- Email: [email protected]
Toxicity of coumarin and various methyl derivatives in cultures of rat hepatocytes and V79 cells
Toxic. in Vitro Vol. 6, No. 1, pp. 21-25, 1992 Printed in Great Britain.All rights reserved
0887-2333/92$5.00+ 0.00 Copyright © 1992PergamonPress plc
TOXICITY OF COUMARIN A N D VARIOUS METHYL DERIVATIVES IN CULTURES OF RAT HEPATOCYTES A N D V79 CELLS J. H. FENTEM,A. H. HAMMOND,M. J. GARLEand J. R. FRY Department of Physiology and Pharmacology, Medical School, Queen's Medical Centre, Nottingham NG7 2UH, UK (Recewed 22 January 1991; revisions received 22 April 1991)
Abstract--The toxicity of coumarin and various simple methyl derivatives in rat hepatocyte and V79 cell cultures was studied to investigate further the mechanism of coumarin hepatotoxicity. Coumarin was six times more toxic in hepatocyte cultures from phenobarbitone (PB)-treated rats than in those from untreated rats. At concentrations below 3 raM,coumarin did not affect the survival of V79 lung fibroblasts. SKF-525A inhibited coumarin-induced toxicity in hepatocytes cultured from PB-treated rats, whereas depletion of hepatocyte glutathione (GSH) levels with buthionine sulphoximine (BSO) significantly increased toxicity. Dihydrocoumarin (DHC) had little effect on the survival of cultured hepatocytes, indicating that the 3,4-double bond is an important determinant of coumarin toxicity. In general, the toxicity of coumarin in hepatocyte cultures was reduced by substitution with one or more methyl groups. 3-Methylcoumarin (MeC), however, was more toxic than coumarin itself in hepatocyte cultures from untreated rats. Except for 3,4-diMeC, the methyl derivativeswere markedly more toxic in rat hepatocytes than in V79 cell cultures. The data obtained for coumarin and 4-MeC, and possibly 6-MeC and 7-MeC, are consistent with bepatocyte toxicity being due to the cytochrome P-450-dependent formation of one or more toxic metabolites that may be detoxified by reacting with GSH. This was less apparent for 3-MeC and 3,4-diMeC, although depletion of GSH levels significantlyincreased the hepatocyte toxicity of both compounds.
INTRODUCTION
Some controversy still exists as to whether coumatin per se, or a metabolite of it, is responsible for hepatotoxicity in the rat (Cohen, 1979; Gibbs et al., 1971; Lake, 1984). Recently, Lake et al. (1989) have shown that the toxicity of coumarin in rat hepatocyte cultures is dependent on its bioactivation by cytochrome P-450, and that glutathione (GSH) appears to protect against this toxicity. A comparison of the in vivo toxicities of coumarin and dihydrocoumarin (DHC; 1,2-benzohydropyrone) in the rat showed that DHC, which lacks the 3,4-double bond, was not hepatotoxic (Lake et al., 1989). These authors postulated that a coumarin 3,4-epoxide intermediate might be responsible for coumarin-induced hepatotoxicity in the rat. To investigate further the mechanism of coumarin toxicity we have studied the effects of various methyl derivatives, in particular 3- and 4-methylcoumarins (MeC) and 3,4-dimethylcoumarin (diMeC), on the survival of rat hepatocytes in primary culture. The toxicity of these coumarins to the V79 Chinese hamster lung fibroblast cell line, which lacks cytochrome P-450, has also been examined.
The coumarins are a group of phenylpropanoid lactones that constitute some of the most important natural products, and exert a wide range of physiological effects (Feuer, 1974; Murray et al., 1982; Soine, 1964). Coumarin (l,2-bcnzopyrone) is added to perfumes, soaps, detergents, toothpastes, and some tobacco products and alcoholic beverages (Opdyke, 1974). It is used clinically in the treatment of highprotein oedema (Jamal et al., 1989), chronic brucellosis and immune suppression (Cox et al., 1989; Moran et al., 1987). In combination with cimetidine, coumarin is currently undergoing clinical trials for the treatment of several malignancies in humans (Marshall et al., 1987 and 1989). Coumarin was banned from use as a food flavouring agent when it was shown to be hepatotoxic in the rat and dog (Hazleton et al., 1956); however, no significant liver damage was observed in studies using the baboon (Evans et al., 1979) and Syrian hamster (Ueno and Hirono, 1981). Species differences in the hepatotoxicity of coumarin are thought to be related to its metabolism (Cohen, 1979). In humans, coumatin is metabolized primarily by 7-hydroxylation (Shilling et al., 1969), whereas this pathway is only minor in the rat, which metabolizes coumarin mainly through 3-hydroxylation (Feuer, 1970b).
MATERIALS AND METHODS
Materials. DL- Buthionine-(S,R)- sulphoximine (BSO), coumarin, glutathione (GSH), MTT and nicotinamide were purchased from Sigma Chemical Abbreviations: BSO = DL-buthionine-(S,R)-sulphoximine; Co. Ltd (Poole, Dorset, UK). Dihydrocoumarin and 7-methylcoumarin were supplied by Aldrich DHC = dihydrocoumarin; GSH = glutathione; Met2 = methylcoumarin; NPS = non- protein sulphhydryl; Chemical Co. Ltd (Poole, Dorset, UK), and phenobarbitone (sodium salt) by BDH Chemicals PB = phenobarbitone; WEC = Williams' medium E.
21
22
J.H. FENTEMet al.
(Dorset, UK). Williams' medium E, Eagle's minimum essential medium, foetal calf serum, glutamine and Nunclon 24-well multidishes (for V79 cultures) were obtained from Gibco Ltd (Paisley, UK); insulin (Actrapid MC) from Novo Industri A/S (Denmark); dexamethasone sodium phosphate (Decadron) from Merck, Sharp and Dohme Ltd (Hoddesdon, UK); and gentamicin from Roussel Laboratories Ltd (Uxbridge, UK). Falcon 24-well Primaria tissue culture plates (for hepatocyte cultures) were purchased from Fahrenheit Lab. Supplies (Nottingham, UK). 6-Methylcoumarin and Proadifen HCl--also known as SKF-525A or 2-diethylaminoethyl-2,2-diphenylvalerate---were the generous gifts of Avon Cosmetics Ltd (USA) and Smith Kline and French Labs. Ltd (Herts., UK), respectively. All other chemicals were of A R grade. 3-Methylcoumarin (m.p. 88-89°C) was synthesized according to the method of Phadke et al. (1984), and recrystallized from ethanol. 4-Methylcoumarin (m.p. 79-80°C) and 3,4-dimethylcoumarin (m.p. 114-115°C) were synthesized by the method reported by Peters and Simonis (1908). They were recrystallized from n-hexane containing a small amount of ethylacetate. Animals andpretreatment. The source and maintenance of the male adult Wistar rats (120-150 g) used in this study have been described previously (Fry, 1981). Phenobarbitone (PB) was given as a 0.1% (w/v) solution of the sodium salt in drinking-water for 7 days. Control animals were untreated. Hepatocyte isolation and culture. Hepatocytes were isolated by the lobe perfusion technique, essentially as described by Reese and Byard (1981). Viability was 93-99%, as determined by trypan blue dye exclusion. Yield and viability were similar for hepatocytes isolated from both control and PB-treated rats. Cells were plated out in WEC medium (Williams' medium E containing 5mi-glutamine, 50/~g/ml gentamicin, 10% (v/v) foetal calf serum, 10mU insulin/ml, 5 mM-nicotinamide and 1 pM-dexamethasone) at a density of 0.25 x 106 cells/well in 24-well Primaria plates. Cultures were maintained at 37°C in an atmosphere of 5% CO2/95% air. The medium was changed after allowing for 2-4 hr cell attachment. Hepatocytes were cultured for a further 22 hr prior to the addition of fresh medium containing the test compounds (i.e. coumarin; DHC; 3-, 4-, 6and 7-MeCs and 3,4-diMeC) at the desired concentrations (0.01-3.0 mM). Coumarins were initially dissolved in dimethylformamide, and then diluted in WEC medium such that the final solvent concentration did not exceed 1%. Cultures were incubated with the compounds for 24hr and cell survival, relative to untreated cultures, was determined by a modification of the MTT reduction method of Denizot and Lang (1986). In outline, medium containing MTT (0.1 mg/ml) was added to the cell cultures. After incubating for 2-3 hr at 37°C, the MTT solution was removed and isopropanol added to each well to solubitize the resulting blue formazan product. The absorbance of the isopropanol solution was determined spectrophotometrically, using an automatic plate reader (Anthos 2001) with a test wavelength of 570 nm and a reference wavelength of 620 nm.
