'SPECIES DIFFERENCES IN THE HEPATIC

AND RENAL RESPONSES TO CIPROFIBRATE.

A thesis presented for the

Degree of Doctor of Philosophy

by

Janet Mary Makowska

March 1988

University of Surrey

Division of Pharmacology and Toxicology

Department of Biochemistry

Guildford, Surrey

England. ProQuest Number: 10804226

All rights reserved

INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted.

In the unlikely event that the author did not send a com plete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. uest

ProQuest 10804226

Published by ProQuest LLC(2018). Copyright of the Dissertation is held by the Author.

All rights reserved. This work is protected against unauthorized copying under Title 17, United States C ode Microform Edition © ProQuest LLC.

ProQuest LLC. 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106- 1346 I would like to thank the personnel of the Toxicology

Department at Sterling-Winthrop Research Centre for dosing the mice, guinea-pigs, hamsters, rabbits and marmosets, and for designing the protocols and dosing the animals in the two six-month rat studies. ACKNOWLEDGEMENTS

I would like to express my profound gratitude to Dr.Gordon Gibson, my supervisor, for his guidance and encouragement throughout the course of this work.

I also wish to acknowledge with special thanks Dr.Frank Bonner for his helpful advice and for always providing extensive opportunity for discussion at Alnwick. I would also like to thank Dr.F. Bonner for making it possible to pursue my studies at Sterling Winthrop Research Centre, Alnwick. My thanks to all in the Toxicology Department at Alnwick, especially to Drs. C.Eason and D.Howells, for the continuous help and patience which was shown to me.

I am indebted to Dr.P. Goldfarb for his useful criticisms and informative discussion. Additional acknowledgement is due to Raj Sharma for his generous donation of cytochrome P-452 and the corresponding antibody as well as to my colleagues in the Department of Biochemistry for their encouragement.

My thanks go to the Science and Engineering Research Council and Sterling Winthrop for their financial support in the form of an SERC-CASE studentship.

I would like to express my appreciation to Professor D.V.Parke and Professor V.Marks for supporting and encouraging my period in the Department of Biochemistry.

I am also very grateful for the help and advice of Dr. Ryszard Makowski, my husband,in the proof reading.

Finally I would to extend heartfelt thanks to Ryszard, my parents, Cyril James and Barbara Hawkins, to my sisters Carol and Elaine, my brother Anthony and to John, Andy and Verity for their unflagging support and encouragement throughout. My special thanks must go to Carol and John for being there through all the "ups" and "downs". Lastly but by no means least, I would like to express my thanks to my "second" family, Professor Zygmunt and Mrs. Cecylia Makowski, Zbyszek, Barbara, Ania and Basia. CONTENTS

Page Number

Summary 1

Chapter 1 Introduction. 3

1.1 Hepatic responses to administration of 5 peroxisome proliferators.

1.2 Renal responses to the administration of 41 peroxisome proliferators.

1.3 Mechanisms of the hypolipidaemic action of 45 peroxisome proliferators.

1.4 Mechanisms of peroxisome proliferation and 49 induction of cytochrome P-452.

1.5 Response of different species to the 54 administration of peroxisome proliferators.

1.6 Molecular approach to study the hepatic 59 responses induced by peroxisome proliferators.

1.7 Aims of the present investigation. 62

Chapter 2 Methodology relating to preparation and 64 characterisation of enzyme content and activity in subcellular fractions.

2.1 Materials. 64

2.2 Animals. 65

2.2.1 Treatment of animals. 66

2.3 Autopsy procedure. 67

2.3.1 Isolation of hepatic and renal microsomes. 67

2.3.2 Preparation of tissue for electron 68 microscopy.

2.4 General analytical procedures. 70

2.4.1 Protein determination. 70

2.5 Assessment of enzyme levels and activities 71 in the microsomal fractions. 2.5.1 Determination of cytochrome P-450 content. 71

2.5.2 Determination of lauric acid hydroxylation. 72

2.5.3 Determination of ethoxyresorufin-0- 74 deethylation.

2.5.4 Determination of benzphetamine 75 demethylation.

2.5.5 Enzyme linked immunosorbent assay (ELISA). 77

2.6 Assessment of enzyme activity in the 78 mitochondrial fraction.

2.6.1 Determination of monoamine oxidase 78 activity.

2.6.2 Determination of succinate dehydrogenase 79 activity.

2.6.3 Determination of alpha-glycerophosphate 80 dehydrogenase activity.

2.6.4 Determination of carnitine transferase 81 activity.

2.7 Assessment of enzyme activity in the 82 peroxisomal fraction.

2.7.1 Determination of catalase activity. 82

2.7.2 Determination of uricase activity. 83

2.7.3 Estimation of cyanide insensitive 84 palmitoyl-CoA beta-oxidation.

2.7.4 Determination of carnitine acetyl- 85 transferase activity.

2.8 Assessment of enzyme activity in the 85 cytosolic fraction.

2.8.1 Determination of glutathione peroxidase 85 activity.

2.8.2 Determination of glutathione-S-transferase 86 activity.

2.9 Methodology relating to the molecular 87 biology of cytochrome P-452 and glutathione peroxidase.

2.9.1 Materials. 88

2.9.2 Isolation of DNA probes to cytochrome P-452 88 and glutathione peroxidase. 2.9.3 Treatment of animals. 90

2.9.4 Extraction of total cellular RNA. 90

2.9.5 Determination of RNA purity and 93 concentration.

2.9.6 RNA agarose gel electrophoresis. 93

2.9.7 Dot blots with nitrocellulose. 95

2.9.8 Nick translation. 95

2.9.9 Removal of unincorporated alpha 32-P dCTP. 96

2.9.10 Filter hybridisation with radioactive 97 probes.

2.9.11 Washing of filters following hybridisation. 98

Chapter 3 Hepatic induction potency of three 100 hypolipidaemic drugs.

3.1 Introduction. 100

3.2 Hepatomegaly. 103

3.3 The microsomal drug metabolising enzyme 105 system.

3.4 Mitochondrial enzyme activities. 106

3.4.1 Succinate dehydrogenase activity. 106

3.4.2 Alpha-glycerophosphate dehydrogenase 108 activity.

3.5 Mitochondrial and peroxisomal carnitine 109 acetyltransferase activity.

3.6 Peroxisomal enzyme activities. 112

3.6.1 Catalase activity. 113

3.6.2 Peroxisomal beta-oxidation. 113

3.6.3 Uricase activity. 115

3.7 Cytosolic glutathione peroxidase activity. 117

3.8 Discussion. 119 Chapter 4 Species and strain differences in the 127 hepatic response to short term administration of ciprofibrate.

4.1 Introduction. 127

4.2 Hepatomegaly. 128

4.3 The microsomal drug metabolising enzyme 133 system.

4.3.1 Cytochrome P-450 content. 133

4.3.2 Benzphetamine-N-demethylase activity. 135

4.3.3 Ethoxyresorufin-O-deethylase activity. 138

4.3.4 Induction of cytochrome P-452. 140

4.4 Mitochondrial enzyme levels. 14 8

4.4.1 Monoamine oxidase activity. 148

4.4.2 Succinate dehydrogenase activity. 150

4.4.3 Alpha-glycerophosphate dehydrogenase. 152

4.5 Mitochondrial and peroxisomal carnitine 153 acetyltransferase activity.

4.6 Peroxisomal enzyme activities. 157

4.6.1 Catalase activity. 157

4.6.2 Peroxisomal beta-oxidation. 160

4.6.3 Uricase activity. 163

4.7 Cytosolic enzyme activities. 164

4.8 The molecular biology of cytochrome P-452 171 and glutathione peroxidase.

4.9 Discussion. 188

Chapter 5 Comparative responses in the rat and 201 marmoset following chronic administration of ciprofibrate.

5.1 Introduction. 201

5.2 Hepatomegaly. 202

5.3 The microsomal drug metabolising enzyme 205 system. 5.3.1 Cytochrome P-450 content. 205

5.3.2 Benzphetamine-N-demethylase activity. 206

5.3.3 Ethoxyresorufin-O-deethylase activity. 209

5.3.4 Induction of cytochrome P-452. 210

5.4 Mitochondrial enzyme levels. 217

5.4.1 Monoamine oxidase activity. 217

5.4.2 Succinate dehydrogenase activity. 217

5.4.3 Alpha-glycerophosphate dehydrogenase 220 activity.

5.5 Mitochondrial and peroxisomal carnitine 220 acetyltransferase activity.

5.6 Peroxisomal enzyme activities. 223

5.6.1 Catalase activity. 223

5.6.2 Peroxisomal beta-oxidation. 223

5.6.3 Uricase activity. 227

5.7 Cytosolic glutathione peroxidase activity. 227

5.8 The molecular biology of cytochrome P-452 228 and glutathione peroxidase.

5.9 Discussion. 236

Chapter 6 Species and rat strain differences in the 266 renal response to ciprofibrate administration.

6.1 Introduction. 266

6.2 The microsomal drug metabolising enzyme 268 system.

6.2.1 Cytochrome P-450 content. 268

6.2.2 Benzphetamine-N-demethylase activity. 272

6.2.3 Ethoxyresorufin-O-deethylase activity. 273

6.2.4 Induction of cytochrome P-452. 277

6.3 Glutathione peroxidase activity. 284

6.4 Discussion. 285 Chapter 7 Discussion. 297

7 .1 Introduction. 297

7.1.1 Influence of pharmacokinetics on the 297 inductive potency of three hypolipidaemic drugs in the Fischer rat.

7.1.2 Species variability in the hepatic response 298 to ciprofibrate administration.

7.1.3 The rat and species renal response to the 299 administration of ciprofibrate.

7.1.4 Extrapolation of animal data to predict the 300 response in non-human primates.

7.2 Mechanisms of toxicity in the rat. 300

7.3 Tissue specificity of the response to 316 peroxisome proliferators in the rat.

7.4 Mechanism of induction in different 319 species.

7.5 Extrapolation of laboratory animal studies 326 to predict the response in non-human primates and man.

7.6 Suggestions for further work. 335

References. 338

Appendix 1 Haematology, clinical biochemistry and 361 autopsy procedures completed in the marmoset studies.

Appendix 2 Taxonomy of the different species. 362 SUMMARY

The influence of pharmacokinetic parameters on the hepatic

biochemical responses to the peroxisome prolif erators,

ciprofibrate, bezafibrate and clofibrate, was studied in the

Fischer rat following a 2 6-week treatment period. With once

daily dosing, the induction profiles of these compounds were

dissimilar and the order of response was ciprofibrate >

bezafibrate > clofibrate. By adjusting the frequency of dosing with respect to drug half life, i.e. clofibrate and

bezafibrate twice daily and ciprofibrate once every 48 hours,

the differences in response were ablated.

The effect of short term ciprofibrate administration on

hepatic enzyme parameters was studied in different rat strains

and species. The rat (all strains), mouse, hamster and rabbit were termed responsive due to a coordinate induction of cytochrome P-452, carnitine acetyltransf erase and peroxisomal beta-oxidation. No treatment related changes were observed in the guinea pig* In the marmoset a slight increase in peroxisomal beta-oxidation was demonstrated with no induction of cytochrome P-452 and carnitine acetyltransf erase. In responsive species increased 12-hydroxylation of lauric acid correlated with an increase in mRNA hybridising to a cytochrome P-452 cDNA probe. The guinea pig and marmoset were designated non-responsive.

In the marmoset and Fischer rat the hepatic enzyme responses to a 14-day and 2 6-week ciprofibrate administration were similar. A 4-week recovery group was included in the marmoset chronic study and the increase in peroxisomal beta-oxidation was found to be readily reversible whereas mitochondrial and microsomal changes were not. Electron microscopy revealed no peroxisome proliferation in the marmoset.

Renal enzyme parameters were examined in ciprofibrate treated animals and considerable rat strain and species differences were observed. Renal enzyme changes were minimal. The response to ciprofibrate appeared to be largely specific for the liver in responsive species.

From these results it is clear that the rat is not a suitable animal model to predict the hepatic response in the marmoset.

If it is assumed that the marmoset resembles man more closely than the rat, extrapolation would indicate that peroxisome prolif erators are not a toxicological hazard to man. 3

CHAP T E R 1

INTRODUCTION

Peroxisome prolif erators are a group of structurally diverse

chemicals which on administration to rodents cause a increase

in peroxisome number and a decrease in plasma triglyceride and

cholesterol levels (Reddy et al.,1975). This group of

compounds may be broadly classified into hypolipidaemic agents

and phthalate ester plasticisers. The term "hypolipidaemic

drug" is used in this thesis to denote hypolipidaemic agents

approved for clinical therapy either in the United States of

America or in Europe. In addition to these two categories of

peroxisome prolif erators, it has been found that certain

nutritional states stimulate the proliferation of peroxisomes

with induction of peroxisomal beta-oxidation enzymes, namely a

high fat diet and a vitamin E deficient diet (Ishii et

al.,1980, Reddy et al.,1981). The phenoxy acid herbicides,

2,4-dichlorophenoxy acetic acid (2,4-D), 2,4,5-trichloro-

phenoxy acetic acid (2,4,5-T) and 4-chloro-2-methyl-phenoxy

acetic acid (MCPA) have also been found to induce hepatic peroxisomal proliferation in rodents in a manner similar to clofibrate (Vainio et al.,1983, Kawashima et al.,1984a).

Peroxisome proliferator drugs are used to treat hyperlipidaemia, a lipid disorder characterised by elevated levels of lipoproteins and/or lipids in the blood

(Lewis,1976). Elevated cholesterol and triglyceride levels have been defined as risk factors for ischaemic heart disease.

