Modification of in vivo heterocyclic amine genotoxicity by dietary flavonoids

Female BALB/c mice were fed diets containing equimolar amounts of quercetin or its glycoside, rutin, for 5 weeks. These mice were used either in host-mediated bacterial mutation assays or as sources of hepatic microsomes. In host-mediated bacterial mutation assays using radiolabelled mutagens, the heterocyclic amines 2-amino-3,5-dimethyl[4,5-f]imidazoquinoline (MeIQ) and 3-amino-1-methyl-5H-pyrido[4,3-b]indole (Trp-P-2) induced greater numbers of revertants in mice fed either of the flavonoid diets compared with control. Experiments using hepatic microsomes revealed that although feeding mice either flavonoid produced slight changes in some parameters of hepatic xenobiotic metabolism (mixed function oxidase and glutathione transferase activities), microsomes from quercetin-fed mice were more potent activators of both MeIQ and Trp-P-2 compared with microsomes from control or rutin-fed mice. This difference in microsomal ability may be due to the different biological availability of the two flavonoids within the gastrointestinal tract.

The lectin binding characteristics of spontaneous and phenobarbitone induced hepatic lesions in C3H/He mice

The surface membrane glycoprotein patterns of spontaneous hepatic nodules, phenobarbitone induced nodules and hepatocellular carcinoma were studied in the C3H mouse using lectin histochemistry. The lectin binding patterns of hepatocellular carcinoma cells were markedly different to those of non-tumour cells and similar to the pattern in chemically induced hepatocellular carcinomas. This supports the hypothesis that changes in surface glycoprotein are a consistent feature associated with malignancy. Similar changes in the distribution of lectin binding sites were also seen in the phenobarbitone induced eosinophilic nodules and in a proportion of spontaneous basophilic nodules. Two populations of early basophilic nodules were identified on the basis of their lectin binding patterns, and this may indicate a link between one nodular type and carcinoma.

Modified mutagen activation in hepatic fractions from rats fed dietary rutin--interaction between gut flora and host metabolism

Male Sprague-Dawley rats were fed a purified diet or one supplemented with the glycosidic plant flavonoid (+)rutin for 14 days. Rutin treatment significantly increased caecal bacterial beta-glucosidase activity (responsible for the conversion of rutin to the flavonoid quercetin) and there was an associated increase in the capacity of hepatic fractions (S-9) to activate the food pyrolysis products IQ, MeIQ and MeIQx to bacterial mutagens in vitro. Hepatic conversion of aflatoxin B1 to a mutagen was unaltered while in vitro activation of quercetin was significantly lower in tissue fractions from the rutin-fed rats compared with those from controls. Rutin treatment was without effect, however, on a number of hepatic cytochrome P-450-dependent mixed-function oxidase activities. The results suggest that products of bacterial metabolism of rutin formed in the hindgut may influence the activity of hepatic enzymes involved in the activation of certain classes of mutagen.

Differential induction of peroxisomal and microsomal fatty-acid-oxidising enzymes by peroxisome proliferators in rat liver and kidney. Characterisation of a renal cytochrome P-450 and implications for peroxisome proliferation

