A highly infectious ‘mycoplasma’ that inhibits meiosis in the fungus Coprinus

We have discovered a cytoplasmically inherited infectious agent that inhibits meiosis in a lower eukaryote, a species of Coprinus. We believe, from the work described below, that this agent is a mycoplasma with a very interesting host-parasite interaction.

Mycoplasmas cause, or are suspected of causing, many chronic diseases of plants and animals, including man (Hayflick, 1972; Sharp, 1970; Smith, 1971; Bové & Duplan, 1974; Chen & Liao, 1975; Williamson & Whitcomb, 1975), yet little is known about the basis of their pathogenicity or the mechanisms of infection. Some of the reasons for this lack of knowledge are the small size (ca. 0·2-1·5 μm), the relative difficulty of obtaining mycoplasmas in pure culture and the difficulty of fulfilling Koch’s postulates in determining their causal relationship with disease symptoms. Mycoplasmas have been implicated in at least 21 human clinical diseases, but only one species has been unequivocally proven to be the cause of a human disease (Hayflick, 1972). Other reasons preventing a clear understanding of the nature and role of mycoplasmas in disease are the complexities of their hosts, with the numerous cell and tissue types preventing the identification of the site of malfunction, and the fact that mycoplasmas, in general, seem to give rise to multiple pathological symptoms (Smith, 1971).

Our mycoplasma is h ighly infectious, has only one pathological effect–the inhibition of meiosis–and the host is a relatively simple organism, a basidiomycetous fungus that possesses only one basic cell type for most of its life-cycle. This report is the first to deal with the biology of a fungal mycoplasma and is only the second report of the existence of mycoplasmas in fungi (Heath & Unestam, 1974), and thus extends the host range of mycoplasmas to another kingdom (Whittaker, 1969).

The host organism, Coprinus congrégalos Fries 1838, designated in our laboratory as Cop I, was isolated from steer manure and grown in pure culture in 1971. It grows well on Emerson’s YpSs agar (Difco). Single-spore-derived clones have been maintained by routine transfer by cutting plugs out of mycelia with a no. 1 cork borer. Details of host structure and life cycle are given below (see Results and Observations).

Infectivity experiments were carried out by placing a plug of a clone carrying the infectious agent inverted on the centre of a fully grown recipient mycelium. At intervals a series of plugs was taken from the recipient mycelium along a radius and transferred to fresh media where such plugs were allowed to form a mycelium and sporulate. Such sporulating cultures were scored for infectivity as described below.

Extracts were made from mycelia grown in liquid YpSs for 6–7 days at 26 °C, harvested by filtration through cheesecloth and washed twice with cold 0·05 M Tris-glycine buffer, pH 8·5 (Ross, Martini & Thoman, 1973). The washed mycelium was pressed dry and the cake weighed and suspended in 3:1 vol/wt cold buffer. The suspension was homogenized in a Virtis 45 homogenizer at 25 000 rev/min for 4 min at 0 °C and then centrifuged at 1000g for 10 min to remove cell debris. The supernatant was filtered by suction through a sterile 0·22-μm Millipore filter. The filtrate was used to infect test mycelia by placing 0·3 ml on the test mycelium and then slashing the hyphae through the drop with a double-bladed instrument so that all cutting was uniform. The treated clones were incubated in the dark for 3–4 days before plugs were removed from the unslashed parts of the mycelium and crossed with appropriate partners. Subsequent mushroom formation was scored for infectivity.

Preliminary light microscopy had shown that there was a correlation between the size of mushroom primordia and the state of meiosis. For electron microscopy mushrooms were harvested and fixed at approximately 5, 10, 15, 20, 25 and 30–40 mm stages. Small pieces of gill tissue were removed and fixed for 2 h in 3 % glutaraldehyde in 0·1 m cacodylate buffer, pH 7·0. The material was washed in 4 changes of cacodylate buffer (30 min/wash) and then postfixed in 1 % osmium tetroxide for 2 h. The material was further washed in 3 changes of distilled water. All of these procedures were carried out at 4 °C. The material was dehydrated in an ascending series of acetone, followed by propylene oxide and embedded in Spurt’s resin (Spurt, 1969). Thin sections were cut with a Porter-Blum MT-2 ultramicrotome using a diamond knife and stained with uranyl acetate and lead citrate (Reynolds, 1963). The preparations were examined with a Siemens Elmiskop I electron microscope at an accelerating voltage of 80 kV.

