Changes in cell morphology and cell adhesion occurred when cultured cells from the rat liver cell strain C3 were exposed to the fungal toxins, sporidesmin or gliotoxin. Both toxins caused loss of attachment of the cells to the plastic of tissue culture plates and this effect was preceded by loss of actin cables. Other changes included cytoplasmic vacuolation and blocked entry into S-phase of the cell cycle. Under these conditions [3H]thymidine incorporation into the cells was also diminished but changes were not detected in the amount of cellular actin, or in the accessibility of cell surface proteins to iodination carried out by the Bolton and Hunter method. The observations suggest that disruption of microfilaments is one of the earliest effects of these toxins on eukaryotic cells.

The mycotoxins sporidesmin and gliotoxin belong to a class of fungal metabolites that contain a disulphide-bridged piperazinedione ring system (Taylor, 1971). This sulphur-containing structure is considered to be responsible for the various cytopathogenic activities of the toxins. Gliotoxin is produced by a range of fungi including Aspergillus fumigatus, is immunosuppressive in vitro (Mullbacher & Eichner, 1984), and has been implicated as the causative agent in tissue damage associated with some opportunistic fungal infections (Eichner & Mullbacher, 1984). Sporidesmin is synthesized only by the saprophytic mould Pithomyces chartarum and is responsible for the hepatogenous photosensitization disease of livestock that has been called facial eczema (Mortimer et al. 1978). P. chartarum has a cosmopolitan distribution and the toxigenic strains that produce sporidesmin are abundant in New Zealand where the mycotoxicosis facial eczema is responsible for substantial losses in sheep and cattle production. In sheep, ingested sporidesmin is concentrated in the liver and bile, alters liver function, and causes the necrotic and inflammatory changes in the liver that result in biliary obstruction and the external signs of photosensitization (Mortimer, 1963).

Altered morphology and function of the plasma membrane appear to be characteristic changes in isolated or cultured cells exposed to gliotoxin or sporidesmin. Both agents induce detachment of cultured adherent cells from tissue culture substrates (Mortimer & Collins, 1968; Mullbacher & Eichner, 1984) and strip microvilli from the plasma membranes of target cells (Cordiner & Jordan, 1983; Eichner & Mullbacher, 1984). Mullbacher & Eichner (1984) demonstrated that gliotoxin completely inhibited phagocytosis by rodent macrophages and we have shown that sporidesmin blocks the ability of neutrophils to kill target cells in vitro (Jordan & Vadas, unpublished). Cytoplasmic vacuolation also appears to be a general feature of the activity of these toxins and occurs in both the periportal hepatocytes and the biliary epithelia of animals dosed orally with sporidesmin, and in cells that are exposed to the toxins in vitro.

Bullock et al. (1974) suggested that microfilament disruption might be responsible for the observed loss of microvilli from canalicular membranes in isolated perfused rat livers exposed to sporidesmin, and Cordiner & Jordan (1983) similarly proposed that microfilament modification might account for the effects of sporidesmin and gliotoxin on isolated hepatocytes. We report here that the two toxins induce detachment and vacuolation of cultured liver cells and that these processes are preceded by disruption of microfilament cables.

Cell culture, cell detachment and cell viability

Cells of the non-tumorigenic rat liver C3 strain (Paraskeva & Gallimore, 1980) were grown in 25 cm2 tissue culture flasks, containing 5 ml of Dulbecco’s modified Eagle’s medium supplemented with 10 % foetal calf serum at 37°C in a 90 % air/10 % CO2 atmosphere. Sporidesmin (unsolvated, A/r474) from Ruakura Animal Research Centre, Hamilton, New Zealand, gliotoxin (Mr326) from Lederle Labs, New York, or other agents were added dissolved in ethanol to give final ethanol concentrations not greater than 0·1 % in the cell culture medium. In all experiments described in this paper ethanol at a concentration of 0·1% in the culture medium produced no observable change when compared with control cultures containing only cells and medium. Dye exclusion was examined after adding eosin dissolved in phosphate-buffered saline (PBS) to a final eosin concentration of 0·1 %. Cell detachment was observed routinely by light microscopy of cultures of approximately 106 confluent C3 cells that had been grown in 5 ml of medium.

