Cytoskeletal changes in hepatocytes induced by Microcystis toxins and their relation to hyperphosphorylation of cell proteins

Chem.-Biol. Interactions, 81 (1992) 181-196

181

Elsevier Scientific Publishers Ireland Ltd.

CYTOSKELETAL CHANGES IN HEPATOCYTES INDUCED BY MICROCYSTIS TOXINS AND THEIR RELATION TO HYPERPHOSPHORYLATION OF CELL PROTEINS

IAN R. FALCONER and DAVID S.K. YEUNG

Department of Biochemistry, Microbiology and Nutrition, University of New England, Armidale, N.S. W., 2351 (Australia) (Received June 26th, 1991) (Revision received November 6th, 1991) (Accepted November 7th, 1991)

SUMMARY

The heptapeptide toxins produced by the blue-green alga (cyanobacterium)

Microcystis aeruginosa are selectively hepatotoxic in mammals. The characteristic post-mortem pathology of the liver is extensive lobular disruption due to sinusoidal breakdown, leakage of blood into the tissue and hepatocyte disintegration. Isolated hepatocytes incubated with toxin show severe structural deformity and surface blebbing. This paper demonstrates the effects of Microcystis toxins on the contraction and aggregation of actin microfilaments, and on the relocation and breakdown of cytokeratin intermediate filaments, in cultured hepatocytes. Earlier work did not show changes in the assembly/disassembly of actin; however, this paper demonstrates the change in cytokeratin from intermediate filaments to distributed granules in the cytoplasm of toxin-affected cells. Acrylamide gel electrophoresis of cytoskeletal fractions from hepatocytes did not show changes in total cytokeratins; however, marked changes in the immunogenicity of cytokeratins at 52 and 58 kDa were seen on toxin exposure of cells. Measurement of 32p-phosphorylation of proteins in toxin-affected cells incubated with [32p]orthophosphate showed a dramatic increase compared to control incubations. This is in agreement with research elsewhere describing phosphatase inhibition in vitro by Microcystis toxins. The data indicate that phosphorylated cytokeratin is a major component of cytoplasmic fraction phosphorylated protein after toxin exposure to hepatocytes. It is concluded that the mechanism of Microcystis toxicity to the hepatocyte is through cytoskeletal damage leading to loss of cell morphology, cell to cell adhesion and finally cellular necrosis. The underlying biochemical lesion is likely to be phosphatase inhibition Correspondence to: I.R. Falconer, Department of Biochemistry, Microbiology and Nutrition, University of New England, Armidale, N.S.W., 2351, Australia. 0009-2797/92/$05.00 © 1992 Elsevier Scientific Publishers Ireland Ltd. Printed and Published in Ireland

182

causing hyperphosphorylation of a number of hepatocyte proteins, including those cytokeratins responsible for microfilament orientation and intermediate filament integrity.

Key words: Blue-green algae -- Microcystin -- Hepatocytes -- Toxicity -Phosphorylation -- Cytoskeleton -- Intermediate filaments

INTRODUCTION

Cyanobacterial contamination of water supplies leads to potential hazards to public health, with the most frequent cause being the organism Microcystis aeruginosa [1,2]. The toxin family produced by this cyanobacterium, the microcystins, have a heptapeptide ring structure containing four D-amino acids with little molecular variation. Two L-amino acids also occur in the ring, which show considerable variation, and the seventh amino acid is a unique hydrophobic H-amino acid abbreviated to Adda (3-amino-9-methoxy-2,6,8 trimethyl-phenyldeca-4,6 dienoic acid) [3]. Toxicity is selectively targeted to the liver, as a result of the presence of the bile acid transporter which carries microcystins into cells [4,5]. The consequence of acute liver intoxication is a rapid and progressive disorganization of the hepatic architecture, a breakdown in the sinusoidal structure and pooling of blood in the liver [6,7]. Hepatocytes rupture into the bloodstream, resulting in rapid increases in liver cell enzymes in peripheral serum [8]. Chronic toxicity due to oral Microcystis ingestion in mice resulted in a generalized hepatocyte degeneration with necrosis, progressive fibrosis and mononuclear leukocyte infiltration [9] illustrative of chronic active liver injury [101. Isolated hepatocytes from rat liver when exposed to microcystins show doseand time-dependent deformation, leading to grotesque multi-blebbed cells which are initially viable [4]. The distortion of the cells is associated with progressive uptake of microcystins from solution, and can be prevented by addition of bile salts or inhibitors of bile acid transport [4,11,12]. The mechanism of microcystin toxicity has been explored extensively, both in intact animals and in isolated cell suspensions and cultures. Studies of the metabolic function of liver tissue and of hepatocytes have shown changes in mitochondrial appearance and respiration in fasted rats after microcystin, which are not seen in fed rats [13]. A clear dose-response relationship has been shown between microcystin concentration and the activation of glycogen phosphorylase in hepatocytes, which, in itself, is not toxic to cells [14]. No clear evidence of direct inhibition of protein or nucleic acid synthesis by microcystin has been reported [15]. Exploration of the possible cation ionophore properties of the cyclic peptide microcystin did not show any major activity, the only changes observed being transiently increased intracellular Ca 2÷ within the physiological range [16]. It was concluded that the observed cytoskeletal changes in the

