- Email: [email protected]
Species differences in the hepatotoxicity of coumarin: A comparison of rat and Mongolian gerbil
Toxicology, 71 (1992) 129-136 Elsevier Scientific Publishers Ireland Ltd.
129
Species differences in the hepatotoxicity of coumarin: A comparison of rat and Mongolian gerbil Julia H. F e n t e m a, Jeffrey R. F r y a and N o r m a n W. T h o m a s b Departments of aphysiology and Pharmacology and hHuman Morphology, Medical School, Queen's Medical Centre, Nottingham NG7 2 UH ( U. K.) (Received July 3rd, 1991; accepted October 23rd, 1991)
Summary The acute hepatic effects o f c o u m a r i n (2H-l-benzopyran-2-one) in male Wistar rats and Mongolian gerbils has been compared. A single dose of coumarin (125 mg/kg, intraperitoneally (i.p.)) was hepatotoxic to rats within 24 h as assessed by its effects on a variety of hepatic parameters. Coumarin-induced hepatotoxicity was associated with significant increases in relative liver weight, plasma alanine and aspartate aminotransferase activities and hepatic non-protein sulphydryl groups. Cytochrome P-450 content and 7-ethoxycoumarin O-deethylase and glucose 6-phosphatase activities were significantly lower in coumarin-treated compared with control rats. Centrilobular necrosis was only observed in two out of six rats at this dose, but was present in all four coumarin-treated rats when the dose was increased to 150 mg/kg. In contrast to the effects observed in the rat, no evidence was found for coumarin-induced hepatotoxicity in gerbils following a single i.p. dose of 125 mg/kg. These data indicate that the gerbil is less sensitive to the hepatotoxic effects of coumarin than the rat.
Key words." Coumarin; Gerbil; Hepatotoxicity; Rat
lntoduction Coumarin (2H-l-benzopyran-2-one) is a naturally occurring plant product which was first isolated from the Tonka bean by Vogel in 1820 [1]. It was banned from use as a food flavouring agent in the mid-1950s when it was shown to cause liver necrosis in rats [2], but is still used in perfumes, toilet soaps, toothpastes and some tobacco products [3]. Coumarin has several potential clinical applications [4] and has been used for the treatment of various cancers [5-7], high protein oedema [8], brucellosis and certain other chronic infections [9]. Liver toxicity in patients receiving relatively high daily doses of coumarin is very rare [9,10]. Species differences in coumarin hepatotoxicity are thought to be metabolism-
Correspondence to: Jeffrey R. Fry. 0300-483X/92/$05.00 © 1992 Elsevier Scientific Publishers Ireland Ltd. Printed and Published in Ireland
130 mediated [3,i1,12]. The rat, in which it is markedly hepatotoxic, primarily metabolizes coumarin via 3-hydroxylation and cleavage of the heterocyclic ring. Coumarin is less toxic in the baboon [131 and DBA/2J mouse [14], species which resemble man [15] in their extensive formation of the 7-hydroxy metabolite. 7-Hydroxylation of coumarin is negligible in the rat, making it a poor experimental model for man with respect to coumarin hepatotoxicity. Studies reported by Ritschel and Hardt [16] indicated that the pharmacokinetic profiles of coumarin, 7-hydroxycoumarin and 7-hydroxycoumarin glucuronide in the blood of Mongolian gerbils were similar to those observed in man. We have previously demonstrated that coumarin 7-hydroxylase activity was over 100-fold greater in gerbil compared with rat liver microsomes [17,18]. On the basis of these findings we decided that a more comprehensive investigation of coumarin metabolism and hepatotoxicity in the gerbil was warranted. In this study we have compared the acute hepatotoxicity of coumarin in male Wistar rats and Mongolian gerbils. Materials and methods
Chemicals Coumarin, l-amino-2-naphthol-4-sulphonic acid (Fiske and Subbarow reducer), glucose 6-phosphate, glucose 6-phosphate dehydrogenase, 7-hydroxy-coumarin and NADP and kits for the determination of alanine aminotransferase (ALT) and aspartate aminotransferase (AST) activities, were purchased from Sigma (Dorset, U.KI). 7-Ethoxycoumarin was synthesized according to the method of Ullrich and Weber [19]. All other chemicals were of analytical grade. Animals and treatment Male Wistar albino rats (8 weeks old; 180-200 g) and male Mongolian gerbils (Meriones unguiculatus; 8 weeks old; 60-65 g) were obtained from the University of Nottingham Joint Animal Breeding Unit (Sutton Bonington, Leics., U.K.). They were housed in small groups at a constant room temperature of 22°C and allowed free access to standard laboratory diet (Diet 41B, Heygates (Pilsbury) York, U.K.) and tap water. Rats and gerbils received single i.p. injections of coumarin (125 or 150 mg/kg); control animals were given corresponding volumes (5 ml/kg) of the arachis oil vehicle. They were killed 24 h later following overnight starvation. Blood was taken, for the determination of plasma ALT and AST activities, by cardiac puncture under haiothane anaesthesia. The livers were removed and weighed; samples were coded and taken for histology. The remaining liver was used for biochemical studies. Histology Thin slices from each coded liver sample were immersed in 10°/,, neutral formalin, Helly's or Bouin Hollande fixatives and, after appropriate treatment, embedded in paraffin wax. Sections (8 t~m) were stained with haematoxylin and eosin and examined by light microscopy. Only when the examination had been completed was the code broken. Biochemical investigations The non-protein sulphydryl groups in whole liver homogenates (0.25 g fresh
131
tissue/ml) were determined by the method of Sedlak and Lindsay [20]. Liver microsomal fractions were prepared by the calcium aggregation technique, as outlined previously [21]. Protein was determined by the method of Lowry et al. [22] using bovine serum albumin as standard. Hepatic microsomal cytochrome P-450 content [23] and 7-ethoxycoumarin O-deethylase [241 and glucose 6-phosphatase [25] activities, were assayed by the methods quoted. Plasma ALT and AST activities were measured using standard diagnostic test kits.