Detachment was quantified using cells that had been pre-loaded with 51Cr before exposure to toxin. For 51Cr loading approximately 106 adherent cells were exposed to 5 ml of medium with 40μCi of 51Cr-labelled sodium chromate (400mCi per mg Cr, from Amersham, UK) for 2h and were then washed free of excess isotope with three changes of medium. 51Cr release from damaged cells was assayed by removing samples of the medium, and then sedimenting any detached cells by centrifugation at 500g for 5 min, before measuring the released radioactivity in the supernatant. To assay cell detachment the culture flasks that contained 51Cr-loaded, control or toxin-exposed C3 cells were washed with three changes of culture medium and the detached cells were sedimented from the pooled wash medium. 51Cr was then determined in the detached washed cell pellets, in the cell-free supernatants, and in the residual adherent cells after addition of 5% Triton X-100, and detachment was calculated as the percentage of the total radioactivity found in detached cells. All results are means of three to six experiments that showed less than 15 % variation from the mean.

Cell cycle analysis

Cell cycle analysis was carried out by flow cytometry in a FACS-II using the propidium iodide staining technique described by Taylor (1980). Excitation was with an argon laser at 488nm and fluorescence was measured at greater than 580 nm.

Indirect immunofluorescence

Cultured cells were grown on glass coverslips and when necessary were transferred to fresh medium containing sporidesmin or gliotoxin. The coverslips were rinsed in PBS, fixed in absolute acetone at –4°C for 1 min, air dried, then exposed to a human autoimmune anti-actin serum, diluted 1:40 in PBS, at room temperature for 20 min in a humidified chamber. The coverslips were washed three times in PBS for 5 min each before incubation for 20 min with FITC-conjugated antihuman immunoglobulin (Ig) antibody (DAKO, Denmark) diluted 1:30 in PBS. After washing a further three times in PBS the coverslips were mounted in veronal-buffered glycerol containing 4-phenylenediamine, pH 8·3, and were examined by fluorescence microscopy.

The specificity of the autoimmune serum has been described (Toh et al. 1976) and has been confirmed as anti-actin by the codistribution of anti-actin staining with that of the F-actin binding dye rhodamine phalloidin in double-label staining of cryostat sections of liver, by diminished staining with the anti-actin serum after pre-exposure of liver sections to a monoclonal anti-actin, by immunoprecipitation of purified skeletal muscle actin, and by the demonstration that a protein migrating with actin on polyacrylamide gels accounted for more than 90 % of the antigen immunoprecipitated by the anti-actin serum from biosynthetically labelled cultured C3 cells (Jordan, 1986).

Rhodamine phalloidin staining

Coverslip cultures of cells were rinsed, fixed and air dried as before. The coverslips were then exposed to a solution of rhodamine phalloidin (50 ng ml-1, obtained from Molecular Probes Inc., USA) at room temperature for 20min before washing and mounting for examination by fluorescence microscopy.

Polyacrylamide gel electrophoresis

One-dimensional sodium dodecyl sulphate (SDS)-polyacrylamide gel electrophoresis on 7|% or 10% resolving gels was carried out as described (Frazer et al. 1985). For two-dimensional electrophoresis cells lysates were adjusted to 9M-urea, 1 % Triton X-100, 1 % pH 3-5–10·0 LKB ampholytes, 25 mM-dithiothreitol, and electrophoresis was for 5 h at 550 V in 120 mm × 2 mm 4 % polyacrylamide tube gels containing 9M-urea, 2% Triton X-100, 1% pH3·5·100 ampholytes. Developed tube gels were equilibrated for 1 h in 2-3% SDS, 10% glycerol, 62·5mM-Tris’HCl, pH 6·8, 5·0 mM-dithiothreitol, and electrophoresis in the second dimension was on 10 % polyacrylamide resolving gels in the presence of SDS.

Coomassie Blue staining and autoradiographic detection of radioactively labelled proteins were performed as described (Frazer et al. 1985).

Radioactive labelling

For biosynthetic labelling of proteins approximately 106 cells were grown for 2h in carrier methionine-free medium with dialysed foetal calf serum. Labelling was initiated by addition of 0·5 mCi of L-[35S]methionine (800 Ci mmol-1, Amersham, UK) and the reaction was terminated by replacement of the medium with ice-cold PBS containing 2mM-methionine. The cells were washed in PBS, harvested by scraping, washed again and vortexed in a lysis buffer, which was ice-cold PBS containing 0·5 % Triton X-100, 1 mM-iodoacetamide, 1 mM-phenylmethylsulphonyl fluoride and Trasylol (200 Kallikrein inhibitor units per ml).