Whether the relationship between hypertriglyceridaemia and atherosclerosis is causal or casual has yet to be

unequivocally ascertained, however the evidence is sufficient

to justify long term treatment as a preventative measure in

severe cases. With familial hyperlipidaemias, drug treatment

probably offers the only prospect of successful therapy. The

prolonged use of lipid lowering drugs has been associated with

occasional side effects (such as nausea, diarrhoea, vomiting

and infrequently, cardiac arrhythmias) of varying magnitude

(Rogers and Spector,1983). There is little information

available on the long term effects of drug therapy with these

hypolipidaemic drugs. However, two trials with prolonged

clofibrate treatment have been completed. In the US Coronary

Drug Project (treatment period of 5 years) it was found that

with patients who already have had a myocardial infarction,

there was no significant reduction in overall mortality, death

from ischaemic heart disease or further infarctions (Coronary

Drug Project Research Group, 1975). There was a small but

significant increase in the incidence of arrhythmias and

angina in patients with no previous history of this clinical condition. In the WH O coordinated trial with a treatment duration of 4-6 years, there was a decrease in non-fatal myocardial infarctions in the treated groups (Oliver, 1978).

However, in this group there was a higher non-cardiac mortality rate than with the control subjects, due mainly to an increased incidence of malignant neoplasms and complications of cholecystectomy. The current use of clofibrate has become increasingly restricted reflecting the doubt of effectiveness of its therapeutic action and the possibility of latent side effects. Two classes of hypolipidaemic agents are recognised as inducers of peroxisome proliferation; clofibrate and its structural analogues and compounds structurally unrelated to clofibrate. Examples of these groups are shown in figure 1 together with examples of phthalate esters inducing peroxisomal proliferation.

In recent years growing concern has developed regarding the possible effects of long term exposure of man to hypolipidaemic drugs (i.e. therapeutic use) and to industrial plasticisers (as a result of leouching from medicinal and food packaging) on human health. The concern is based largely on the ever increasing evidence that compounds capable of inducing peroxisome proliferation in rodent liver also induce hepatocellular carcinomas in rats and mice (Reddy, J.K. et al.,1980).

From the onset it must be emphasized that although the rodent responses to peroxisome prolif erators are amply documented, there is intense controversy regarding the relevance of rodent data to other species, particularly non-human primates and man.

1.1. HEPATIC RESPONSES TO ADMINISTRATION OF PEROXISOME

PROLIFERATORS

The administration of clofibrate and other peroxisome proliferators to rodents results in several characteristic changes. The induced hepatic responses have been studied 6

Figure 1.

Chemical Structures of Some Peroxisome

P rolif erators

I Clofibrate and its structural analogues

CM- ci-e V o - c - c o o c2h5 Clofibrate

0 - C - C 0 0 H Nafenopin

Methyl o - c - c o o c h 3 Clofenapate

c h 3 c h 3 i i O-C-COO-CH Fenofibra te I I c h 3 c h 3 Figure 1 (cont.).

I. Clofibrate and its structural analogues.

CH’ c h 3 // 0 c h 2c h 2c h 2-c - c o o h Gemfibrozil

CH, CH3 CH3

Cl~\ VcO-NH-Ch'2-CH2- / Vo-C-COCHC ch , Bezafibrate

o- c - cooh Ciprofibrate

ch3 N

Cl- / Vo-c-o-f Vet SaH-42,348 8

figure 1 (cont.).

II Compounds structurally unrelated to clofibrate.

Wy- 14,643

ch 3 ch 3 s - ch 2cooh

N—( N ... BR- 931

ch 3 c h 3 s - ch 2c - n h - c h 2ch ; OH

^n - s o 2— Tibric acid

COOH

III Industrial plasticisers.

c h 3 I c h 2 / ^ 5r.C00CH2-CH(CH2)3CH3 DEHP U-c o o c h 2- c h (ch 2) 3ch : c h 2 I c h 3

c h 3 c h 2

c o o c h 2- gh ( ch 2) 3ch 3

( c h 2)<, DEHA COOCH2-CH(CH2)3CH3 c h 2 I c h 3 extensively and can be sub-divided into ;

1. hepatomegaly,

2. proliferation of smooth endoplasmic reticulum and

induction of cytochrome P-450,

3. peroxisome proliferation and associated changes in

enzyme composition, and

4. alteration in mitochondrial number and structure with

concomitant increases in certain enzyme levels.

1.1.1 Hepatomegaly

Hepatomegaly is liver enlargement measured as an increase in the liver to body weight ratio and it appears to be a characteristic response in laboratory animals exposed to a variety of xenobiotics (Reddy and Lalwani, 1983).

The administration of peroxisome prolif erators to rodents is predictably associated with hepatomegaly, the extent of which is dependent upon the agent involved (Lake et al.,1975,

Reddy, J.K., 1980). Liver enlargement is observed in both male and female rats and this has been attributed, in part, to an increase in phospholipid and total protein levels with an associated increase in microsomal and cell sap protein concentration (Azarnoff et al.,1965, Platt and Cockrill,1969).

It is the result of hypertrophy and hyperplasia which correlates to increased hepatic DNA content and may be dependent on the dose and duration of treatment (Beckett et al.,1972). The hypertrophy is associated with a marked 10

proliferation of the smooth endoplasmic reticulum and

peroxisomes (Reddy et al.,1973). Chronic administration

results in an accumulation of protein, phospholipids and RNA

with the absence of cholesterol or triglycerides indicating an

adaptive hepatomegaly as opposed to a pathological condition

associated with fat droplets and high cholesterol and/or

triglyceride levels (Dalton et al.,1974). This liver

enlargement may be associated with mid zonal and periportal

accumulation of lipid droplets (Mann et al.,1985). Persistent

hepatomegaly has been observed in rats dosed over a 25 month

period with nafenopin (Reddy and Rao,1977).

Short-term administration of Wy-14,643 stimulated DNA

replication and cell division and the increase in tritiated

thymidine incorporation was apparent at 24 hrs following a

single dose (Reddy et al.,1979). Treatment of primary cultures

of rat hepatocytes with nafenopin caused an increased level of

thymidine incorporation into cultures as a consequence of

replicative DNA synthesis rather than DNA repair (Bieri et

al.,1984). Semi-conservative DNA synthesis and not unscheduled

DNA synthesis was stimulated with replicative DNA synthesis

occurring between the 2nd and 3rd day in all cultures (Bieri

et al.,1984). A single treatment of primary neonatal rat

hepatocytes with nafenopin enhanced hepatocyte DNA synthesis

irrespective of the environmental calcium level, whereas in normal untreated cells the Gl/S transition is physiologically controlled by the level of extracellular calcium (Armato et al.,1984). This would infer that the mitogenic effect exerted by nafenopin occurred independently of normal physiologically 11

controlled replication.

Liver enlargement occurs rapidly in response to administration

reaching a steady state level within 10-14 days which is

maintained throughout treatment. Hypertrophy may be associated

with a combined proliferation of the smooth endoplasmic

reticulum and peroxisomes and in some instances, a

contributory increase in mitochondrial number. The mitogenic

effect of nafenopin and Wy-14,643 has been amply described

(Bieri et al.,1984, Levine et al.,1977 and Reddy et al.,1979)

and it is probably common to all peroxisome prolif erators

causing hepatomegaly. It should be noted that the observed

increase in DNA synthesis occurs in the absence of cyto­

toxicity/hepatocellular necrosis and therefore represents a

replicative event and not reparative hyperplasia (Levine et

al.,1977 and Reddy et al.,1979).

1.1.2 Proliferation of Smooth Endoplasmic Reticulum and

Induction of Cytochrome P-450.

No introductory section on cytochrome P-450 has been included

as this area has been well documented. The physical and

catalytic properties, induction and multiplicity of cytochrome

P-450 have been reviewed (Lewis,1986, Guengerich,1987, Nebert

et al.,1987) and the molecular biology and induction of

cytochrome P-450 with respect to genetic variability and gene

regulation has been described at length (Adesnik and

Atchison,1986, Makowski et al.,1982, Nebert and Gonzalez,1987

and Whitlock, 1986). 12 Cytochrome P-450 Nomenclature.

Cytochromes P-450 are important in the oxidative metabolism of

many endogenous substrates and innumerable drugs, chemical

carcinogens and environmental pollutants. More than 300

chemicals, which induce their own metabolism or metabolism of,

compounds with chemically related structures, have been

described (Conney,1967). Over the last two decades workers in

the field have used a variety of different nomenclatures for

an increasing number of cytochrome P-450 forms isolated in

their laboratories. These trivial names may reflect the nature

of the inducing agent, the substrate specificity, the reaction

catalysed, the absorption maximum or may merely be

alphabetical or numerical (Guengerich,1987).

Recently a standardised nomenclature for the cytochrome P-450 gene superfamily was introduced (Nebert et al.,1987). This system is based on evolution and the recommendations include

Roman numerals for distinct gene families, capital letters for subfamilies and Arabic numerals for individual genes. Using this nomenclature, the peroxisome-prolif erator inducible isoenzyme was assigned to the P450IV gene family containing one subfamily (Nebert et al.,1987). This form was designated cytochrome P450IVA1. In this thesis, the trivial terms cytochrome P-452 (Tamburini et al.,1984) and cytochrome P-450

LAw (Hardwick et al.,1987) - are used interchangeably and are considered synonymous with P450IVA1. 13

1.1.2.1 Smooth Endoplasmic Reticulum and Cytochrome P-450

Changes

Proliferation of smooth endoplasmic reticulum (S.E.R) in

hepatocytes of rodents treated with peroxisome proliferators

is associated with a concomitant increase in certain microsomal enzymes, in particular cytochrome P-450.

Proliferation of the S.E.R in rat liver is apparent as early

as the third day of clofibrate treatment and is sustained after 14 and 30 days continuous administration with an associated increase in liver protein content accompanying hepatomegaly (Platt and Thorp,1966, Svoboda and

Azarnoff ,1966). This was found to be readily reversible, rapidly regressing after treatment was discontinued for 14 days (Hess et al., 19-65). W y - 14,643 and tibric acid induced a similar proliferation of the S.E.R (Reddy and Krishnakantha,

1975). Fenofibrate and methyl clofenapate induce a dose-dependent increase in the liver to body weight ratio and cytochrome P-450 content, yet metabolism of model substrates was either inhibited or unaffected (Orton and Higgins, 1979).

The induction of cytochrome P-450 by clofibrate has been attributed to the synthesis of specific haemoproteins based on the hepatic microsomal metabolism of model substrates (Odum and Orton,1980).

The involvement of cytochrome P-450 in the metabolism of fatty acids was implicated with the observed catalytic conversion of laurate (dodecanoic acid) to omega-hydroxylaurate using a 14

reconstituted system (Lu and Coon,1968). The omega-oxidation

of fatty acids is defined as the oxidation of fatty acids at

the omega-carbon atom (the terminal methylene carbon) whereas

(omega-1)-oxidation is the oxidation of the penultimate

methylene carbon. Extensive research has demonstrated the

involvement of cytochrome P-450 in omega- and (omega-1)-

hydroxylation of saturated fatty acids such as laurate and

unsaturated fatty acids such as oleic acid in the liver.

The first definitive report of a cytochrome P-450 species

induced by a peroxisome proliferator with substrate

specificity towards omega- and (omega-1)-hydroxy lation of

fatty acids namely lauric acid, was reported by Gibson et

al.,(1982). Four cytochrome P-450 fractions were solubilised

from hepatic microsomes of animals pretreated with clofibrate.

Purified cytochrome P-450 fraction 1 had an absorbance maxima

at 451.8nm and exhibited substrate specificity towards omega-

and (omega-l)-hydroxylation with a ratio of 12- to 11-hydroxyl products of 4:1, yet was relatively inactive with respect to the metabolism of benzphetamine (Gibson et al.,1982). Chronic

administration of clobuzarit and methyl clofenapate to male rats was found to induce liver microsomal cytochrome(s) P-450 that hydroxy late lauric acid (Parker and Orton, 1980).

Clofibrate administration was associated with a specific increase, approximately 28-fold, in the hepatic microsomal omega-hydroxylation of lauric acid (Orton and Parker, 1982).

Clofibrate at a concentration of 0.5mM in culture media maintains the cytochrome P-450 content of rat hepatocytes for 15

up to 96 hours and this was associated with a marked induction

of lauric acid hydroxylation (Lake et al.,1983). The induction

of lauric acid hydroxylation in primary hepatocyte cultures by

Wy-14,643, BR-931, tiadenol and nafenopin has also been

demonstrated (Lake et al.,1984a).

The clofibrate-induced cytochrome P-450 haemoprotein with an

absorption maximum of 451.8nm has been identified on the basis

of current criteria for homogeneity as a novel form, and designated cytochrome P-452 (Tamburini et al.,1984). Purified cytochrome P-452 isolated from the hepatic microsomes of rats pretreated with clofibrate was compared to the major phenobarbitone (PB)-induced form (cytochrome P-450b) and the major beta-napthoflavone (BNF)-induced isoenzyme (cytochrome

P-450c). The spectral properties of cytochrome P-452 were determined and using SDS PAGE, the monomeric molecular weight was ascertained to be 51,530 daltons. Peptide mapping studies using limited proteolysis of the above 3 cytochromes showed that there was no similarity between the isoenzymes induced by the three compounds, indicating fundamental differences in the amino acid sequences of the three forms. Ouchterlony double immunodiffusion analyses were used to characterise the specificity of the antibodies raised to the three haemoprotein species. No immunological cross reactivity was observed between the different antibodies suggesting a structural dissimilarity between the three forms. Cytochrome P-452 was found to have a type 1 interaction (i.e cause a low to high spin transition of the haem iron) with lauric acid. The combined 11- and 12-hydroxylation of lauric acid by this 16

haemoprotein was the highest reported value for any cytochrome

P-450 isoenzyme. Accordingly, lauric acid was considered to be

a suitable substrate marker for cytochrome P-452 (Tamburini et

al.,1984). It is possible that the increase in cytochrome

P-450 observed with administration of peroxisome

prolif erators, which in some cases has been shown to be

associated with an overall increase in lauric acid

hydroxylation, may represent the same haemoprotein species,

namely cytochrome P-452.

The involvement of cytochrome P-450 monooxygenases in the

oxidation of prostaglandins to hydroxy derivatives has also

been shown (Kupfer and Navarro, 1976). Hepatic hydroxylation of

prostaglandin alpha-1 and prostaglandin alpha-2, is primarily

at the (omega-1)-position (95%) and to a lesser extent,

omega-oxidation (5%) yielding (omega-1) (19-) and omega-(20-) hydroxy derivatives (Kupfer et al.,1978). Cytochrome P-450 prostaglandin alpha-1 omega-hydroxylase activity was induced in rabbit liver by clofibrate and DEHP (Yamamoto et al.,1986).