In some experiments, SKF-525A (10pM), an inhibitor of cytochrome P-450 activity (Netter, 1980), or BSO (5 mM), in WEC medium, was added to the cultures after 2 hr. Hepatocytes were incubated with SKF-525A or BSO for 22 hr prior to addition of the test compounds and determination of cell survival. BSO, an irreversible inhibitor of y-glutamyl cysteine synthetase, depletes cellular GSH levels. It has been reported that this compound has no effect on cytochrome P-450 levels or on P-450-dependent and conjugating enzyme activities (Drew and Miners, 1984). Culture of V79 Chinese hamster lung fibroblasts. V79 cells were cultured in Eagle's minimum essential medium as described by Horner and Shah (1984). Cells were plated out at a density of 5 x 103 cells/well in 24-well multidishes and cultured for 3-4 days until a confluent monolayer had formed. They were then exposed to the coumarins (0.5-3.0 mM) for 24 hr and cell survival, relative to untreated cultures, was determined by the MTT assay. In some experiments, cells approaching confluency were incubated with BSO (100/tM) for 24hr prior to exposure to the test compounds. Presentation of results and statistical analysis. For each 24-well plate the percentage of cell survival in treated, relative to untreated, wells was calculated for each concentration of compound from the average absorbance values of 3 or 4 wells per concentration. Dose-response curves were constructed and TCs0 values (i.e. the concentrations producing a 50% reduction in cell survival) were determined from the graphs. Each TCs0 value is the mean + SEM of 3-15 plates of hepatocytes cultured from different animals. Statistical analysis was undertaken using unpaired t-tests or ANOVA/Dunnett's test as appropriate. A value of P < 0.05 was considered to be significant. RESULTS There were no differences in the survival in cultures of hepatocytes isolated from PB-treated rats compared with controls (i.e. untreated rats). The TCs0 values determined for each compound in hepatocyte cultures are shown in Table 1. The toxicity of coumarin and its methyl derivatives was concentration dependent. In hepatocyte cultures from control rats 3-MeC was 2.7-fold more toxic than coumarin itself, whereas the other methylcoumarins were less toxic than the parent compound. PB treatment of rats increased the toxicity of coumarin (6-fold), 3-MeC (l.5-fold), 4-MeC (3.1-fold), 6-MeC (4.3-fold) and 7-MeC (11.5-fold). TCs0 values for 3,4-diMeC were similar in hepatocytes cultured from control and PB-treated animals. DHC, at concentrations ~<3 mM, did not induce a sufficient loss of viability of hepatocytes from either untreated or PB-treated rats to allow the determination of TCs0 values. The incubation of hepatocytes isolated from PBtreated rats with various concentrations of BSO (~< 10 mM) for 22 hr, prior to the addition of the test compounds, resulted in a dose-dependent depletion of cellular non-protein sulphhydryl (NPS) levels (mainly GSH). This depletion mainly affected cellular GSH, as determined by the method of Saville (1958). BSO at 5 m i depleted the NPS content to 20-25%
In vitro toxicity of coumarins
23
Table 1. Toxicity of coumarins in rat hepatoeyte cultures TC~ (gM) PB
Compound Coumarin Dihydrocoumarin 3-Methylcoumarin 4-Methylcoumarin 6-Methylcoumarin 7-Methylcoumarin 3,4-Dimethylcoumarin
Control Medium BSO SKF-525A 615 + 69 (8)*** 102 __.15 ( 15)t 23 __.4 (4)* 370 + 86 (4)** > 3000 (4) > 3000 (4) ND ND 226 __+43 (7) 150 -I- 20 (15)t 5 + 0 (4)* 292 +_127 (4)* 1305 __.316 (6)** 420 + 76 (13)~" 61 5:18 (4)* 1008 + 122 (4)** 867 + 122 (3)*** 201 + 17 (7) ND ND 1920 +_873 (3)* 167 + 40 (6) ND ND 953 _ 236 (4) 1354 + 246 (1 l)t 70 5- 35 (4)** 1257 __.225 (3) ND = not determined Hepatocytes were cultured for 22 hr in either WEC medium, or WEC medium containing BSO (5 raM)or SKF-525A (10/~M). They were incubated for a further 24 hr with the eoumarins (0.01-3.0 mM), and survival, relative to untreated cultures, was determined by the MTT reduction method. TCs0 values are means + SEM (n), where n is the number of rats from which hepatocytes were isolated. Asterisks indicate significant difference from the corresponding PB-treated (Medium) values [*P < 0.05; **P < 0.01; ***P < 0.001; unpaired t-test (control v. PB) or ANOVA/Dunnett's test (effects of BSO and SKF-525A)]. tPooled values from 3 separate experiments. t h a t of u n t r e a t e d cultures, b u t did n o t affect cell survival. The N P S level remained below 2 5 % t h a t of u n t r e a t e d cultures t h r o u g h o u t the 48-hr incubation period. Pre-treatment o f hepatocyte cultures with 5 mM-BSO significantly increased the toxicity of coum a t i n (4.4-fold), 3-MeC (30-fold), 4 - M e C (6.9-fold) a n d 3,4-diMeC (19.3-fold) in hepatocyte cultures f r o m PB-treated rats (Table 1). I n c u b a t i o n of hepatocytes from PB-treated rats with S K F - 5 2 5 A (10 #M) for 22 hr, prior to the addition of the test c o m p o u n d s , significantly decreased the toxicity of c o u m a r i n (3.6-fold), 3-MeC (1.9-fold) a n d 4 - M e C (2.4-fold). The toxicity of 3,4-diMeC was n o t affected by S K F - 5 2 5 A (Table 1). S K F - 5 2 5 A was used at the highest c o n c e n t r a t i o n that did not affect cell survival in hepatocyte cultures from PB-treated rats, as m e a s u r e d by the M T T reduction method. A p a r t from 3,4-diMeC (TCs0 = approx. 1 mM), the c o u m a r i n s were all less toxic in V79 cells t h a n in hepatocyte cultures (Table 2). 3-MeC was markedly more toxic (10-fold difference in TCs0 values) in rat hepatocyte cultures (approx. 0.2mM) t h a n in V79 cells (approx. 2 mM). C o u m a r i n , D H C a n d 7-MeC h a d little effect o n V79 cell survival, with TCs0 values all greater t h a n 3 mM. The incubation of V79 cells a p p r o a c h i n g confluency with BSO (100/~M) for 24 hr depleted the N P S c o n t e n t to 10-20% that of u n t r e a t e d cultures. The same result has been previously reported by Hodgkiss a n d M i d d l e t o n (1985). The c o n c e n t r a t i o n of BSO used did n o t affect cell survival as determined by the M T T reduction assay. The N P S c o n t e n t remained below 20% t h a t of Table 2. Toxicity of coumarins in V79 cell cultures TCso (mM) Compound Medium BSO Coumarin > 3 (14) > 3 (8) Dihydrocoumarin > 3 (10) > 3 (4) 3-Methylcoumarin 1.9 _ 0.2 (13) 2.4 +_0.1 (7) 4-Methylcoumarin 2.3 + 0.1 (14) 2.1 + 0.1 (8) 6-Methylcoumarin 1.4 + 0.1 (10) 1.2 + 0.3 (4) 7-Methylcoumarin > 3 (14) > 3 (8) 3,4-Dimethylcoumarin 1.1 _ 0.1 (13) 0.9 __.0.1 (8) Confluent V79 cell cultures, with or without prior incubation with BSO (100/~M)for 24 hr, were exposed to coumarins (0.5-3.0 raM) for 24 hr. Cell survival, relative to untreated cultures, was determined by the MTT reduction method. TCs0 values are means +_SEM (n), where n is the number of independent cell cultures.