Cell surface proteins of cultured cells were labelled (Bolton & Hunter, 1973) with 125I-labelled Bolton & Hunter reagent (2000 Ci mmol-1, Amersham, UK). Cells for labelling were harvested by scraping, washed twice in PBS, resuspended (106 cells in 0·5 ml ice-cold PBS) and added to 0·2 mCi of dry labelling reagent. After 10 min at 4°C the reaction was terminated by adding 2 ml of ice-cold PBS containing 0·2M-glycine and the cells were harvested by centrifugation and washed three times in PBS-glycine. Washed cell pellets were vortexed in lysis buffer and prepared for electrophoresis.

Thymidine uptake was measured using samples of approximately 104 C3 cells that had been cultured overnight in 200 μ(1 of serum-containing medium in the wells of 96-well plates (Limbro, Flow Labs Ltd, UK); 10μCi of [6-3H]thymidine (25 Ci mmol-1, Amersham, UK) was distributed to each well and at intervals up to 2 h after the addition of thymidine, incorporated radioactivity was measured by liquid scintillation counting after harvesting on a MASH apparatus (Skatron A.S., Norway) as described by Burnset al. (1982).

Cell detachment and viability

Sporidesmin and gliotoxin rapidly induced morphological change and detachment when added to cultured adherent C3 cells. Addition of sporidesmin at 1 jugml-1 in medium resulted after 30–60min in strands of detaching cells (Fig. 1), and after 2–4 h there were only single rounded-up cells as a result of cell-substratum and cell-cell detachment. Detachment was slower in C3 cells exposed to 0·01·0’1 μg sporidesmin per ml culture medium. Complete detachment required 4–5 days continuous exposure to toxin at 0’01 pg sporidesmin per ml medium. Because gliotoxin appeared to be more potent than sporidesmin in promoting cell detachment the process was quantified using cells preloaded with 51Cr. When compared with control C3 cells, which were incubated in the absence of toxin, or harvested after 1 h exposure to 0’25 % trypsin, neither sporidesmin nor gliotoxin substantially damaged plasma membranes as measured by the criteria of 51Cr retention (Zawydiwski & Duncan, 1978) or of eosin exclusion. Cell detachment was progressive with time and was more effectively promoted by gliotoxin than by sporidesmin (Fig. 2). At 0’1 μg toxin per ml of culture medium gliotoxin caused approximately 50% detachment within 4h although under similar conditions sporidesmin produced less than 15% cell detachment. Interaction of sporidesmin with the tissue culture plastic did not appear to be an important component of the detachment process, as flasks exposed to 1 jtigml-1 sporidesmin for 1 h, and then washed in PBS before the addition of 105 cells in fresh medium, supported normal growth and adherence.

Fig. 1.

Sporidesmin-induced changes in cultured C3 cells. C3 cells were cultured in the presence or absence of sporidesmin. A. Control cells, no toxin; B, sporidesmin, 1 pg ml-1 for 1 h; C, sporidesmin, 0·1 μgml-1 for 1 day; D, sporidesmin, 0·1 μgml”1 for 2 days; E, sporidesmin, 0·01 μgml-1 for 4 days; F, cells exposed to 1 μg ml-1 sporidesmin for 1 h, washed in PBS and cultured for a further 3 days. The major changes were detachment and rounding up of cells plus vacuolation and changes of form at the longer exposure times.

Fig. 1.

Sporidesmin-induced changes in cultured C3 cells. C3 cells were cultured in the presence or absence of sporidesmin. A. Control cells, no toxin; B, sporidesmin, 1 pg ml-1 for 1 h; C, sporidesmin, 0·1 μgml-1 for 1 day; D, sporidesmin, 0·1 μgml”1 for 2 days; E, sporidesmin, 0·01 μgml-1 for 4 days; F, cells exposed to 1 μg ml-1 sporidesmin for 1 h, washed in PBS and cultured for a further 3 days. The major changes were detachment and rounding up of cells plus vacuolation and changes of form at the longer exposure times.

Fig. 2.

Toxin-induced detachment of cultured C3 cells. Cultured adherent cells were loaded with 51Cr by exposure to medium containing slCr-labelled sodium chromate for 2h, washed free of excess isotope, and supplemented “with fresh medium. Trypsin (0·25 %), sporidesmin (Spd) or gliotoxin (Glio) were added at the stated concentrations and after 0, 1 or 4 h detached cells were removed in the culture medium and harvested by centrifugation. The amount of 51Cr present in adherent and detached cells and released into the culture supernatants was determined by gamma counting. Detachment was progressive with time and gliotoxin was more effective than sporidesmin in promoting detachment.

Fig. 2.