The induction of renal fatty-acid-oxidising enzymes has been investigated following short-term exposure to a group of structurally diverse peroxisome proliferators and compared to the more extensively documented hepatic responses in the rat. There was a marked compound dependence on induction of both cytochrome P-450-IVA1-dependent omega-hydroxylation of lauric acid and enzymes of the peroxisomal fatty acid beta-oxidation pathway (measured as cyanide-insensitive palmitoyl-CoA oxidation and enoyl-CoA hydratase). Cytochrome P-450 IVA1 (or a very closely related isoenzyme in the same gene family) was a major constitutive haemoprotein in rat kidney microsomes and actively supported the omega-hydroxylation of lauric acid. This activity was induced 2-3-fold by peroxisome proliferators such as clofibrate, di-(2-ethylhexyl)phthalate, bezafibrate and nafenopin. By using a cDNA probe to the cytochrome P-450 IVA1 gene in Northern blot analysis, we have shown that increased renal and hepatic omega-hydroxylation of lauric acid, after treatment with peroxisome proliferators is a consequences of a substantial increase in the mRNA coding for this haemoprotein. In addition, programming of an in vitro rabbit reticulocyte translation system with both renal and hepatic RNA resulted in the synthesis of similar (if not identical) cytochrome-P-450-IVA1-related polypeptides. Furthermore, we have provided Western blot evidence that both rat liver and kidney microsomes contain two closely related cytochrome P-450 IVA1 polypeptides, the major one characterised by a monomeric molecular mass of 51.5 kDa (identical to authentic, purified hepatic cytochrome P-450 IVA1) and a minor one of 52 kDa. The kidney-supported fatty acid omega-hydroxylase activity was refractory to inhibition by a polyclonal antibody to liver cytochrome P-450 IVA1, which may be related to the existence of two closely related (but immunochemically distinct) fatty acid hydroxylases in this tissue. Our studies have also demonstrated that certain of the compounds tested (including clofibrate, bezafibrate and nafenopin) induced renal fatty acid beta-oxidation, mirroring the increased omega-hydroxylase activity in the endoplasmic reticulum. Our studies have also indicated that the kidney was more refractory to induction of the endoplasmic reticulum and peroxisomal fatty-acid-oxidising enzymes than the liver. Taken collectively, our data is strongly suggestive of a possible linkage of the renal fatty acid oxidative enzymes in these two organelles, a situation that also occurs in the liver. In addition, our studies have provided a possible conceptual framework that may rationalise the decreased susceptibility of the k

Comparative studies on nafenopin-induced hepatic peroxisome proliferation in the rat, Syrian hamster, guinea pig, and marmoset

Nafenopin was administered orally for 21 days to male Sprague-Dawley rats (0.5-50 mg/kg/day), Syrian hamsters (5-250 mg/kg/day), Dunkin-Hartley guinea pigs (50 and 250 mg/kg/day), and marmosets (Callithrix jacchus, 50 and 250 mg/kg/day). With the rat, and to a lesser extent in the hamster, nafenopin treatment produced dose-related increases in liver size and induction of peroxisomal (palmitoyl-CoA oxidation) and microsomal (lauric acid 12-hydroxylase) fatty acid oxidizing enzyme activities. In contrast, in the guinea pig and marmoset, there was no effect on liver size and only comparatively small changes were observed in these enzyme activities. Ultrastructural examination of liver sections from nafenopin-treated rats and hamsters revealed increased numbers of peroxisomes many of which lacked the characteristic crystalline nucleoid. While nafenopin had little effect on peroxisome numbers in either the guinea pig or marmoset, increases in microsomal cytochrome P450 content and mixed function oxidase activities were observed in these species. These results demonstrate marked species differences in nafenopin-induced hepatic peroxisome proliferation with the Syrian hamster being less responsive than the rat and the guinea pig and marmoset being only weakly responsive. As nafenopin is a known hepatocarcinogen in the rat, comparative long-term studies in poorly responsive species, such as the guinea pig and marmoset, may help clarify the role of organelle proliferation in the hepatocarcinogenicity of certain peroxisome proliferators.

Studies on the induction of cholangiofibrosis by coumarin in the rat

The histogenesis of coumarin-induced cholangiofibrosis in the rat has been determined. Proliferation of ductal structures was preceded by extensive damage to hepatocytes in the centrilobular region. Focal proliferation of ducts and fibrous tissue was present at 3 months and typical areas of cholangiofibrosis at 6 months. By 18 months the lesion was extensive and contained areas showing bizarre histological features suggestive of malignancy although no evidence of extra-hepatic metastasis was found. The lesion in animals returned to standard diet showed varying degrees of involution with extensive atrophy and fibrosis. A number of parameters of hepatic mixed function oxidase activity were reduced during the initial treatment period, at later times there was recovery of some microsomal enzyme activities. The activity of gamma-glutamyltransferase and the hepatic content of non-protein sulphydryl groups, in contrast, were raised throughout the treatment period.