untreated cultures t h r o u g h o u t the subsequent 24-hr incubation with the test compounds. BSO t r e a t m e n t h a d no significant effect o n the toxicity of any of the c o u m a t i n s in V79 cell cultures (Table 2). DISCUSSION Primary cultures of rat hepatocytes have been widely used for the study of metabolism-mediated cytotoxicity. In this study hepatocytes were cultured for 24 hr prior to exposure to the test c o m p o u n d s ; this allowed for acclimatization of the cells to the culture conditions. The suitability of this particular in vitro system for investigating metabolism-mediated cytotoxcity has been d e m o n s t r a t e d in a n earlier study ( H a m m o n d a n d Fry, 1991). The effectiveness of the procedure used to induce animals in vivo with PB, as well as the stability of a range of cytochrome P - 4 5 0 - d e p e n d e n t enzyme activities in hepatocytes from untreated a n d PB-treated rats, over a 24-hr culture period, have also been reported ( H a m m o n d a n d Fry, 1990). In agreement with the results of Lake et al. (1989), the d a t a show t h a t c o u m a r i n was moderately toxic in hepatocytes cultured from untreated rats. In contrast, D H C - - w h i c h lacks the 3,4-double b o n d and, unlike coumarin, is n o t hepatotoxic in vivo ( H a g a n et al., 1967; Lake et al., 1989)---had little effect o n the survival of hepatocytes in culture. In vivo induction with PB of cytochrome P-450 a n d other related enzyme activities, prior to cell isolation, markedly increased the toxicity o f c o u m a r i n in hepatocytes. This, together with the absence of c o u m a r i n toxicity in V79 cells, which do n o t possess c y t o c h r o m e P-450, a n d the inhibition o f c o u m a r i n - i n d u c e d toxicity in hepatocytes from PB-treated rats by S K F - 5 2 5 A confirm t h a t the toxicity of c o u m a r i n is due to its metabolism by cytochrome P-450 isozymes. The significant increase in c o u m a r i n toxicity subsequent to the depletion of hepatocyte G S H levels by BSO indicates t h a t G S H is i m p o r t a n t in protecting against c o u m a r i n - i n d u c e d cytotoxicity. The depletion of G S H levels in V79 cell cultures did not affect c o u m a r i n toxicity. This is consistent with the suggestion t h a t the toxicity results from a n initial bioactivation of the c o m p o u n d by cytochrome P-450d e p e n d e n t enzymes; this leads to the f o r m a t i o n o f a toxic metabolite(s), which m a y then be detoxified
24
J.H. FEr~TEMet al.
by reacting with GSH (Lake et al., 1989). The marked differences between the hepatocyte toxicity of coumarin and DHC indicates that the 3,4-double bond is an important determinant of coumarin toxicity, and it has been suggested that coumarin 3,4-epoxide may be responsible for the observed toxicity (Lake et al., 1989). The cytotoxicity of several simple methyl derivatives of coumarin, in particular 3-MeC, 4-MeC and 3,4-diMeC, which are expected to affect the formation of a putative 3,4-epoxide intermediate, was therefore investigated. In general, the toxicity of coumarin in hepatocyte cultures was reduced by substitution with one or more methyl groups. However, 3-MeC was more toxic than the parent compound in hepatocytes isolated from untreated rats. This could be due to differences in lipid solubility (the partition coefficients for coumarin and 3-MeC are 0.75 and 1.24, respectively; Feuer, 1970a) and/or metabolism. The TCs0 value for 4-MeC in hepatocytes from untreated rats was approximately two-fold that for coumarin. These data are in agreement with in vivo results showing that administration of 4-MeC to rats produces slight hepatomegaly and marked induction of hepatic microsomal drug-metabolizing activity (Feuer, 1970a,b), but no liver damage (Gibbs et al., 1971), whereas coumarin itself causes hepatic necrosis (Cohen, 1979). 4-, 6- and 7-Methylcoumarins were more toxic in hepatocytes from PB-treated than from untreated rats, but no significant differences were observed with 3-MeC and 3,4-diMeC. These latter two compounds may be poor substrates for the PB-inducible cytochrome P-450 isozymes. There was evidence for some basal cytotoxicity caused by 3-, 4- and 6-methylcoumarins, and 3,4diMeC, in V79 cell cultures. However, apart from 3,4-diMeC, all the methyl derivatives were more toxic in hepatocyte cultures than in V79 lung fibroblasts, suggesting that cytochrome P-450-dependent metabolism was involved in the toxicity of these compounds. Pretreatment of the cultures with BSO led to the depletion of GSH levels causing a significant increase in the toxicity of 3-MeC, 4-MeC and 3,4diMeC in hepatocytes but not in V79 cell cultures. This indicates a protective role for GSH with respect to the toxicity of a metabolite(s) rather than the unchanged compound itself. SKF-525A, an inhibitor of several cytochrome P-450-mediated activities (Netter, 1980), decreased the toxicity of 3-MeC and 4-MeC in hepatocyte cultures from PB-treated rats, but did not significantly affect the toxicity of 3,4diMeC. The data obtained suggest that hepatocyte toxicity caused by coumarin and 4-MeC, and possibly 6-MeC and 7-MeC is due to the cytochrome P-450-dependent formation of one or more toxic metabolites that may be detoxified by reacting with GSH. The same suggestion, however, cannot be applied to 3-MeC and 3,4-diMeC. In fact, a methyl substituent at the 3-position may actually enhance, rather than inhibit, the formation of a 3,4-epoxide. Moreover, a toxic free radical species of the two compounds may be produced resulting in toxicity by way of an oxidative stress mechanism. Thus, substitution of coumarin with methyl groups at both the 3- and 4- positions protects against metabolism-mediated cytotoxicity to