Toxin-induced detachment of cultured C3 cells. Cultured adherent cells were loaded with 51Cr by exposure to medium containing slCr-labelled sodium chromate for 2h, washed free of excess isotope, and supplemented “with fresh medium. Trypsin (0·25 %), sporidesmin (Spd) or gliotoxin (Glio) were added at the stated concentrations and after 0, 1 or 4 h detached cells were removed in the culture medium and harvested by centrifugation. The amount of 51Cr present in adherent and detached cells and released into the culture supernatants was determined by gamma counting. Detachment was progressive with time and gliotoxin was more effective than sporidesmin in promoting detachment.

Cells detached from tissue culture plastic by exposure to 0·1 or 1·0μgml-1 sporidesmin or gliotoxin were not able to re-establish dividing cell cultures. When the cells detached by exposure to toxin were washed three times in PBS to remove extraneous toxin and then added to fresh growth medium in tissue culture flasks, normal growth did not occur, so that by 3 days after seeding less than 105 cells were present in the toxin-exposed cultures whereas control cultures that had been detached with 0-25 % trypsin had grown to confluence at about 106 cells per flask. In addition, the cells that had been exposed to toxin retained the ability to exclude eosin but were enlarged, had extensive cytoplasmic vacuolation and showed abnormal forms (Fig. 1). Similar changes were seen in the period prior to detachment for cells that had been grown in 0·01 or 0·1 μgml-1 sporidesmin for 1–4 days.

Potential modes of action of gliotoxin and sporidesmin were assessed by comparing the effects of other agents on cell detachment. The compounds used were the sulph-hydryl reagent dithiothreitol, the calcium ionophore A23187, tunicamycin, which blocks peptide glycosylation, and the microfilament-disrupting agent cytochalasin B. Cells cultured in medium containing 0’03 mM-dithiothreitol, or 1–10 /1M-A23187 for 1 h, or 0·1–1 μgml-1 tunicamycin for 24h, did not show signs of morphological change or detachment. However, overnight exposure of cells to medium containing 1 μgml-1 cytochalasin B produced swelling and cytoplasmic vacuolation without loss of cell adherence.

Cell cycle analysis and thymidine uptake

Because the toxins apparently blocked the growth of the cultured cells, the effect of sporidesmin on cell cycling was examined using flow microfluorimetry of propidium-iodide-stained cells. Analysis of cells from control, toxin-free populations indicated that approximately 90% of the cells were found in the Gi peak. Cells with increased staining distributed as an S-phase shoulder upstream from the main peak and as a small G2 population (Fig. 3). A similar distribution of DNA occurred in cells that had been exposed to 1 μg ml-1 sporidesmin for 1 h, but cells that had been cultured in 0·1 μgml-1 sporidesmin for 1–2 days did not enter S-phase, although those cells that were in G2 before sporidesmin was added apparently went through a normal round of cell division. By 1 day after exposure to O-lμgml’1 sporidesmin the S-phase shoulder was diminished and after the second day more than 95 % of the cells that stained were in the G1 peak.

Fig. 3.

The effect of sporidesmin on cell cycling. Cell cycle analysis was carried out by flow microfluorimetry of the nuclei of propidium-iodide-stained cells. A. Control cells, no toxin; B, cells exposed to 1 μgml-1 sporidesmin for 1 h; C, cells exposed to 0·1 ftgml-1 sporidesmin for 1 day; D, cells exposed to 0·1 fig ml-1 sporidesmin for 2 days.

Fig. 3.

The effect of sporidesmin on cell cycling. Cell cycle analysis was carried out by flow microfluorimetry of the nuclei of propidium-iodide-stained cells. A. Control cells, no toxin; B, cells exposed to 1 μgml-1 sporidesmin for 1 h; C, cells exposed to 0·1 ftgml-1 sporidesmin for 1 day; D, cells exposed to 0·1 fig ml-1 sporidesmin for 2 days.

As it appeared that sporidesmin blocked entry into S-phase without blocking cell division the effect of the toxin on [3H]thymidine uptake into cultured C3 cells was measured (Table 1). Sporidesmin at 0·4–1·0μgml-1 in the culture medium partially inhibited thymidine incorporation. Cells exposed to 04 μgml-1 sporidesmin overnight prior to the addition of [3H] thymidine showed a much reduced isotope incorporation while 1μgml-1 sporidesmin added to cultured cells 5 min before addition of the thymidine diminished 3H-labelling to about 40 % of control values.

Table 1.