Since this is the first detailed report of the biology of a fungal mycoplasma, a description of the host and the location of the affected cells may be warranted. Figs, 1 and 2 show the life cycle and important stages in this species of Coprinus. The organism was isolated as a dikaryon mycelium and allowed to sporulate. Seventeen monokaryon (haploid) clones were grown from single-spore isolates and these were crossed in all combinations of 2 to determine the incompatibility system of this species (Raper, 1966). The mating results indicated that the species had a bipolar compatibility system and the symbols a_x and a₂ were assigned to the 2 mating types present.

This fungus has certain restrictions and abilities that are important to an understanding of the infectious agent. The monokaryon state is functionally the haploid state and is wholly vegetative and unable to differentiate into mushrooms. No monokaryon mycelium has produced mushrooms since first isolation. Only the dikaryon, the functional diploid state with 2 nuclei of different mating type in each cell, is capable of differentiation. As shown in Fig. 2, the dikaryon is established by the somatic fusion (anastomosis or somatogamy) of vegetative cells of compatible monokaryons and the subsequent migration of the nuclei from one monokaryon into the other. Migration is usually reciprocal in this species. No nuclear fusion occurs, and the 2 nuclei in each cell remain haploid. Monokaryons of the same mating type, and therefore sexually incompatible, will still undergo anastomosis and establish cytoplasmic continuity, but the nuclei do not migrate.

Differentiation of the dikaryon mycelium requires a maturation of at least 4 days after transfer followed by a light stimulus necessary to induce further development. Dark-grown cultures do not differentiate. After light induction, a further 5–8 days are required before differentiation can be detected by the formation of mushroom primordia. The time between light induction and primordial initiation can be reduced by the application of exogenous glycine (Ross, unpublished). The primordia enlarge quite rapidly and develop into mushrooms. These mushrooms produce localized areas, gills, on which premeiotic cells differentiate from the genetically identical ground tissue (Fig. i b). After meiosis, spores are formed (Fig. 1 c) that rapidly turn black from the synthesis of dark pigments. The abundance of spores on each gill and the thinness of the overlying tissue cause each mature mushroom to become jet black (Fig. 3, right), the normal dark. If meiosis is inhibited, spores do not form and the resulting mushroom retains the pale, golden brown colour of the cap tissue, the pale form (Fig. 3, left). In this species, as in most Coprinus species, the mature mushrooms undergo a programmed autolysis after spore formation.

During the initial mating-type determination crosses, it was noted that all compatible crosses produced dark mushrooms with one exception. All compatible crosses with clone 8 produced mature, autolysing mushrooms that were pale. Examination showed that spores were almost completely absent from the pale mushrooms, there being fewer than 10 tetrads per gill compared to the normal count of ca. 2 × 10³ mm ⁻².

An experiment to determine whether the cause of the paleness was due to a nuclear gene or to some other factor is shown in Fig. 4. A 9-day-old mycelium of clone 16 was inoculated in the centre with an inverted plug of clone 8 (a₂). Every 2 h plugs were taken on 0·5-001 centres along a radius and plated out separately on fresh media; 5 replica plates were used. Only if the nuclei of clone 8 had entered clone 16 and migrated to the area of a plug would that plug give rise to mycelia on which mushrooms could form. This species has an extremely fast rate of somatic fusion and of nuclear migration (Ross, 1976). In this experiment, of the plugs taken at the first time interval of 2 h, only those taken up to 2 cm from the centre produced mushrooms, all dark. Plugs taken at successive 2-h intervals all produced mushrooms. Pale mushrooms were produced at first only by the plug closest to the centre, and after a lag, by successively more distant plugs. All mycelia have continued to produce pale or dark mushrooms respectively.