a certain extent. The results suggest that 3,4-diMeC is a relatively poor substrate for metabolism by cytochrome P-450 isozymes. However, small amounts of a reactive intermediate(s) may be produced that can cause hepatocyte toxicity when the protective effect of GSH is removed. Further studies on the mechanisms responsible for the toxicity of the methylcoumarins, and their relevance to the hepatotoxicity of coumarin itself, are needed in order to assess their toxicological implications to humans. Acknowledgement--This work was supported by the industrial sponsors of the FRAME Research Programme. REFERENCES
Cohen A. J. (1979) Critical review of the toxicology of coumarin with special reference to interspeciesdifferences in metabolism and hepatotoxic response and their significance to man. Food and Cosmetics Toxicology 17, 277-289. Cox D., O'Kennedy R. and Thornes R. D. (1989) The rarity of liver toxicity in patients treated with coumarin (1,2benzopyrone). Human Toxicology 8, 501-506. Denizot F. and Lang R. (1986) Rapid colourimetric assay for cell growth and survival. Modifications to the tetrazolium dye procedure giving improved sensitivity and reliability.Journal of lmmunological Methods 89, 271-277. Drew R. and Miners J. O. (1984) The effects of buthionine sulphoximine (BSO) on glutathione depletion and xenobiotic biotransformation. Biochemical Pharmacology 33, 2989-2994. Evans J. G., Gaunt I. F. and Lake B. G. (1979) Two-year toxicity study on coumarin in the baboon. Food and Cosmetics Toxicology 17, 187-193. Feuer G. (1970a) Induction of drug-metabolizing enzymes of rat liver by derivatives of coumarin. Canadian Journal Physiology and Pharmacology 48, 232-240. Feuer G. (1970b) 3-Hydroxylation of coumarin Or 4-methylcoumarin by rat liver microsomes and its induction by 4-methylcoumaringiven orally. Chemico-Biological Interactions 2, 203-216. Feuer G. (1974) The metabolism and biological actions of coumarins. Progress in Medicinal Chemistry 10, 85-158. Fry J. R. (1981) A comparison of biphenyl 4-hydroxylation and 4-methoxybiphenyl O-demethylation in rat liver microsomes. Biochemical Pharmacology 30, 1915-1919. Gibbs P. A., Janakidevi K. and Feuer G. (1971) Metabolism of coumarin and 4-methylcoumarin by rat liver microsomes. Canadian Journal of Biochemistry 49, 177-184. Hagan E. C., Hansen W. H., Fitzhugh O. G., Jenner P. M., Jones W. I., Taylor J. M., Long E. L., Nelson A. A. and Brouwer J. B. (1967) Food flavourings and compounds of related structure. II. Subacute and chronic toxicity. Food and Cosmetics Toxicology 5, 141-157. Hammond A. H. and Fry J. R. (1990) The in vivo induction of rat hepatic cytochrome P450-dependentenzyme activities and their maintenance in culture. Biochemical Pharmacology 40, 637-642. Hammond A. H. and Fry J. R. (1991) The use of hepatocytes cultured from inducer-treated rats in the detection of cytochrome P-450-mediated cytotoxicity. Toxicology in Vitro 5, 133-137. Hazleton L. W., Tusing T. W., Zeitlin B. R., Theissen R. and Murer H. K. (1956) Toxicity of coumarin. Journal of Pharmacology and Experimental Therapeutics 118, 348-358. Hodgkiss R. J. and Middleton R. W. (1985) Effects of glutathionedepletionusing buthionine sulphoximineon the cytotoxicity of nitroaromatic compounds in mammalian cells in vitro. Biochemical Pharmacology 34, 2175-2178.
In vitro toxicity of coumarins Homer S. A. and Shah M. K. (1984) Effects of fungizone on the in vitro cytotoxicity of other chemicals ATLA 11, 147-i53. Jamal S., Casley-smith J. R. and Casley-Smith Judith R. (1989) The effects of 5,6-benzo[a]pyrone (coumarin) and DEC on filaritic lymphoedema and elephantiasis in India. Preliminary results. Annals of Tropical Medicine and Parasitology 83, 287-290. Lake B. G. (1984) Investigations into the mechanism of coumarin-induced hepatotoxicity in the rat. Archives of Toxicology 7, (Suppl.), 16-29. Lake B. G., Gray T. J. B., Evans J. G., Lewis D. F. V., Beamand J. A. and Hue K. L. (1989) Studies on the mechanism of coumarin-induced toxicity in rat hepatocytes: comparison with dihydrocoumarin and other coumatin metabolites. Toxicology and Applied Pharmacology 97, 311-323. Marshall M. E., Mendelsohn L., Butler K., Riley L. K., Cantrell J., Harvey J., Wiseman C., Taylor T. and Macdonald J. (1987) Treatment of metastatic renal cell carcinoma with coumarin (1,2-benzopyrone) and cimetidine: a pilot study. Journal of Clinical Oncology 5, 862-866. Marshall M. E., Riley L. K., Rhodes J., Eichom T., Jennings C. D., Cibull M. and Thompson J. (1989) Effects of coumatin (1,2-benzopyrone) and cimetidine on peripheral blood lymphocytes, NK cells and monocytes in patients with advanced malignancies. Journal of Biological Response Modifiers 8, 62-69. Moran E., O'Kennedy R. and Thornes R. D. (1987)
25
Analysis of coumarin and its urinary metabolites by HPLC. Journal of Chromatography 416, 165-169. Murray R. D. H., Mendez J. and Brown S. A. (1982)
The Natural Coumarins: Occurrence, Chemistry and Biochemistry. John Wiley and Sons Ltd, London. Netter K. J. (1980) Inhibition of oxidative drug metabolism in microsomes. Pharmacology and Therapeutics 10, 515-535. Opdyke D. L. J. (1974) Monographs on fragrance raw materials: coumarin. Food and Cosmetics Toxicology 12, 385-388. Peters F. and Simonis H. (1908) ~Iber das 2-methylcumarin. Berichte 41, 830-837. Phadke C. P., Kelkar S. L. and Wadia M. S. (1984) A one step general synthesis of 3-substituted coumarins. Synthetic Communications I4, 407-411. Reese J. A. and Byard J. L. (1981) Isolation and culture of adult hepatocytes from liver biopsies. In Vitro 17, 935-940. Saville B. (1958) A scheme for the colourimetricdetermination of microgram amounts of thiols. Analyst 83, 67(b672. Shilling W. H., Crampton R. F. and Longland R. C. (1969) Metabolism of coumarin in man. Nature, London 221, 664-665. Soine T. O. (1964) Naturally-occurting coumatins and related physiological activities. Journal of Pharmacological Sciences 53, 231-264. Ueno I. and Hirono I. (1981) Non-carcinogenic response to coumarin in Syrian golden hamsters. Food and Cosmetics Toxicology 19, 353-355.