Statistical analysis Values for control and coumarin-treated animals were compared using Student's t-tests or Mann-Whitney U-tests (plasma ALT and AST activities) as appropriate. Results
A single dose of coumarin (125 mg/kg, i.p.) was hepatotoxic to rats within 24 h, as assessed by its effects on a variety of hepatic parameters (Table 1). Coumarininduced hepatotoxicity in the rat was manifested by significant, dose-dependent, in-
TABLE I EFFECT OF C O U M A R I N T R E A T M E N T ON SOME H E P A T I C P A R A M E T E R S IN THE RAT AND GERBIL Results are expressed as mean + S.E.M. or medians (range) for groups of four or six animals 01). ALT, alanine aminotransferase; AST, aspartate aminotransferase; b.w., body wt; ECOD, 7-ethoxycoumarin Odeethylase; NPS, non-protein sulphydryl: n.d., not determined. Parameter
Relative liver weight (g liver/100 g b.w.) ALT (#mol/min per I) AST (#mol/min per 1) Cytochrome P-450 (nmol/mg protein) ECOD (nmol/min per mg protein) Glucose 6-phosphatase (nmol/min per mg protein) NPS groups (#mol/g liver)
Rat
Gerbil
Control (n = 6)
125 mg/kg (n = 6)
150 mg/kg (n = 4)
Control (n = 6)
125 mg/kg (n = 6t
3.6 + 0.1
4.2 + 0.1"*
4.5 + 0.1"**
3.5 4. 0.2
3.7 + 0.1
25 (21-28) 41 (37-81) 1.1 + 0.0
87** (42-4632) 194" (57-7585) 0.7 4. 0.1"**
939** (518-2970) 2184"* (1167-6824) n.d.
65 (55-104) 157 (111-311) 1.1 4. 0.1
60 (37-1 I1) 119 (101-179) 1.0 -4- 0.0
2.9 4. 0.2
1.5 4- 0.3**
n.d.
5.6 4. 0.4
4.7 4- 0.3
n.d.
362 + 20
333 4. 20
415 4. 11
4.3 4. 0.4
292 4- 7***
5.8 ± 0.2**
n.d.
5.8 + 0.4
6.6 + 0.3
Values are significantly different from the corresponding controls (arachis oil-treated) at *P < 0.05, **P < 0.01 and ***P < 0.001 (Student's t-test or Mann -Whitney U-test).
132
Fig. 1. Photomicrographs of rat liver illustrating the effect of different doses of coumarin. (A) illustrates the typical appearance of the control liver with a radial disposition of sinusoids about a central vein (c). (B,C) illustrate the range of response which was observed in different rats following a coumarin dose of 125 mg/kg. In (B) cellular infiltration without hepatic necrosis characterizes the tissue surrounding a central vein (c), while in (C) the small arrows define the junction between a region of centrilobular necrosis and the normal disposition of sinusoids about a portal triad (p). (DI illustrates the appearance of the liver in rats receiving a dose of 150 mg/kg; the small arrows define the boundary of a region of centrilobular necrosis around the axis of a central vein (c), while the parenchyma about the axis of a portal triad (p) retains a normal appearance; original magnification, x 150.
133 creases in relative liver weight and plasma A L T and A S T activities. C y t o c h r o m e P-450 content and 7-ethoxycoumarin O-deethylase and glucose 6-phosphatase activities were all significantly lower in coumarin-treated c o m p a r e d with control rats. Hepatic non-protein sulphydryl groups (mainly glutathione) were significantly increased 24 h after coumarin administration. In contrast to the effects o f coumarin in the rat, none o f the hepatic parameters measured were significantly altered by coumarin treatment of gerbils. Histological examination o f liver sections indicated a range o f responses amongst the group o f six rats which received a coumarin dose o f 125 mg/kg (Fig. 1). N o histopathological changes were detected in three rats, cellular infiltration without hepatic necrosis was present in one rat and centrilobular necrosis was found in two rats. In one o f these the necrosis was extensive and at the time o f tissue collection it was noted that the liver o f this particular rat was grossly enlarged (4.9 g liver/100 g b o d y wt) and had a mottled appearance. All tissue samples from rats which were given a coumarin dose o f 150 mg/kg exhibited centrilobular hepatic necrosis (Fig. 1) although in one rat the necrosis was not as extensive as in the rest o f the group. There was no evidence for coumarin-induced histopathological changes in the livers o f gerbils following a single dose of 125 mg/kg (Fig. 2).
Fig. 2. Photomicrographs of the liver from control (A) and coumarin-treated (B) gerbils. At a dose of 125 mg/kg there was no detectable histological difference between the control and experimental groups; (c) central vein; × 150.