183

deformed hepatocytes demonstrated the most probable cellular basis for toxicity, rather than changes in membrane permeability or cell metabolism [16]. The present study reports an investigation of the action of Microcystis toxins on the hepatocyte cytoskeleton, focussing on changes in microfilaments and intermediate filaments. These changes are linked with protein phosphorylation, which is accentuated in cells exposed to microcystins and appears to be the consequence of inhibition of intracellular phosphatases [17- 19]. MATERIALS AND METHODS

Microcystis extracts Blooms of the cyanobacterium Microcystis aeruginosa were collected from Malpas Dam, NSW, Australia and the toxicity evaluated. Purification of the toxic fraction proceeded to a peptide concentrate [9] prior to analytical examination by HPLC [20]. The toxic peaks were microcystin-YM and microcystin-YM sulphoxide [21]. Reagents. Cell culture materials were purchased from Commonwealth Serum Laboratories (Melbourne, Australia). Immunoreagents used for cytoskeleton visualisation were purchased from Dakopatts (Glostrup, Denmark) and from Amersham International (Amersham, U.K.). Rhodamine-linked phalloidin was obtained from Molecular Probes (Eugene, OR). Other biochemicals were purchased from BDH Chemicals (U.K.) and the Sigma Chemical Co. (U.S.A.).

Cells examined Isolated hepatocytes were obtained by perfusion of livers of rats of 200-300 g body weight with collagenase solution as earlier described [4]. Cell suspensions were used within 1 - 4 h of preparation. For culture of monolayers of primary hepatocytes, slides coated with air dried rat-tail collagen [22] were spread with 3 × 106 hepatocytes, and after 4 h washed and the medium replaced. The cell culture medium employed was Leibovitz L15 containing 8 mM NaHCO3, 20 mM Hepes, 10% v/v heat deactivated newborn calf serum, 2 rag/1 insulin, 1 mg/1 hydrocortisone, 1.5 g/1 penicillin and 1.5 g/1 streptomycin. After 24 h of culture, the attached monolayer of hepatocytes had formed an intercom necting network of high viability. Detergents used were Triton X-100 and Nonidet P40 from BDH Chemicals (U.K.), Tween 20 was obtained from Aldrich Chemical Co. (Milwaukee, WI). Reagents were Analytical Reagent grade.

Immunofluorescent staining After exposure to Microcystis toxins, or after control incubations, slides carrying hepatocyte monolayers were air dried in a laminar flow cabinet. Dry slides were stored at - 1 5 ° C in sealed containers until staining. Identification of cytokeratin employed a procedure for immunofluorescent staining from Amersham [23]. The first antibody against cytokeratin (monoclonal mouse anti-human cytokeratin, Dako.CK MNF 116 code No M821 lot 10a) was diluted 1:50 in phosphate buffered saline (pH 7.45) containing Nonidet P40 (10 #1 per 10 ml),

184 Triton-X100 (30 #l in 10 ml) and sheep serum (0.5 ml in 10 ml). The second biotinylated antibody (Amersham RPM 1001) was used at 1:200 dilution and Texas Red Streptavidin (Amersham RPM 1233) at 1:26 dilution, in the same buffer. Actin staining of air-dried hepatocyte monolayers was carried out by a similar procedure, using the same permeabilization technique, but staining the cell cytoskeleton with rhodamine-linked phalloidin (20 ~1) diluted in 400 td of phosphate-buffered saline containing Nonidet P-40 and Triton X-100, as above.