0887-2333/92$5.00+ 0.00 Copyright © 1992PergamonPress plc
TOXICITY OF COUMARIN A N D VARIOUS METHYL DERIVATIVES IN CULTURES OF RAT HEPATOCYTES A N D V79 CELLS J. H. FENTEM,A. H. HAMMOND,M. J. GARLEand J. R. FRY Department of Physiology and Pharmacology, Medical School, Queen's Medical Centre, Nottingham NG7 2UH, UK (Recewed 22 January 1991; revisions received 22 April 1991)
Abstract--The toxicity of coumarin and various simple methyl derivatives in rat hepatocyte and V79 cell cultures was studied to investigate further the mechanism of coumarin hepatotoxicity. Coumarin was six times more toxic in hepatocyte cultures from phenobarbitone (PB)-treated rats than in those from untreated rats. At concentrations below 3 raM,coumarin did not affect the survival of V79 lung fibroblasts. SKF-525A inhibited coumarin-induced toxicity in hepatocytes cultured from PB-treated rats, whereas depletion of hepatocyte glutathione (GSH) levels with buthionine sulphoximine (BSO) significantly increased toxicity. Dihydrocoumarin (DHC) had little effect on the survival of cultured hepatocytes, indicating that the 3,4-double bond is an important determinant of coumarin toxicity. In general, the toxicity of coumarin in hepatocyte cultures was reduced by substitution with one or more methyl groups. 3-Methylcoumarin (MeC), however, was more toxic than coumarin itself in hepatocyte cultures from untreated rats. Except for 3,4-diMeC, the methyl derivativeswere markedly more toxic in rat hepatocytes than in V79 cell cultures. The data obtained for coumarin and 4-MeC, and possibly 6-MeC and 7-MeC, are consistent with bepatocyte toxicity being due to the cytochrome P-450-dependent formation of one or more toxic metabolites that may be detoxified by reacting with GSH. This was less apparent for 3-MeC and 3,4-diMeC, although depletion of GSH levels significantlyincreased the hepatocyte toxicity of both compounds.
INTRODUCTION
Some controversy still exists as to whether coumatin per se, or a metabolite of it, is responsible for hepatotoxicity in the rat (Cohen, 1979; Gibbs et al., 1971; Lake, 1984). Recently, Lake et al. (1989) have shown that the toxicity of coumarin in rat hepatocyte cultures is dependent on its bioactivation by cytochrome P-450, and that glutathione (GSH) appears to protect against this toxicity. A comparison of the in vivo toxicities of coumarin and dihydrocoumarin (DHC; 1,2-benzohydropyrone) in the rat showed that DHC, which lacks the 3,4-double bond, was not hepatotoxic (Lake et al., 1989). These authors postulated that a coumarin 3,4-epoxide intermediate might be responsible for coumarin-induced hepatotoxicity in the rat. To investigate further the mechanism of coumarin toxicity we have studied the effects of various methyl derivatives, in particular 3- and 4-methylcoumarins (MeC) and 3,4-dimethylcoumarin (diMeC), on the survival of rat hepatocytes in primary culture. The toxicity of these coumarins to the V79 Chinese hamster lung fibroblast cell line, which lacks cytochrome P-450, has also been examined.
The coumarins are a group of phenylpropanoid lactones that constitute some of the most important natural products, and exert a wide range of physiological effects (Feuer, 1974; Murray et al., 1982; Soine, 1964). Coumarin (l,2-bcnzopyrone) is added to perfumes, soaps, detergents, toothpastes, and some tobacco products and alcoholic beverages (Opdyke, 1974). It is used clinically in the treatment of highprotein oedema (Jamal et al., 1989), chronic brucellosis and immune suppression (Cox et al., 1989; Moran et al., 1987). In combination with cimetidine, coumarin is currently undergoing clinical trials for the treatment of several malignancies in humans (Marshall et al., 1987 and 1989). Coumarin was banned from use as a food flavouring agent when it was shown to be hepatotoxic in the rat and dog (Hazleton et al., 1956); however, no significant liver damage was observed in studies using the baboon (Evans et al., 1979) and Syrian hamster (Ueno and Hirono, 1981). Species differences in the hepatotoxicity of coumarin are thought to be related to its metabolism (Cohen, 1979). In humans, coumatin is metabolized primarily by 7-hydroxylation (Shilling et al., 1969), whereas this pathway is only minor in the rat, which metabolizes coumarin mainly through 3-hydroxylation (Feuer, 1970b).
MATERIALS AND METHODS
Materials. DL- Buthionine-(S,R)- sulphoximine (BSO), coumarin, glutathione (GSH), MTT and nicotinamide were purchased from Sigma Chemical Abbreviations: BSO = DL-buthionine-(S,R)-sulphoximine; Co. Ltd (Poole, Dorset, UK). Dihydrocoumarin and 7-methylcoumarin were supplied by Aldrich DHC = dihydrocoumarin; GSH = glutathione; Met2 = methylcoumarin; NPS = non- protein sulphhydryl; Chemical Co. Ltd (Poole, Dorset, UK), and phenobarbitone (sodium salt) by BDH Chemicals PB = phenobarbitone; WEC = Williams' medium E.
21
22
J.H. FENTEMet al.
(Dorset, UK). Williams' medium E, Eagle's minimum essential medium, foetal calf serum, glutamine and Nunclon 24-well multidishes (for V79 cultures) were obtained from Gibco Ltd (Paisley, UK); insulin (Actrapid MC) from Novo Industri A/S (Denmark); dexamethasone sodium phosphate (Decadron) from Merck, Sharp and Dohme Ltd (Hoddesdon, UK); and gentamicin from Roussel Laboratories Ltd (Uxbridge, UK). Falcon 24-well Primaria tissue culture plates (for hepatocyte cultures) were purchased from Fahrenheit Lab. Supplies (Nottingham, UK). 6-Methylcoumarin and Proadifen HCl--also known as SKF-525A or 2-diethylaminoethyl-2,2-diphenylvalerate---were the generous gifts of Avon Cosmetics Ltd (USA) and Smith Kline and French Labs. Ltd (Herts., UK), respectively. All other chemicals were of A R grade. 3-Methylcoumarin (m.p. 88-89°C) was synthesized according to the method of Phadke et al. (1984), and recrystallized from ethanol. 4-Methylcoumarin (m.p. 79-80°C) and 3,4-dimethylcoumarin (m.p. 114-115°C) were synthesized by the method reported by Peters and Simonis (1908). They were recrystallized from n-hexane containing a small amount of ethylacetate. Animals andpretreatment. The source and maintenance of the male adult Wistar rats (120-150 g) used in this study have been described previously (Fry, 1981). Phenobarbitone (PB) was given as a 0.1% (w/v) solution of the sodium salt in drinking-water for 7 days. Control animals were untreated. Hepatocyte isolation and culture. Hepatocytes were isolated by the lobe perfusion technique, essentially as described by Reese and Byard (1981). Viability was 93-99%, as determined by trypan blue dye exclusion. Yield and viability were similar for hepatocytes isolated from both control and PB-treated rats. Cells were plated out in WEC medium (Williams' medium E containing 5mi-glutamine, 50/~g/ml gentamicin, 10% (v/v) foetal calf serum, 10mU insulin/ml, 5 mM-nicotinamide and 1 pM-dexamethasone) at a density of 0.25 x 106 cells/well in 24-well Primaria plates. Cultures were maintained at 37°C in an atmosphere of 5% CO2/95% air. The medium was changed after allowing for 2-4 hr cell attachment. Hepatocytes were cultured for a further 22 hr prior to the addition of fresh medium containing the test compounds (i.e. coumarin; DHC; 3-, 4-, 6and 7-MeCs and 3,4-diMeC) at the desired concentrations (0.01-3.0 mM). Coumarins were initially dissolved in dimethylformamide, and then diluted in WEC medium such that the final solvent concentration did not exceed 1%. Cultures were incubated with the compounds for 24hr and cell survival, relative to untreated cultures, was determined by a modification of the MTT reduction method of Denizot and Lang (1986). In outline, medium containing MTT (0.1 mg/ml) was added to the cell cultures. After incubating for 2-3 hr at 37°C, the MTT solution was removed and isopropanol added to each well to solubitize the resulting blue formazan product. The absorbance of the isopropanol solution was determined spectrophotometrically, using an automatic plate reader (Anthos 2001) with a test wavelength of 570 nm and a reference wavelength of 620 nm.