134
Discussion A single i.p. injection o f c o u m a r i n at a dose of 125 mg/kg was hepatotoxic to rats as judged by a variety of blood and liver parameters. Thus, relative liver weight was significantly increased compared with controls, plasma A L T and AST activities were elevated and the hepatic microsomal parameters were all significantly decreased. Relative liver weight and plasma A L T and AST activities have previously been demonstrated to represent sensitive markers of xenobiotic-mediated hepatotoxicity [26]. These data are entirely consistent with those reported by Lake et al. [12]. In the present study a range of histopathological changes was observed within the group of young male Wistar rats administered coumarin at this dose. Three out of the six rats showed no histological evidence of liver damage and in one there was cellular infiltration but no hepatic necrosis. Centrilobular necrosis was observed in the other two rats, with this being very widespread in one animal. The plasma A L T (4632 #mol/min per 1) and AST (7585 #mol/min per 1) activities for this rat were markedly elevated compared to those for the other rats in the same group (ALT: 42-143; AST: 57-375 t~mol/min per l). The reason for the extreme sensitivity of this particular rat to the same dose of coumarin as the other treated animals is unknown, but may be related to it having a lower non-protein sulphydryl content or being compromised in some other way prior to treatment. The range of histopathological responses observed was in close agreement with the results of the biochemical investigations; there was a good correlation between the rankings of individual rats on the basis of histopathoiogy and the hepatic parameters measured. Interestingly, the histopathological observations in this study contrast with those of Lake et al. [12] who reported that, in young male Sprague-Dawley rats, a single i.p. injection of coumarin at 125 mg/kg induced necrosis of the centrilobular hepatocytes together with some vacuolation of cells in the periportal region. On increasing the dose to 150 mg/kg centrilobular hepatic necrosis was found in all rats and this was accompanied by further increases in A L T and AST activities compared with rats given 125 mg coumarin/kg. In one rat the necrosis was less extensive than in the rest of the group; the values for relative liver weight and A L T and AST activities were correspondingly lower than those for the other rats similarly treated. The coumarin dose was not increased above 150 mg/kg to avoid other symptoms of clinical toxicity, including narcotic effects. It is apparent from these data that a dose of 125 mg/kg is on the lower end of the coumarin dose-response relationship for the rat. In contrast to the effects observed in the rat, coumarin was not hepatotoxic to gerbils following a single i.p. dose of 125 mg/kg, so demonstrating a relative insensitivity of the gerbil to coumarin-induced hepatotoxicity. At least two possible explanations can be proposed for t h i s species difference in sensitivity to coumarin hepatotoxicity - one based on coumarin metabolism and one based on the abundance of protective nucleophiles. The two principal routes of hepatic metabolism of coumarin involve oxidation on the aromatic ring (particularly at the 7-position) and oxidation on the pyrone ring (at the 3,4-double bond). Oxidation on the pyrone ring accounts for the bulk of coumarin metabolism in rat liver microsomes and Lake et al. [12] have proposed that
135 coumarin-induced hepatotoxicity in the rat is due to its bioactivation by cytochrome P-450-dependent enzymes to a toxic metabolite(s), postulated to be a 3,4-epoxide intermediate. Aromatic hydroxylation is a major metabolic route in gerbil liver microsomes [17,18] and it is possible that this represents the principal route of coumarin metabolism in gerbils in vivo. Thus, differences between rat and gerbil in their susceptibility to coumarin-induced hepatotoxicity may be related to differences in routes of coumarin metabolism. Further studies are required to test this hypothesis. Lake et al. [12] demonstrated an initial decrease in the non-protein sulphydryl content of liver homogenates measured 2 h after dosing rats with coumarin. This then recovered and by 24 h had increased to a level significantly greater than that found with control rats; this latter aspect was confirmed in the present study (Table I). Glutathione protects against coumarin-induced hepatotoxicity in the rat, possibly by the formation of conjugates with the reactive coumarin metabolite(s). The nonprotein sulphydryl content (principally glutathione) of liver homogenates prepared from control gerbils is considerably higher than that for control rats ([17] and Table I). This may afford the gerbil greater protection than the rat from any toxic intermediates formed during coumarin metabolism. In summary, no evidence was found for coumarin-induced hepatotoxicity in the Mongolian gerbil at a dose of coumarin that was hepatotoxic in the rat. The rarity of liver toxicity in patients receiving relatively high daily doses of coumarin [9,10] and the similarity of gerbil and man in the extent of coumarin 7-hydroxylation, both in vivo [16] and in vitro [17,18], point to the gerbil being a more appropriate species than the rat for studies aimed at assessing the toxicological hazard, if any, of coumarin to man.
Acknowledgements We would like to thank Mr. I. Janson and Ms. S. Ecob for their help with the animal procedures and Ms. P. Barnes for the preparation of liver sections for histological examination. Financial support for this study was provided by the University of Nottingham in the form of a studentship to J.H. Fentem.
