. 105]. Also in the USA, between 1949 and 1960, Edward Evans23 and his colleagues published a series of papers extending Benham's antigenic studies. Using tube agglutination tests24, they found three serotypes, A, B and C, based on antigenic differences in the capsular polysaccharide92-95, thus providing the first clear evidence that Cr. neoformans was taxonomically heterogeneous. Nearly 21 years after Evans' initial discovery, a group working at the National Institutes of Health, Bethesda, demonstrated a fourth serotype, D. This was done when investigating 106 isolates of Cr. neoformans by agglutination and absorption studies with anti-cryptococcal sera237 and the use of serological methods for diagnosing cryptococcosis was reviewed in 1964200. At the time Evans was making his fundamental contributions to the knowledge of the serotypes of Cr. neoformans, Drouhet with his fellow workers at the Institut Pasteur in Paris found the capsule polysaccharide of serotype A to contain xylose, mannose and some uronic acid bound to mannose82. They also reported that the polysaccharide inhibited leucocyte migration in vitro81, which observation received significant support more than 40 years later from the work of Zhao Dong and Juneann Murphy79. Knowing that there is a high concentration of the capsular polysaccharide in the serum of infected patients77, Dong and Murphy injected mice with the antigen of Cr. neoformans and found that it did, indeed, inhibit the migration of leucocytes. Further evidence of the rôle of the capsule polysaccharide in protecting Cr. neoformans was given in 1967 and 1968, when Glenn Bulmer and DeLaine Sans at the University of Oklahoma published photomicrographs, reproduced in Figure 10, of in vitro phagocytosis of Cr. neoformans cells by human leucocytes44. These workers isolated non-encapsulated (avirulent) mutants and found that more mutants than capsulated cells were taken up by the leucocytes: the higher the concentration of capsular material, the fewer yeast cells were taken up by the leucocytes (Figure 11)45. Photomicrographs of phagocytosis of non-encapsulated Cryptococcus neoformans by human leucocytes in vitro, published by Glenn Bulmer and DeLaine Sans in 1967 [44, p. 1482]. (A–D) Incorporation of a yeast cell into a leucocyte; (E) leucocytes, each containing several yeast cells; (F) a leucocyte containing about nine yeast cells. Reproduced with permission from the American Society for Microbiology © 1067. Note: this figure is reproduced for its historical interest, despite its poor quality Inhibition of phagocytosis in vitro of non-capsulated mutants of Cryptococcus neoformans by suspensions of capsular material. Results of Glenn Bulmer and DeLaine Sans, published in 1968 [45, p. 6]. Reproduced with permission from the American Society for Microbiology © 1968 The medical importance of the Cr. neoformans capsule was clinched once and for all by the finding in 1994, by Yun Chang and Kwon-Chung, that capsule-deficient mutants were markedly less virulent than the wild-type (as had been reported previously in 1982103) and that complementation of the capsule-deficient mutant restored virulence57. These findings established that the characteristics of the capsule were clearly well worth investigating. In 1961, Toshio Miyazaki of the Tokyo College of Pharmacy published three papers on the chemical structure of the capsular polysaccharide, finding that it consisted of D-xylose, D-glucuronic acid and D-mannose in the ratio 1 : 1 : 3161, 162. He proposed a structure for the polysaccharide, shown in Figure 12A163, having a mannan backbone with branches of D-xylose and D-glucuronic acid. Twenty years later, Apurba Bhattacharjee and colleagues at the National Institutes of Health in Bethesda published further structural studies32-34, and Figure 12B shows the structure they suggested for the polysaccharide for serotype D32. Earlier studies of the capsular material involved chemical degradation and techniques of paper, column or gas–liquid chromatography and mass spectrometry. Then, in the 1990s, Robert Cherniak and his colleagues at Atlanta, Georgia, applied nuclear magnetic resonance to identify the structures more precisely63, 220, 221. Their description of the serotype B polysaccharide, glucuronoxylomannan (GXM), is represented in Figure 12C and that of galactoxylomannan (GalXM), which constitutes about 7% of the capsule, in Figure 12D. Structures described for the capsular polysaccharide