JOURNAL OF APPLIED TOXICOLOGY, VOL. 11(1), 65-74 (1991)
Evaluation of Potential Chemoprotectants against Microcystin-LR Hepatotoxicity in Mice S. J. Hermansky, S. J. Stohs,? Z. M. Eldeen and V. F. Roche School of Pharmacy and Allied Health Professions, Creighton University, Omaha, NE 68178, USA
K. A. Mereish US Army Medical Research Institute of Infectious Diseases, Fort Detrick, Frederick, M D 21701. USA
Key words: microcystin-LR; hepatotoxicity; chemoprotection; cyclosporin A ; rifampin; silymarin.
Microcystin-LR (MCLR) is a potent cyclic heptapeptide hepatotoxin produced by the blue-green algae, Microcystis aeruginosa. Toxic blooms of this cyanobacteria have been reported throughout the temperate world. In spite of the potential economic loss and health hazard posed by this toxin, few studies on the development of an antidote have been conducted. Thus, a number of biologically active compounds were tested in mice for effectiveness in preventing the toxicity of a lethal dose of MCLR (100 pg kg-I). Efficacy was evaluated based upon the percentage of surviving mice, time to death and serum lactate dehydrogenase activity 45 min after treatment with the toxin. The biologically active compounds were separated into groups based upon proposed mechanisms of action. Enzyme induction by phenobarbital but not by 2,3,7,8tetrachlorodibenzo-pdioxin (TCDD) resulted in partial protection against toxicity. Calcium channel blockers, free-radical scavengers and water-soluble antioxidants produced little protection against toxicity. The membrane-active antioxidants vitamin E and silymarin, as well as glutathione and the monoethyl ester of glutathione, produced significant protection from lethality. Rifampin and cyclosporin-A, both immunosuppressive and membrane-active agents, which also block the bile acid uptake system of hepatocytes, produced complete protection from the toxicity of MCLR. Thus, lipophilic antioxidants provide partial protection against MCLR toxicity while cyclosporin-A and rifampin are highly effective and potentially useful antidotes. The toxicity of MCLR may depend upon stimulation of the immune system and may be mediated by membrane alterations .
INTRODUCTION
The most widely recognized toxic blue-green algae is Microcystis aeruginosa. Blooms of this organism occur primarily during mid to late summer in freshwater lakes and ponds throughout the temperate world.' This cyanobacterium often forms blooms in waters where the nutrient concentrations are elevated. While the blooms are only occasionally toxic, increased concentrations of phosphorus and nitrogen or ammonia (often associated with fertilizer run-off) combined with warm temperatures may facilitate toxin production.2 Several toxins produced by M . aeruginosa are identical in structure except for two variant L-amino acids.3 The toxin containing L-leucine and L-arginine is the most extensively studied and has been designated microcystin-LR (MCLR, ~yanoginosin-LR).~ In spite of the potential for economic loss due to livestock poisoning and the hazard posed to humans by these toxins, only limited studies on the development of an antidote have been conducted. The microsomal (cytochrome P-450) enzyme inhibitors SKF-525A and cobalt chloride do not alter the lethality of the toxin.' However, the cytochrome P-450 enzyme-inducers pnapthoflavone, 3-methylcholanthrene (3-MC) and phenobarbital (Pb) provided some protection against
t
Author to whom correspondence should be addressed.