References 1 G. Feuer, The metabolism and biological actions ofcoumarins, in G.P. Ellis and G.B. West (Eds.), Progress in Medicinal Chemistry, Vol. 10, Elsevier, Amsterdam, 1974, p, 85. 2 L.W. Hazleton, T.W. Tusing, B.R. Zeitlin, R. Thiessen and H.K. Murer, Toxicity of coumarin. J. Pharmacol. Exp. Ther., 118 (1956) 348. 3 A.J. Cohen, Critical review of the toxicologyof coumarin with special reference to interspecies differences in metabolism and hepatotoxic response and their significance to man. Food Cosmet, Toxicol., 17 (1979) 277. 4 D. Egan, R. O'Kennedy, E. Moran, D. Cox, E. Prosser and R.D. Thornes, The pharmacology,metabolism, analysis and applications of coumarin and coumarin-related compounds. Drug Metab. Rev., 22 (1990) 503. 5 M.E. Marshall, L. Mendelsohn, K. Butler, L. Riley, J. Cantrell, J. Harvey, C. Wiseman, T. Taylor and J,S. Macdonald, Treatment of metastatic renal cell carcinoma with coumarin (l,2-benzopyrone) and cimetidine: a pilot study. J. Clin. Oncol., 5 (1987) 862.
136 6
7 8
9 10
11 12
13 14 15 16
17
18 19 20 21 22 23 24 25
26
M.E. Marshall, K. Butler, J. Cantrell, C. Wiseman and L. Mendelsobn, Treatment of advanced malignant melanoma with coumarin and cimetidine: a pilot study. Cancer Chemother. Pharmacol., 24 (1989) 65. M.E. Marshall, K. Butler and D. Hermansen, Treatment of hormone-refractory stage D carcinoma of prostate with coumarin (I,2-benzopyrone) and cimetidine: a pilot study. Prostate, 17 (1990) 95. S. Jamal, J.R. Casley-Smith and Judith R. Casley-Smith, The effects of 5,6-benzo[a]pyrone (coumarin) and DEC on filaritic lymphoedema and elephantiasis in India. Preliminary results. Ann. Trop. Med. Parasitol., 83 (1989) 287. D. Cox, R. O'Kennedy and R.D. Thornes, The rarity of liver toxicity in patients treated with coumarin (1,2-benzopyrone). Hum. Toxicol., 8 (1989) 501. V.S. Gallicchio, B.C. Hulette, C. Harmon and M.E. Marshall, Toxicity of coumarin (1,2-benzopyrone) on human peripheral blood mononuclear cells and human and murine bone marrow progenitor stem cells. J. Biol. Resp. Mod., 8 (1989) 116. B.G. Lake, Investigations into the mechanism ofcoumarin-induced hepatotoxicity in the rat. Arch. Toxicol. Suppl., 7 (1984) 16. B.G. Lake, T.J.B. Gray, J.G. Evans, D.F.V. Lewis, J.A. Beamand and K.L. Hue, Studies on the mechanisms ofcoumarin-induced toxicity in rat hepatocytes: comparison with dihydrocoumarin and other coumarin metabolites. Toxicol. Appl. Pharmacol., 97 (1989) 31 I. J.G. Evans, I.F. Gaunt and B.G. Lake, Two-year toxicity study on coumarin in the baboon. Food Cosmet. Toxicol., 17 (1979) 187. W. Endell and G. Seidel, Coumarin toxicity in different strains of mice. Agents Actions, 8 (1978) 299. W.H. Shilling, R.F. Crampton and R.C. Longland, Metabolism ofcoumarin in man. Nature, 221 (1969) 664. W.A. Ritschel and T.J. Hardt, Pharmacokinetics of coumarin, 7-hydroxycoumarin and 7-hydroxycoumarin glucuronide in the blood and brain of gerbils following intraperitoneal administration of coumarin. Arzneim.-Forsch., 33(ii) (1983) 1254. K. Dominguez, J.H. Fentem, M.J. Garle and J.R. Fry, Comparison of Mongolian gerbil and rat hepatic microsomal monooxygenase activities: high coumarin 7-hydroxylase activity in the gerbil. Biochem. Pharmacol., 39 (1990) 1629. J.H. Fentem and J.R. Fry, Comparison of the effects of inducers of cytochrome P450 on Mongolian gerbil and rat hepatic microsomal monooxygenase activities. Xenobiotica. 21 (1991) 895. V. Ullrich and P. Weber, The O-dealkylation of 7-ethoxycoumarin by liver microsomes. HoppeSeyler's Z. Physiol. Chem., 353 (1972) 1171. J. Sedlak and R.H. Lindsay, Estimation of total, protein-bound and non-protein sulfhydryl groups in tissue with Ellman's reagent. Anal. Biochem., 25 (1968) 192. M.J. Garle and J.R. Fry, Detection of reactive metabolites in vitro. Toxicology, 54 (1989) 101. O.H. Lowry, N.J. Rosebrough, A.L. Farr and R.J. Randall, Protein measurement with the Folin phenol reagent. J. Biol. Chem., 193 (1951) 265. T. Omura and R. Sato, The carbon monoxide binding pigment of liver microsomes. I. Evidence for its hemoprotein nature. J. Biol. Chem., 239 (1964) 2370. A. Aitio, A simple and sensitive assay of 7-ethoxycoumarin deethylation. Anal. Biochem., 85 (1978) 488. N.N. Aronson Jr. and O. Touster, Isolation of rat liver plasma membrane fragments in isotonic sucrose, in S. Fleischer and L. Packer (Eds.), Methods in Enzymology, Vol. 31 (Part A), Academic Press, New York, 1974, p.90. G.O. Korsrud, H.C. Grice and J.M. McLaughlan, Sensitivity of several serum enzymes in detecting carbon tetrachloride-induced liver damage in rats. Toxicol. Appl. Pharmacol., 22 (19721 474.