In some experiments, SKF-525A (10pM), an inhibitor of cytochrome P-450 activity (Netter, 1980), or BSO (5 mM), in WEC medium, was added to the cultures after 2 hr. Hepatocytes were incubated with SKF-525A or BSO for 22 hr prior to addition of the test compounds and determination of cell survival. BSO, an irreversible inhibitor of y-glutamyl cysteine synthetase, depletes cellular GSH levels. It has been reported that this compound has no effect on cytochrome P-450 levels or on P-450-dependent and conjugating enzyme activities (Drew and Miners, 1984). Culture of V79 Chinese hamster lung fibroblasts. V79 cells were cultured in Eagle's minimum essential medium as described by Horner and Shah (1984). Cells were plated out at a density of 5 x 103 cells/well in 24-well multidishes and cultured for 3-4 days until a confluent monolayer had formed. They were then exposed to the coumarins (0.5-3.0 mM) for 24 hr and cell survival, relative to untreated cultures, was determined by the MTT assay. In some experiments, cells approaching confluency were incubated with BSO (100/tM) for 24hr prior to exposure to the test compounds. Presentation of results and statistical analysis. For each 24-well plate the percentage of cell survival in treated, relative to untreated, wells was calculated for each concentration of compound from the average absorbance values of 3 or 4 wells per concentration. Dose-response curves were constructed and TCs0 values (i.e. the concentrations producing a 50% reduction in cell survival) were determined from the graphs. Each TCs0 value is the mean + SEM of 3-15 plates of hepatocytes cultured from different animals. Statistical analysis was undertaken using unpaired t-tests or ANOVA/Dunnett's test as appropriate. A value of P < 0.05 was considered to be significant. RESULTS There were no differences in the survival in cultures of hepatocytes isolated from PB-treated rats compared with controls (i.e. untreated rats). The TCs0 values determined for each compound in hepatocyte cultures are shown in Table 1. The toxicity of coumarin and its methyl derivatives was concentration dependent. In hepatocyte cultures from control rats 3-MeC was 2.7-fold more toxic than coumarin itself, whereas the other methylcoumarins were less toxic than the parent compound. PB treatment of rats increased the toxicity of coumarin (6-fold), 3-MeC (l.5-fold), 4-MeC (3.1-fold), 6-MeC (4.3-fold) and 7-MeC (11.5-fold). TCs0 values for 3,4-diMeC were similar in hepatocytes cultured from control and PB-treated animals. DHC, at concentrations ~<3 mM, did not induce a sufficient loss of viability of hepatocytes from either untreated or PB-treated rats to allow the determination of TCs0 values. The incubation of hepatocytes isolated from PBtreated rats with various concentrations of BSO (~< 10 mM) for 22 hr, prior to the addition of the test compounds, resulted in a dose-dependent depletion of cellular non-protein sulphhydryl (NPS) levels (mainly GSH). This depletion mainly affected cellular GSH, as determined by the method of Saville (1958). BSO at 5 m i depleted the NPS content to 20-25%
In vitro toxicity of coumarins
23
Table 1. Toxicity of coumarins in rat hepatoeyte cultures TC~ (gM) PB
Compound Coumarin Dihydrocoumarin 3-Methylcoumarin 4-Methylcoumarin 6-Methylcoumarin 7-Methylcoumarin 3,4-Dimethylcoumarin
Control Medium BSO SKF-525A 615 + 69 (8)*** 102 __.15 ( 15)t 23 __.4 (4)* 370 + 86 (4)** > 3000 (4) > 3000 (4) ND ND 226 __+43 (7) 150 -I- 20 (15)t 5 + 0 (4)* 292 +_127 (4)* 1305 __.316 (6)** 420 + 76 (13)~" 61 5:18 (4)* 1008 + 122 (4)** 867 + 122 (3)*** 201 + 17 (7) ND ND 1920 +_873 (3)* 167 + 40 (6) ND ND 953 _ 236 (4) 1354 + 246 (1 l)t 70 5- 35 (4)** 1257 __.225 (3) ND = not determined Hepatocytes were cultured for 22 hr in either WEC medium, or WEC medium containing BSO (5 raM)or SKF-525A (10/~M). They were incubated for a further 24 hr with the eoumarins (0.01-3.0 mM), and survival, relative to untreated cultures, was determined by the MTT reduction method. TCs0 values are means + SEM (n), where n is the number of rats from which hepatocytes were isolated. Asterisks indicate significant difference from the corresponding PB-treated (Medium) values [*P < 0.05; **P < 0.01; ***P < 0.001; unpaired t-test (control v. PB) or ANOVA/Dunnett's test (effects of BSO and SKF-525A)]. tPooled values from 3 separate experiments. t h a t of u n t r e a t e d cultures, b u t did n o t affect cell survival. The N P S level remained below 2 5 % t h a t of u n t r e a t e d cultures t h r o u g h o u t the 48-hr incubation period. Pre-treatment o f hepatocyte cultures with 5 mM-BSO significantly increased the toxicity of coum a t i n (4.4-fold), 3-MeC (30-fold), 4 - M e C (6.9-fold) a n d 3,4-diMeC (19.3-fold) in hepatocyte cultures f r o m PB-treated rats (Table 1). I n c u b a t i o n of hepatocytes from PB-treated rats with S K F - 5 2 5 A (10 #M) for 22 hr, prior to the addition of the test c o m p o u n d s , significantly decreased the toxicity of c o u m a r i n (3.6-fold), 3-MeC (1.9-fold) a n d 4 - M e C (2.4-fold). The toxicity of 3,4-diMeC was n o t affected by S K F - 5 2 5 A (Table 1). S K F - 5 2 5 A was used at the highest c o n c e n t r a t i o n that did not affect cell survival in hepatocyte cultures from PB-treated rats, as m e a s u r e d by the M T T reduction method. A p a r t from 3,4-diMeC (TCs0 = approx. 1 mM), the c o u m a r i n s were all less toxic in V79 cells t h a n in hepatocyte cultures (Table 2). 3-MeC was markedly more toxic (10-fold difference in TCs0 values) in rat hepatocyte cultures (approx. 0.2mM) t h a n in V79 cells (approx. 2 mM). C o u m a r i n , D H C a n d 7-MeC h a d little effect o n V79 cell survival, with TCs0 values all greater t h a n 3 mM. The incubation of V79 cells a p p r o a c h i n g confluency with BSO (100/~M) for 24 hr depleted the N P S c o n t e n t to 10-20% that of u n t r e a t e d cultures. The same result has been previously reported by Hodgkiss a n d M i d d l e t o n (1985). The c o n c e n t r a t i o n of BSO used did n o t affect cell survival as determined by the M T T reduction assay. The N P S c o n t e n t remained below 20% t h a t of Table 2. Toxicity of coumarins in V79 cell cultures TCso (mM) Compound Medium BSO Coumarin > 3 (14) > 3 (8) Dihydrocoumarin > 3 (10) > 3 (4) 3-Methylcoumarin 1.9 _ 0.2 (13) 2.4 +_0.1 (7) 4-Methylcoumarin 2.3 + 0.1 (14) 2.1 + 0.1 (8) 6-Methylcoumarin 1.4 + 0.1 (10) 1.2 + 0.3 (4) 7-Methylcoumarin > 3 (14) > 3 (8) 3,4-Dimethylcoumarin 1.1 _ 0.1 (13) 0.9 __.0.1 (8) Confluent V79 cell cultures, with or without prior incubation with BSO (100/~M)for 24 hr, were exposed to coumarins (0.5-3.0 raM) for 24 hr. Cell survival, relative to untreated cultures, was determined by the MTT reduction method. TCs0 values are means +_SEM (n), where n is the number of independent cell cultures.