Transmission electron microscopy For transmission electron microscopy suspensions of hepatocytes were fixed at 4°C for 60 min in 2% v/v glutaraldehyde in 0.1 M sodium phosphate buffer (pH 7.4). After four washings with 0.1 M phosphate buffer the cells were postfixed in 1% osmium tetroxide (w/v) in 0.1 M sodium phosphate buffer pH 7.4, for 60 min at room temperature, dehydrated with ethanol, and embedded in Spurr's low viscosity resin. Sections stained with lead citrate were examined in a Phillips EM 300 transmission electron microscope.

Gel electrophoretic protein separation and immunoblotting Freshly isolated hepatocytes were continuously gassed with 95% 02/5% C02 in a rotating flask at 37°C for 3 h in Leibovitz L15 medium to which 5 mg of DNase (Sigma, U.S.A.) had been added to prevent clumping due to DNA strands in solution. After incubation with or without Microcystis toxins, cells were spun off at 200 × g for 5 min and washed twice. Intermediate filament fractions were obtained by the method of Franke et al. [23], employing a high-salt concentration to extract actin. Gels were run on solubilized cytoskeletal fractions by the method of Laemmli [25]. Immunoblotting used a Bio Rad Transbtot electrophoretic transfer cell and Bio Rad nitrocellulose membrane, using the manufacturer's procedures. Immunoidentification was carried out using the primary antibody against cytokeratin (Dakopatts) at 1:100 dilution in phosphate-buffered saline containing 0.1% v/v Tween 20. Bovine serum albumin (2.5% w/v) in antibody dilution buffer was used for blocking of adsorption sites on the nitrocellulose. Staining of the nitrocellulose sheets was carried out using a peroxidase-linked anti-mouse antibody (Amersham, NA 931) at 1:300 dilution, followed by peroxide/diaminobenzidine reaction (Amersham) using manufacturer's procedures.

s2p-phosphorylation of proteins in cultured hepatocytes After 24 h of culture in Leibovitz L 15 medium monolayers of rat hepatocytes were incubated with 40 ~Ci [32p]orthophosphate in 10 ml medium for 3 h. Toxin-exposed cultures were given Microcystis toxins (20 ~g/ml) for the last 2.5 h of culture. The 32P-labelled cell monolayers were washed and displaced from the collagen-coated surface. After further washing the cells were lysed in phosphate buffered saline containing Triton X-100 (0.5%) and phenyl methyl sulphonyl fluoride (2 raM) and centrifuged to obtain cytoplasmic proteins. The pellet was extracted for cytoskeletal protein enrichment as described earlier.

185 Total radioactivity was measured by scintillation counting. Gel electrophoresis was carried out as described earlier. Autoradiography on the gels used Fuji Xray film exposed within cassettes for approximately 3 days. RESULTS

Cell-cell adhesion in hepatocyte monolayers and effects of Microcystis toxins Monolayers of rat hepatocytes cultured for 24 h on glass slides coated with rattail collagen were exposed to Microcystis toxins at 20 t~g/ml of culture medium for 0, 30 or 90 min. The cells were then incubated in the presence of Trypan Blue (0.15%) for 30 rain and observed on a Zeiss phase-contrast microscope. As can be seen in Fig. 1, few dead cells have stained with Trypan Blue in any of the control or toxin-exposed cultures. Control cultures showed a linked network of adherent cells in strings and clusters. No blebbing was seen and many cells showed a polygonal shape (Fig. la). After 30 min exposure to toxin the adherent network of hepatocytes had dissociated into largely isolated cells which now showed blebbing of the cell membrane (Fig. lb). Cell-cell adhesion had been lost, presumably as a result of desmosomal attachments between cells being broken. After 90 min the cell population had been reduced on the collagen-coated surface due to cell detachment into free suspension. Many cells showed blebs almost as large as the residual cell body.