The assumption was made that the agent responsible for producing paleness had migrated through the cells of clone 16 at a slower rate than the nuclei and that the agent was an extra-nuclear, cytoplasmically inherited factor and not a nuclear gene. We have since found several parameters that will affect the ability of the pale factor to migrate and these will be reported elsewhere (Ross & Damm, unpublished).

Cell-free extracts obtained from clone 8 as described above have been used to infect clone 16 and clone 9 (a₂). Each infected mycelium was incubated for 3–4 days in the dark before plugs were taken from non-slashed areas of the culture. These plugs were crossed with compatible clones and allowed to sporulate. So far, every mycelium of clone 16 infected by extracts of clone 8 has given rise to pale mushrooms after crossing with normal compatible clones and has since retained this infectivity. Infected clone 9 routinely produced 100% pale mushrooms when crossed with normal clones for the first 2 or 3 transfers. After 4–6 transfers, dark and mosaic mushrooms began to appear and after 10–15 transfers of the original infected mycelium, the culture became wholly dark. The implications of clonal variation in response to infectivity will be discussed more fully elsewhere (Ross & Damm, unpublished).

If the 0·22-μm filtrate is centrifuged at 10000g for 1 h, all infectivity is found in the pellet. Extracts frozen at —40 °C have retained full infectivity for over 5 months.

In all infection experiments, control cultures were slashed through the Tris-glycine buffer alone or through extracts of normal clones. In no case has any of the controls produced pale mushrooms. In control experiments without slashing, infection was far more erratic. In any series of plates so treated, infection would range from 10 to 100 % of the plates. Any plate that did become infected retained the infection. Slashing alone has the property of increasing the number of primordia formed, but does not affect the colour of the mushrooms (Ross, unpublished).

We have had the pale factor carrying clone 8 in culture for 4 years and have not seen any well defined, consistent pathological symptoms except the failure to produce spores. We have found no significant differences in growth rate or appearance between clone 8 and its derived dikaryons and the normal monokaryons and dikaryons. All dikaryons, with and without clone 8 as a member, are identical in light-induction requirements, in rates and numbers of primordial differentiation and in development and timing of mushroom maturation. General protein and isozyme patterns on polyacrylamide gels do not show any differences that may be attributed at this time to the pale factor (Ross et al. 1973). The only overt symptom of the presence of the pale factor is the absence of spores.

Preliminary examination did not reveal any unusual cytoplasmic contents of vegetative hyphae but did indicate a difference in the cytology of basidia of infected dikaryons. Figs. 5–24 represent a series of maturing basidia from both dark (6 × 16) and pale (8 × 16) mushrooms. Figs. 5 and 6 show the basidia of the 5-mm primordia. Most of the pre-basidial apical cells are beginning to enlarge and all still contain the 2 nuclei of the dikaryon. No differences can be seen at this stage between the pale and the dark forms. Both contain cytoplasmic vacuoles that appear electron-transparent or contain amorphous, fibrous material. By the 10-mm stage, Figs. 7 and 8, nearly all basidia contain the diploid fusion nucleus. In the dark form, most of the diploid nuclei at this stage have entered prophase of meoisis I and are forming synaptinemal complexes (Figs. 7, 9). The synaptinemal complexes are well defined, long, and appear normal. In the pale form, synaptinemal complexes do not appear normal. They are shorter and more diffuse than in the dark form and have not achieved the delimitation of the normal synaptinemal complexes (Figs. 8, 10). The cytoplasmic vacuoles of the dark form (Fig. 11) are still electron-transparent, but the vacuoles of the pale form now contain numerous membrane-bound bodies, some with swollen heads and long, taillike, projections that appear to curve in and out of the plane of section (Fig. 12).