Complimentary DNA (cDNA) clones to cytochrome P-450 p-2

(prostaglandin omega-hydroxylase) were isolated from rabbit lung (Matsubara et al.,1987). This haemoprotein was found to share 74% amino acid similarity with rat hepatic cytochrome

P-450 lauric acid omega-hydroxylase (LAw) (Hardwick et al.,1987), suggesting that cytochrome P-450 p-2 and cytochrome

P-450 LAw are members of the same gene family (P-450 IV)

(Nebert and Gonzalez,1987). Cytochrome P-452, purified and characterised by Tamburini et al. (1984), is considered to be identical to cytochrome P-450 LAw (Hardwick et al.,1987). 17

The metabolism of arachidonic acid by rat hepatic microsomes

in the presence of NADPH and oxygen was reported by Capdevila

et al./(1981). Studies using various inhibitors indicated that

metabolism of arachidonic acid to hydroxy derivatives was by a

lipoxygenase-like function of the membrane bound cytochrome

P-450. The involvement of cytochrome P-450 in the metabolism

of arachidonic acid to a variety of products including 19- and

20-hydroxy derivatives has been established (Oliw and

Oates, 1981). Further investigation led to the identification

of several new metabolites from hepatic microsomal metabolism

of arachidonic acid including four novel epoxy acid

derivatives (Chacos et al.,1982), hydroxyicosatetraenoic acids

(HETES), (Capdevila et al.,1982), four epoxide intermediates

(Oliw et al.,1982) and epoxyeicosatrienoic acids (Chacos et

al. 1983a). It has been demonstrated that the hepatic metabolism of docosahexaenoic acid, presumably by cytochrome

P-450 monooxygenases, results in the formation of 19,20-,

16,17-, 13,14-, 10,11- and 7,8-dihydroxydocosapentaenoic acids, 22-hydroxydocosahexaenoic acid and 21-hydroxy- docosahexaenoic acid (VanRollins et al.,1984).

Epoxyeicosatrienoic acids, the 8,9- and 14,15- isomers, have been detected in human urine (Toto et al.,1987) and purified from female rabbit kidneys (Falck et al.,1987). In addition 5- and 8-HETE metabolites of arachidonic acid have been isolated

(Falck et al.,1984). The metabolism of leukotriene B4 by rat liver microsomes in the presence of oxygen and NADPH and in a reconstituted system with cytochrome P-450 LM2, has been reported (Bosterling and Trudell,1983). 18

It has been demonstrated that pretreatment of rats with the

fibric acid type hypolipidaemic drug, ciprofibrate results in

a 7-fold stimulation of omega- and (omega-1)-hydroxylation of

arachidonic acid (Capdevila et al.,1985). The omega- and

(omega-1)-oxidation products predominated accounting for 70%

of the total products whereas in PB-treated animals, omega-

and (omega-l)-products accounted for 9%. The percentage of

omega- and (omega-1)-products in ciprofibrate-treated rats was

in the ratio of 80:20 in contrast to 30:70 in PB-treated

animals. Further research in our laboratory has demonstrated

that clofibrate, bezafibrate, Wy-14,643 and clobuzarit

administration to the rat results in induction of lauric acid

hydroxylation, accompanied by a concomitant induction of

cytochrome P-452 (Bains et al.,1985). It has also been

established that the profile of arachidonic acid metabolites

from clofibrate treated rats is markedly different to that of

control animals (Bains et al.,1985). The addition of lauric

acid and arachidonic acid to hepatic microsomes from control, clofibric acid and DEHP treated animals results in the

formation of typical type 1 spectra, indicative of a substrate-cytochrome P-450 interaction (Lake et al.,1984c).

The assumption is made that the cytochrome P-450 induced by ciprofibrate is identical to that induced by clofibrate treated animals (cytochrome P-452). As such, it is thought that the major physiological role of cytochrome P-452 is in the hepatic metabolism of endogenous substrates, namely fatty acids for example arachidonic acid. It is logical to propose that cytochrome P-452 represents a constitutive form and that 19 its induction by peroxisome proliferators may be related to their hypolipidaemic effects. The former proposal has been substantiated in our laboratory where quantitation of cytochrome P-452 using a single radial immunodiffusion technique, has indicated that this enzyme is present in uninduced rat liver microsomes,.constituting 22% of the total cytochrome P-450 content. In induced rat liver microsomes the levels are proportionally higher (53.5-66.7%) (Bains et al.,1985). More recently, results from our laboratory (Sharma et al.,1988) using a more definitive ELISA technique for the immut*ioquantitation of hepatic cytochrome P-452, have revised the above figures. A more realistic estimation of cytochrome

P-452 levels is lower than previously reported (Bains et al.,1985) in that constitutive levels are approximately 5% of the total cytochrome P-450 population and clofibrate pretreatment raises this level to approximately 45%. However, it is absolutely clear that peroxisome proliferators are functioning as inducers of a constitutive (endogenous) isoenzyme of cytochrome P-450. No other class of compounds has been found to induce this isoenzyme, a characteristic seemingly exclusive to the peroxisome proliferators.

The involvement of cytochrome P-452 in the omega- hydroxylation of saturated fatty acids has been established and it has been unequivocally demonstrated that the same isoenzyme actively hydroxylates arachidonic acid. It would seem plausible that the isoenzyme cytochrome P-452 with omega- hydroxylase activity is capable on the basis of broad substrate specificity to metabolise other endogenous 20

substrates such as saturated and unsaturated medium to long

chain fatty acids. The elucidation of the precise substrate

specificity and catalytic properties of cytochrome P-452 will

be of fundamental importance in clarifying the physiological

role of this haemoprotein.

1.1.3 Peroxisome Proliferation and Associated Changes

in Enzyme Composition.

Peroxisomes are cytoplasmic organelles widely distributed in animal and plant cells (De Duve,1969). They are readily identified using the cytochemical procedure introduced by

Novikoff and Goldfischer,( 1969), to localise catalase, a peroxisome enzyme marker (De Duve,1970). Hepatic peroxisomes are spherical or oval in shape (approximately 0.2-0.6um diameter) and are surrounded by a single limiting membrane, containing a finely ground matrix with a central inclusion

(Cohen and Grasso,1981). This crystalloid core has been associated with the presence of urate oxidase (Angermuller and

Fahimi,1986). The peroxisome membrane is permeable to small solutes including sucrose, inorganic ions and the cofactors for fatty acid oxidation, NAD+, CoA, ATP and carnitine (Van

Veldhoven et al.,1987). The nonspecific permeability of the peroxisomal membrane to small solutes is based on the presence of a nonselective pore-forming protein in the membrane (Van

Veldhoven et al.,1987).

It was originally belived that peroxisomes were formed by budding from the endoplasmic reticulum (De Duve and 21

Baudhuin, 1966). However, the polypeptide composition of

peroxisomal, mitochondrial and microsomal membranes are

markedly dissimilar (Fujiki et al.,1982). Clofibrate

administration causes a marked perturbation of the

phospholipid composition of peroxisomal membranes (Crane and

Masters,1986). The current model of peroxisome biogenesis

suggests that peroxisomes form by fission from pre-existing

organelles (Fujiki et al.,1984). Peroxisomal major

polypeptides are synthesised on free polysomes and transported

to the peroxisomal membrane without any apparent proteolytic

processing (Suzuki et al.,1987). The apparent size of catalase

does not alter when it enters peroxisomes (Robbi and

Lazarow, 1978), and both uncase and catalase are transported

to the interior of peroxisomes by a post-translational mechanism (Goldman and Blobel,1978). Post-translation import

into peroxisomes in vitro of two cell-free translation

products has been demonstrated (Fujiki and Lazarow,1985).

Thiolase is synthesised as a large precursor implying post-translation import of thiolase into peroxisomes with proteolytic processing (Fujiki et al.,1985).

Peroxisomes participate in respiration, energy production and oxidative metabolism (De Duve and Baudhuin, 1966). Subsequent

identification of additional peroxisomal enzymes led to

further speculation on the metabolic roles of these organelles

(Masters and Crane, 1984). The presence of a peroxisomal fatty acid beta-oxidation pathway and identification of three enzymes involved has since been established (Lazarow and De

Duve,1976, Lazarow,1978). The involvement of rat liver 22

peroxisomes in cholesterol biosynthesis (Keller et al.,1985),

in the shortening of acyl side chains of drugs by beta-

oxidation (Yamada et al.,1984) and in the synthesis of

ether-linked glycerolipids (Balias et al.,1984) has been

demonstrated. A comprehensive list of peroxisomal enzymes has

been compiled by Tolbert, (1981). The oxidase enzymes reduce

oxygen to hydrogen peroxide and catalase utilises hydrogen

peroxide catalatically and ethanol, methanol and nitrite

peroxidatically. All the oxidases have a high affinity for

oxygen with the exception of glycolate oxidase (Reddy and

Lalwani,1983).

Peroxisomal and Mitochondrial Beta-Oxidation of Fatty Acids

The enzymes of peroxisomal beta-oxidation differ from

mitochondrial enzymes with respect to catalytic and molecular

properties. Peroxisomal beta-oxidation enzymes are KCN

insensitive in vitro and do not require carnitine unlike the

carnitine dependent KCN sensitive mitochondrial system

(Lazarow, 1982). The peroxisomal and mitochondrial beta-

oxidation pathways are depicted in figures 1.1 and 1.2 with the enzymes listed. The oxidation of fatty acyl-CoA esters in the peroxisome is catalysed by fatty acyl-CoA oxidase (a purely peroxisomal enzyme) yielding hydrogen peroxide

(Hashimoto, 1982). In peroxisomes the activities of enoyl-CoA hydratase and 3-hydroxyacyl-CoA dehydrogenase have been shown to copurify on one single polypeptide whereas in mitochondria the enzyme activities reside on separate protein molecules

(Bremer and Osmundsen,1984). Peroxisomal thiolase is Figure 1.1

The Fatty Acid Oxidation Cycle of the Peroxisome Matrix

R-CH CH CH -C

ATP ■Co ASH AMP + 2 (P) Acyl-CoA Synthetase

R-CH CH CH -C ‘SCoA

Fatty Acyl-CoA Oxidase

RCH CH=CHCSCoA

RCH C-SCoA

RCH CHOHCH C-SCoA

NAD

NADH CH C-SCoA

RCH CCH COSCoA

CoASH

[3] Enoyl-CoA hydratase [4] 3-hydroxyacyl-CoA DH [5] Thiolase Figure 1.2

The Fatty Acid Oxidation Cycle of the Mitochondrial Matrix

R-CH CH CH -C 2 2 2 'cf ATP CoASH AMP [1] Enzyme: Thiokinase + 2 (P

R-CH CH CH -C 2 2 2 SCoA [21 Acyl DH

■FAD

-FADH

RCH C RCH CH=CHCOSCoA

SCoA

[3] Enoylhydratase

RCH CHOHCH COSCoA

NAD B-OH acyl DH NADH + H

CH C

SCoA

RCH COCH COSCoA

CoASH-

[5] B Ketothiolase

Each turn of the cycle shortens the fatty acid by 2 carbons and generates one molecule of acetyl CoA and one molecule each of NADH and FADH,. 2. 25 immunologically distinct from the mitochondrial and cytosolic forms (Lazarow, 1982, Bremer and Osmundsen, 1984). The existence of a peroxisomal system for beta-oxidation of dicarboxylic acids has been reported (Mortensen et al.,1982).

1.1.3.1 Peroxisome Proliferation and Induction of

Enzymes

The in vivo effects of peroxisome proliferators on peroxisome proliferation and enzyme induction are shown in tables 1.0 -

1.2. Induction of enzymes not included in the main tables and the in vitro changes following administration of these compounds are discussed as follows. The activity of peroxisomal 2,4-dienoyl-CoA reductase is enhanced by clofibrate (Dommes et al.,1981) and the rate of C12-C19 dicarboxylic acid beta- oxidation is stimulated (Mortensen et al.,1982). The existence of two beta-oxidation pathways for unsaturated fatty acids has been suggested (Mizugaki et al.,1982).