Studies on the mechanism of coumarin-induced toxicity in rat hepatocytes: comparison with dihydrocoumarin and other coumarin metabolites

Single doses of coumarin (125 mg/kg, ip) produced a depletion of hepatic nonprotein sulfhydryl groups (mainly reduced glutathione; GSH) in young male Sprague-Dawley rats after 2 hr and increased liver weight and produced hepatic centrilobular necrosis after 24 hr. Coumarin also produced time- and dose-dependent toxic effects in primary rat hepatocyte cultures. A marked reduction of GSH levels was also observed in vitro and this was not due either to the formation of oxidized glutathione (GSSG) or to the leakage of GSH and/or GSSG from the hepatocytes. Coumarin-induced toxicity in rat hepatocytes could be inhibited by the cytochrome P450 inhibitors ellipticine and metyrapone and potentiated by depleting hepatocyte GSH levels with diethyl maleate. In contrast to coumarin, dihydrocoumarin--which lacks the 3,4-double bond--produced little toxicity in rat hepatocytes either in vivo (127 and 254 mg/kg, ip) or in vitro. Similarly, coumarin was more toxic to rat hepatocytes than a number of known coumarin metabolites including 3- and 7-hydroxycoumarin and o-hydroxyphenylacetic acid. The results of these studies demonstrate a good in vivo/in vitro correlation for the effects of coumarin and dihydrocoumarin in rat hepatocytes. Furthermore, the data suggest that coumarin hepatoxicity in the rat is due to coumarin bioactivation by cytochrome P450-dependent enzymes to a toxic metabolite(s), which may be a coumarin 3,4-epoxide intermediate. GSH appears to protect against coumarin-induced toxicity possibly by the formation of conjugates with the toxic coumarin metabolite(s).

Nafenopin, a peroxisome proliferator, depletes hepatic vitamin E content and elevates plasma oxidised glutathione levels in rats

Male Sprague-Dawley rats were given oral doses of nafenopin (80 mg/kg/d) for up to 28 d. Nafenopin administration resulted in liver enlargement and induction of peroxisomal fatty acid beta-oxidation enzymes (which generate hydrogen peroxide), but little effect was observed on catalase and cytosolic GSH peroxidase was decreased. Hepatic vitamin E levels were depleted to around 50% of control. A small increase in hepatic oxidised glutathione (GSSG) content was paralleled in a time-dependent increase in plasma GSSG. These changes in vitamin E and GSSG levels may represent early indicators of oxidative stress produced in rat hepatocytes by nafenopin and other peroxisome proliferators as a consequence of the increased production of hydrogen peroxide.

Effect of dietary fibre on the mutagenicity and distribution of 2-amino-3,4-dimethylimidazo[4,5-f]quinoline (MeIQ)

Female BALB/c mice were fed either a fibre-free diet or one supplemented with 30% wheat-bran for 5 weeks. The ability of these mice to convert MeIQ to a bacterial mutagen in vivo was determined using intrasanguinous host-mediated bacterial mutation assays. Less mutagenic activity was detected in the livers of mice fed the bran-supplemented diet compared with those fed the fibre-free diet. Subsequent experiments demonstrated that the effect of brain was not due to modifications in hepatic metabolism, but to changes in uptake of MeIQ from the gastrointestinal tract.

Comparison of the metabolism and disposition of [3-14C]coumarin in the rat and marmoset (Callithrix jacchus)

Male Sprague-Dawley rats and marmosets were given a single oral 25 mg/kg dose of [3-14C]coumarin and the excretion of radioactivity in the expired air, urine and faeces monitored up to 96 h. Excretion profiles were similar in both species with the bulk of the dose being excreted in the urine and faeces within 24 h. Chromatographic analysis of 0-48 h urine samples revealed similar metabolic profiles with only small amounts of unchanged coumarin and very little 7-hydroxycoumarin. Coumarin 7-hydroxylase activity was not detectable in hepatic microsomes from either species. These results demonstrate that the disposition of [3-14C]coumarin was similar in the rat and marmoset, a New World primate, and that both species, unlike man, are poor 7-hydroxylators of coumarin.

Comparison of the effects of ascorbyl palmitate and L-ascorbic acid on paracetamol-induced hepatotoxicity in the mouse