Actin distribution in toxin-affected hepatocytes The experimental approach described above was used to examine, by fluorescent visualization of actin filaments using rhodamine-phalloidin staining, the effects of microcystis toxins after 0, 30 and 180 rain. Figure 2 illustrates these results; control cultures showed an uneven distribution of actin filaments within the cells (Fig. 2a) with particular concentrations of actin at the cell periphery in regions of cell-cell adhesions. These regions contain desmosomal attachments and bile canaliculi, seen as small rings of actin, in cultured monolayers of hepatocytes. After 30 min actin is largely concentrated into aggregations within the cells and the structured and interconnected cell monolayer has dispersed into a random mass of cells. Some appearance of fibrillar but disorganised actin also occurs within the cells, better seen at higher magnification (Fig. 2b inset). At 180 min of toxin exposure, the cultured cells have lost actin altogether in many cases, with almost spherical actin aggregates in the other cells (Fig. 2c). The few cells which showed the actin distribution seen in control cultures, are cells which have been dead throughout toxin exposure.

Cytokeratin distribution in toxin-affected hepatocytes Hepatocyte cytokeratins were stained using a monoclonal antibody to cytokeratin, and a second antibody coupled to streptavidin--Texas Red. Control hepatocyte cultures showed cytokeratins as a fibrous diffuse network, with components at the cell periphery, surrounding and adjacent to the nucleus, and concentrated at regions of cell-cell adhesion (Fig. 3a). At higher magnification a

186

la

lb

1c Fig. 1. Monolayer culture of primary rat hepatocytes on collagen after 24 h. Control cells (la); after 30 min toxin exposure (lb); after 90 min toxin exposure (lc). Bar 10~m.

187

2a

2b

2c Fig. 2. Actin microfilaments observed by fluorescence microscopy in cultured hepatocyte monolayers. Control cells (2a); after 30 min toxin exposure (2b); after 180 min toxin exposure (2c). Bar 10 tLm, inset bar 3 ttm.

188

focus of cytokeratin filaments can be seen next to the nucleus, and at the regions of cell contact (Fig. 3a inset). After 30 min toxin exposure cytokeratin filaments had withdrawn from the cell periphery and now form a distorted hollow sphere around the nucleus in most cells (Fig. 3b). At higher magnification the cytokeratin appeared to have contracted around the perinuclear focus seen in control cells, leaving the cytoplasm free of stained filaments (Fig. 3b inset). After 180 min a marked change in cytokeratin distribution was seen. In place of the hollow sphere, the staining occurred as small spherical aggregates dispersed into the cytoplasm (Fig. 3c). A few cells showed a condensed focus of cytokeratin in the periphery of the cell in addition to the small aggregates (Fig. 3c inset). No filamentous cytokeratin has survived 180 min of toxin exposure. Dead cells are seen to be unaffected by the presence of toxin, with the distribution of cytokeratin as seen in control cells (Fig. 3c).

Electron microscopy of toxin-affected hepatocytes Isolated hepatocytes were incubated with or without HPLC-purified microcystin-YM (1 ~g/ml) for 20 min, and thin sections examined for morphological changes. The blebs seen in whole cells are observed as radiating projections around discrete loci in thin sections (Fig. 4). The mitochondria and endoplasmic reticulum appear normal, as does the nucleus. Cells may have more than one focus of blebbing, as seen in the cell illustrated which shows two loci. The appearance of lightly stained 'granular' holes throughout the cytoplasm reflects glycogen hydrolysis, as at 0 or 5 rain of toxin exposure dark-staining glycogen granules are similarly distributed.

Gel electrophoretic separation and immunological identification of cytoskeletal proteins from toxin-affected hepatocytes Enriched cytoskeletal proteins from hepatocytes, after actin extraction in high-salt concentration buffer, were dissociated and separated on acrylamide gel electrophoresis. Control suspensions of freshly isolated hepatocytes were compared with similar hepatocytes incubated with microcystis toxins (20 ~g/ml) for 60 rain. Visual examination of silver-stained gels run at the same protein concentrations did not show any marked qualitative or quantitative changes in cytoskeletal proteins, or redistribution of comparable molecular weight bands (Fig. 5). Laser scans of representative gels confirmed the similarities of protein quantity and molecular weight distribution. Immuno-blotting of unstained gels onto nitrocellulose, followed by cytokeratin identification by reaction with a monoclonal antibody to cytokeratin, followed by a peroxidase-linked second antibody, gave a different outcome. Clear cytokeratin bands were seen at 52 and 58 kDa molecular weight in cytoskeletal extracts from both control and toxin-treated hepatocytes. However, visually and by laser scanning densitometry, the staining of 52 and 58 kDa cytokeratins in the toxintreated cell extract was reduced by approximately 60%. The low molecular weight material reacting with the anti-cytokeratin on the nitrocellulose sheet was markedly increased in extracts from toxin-treated cells whereas other bands appeared similar (Fig. 6).