The use of rat hepatocyte primary cultures as a screening system to investigate the effects of peroxisome proliferators, has been a recent development. Culturing adult rat hepatocytes with clofibrate revealed an increase in peroxisome number associated with a dose related stimulation of carnitine acetyltransferase activity (Gray et al.,1982). Within 12-24 hours palmitoyl Co-A oxidation was markedly increased (Gray et al.,1983). Induction of carnitine acetyltransferase and palmitoyl-CoA oxidation has been observed in primary rat hepatocyte cultures with a concomitant increase in peroxisome E-* < CO i-3(3

In Vivo Peroxisomal Proliferation and Induction of Peroxisomal Beta-Oxidation Enzymes O !—■ 1 CO. OS < E-t C3 Z os t—l U m O (3 o 1 < Q CO E-* ra O E-« ra os <5 iJ < ►a o u :=> (3 CO 3 Oi 1 >—(1 Q 1 § : M 1 1 i i 1 1 1 ac t Q 1 O 1 1—11 i i i i i i Oi 1 ca i OS 1 o t OS CO 1 1 1 X 1 1—11 3 : z o z o O co X m 01 03 01 rtj 0) Z i— i o < (3 »-3 X X Q ae E- 0 >-« i-3 z O X O ae a (3 osas x Q 0 X x x os Z X O < C3 < o < ca 0 E-* i— t i-3 CO J J O O 1 1 o 3 j

ca z Ei *4 < o Ch OS m t—i (3 ca E n o z s MH O Z O 0 X HH a O co co < < C3 ra h 1 3

OS § m o C3 HH CO C3 EC ca ca E < m iJ CO«H > H t>3 I* H r a t C os co O DR 3 3 h Q 3 3 • ra cn on CO lt r- uo CO cr* C^ rH >

HH 03 (3 t>3 CO OS S E3 on J5 m m *< OO < Z IHJQ ca m3 m3 E m £5on c*on E 1-2 HH JQ HH ca E-4 c co Z co Z De on De < 3 rH 3 1-3 1-3 rH < co Z ih Ed 3 3 3 3 h h • « • • • K • • XJ co rH OO OO cu O C3 HH CO OS S C3 m OS m < £* *< Z ca < m3 m3 E I* on I* aa <3 3 m m *< m3 E < HJ IH jQ MHJQ CO Z < ih ca E-* rtjOO HHJQ (3 3 3 h h 3 3 S. ja on rH o« o 3 o X oo *3 oo «C co co rH OS C3 1-4CO < E (3 m m m E (3 z O C3 ca E *2 m3 m3 «< m3 J5On < m < Z

• AL • X O C3 O Oi os ca E s ■* os X z Q z z o § Q ca os a On Q OO X ca a O a ca a X (3 El h • • • • • on rH rH o CO cn o m T— I CO < 3 AL ■w N C co nr oo a X CO S3 o ca os c-~ X § a Z o Cx O 03 I

u> rH on

METHYL LALWANI ET AL., LALWANI ET AL CLOFENAPATE 1983b 1983b E ca w < X h

In Vivo Peroxisomal Proliferation and Induction of Peroxisomal Beta-Oxidation Enzymes o x o ra E OS D E X o x 25 a ca CO CD ra as >-3 ra X Q E X fa as fa X < E ca o cd o hh ca x S W h h h h X X X ca i x < < o I H H X X H H O 1 1 < 1 fa X H H ca o H H O x hh ca I O x x X X CO o <3X s a I H H x x m X Eh O HH X s a O X Q X u s < < O Ed X X o x ra o X a < x x <3 q os ca re 0 X X >H Q o c 0 ca x X X E re HH O X ca fa < O O 0 1 o 1 1 OS h

o fa E fa x X X H H cd <3 ca ca h

1 1 1 1 1 1 1 1 I 1 1 E X < CD fa x ca H H H H o X CD X Q < X X X X X X X co X fa E *3 ca X < Q h h • 03 O C rH < O hh c-~ cn cn rH ca ra CD <3 . H. 03 ra rH X X X rH DS co X j t r 03 < H H W E X < X r 3 0 H H ca E l h h • ■ rH O G CD X o X fa E ca X o *3 X < X H H fa X X < ca < X fa X fa X nr Q X co X CD «} fa h » oa Hi— 1 rH X H H O o E HH X HH O HH a X X X X fa X fa HP Q o < ca h •» • cn CO < o ca X O X X X E X HH a f <3 < a s a E fa ja fa H H a a & h i-3 <3 HH fa E X <3 X X X faGO X Q o CO X ca E h h h h 3 • • • * , X a r X H » 3 0 CO CO 03 CO ca XJ < t > ~ c o h — H 3 0 . X 1 H H E HH X o < O X HH a <3 X o h ^ Q Q «3 X fa X X X Q rH Q CQ C Q o H fa X X <3 X X X X Q X < O O <3 X X r • “ Q X fa rH O a X E a f HO X NS X rtj X c x. X O fa «*HO cn Q a f E h h h . i • X • X • • • X • *

3 0 - r a u h E X X X X H H ca rH en e'­ ID H H X CO X «! O cn E CO H o h h 3 * * • • » . * • •

rH Hf CO r n ra X fa fa X X o h 5 r E O x a x < X X HO CJ rH CJ X X ca X a < Q X X >"3 X a f - a c-~X i a C < a X 3 X * fa X Q C- Q c- x n < 3 < 1 * * • • X • * aX • • ra • • X

3 0 > r i * * r C" t— a on X X ca H H X <3 rH X X < X h

X X Q fa X < X E W a fa o ra X ca as O Q H H ca ra o § ca o Q X o ra ca x X H H H H ca *< H H H H 3 O < E X O Q x o ra ca X X SB E O Q h

o t 03 ~ r rH 03 03 x —1 T— cn ca X X H H cn X O i—1 *3 E X O X h h h •* * •

MEHP LAKE ET AL 1975 TABLE 1.2 IN VIVO INDUCTION OF NON-Beta-OXIDATION PEROXISOMAL ENZYMES D C O C - J _ CD >- C 3 ca ca 3 C O C D C - x

- C < 2= ca O i LU C LU i O >- ca S I 2S U L D «C CD ca LU Oi CD LU CO CO LU U *ch LU J —

ca

L U CO CO U L — i C om >- CO 1 _i ^ Qi a O C •c u l 1 — *c — r-* — •J cn

CM* O C cm>- O L U L - n m c r-* D U

u l D C cutu C 2 U L C I— 03 m C O C

Oi •—«Oi 0-» D 3 C 0 U L < Oi D C m c u l LU

i O D C 2 a c a c U L D C d c ci = a c cn - r o c m c LO m C 03 r—

HY-14,643 J.K.REDDY AND WW AND KRISHNAKANTHA., 1975 Ktiw*, 1978 29 number following the administration of nafenopin, BR-931, tiadenol and acetylsalicylic acid (Gray et al.,1983).

In general it would appear that in vitro experimentation provides a sensitive and rapid means of examining the hepatic responses to peroxisome proliferators.

1.1.4 Mitochondrial Changes

Treatment of rats with many structurally diverse peroxisome proliferators results in several observable mitochondrial changes which may include proliferation, structural changes and induction of certain enzymes.

The administration of clofibrate to rats results in increased specific activity of cytochrome oxidase and glycerol-1- phosphate dehydrogenase (Hess et al.,1965, Westerfeld et al.,1968). Structural changes in hepatic mitochondria of rats and mice have been observed (Svoboda and Azarnoff,1966) with marked organelle enlargement (Reddy et al.,1969). A 50-100% increase in the hepatic content of mitochondria has been reported (Kurup et al.,1970). The maximum increase was seen at

2-4 days after administration (Gear et al.,1974); no change in particle size or specific content of DNA or RNA was observed and the organelle half-life was unaltered. A 30-40% decrease in the specific activity of monoamine oxidase, kyuramine hydroxylase and. rotenone-insensitive NADH-cytochrome c reductase was observed from 2-15 days. Neutral protease activity was inhibited (Gear et al.,1974). Chronic clofibrate 30 treatment resulted in a 30% decrease in oxygen consumption and carbon dioxide production associated with the oxidation of fatty acids, indicative of impaired oxidation with a reduction in ketone bodies produced (Cederbaum et al.,1976).

A two to three-fold increase in hepatic total acyl-CoA synthetase activities was demonstrated and it is assumed that clofibrate treatment results in an induction of mitochondrial acyl-CoA synthetases (Krisans et al.,1980). Carnitine acetyltransferase is found primarily in the hepatic mitochondrial fractions and it constitutes 54% of the total enzyme activity whereas it is 2-3 times lower in peroxisomes

(14%) (Markwell et al.,1973). This enzyme has been purified and characterised (Mittal and Kurup,1980). The substrate specificity pattern for peroxisomal carnitine acyltransferases is bi-modal with maxima at C4 and C12, however in mitochondria the specific activity appears to correspond to two different enzymes with maximal specificities for C4 and C12 fatty acids

(Leighton et al.,1982). In ciprofibrate treated rats a 21- and

7-fold increase was observed in mitochondrial C4 and C12 transferase activities respectively (Leighton et al.,1982). A general 4 to 5-fold increase in peroxisomal acyltransferase activity was evident following ciprofibrate administration

(Leighton et al.,1982). A variable increase in carnitine acyltransf erase activities, dependent on the chain length of the respective substrates, was also demonstrated in clofibrate treated animals (Solberg et al.,1972). Induction of carnitine acyltransferases exhibited an apparent dose-response increase 31

which displayed a time lag during which the activity increased

in microsomal and peroxisomal fractions (Mittal and

Kurup,1981). The increase in the specific activity of the

acyltransf erases was in the order of 5:3:2 for short-, medium-

and long-chain transferases, respectively (Solberg et

al.,1972). The activity of carnitine palmitoyltransferase, an

enzyme exclusive to mitochondria, increased to 260% of the

control activity following clofibrate treatment (Solberg et

al.,1972).

The isolated hepatic mitochondrial fraction from clofibrate

treated animals exhibited an increased capacity for the

oxidation of short chain acyl carnitines (including acetyl-

carnitine) whereas the oxidation of palmitoyl- and erucoyl-

carnitine showed little change (Christiansen et al.,1978). The

activity of 2,4-dienoyl-CoA reductase was increased (Dommes et

al.,1981). The ratio of mitochondrial protein to cellular DNA was increased 2-fold with no alteration to the apparent rates

of degradation of protein (Lipsky and Pedersen, 1982). The total activity of carnitine acetyltransferase was induced whereas sensitivity to malonyl-CoA was decreased; a similar pattern of increased activities was observed with palmitoyl-CoA and octanoyl-CoA as substrates (Lund and

Bremer,1983). SaH-42,348 and nafenopin treatment resulted in a large increase in mitochondrial protein and phospholipid in accordance with the observed increase in organelle size and number (Dalton et al.,1974, Kurup et al.,1970). Tibric acid and fenofibrate administration resulted in a dose dependent increase in alpha-glycerophosphate dehydrogenase activity 32

(Holloway and Thorp,1979), reflecting increased thyroxine

activity in the liver (Holloway and Orton, 1980). Gemfibrozil

increased the activity of this enzyme in addition to carnitine

acyltransf erase (Kahonen and Ylikahri, 1979). Bezafibrate,

ciprofibrate, fenofibrate and tiadenol also increased the

activity of hepatic carnitine palmitoyltransferase in rats

(Halvorsen,1983). Fenofibrate administration significantly

increased glutamate dehydrogenase, carnitine palmitoyl­

transferase, carnitine acetyltransferase activities and

mitochondrial beta-oxidation (Van Veldhoven et al.,1984).

Methyl clofenapate induced alpha-glycerophosphate

dehydrogenase activity (Krishnakantha and Kurup,1974).

The induction of mitochondrial enzymes involved in beta-

oxidation of fatty acids would suggest, in conjunction with

other known cellular responses, that a common mechanism is

involved in the hypolipidaemic effect of peroxisome proliferators.

1.1.5 Changes in Cytosolic Enzyme Levels

The influence of peroxisomal proliferators on cytosolic enzymes has not been studied extensively. However, the major changes appear to be at the level of cellular defence, namely glutathione peroxidase, superoxide dismutas.e and epoxide hydrolasej

!enzyme activities.// L—■■ -— v

Glutathione peroxidase is a selenium containing tetrameric protein located in mammalian cell cytosol and mitochondria. 33

The enzyme consists of four identical sub-units each of

molecular weight of 21,000 daltons. It is present at the

highest levels in the cytosol of liver, kidney and erythroid

cells (Wendel,1980). This enzyme converts glutathione to

glutathione disulphide during hydroperoxide metabolism and

utilises hydrogen peroxide, lipid and steroid peroxides as

substrates (Wendel,1980, Halliwell and Gutteridge,1985). A decrease in hepatic glutathione peroxidase and superoxide dismutase activities has been observed in the rat following administration of clofibrate and fenofibrate for 30 days

(Ciriolo et al.,1982 and 1984). Superoxide dismutase catalyses the conversion of superoxide radicals into oxygen and hydrogen peroxide. The superoxide radical is a highly reactive species of oxygen which is likely to cause irreversible damage at the site of origin unless it is removed. The hydrogen peroxide formed is less reactive than the radical and may undergo further metabolism by glutathione peroxidase and catalase

(Lehninger,1975).

It has been reported that following a 14 day clofibrate treatment, glutathione peroxidase levels remain unchanged

(Hietanen et al.,1985). Glutathione peroxidase activity was decreased 40-60% in rats and mice pretreated with nafenopin and DEHP (Tomaszewski et al.,1986). Nafenopin also decreased glutathione transferase activity (Furukawa et al.,1985). Epoxide hydrolase is induced in microsomal and cytosolic fractions by clofibrate

i(Crane and Masters,1986). 34

1.1.6 Hepatocarcinoqenesis Induced by Peroxisome

P rolif erator s

The extrapolation from rodent carcinogenesis studies to other

species is rarely satisfactory as qualitative and quantitative

differences exist between the respective responses. It should

be noted that predictions are based on a direct interspecies

comparison and it is assumed that the mechanism of action in

the model is both identical and reproducible in man. Rodent

studies completed on several structurally diverse peroxisome

proliferators have established that long term treatment

results in hepatocarcinogenesis (Reddy,J.K. et al.,1980). It

is now believed that that these compounds in rodents, are as a class, non-genotoxic carcinogens.

1.1.6.1 Short Term Carcinogenicity Tests

The potential of clofibrate, nafenopin, SaH-42,348, tibric acid, Wy-14,643 and BR-931 to produce DNA damage has been tested using the lymphocyte thymidine incorporation and

Salmonella/microsomal assays (Warren et al.,1980). In the absence and presence of liver 9,000g supernatant, all of the above compounds were negative in 5 Salmonella typhimurium strains and in both assays (Warren et al.,1980). Using the same strains (TA 98, TA 100, TA 1535, TA 1537 and TA 1538), gemfibrozil and 5 urinary metabolites failed to induce a significant increase in revertant bacterial colonies

(Fitzgerald et al.,1981). Similarly, other peroxisome proliferators such as methyl clofenapate and bezafibrate were 35

negative in Ames tests (Reddy and Lalwani, 1983, Reddy,J.K. et al.,1982). Wy-14,643 induced unscheduled DNA synthesis in rat hepatocytes, whereas ciprofibrate and nafenopin exhibited no apparent effect (Glauert et al.,1984). BR-931 mediated unscheduled DNA synthesis was observed only at the maximum non-toxic dose. It should be noted that all 4 compounds were negative in the Ames/Salmonella assay using the newly developed strains TA 102 and TA 104 which are more sensitive to a variety of agents including oxidative mutagens (Glauert et al.,1984). The induction of sister chromatid exchanges by

2,4-D, M C P A and clofibrate, in vivo and i n vitro has been studied (Linnainmaa,1984). No significantly increased induction occurred and there was no apparent dose-related effects.

1.1.6.2. Long Term Carcinogenicity Studies

Clofibrate administered in the diet at a concentration of

0.5% (w/w) to male F344 rats resulted in the development of hepatocellular carcinomas in 91% of the treated animals sacrificed between 24 and 28 months (Reddy and Qureshi, 1979).