untreated cultures t h r o u g h o u t the subsequent 24-hr incubation with the test compounds. BSO t r e a t m e n t h a d no significant effect o n the toxicity of any of the c o u m a t i n s in V79 cell cultures (Table 2). DISCUSSION Primary cultures of rat hepatocytes have been widely used for the study of metabolism-mediated cytotoxicity. In this study hepatocytes were cultured for 24 hr prior to exposure to the test c o m p o u n d s ; this allowed for acclimatization of the cells to the culture conditions. The suitability of this particular in vitro system for investigating metabolism-mediated cytotoxcity has been d e m o n s t r a t e d in a n earlier study ( H a m m o n d a n d Fry, 1991). The effectiveness of the procedure used to induce animals in vivo with PB, as well as the stability of a range of cytochrome P - 4 5 0 - d e p e n d e n t enzyme activities in hepatocytes from untreated a n d PB-treated rats, over a 24-hr culture period, have also been reported ( H a m m o n d a n d Fry, 1990). In agreement with the results of Lake et al. (1989), the d a t a show t h a t c o u m a r i n was moderately toxic in hepatocytes cultured from untreated rats. In contrast, D H C - - w h i c h lacks the 3,4-double b o n d and, unlike coumarin, is n o t hepatotoxic in vivo ( H a g a n et al., 1967; Lake et al., 1989)---had little effect o n the survival of hepatocytes in culture. In vivo induction with PB of cytochrome P-450 a n d other related enzyme activities, prior to cell isolation, markedly increased the toxicity o f c o u m a r i n in hepatocytes. This, together with the absence of c o u m a r i n toxicity in V79 cells, which do n o t possess c y t o c h r o m e P-450, a n d the inhibition o f c o u m a r i n - i n d u c e d toxicity in hepatocytes from PB-treated rats by S K F - 5 2 5 A confirm t h a t the toxicity of c o u m a r i n is due to its metabolism by cytochrome P-450 isozymes. The significant increase in c o u m a r i n toxicity subsequent to the depletion of hepatocyte G S H levels by BSO indicates t h a t G S H is i m p o r t a n t in protecting against c o u m a r i n - i n d u c e d cytotoxicity. The depletion of G S H levels in V79 cell cultures did not affect c o u m a r i n toxicity. This is consistent with the suggestion t h a t the toxicity results from a n initial bioactivation of the c o m p o u n d by cytochrome P-450d e p e n d e n t enzymes; this leads to the f o r m a t i o n o f a toxic metabolite(s), which m a y then be detoxified
24
J.H. FEr~TEMet al.
by reacting with GSH (Lake et al., 1989). The marked differences between the hepatocyte toxicity of coumarin and DHC indicates that the 3,4-double bond is an important determinant of coumarin toxicity, and it has been suggested that coumarin 3,4-epoxide may be responsible for the observed toxicity (Lake et al., 1989). The cytotoxicity of several simple methyl derivatives of coumarin, in particular 3-MeC, 4-MeC and 3,4-diMeC, which are expected to affect the formation of a putative 3,4-epoxide intermediate, was therefore investigated. In general, the toxicity of coumarin in hepatocyte cultures was reduced by substitution with one or more methyl groups. However, 3-MeC was more toxic than the parent compound in hepatocytes isolated from untreated rats. This could be due to differences in lipid solubility (the partition coefficients for coumarin and 3-MeC are 0.75 and 1.24, respectively; Feuer, 1970a) and/or metabolism. The TCs0 value for 4-MeC in hepatocytes from untreated rats was approximately two-fold that for coumarin. These data are in agreement with in vivo results showing that administration of 4-MeC to rats produces slight hepatomegaly and marked induction of hepatic microsomal drug-metabolizing activity (Feuer, 1970a,b), but no liver damage (Gibbs et al., 1971), whereas coumarin itself causes hepatic necrosis (Cohen, 1979). 4-, 6- and 7-Methylcoumarins were more toxic in hepatocytes from PB-treated than from untreated rats, but no significant differences were observed with 3-MeC and 3,4-diMeC. These latter two compounds may be poor substrates for the PB-inducible cytochrome P-450 isozymes. There was evidence for some basal cytotoxicity caused by 3-, 4- and 6-methylcoumarins, and 3,4diMeC, in V79 cell cultures. However, apart from 3,4-diMeC, all the methyl derivatives were more toxic in hepatocyte cultures than in V79 lung fibroblasts, suggesting that cytochrome P-450-dependent metabolism was involved in the toxicity of these compounds. Pretreatment of the cultures with BSO led to the depletion of GSH levels causing a significant increase in the toxicity of 3-MeC, 4-MeC and 3,4diMeC in hepatocytes but not in V79 cell cultures. This indicates a protective role for GSH with respect to the toxicity of a metabolite(s) rather than the unchanged compound itself. SKF-525A, an inhibitor of several cytochrome P-450-mediated activities (Netter, 1980), decreased the toxicity of 3-MeC and 4-MeC in hepatocyte cultures from PB-treated rats, but did not significantly affect the toxicity of 3,4diMeC. The data obtained suggest that hepatocyte toxicity caused by coumarin and 4-MeC, and possibly 6-MeC and 7-MeC is due to the cytochrome P-450-dependent formation of one or more toxic metabolites that may be detoxified by reacting with GSH. The same suggestion, however, cannot be applied to 3-MeC and 3,4-diMeC. In fact, a methyl substituent at the 3-position may actually enhance, rather than inhibit, the formation of a 3,4-epoxide. Moreover, a toxic free radical species of the two compounds may be produced resulting in toxicity by way of an oxidative stress mechanism. Thus, substitution of coumarin with methyl groups at both the 3- and 4- positions protects against metabolism-mediated cytotoxicity to