026@437W91/01006599$05.00 0 1991 by John Wiley & Sons, Ltd.
liver damage and lethality.6 Hydrocortisone also was shown to protect mice against acute and delayed microcystin-induced deaths.' In addition to the obvious economic and health benefits obtained from developing an antidote, modulation of biochemical systems by potential antidotes would provide insight into the mechanism of action of the toxin. Thus, the effects of a number of potential chemoprotectants and antidotes on a lethal dose of MCLR were assessed in mice. ~~
MATERIALS AND METHODS Animals and treatment
Female NIH non-Swiss outbred mice, weighing 20-24 g (Amitech, Omaha, NE), were used in all studies. Animals were housed under conditions of controlled temperature (24 "C) and lighting (12-h light-dark cycle) and were allowed free access to Purina Rodent Chow (Ralston Purina Co., St. Louis, MO) and tapwater. All mice were allowed to acclimatize for 3-5 days before the initiation of experiments. Microcystin-LR was provided by Dr Wayne W. Carmichael, who isolated the toxin from cultured Microcystis aeruginosa strain 7820. The toxin had a purity of >%YO, as verified by HPLC. The MCLR was stored at -80°C as a lyophilized powder. At the time of use, it was dissolved in 20% methanol in distilled Received 10 July I990 Revised 26 September 1990
66
S. .I.HERMANSKY E T A L .
water. The methanol was subsequently removed with nitrogen gas, and the resulting solution of toxin was diluted to a concentration that allowed a volume of 0.20-0.25 ml to be administered per mouse. The MCLR was administered i.p. in all studies. All biochemical modifiers were given intraperitoneally (i.p.), except where noted. The doses of biochemical modifiers were chosen based upon previously published doses that resulted in the desired effect in mice. In the event that no previously published doses were available, the dose was established based upon preliminary studies in our laboratories. All chemical modifiers were given in a volume of 0.20-0.25 ml, except where noted. Phenobarbital (PB) as the sodium salt, dimethyl sulfoxide (DMSO), buthionine sulfoximine (BSO), dithiothreitol (DTT), mannitol, allopurinol, glycerine, glucose, ethanol, zinc chloride, penicillin G, heparin, glutathione and glutathione monoethyl ester (GMEE) hydrochloride was given i.p. as solutions in distilled water. The pH of the %MEE solution was adjusted with NaOH solution to pH 7.0. Oltipraz, ellagic acid, silymarin (except where noted), verapamil, nifedipine, phenylmethylsulfonylfluoride (PMSF). indomethacin and rifampin were given i.p. as suspensions in 30% glycerine in water. Cyclosporin-A (CsA) was dissolved in a minimal amount of 95% ethanol and precipitated as a fine suspension by adjusting to a final volume with 30% glycerine in water and given i.p. Vitamin E was dissolved in olive oil and given i.p. in 0.5 ml per mouse. 2.3,7.8-Tetrachlorodibenzo-p-dioxin (TCDD) was dissolved in 10% acetone in corn oil and given i.p. Vitamin C was dissolved in water and given intragastrically. Butylated hydroxyanisole (BHA) was suspended in 30% glycerine and given intragastrically. Silymarin was prepared in two different solvent systems: 30% glycerine in water; and as a colloidal suspension in 75% propylene glycol in water, which was produced by ultrasonicating a suspension of silymarin until it cleared (30-60 min). With the exception of propylene glycol-water, the vehicles of all biochemical modifiers were shown to have no effect on the parameters measured in this study in the presence and absence of MCLR. All mice treated with a biochemical modifier received the MCLR vehicle at the appropriate time. For the sake of convenience, the biochemical modifiers were grouped into categories (Tables 1-6). The investigators recognize that some of the modifiers might be placed in more than one category based on diverse biochemical effects. Synthesis of glutathione ester Glutathione monoethyl ester (GMEE) hydrochloride was synthesized by a modification of the method of Anderson et al. Absolute ethanol was saturated with anhydrous hydrogen chloride gas and reduced glutathione was added. The mixture was maintained at &5"C (ice bath) for 4.5-5 h. Ice-cold anhydrous ethyl ether was added and the precipitate that formed was collected. The product was washed twice with cold ethyl ether and dried in vucuo. TLC analysis indicated that the product was predominantly GMEE. A yield of ca. 90"/" was obtained. This product was > 95% pure. The GMEE was purified further by column chromatography on Amberlite IRC-50 ( H + ) , using