The effects of ascorbyl palmitate (ASCP) and free L-ascorbic acid (LAA) on the hepatotoxicity of paracetamol (acetaminophen) and the in vivo covalent binding of reactive paracetamol metabolites to hepatic proteins has been studied in male MF1 mice. The oral administration of [3H(G)]paracetamol (600 mg/kg) resulted in covalent binding to hepatic proteins, a depletion of hepatic non-protein sulphydryl (NPS) groups after 2 h, and a marked elevation of plasma alanine aminotransferase (ALAT) activity after 24 h. The co-administration of paracetamol and ASCP (1412 mg/kg, equivalent to 600 mg/kg free LAA), but not paracetamol and LAA (600 mg/kg), significantly reduced covalent binding of paracetamol metabolites at 2 and 4 h after treatment. In addition ASCP, but not LAA, significantly reduced the depletion of NPS groups and the elevation of plasma ALAT activity. ASCP also completely prevented the 35% mortality observed at 24 h in paracetamol treated mice. These results demonstrate that ASCP, but not LAA, when co-administered orally with the analgesic is an effective inhibitor of paracetamol-induced hepatotoxicity in the mouse. The mechanism by which ASCP prevents liver injury appears to involve destruction of reactive paracetamol metabolites which is associated with a sparing action on hepatic reduced glutathione levels.

Co-induction of microsomal cytochrome P-452 and the peroxisomal fatty acid beta-oxidation pathway in the rat by clofibrate and di-(2-ethylhexyl)phthalate. Dose-response studies

Male Wistar rats have been pretreated with either clofibrate or diethylhexylphthalate and the dose-dependency of induction of the microsomal, cytochrome P-452-driven fatty acid hydroxylase and peroxisomal fatty acid beta-oxidation system investigated. Both clofibrate and DEHP specifically induced (approximately 10-fold) the 12-hydroxylation of lauric acid in a dose-dependent manner and only marginally increased the associated 11-hydroxylase activity. This dose-dependent increase in fatty acid hydroxylase activity was accompanied by a similar ten-fold increase in the specific content of the cytochrome P-452 isoenzyme responsible for this activity, as assessed by an immunochemical-based ELISA method. Similarly, both clofibrate and DEHP induced the peroxisomal fatty acid beta-oxidation pathway in a dose-dependent manner. Furthermore, our results provide evidence that, after oral administration, clofibrate has a higher in vivo potency in inducing the above enzymes of fatty acid metabolism than is exhibited by DEHP. A correlation matrix analysis of the above data indicated a close association between the induction of microsomal cytochrome P-452 (and its associated fatty acid hydroxylase activity) and peroxisomal beta-oxidation enzymes, implicating a mechanistic inter-relationship between changes in fatty acid metabolising enzymes in these two hepatic subcellular organelles.

Microsomal cytochrome P-452 induction and peroxisome proliferation by hypolipidaemic agents in rat liver. A mechanistic inter-relationship

Eight structurally diverse hypolipidaemic agents have been examined for their ability to induce the microsomal cytochrome P-452-dependent fatty acid hydroxylase system and the enzymes of peroxisomal beta-oxidation in rat liver. Using a specific ELISA method, we have shown that the cytochrome P-452 isoenzyme is induced up to ten fold by hypolipidaemic challenge, concomitant with a pronounced elevation of the peroxisomal beta-oxidation enzymes, mirrored by an increase in peroxisomal volume as determined morphometrically. In addition, the induction of cytochrome P-452 is accompanied by a decrease in the activities of cytochromes P-450b and P-450c as measured by benzphetamine N-demethylase and ethoxyresorufin O-deethylase activities respectively, the latter being more extensively reduced by hypolipidaemic treatment. A hypothesis is presented whereby an early biological response is the hypolipidaemic induction of microsomal cytochrome P-452 resulting in omega-hydroxy fatty acids and their subsequent further oxidation to dicarboxylic acids, the latter providing the proximal stimulus for peroxisomal proliferation.

High levels of dietary fat: alteration of hepatic promutagen activation in the rat

Male Ola:Sprague-Dawley rats were fed diets containing either low or high levels of fats. After being fed these diets for 4 weeks, the rats were killed and individual hepatic postmitochondrial (S9) fractions were prepared. The ability of these fractions to convert the heterocyclic aromatic amines (HAAs)--2-amino-3-methylimidazo[4,5-f]quinoline; 2-amino-3,4-dimethylimidazo[4,5-f] quinoline; and 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline (compounds produced during the cooking of proteinaceous food)--to bacterial mutagens was studied, with the use of Salmonella typhimurium TA98 as indicator. Fractions from rats fed high-fat diets exhibited a greater ability to activate the HAAs than did those from rats fed the low-fat diet. The magnitude of the increase was dependent on the type of fat used.