By the 15-mm primordial stage, meiosis I is under way in the basidia of the dark form with well-defined microtubules extending from the spindle pole bodies (Fig. 13). This meiosis seems typical of those described for Coprinus by others (Raju & Lu 1970; Gull & Newsam, 1975 a, b). In the pale form, however, basidial nuclei in the 15-mm primordia do not progress through meiosis. The large diploid nuclei appear unchanged from the 10-mm stage (Fig. 14) with vague areas of condensation representing possible sites of synaptinemal complex formation. The vacuoles of the pale basidia are quite full of vesiculate, membrane-bound bodies of many shapes, particularly elongate beaded forms and those with swollen areas attached to long sinuate projections (Figs. 15, 16). At the 20-mm stage, the dark basidia contain the 4 daughter nuclei of the completed meiosis (Fig. 17) with vacuoles essentially unchanged. There are occasional membrane fragments in the vacuoles of the dark basidia, but this may result from the continued autophagy that occurs in both forms. The pale form basidia still contain a single diploid nucleus, but there are now larger and more delimited synaptinemal complexes (Fig. 18), although these are usually closely associated with the nucleoli that are still present and prominent in the nuclei. The vacuoles of the pale form still contain the same membrane-bound bodies (Figs. 19, 20).

There is a rapid increase in the size of the basidia of the dark form by the 25-mm stage (Fig. 21) and the apex of the basidia has begun to differentiate into the pre-sterigmatal horns. The basidia of the pale form still contain the single diploid nucleus in which short synaptinemal complexes appear to be disintegrating (Fig. 22) and the vacuoles remain full of membrane-bound bodies (Fig. 23). Normal dark basidia continue to develop and to produce spores (Fig. 24) with normal autolysis occurring soon afterwards. The pale basidia do not mature beyond the point of Fig. 22 and are eventually destroyed by the autolysis of the mushroom.

The failure of the basidia in the infected dikaryons to complete meiosis is the cause of the overt symptom, the absence of spores, and the basis for the pale colour of the infected mushrooms.

The membrane-bound bodies that appear at the beginning of meiosis in the basidia of the pale form range in size from 130 to 200 nm, with elongate tails up to 2 /im in length. These bodies resemble electron micrographs of mycoplasmas published by others in animals, insects, and plants (Smith, 1971; Ehrman & Kirnaghan, 1972; Heath & Unestam, 1974; Chen & Liao, 1975; Williamson & Whitcomb, 1975), though their dimensions are slightly smaller. Difficulties of fixation encountered with this species of Coprinus may account for the apparent lack of contents in some of the mycoplasma-like bodies, although some micrographs do indicate some interior detail.

We are basing our identification of the infectious agent as a mycoplasma on the combined evidence obtained from infectivity, cell-free extract and ultrastructural studies. Since the rate of migration of the pale factor through a recipient mycelium is slower than the nuclei of the pale factor-carrying clone, we are regarding this as evidence that the pale factor is not a nuclear gene, an assumption strengthened by the fact that cell-free extracts filtered through a 0·22-μm Millipore filter, which would exclude nuclei, are highly infectious. The filtration work strongly implicates entities that are known to pass through such small-pore filters: soluble compounds (proteins, nucleic acids, etc.), viruses, mitochondria, mycoplasmas and various membrane vesicles and fragments. The fact that all infectivity pellets at only 10000g mitigates against unbound soluble materials and unaggregated virus. The ability of cultures infected with the pale factor to retain their paleness for several generations implies that it is capable either of replicating in the host or of being replicated by the host. We have found no evidence of any virus-like particles in the basidia or any other developmental stage of this Coprinus”, consequently, the evidence points to mycoplasmas, or, possibly, to membrane-bound macromolecules capable of being replicated by host machinery, such as RNA and DNA. The appearance of mycoplasmalike bodies in only the cells of the infected host that display any disease symptoms is correlated with the failure of those cells to undergo meiosis and thus implies a causal relationship and suggests that the infectious agent is indeed a mycoplasma. Even though this mycoplasma differs from those found in other organisms in several ways, it is difficult to implicate any other kind of known infectious agent in the face of the accumulated evidence. Mycoplasmal infections are usually shown in micrographs as filling large areas of the infected cells or adhering to the cell surface in large numbers (Heath & Unestam, 1974; Smith, 1971), whereas the mycoplasma described here is not found in great abundance even in those cells most affected by its presence. If it is not a mycoplasma, it is an infectious agent that is less than 0-22 /im in size, that sediments at 10000g and that causes proliferation of membrane-bound bodies inside cytoplasmic vacuoles of certain specialized cells of the host, and is of a type unknown to the authors. If it is a mycoplasma, it is one of very intriguing behaviour and interesting potential.