189

3a

3b

Fig. 3. Cytokeratin intermediate filaments observed by fluorescence microscopy in cultured hepatocyte monolayers. Control cells (3a); after 30 rain toxin exposure (3b); after 180 min toxin exposure (3c). Bar 10 ~m, inset bar 3 ~m.

190

4 Fig. 4. Isolated hepatocyte exposed to toxin for 30 rain showing blebs focussed round fibrous aggregates (I). Bar 10 #m.

Effect of Microeystis toxins on 32p-incorporation into proteins in cultured hepatocytes Control and toxin-treated hepatocytes incorporated [32P]orthophosphate during the last 3 h of culture. Simultaneous exposure to toxins for 2.5 h of culture was followed by displacement of the cells from the collagen-coated surface, washing and lysis by Triton X 100. Radioactivity measurement of cytoplasmic lysates from toxin-treated cultures showed seven-fold higher 32p content than control cell lysates, at the same protein concentration. This difference was also reflected in autoradiographs of 32p-labelled proteins from the two sources, as shown in Fig. 7. Laser densitometer scans quantitated this increase in protein phosphorylation resulting from toxin exposure of cells (Fig. 7).

191

1.0

Relative Absorbance (633 nm)

5a1

0.5

47k Dalton

52k Dalton

58k Dalton

1.0

0.5

47k Dalton

52k Dalton 58k Dalton

Fig. 5. L a s e r scan absorbance m e a s u r e m e n t s on SDS-gel electrophoretic s e p a r a t i o n s of i n t e r m e d i a t e filament enriched fractions from control h e p a t o c y t e s (5a); and h e p a t o c y t e s exposed to toxin for 60 min (5b).

Measurement of 32p in the cytoskeletal proteins of the enriched intermediate filament fraction after actin extraction showed only a small increase in toxintreated samples. This was not clearly detectable by autoradiography of acrylamide gel separations of cytokeratins. DISCUSSION

Studies on the effects of Micvocystis toxins on the liver have consistently shown the breakdown of hepatic architecture and disorganization of hepatocytes [6,13]. In culture, primary hepatocytes which formed an interlinked network were similarly dissociated and deformed in the presence of toxins (Fig. 1). The deformations of the hepatocyte are evidence of major changes in the cytoskeleton, both within each cell and in cell-cell interactions.

192

1.0

Relative 0.5 Absorbance (633nm)

10[

47

52 k Dalton

6b

0.5

47 k Dalton

Fig. 6. Laser scan absorbance measurements of an immunoblot against anti-cytokeratin, from an S D S acrylamide gel separatio n of intermediate filament enriched fractions. Control hepatocytes (6a); hepatocytes exposed to toxin for 60 min (6b)

The hepatocyte cytoskeleton has been extensively characterised [26]. The actin microfilament network is highly labile, in continuous change of cellular location and in actin filament association-dissociation [27]. In the region of the bile canaliculi, actin filaments have been observed to form a contractile ring [26]. In this study, the actin network within the hepatocyte can be clearly seen round the periphery of the cell, with particular concentration at points of cell to cell contact. Microcystis toxins have not been shown to cause changes in association/dissociation of actin, filaments [28,29]. However, evidence in this paper and in the paper by Eriksson et al. [27] shows a clear redistribution of actin microfilaments on toxin exposure (Fig. 2). Actin appears to be progressively aggregated into a single focus in each cell by withdrawal of filaments from the peripheral network, and from cell-cell attachment and bile canicular regions. The observed cell blebbing occurs at the same time as actin withdrawal (Fig. 1).