0.05 mM EDTA (pH 5.3) as the mobile phase. Unreacted glutathione was first eluted, followed by GMEE and a small quantity of the glutathione diethyl ester. The various fractions were examined by thin-layer chromatography using silica-gel K6 plates (Whatman) and the solvent systems n-propanollacetic acid/water (10:1:5) or n-butanoUacetic acidlwater (16:3:5). The fractions from columns containing only GMEE were pooled and lyophilized, affording needle-like white crystals. The final produce was characterized by melting point, infrared spectrometry and mass spectrometry. Blood collection Blood was collected from the orbital sinus 45 min after treatment with MCLR or the vehicle using a heparinized microcapillary tube (Fisher Scientific, Pittsburgh, PA). A toxic dose of MCLR consistently produced a significant increase in lactate dehydrogenase (LDH) activity at this time point.x A minimum of 80 4 of blood was drawn from each mouse. Serum was separated via centrifugation for 4 min in a Fischer microcentrifuge, and the serum was frozen at -80°C. Lactate dehydrogenase (LDH) assay was performed on the serum within one week of collection. Preliminary studies by us and Ward el al." have shown no loss of enzyme activity under these conditions. Assessment of toxicity Lactate dehydrogenase (LDH) activity was assayed as described by Moss et al."' using 10 pI of serum. Chemicals used for the assay were of standard laboratory grade and were purchased from Sigma Chemical Company (St. Louis, MO). Studies in our laboratories have shown that LDH isoenzyme 5 increases by >300% 45 min after administration of 100 pg kg-' MCLR, indicating hepatic damage (unpublished observations). No changes were observed in the other LDH isoenzymes. These results confirm that the liver is the target of MCLR toxicity and that the increase in serum LDH is a useful indicator of MCLR hepatotoxicity. Because MCLR treatment produces acute distress in mice, presumably due to hypovolemic shock secondary to hemorrhage into the liver," 'death' was defined as the time that mice ceased to struggle, had no observable respiration and no discernable heart beat. Previous studies have shown that mice that did not exhibit signs of toxicity within 24 h after treatment with MCLR would not develop adverse effects thereafter.x Therefore, survival to 24 h was chosen as an acceptable end-point for studies using potential chemoprotectants and antidotes. Statistical methods Multiple group comparisons were made by analysis of variance (ANOVA). Posr hoc comparisons were made using Scheffe's S method. P < 0.05 was considered significant for all analyses. RESULTS Microcystin-LR consistently produced a significant increase in serum LDH activity of 15&250% 45 min
67
MICROCYSTIN-LR HEPATOTOXICITY
Table 1. Effect of enzyme inducers on the toxicity of microcystin-LR (MCLR)P Treatment group Control MCLR TCDD TCDD
+
Dose and schedule
100 p g kg-' 25 pg kg-' p.0. 48 h MCLR
before MCLR
Phenobarbital MCLR PB
80 mg kg-' i.p. for 4 days before MCLR
Phenobarbital PB + MCLR
150 m g kg-' i.p. 48 h before MCLR
Phenobarbital PB + MCLR
150 m g kg-' i.p. 1 h before MCLR
+
Time to death (min)
% Survival
61 2 8 66 2 7 248 2 38
100 0 100 0 100 0 100 50 100
-
72 and 90 65 4 7
to 24 h
0
LDH (U) 45 min after MCLR
4.03 2 1.60 1 5.1822.54 6.2721.1 2** 19.60+9.98* 4.88+0.95** 6.46+0.78** 6.31~ 2 . 3 5 ~ ~ 6.2422.11 ** 4.1420.98** 18.02+-2.01*
Effect of in vivo administration of selected enzyme inducers to female mice on the toxicity of a 100 pg kg ' i.p. dose of microcystinLR. Results are expressed as the meankSD of four mice. TCDD = 2,3,7,8-tetrachlorodibenzo-pdioxin. PB = phenobarbital. LDH = lactate dehydrogenase. * Significantly different from control group; ** significantly different from microcystin-LR-treated group (P
Evaluation of Potential Chemoprotectants against Microcystin-LR Hepatotoxicity in Mice S. J. Hermansky, S. J. Stohs,? Z. M. Eldeen and V. F. Roche School of Pharmacy and Allied Health Professions, Creighton University, Omaha, NE 68178, USA
K. A. Mereish US Army Medical Research Institute of Infectious Diseases, Fort Detrick, Frederick, M D 21701. USA
Key words: microcystin-LR; hepatotoxicity; chemoprotection; cyclosporin A ; rifampin; silymarin.