The long term administration of nafenopin to acatalasemic mice produced a 100% incidence of hepatocellular carcinomas between

18 and 20 months in male and female animals (Reddy et al.,1976a). Between 18 and 25 months, hepatocellular carcinomas developed in 7 3% of nafenopin treated male F-344 rats (Reddy and Rao, 1977a). Dietary administration of

Wy-14,643 to male acatalasemic CSb mice and F-344 rats resulted in a 100% incidence of liver tumours in animals surviving chronic administration at 14.5 and 16 months,

respectively (Reddy et al.,1979). Using a semi-purified diet

to avoid possible xenobiotic contamination, Wy-14,643 was

administered to male F-344 rats for 65 weeks (Lalwani et

al.,1981a). Fourteen out of fourteen treated animals developed

liver tumours between 40 and 65 weeks. A significant increase

in the incidence of liver tumours in female CSb mice and F-344

rats has been observed with long term treatment with BR-931

(Reddy,J.K. et al.,1980). A 97% incidence of liver tumours in

male F-344 rats treated for a period of 16.5 months with

tibric acid has also been reported (Reddy, J.K. et al.,1980).

Gemfibrozil was administered in the diet to groups of non

inbred CD/1 mice and non inbred CD rats for 78 and 104 weeks

respectively (Fitzgerald et al.,1981). In mice there was no

significant increase in the frequency of mean latency period

of tumours. Statistically significant increases in the

incidence of liver tumours were observed in rats and these

appeared to be dose related.

The administration of bezafibrate at concentrations effective

in decreasing serum triglycerides did not cause an increased

incidence of liver tumours in rats (Fahimi et al.,1982). At

substantially higher dose levels, an enhanced rate of

hepatocellular carcinoma development was observed (Hartig et

al.,1982). Dietary administration of 0.1%(w/w) (equivalent to

lOmg/kg) ciprofibrate for a period of 60 weeks to male F-344

rats resulted on the basis of histological criteria, in a 100%

incidence of developing liver tumours (Rao et al.,1984).

Methyl clofenapate administration to male F-344 rats caused 37

the development of hepatocellular carcinomas between 65 and 75

weeks in fourteen out of fourteen animals (Reddy,J.K. et

al.,1982). A slight increase in carcinoma incidence was seen

in fenofibrate treated animals (Reddy and Lalwani, 1983).

In conclusion, it should be stated that all the peroxisome

proliferators tested, produce hepatic tumours in rodents.

1.1.6.3 Possible Mechanisms of Hepatocarcinoqenesis

In general in carcinogenesis, both initiation and promotion

are important phases (Berenblum,1974). Chemically induced

initiation involves a direct interaction of the chemical or

electophilic metabolite with cellular DNA with a subsequent

alteration to the genome, whereas promotion is a stimulation

of the transformed cell with ultimate tumour production

(Taussig, 1984). Epigenetic or non-genotoxic carcinogens are

defined as compounds which do not damage DNA but act by

indirect mechanisms (Williams,1981). All the peroxisome

proliferators tested produce hepatic tumours in rodents

following prolonged administration at high dose levels and yet

exhibit no apparent capacity to induce genetic damage using the range of mutagenicity tests currently available. This would infer that this group of compounds are epigenetic carcinogens, consistent with reports that these compounds enhance the carcinogenic effect of previously administered genotoxic carcinogens. For example, administration of

Wy-14,643 to male F-344 rats following a prior "initiating" period of 2 weeks with diethyl-nitrosamine (DEN) resulted in the development of a significantly higher number of liver tumours in comparison to the DEN group (Reddy and Rao,1978).

That hepatocellular carcinomas develop in animals treated with peroxisome proliferators alone indicates that these compounds possess both promoting and initiating ability. It is possible that peroxisome proliferators can produce liver tumours, if they promote tumour development from spontaneously arising pre-neoplastic foci, susceptible to endogenous and exogenous growth stimuli (Schulte-Hermann et al.,1983).

The development of gamma-glutamyltransf erase (gamma-GT) negative altered foci, neoplastic nodules and hepatocellular carcinomas between 21 and 70 weeks has been demonstrated with

Wy-14,643 (Rao et al.,1982). This indicated that gamma-GT cannot be considered as a general phenotypic marker of all pre-neoplastic cells. This would suggest that the precursor lesions of induced hepatocellular carcinomas may differ from the gamma-GT enzyme positive altered hepatocyte foci induced by genotoxic agents. It would appear that this group of compounds induce neoplasia by initial non-genotoxic processes.

However, it is possible that initial genotoxic events occur and that the tests currently used are not sufficiently sensitive to detect the changes. It is apparent that prior to the established promoting activity of these compounds, initiation must occur, whether by non-mutational events or by the indirect action of the chemical.

It has been suggested that the peroxisome proliferators may 39

interact with a non DNA target producing active oxygen species with subsequent DNA attack (Reddy and Lalwani, 1983). In such

an event, the normal cellular protective mechanisms utilising glutathione would operate with the probable removal of the

short lived oxygen radicals. Reddy has proposed that the marked peroxisome proliferation produced by the administration of these compounds acts as the endogenous initiator of neoplastic change (Reddy,J.K. et al.,1980). Administration to rodents results in peroxisome proliferation with increased activity of fatty acyl-CoA oxidase and a marked induction of fatty acid beta-oxidation pathway enzymes, resulting in a 2 to

22-fold stimulation of peroxisomal beta-oxidation (Cohen and

Grasso,1981, Inestrosa et al.,1979, Lalwani et al.,1981a).This is thought to result in the excessive production of hydrogen peroxide, which on the basis of a 2.5-fold induction of catalase, cannot be adequately removed (Reddy and

Lalwani, 1983). This surfeit of hydrogen peroxide could lead to the formation of reactive oxygen species leading to lipid peroxidation. Persistent lipid peroxidation may ultimately lead to DNA attack and damage providing the initiating mutational event.

However, the beta-oxidation step which generates hydrogen peroxide in peroxisomes appears to be rate-limiting as the specific activity of fatty acyl-CoA oxidase is lower than that of the other beta-oxidation enzymes (Hashimoto, 1982). Under normal physiological conditions the hydrogen peroxide produced in peroxisomes is largely disposed of by catalase whereas hydrogen peroxide arising from mitochondria, the endoplasmic 40

reticulum or cytosolic enzymes such as superoxide dismutase is

acted upon by glutathione peroxidase (Halliwell and

Gutteridge,1985). Catalase has a relatively minor role in

hydrogen peroxide catabolism at low rates of hydrogen peroxide

generation. Xn the endoplasmic reticulum however the catalase

function increases as the rate of hydrogen peroxide production

is enhanced (Jones et al.,1981). It should be noted that the

specific activity of catalase remains constant throughout

treatment with peroxisome proliferators (Goldenberg et

al.,1976), which combined with its extremely high turnover

number for hydrogen peroxide may be capable of removing the

excess hydrogen peroxide produced during peroxisome

proliferation. One molecule of bovine liver catalase

metabolises 2.5 million molecules of hydrogen peroxide per

minute under optimal conditions (Lehninger, 1975).

Consistent with the lipid peroxidation proposal of neoplastic

transformation is the observed accumulation of lipofuscin in

the livers of treated animals (Reddy, J.K. et al.,1982). In a

recent study it has been reported that the antioxidants,

ethoxyquin and BHA, which have free radical scavenging properties exert an inhibitory influence on ciprofibrate

induced hepatic tumourigenesis (Rao et al.,1984). This

indicates that hydrogen peroxide and/or free radicals may be

involved in the hepatocarcinogenesis induced by this compound.

However, it has been estimated that the extent of radical production in the livers of rats exposed to peroxisome proliferators is insufficient to result in a biologically significant degree of DNA damage (Elliot and Elcombe,1987). 41

It is apparent that peroxisome proliferators possess both

initiating and promoting abilities and it is possible that the

initiating effect may result from a combination of non-

mutagenic events or as yet, a not clearly defined mutagenic

effect, mediated by the indirect action of these chemicals.

1.2 RENAL RESPONSES TO ADMINISTRATION OF PEROXISOME

PROLIFERATORS

Unlike the liver, no kidney enlargement and renal proliferation of the endoplasmic reticulum has been observed in animals treated with peroxisome proliferators. The major cellular response of the kidney to treatment is peroxisome proliferation with a corresponding increase in peroxisomal beta-oxidation enzymes and induction of the polypeptide PPA

80,000 (peroxisomal-proliferation-associated 80,000 mol.wt. polypeptide, immunoc hemic ally identical to the bifunctional protein).

The distribution of the peroxisomes in the kidney of untreated rats is primarily in the P-3 segment, i.e the cells of the par descendens of the proximal convolution located in the outer medulla and medullary rays (Le Hir and Dubach,1982). In the cells of the pars convoluta in the cortex (P-l and P-2 segments), peroxisomes are smaller and less reactive. In the

P-3 cell, mitochondrial activity is less than that in P-2 and

P-l, suggesting that P-3 cells possess an extra mitochondrial means of oxidation in which peroxisomal oxidases play an important role. Fatty acyl-CoA oxidase activity is localised 42

in the matrix of renal peroxisomes (Yokota and Fahimi, 1982).

In mice fed . BR—931, Wy-14,643, fenofibrate and methyl

clofenapate, a marked peroxisome proliferation was induced in

the proximal convoluted tubular epithelium of the kidney

(Lalwani et al.,1981). This peroxisome proliferation was

associated with an enhanced activity of the peroxisomal

palmitoyl-CoA oxidase system in the renal cortical

homogenates. The activity of peroxisomal heat labile enoyl-CoA

hydratase in the kidney cortex increased 3 to 5-fold on

treatment with peroxisome proliferators. Enoyl-CoA hydratase

and PPA-80 which were also substantially induced upon

treatment, were found to have an identical distribution,

limited exclusively to the cytoplasm of the proximal

convoluted tubular epithelium (Lalwani et al.,1981). A rapid

increase in renal palmitoyl-CoA oxidase and palmitoyl-CoA

oxidation was seen with clofibrate with a minimal elevation in

the activity of catalase (Small et al.,1982). Administration of SaH-42,348 increased the activity of catalase in male wild-type mice and caused a mild to moderate increase in peroxisomal number in the proximal convoluted tubular epithelium (Reddy et al.,1975).

Peroxisomal fatty acid acyl-CoA oxidase activity is restricted to the proximal tubule indicating that this is the location of the renal peroxisomal beta-oxidation system (Le Hir and

Dubach, 1982). After 48 hours of starvation the peroxisomal enzyme activity increased in glomeruli, proximal convoluted tubules and collecting ducts. 3-Hydroxy-acyl-CoA, 43

dehydrogenase, representing the mitochondrial pathway, is

distributed in all cortical proximal and distal segments. It's

activity is much lower in glomeruli and collecting ducts and

is unaffected by starvation. It would appear that the proximal

tubule possesses a peroxisomal pathway for beta-oxidation with

a capacity of the same order of magnitude as in hepatocytes

(Le Hir and Dubach,1982). The induction of polypeptide PPA-80,

heat labile enoyl-CoA hydratase has been shown to parallel the

increase in peroxisome number in the proximal convoluted

tubular epithelium (Lalwani et al.,1981). The induction of

mitochondrial carnitine acetyltransferase in the kidney as the

result of clofibrate treatment has been reported (Mittal and

Kurup,1981).

The existence of at least one haemoprotein species exhibiting

a carbon monoxide adduct absorption maxima of 452-454nm with

omega-hydroxylase activity, has been reported in rat kidney

cortex microsomes (Jakobsson et al.,1970). A cytochrome P-450

species, designated P-450 K-5, with a monomeric molecular

weight of 52,000 and a CO-reduced absorption maximum at 452nm,

was purified from untreated rat kidney microsomes. This

haemoprotein catalysed the omega- and (omega-l)-hydroxylation

of lauric acid (Imaoka and Funae,1986). Cytochrome b5 plays a

role in both NADPH- and NADH-dependent hydroxylation of fatty

acids in kidney microsomes (Sasame et al.,1974). The

involvement of renal cytochrome P-450 in the omega- and

(omega-l)-hydroxylation of prostaglandins has also been established (Okita et al.,1981). Two renal cytochrome P-450 forms, P-450 K-l and P-450 K-2 were purified (Yoshimoto et 44

al.,1986). The monomeric molecular weights were 51,500 and

52,000 respectively, with absorption maxima of 450.5 and

451nm. Both forms catalysed the omega- and (omega-1)-

hydroxylation of fatty acids; P-450 K-l exhibited a higher

specific activity with all fatty acids and catalysed the

omega-hydroxylation of PGA-1 and PGA-2. It is generally

accepted that renal cytochrome P-450 monooxygenase enzyme

activity is directed towards the omega- and (omega-1)-

oxidation of fatty acids and that there may be one or several

haemoprotein species involved. It appears that one cytochrome

P-450 is involved in both omega- and (omega-l)-hydroxylation

of saturated fatty acids and that omega-hydroxylation

predominates. The ratio between omega- and (omega-1)-

hydroxylation depends on chain length, the ratio decreasing

with increasing chain length (Ellin et al.,1973).

The metabolism of arachidonate primarily to 20-(omega) and

19-(omega-1)-hydroxy arachidonate and additionally 19-keto

arachidonate and a dicarboxylic acid has been demonstrated

(Morrison and Pascoe,1981). The stimulatory effect of the

hypolipidaemic agent clobuzarit on kidney cortex microsomal cytochrome P-450 (2 to 3-fold induction) has been reported.

The induction of renal cytochrome P-450 was associated with a similar fold increase in total lauric acid omega- and

(omega-l)-hydroxylation; PB and saline had no effect on fatty acid hydroxylation (Parker and Orton, 1980) .The relationship

(if any) between hepatic cytochrome P-452 and peroxisome proliferator induced kidney cytochrome P-450-dependent fatty acid hydroxylase is not clear at present. 45

1.3 MECHANISMS OF THE HYPOLIPIDAEMIC ACTION OF

PEROXISOME PROLIFERATORS

The hepatic subcellular, biochemical and ultrastructural

responses observed in rodents following treatment with

peroxisome proliferators has been clearly established.

However, the relationship between these changes and the

observed hypolipidaemic activity has yet to be clarified.