Effect of sporidesmin on [3H]thymidine uptake by cultured C3 cells

Effect of sporidesmin on [3H]thymidine uptake by cultured C3 cells
Effect of sporidesmin on [3H]thymidine uptake by cultured C3 cells

Histochemical staining of actin microfilaments

Because the cytoplasmic vacuolation and cell swelling caused by cytochalasin B was similar to that seen in the cells that had been exposed to sporidesmin or gliotoxin, the distribution of actin microfilaments was examined by indirect immunofluorescence using a polyclonal anti-actin serum. Sporidesmin and gliotoxin were found to modify microfilament distribution in the period prior to cell detachment (Fig. 4). In control C3 cells cytoskeletal microfilaments were concentrated as actin cables in the cortical cytoplasm underlaying the plasma membrane. Exposure of C3 cells to 0·1 μUgml-1 sporidesmin produced substantial reduction in cable number and fluorescence intensity within 2h. By 6–8 h, when morphological change and detachment were pronounced, actin staining was concentrated in retracting plasma membrane filipodal projections and in cytoplasmic spots, both presumably at remaining sites of cell-substratum contact (Small & Langanger, 1981; WangeZ al. 1984). The effects of gliotoxin were similar but occurred more rapidly with substantial loss of actin cables 20–40 min after exposure to 0·1 μg l-1 gliotoxin so that the changes observed with cells exposed to gliotoxin for 2 h were similar to those found 6–8 h after addition of sporidesmin.

Fig. 4.

Sporidesmin- and gliotoxin-induced disruption of actin microfilaments. Fixed permeabilized C3 cells were subjected to indirect immunofluorescence using human antiactin serum (A-D) or to direct staining of actin filaments with rhodamine phalloidin (E-F). The stained control cells are shown in A and E. Toxin exposures were: B, 0 1 μg ml-1 sporidesmin for 2h; C,F, 0T μg ml-1 sporidesmin for 7h; D, 0·4 μgml-1 gliotoxin for 2 h. Control cells contained abundant microfilament cables. Exposure to the toxins resulted in loss of cables and concentration of actin in filamentous extensions of the plasma membrane.

Fig. 4.

Sporidesmin- and gliotoxin-induced disruption of actin microfilaments. Fixed permeabilized C3 cells were subjected to indirect immunofluorescence using human antiactin serum (A-D) or to direct staining of actin filaments with rhodamine phalloidin (E-F). The stained control cells are shown in A and E. Toxin exposures were: B, 0 1 μg ml-1 sporidesmin for 2h; C,F, 0T μg ml-1 sporidesmin for 7h; D, 0·4 μgml-1 gliotoxin for 2 h. Control cells contained abundant microfilament cables. Exposure to the toxins resulted in loss of cables and concentration of actin in filamentous extensions of the plasma membrane.

In addition to the studies with the anti-actin antibody the microfilament changes produced by the toxins were also examined using the F-actin binding cyclic peptide rhodamine phalloidin (Fig. 4). The results obtained were qualitatively similar to the changes seen with the anti-actin serum, although changes detected with the immune serum preceded those seen when rhodamine phalloidin was used for staining. Substantial loss of actin cables required 3–4 h exposure to 0–1 μgml-1 sporidesmin although the changes at 6–8 h were similar to those detected by anti-actin staining.

Polyacrylamide gel electrophoresis

Polyacrylamide gel electrophoresis was carried out to seek evidence of toxin-induced change in amounts of proteins responsible for cell adhesion or cytoskeletal structure. However, no such change could be detected as judged by the following studies.

C3 cells grown in sporidesmin for 1 h at 1 μg toxin per ml of medium or for 24 h at 0·1 μg toxin per ml did not show obvious changes from control unexposed cells on the Coomassie-Blue-stained gels of total cell protein that had been separated by onedimensional (Fig. 5) or two-dimensional electrophoresis. Similarly, there were no apparent changes in the accessibility of the major cell surface proteins to iodination that had been carried out using the Bolton & Hunter reagent (Fig. 5). Exposure of the cells to Ijiigml-1 sporidesmin for 1 h or to 0·lμUgml-1 sporidesmin overnight prior to addition of [35S]methionine did not qualitatively alter the incorporation of radioactivity into the major proteins, including actin, which could be resolved by two-dimensional gel electrophoresis (Fig. 6).

Fig. 5.

Polyacrylamide gel electrophoresis of proteins from C3 cells that had been labelled with 125I-labelled Bolton & Hunter reagent. A,B,C. Show a 7½% gel stained with Coomassie Blue; A’,B’,C’, show the autoradiograph of the same gel. A,A’. Control cells not exposed to toxin; B,B’, cells exposed to 1 μgml-1 sporidesmin for 1 h; C,C’, cells exposed to 0·1 ugml-1 sporidesmin for 1 day.