Microcystin-LR (MCLR) is a potent cyclic heptapeptide hepatotoxin produced by the blue-green algae, Microcystis aeruginosa. Toxic blooms of this cyanobacteria have been reported throughout the temperate world. In spite of the potential economic loss and health hazard posed by this toxin, few studies on the development of an antidote have been conducted. Thus, a number of biologically active compounds were tested in mice for effectiveness in preventing the toxicity of a lethal dose of MCLR (100 pg kg-I). Efficacy was evaluated based upon the percentage of surviving mice, time to death and serum lactate dehydrogenase activity 45 min after treatment with the toxin. The biologically active compounds were separated into groups based upon proposed mechanisms of action. Enzyme induction by phenobarbital but not by 2,3,7,8tetrachlorodibenzo-pdioxin (TCDD) resulted in partial protection against toxicity. Calcium channel blockers, free-radical scavengers and water-soluble antioxidants produced little protection against toxicity. The membrane-active antioxidants vitamin E and silymarin, as well as glutathione and the monoethyl ester of glutathione, produced significant protection from lethality. Rifampin and cyclosporin-A, both immunosuppressive and membrane-active agents, which also block the bile acid uptake system of hepatocytes, produced complete protection from the toxicity of MCLR. Thus, lipophilic antioxidants provide partial protection against MCLR toxicity while cyclosporin-A and rifampin are highly effective and potentially useful antidotes. The toxicity of MCLR may depend upon stimulation of the immune system and may be mediated by membrane alterations .
INTRODUCTION
The most widely recognized toxic blue-green algae is Microcystis aeruginosa. Blooms of this organism occur primarily during mid to late summer in freshwater lakes and ponds throughout the temperate world.' This cyanobacterium often forms blooms in waters where the nutrient concentrations are elevated. While the blooms are only occasionally toxic, increased concentrations of phosphorus and nitrogen or ammonia (often associated with fertilizer run-off) combined with warm temperatures may facilitate toxin production.2 Several toxins produced by M . aeruginosa are identical in structure except for two variant L-amino acids.3 The toxin containing L-leucine and L-arginine is the most extensively studied and has been designated microcystin-LR (MCLR, ~yanoginosin-LR).~ In spite of the potential for economic loss due to livestock poisoning and the hazard posed to humans by these toxins, only limited studies on the development of an antidote have been conducted. The microsomal (cytochrome P-450) enzyme inhibitors SKF-525A and cobalt chloride do not alter the lethality of the toxin.' However, the cytochrome P-450 enzyme-inducers pnapthoflavone, 3-methylcholanthrene (3-MC) and phenobarbital (Pb) provided some protection against
t
Author to whom correspondence should be addressed.
026@437W91/01006599$05.00 0 1991 by John Wiley & Sons, Ltd.