As discussed previously, the proliferation of S.E.R. and the

concomitant induction of cytochrome P-452 displaying

specificity towards omega- and (omega-1)-hydroxylation of

lauric acid, has been demonstrated (Tamburini et al.,1984). An

increase in cytochrome P-452 omega-hydroxylation of fatty

acids would reflect the overall enhanced lipid catabolism

resulting from clofibrate treatment although omega-oxidation

has been calculated to account for only 5-10% of the total

oxidation of fatty acids in starved rats (i.e. in the non

drug-induced state) (Bjorkhem,1978). However in

normolipidaemic and hyperlipidaemic fats, the antilipidaemic

action of clofibrate does not depend critically on the

enhanced omega-hydroxylation of fatty acids (Reich and Ortiz

de Montellano, 1986). In starved rats, competition exists

between the acyl-CoA synthetases involved in the biosynthesis

of glycerides and the microsomal hydroxylase(s) involved in

omega-oxidation of fatty acids (Bjorkhem,1976). It is possible that elevated omega-oxidation in treated animals may be inhibitory to triglyceride synthesis, competing with this pathway for fatty acid substrates. This is substantiated by 46

the observed inhibitory effect of clofibrate on fatty acid

synthesis (Landriscina et al.,1975). The physiological

significance of omega-oxidation may be to produce succinyl-CoA

from fatty acids in mitochondria stimulating oxidation of

accumulated acetyl-CoA thereby conserving carbohydrate stores

(Wada et al.,1971). The hypocholesterolaemic action of DEHP

arises largely from its ability in vivo to stimulate the

degradative removal of the sterol as bile acids (Nair and

Kurup,1986).

Mitochondrial changes have been observed in rodents on

treatment with peroxisome proliferators and these include

proliferation and alterations to the inner membrane. The major

biochemical feature is the induction of beta-oxidation enzymes

which is possibly the major contribution to the hypolipidaemic

action exhibited by these compounds. The flux of palmitate

through mitochondrial and peroxisomal beta-oxidation pathways

was studied in isolated untreated rat hepatocytes (Kondrup and

Lazarow,1985). It was estimated that approximately 32% of

palmitate oxidation began in peroxisomes. The effect of

peroxisome proliferators on the flux of fatty acids through

the two systems would be of obvious interest.

It has been postulated that the reduction in serum triglycerides seen with peroxisome proliferators is due to the displacement of thyroxine from its serum-binding proteins by these compounds, resulting in a hyperthyroid effect in the liver (indicated by increased mitochondrial alpha-glycero- phosphate dehydrogenase activity) and a hypothyroid effect in 47

other organs (based on the lack of change in metabolic rate

and thyroid function) (Ruegamer et al.,1969). In the

hyperthyroid state, fatty acid oxidation and ketogenesis are

stimulated simultaneously and esterification of fatty acids to

triglycerides reduced as is the secretion of very low

density lipoproteins (Heimberg et al.,1985). The hepatic

mitochondria of rats fed dessicated thyroid oxidise

alpha-glycerophosphate 5 times more rapidly than mitochondria

from untreated animals (Lee et al.,1959). A 2 to 5-fold

increase in peroxisomal palmitoyl-CoA oxidation has been found

following treatment of male rats with daily doses of

L-thyroxine (Just and Hartl,1983).

Hepatic peroxisomes adapt in a structural and proliferative

manner to 3,3,5 triiodothyronine i.e. hyperthyroidism (Fringes

Reith,1982). Thyroid hormones may regulate the rate of fatty

acid oxidation in the liver by changing the activity of the

outer membrane carnitine palmitoyltransferase (Stakkestad and

Bremer,1983) The hypotriglyceridaemic effect of clofibrate and thyroxine may be via the activation of carnitine-mediated transport of fatty acids in hepatic mitochondria (Pande and

Parvin,1980). Thyroidectomy abolishes the hypolipidaemic effect of clofibrate in male rats yet peroxisome proliferation and increased catalase activity persists (Svoboda et al,1969).

The hypocholesterolaemic action of DEHP is unaffected by thyroidectomy yet elevated alpha-glycerophosphate dehydrogenase activity is largely abolished (Nair and

Kurup, 1986a). The role of thyroxine in peroxisome proliferation and the observed hypolipidaemic action has yet 48

to be clarified. One of the physiological roles attributed to

peroxisomes is beta-oxidation whereby the accumulation of

fatty acids poorly metabolised by other metabolic pathways is

prevented. These include polyunsaturated fatty acids and very

long chain fatty acids (Karki et al.,1987, Singh et al.,1987)

Additionally, trans-fatty acids are readily degraded enhanced by a channelling mechanism on the peroxisomal bifunctional enzyme polypeptide (Yang et al.,1986). The hypolipidaemic action of clofibrate appears to be dissociated from the induction of peroxisomal beta-oxidation in normolipidaemic animals (Harrison,1984).

In bezafibrate, gemfibrozil and nafenopin normolipidaemic treated rats, the serum hypolipidaemic effect is not dose-dependent and does not correlate with an increased peroxisome volume and associated increased enzymes, implying a minor role of peroxisomes in the mode of action of hypolipidaemic drugs (Stegmeier et al.,1982). Whether this is true in the hyperlip id aemic animal has yet to be ascertained.

It has been estimated that the contribution of peroxisomal fatty acid oxidation is less than 10% in control and clofibrate treated animals (Mannaerts et al.,1979). This would suggest that peroxisome proliferation in rodents is a toxic side-effect of these compounds with a minimal contribution to their pharmacological action. However, a decrease in serum lipids by various plasticisers occurs only with peroxisomal proliferation (Moody and Reddy,1982). The relationship between peroxisome proliferation and the observed hypolipidaemic effect was examined by treating animals bearing non-peroxisome 49

prolif erator-induced tumours with tibric acid or W y - 14,64 3.

Peroxisomal enzyme induction was found to be dissociated from

the decrease in serum cholesterol while an inverse biological

association existed between liver catalase and serum cholesterol (Moody et al.,1984). It was suggested that carnitine acetyltransferase was associated with serum triglycerides during the perturbation state only.

The rate of incorporation of oleate into perfusate triglycerides is reduced by clofibrate (Ide and Sugano,1983), and methyl clofenapate causes an increased palmitate utilisation through oxidation and esterification (Ghis et al.,1984). This suggests that the hypolipidaemic effect results from both increased catabolism and decreased anabolism of triglycerides. In control animals, microsomal omega-oxidation and peroxisomal beta-oxidation contribute between 5 and 10% to the total oxidation of fatty acids respectively (Bjorkhem,1978, Mannaerts et al.,1979). The major intracellular location of increased triglyceride catabolism is probably the mitochondria, reflecting the induction of beta- oxidation enzymes following treatment. The mechanisms accounting for the hypolipidaemic action have yet to be clearly defined.

1.4 MECHANISMS OF PEROXISOMAL PROLIFERATION AND INDUCTION

OF CYTOCHROME P-452

It is clear that administration of peroxisome proliferators to rodents results in an increased activity of certain 50

peroxisomal enzymes and cytochrome P-450 with respect to

lauric acid hydroxylation. In general, these changes reflect a

dual proliferation of peroxisomes and S.E.R.. Immunological

quantitation of cytochrome P-452 (Bains et al.,1985) and

PPA-80/enoyl CoA hydratase in treated animals indicate that as

with catalase, increased activity is due to an overall

enhanced synthesis of the proteins as opposed to alterations

to catalytic activity and/or protein degradation (Reddy and

Lalwani, 1983). With respect to peroxisomal enzymes it has been

established that Wy-14,643 administration resulted in

peroxisome proliferation and a marked increase in CoA-thiolase

and enoyl-CoA hydratase, as determined using immuno- precipitation studies (Chatterjee et al.,1983).

It is possible that peroxisome proliferators and/or their metabolites act as substrates for the peroxisomal beta- oxidation pathway thereby initiating enzyme induction as a result of substrate overload (Reddy and Lalwani, 1983). It has long been established that the activities of drug metabolising enzymes in liver microsomes are markedly increased when animals are treated with a variety of xenobiotics, of which many stimulate their own metabolism (Conney,1967). This enhanced activity is termed enzyme induction and represents new enzyme synthesis. It is possible that peroxisome proliferators containing an omega-carbon group may serve as substrates for cytochrome P-452 hydroxylase activity and act as inducers in a similar manner to the peroxisomal system. The limited formation of hydroxybezafibrate by a cytochrome P-450 mediated process has been demonstrated in an in vitro system 51

(Facino et al.,1981). This is consistent with a substrate linked induction mechanism. The in vitro metabolism of ME HP to omega-, (omega-1)- and (omega-2)- hydroxylation products has been demonstrated and it is believed to be mediated by the cytochrome P-450 isoenzymes associated with omega- and

(omega-1)- and (omega-2)-hydroxylation of fatty acids (Albro et al.,1984).

It has been suggested that an influx of fatty acids into the liver, as a result of drug treatment leads to a lipid excess, thereby triggering peroxisomal beta-oxidation. Fatty acids serve as substrates for the microsomal cytochrome P-450 hydroxylase system and for peroxisomal beta-oxidation, and both activities undergo substantial induction in peroxisome proliferator treated animals. A similar mechanism may account for the increased enzyme activity in both organelles. This concept has been, in part substantiated by induction of peroxisomal enzymes and peroxisomal beta-oxidation by high fat diets (Neat et al.,1980). Peroxisomal proliferation was less apparent, yet peroxisomal beta-oxidation more marked in diets containing a high amount of fatty acids which are poorly oxidised by mitochondria (Osmundsen, 1982). The presence of a haemoprotein with a CO-adduct absorption maximum at 454 nm, designated cytochrome P-454 has been identified in rat kidney cortex microsomes (Jakobsson et al.,1970). This cytochrome catalysed the omega-oxidation of fatty acids whereas metabolism of model substrates was minimal. A diet supplemented with 10% lauric acid resulted in a marked increase in cytochrome P-454 concentration, a type 1 spectral 52

change and a moderate enhancement of total omega-oxidation of

laurate (Jakobsson et al.,1970). It would appear that the

concentration of endogenous substrates such as fatty acids may

have a regulatory effect on the levels of the cytochrome P-454

omega-hydroxylase system in the kidney cortex. The influence

of a lauric acid supplemented diet on the hepatic microsomal

system was not studied, but it is tempting to speculate that a

similar response may exist.

It has been demonstrated that rats maintained on a high fat diet exhibit a significantly increased hepatic microsomal membrane phospholipid to cholesterol ratio (Narbonne et al.,1984). This effect tends to increase the microsomal membrane fluidity and is associated with an increased ability of cytochrome P-450 to hydroxylate aniline. The increased cytochrome P-450 hydroxylase activity observed in the kidney and liver following the administration of peroxisome proliferators and high fat diets may in part be attributed to the decreased viscosity and increased mobility of the microsomal membrane which may either:

1. facilitate the interaction of cytochrome P-450 and the

NADPH-cytochrome P-450 reductase (Jakobsson et al.,1970),

or,

2. favourably affect the accessibility of the respective

components to the substrate.

Induction may result from an interaction of these compounds and/or their metabolites with a receptor in the cytoplasm of hepatocytes (and renal proximal tubular epithelium) (Reddy and

Lalwani, 1983). Recently, tritiated nafenopin was shown to bind 53 to a specific saturable pool of binding sites in cytosols from

rat liver and kidney cortex (Lalwani et al.,1983). Tissue

levels of this protein correlated with the ability of

nafenopin to induce peroxisomal enzymes in these organs.

Ciprofibrate and clofibrate competitively blocked the specific

binding of nafenopin. This peroxisome-prolif erator binding

protein has been purified using nafenopin, clofibric acid and

ciprofibrate as affinity ligands and eluting agents (Lalwani

et al.,1987). It was suggested that this protein probably

plays an important role in the regulation of the induced

pleiotropic response. In this context it should be noted that

a cytosolic receptor has been identified and implicated in the

induction process of certain cytochrome P-450 species, namely the murine Ah locus with polycyclic aromatic compounds

(Whitlock,1986). The molecular events involve the passive transfer of the inducer across the cell membrane with specific binding to the cytosolic Ah receptor, the major regulatory gene product. The pleiotropic response, including the activation of numerous structural genes follows the apparent translocation of the inducer-receptor complex into the nucleus. The induction of several drug metabolising enzymes, the structural gene products, is the final outcome

(Whitlock, 1986).

It is both plausible and tempting to speculate that a cytosolic receptor binding protein is responsible for the induction of microsomal and/or peroxisomal parameters; binding occurring via a functional group or structural orientation common to all the agents producing peroxisome and S.E.R proliferation. This would by its presence or absence readily

explain tissue, sex and species differences in the

responsiveness to these agents. The existence of the peroxisome-proliferator binding receptor has not been universally accepted with the original receptor binding studies having yet to be verified in other laboratories. Our laboratory has been unable to verify the presence of this putative receptor and found no evidence of specific binding of peroxisome proliferators to hepatic cytosolic proteins or receptor material (Milton et al.,1988). Further research is necessary to identify and characterise this postulated binding protein and receptor inhibitor studies will clarify the relationship between the induction of both systems and this

"cytosolic receptor".

1.5 RESPONSE OF DIFFERENT SPECIES TO THE ADMINISTRATION

OF PEROXISOME PROLIFERATORS

The hepatic changes induced by the administration of peroxisome proliferators to rats and mice has been reviewed in depth herein. It is important to determine whether the changes induced in rodents represents a rodent specific phenomenon or a species common response. Clofibrate induced hepatomegaly has been studied in the dog and Rhesus monkey following various dose regimens (Platt and Thorp, 1966). An increased liver weight was observed in dogs and monkeys following 3 months of treatment. No increase was demonstrated in the latter group as a result of administration for a 2 year period (Platt and

Thorp,1966). Hamsters, guinea pigs and rabbits were fed 0.25% 55

(w/w) clofibrate in the diet for 14 days (Svoboda et

al.,1967). An increased peroxisome number was observed in

hamsters only, the peroxisomes exhibiting marked structural

variation. No peroxisome changes were detected in chickens fed

0.25% (w/w) clofibrate in the diet for 3 weeks. The number of

peroxisomes increased in the dog following a daily dose of

25mg/kg clofibrate for 26 days. No changes were detected in

the squirrel monkey treated with clofibrate (75mg/kg/day for

22 days) (Svoboda et al.,1967).