Studies on the metabolism of deoxynivalenol in the rat

The metabolism and tissue distribution of [14C]deoxynivalenol have been studied in male PVG rats. Following administration of a single oral 10-mg/kg dose, radioactivity excreted in the urine and faeces accounted, respectively, for 25 and 64% of the administered dose within 96 hr. Less than 0.15% of the dose was detected in the respired air. Very little radioactivity appeared to be retained in any of the tissues examined after 96 hr. HPLC separation of several urinary and faecal metabolites was achieved on a reversed-phase column, using two different elution systems, one at neutral pH and one acidified. Two of the major non-polar HPLC peaks were identified by gas chromatography-mass spectrometry as unchanged deoxynivalenol and 3 alpha,7 alpha,15-trihydroxytrichothec-9,12-dien-8-one.

Structure-activity relationships for induction of peroxisomal enzyme activities by phthalate monoesters in primary rat hepatocyte cultures

Rat hepatocytes were cultured for 70 hours with a series of four isomeric octyl and five isomeric hexyl phthalate monoesters, and their effects on peroxisomal fatty acid beta-oxidation (palmitoyl-CoA oxidation) and carnitine acetyltransferase activities determined. All nine monoesters produced dose-related increases in enzyme activities and marked quantitative compound potency differences were observed. Generally octyl isomers were more potent than hexyl isomers and 2- and 3-ethyl substituted isomers were more potent than their straight chain and 1-ethyl substituted analogs. For example, mono(2-ethylhexyl)phthalate was more potent than mono(1-ethylhexyl)phthalate, and this was also observed after oral administration of the two isomers to rats for seven days. The cell culture data for induction of palmitoyl-CoA oxidation were used to generate quantitative structure-activity relationships. Relatively poor correlations were observed between biological activity and simple hydrophobic parameters, but a good correlation was obtained when compound electronic structural parameters, obtained by molecular orbital calculations, were employed. These studies demonstrate relationships between biological activity and chemical structure for a series of phthalate monoesters and indicate the potential usefulness of primary rat hepatocyte cultures to screen compounds for peroxisome proliferation.

Peroxisome induction studies on seven phthalate esters

Seven phthalate esters, representing a variety of chain lengths and degrees of branching in the alcohol moiety, were tested for their ability to produce peroxisome proliferation in the Fischer 344 rat. Di(2-ethylhexyl)adipate (DEHA) was tested using the same protocol and di(2-ethylhexyl)phthalate (DEHP) was run with each study as an internal control. Each ester was administered in the feed for a period of 21 days at levels of 2.5%, 1.2% and either 0.6% or 0.3%. DEHP and DEHA were also fed at levels of 0.1% and 0.01%. The animals were sacrificed and samples of liver were prepared for both light and electron microscopy. Serum samples were assayed for both triglyceride and cholesterol. The remaining portion of the liver was homogenized and assayed for cyanide-insensitive palmitoyl-CoA oxidation, lauric acid 11-hydroxylase and lauric acid 12-hydroxylase. The results show that there is approximately a ten-fold difference between the weakest and strongest esters in terms of their potency to induce changes in relative liver weight and in several of the biochemical parameters. In general, the longer chain esters were more potent than the shorter chain ones, and branched chain esters seemed more potent than straight. Several statistical analyses of the dataset have been performed and all render similar conclusions. The results of one of these evaluations are presented elsewhere in this volume (Lin, 1987).