of Cr. neoformans: (A) by Toshio Miyazaki in 1961 [163, p. 830]; (B) by Apurba Bhattacharjee, June Kwon-Chung and Cornelis Glaudemans in 1979 for serotype D32. (C, D) Are from the descriptions by Robert Cherniak and his colleagues in 1991 and 1998, respectively: (C) glucuronoxylomannan of serotype B220; (D) galactoxylomannan221. Residues are represented as follows: D-Gal, D-Galactose; D-GlcA, β-D-glucopyranosyluronic acid; D-Man, D-mannopyranose, D-Xyl, D-xylopyranose For Cr. neoformans to bring about extensive invasion of cerebral tissue (disseminated meningoencephalitis), the yeast cells must cross the blood–brain barrier (BBB)64. Experimenting on mice, Caroline Charlier and her colleagues in Paris have recently found that crossing the BBB occurs at the capillaries of the brain cortex, severely damaging the fine vessels and this invasion is associated with changes in the structure of the yeast capsules60. In 2009, Oscar Zaragoza and his colleagues reviewed what is known about the capsule of Cr. neoformans238. The capsular components are major virulent determinants and are: (a) glucuronoxylomannan (GXM), 90% of the capsule; (b) galactoxylomannan (GalXM), which has important immunological properties; and (c) mannoprotein. GXM and GalXM have harmful effects on the immune response and their molecular structures have not been fully worked out. Melanins are dark-brown to black pigments found in animals, plants and microorganisms. These pigments are not essential for growth and development, but rather they enhance the survival and competitive abilities of species in certain environments [26, p. 411]. In 1962, Friedrich Staib reported that Cr. neoformans produces brown colonies when grown on media made from seeds of the composite plant, Guizotia abyssinica, or from various bird droppings183, 209, 210, whereas six other species of Cryptococcus do not do so. Hence, this pigment production has seemed to be a suitable characteristic for identifying Cr. neoformans61, 62, 127, 173, 203, 214. Ten years after Staib's finding, Carol Shaw and L. Kapica of McGill University suggested that the pigment is a melanin203,25 having found the pigment to be produced on media containing extracts of potatoes, which were known to contain both tyrosine, a precursor of melanin, which is the substance of potato blackening [166, p. 118]. In 1895, working with mushrooms (Boletus spp.), Émile Bourquelot26 and Gabriel Bertrand showed that 'tyrosinase' oxidized tyrosine to form a dark pigment31, 41. There is now extensive knowledge about many of the complexities of melanin formation, because knowledge of its biochemistry is interesting medically1, as it is the dark brown pigment of the hair, skin and eyes. In brief, L-tyrosine is a precursor of melanin: 'tyrosinase' or 'phenoloxidase' [either catechol oxidase (EC 1.10.3.1) or monophenol monooxygenase (EC 1.14.18.1)] (a) converts tyrosine to L-3,4-dihydroxyphenylalanine (L-dopa) and (b) oxidizes L-dopa further to the reactive intermediate L-dopaquinone, followed by reactions leading to the formation of melanin (Figure 13)144. Pathway of the oxidation of tyrosine to melanin: a simplified diagram, based on an illustration by Aaron Lerner144 In summary, C. neoformans has a phenoloxidase enzyme system that catalyzes pigment production from structurally diverse phenolic substrates.The pigments generated from L-dopa and epinephrine28 have been shown to be melanin. The biochemistry of melanin and the assembly of melanin on the cell wall remain poorly understood. Melanogenesis is interesting because of its association with virulence and because it is a potential target for antifungal drug design [52, p. 91]. Various publications have reviewed fungal melanins26, 50, 120 and quite recently Sols and his colleagues published evidence that a very different pathogenic yeast, Candida glabrata, undergoes reversible switching between phenotypes that include dark brown, light brown and white forms, the dark brown form colonizing most readily mouse spleen and liver208. Cryptococcus neoformans var. neoformans. Cryptococcus neoformans var. grubii [sexual state of both (a) and (b) called Filobasidiella neoformans]. Cryptococcus gattii [sexual state called Filobasidiella bacillispora]. In order to be certain about the varietal status of each C. neoformans strain, one cannot rely on its serotype alone. Unfortunately, the literature is full of inaccurate statements such as 'serotype A strains are classified