The administration of BR-931, fenofibrate and methyl

clofenapate to male Syrian golden hamsters resulted in a marked peroxisome proliferation with a concomitant induction

of enoyl-CoA hydratase and PPA-80 and elevation of peroxisomal

beta-oxidation (Reddy,M.K. et al.,1982). The effect of clofibrate was less marked. The influence of gemfibrozil on peroxisomes in Rhesus monkeys, dogs and hamsters of both sexes has been studied (Gray and De La lglesia,1984). In hamsters

(400mg/kg/day for 2 weeks), peroxisome proliferation with a correspondingly decreased mean volume of individual peroxisomes was observed. In dogs, the number of peroxisomes per hepatocyte was increased and the volume fraction was significantly increased in females only. Peroxisome numbers were unchanged in young monkeys (3-4 years old) with sex dependent alterations in organelle volume. Treatment of aged monkeys (6-13 years old) resulted in an increase in peroxisome number and volume in both sexes. Both groups were administered gemfibrozil at 300mg/kg/day for 3 months (Gray and De La

Iglesia,1984). It would appear that a species dependent 56

organelle response occurs with gemfibrozil treatment, the

order of susceptibility being dog < monkey < hamster < rat.

The oral administration of DEHP at 25-lOOOmg/kg/day for 14

days to male Syrian golden hamsters has also been studied

(Lake et al.,1984). Liver enlargement and peroxisome

proliferation was observed in response to this compound,

however the magnitude of response was considerably less than

that of rats at the same dose levels. A similar response was

demonstrated with the administration of ME HP and clofibrate,

indicating that a marked species difference exists (Lake et

al.,1984). Acute toxicity studies with DEHP in the rat and

marmoset (oral administration, 2000mg DE HP/kg body weight for

14 days) demonstrated that none of the hepatic changes found

in the rat were observed in the marmoset (Jackh et al.,1984).

However, bioavailability of DEHP is much lower in the marmoset

than in the rat and the marmoset has also been shown to have a

different physiological and pharmacological response to DEHP

and its metabolites (Jackh et al.,1984).

The effect of clobuzarit, a structural analogue of clofibrate,

has been examined in various species following a 14 or 15 day

treatment period (Orton et al.,1984). A limited peroxisome

proliferation was detected in hamsters with an increase in

endoplasmic reticulum fatty acid oxidase. No proliferative

effects were observed in the liver of the dog or marmoset at dose levels of 15 and 45mg/kg. In a recent study, the

administration of ciprofibrate resulted in hepatic peroxisome proliferation in cats, pigeons, chickens, Rhesus monkeys and 57

Cynomolgus monkeys (Reddy et al.,1984). The dose levels were as follows;

1) Cats; morphology (50mg/kg x 1 week, lOOmg/kg x 2 weeks),

biochemistry (lOmg/kg x 1 week, 20mg/kg x 2 weeks,

40mg/kg x 1 week),

2) Pigeons; morphology (300mg/kg x 3 weeks),

biochemistry (300mg/kg x 2 weeks),

3) Chickens; morphology (lOOmg/kg x 6 weeks),

biochemistry (50mg/kg x 4 weeks),

4) Rhesus monkeys; morphology (50mg/kg x 1 week, lOOmg/kg x 3

weeks, 200 mg/kg x 3 weeks),

biochemistry (50mg/kg x 1 week, lOOmg/kg x

3 weeks, 200 mg/kg x 3 weeks),

5) Cynomolgus monkeys; morphology (400mg/kg x 3 weeks),

biochemistry (400mg/kg x 2 weeks).

A marked but variable increase in the activities of peroxisomal catalase, carnitine acetyltransf erase, heat labile enoyl-CoA hydratase and the fatty acid beta-oxidation system was observed, indicating that peroxisome proliferation is a dose-dependent phenomenon. A significant increase in mitochondrial volume density was observed in the cat and chicken.

The effect of long term clofibrate therapy on human liver has been studied by Hanefeld et al.,(1980). With the short term use of clofibrate, an increase in E.R especially in the agranular membranes of hepatocytes of all patients was observed. Selective alterations in peroxisomes were not seen but an increase in liver cell mitochondrial inner membranes 58

was detected. With long term treatment the above changes were

more marked; structural alterations in mitochondria were

observed and a further increase in agranular and granular E.R

and peroxisomes was demonstrated. These ultrastructural

alterations observed in hepatocytes (proliferation of E.R and peroxisome populations) persisted for several months after cessation of treatment (Hanefeld et al.,1980).

The morphological effects of clofibrate in man were investigated more thoroughly in a later study (Hanefeld et al.,1983). Clofibrate treatment resulted in a significant increase in the numerical density of mitochondria and peroxisomes (38% and 50%) respectively. The volume density similarly increased by 34% and 23% for mitochondria and peroxisomes. A marked proliferation of both organelles occurred during the first months after which no further significant increases were observed. There was a significant reverse correlation between the increase in volume density of mitochondria and the decrease in triglyceride levels, whereas no relation between the proliferation of peroxisomes and hypolipidaemic action appeared to exist. The increase in peroxisomes was less pronounced in man in contrast to the marked proliferation observed in rodents and this was proportional to the increase in mitochondria (Hanefeld et al.,1983).

Fenofibrate treatment of patients for a 2 year period revealed no increase in peroxisomes nor were any other peroxisomal morphological alterations observed (Blumcke et al.,1983). 59

Mitochondria containing paracrystalline inclusions were more

common in treated subjects and dilated S.E.R associated with

reduced amounts of R.E.R appeared to be more evident in

patients receiving fenofibrate. Light and electron microscopy

of liver biopsies from fenofibrate treated and untreated

patients showed no morphological differences in peroxisome

numbers or volume densities (Gariot et al.,1983 and 1984).

Biopsies from subjects on long term gemfibrozil therapy (17-27 months) revealed peroxisome populations varying from 656 to

1452 per hepatocyte with a mean of 830 per group (De La

lglesia et al.,1981). Considerable variation in peroxisome volume was detected with some peroxisomes containing marginal plates and spurious densities. No peroxisome proliferation was noted. Gemfibrozil treatment was found to cause no observable alterations in nuclei and mitochondria whereas S.E.R membranes appeared proliferated when EM sections were examined subjectively (De La lglesia et al.,1982). No induction of peroxisomes occurred although similar ultrastructural changes were observed.

1.6 MOLECULAR APPROACH TO STUDY THE HEPATIC RESPONSES

INDUCED BY PEROXISOME PROLIFERATORS

The hepatic responses induced in rats by peroxisome proliferators have been studied extensively at the biochemical and ultrastructural levels. However, the use of recombinant

DNA technology to examine gene expression in treated animals has been limited, although this approach offers the opportunity to gain a fundamental insight into the process of 60 induction.

Recombinant DNA technology has provided a novel approach to investigating the regulation of gene expression. Any region of a cell's DNA can be excised with restriction nucleases, replicated by DNA cloning and sequenced. Similarly complimentary DNA (cDNA) copies of any messenger RNA can be isolated. Using nucleic acid hybridisation methods, mRNA molecules corresponding to the cloned DNA molecules can be detected, quantitated and characterised. It is also possible to use the specific antisera to pure proteins to isolate the gene encoding for them. Alternatively, specific DNA probes

(oligonucleotide probes) corresponding to a short length of amino acid sequence from the protein, can be synthesised which will hybridise with the mRNA and DNA encoding the protein

(Alberts et al.,1983).

Much of the current research on the molecular biology of the cytochrome P-450 system is concerned with the elucidation of the genetic, molecular and evolutionary mechanisms responsible, for the existence of a multiplicity of forms of cytochrome

P-450 and also for the induction of distinct forms by specific inducing agents (Adesnik and Atchison, 1986). Using an antibody to cytochrome P-450 LAw, thought to be identical to cytochrome

P-452 purified by Tamburini et al.(1984), the haemoprotein was immunochemically detected in non-induced rat liver and kidney.

The level of this isoenzyme was markedly elevated after clofibrate treatment due to an activation of cytochrome P-450

LAw mRNA synthesis, as determined using the corresponding cDNA 61

probe (Hardwick et al.,1987). This rapid transcriptional

increase is similar to the transcriptional activation of fatty

acyl-CoA oxidase and enoyl-CoA hydratase/3-hydroxy acyl CoA

dehydrogenase induced by clofibrate in rat liver (Reddy et

al.,1986). It appears that cytochrome P-450 LAw is regulated

by the same mechanism that regulates these peroxisomal

enzymes. A lKbp (kilo base pair) cDNA probe to cytochrome

P-452 was recently isolated in our laboratory from a

clofibrate-induced rat liver cDNA library in the bacteriophage

expression vector lambda gtll using anti-cytochrome P-452 sera

and a biotin-streptavidin linked immunoassay (Earnshaw et

al.f1988). This cDNA probe is being used in our laboratory to

examine the corresponding mRNA levels in untreated and treated

animals utilising hybridisation methodology.

It has been reported that clofibrate and fenofibrate treatment

decrease glutathione peroxidase (GPX) enzyme activity (Ciriolo

et al.,1982 and 1984). The mouse GPX-1 gene has been isolated

using recombinant DNA technology (Goldfarb et al.,1983,

Chambers et al.,1986). A genomic DNA probe (p2D6) derived from the 3' end of the gene, has very strong homology with

sequences in human and rat DNA and mRNA. This probe can therefore be used to study GPX gene expression in a variety of

species. The tissue distribution of glutathione peroxidase varies between species, and in the mouse there is good evidence that tissue specificity of gene expression is controlled at the transcriptional level (Goldfarb et al.,1985,

Affara et al.,1985, Chambers et al.,1986). The techniques of molecular biology, namely the use of

hybridisation probes, were applied in this study to

rationalise the ciprofibrate-dependent modulation of the

cytochrome P-452 and GPX-1 gene expression. Use of the cloned

DNA probes combined with immunochemical and catalytic analyses

should clarify the level at which the regulation of enzyme

changes occurs. The inter-relationships between enzyme levels

and gene expression will be delineated in susceptible and non-

susceptible species. It is hoped that this will define the

hepatic response in rodents as species common or species

specific.

1.7 AIMS OF THE PRESENT INVESTIGATION

The aim of the present study was twofold;

1. to endeavour to clarify the rodent hepatic response to

peroxisome proliferators as species specific or species

common, and;

2. to assess the validity of extrapolating animal toxicity data to predict human risk.

The effect of xenobiotic pretreatment was investigated at a biochemical and molecular level. Enzyme levels were determined in the cytosolic, mitochondrial, microsomal and peroxisomal fractions. The influence of pharmacokinetic differences in metabolism on the relative potency of peroxisome proliferators was examined in the rat (Chapter 3). The hepatic and renal 63 response was studied in five rat strains and six species following the short term administration of ciprofibrate

(Chapter 4 and 6). The effect of long term ciprofibrate treatment in the marmoset and Fischer rat was also undertaken

(Chapter 5). 64

Chapter 2

METHODOLOGY RELATING TO PREPARATION AND CHARACTERISATION

OF ENZYME CONTENT AND ACTIVITY IN SUBCELLULAR FRACTIONS

2.1 MATERIALS

Acetylacetone, ammonium acetate, citric acid, copper sulphate,

EDTA, Folin-Ciocalteau phenol reagent, gelatin, glycine, hydrogen peroxide, magnesium chloride, mercaptoethanol, potassium cyanide, potassium phosphate, sodium carbonate, sodium chloride, sodium dithionite, sodium potassium tartrate, sucrose and Triton-X-100 were purchased from British Drug

Houses Ltd. (Poole, Dorset). Emulgen 911 was obtained from

K.A.O Atlas Co. (Tokyo, Japan).

Analytical grade acetyl-CoA, bovine serum albumin, carnitine, coenzyme A, 5, 5'-dithiobis (2-nitro-ben zoic acid), dithio- threitol, flavin adenine dinucleotide, 4-hydroxyquinoline, kynuramine, lauric acid, NAD + , NADH, NADPH, nicotinamide, palmitoyl-CoA, sodium cholate, sodium alpha-glycerophosphate, sodium malonate, sodium succinate, Tris (Trizma base), Tween

20 and uric acid were purchased from the Sigma Chemical Co.

Ltd. (Poole, Dorset). Benzphetamine was obtained from the

Upjohn Co. (Kalamazoo, Michigan, USA), ciprofibrate, bezafibrate, clofibric acid and gum tragacanth from Sterling

Winthrop Research Centre (Alnwick,Northumberland), ethoxyresorufin and resorufin from Molecular Probes Inc.

(Junction City, Oregon, USA) and 14-C-lauric acid from the

Radiochemical Centre (Amersham, Bucks.). Glycerol was obtained 65

from the Fisons Scientific Equipment Co. (Loughborough

Leics.).

Nitrocellulose filters (0.45um) were purchased from Anderman

and Co. Ltd. (Kingston Upon Thames, Surrey). Control donkey ..... anti-sheep IgG enzyme lab el and orthophenylenediamine were kindly

supplied from the Guildhay unit (University of Surrey.)

horseradish peroxidase.

All other chemicals were obtained as Analar grade wherever possible.

2.2. ANIMALS

Male Wistar albino rats were obtained from the University of

Surrey breeders and male Gunn rats from the University of 344 Dundee breeders. Male Sprague Dawley, Fischer|and Long Evans rats, mice (CD-I) and marmosets were purchased from Charles

River UK (Margate, Kent). The rats, approximate body weight of

140-160g, were housed at the University of Surrey in cages with sawdust bedding, being allowed food (Spratts Animal Diet

No.l) and water ad libitum. A 12 hr light-dark cycle was in operation (0700-1900 light) at 22 degrees centigrade and 50% humidity.