129
Species differences in the hepatotoxicity of coumarin: A comparison of rat and Mongolian gerbil Julia H. F e n t e m a, Jeffrey R. F r y a and N o r m a n W. T h o m a s b Departments of aphysiology and Pharmacology and hHuman Morphology, Medical School, Queen's Medical Centre, Nottingham NG7 2 UH ( U. K.) (Received July 3rd, 1991; accepted October 23rd, 1991)
Summary The acute hepatic effects o f c o u m a r i n (2H-l-benzopyran-2-one) in male Wistar rats and Mongolian gerbils has been compared. A single dose of coumarin (125 mg/kg, intraperitoneally (i.p.)) was hepatotoxic to rats within 24 h as assessed by its effects on a variety of hepatic parameters. Coumarin-induced hepatotoxicity was associated with significant increases in relative liver weight, plasma alanine and aspartate aminotransferase activities and hepatic non-protein sulphydryl groups. Cytochrome P-450 content and 7-ethoxycoumarin O-deethylase and glucose 6-phosphatase activities were significantly lower in coumarin-treated compared with control rats. Centrilobular necrosis was only observed in two out of six rats at this dose, but was present in all four coumarin-treated rats when the dose was increased to 150 mg/kg. In contrast to the effects observed in the rat, no evidence was found for coumarin-induced hepatotoxicity in gerbils following a single i.p. dose of 125 mg/kg. These data indicate that the gerbil is less sensitive to the hepatotoxic effects of coumarin than the rat.
Key words." Coumarin; Gerbil; Hepatotoxicity; Rat
lntoduction Coumarin (2H-l-benzopyran-2-one) is a naturally occurring plant product which was first isolated from the Tonka bean by Vogel in 1820 [1]. It was banned from use as a food flavouring agent in the mid-1950s when it was shown to cause liver necrosis in rats [2], but is still used in perfumes, toilet soaps, toothpastes and some tobacco products [3]. Coumarin has several potential clinical applications [4] and has been used for the treatment of various cancers [5-7], high protein oedema [8], brucellosis and certain other chronic infections [9]. Liver toxicity in patients receiving relatively high daily doses of coumarin is very rare [9,10]. Species differences in coumarin hepatotoxicity are thought to be metabolism-
Correspondence to: Jeffrey R. Fry. 0300-483X/92/$05.00 © 1992 Elsevier Scientific Publishers Ireland Ltd. Printed and Published in Ireland
130 mediated [3,i1,12]. The rat, in which it is markedly hepatotoxic, primarily metabolizes coumarin via 3-hydroxylation and cleavage of the heterocyclic ring. Coumarin is less toxic in the baboon [131 and DBA/2J mouse [14], species which resemble man [15] in their extensive formation of the 7-hydroxy metabolite. 7-Hydroxylation of coumarin is negligible in the rat, making it a poor experimental model for man with respect to coumarin hepatotoxicity. Studies reported by Ritschel and Hardt [16] indicated that the pharmacokinetic profiles of coumarin, 7-hydroxycoumarin and 7-hydroxycoumarin glucuronide in the blood of Mongolian gerbils were similar to those observed in man. We have previously demonstrated that coumarin 7-hydroxylase activity was over 100-fold greater in gerbil compared with rat liver microsomes [17,18]. On the basis of these findings we decided that a more comprehensive investigation of coumarin metabolism and hepatotoxicity in the gerbil was warranted. In this study we have compared the acute hepatotoxicity of coumarin in male Wistar rats and Mongolian gerbils. Materials and methods
Chemicals Coumarin, l-amino-2-naphthol-4-sulphonic acid (Fiske and Subbarow reducer), glucose 6-phosphate, glucose 6-phosphate dehydrogenase, 7-hydroxy-coumarin and NADP and kits for the determination of alanine aminotransferase (ALT) and aspartate aminotransferase (AST) activities, were purchased from Sigma (Dorset, U.KI). 7-Ethoxycoumarin was synthesized according to the method of Ullrich and Weber [19]. All other chemicals were of analytical grade. Animals and treatment Male Wistar albino rats (8 weeks old; 180-200 g) and male Mongolian gerbils (Meriones unguiculatus; 8 weeks old; 60-65 g) were obtained from the University of Nottingham Joint Animal Breeding Unit (Sutton Bonington, Leics., U.K.). They were housed in small groups at a constant room temperature of 22°C and allowed free access to standard laboratory diet (Diet 41B, Heygates (Pilsbury) York, U.K.) and tap water. Rats and gerbils received single i.p. injections of coumarin (125 or 150 mg/kg); control animals were given corresponding volumes (5 ml/kg) of the arachis oil vehicle. They were killed 24 h later following overnight starvation. Blood was taken, for the determination of plasma ALT and AST activities, by cardiac puncture under haiothane anaesthesia. The livers were removed and weighed; samples were coded and taken for histology. The remaining liver was used for biochemical studies. Histology Thin slices from each coded liver sample were immersed in 10°/,, neutral formalin, Helly's or Bouin Hollande fixatives and, after appropriate treatment, embedded in paraffin wax. Sections (8 t~m) were stained with haematoxylin and eosin and examined by light microscopy. Only when the examination had been completed was the code broken. Biochemical investigations The non-protein sulphydryl groups in whole liver homogenates (0.25 g fresh
131
tissue/ml) were determined by the method of Sedlak and Lindsay [20]. Liver microsomal fractions were prepared by the calcium aggregation technique, as outlined previously [21]. Protein was determined by the method of Lowry et al. [22] using bovine serum albumin as standard. Hepatic microsomal cytochrome P-450 content [23] and 7-ethoxycoumarin O-deethylase [241 and glucose 6-phosphatase [25] activities, were assayed by the methods quoted. Plasma ALT and AST activities were measured using standard diagnostic test kits.