Fig. 5.

Polyacrylamide gel electrophoresis of proteins from C3 cells that had been labelled with 125I-labelled Bolton & Hunter reagent. A,B,C. Show a 7½% gel stained with Coomassie Blue; A’,B’,C’, show the autoradiograph of the same gel. A,A’. Control cells not exposed to toxin; B,B’, cells exposed to 1 μgml-1 sporidesmin for 1 h; C,C’, cells exposed to 0·1 ugml-1 sporidesmin for 1 day.

Fig. 6.

Autoradiography of lysates of [3sS] methionine-labelled cells after electrophoresis on two-dimensional polyacrylamide gels. Cultured cells were labelled for 1 h using 0-5 mCi of [3sS] methionine in either the absence (A) or presence (B) of 1μgml-1 sporidesmin. Arrows show position of monomeric actin determined by electrophoresis of reference actin.

Fig. 6.

Autoradiography of lysates of [3sS] methionine-labelled cells after electrophoresis on two-dimensional polyacrylamide gels. Cultured cells were labelled for 1 h using 0-5 mCi of [3sS] methionine in either the absence (A) or presence (B) of 1μgml-1 sporidesmin. Arrows show position of monomeric actin determined by electrophoresis of reference actin.

Various suggestions have been advanced to explain the way in which the toxicologically active sulphurs of gliotoxin and sporidesmin might interact with target tissue molecules. The hypotheses include interaction with the thiol groups of membrane proteins (Middleton, 1974) and cytoskeletal microfilaments (Bullock et al. 1974) and production of oxygen-free radicals (Munday, 1982, 1984). We have chosen to investigate the cytological effects of the toxins on isolated and cultured liver cells prior to biochemical investigation of any observed morphological change. Many of the changes that we have seen (Cordiner & Jordan, 1983; Cordiner et al. 1983; and this study) have been located in the region of the plasma membrane and microfilament cytoskeleton. The initial biochemical observations suggest that synthesis of actin or other proteins is not altered under these conditions.

The perturbation of cytoskeletal microfilament cables described in this study appears to be one of the earliest effects of gliotoxin and sporidesmin on tissue culture cells, and precedes other morphological change including detachment and vacuolation. We have repeated these experiments (unpublished) using 3T3 cells, human lung embryo fibroblasts, a rat glioma cell line and human hepatocellular carcinomas. Actin cable disruption always preceded cell detachment and there was no change in the distribution of intermediate filaments or microtubules when these structures were detected by immunofluorescent staining with monoclonal antibodies against vimentin and tubulin. Diminished immunohistochemical staining of microfilament actin has also been observed in the liver of mice 3 h after exposure to an oral dose of sporidesmin and there were no accompanying changes in the staining of intermediate filaments detected by immunohistochemical reaction with an antiserum to bovine hoof prekeratin (Jordan, 1986).

Microfilament organization including the association of actin with plasma membranes provides an important component of cultured cell morphology (Lanks & Kasambalides, 1983) and cell-substratum contact (Small & Langanger, 1981; Wang et al. 1984). Loss of microfilament bundles and aggregation of actin with accompanying changes in cell-adhesive properties and plasma membrane function occurs in cells transformed with Rous sarcoma virus and has been related to modification of actin-organizing proteins including vinculin and gelsolin (David-Pfeuty & Singer, 1980; Shriver & Rohrschneider, 1981; Wang et al. 1984). Similarly, formation of vinculin plaques and actin stress fibres is accompanied by morphological change during retinoic-acid-induced differentiation of teratocarcinoma cells (Lehtonen et al. 1983). These changes differ from the intermediate filament redistribution induced by agents including acrylamide (Eckert, 1985), and protein synthesis inhibitors such as diptheria toxin or Pseudomonas aeruginosa exotoxin A (Sharpe et al. 1980), which produce specific disruption of vimentin intermediate filaments without modification of microfilament or microtubule organization.

The cell detachment induced by sporidesmin and gliotoxin that has been described by us and by others (Mortimer & Collins, 1968; Mullbacher & Eichner, 1984) may thus be the result of microfilament disruption. Such an effect could be due to direct action of the toxins on microfilament proteins including actin or the actin-binding proteins that regulate microfilament bundling and capping or binding to plasma membranes (Weeds, 1982). However, as cell adhesion is a complex process it is possible that the cell detachment and microfilament disruption resulted from otherwise undetected earlier changes to plasma membrane, cell surface or intracellular components. In an attempt to investigate whether sporidesmin caused the loss of particular plasma membrane proteins we examined the cell surface proteins that were accessible to Bolton & Hunter iodination, but could not show any changes in the cells that were exposed to the toxin (Fig. 5).