in var. grubii and strains of serotype D are classified in var. neoformans' [140, p. 585]. Analysis of GXM structure is complicated by variability on several levels. One instructive study involved isolates from separate episodes of cryptococcosis in individual patients. In several cases, the strains remained the same … but alterations in GXM structure resulted in their assignment to different serotypes [39, p. 656]. C. neoformans strains are capable of sexual reproduction. Sexual reproduction has been observed only in the laboratory, and its role in pathogenesis is uncertain (Arturo Casadevall and John Perfect, 1998 [52, p. 11]). … the rapid loss of this trait [hypha production] following culture on artificial media suggests that many such strains may have been overlooked … [98, p. 33] (see also154, 155). Jean Shadomy's photomicrographs of clamp-connections of Cryptococcus neoformans, published in 1970201 Photomicrograph of hyphae of Cryptococcus neoformans, published in 1966 by Jean Shadomy and John Utz202 Working at the Medical College of Virginia, Shadomy's findings were conclusive and consistent with the suggestion of Takashi Nakase and Kazuo Komagata, based on their analysis of the nDNA (GC content), that Cryptococci are heterobasidiomycetes30168 or hymenomycetes. And in the 1970s Kwon-Chung provided further evidence that this yeast is basidiomycetous: subjecting the hyphae to electron microscopy, which revealed the presence of dolipores31 in the hyphal septa (Figure 16)139. This finding provided further evidence that this yeast is basidiomycetous, as dolipores are characteristic of basidiomycetous fungi164. Three dolipores of Filobasidiella neoformans. Indicated by arrows, the dolipores are centrally in the septa between hyphal cells. Electron micrograph of an ultrathin section made by June Kwon-Chung and Terry Popkin in 1976 [139, p. 525]. Reproduced with permission from the American Society for Microbiology © 1976 Cells of the opposite mating type fuse to form a heterokaryon, which almost immediately develops into dikaryotic hyphae. Hyphal cells possess unfused MATa and MATα nuclei and clamp connections typical of basidiomycetes. Paired nuclei divide in synchrony until a basidium forms at the hyphal tip. Karyogamy occurs in the basidium followed by meiosis, and sporogenesis occurs at the basidial apex. The original four nuclei resulting from meiosis remain in the basidium, whereas repeated post-meiotic mitosis generates four long chains of up to 40 spores [159, p. 208]. Filobasidiella neoformans: June Kwon-Chung's illustration in 1975 of basidia with terminal basidiospores130. Reprinted with permission from Mycologia, © 1975 The Mycological Society of America Filobasidiella neoformans: June Kwon-Chung's illustration in 1980 of nuclear migration in the formation of basidiospores at the apex of each basidium [133, p. 421] Filobasidiella neoformans: June Kwon-Chung's illustration in 1984 of the life cycle [137, p. 475] Genetic analysis of C. neoformans over the past two decades has revolutionized what we know about its life cycle and virulence properties. As we proceed, the value of having defined the sexual cycle will continue to be high. Although scientists have found the C. neoformans sexual cycle to be invaluable, it is much less clear what its value is to C. neoformans. Even though it can be seduced to mate in the laboratory, there is limited evidence that C. neoformans undergoes sexual reproduction in nature (Christina Hull and Joseph Heitman, 2002) [119, p. 604]. Kwon-Chung laid the foundations for research on the genetics of Cr. neoformans in the 1970s, when she found that (a) crossing serotypes A and D produces Filobasidiella neoformans130 and (b) crossing B and C produces F. bacillispora132. In both cases, she described the control of mating type by two alleles, MATa and MATα, at one locus131. This analysis assumed special medical interest when, in 1978, working with John Bennett, Kwon-Chung reported the predominance of α over a in those isolates from clinical sources (about 30 : 1 for 233 cases; the ratio was about 40 : 1 for 105 non-clinical isolates—most from pigeon excrement or soil)134. Chester Emmons32 had isolated Cr. neoformans from soil and from pigeon (Columbia livia) nests and excreta in the 1950s89, 91. From the 1990s onwards, efforts have been made to identify and isolate the genes which control the mating system. The α-mating type locus of Cr. neoformans has been