Statistical analysis Values for control and coumarin-treated animals were compared using Student's t-tests or Mann-Whitney U-tests (plasma ALT and AST activities) as appropriate. Results
A single dose of coumarin (125 mg/kg, i.p.) was hepatotoxic to rats within 24 h, as assessed by its effects on a variety of hepatic parameters (Table 1). Coumarininduced hepatotoxicity in the rat was manifested by significant, dose-dependent, in-
TABLE I EFFECT OF C O U M A R I N T R E A T M E N T ON SOME H E P A T I C P A R A M E T E R S IN THE RAT AND GERBIL Results are expressed as mean + S.E.M. or medians (range) for groups of four or six animals 01). ALT, alanine aminotransferase; AST, aspartate aminotransferase; b.w., body wt; ECOD, 7-ethoxycoumarin Odeethylase; NPS, non-protein sulphydryl: n.d., not determined. Parameter
Relative liver weight (g liver/100 g b.w.) ALT (#mol/min per I) AST (#mol/min per 1) Cytochrome P-450 (nmol/mg protein) ECOD (nmol/min per mg protein) Glucose 6-phosphatase (nmol/min per mg protein) NPS groups (#mol/g liver)
Rat
Gerbil
Control (n = 6)
125 mg/kg (n = 6)
150 mg/kg (n = 4)
Control (n = 6)
125 mg/kg (n = 6t
3.6 + 0.1
4.2 + 0.1"*
4.5 + 0.1"**
3.5 4. 0.2
3.7 + 0.1
25 (21-28) 41 (37-81) 1.1 + 0.0
87** (42-4632) 194" (57-7585) 0.7 4. 0.1"**
939** (518-2970) 2184"* (1167-6824) n.d.
65 (55-104) 157 (111-311) 1.1 4. 0.1
60 (37-1 I1) 119 (101-179) 1.0 -4- 0.0
2.9 4. 0.2
1.5 4- 0.3**
n.d.
5.6 4. 0.4
4.7 4- 0.3
n.d.
362 + 20
333 4. 20
415 4. 11
4.3 4. 0.4
292 4- 7***
5.8 ± 0.2**
n.d.
5.8 + 0.4
6.6 + 0.3
Values are significantly different from the corresponding controls (arachis oil-treated) at *P < 0.05, **P < 0.01 and ***P < 0.001 (Student's t-test or Mann -Whitney U-test).
132
Fig. 1. Photomicrographs of rat liver illustrating the effect of different doses of coumarin. (A) illustrates the typical appearance of the control liver with a radial disposition of sinusoids about a central vein (c). (B,C) illustrate the range of response which was observed in different rats following a coumarin dose of 125 mg/kg. In (B) cellular infiltration without hepatic necrosis characterizes the tissue surrounding a central vein (c), while in (C) the small arrows define the junction between a region of centrilobular necrosis and the normal disposition of sinusoids about a portal triad (p). (DI illustrates the appearance of the liver in rats receiving a dose of 150 mg/kg; the small arrows define the boundary of a region of centrilobular necrosis around the axis of a central vein (c), while the parenchyma about the axis of a portal triad (p) retains a normal appearance; original magnification, x 150.
133 creases in relative liver weight and plasma A L T and A S T activities. C y t o c h r o m e P-450 content and 7-ethoxycoumarin O-deethylase and glucose 6-phosphatase activities were all significantly lower in coumarin-treated c o m p a r e d with control rats. Hepatic non-protein sulphydryl groups (mainly glutathione) were significantly increased 24 h after coumarin administration. In contrast to the effects o f coumarin in the rat, none o f the hepatic parameters measured were significantly altered by coumarin treatment of gerbils. Histological examination o f liver sections indicated a range o f responses amongst the group o f six rats which received a coumarin dose o f 125 mg/kg (Fig. 1). N o histopathological changes were detected in three rats, cellular infiltration without hepatic necrosis was present in one rat and centrilobular necrosis was found in two rats. In one o f these the necrosis was extensive and at the time o f tissue collection it was noted that the liver o f this particular rat was grossly enlarged (4.9 g liver/100 g b o d y wt) and had a mottled appearance. All tissue samples from rats which were given a coumarin dose o f 150 mg/kg exhibited centrilobular hepatic necrosis (Fig. 1) although in one rat the necrosis was not as extensive as in the rest o f the group. There was no evidence for coumarin-induced histopathological changes in the livers o f gerbils following a single dose of 125 mg/kg (Fig. 2).
Fig. 2. Photomicrographs of the liver from control (A) and coumarin-treated (B) gerbils. At a dose of 125 mg/kg there was no detectable histological difference between the control and experimental groups; (c) central vein; × 150.