liver damage and lethality.6 Hydrocortisone also was shown to protect mice against acute and delayed microcystin-induced deaths.' In addition to the obvious economic and health benefits obtained from developing an antidote, modulation of biochemical systems by potential antidotes would provide insight into the mechanism of action of the toxin. Thus, the effects of a number of potential chemoprotectants and antidotes on a lethal dose of MCLR were assessed in mice. ~~
MATERIALS AND METHODS Animals and treatment
Female NIH non-Swiss outbred mice, weighing 20-24 g (Amitech, Omaha, NE), were used in all studies. Animals were housed under conditions of controlled temperature (24 "C) and lighting (12-h light-dark cycle) and were allowed free access to Purina Rodent Chow (Ralston Purina Co., St. Louis, MO) and tapwater. All mice were allowed to acclimatize for 3-5 days before the initiation of experiments. Microcystin-LR was provided by Dr Wayne W. Carmichael, who isolated the toxin from cultured Microcystis aeruginosa strain 7820. The toxin had a purity of >%YO, as verified by HPLC. The MCLR was stored at -80°C as a lyophilized powder. At the time of use, it was dissolved in 20% methanol in distilled Received 10 July I990 Revised 26 September 1990
66
S. .I.HERMANSKY E T A L .
water. The methanol was subsequently removed with nitrogen gas, and the resulting solution of toxin was diluted to a concentration that allowed a volume of 0.20-0.25 ml to be administered per mouse. The MCLR was administered i.p. in all studies. All biochemical modifiers were given intraperitoneally (i.p.), except where noted. The doses of biochemical modifiers were chosen based upon previously published doses that resulted in the desired effect in mice. In the event that no previously published doses were available, the dose was established based upon preliminary studies in our laboratories. All chemical modifiers were given in a volume of 0.20-0.25 ml, except where noted. Phenobarbital (PB) as the sodium salt, dimethyl sulfoxide (DMSO), buthionine sulfoximine (BSO), dithiothreitol (DTT), mannitol, allopurinol, glycerine, glucose, ethanol, zinc chloride, penicillin G, heparin, glutathione and glutathione monoethyl ester (GMEE) hydrochloride was given i.p. as solutions in distilled water. The pH of the %MEE solution was adjusted with NaOH solution to pH 7.0. Oltipraz, ellagic acid, silymarin (except where noted), verapamil, nifedipine, phenylmethylsulfonylfluoride (PMSF). indomethacin and rifampin were given i.p. as suspensions in 30% glycerine in water. Cyclosporin-A (CsA) was dissolved in a minimal amount of 95% ethanol and precipitated as a fine suspension by adjusting to a final volume with 30% glycerine in water and given i.p. Vitamin E was dissolved in olive oil and given i.p. in 0.5 ml per mouse. 2.3,7.8-Tetrachlorodibenzo-p-dioxin (TCDD) was dissolved in 10% acetone in corn oil and given i.p. Vitamin C was dissolved in water and given intragastrically. Butylated hydroxyanisole (BHA) was suspended in 30% glycerine and given intragastrically. Silymarin was prepared in two different solvent systems: 30% glycerine in water; and as a colloidal suspension in 75% propylene glycol in water, which was produced by ultrasonicating a suspension of silymarin until it cleared (30-60 min). With the exception of propylene glycol-water, the vehicles of all biochemical modifiers were shown to have no effect on the parameters measured in this study in the presence and absence of MCLR. All mice treated with a biochemical modifier received the MCLR vehicle at the appropriate time. For the sake of convenience, the biochemical modifiers were grouped into categories (Tables 1-6). The investigators recognize that some of the modifiers might be placed in more than one category based on diverse biochemical effects. Synthesis of glutathione ester Glutathione monoethyl ester (GMEE) hydrochloride was synthesized by a modification of the method of Anderson et al. Absolute ethanol was saturated with anhydrous hydrogen chloride gas and reduced glutathione was added. The mixture was maintained at &5"C (ice bath) for 4.5-5 h. Ice-cold anhydrous ethyl ether was added and the precipitate that formed was collected. The product was washed twice with cold ethyl ether and dried in vucuo. TLC analysis indicated that the product was predominantly GMEE. A yield of ca. 90"/" was obtained. This product was > 95% pure. The GMEE was purified further by column chromatography on Amberlite IRC-50 ( H + ) , using