found to contain a gene which encodes a pheromone33165 and, having identified the mating types, it became practicable to obtain by back-crossing haploid congenic serotype D strains115, which could be used conveniently for genetic studies. In 2005, the genome of Cr. neoformans was sequenced and found to span 14 chromosomes153. Table 4 summarizes many of the gene isolations, mostly in the 1990s, of genes of Cr. neoformans, which is in many ways similar genetically to Saccharomyces cerevisiae and Schizosaccharomyces pombe119. The various auxotrophic mutants of Cr. neoformans, obtained in the 1970s and 1980s, such as those for lysine or pantothenate121, have been useful for studing the life-cycle231. Since 1967, a number of the mutants studied have concerned the formation of the capsule (e.g.46, 57). Other mutants concerned enzymes which are important for the yeast's physiology, such as orotate phosphoribosyltransferase51, 84, which is involved in nucleotide synthesis146; and Anja Forche and her colleagues have produced a linkage map, using amplified fragment length polymorphisms96, while Robert Marra and others constructed a linkage map by crossing two strains and analysing 94 progeny for 301 markers158. Table 5 indicates some of the niches of Cr. neoformans (other than Man) which include various mammals, from horses223 to koalas128. Although this yeast has been found mainly associated with animals, it was isolated from the juices of various fruits in 1894194, 195 and 100 years later from eucalyptus trees in Australia88 and in Spain24. Cr. neoformans has been reported many times as occurring in pigeon droppings (e.g.90), as well as inside the pigeons themselves216. Such observations are notable because pigeons have high body temperatures of > 40 °C160 and Cr. neoformans will grow at such temperatures [21, p. 382], whereas other Cryptococcus species do not do so [21, pp. 60–63] and are not reported as being found in pigeons [21, pp. 289–326]. See also177. Since the early 1980s, Cr. neoformans has become highly significant clinically as a cause of deaths of immunosuppressed patients, especially those with AIDS. Hence it is not astonishing that between 1980 and 1995 the numbers of papers published about this yeast increased exponentially to more than 300 a year (Figure 20). This mass of publications—PubMed lists more than 9000 since 1947—has necessitated a great deal of severe selection, both of the papers and of the topics covered, in writing this article.34 As with Candida albicans17, the vast majority of these papers were written with reference to the clinical significance of the work they described. And, since much of the most interesting research done on Cr. neoformans has concerned its capsule, because of its various effects on the human systems of defence against invading microbes, a large part of this article has discussed the capsule. Increase in numbers of publications concerning Cryptococcus neoformans in the second half of the twentieth century. The numbers of publications for each 5 year period were taken from PubMed. Note: (a) transplanting organs clinically, employing long-term immunosuppression, took off after 1962, when azathioprine [6-(3-methyl-5-nitroimidazol-4-yl)sulfanyl-7H-purine] was used to prevent rejection [167, p. 1445]; (b) the early history of the AIDS epidemic is given in reference118 Because C. neoformans is one of only a few microorganisms that can cross the BBB [blood–brain barrier] and invade the brain parenchyma, defining the molecules responsible for this process will further an understanding not only of cryptococcosis, but also of the general mechanisms operating during meningitis [147, p. 75]. It is only proper to end this conclusion by paying respectful tribute to the outstanding contribution to the study of Cr. neoformans of June Kwon-Chung, of the National Institutes of Health at Bethesda. Since describing its sexual reproduction in 1975, her name has appeared among the authors of at least 100 publications which have been concerned with this yeast, covering its taxonomy, physiology, genetics, molecular biology and pathology. It is a pleasure to thank the following for their help and kindness: Nick Hopwood, Boualem Sendid, Michael Vinegrad and M. A. Viviani. However, none of them is responsible for faults in the article; neither is L. K. Barnett, who has done much to improve both the writing and the illustrations. I also thank the Royal Society for a research grant.