a certain extent. The results suggest that 3,4-diMeC is a relatively poor substrate for metabolism by cytochrome P-450 isozymes. However, small amounts of a reactive intermediate(s) may be produced that can cause hepatocyte toxicity when the protective effect of GSH is removed. Further studies on the mechanisms responsible for the toxicity of the methylcoumarins, and their relevance to the hepatotoxicity of coumarin itself, are needed in order to assess their toxicological implications to humans. Acknowledgement--This work was supported by the industrial sponsors of the FRAME Research Programme. REFERENCES
Cohen A. J. (1979) Critical review of the toxicology of coumarin with special reference to interspeciesdifferences in metabolism and hepatotoxic response and their significance to man. Food and Cosmetics Toxicology 17, 277-289. Cox D., O'Kennedy R. and Thornes R. D. (1989) The rarity of liver toxicity in patients treated with coumarin (1,2benzopyrone). Human Toxicology 8, 501-506. Denizot F. and Lang R. (1986) Rapid colourimetric assay for cell growth and survival. Modifications to the tetrazolium dye procedure giving improved sensitivity and reliability.Journal of lmmunological Methods 89, 271-277. Drew R. and Miners J. O. (1984) The effects of buthionine sulphoximine (BSO) on glutathione depletion and xenobiotic biotransformation. Biochemical Pharmacology 33, 2989-2994. Evans J. G., Gaunt I. F. and Lake B. G. (1979) Two-year toxicity study on coumarin in the baboon. Food and Cosmetics Toxicology 17, 187-193. Feuer G. (1970a) Induction of drug-metabolizing enzymes of rat liver by derivatives of coumarin. Canadian Journal Physiology and Pharmacology 48, 232-240. Feuer G. (1970b) 3-Hydroxylation of coumarin Or 4-methylcoumarin by rat liver microsomes and its induction by 4-methylcoumaringiven orally. Chemico-Biological Interactions 2, 203-216. Feuer G. (1974) The metabolism and biological actions of coumarins. Progress in Medicinal Chemistry 10, 85-158. Fry J. R. (1981) A comparison of biphenyl 4-hydroxylation and 4-methoxybiphenyl O-demethylation in rat liver microsomes. Biochemical Pharmacology 30, 1915-1919. Gibbs P. A., Janakidevi K. and Feuer G. (1971) Metabolism of coumarin and 4-methylcoumarin by rat liver microsomes. Canadian Journal of Biochemistry 49, 177-184. Hagan E. C., Hansen W. H., Fitzhugh O. G., Jenner P. M., Jones W. I., Taylor J. M., Long E. L., Nelson A. A. and Brouwer J. B. (1967) Food flavourings and compounds of related structure. II. Subacute and chronic toxicity. Food and Cosmetics Toxicology 5, 141-157. Hammond A. H. and Fry J. R. (1990) The in vivo induction of rat hepatic cytochrome P450-dependentenzyme activities and their maintenance in culture. Biochemical Pharmacology 40, 637-642. Hammond A. H. and Fry J. R. (1991) The use of hepatocytes cultured from inducer-treated rats in the detection of cytochrome P-450-mediated cytotoxicity. Toxicology in Vitro 5, 133-137. Hazleton L. W., Tusing T. W., Zeitlin B. R., Theissen R. and Murer H. K. (1956) Toxicity of coumarin. Journal of Pharmacology and Experimental Therapeutics 118, 348-358. Hodgkiss R. J. and Middleton R. W. (1985) Effects of glutathionedepletionusing buthionine sulphoximineon the cytotoxicity of nitroaromatic compounds in mammalian cells in vitro. Biochemical Pharmacology 34, 2175-2178.
In vitro toxicity of coumarins Homer S. A. and Shah M. K. (1984) Effects of fungizone on the in vitro cytotoxicity of other chemicals ATLA 11, 147-i53. Jamal S., Casley-smith J. R. and Casley-Smith Judith R. (1989) The effects of 5,6-benzo[a]pyrone (coumarin) and DEC on filaritic lymphoedema and elephantiasis in India. Preliminary results. Annals of Tropical Medicine and Parasitology 83, 287-290. Lake B. G. (1984) Investigations into the mechanism of coumarin-induced hepatotoxicity in the rat. Archives of Toxicology 7, (Suppl.), 16-29. Lake B. G., Gray T. J. B., Evans J. G., Lewis D. F. V., Beamand J. A. and Hue K. L. (1989) Studies on the mechanism of coumarin-induced toxicity in rat hepatocytes: comparison with dihydrocoumarin and other coumatin metabolites. Toxicology and Applied Pharmacology 97, 311-323. Marshall M. E., Mendelsohn L., Butler K., Riley L. K., Cantrell J., Harvey J., Wiseman C., Taylor T. and Macdonald J. (1987) Treatment of metastatic renal cell carcinoma with coumarin (1,2-benzopyrone) and cimetidine: a pilot study. Journal of Clinical Oncology 5, 862-866. Marshall M. E., Riley L. K., Rhodes J., Eichom T., Jennings C. D., Cibull M. and Thompson J. (1989) Effects of coumatin (1,2-benzopyrone) and cimetidine on peripheral blood lymphocytes, NK cells and monocytes in patients with advanced malignancies. Journal of Biological Response Modifiers 8, 62-69. Moran E., O'Kennedy R. and Thornes R. D. (1987)
25
Analysis of coumarin and its urinary metabolites by HPLC. Journal of Chromatography 416, 165-169. Murray R. D. H., Mendez J. and Brown S. A. (1982)
The Natural Coumarins: Occurrence, Chemistry and Biochemistry. John Wiley and Sons Ltd, London. Netter K. J. (1980) Inhibition of oxidative drug metabolism in microsomes. Pharmacology and Therapeutics 10, 515-535. Opdyke D. L. J. (1974) Monographs on fragrance raw materials: coumarin. Food and Cosmetics Toxicology 12, 385-388. Peters F. and Simonis H. (1908) ~Iber das 2-methylcumarin. Berichte 41, 830-837. Phadke C. P., Kelkar S. L. and Wadia M. S. (1984) A one step general synthesis of 3-substituted coumarins. Synthetic Communications I4, 407-411. Reese J. A. and Byard J. L. (1981) Isolation and culture of adult hepatocytes from liver biopsies. In Vitro 17, 935-940. Saville B. (1958) A scheme for the colourimetricdetermination of microgram amounts of thiols. Analyst 83, 67(b672. Shilling W. H., Crampton R. F. and Longland R. C. (1969) Metabolism of coumarin in man. Nature, London 221, 664-665. Soine T. O. (1964) Naturally-occurting coumatins and related physiological activities. Journal of Pharmacological Sciences 53, 231-264. Ueno I. and Hirono I. (1981) Non-carcinogenic response to coumarin in Syrian golden hamsters. Food and Cosmetics Toxicology 19, 353-355.