Peroxisome induction studies on di(2-ethylhexyl)terephthalate

Groups of five male and five female rats were fed diets containing from 0% to 2.5% di(2-ethylhexyl)terephthalate (DEHT) or 1.2% di(2-ethylhexyl)phthalate (DEHP) for 21 days. Feed consumption and body weight gains were collected and, at study termination, animals were examined for alterations in body weight, differences in serum lipids, changes in the activities of certain enzymes associated with fat metabolism, and proliferation of hepatic peroxisomes. Feed consumption and weight gain were greatly decreased in DEHT-fed animals only at 2.5%. No biologically significant alterations in absolute liver weight occurred with DEHT. Relative liver weights were increased at 2.5% in both sexes and at 1.0% and 1.2% in females. The alterations were due wholly to decreased terminal body weights. Serum triglyceride and cholesterol levels were not found useful in interpreting the effects of DEHT. Cyanide-insensitive palmitoyl CoA oxidation and lauric acid 11- and 12-hydroxylation were increased in animals consuming 2.5%, but no lower levels of DEHT. Induction of hepatic peroxisomes did not occur at 1.2% DEHT. Interpretation of minimal peroxisomal effects with 2.5% DEHT was confounded by reduced feed consumption. Slight decreases in weight gain occurred in males consuming the 1.2% DEHP diet, but differences were minor relative to effects observed at 2.5% DEHT. Results with DEHP contrasted with those obtained with DEHT. Absolute and relative liver weights, activities of enzymes of lipid metabolism, and peroxisome content were all significantly increased at 1.2% DEHP. Reduction of feed intake was implicated in the effects observed at 2.5% DEHT, since the amount of DEHT consumed by 2.5% animals was only 1.4 times as much as by 1.2% animals. A possible explanation for the observed differences between DEHP and DEHT was related to the results of a metabolic fate study on DEHT. Metabolism of DEHT by the rat appears to occur via rapid hydrolysis of both ester linkages to give two moles of 2-ethylhexanol and one mole of terephthalic acid. Although 2-ethylhexanol has been shown to induce peroxisome proliferation, it appears to be less active in this respect than the monoester of DEHP. The relatively smaller amounts of monoester produced during the metabolism of DEHT may explain the differences seen in these experiments.

Effect of prolonged administration of clofibric acid and di-(2-ethylhexyl)phthalate on hepatic enzyme activities and lipid peroxidation in the rat

Male Sprague-Dawley rats were fed diets containing either 0.5% clofibric acid (CA) or 2% di-(2-ethylhexyl)phthalate (DEHP) for 2 years. Both compounds produced liver enlargement which was accompanied by the formation of liver nodules. Hepatic peroxisomal and microsomal fatty acid oxidising enzyme activities were induced in both large nodules and host tissue (i.e. tissue remaining after removal of large nodules) preparations from CA and DEHP treated rats. In contrast, little change in catalase activity was observed and the activities of cytosolic GSH peroxidase and GSH S-transferases were markedly reduced. Increased lipid peroxidation was observed by measurement of conjugated dienes in host tissue homogenates from CA and DEHP treated rats. Microsomal NADPH-dependent lipid peroxidation was also stimulated. Histological examination revealed extensive lipofuscin deposition in non-nodular, but not in nodular, tissue sections from treated rats. These results demonstrate that prolonged peroxisome proliferation can result in lipid peroxidation and that certain enzymes which metabolise hydrogen peroxide and organic hydroperoxides are either little affected or markedly inhibited.

The histopathology and biochemistry of phenobarbitone-induced liver nodules in C3H mice

The sequential development of hepatic nodules induced by phenobarbitone (PB) has been studied in the C3H/He strain of mouse, a strain prone to the development of spontaneous liver tumours. PB was administered in the diet to young male animals at a dose level of 85 mg/kg/day for up to 91 weeks. Control and PB-treated mice developed hepatic nodules within 60 weeks. In control animals the nodules consisted of small basophilic cells, and by 80 weeks a small proportion of the animals had developed unequivocal hepatocellular carcinoma. The basophilic nodules were similar in many respects to those induced by N-nitrosodiethylamine. In PB-treated mice the incidence of basophilic nodules was similar to that in controls. These animals also developed hepatic nodules formed of large eosinophilic cells which were readily dissectable from the surrounding host tissue by 60 weeks. Biochemical investigations into the large eosinophilic nodules from PB treated mice showed that both Phase I and Phase II types of xenobiotic metabolizing enzyme activities were induced to levels either similar to, or greater than in the surrounding host tissue. In contrast, enzyme activities in basophilic nodules from untreated mice killed at 91 weeks, were generally similar to or lower than in the surrounding host tissue. The presence of eosinophilic nodules did not lead to an increase in the incidence of hepatocellular carcinoma in the PB-treated C3H/He mice. Concurrent experiments conducted in the C57BL/6 strain of mouse did not result in the development of liver nodules at 60 weeks. Thus, the eosinophilic nodules induced by phenobarbitone in C3H/He mice appear to be distinctly different from the basophilic liver nodules arising spontaneously or to basophilic nodules produced by the hepatocellular carcinogen N-nitrosodiethylamine.