134
Discussion A single i.p. injection o f c o u m a r i n at a dose of 125 mg/kg was hepatotoxic to rats as judged by a variety of blood and liver parameters. Thus, relative liver weight was significantly increased compared with controls, plasma A L T and AST activities were elevated and the hepatic microsomal parameters were all significantly decreased. Relative liver weight and plasma A L T and AST activities have previously been demonstrated to represent sensitive markers of xenobiotic-mediated hepatotoxicity [26]. These data are entirely consistent with those reported by Lake et al. [12]. In the present study a range of histopathological changes was observed within the group of young male Wistar rats administered coumarin at this dose. Three out of the six rats showed no histological evidence of liver damage and in one there was cellular infiltration but no hepatic necrosis. Centrilobular necrosis was observed in the other two rats, with this being very widespread in one animal. The plasma A L T (4632 #mol/min per 1) and AST (7585 #mol/min per 1) activities for this rat were markedly elevated compared to those for the other rats in the same group (ALT: 42-143; AST: 57-375 t~mol/min per l). The reason for the extreme sensitivity of this particular rat to the same dose of coumarin as the other treated animals is unknown, but may be related to it having a lower non-protein sulphydryl content or being compromised in some other way prior to treatment. The range of histopathological responses observed was in close agreement with the results of the biochemical investigations; there was a good correlation between the rankings of individual rats on the basis of histopathoiogy and the hepatic parameters measured. Interestingly, the histopathological observations in this study contrast with those of Lake et al. [12] who reported that, in young male Sprague-Dawley rats, a single i.p. injection of coumarin at 125 mg/kg induced necrosis of the centrilobular hepatocytes together with some vacuolation of cells in the periportal region. On increasing the dose to 150 mg/kg centrilobular hepatic necrosis was found in all rats and this was accompanied by further increases in A L T and AST activities compared with rats given 125 mg coumarin/kg. In one rat the necrosis was less extensive than in the rest of the group; the values for relative liver weight and A L T and AST activities were correspondingly lower than those for the other rats similarly treated. The coumarin dose was not increased above 150 mg/kg to avoid other symptoms of clinical toxicity, including narcotic effects. It is apparent from these data that a dose of 125 mg/kg is on the lower end of the coumarin dose-response relationship for the rat. In contrast to the effects observed in the rat, coumarin was not hepatotoxic to gerbils following a single i.p. dose of 125 mg/kg, so demonstrating a relative insensitivity of the gerbil to coumarin-induced hepatotoxicity. At least two possible explanations can be proposed for t h i s species difference in sensitivity to coumarin hepatotoxicity - one based on coumarin metabolism and one based on the abundance of protective nucleophiles. The two principal routes of hepatic metabolism of coumarin involve oxidation on the aromatic ring (particularly at the 7-position) and oxidation on the pyrone ring (at the 3,4-double bond). Oxidation on the pyrone ring accounts for the bulk of coumarin metabolism in rat liver microsomes and Lake et al. [12] have proposed that
135 coumarin-induced hepatotoxicity in the rat is due to its bioactivation by cytochrome P-450-dependent enzymes to a toxic metabolite(s), postulated to be a 3,4-epoxide intermediate. Aromatic hydroxylation is a major metabolic route in gerbil liver microsomes [17,18] and it is possible that this represents the principal route of coumarin metabolism in gerbils in vivo. Thus, differences between rat and gerbil in their susceptibility to coumarin-induced hepatotoxicity may be related to differences in routes of coumarin metabolism. Further studies are required to test this hypothesis. Lake et al. [12] demonstrated an initial decrease in the non-protein sulphydryl content of liver homogenates measured 2 h after dosing rats with coumarin. This then recovered and by 24 h had increased to a level significantly greater than that found with control rats; this latter aspect was confirmed in the present study (Table I). Glutathione protects against coumarin-induced hepatotoxicity in the rat, possibly by the formation of conjugates with the reactive coumarin metabolite(s). The nonprotein sulphydryl content (principally glutathione) of liver homogenates prepared from control gerbils is considerably higher than that for control rats ([17] and Table I). This may afford the gerbil greater protection than the rat from any toxic intermediates formed during coumarin metabolism. In summary, no evidence was found for coumarin-induced hepatotoxicity in the Mongolian gerbil at a dose of coumarin that was hepatotoxic in the rat. The rarity of liver toxicity in patients receiving relatively high daily doses of coumarin [9,10] and the similarity of gerbil and man in the extent of coumarin 7-hydroxylation, both in vivo [16] and in vitro [17,18], point to the gerbil being a more appropriate species than the rat for studies aimed at assessing the toxicological hazard, if any, of coumarin to man.
Acknowledgements We would like to thank Mr. I. Janson and Ms. S. Ecob for their help with the animal procedures and Ms. P. Barnes for the preparation of liver sections for histological examination. Financial support for this study was provided by the University of Nottingham in the form of a studentship to J.H. Fentem.