193

1.0

Relative

0.5

Absorbance (6~nm)

40

60 70 k Dalton

90

10

40

60 70 k Dalton

90

Fig. 7. Laser scan absorbance measurements from a ~2p-autoradiograph of an SDS acrylamide gel separation of cytoplasmic proteins from control hepatocytes incubated with [32p]orthophosphate (7a); and hepatocytes exposed to [32p]orthophosphate and toxin for 150 min (7b).

The cytokeratin intermediate filaments are generally regarded as highly stable, acting as a scaffolding within the cytoplasm [26]. They occur in high concentrations round the periphery of the nucleus, and appear to relate to an organising-centre in this region (Fig. 3). From the perinuclear sphere of filaments, fibres cross the cytoplasm and occur in large numbers near bile canaliculi and desmosomes [26]. These filaments are regarded as providing the support for the actin microfilaments which generate the contractility of the canaliculi. On exposure to toxins the cytokeratin filaments contract away from the cell-cell contact areas into a dense ragged sphere around the nucleus, with a focus on the organising centre (Fig. 3). Later during toxin exposure, these filamentous aggregations break up into a large number of small spheres of cytokeratin, with no filamentous structure (Fig. 3c). This rearrangement of cytokeratins has been observed previously, during mitosis in transformed cell

194

cultures [30]. This reorganisation is associated with the phosphorylation of cytokeratin and other polypeptides which occurs during mitosis [31,32]. The withdrawal of the intermediate filaments from the region of the desmosomes and bile canaliculi can be expected to weaken the attachments between cells, leading to structural breakdown of the tissue and deformation of isolated cells or cell monolayers. The ultrastructure of the toxin-deformed hepatocytes shows an obvious focus (or loci) of material of fibrous appearance forming the base of the protruding blebs of cytoplasm. These foci are considered likely to be the contracted aggregations of actin microfilaments which were also demonstrated by immunofluorescent staining. Recent studies on phosphatase inhibition in vitro [17,18] and in vivo [19] by Microcystis toxins demonstrate a highly specific focus of toxic action. The results presented here show increases in intracellular protein phosphorylation in cultured hepatocytes exposed to toxins. The increased phosphorylation within the hepatocyte is coincident with cell deformation, loss of cell-cell contact and contraction of actin and cytokeratin filaments. This research supports the hypothesis that phosphorylation of intermediate filament cytokeratins following toxin exposure results in filament disassembly and consequent appearance of phosphorylated cytokeratins in the 'cytoplasmic' fraction (Fig. 7). This is in agreement with studies on mitosis in hamster kidney cells, which show coincident phosphorylation of intermediate filament proteins and filament dissociation into cytoplasmic aggregates [32]. It is also supported by evidence that microcystin dosing to rats results in a loss of cytokeratins from the particulate desmosome/plasma membrane fraction obtained from liver [33]. It is concluded that the primary mechanism of Microcystis toxin action is the inhibition of phosphatases, leading to hyperphosphorylation of a variety of cell proteins, which in turn affects cytoskeletal coherence. In particular actin microfilaments contract, and cytokeratin intermediate filaments withdraw to around the nucleus prior to filament loss and the appearance of granular cytokeratin in the cytoplasm. The loss of cytokeratin intermediate filaments associated with phosphorylation in toxin-treated cells, and at mitosis, may link the potential tumour-promoting activity of Microcystis-toxins with that of okadaic acid, a known tumour promoter [34]. ACKNOWLEDGEMENTS

We wish to thank the Australian Research Council for financial support for this investigation, Mr Stuart Forsyth, Mrs Mandy Choice and Dr Robin Hesketh (of Cambridge University, U.K.) for help and advice, Mrs Shelley Harvey for preparation of this manuscript and the Electron Microscope Unit and Media Resources Unit for photographs of ultrastructural and optical microscopy. REFERENCES 1 P.R. Hawkins, M.T.C. Runnegar, A.R.B. Jackson and I.R. Falconer, Severe hepatotoxicity caused by the tropical cyanobacterium (blue-green alga) Cylindrospermopsis racborskii