Of considerable interest is the appearance of the mycoplasma in the vacuoles of the meiotic cells. These bodies are not present in the young basidia until after karyogamy. They are first seen at the stage where processes specific for meiosis begin to occur: pairing of homologous chromosomes and the formation of synaptinemal complexes. What causes the appearance of the mycoplasma at this time? Quite apart from the possible effect of the mycoplasma on meiosis, does the host meiotic cell have an influence on the mycoplasma? Is there something specifically different about the interior environment of a meiotic cell that particularly causes this mycoplasma to replicate? In what form does the mycoplasma exist in the vegetative hyphae? Preliminary studies have not found any mycoplasma-like bodies in the vegetative hyphae, and yet extracts of vegetative hyphae are highly infectious.

The failure of the infected cells to complete meiosis is also of great interest. Since normal growth of the host is unaffected by the infection, we are starting with the working hypothesis that the mycoplasma is affecting whatever differentiates the meiotic divisions from those of mitosis. This could be as general as the reduction of available energy because of the growth of the mycoplasma in the meiotic cells or as highly specific as the blocking of synaptinemal complex formation by the production of analogues to essential synaptinemal complex components (e.g. the synaptomeres and zygosomes of King, 1970). Since mitosis is not affected, it is doubtful if there would be any involvement with the spindle microtubules or with tubulin itself, although the production of a tubulin depolymerizer by high concentrations of the mycoplasma cannot be discounted.

The association of the synaptinemal complexes of the 20-mm-stage pale mushrooms with the still persistent nucleolus resembles the situation reported by Gull & Newsam (1975b) as a normal occurrence in several basidiomycetes, but regarded by Moses (1968) as indicating an abnormal situation in spermatocytes. Such associations of synaptinemal complexes with the nucleolus have not yet been seen in the normal dark-form basidia.

A final point at this time is the question of the origin of the pale factor infection. Since all clones were originally derived from the same dikaryon and only one spore, that which gave rise to clone 8, resulted in a carrier clone, there is the distinct possibility that the pale factor is actually endemic to all clones, but is expressed in a pathogenic state only in clone 8 and its derived dikaryons. Since clone 8 extracts are infective, the agent must be present in clone 8 in an infective form and is therefore probably repressed or inhibited in the other clones, a possibility of intriguing potential in terms of the regulation of meiotic inhibition. The gradual curing of infection by clone 9 does suggest a variation in clonal responses to the pale factor and we have recorded several similar instances of host responses (Ross and Damm, unpublished). Additional evidence for the presence of the pale factor in other clones comes indirectly from attempts to obtain mutant strains. Spores from a normal 6 ×16 dikaryon were irradiated with 95 % lethal levels of ultraviolet light. Surviving spores were isolated, grown and crossed. In several cases, dikaryons from such crosses gave rise to pale mushrooms that appear identical to the 8 × 16 pale mushrooms. There is, therefore, the possibility that nuclear genes may regulate the growth and replication of the pale factor and/or its effect on meiosis. Such questions are now under investigation.

The high rate of infection and migration through the host, the simplicity of the host, and the singularity of the pathological effect all make this mycoplasma-host association an ideal one in which to study the basis of infection and pathogenicity. It is also possible that the effect on meiosis may result in a new tool to investigate the biochemistry and regulation of meiosis.

We thank G. Shipley and P. Robertson for technical assistance and Laurie Marx for Figs. 1, 2. This work was supported by National Science Foundation grant GB-43819 and a grant from the Faculty Research Committee, U.C.S.B.