References 1 G. Feuer, The metabolism and biological actions ofcoumarins, in G.P. Ellis and G.B. West (Eds.), Progress in Medicinal Chemistry, Vol. 10, Elsevier, Amsterdam, 1974, p, 85. 2 L.W. Hazleton, T.W. Tusing, B.R. Zeitlin, R. Thiessen and H.K. Murer, Toxicity of coumarin. J. Pharmacol. Exp. Ther., 118 (1956) 348. 3 A.J. Cohen, Critical review of the toxicologyof coumarin with special reference to interspecies differences in metabolism and hepatotoxic response and their significance to man. Food Cosmet, Toxicol., 17 (1979) 277. 4 D. Egan, R. O'Kennedy, E. Moran, D. Cox, E. Prosser and R.D. Thornes, The pharmacology,metabolism, analysis and applications of coumarin and coumarin-related compounds. Drug Metab. Rev., 22 (1990) 503. 5 M.E. Marshall, L. Mendelsohn, K. Butler, L. Riley, J. Cantrell, J. Harvey, C. Wiseman, T. Taylor and J,S. Macdonald, Treatment of metastatic renal cell carcinoma with coumarin (l,2-benzopyrone) and cimetidine: a pilot study. J. Clin. Oncol., 5 (1987) 862.
136 6
7 8
9 10
11 12
13 14 15 16
17
18 19 20 21 22 23 24 25
26
M.E. Marshall, K. Butler, J. Cantrell, C. Wiseman and L. Mendelsobn, Treatment of advanced malignant melanoma with coumarin and cimetidine: a pilot study. Cancer Chemother. Pharmacol., 24 (1989) 65. M.E. Marshall, K. Butler and D. Hermansen, Treatment of hormone-refractory stage D carcinoma of prostate with coumarin (I,2-benzopyrone) and cimetidine: a pilot study. Prostate, 17 (1990) 95. S. Jamal, J.R. Casley-Smith and Judith R. Casley-Smith, The effects of 5,6-benzo[a]pyrone (coumarin) and DEC on filaritic lymphoedema and elephantiasis in India. Preliminary results. Ann. Trop. Med. Parasitol., 83 (1989) 287. D. Cox, R. O'Kennedy and R.D. Thornes, The rarity of liver toxicity in patients treated with coumarin (1,2-benzopyrone). Hum. Toxicol., 8 (1989) 501. V.S. Gallicchio, B.C. Hulette, C. Harmon and M.E. Marshall, Toxicity of coumarin (1,2-benzopyrone) on human peripheral blood mononuclear cells and human and murine bone marrow progenitor stem cells. J. Biol. Resp. Mod., 8 (1989) 116. B.G. Lake, Investigations into the mechanism ofcoumarin-induced hepatotoxicity in the rat. Arch. Toxicol. Suppl., 7 (1984) 16. B.G. Lake, T.J.B. Gray, J.G. Evans, D.F.V. Lewis, J.A. Beamand and K.L. Hue, Studies on the mechanisms ofcoumarin-induced toxicity in rat hepatocytes: comparison with dihydrocoumarin and other coumarin metabolites. Toxicol. Appl. Pharmacol., 97 (1989) 31 I. J.G. Evans, I.F. Gaunt and B.G. Lake, Two-year toxicity study on coumarin in the baboon. Food Cosmet. Toxicol., 17 (1979) 187. W. Endell and G. Seidel, Coumarin toxicity in different strains of mice. Agents Actions, 8 (1978) 299. W.H. Shilling, R.F. Crampton and R.C. Longland, Metabolism ofcoumarin in man. Nature, 221 (1969) 664. W.A. Ritschel and T.J. Hardt, Pharmacokinetics of coumarin, 7-hydroxycoumarin and 7-hydroxycoumarin glucuronide in the blood and brain of gerbils following intraperitoneal administration of coumarin. Arzneim.-Forsch., 33(ii) (1983) 1254. K. Dominguez, J.H. Fentem, M.J. Garle and J.R. Fry, Comparison of Mongolian gerbil and rat hepatic microsomal monooxygenase activities: high coumarin 7-hydroxylase activity in the gerbil. Biochem. Pharmacol., 39 (1990) 1629. J.H. Fentem and J.R. Fry, Comparison of the effects of inducers of cytochrome P450 on Mongolian gerbil and rat hepatic microsomal monooxygenase activities. Xenobiotica. 21 (1991) 895. V. Ullrich and P. Weber, The O-dealkylation of 7-ethoxycoumarin by liver microsomes. HoppeSeyler's Z. Physiol. Chem., 353 (1972) 1171. J. Sedlak and R.H. Lindsay, Estimation of total, protein-bound and non-protein sulfhydryl groups in tissue with Ellman's reagent. Anal. Biochem., 25 (1968) 192. M.J. Garle and J.R. Fry, Detection of reactive metabolites in vitro. Toxicology, 54 (1989) 101. O.H. Lowry, N.J. Rosebrough, A.L. Farr and R.J. Randall, Protein measurement with the Folin phenol reagent. J. Biol. Chem., 193 (1951) 265. T. Omura and R. Sato, The carbon monoxide binding pigment of liver microsomes. I. Evidence for its hemoprotein nature. J. Biol. Chem., 239 (1964) 2370. A. Aitio, A simple and sensitive assay of 7-ethoxycoumarin deethylation. Anal. Biochem., 85 (1978) 488. N.N. Aronson Jr. and O. Touster, Isolation of rat liver plasma membrane fragments in isotonic sucrose, in S. Fleischer and L. Packer (Eds.), Methods in Enzymology, Vol. 31 (Part A), Academic Press, New York, 1974, p.90. G.O. Korsrud, H.C. Grice and J.M. McLaughlan, Sensitivity of several serum enzymes in detecting carbon tetrachloride-induced liver damage in rats. Toxicol. Appl. Pharmacol., 22 (19721 474.