195

2 3

4

5

6 7 8

9

10

11 12 13 14

15

16

17

18

19

20

21

(Wolosz.) Seenaya and Subba Raju isolated from a domestic water supply reservoir, Appl. Environ. Microbiol., 50 (1985) 1292-1295. I.R. Falconer, A. Beresford and M.T.C Runnegar, Evidence of liver damage by toxin from a bloom of the blue-green alga, Microcystis aeruginosa, Med. J. Aust., 1 (1983) 511-514. W.W. Carmichael, V. Beasley, D.L. Bunner, J.N. Eloff, I. Falconer, P. Gorham, K. Harada, Y. Min-Juan, T. Krishnamurthy, R.E. Moore, K. Rinehart, M. Runnegar, O.M. Skulberg and M. Watanabe, Naming of cyclic heptapeptide toxins of cyanobacteria (blue-green algae), Toxicon 26 (1988) 971-973. M.T.C. Runnegar, I.R. Falconer and J. Silver, Deformation of isolated rat hepatocytes by a peptide hepatotoxin from the blue-green alga Microcystis aeruginosa, Naunyn-Schmiedeberg's Arch. Pbarmakol., 317 (1981) 268-272. W.L. Thompson, K.A. Bostian, N.A. Robinson and J.G. Pace, Protective effects of bile acids on cultured hepatocytes exposed to the hepatotoxin Microcystis, FASEB J. 2 (1988) A825, No. 3077. I.R. Falconer, A.R.B. Jackson, J. Langley and M.T. Runnegar, Liver pathology in mice in poisoning by the blue-green alga Microcystis aeruginosa, Aust. J. Biol. Sci., 34 (1981) 179-187. W.H. Adams, R.D. Stoner, D.G. Adams, D.N. Slatkin and H.W. Siegelman, Pathophysiologic effects of a toxic peptide from Microcystis aeruginosa, Toxicon, 23 (1985) 441- 447. M.T. Runnegar and I.R. Falconer, The in vivo and in vitro biological effects of the peptide hepatotoxin from the blue-green alga Microcystis aeruginosa, S. Afr. J. Sci., 78 {1982) 363 - 366. T.C. Elleman, I.R. Falconer, A.R.B. Jackson and M.T. Runnegar, Isolation, characterization and pathology of the toxin from a Microcystis aeruginosa (= Anacystis cyanea) bloom, Aust. J. Biol. Sci., 31 {1978) 209-218. I.R. Falconer, J.V. Smith, A.D. Jackson, A. Jones and M. Runnegar, Oral toxicity of a bloom of the cyanobacterium Microcystis aeruginosa administered to mice over periods up to one year. J. Toxicol. Environ. Health, 24 (1988) 291-305. M.T. Runnegar, R.G. Gerdes and I.R. Falconer, The uptake of the cyanobacterial hepatotoxin microcystin by isolated rat hepatocytes, Toxicon, 29 (1991) 43-51. M. Frimmer and K. Ziegler, The transport of bile acids in liver cells, Biochim. Biophys. Acta, 947 (1988) 75-99. G.A. Miura, N.A. Robinson, W.B. Lawrence and J.G. Pace, Hepatoxicity of microcystin-LR fed and fasted rats, Toxicon, 29 (1991) 337-346. M.T.C. Runnegar, J. Andrews, R.G. Gerdes, and I.R. Falconer, Injury to hepatocytes induced by a peptide toxin from the cyanobacterium Microcystis aeruginosa, Toxicon, 25 (1987) 1235 - 1239. M.T. Runnegar and I.R. Falconer, Isolation, characterization and pathology of the toxin from the blue-green alga Microcystis aeruginosa, in: W.W. Carmichael (Ed.), The Water Environment -- Algal Toxins and Health, Plenum Press, New York, 1981, 325-342. I.R. Falconer and M.T. Runnegar, Effects of the peptide toxin from Microcystis aeruginosa on intracellular calcium, pH and membrane integrity in mammalian cells, Chem.-Biol. Interact., 63 (1987) 215- 225. C. MacKintosh, K.A. Beattie, C. Klumpp, C. Cohen and G.A. Codd, Cyanobacterial microcystinLR is a potent and specific inhibitor of protein phosphatases 1 and 2A from both mammals and higher plants, FEBS Lett., 264 (1990) 187-192. S. Yoshizawa, R. Matsushima, M.F. Watanabe, K. Harada, A. Ichihara, W.W. Carmichael and H. Fujiki, Inhibition of protein phosphatases by microcystis and nodularin associated with hepatotoxicity, J. Cancer Res. Clin. Oncol., 116 (1990) 609-614 J.E. Eriksson, D. Toivola, J.A.O. Meriluoto, H. Karaki, Y.-G. Han and D. Hartshorne, Hepatocyte deformation induced by cyanobacterial toxins reflects inhibition of protein phosphatases, Biochem. Biophys. Res. Commun., 173 (1990) 1347-1353 M.T.C. Runnegar, I.R. Falconer, T. Buckley and A.R.B. Jackson, Lethal potency and tissue distribution of 125I-labelled toxic peptides from the blue-green alga Microcystis aeruginosa, Toxicon, 24 (1986) 506-509. D.P. Botes, P.L. Wessels, H. Kruger, M.T.C. Runnegar, S. Santicarn, R.J. Smith, J.C.J. Barna