Many of the effects of sporidesmin and gliotoxin, including inhibition of phagocytosis, induction of microvilli loss, reorganization of cortical cytoplasm, altered bile acid uptake, and efflux and modified hormone responsiveness (Bullock et al. 1974; Cordiner & Jordan, 1983; Cordiner et al. 1983; Mullbacher & Eichner, 1984), are explicable in terms of modification of microfilaments or plasma membrane. Whether the effects of the toxins on cell cycling and thymidine incorporation are secondary to changes in plasma membrane and microfilaments or are due to other processes remains to be investigated. We are currently examining the effects of sporidesmin, gliotoxin and their analogues on the organization of actin and actin-binding proteins. If the various effects of the toxins are due directly to microfilament perturbation it will be of interest to elucidate how the biologically active sulphur-bridged toxin ring system specifically interacts with actin or actin-binding proteins to produce the changes described in this paper.

The authors are indebted to Dr F. Battye for expertly carrying out the FACS analysis and to Dr I. R. Mackay and Professor B. H. Toh for their interest and advice during the course of this work.

Bolton
,
A. E.
&
Hunter
,
W. M.
(
1973
).
The labelling of proteins to high specific radioactivities by conjugation to a 125I-containing acylating agent. Application to the radioimmunoassay
.
Biochem.J
.
133
,
529
538
.
Bullock
,
G.
,
Eakins
,
M. N.
,
Sawyer
,
B. C.
&
Slater
,
T. F.
(
1974
).
Studies on bile secretion with the aid of the isolated perfused rat liver. I. Inhibitory action of sporidesmin and icterogenin
.
Proc. R. Soc. Lond. B
186
,
333
356
.
Burns
,
G. F.
,
Boyd
,
A. W.
&
Beverly
,
P. C. L.
(
1982
).
Two monoclonal anti-human T- lymphocyte antibodies have similar biologic effects and recognise the same cell surface antigen
.
J. Immun
.
129
,
1451
1457
.
Cordiner
,
S. J.
&
Jordan
,
T. W.
(
1983
).
Inhibition by sporidesmin of hepatocyte bile acid transport
.
Biochem. J
.
212
,
197
204
.
Cordiner
,
S. J.
,
Moore
,
C. H.
,
Vintiner
,
S. K.
&
Jordan
,
T. W.
(
1983
).
Does sporidesmin act on membranes?
In
Proc. Seventh Int. Cong. Soc. Human Animal Mycol
. (ed.
M.
Baxter
), pp.
456
-
459
. Palmerston North, New Zealand: Massey University Press.
David-Pfeuty
,
T.
&
Singer
,
S. J.
(
1980
).
Altered distributions of the cytoskeletal proteins vinculin and cr-actinin in cultured fibroblasts transformed by Rous sarcoma virus
.
Proc. natn. Acad. Sci. U.SA
.
77
,
6687
6691
.
Eckert
,
B. S.
(
1985
).
Alteration of intermediate filament distribution in PtKl cells by acrylamide
.
Eur.J. Cell Biol
.
37
,
169
174
.
Eichner
,
R. D.
&
Mullbacher
,
A.
(
1984
).
Hypothesis: Fungal toxins are involved in Aspergillosis and AIDS
.
Aust. J. exp. Biol. med. Sci
.
62
,
479
484
.
Frazer
,
I. H.
,
Mackay
,
I. R.
,
Jordan
,
T. W.
,
Whittingham
,
S.
&
Marzuki
,
S.
(
1985
).
Reactivity of antimitochondrial autoantibodies in primary biliary cirrhosis. Definition of two novel mitochondrial polypeptide autoantigens
.
J. Immun
.
135
,
1739
1745
.
Jordan
,
T. W.
(
1986
).
Use of indirect immunofluorescence to show changes in liver actin microfilament staining in inbred mice strains exposed to the mycotoxin sporidesmin
.
Liver
6 (in press
).
Lanks
,
K. W.
&
Kasambalides
,
E. J.
(
1983
).
Dexamethasone induces gelsolin synthesis and altered morphology in L929 cells
.
J. Cell Biol
.
96
,
577
581
.
Lehtonen
,
E.
,
Lehto
,
V.-P.
,
Bradley
,
R. A.
&