0.05 mM EDTA (pH 5.3) as the mobile phase. Unreacted glutathione was first eluted, followed by GMEE and a small quantity of the glutathione diethyl ester. The various fractions were examined by thin-layer chromatography using silica-gel K6 plates (Whatman) and the solvent systems n-propanollacetic acid/water (10:1:5) or n-butanoUacetic acidlwater (16:3:5). The fractions from columns containing only GMEE were pooled and lyophilized, affording needle-like white crystals. The final produce was characterized by melting point, infrared spectrometry and mass spectrometry. Blood collection Blood was collected from the orbital sinus 45 min after treatment with MCLR or the vehicle using a heparinized microcapillary tube (Fisher Scientific, Pittsburgh, PA). A toxic dose of MCLR consistently produced a significant increase in lactate dehydrogenase (LDH) activity at this time point.x A minimum of 80 4 of blood was drawn from each mouse. Serum was separated via centrifugation for 4 min in a Fischer microcentrifuge, and the serum was frozen at -80°C. Lactate dehydrogenase (LDH) assay was performed on the serum within one week of collection. Preliminary studies by us and Ward el al." have shown no loss of enzyme activity under these conditions. Assessment of toxicity Lactate dehydrogenase (LDH) activity was assayed as described by Moss et al."' using 10 pI of serum. Chemicals used for the assay were of standard laboratory grade and were purchased from Sigma Chemical Company (St. Louis, MO). Studies in our laboratories have shown that LDH isoenzyme 5 increases by >300% 45 min after administration of 100 pg kg-' MCLR, indicating hepatic damage (unpublished observations). No changes were observed in the other LDH isoenzymes. These results confirm that the liver is the target of MCLR toxicity and that the increase in serum LDH is a useful indicator of MCLR hepatotoxicity. Because MCLR treatment produces acute distress in mice, presumably due to hypovolemic shock secondary to hemorrhage into the liver," 'death' was defined as the time that mice ceased to struggle, had no observable respiration and no discernable heart beat. Previous studies have shown that mice that did not exhibit signs of toxicity within 24 h after treatment with MCLR would not develop adverse effects thereafter.x Therefore, survival to 24 h was chosen as an acceptable end-point for studies using potential chemoprotectants and antidotes. Statistical methods Multiple group comparisons were made by analysis of variance (ANOVA). Posr hoc comparisons were made using Scheffe's S method. P < 0.05 was considered significant for all analyses. RESULTS Microcystin-LR consistently produced a significant increase in serum LDH activity of 15&250% 45 min
67
MICROCYSTIN-LR HEPATOTOXICITY
Table 1. Effect of enzyme inducers on the toxicity of microcystin-LR (MCLR)P Treatment group Control MCLR TCDD TCDD
+
Dose and schedule
100 p g kg-' 25 pg kg-' p.0. 48 h MCLR
before MCLR
Phenobarbital MCLR PB
80 mg kg-' i.p. for 4 days before MCLR
Phenobarbital PB + MCLR
150 m g kg-' i.p. 48 h before MCLR
Phenobarbital PB + MCLR
150 m g kg-' i.p. 1 h before MCLR
+
Time to death (min)
% Survival
61 2 8 66 2 7 248 2 38
100 0 100 0 100 0 100 50 100
-
72 and 90 65 4 7
to 24 h
0
LDH (U) 45 min after MCLR
4.03 2 1.60 1 5.1822.54 6.2721.1 2** 19.60+9.98* 4.88+0.95** 6.46+0.78** 6.31~ 2 . 3 5 ~ ~ 6.2422.11 ** 4.1420.98** 18.02+-2.01*
Effect of in vivo administration of selected enzyme inducers to female mice on the toxicity of a 100 pg kg ' i.p. dose of microcystinLR. Results are expressed as the meankSD of four mice. TCDD = 2,3,7,8-tetrachlorodibenzo-pdioxin. PB = phenobarbital. LDH = lactate dehydrogenase. * Significantly different from control group; ** significantly different from microcystin-LR-treated group (P