196

22 23 24

25 26

27 28

29

30

31

32 33

34

and D.H. Williams, Structural studies on cyanoginosins-LR, -YR, -YA, and -YM, peptide toxins from Microcystis aeruginosa. Ji Chem. Soc. Perkin Trans., 1 (1985) 2747-2748. K. Rubin, A. Oldberg, M. Hook and B. Obrink, Adhesion of rat hepatocytes to collagen. Exp. Cell Res., 117 (1978) 165-177. Amersham International plc, Protocol for Use of the Biotin-Streptavidin System, 1990. W.W. Franke, H. Denk, R. Kalt and E. Schmid, Biochemical and immunological identification of cytokeratin proteins present in hepatocytes of mammalian liver tissues, Exp. Cell Res., 131 (1981) 299- 318. U.K. Laemmli, Cleavage of structural proteins during the assembly of the head of bacteriophage T4, Nature (Lond.), 227 (1970) 680-685. M.J. Phillips and P. Satir, The cytoskeleton of the hepatocyte: organisation, relationships and pathology, in: I.M. Arias, W.B. Jackoby, H. Popper, D. Schachter and D.A. Shafritz (Eds.), The Liver: Biology and Pathobiology, Rouen Press, N.Y., 1988, p. 11-26. Y.-L. Wang, Mobility of filamentous actin in living cytoplasm, J. Cell. Biol., 105 (1987) 2811-2816. J.E. Eriksson, G.I.L. Paatero, J.A.O. Meriluoto, G.A. Codd, G.E.N. Kass, P. Nicotera and S. Orrenius, Rapid microfilament reorganisation induced in isolated rat hepatocytes by microcystin-LR, a cyclic peptide toxin, Exp. Cell Res., 185 (1989) 86-100. M.T. Runnegar and I.R. Falconer, Effect of toxin from the cyanobacterium Microcystis aeruginosa on ultrastructural morphology and actin polymerization in isolated hepatocytes, Toxicon, 24 {1986) 109- 115. W.W. Franke, E. Schmid, C. Grund and G.B. Geiger, Intermediate filament proteins in nonfilamentous structures: transient disintegration and inclusion of subunit proteins in granular aggregates, Cell, 30 (1982) 103-113. J.E. Celis, P.M. Larsen, S.J. Fey and A. Celis, Phosphorylation of keratin and vimentin polypeptides in normal and transformed mitotic human epithelial amnion cells: behaviour of keratin and vimentin filaments during mitosis. J. Cell. Biol., 97 (1983) 1429- 1434. Y.-H. Chou, E. Rosevear and R.D. Goldman, Phosphorylation and disassembly of intermediate filaments in mitotic cells, Proc. Natl. Acad. Sci. U.S.A., 86 (1989) 1885-1889. G.A. Miura, N.A. Robinson, T.W. Geisbert, K.A. Bostian, J.D. White and J.G. Pace, Comparison of in vivo and in vitro toxic effects of microcystin-LR in fasted rats, Toxicon, 27 (1989) 1229 - 1240. M. Suganuma, H. Fujiki, H. Suguri, S. Yoshizawa, M. Hirota, M. Nakayasu, M. Ojika, K. Wakamatsu, K. Yamada and T. Sugimura, Okadaic acid: An additional nonphorbol-12-tetradecanoate-13-acetate-type tumor promoter in mouse skin. Proc. Nat]. Acad. Sci. U.S.A., 85 (1988) 1768-1771.