Virtanen
,
I.
(
1983
).
Formation of vinculin plaques precedes other cytoskeletal changes during retinoic acid-induced teratocarcinoma cell differentiation
.
Expl Cell Res
.
144
,
191
197
.
Middleton
,
M. C.
(
1974
).
The involvement of the disulphide group of sporidesmin in the action of the toxin on swelling and respiration of liver mitochondria
.
Biochem. Pharmac
.
23
,
811
820
.
Mortimer
,
P. H.
(
1963
).
The experimental intoxication of sheep with sporidesmin a metabolic product of Pithomyces chartarum. IV - Histological and histochemical examination of orally dosed sheep
.
Res. vet. Sci
.
4
,
166
185
.
Mortimer
,
P. H.
&
Collins
,
B. S.
(
1968
).
The in vitro toxicity of the sporidesmins and related compounds to tissue culture cells
.
Res. vet. Sci
.
9
,
136
142
.
Mortimer
,
P. H.
,
White
,
E. P.
&
Di Menna
,
M. E.
(
1978
).
Pithomycotoxicosis ‘facial eczema’ in sheep
. In
Mycotoxic Fungi, Mycotoxins, Mycotoxicoses
, vol.
2
(ed. T. D. Wylie &
L. G.
Morehouse
), pp.
195
203
.
New York
:
Marcel Dekker
.
Mullbacher
,
A.
&
Eichner
,
R. D.
(
1984
).
Immunosuppression in vitro by a metabolite of a human pathogenic fungus
.
Proc. natn. Acad. Sci. U.S A
.
81
,
3835
3837
.
Munday
,
R.
(
1982
).
Studies on the mechanism of toxicity of the mycotoxin sporidesmin. Generation of superoxide radical by sporidesmin
.
Chem. Biol. Int
.
41
,
361
374
.
Munday
,
R.
(
1984
).
Studies on the mechanism of toxicity of the mycotoxin sporidesmin. 2. Evidence for intracellular generation of superoxide radical from sporidesmin
.
J. appl. Toxicol
.
4
,
176
181
.
Paraskeva
,
C.
&
Gallimore
,
P. H.
(
1980
).
Tumorigenicity and in vitro characteristics of rat liver epithelial cells and their adenovirus-transformed derivatives
.
Int. J. Cancer
25
,
631
639
.
Sharpe
,
A. H.
,
Chen
,
L. B.
,
Murphy
,
J. R.
&
Fields
,
B. N.
(
1980
).
Specific disruption of vimentin filament organisation in monkey kidney CV-1 cells by diptheria toxin, exotoxin A, and cycloheximide
.
Proc. natn. Acad. Sci. U.S A
.
77
,
7267
7271
.
Shriver
,
K.
&
Rohrschneider
,
L.
(
1981
).
Organization of pp6Osrc and selected cytoskeletal proteins within adhesion plaques and junctions of virus transformed rat cells
.
J. Cell Biol
.
89
,
525
535
.
Small
,
I. V.
&
Langanger
,
G.
(
1981
).
Organization of actin in the leading edge of cultured cells: Influence of osmium tetroxide and dehydration on the ultrastructure of actin meshworks
.
J. Cell Biol
.
91
,
695
705
.
Taylor
,
A.
(
1971
).
The toxicology of the sporidesmins and other epipolythiadixopiperazines
. In
Microbial Toxins
, vol.
7
(ed.
S.
Kadis
,
A.
Ciegler
&
S. J.
Ajl
), pp.
337
376
.
New York
:
Academic Press
.
Taylor
,
I. W.
(
1980
).
A rapid single step staining technique for DNA analysis by flow microfluorimetry
.
J. Histochem. Cytochem
.
28
,
1021
1024
.
Toh
,
B. H.
,
Gallichio
,
H. A.
,
Jeffrey
,
P. L.
,
Livett
,
B. G.
,
Muller
,
H. K.
&
Cauchi
,
M. N.
(
1976
).
Anti-actin stains synapses
.
Nature, Land
.
264
,
648
650
.
Wang
,
E.
,
Yin
,
H. L.
,
Krueger
,
J. G.
,
Caliguiri
,
L. A.
&
Tamm
,
I.
(
1984
).
Unphosphorylated gelsolin is localized in regions of cell-substratum contact or attachment in Rous sarcoma virus- transformed rat cells
.
J. Cell Biol
.
98
,
761
771
.
Weeds
,
A.
(
1982
).
Actin-binding proteins - regulators of cell architecture and mobility
.
Nature, Lond
.
296
,
811
816
.
Zawydiwski
,
R.
&
Duncan
,
G.
(
1978
).
Spontaneous 51Cr release by isolated rat hepatocytes: an indicator of membrane damage
.
In Vitro
14
,
707
714
.