Male albino (Duncan Hartley) guinea pigs were obtained from the Porcellus Animal Breeding Ltd. (Heathfield, Sussex), male half-lop rabbits from Froxfield Research Suppliers of

Laboratory Animal Breeders (Froxfield, Hants) and male golden Syrian hamsters from the National Institute of 66

Medical Research (Mill Hill, London) as young adults (post­

juvenile) .The bodyweight at the start of the experiments were

as follows ; mice (25-35g), hamsters (85-130g), guinea pigs

(450-550g) and rabbits (2750-3200g). Marmosets, mice,

hamsters, guinea pigs and rabbits were housed in the barrier

maintained animal unit at Sterling Winthrop Research Centre

(Alnwick, Northumberland) with a 12 hr light/dark cycle

(0700-1900 light) at 21 (+/-1) degrees centigrade, 50 (+/-5) %

humidity with 30 air changes per hour during the day, 15

changes per hour at night.

2.2.1 Treatment of Animals

Ciprofibrate was administered as a suspension in 0.25%(w/v)

gum tragacanth,.the vehicle. Rats, hamsters, mice, guinea pigs

and rabbits received 2 mg/kg or 20 mg/kg per day ciprofibrate

for a duration of 14 days and were sacrificed 24 hours

following the final dose. Marmosets were dosed with 20 mg/kg,

80 mg/kg or 100 mg/kg for a period of 2 weeks or 6 months. An additional group of rats received varying doses of ciprofibrate, bezafibrate, or clofibrate for a treatment period of 6 months. Bezafibrate and clofibrate were also administered as a suspension in 0.25% (w/v) gum tragacanth. In each case the drugs and vehicle were administered by oral gavage in a volume of 5ml/kg. The concentration and homogeneity of drug suspensions used in studies at Sterling

Winthrop Research Centre were assessed prior to use by Quality

Control Laboratories. 67

2.3 Autopsy Procedure

The animals were weighed prior to sacrifice and the body

weights recorded. Marmosets were anaesthetised with an i.p. 5k injection of sagatal (intravenous anaesthetic) and were bled

from the posterior vena cava, death resulting from

exsanguination. With other species a single lethal dose of

sagatal was used and no blood samples were collected. The

livers and kidneys were removed and weighed as rapidly as

possible and as required, sections for light and electron

microscopy were taken. Frozen sections were also taken from

marmoset livers for enzyme histochemistry. The remaining

tissue was maintained on ice in 0.9% (w/v) sodium chloride. 5k Pentobarbitone

2.3.1 Isolation of Hepatic and Renal Microsomes

Microsomes were prepared by ultracentrifugation using a modification of the method of Omura and Sato, (1964). The

livers were removed and perfused with 0.9% (w/v) aqueous

sodium chloride to remove contaminating haemoglobin, blotted dry and placed in ice cold 0.25M sucrose. All subsequent steps were carried out at 4 degrees centigrade. The kidneys from animals of the same group were pooled and decapsulated.

Likewise livers were pooled from hamsters and mice. The livers and kidneys were scissor minced and homogenised in 0.25M sucrose using a Potter Elvehjem glass teflon homogeniser (3 return strokes). The liver homogenate was adjusted to 25-33%

(w/v) and kidney homogenate to 15% (w/v) by the addition of

0.25M sucrose. Aliquots of homogenate were taken, with 1ml of 68

the liver samples being retained on ice for catalase and

monoamine oxidase assays. The remaining aliquots were frozen at

-80 degrees centigrade.

The homogenates were spun at 10,000rpm for 30 minutes using a 14

x 50 ml rotor in a Beckman J2-21 centrifuge at 4 degrees

centigrade, or a 12 x 30ml rotor in a Sorvall ultracentrifuge at

4 degrees centigrade. The supernatant was decanted and

centrifuged at 40,000rpm for 1 hr using a8 x 25ml rotor in a

Beckman LK65 refrigerated ultracentrifuge at 4 degrees

centigrade or a 12 x 30ml rotor in a Sorvall ultracentrifuge at 4

degrees centigrade. Aliquots of the supernatant, designated the

cytosol, were frozen at -80 degrees centigrade. The microsomal

pellet was suspended in ice-cold 50mM potassium phosphate

buffer, pH 7.25, containing 20% (v/v) glycerol using a motor

driven Potter Elvehjem glass teflon homogeniser. A final volume

of 8-10ml per liver sample (equivalent to a protein

concentration of 10-15 mg/ml) and 6-8ml per kidney sample

(corresponding to a protein concentration of 3-8 mg/ml) was

achieved.

The microsomal suspension was stored as 1ml aliquots at -80

degrees centigrade and under these conditions was shown to be

stable for a period of several months.

2.3.2 Preparation of Tissue for Electron Microscopy

The methodology employed was one routinely used in the

Toxicology Department at Sterling Winthrop Research Centre 69

(Alnwick, Northumberland). All chemicals were purchased as EM

grade from Taab Laboratories (Reading, Berks.).

After removal of the livers and kidneys, sections were cut

from the median lobe of the liver and the cortex of the

kidney. These sections were then sliced into 1mm cubes and

fixed at room temperature for 2-4 hrs in Karnowskyl fixative

(0.1M sodium phosphate bufffer, pH 7.4, containing 2.5% (v/v)

glutaraldehyde, 2% (w/v) paraformaldehyde and 0.01% (w/v)

magnesium chloride). After fixation, this solution was

decanted and replaced with 0.1M sodium phosphate buffer, pH

7.4, and the samples stored at 4 degrees centigrade prior to

processing.

The tissue blocks were counterstained in 1% (w/v) osmium

tetroxide in 0.1M sodium phosphate buffer, pH 7.4 for 1 hr,

rinsed in distilled water and stained for 30 minutes with 2%

(w/v) uranyl acetate in distilled water. The tissue blocks

were rinsed in distilled water and dehydrated by means of a

70% (v/v) ethanol wash, two 95% (v/v) ethanol washes, and

three absolute alcohol washes. Each wash allowed for a minimum

tissue contact time of 10 minutes. The tissue blocks were

subsequently rinsed in propylene oxide and immersed in a 1:1

solution of propylene oxide and Epon 812 resin for 30 minutes.

The blocks were left overnight in resin, prior to embedding in

Epon 812 resin-filled truncated capsules. The capsules were

incubated at 58 degrees centigrade for 48 hrs to allow the resin to polymerise. 70

Silver-grey sections were cut on a Reichardt Ultramicrotome

(Reichardt-Jung, Austria), counterstained with uranyl acetate

and lead citrate (Lewis and Knight, 1977) and examined on a

JEOL 100B electron microscope (JEOL, Japanese Optical Co,

Japan);(courtesy of the Microstructural Studies Unit, University of Surrey).

2.4 GENERAL ANALYTICAL PROCEDURES

2.4.1 Protein Determination

Protein concentrations of liver and kidney homogenates,

cytosolic and microsomal samples were routinely determined by

a modified method of Lowry et al., (1951) using bovine serum

albumin as the standard.

REAGENTS

2% (w/v) sodium carbonate in 0.1M sodium hydroxide.

1% (w/v) hydrated copper sulphate.

2% (w/v) sodium potassium tartrate.

A stock solution of bovine serum albumin (20 mg/ml) was

diluted in distilled water to give 0, 8, 16, 24, 32 and 40 ug/ tube in duplicate. Protein samples were diluted as necessary

and a range of 2-4 dilutions were used in duplicate assays.

With microsomal samples (which routinely contained glycerol) compensatory appropriate ) additions of glycerol were made to the standard curve. Copper sulphate, sodium potassium tartrate and sodium carbonate/hydroxide solution were mixed sequentially to give the ratio (1:1:100 by volume) immediately before use. The 71

protein standards, samples and blanks (0.5ml) were mixed with

2.5ml of the above reagent and allowed to stand for 10 minutes

at room temperature. Folin-Ciocalteu phenol reagent, diluted

1:1 with distilled water, was added (0.25ml) and the solutions

were immediately vortexed. After a thirty minute incubation

period at room temperature the absorbance was recorded at

750nm using a Cecil 292 Ultraviolet spectrophotometer. The

extent of colour development is a partial function of the

amino acid composition, due to the reaction of tyrosine

residues with the Lowry reagent.

2.5. ASSESSMENT OF ENZYME LEVELS AND ACTIVITIES IN THE

MICROSOMAL FRACTIONS

2.5.1 Determination of Cytochrome P-450 Content

A). LIVER MICROSOMES

Cytochrome P-450 was routinely measured as the reduced carbon monoxide adduct (Omura and Sato, 1964). Microsomal samples were diluted to give a protein concentration of 1-2 mg/ml with 50mM potassium phosphate buffer (pH 7.25), 20% (v/v) glycerol.

Reduction of the ferric to the ferrous form of the haemoprotein was achieved by the addition of sodium dithionite. The reduced sample was divided between two cuvettes and the baseline recorded between 400nm and 500nm using a Kontron-Uvikon 860 spectrophotometer or Varian 2200 split beam spectrophotometer. Carbon monoxide (40 bubbles at approximately 1 bubble per second) was gassed through the sample cuvette only. The spectrum was repetitively scanned 72

until there was no further absorbance development at 450nm.

The concentration of cytochrome P-450 was calculated using the

difference millimolar extinction coefficient (450nm minus

490nm) of 91.

B). KIDNEY MICROSOMES

Because of the presence of mitochondrial contamination a

different method was utilised to determine the cytochrome

P-450 content in kidney microsomes. In this instance, sodium

succinate was added to the assay solution to reduce

mitochondrial electron transport enzymes and NADH to reduce

any cytochrome b5 present.

To 0.6ml of kidney microsomal suspension, 0.6ml sodium

succinate (lOOmM) and 4.8ml potassium phosphate buffer (50mM, pH 7.25, containing 20% (v/v) glycerol) were added. The sample was divided between two cuvettes and the baseline (4 00nm to

500nm) recorded. N A D H (25 ul, 2% (w/v) solution) and a few grains of sodium dithionite were added to each cuvette. Carbon monoxide was bubbled through the sample cuvette for 40 seconds, the spectrum was rescanned and concentration measurements calculated as for liver microsomes.

2.5.2 Determination of Laurie Acid Hydroxylation

The metabolism of 14-C lauric acid to 12- and 11- hydroxylated products was evaluated using HPLC, utilising the method described by Parker and Orton, (1980). 73

The 2ml incubation volume contained 14-C lauric acid (0.1 uCi

in 10 ul methanol) and a final concentration of O.lmM 12-C

lauric acid (200mM stock solution in methanol diluted to ImM

in 0.5M Tris-HCl buffer, pH 7.4). One mg of protein from

control and kidney samples was used per assay or 0.5 mg

protein from induced microsomes. The tubes were incubated for

5 minutes at 37 degrees centigrade prior to the addition of 40 ul of 40mM NADPH to initiate the reaction. NADPH blanks were completed using distilled water in place of NADPH.

Samples from untreated animals were incubated for 10 minutes and samples from treated animals for 5 minutes. The reaction was terminated by the addition of 0.2ml 3M HC1. Incubations were carried out in polypropylene stoppered tubes.

Extraction and Analysis of 11- and 12- Hydroxylauric Acid

A total of 10 ml of diethyl ether was added to the incubation mixtures and the tubes were subsequently rotated end over end for 15 minutes. The tubes were allowed to stand for 10 minutes to clarify the phases and the upper (ether) layer

(approximately 9ml) was transferred to a test tube and evaporated to dryness under a stream of nitrogen. The test tubes were sealed and stored at -20 degrees centigrade until used.

High Performance Liquid Chromatography (HPLC) Procedure

for the Quantitation of 11- and 12- Hydroxylauric Acid

The dried ether extracts were reconstituted in 150 ul of 74

eluting solvent (water: methanol: glacial acetic acid,

45:55:0.05 by volume). A 100 ul aliquot was then injected into

a HPLC column of the following specifications (column :

ultrasphere ODS 15 x 0. 46cm). The HPLC column was calibrated

with the NADPH blanks to check that the metabolism of

lauric acid was NADPH-dependent.

Elution of the products was achieved by using a gradient of

0.1% (v/v) glacial acetic acid in distilled water : methanol

(45:55 by volume) to 35% (v/v) 0.1% (v/v) glacial acid in

water and 65% (v/v) methanol over a 30 minute period. The

percentage of methanol was increased to 100% over 5 minutes

and maintained to the end of the programme, the duration of

which was 40 minutes. A constant flow rate of 1ml/min was

used. The 11- and 12- hydroxylauric acids were successively

eluted followed by the unmetabolised lauric acid. The elution profile of the radioactive products was monitored using a

Berthold LB 503 HPLC radioactivity monitor interfaced (Rexagan

interface,ICI) with a PET 4032 microcomputer linked to a chart recorder for data analysis. Results were expressed as nmoles hydroxyproduct formed/minute/mg protein or nmoles hydroxyproduct formed/ minute/nmol cytochrome P-450.

2.5.3 Determination of Ethoxyresorufin-O-Deethylation

The O-deethylation of ethoxyresorufin was routinely determined by the method of Burke et al., (1977), using the difference in fluorescent properties of ethoxyresorufin (excitation wavelength = 456nm, emission wavelength = 560nm) and the 75

product resorufin (excitation wavelength = 510nm, emission

wavelength = 586nm).

A standard assay system contained 2ml of 0.1M Tris-HCl buffer,

pH 7.8, a final concentration of 50uM ethoxyresorufin (stock

solution in methanol) and an appropriate volume of microsomes;

25-50 ul (equivalent to 0.5-1 mg protein) of liver microsomes

or 100-150 ul (equivalent to 1-2 mg protein) of kidney sample.

The complete system was transferred to a fluorimeter cuvette

and placed in a thermostatically controlled cuvette housing of

a Perkin-Elmer MPF-3 fluorimeter, set at an excitation

wavelength of 510nm and an emission wavelength of 586nm. After

preincubation at 37 degrees centigrade for 2 minutes, a steady

baseline with time was established. The reaction was initiated

with 10 ul of 50mM NADPH and the production of resorufin with

time monitored as the increase in fluorescence at 586nm. The

fluorimeter was calibrated with multiples of 2 ul of resorufin

standard (lOuM in methanol).

2.5.4 Determination of Benzphetamine Demethvlase Activity

The hydroxylation of the benzphetamine-N-methyl group results

in the formation of an unstable amino-carb in ol intermediate which breaks down to release formaldehyde which can subsequently be determined by the method of Nash (1953).

The 1ml incubation volume contained 0.1ml microsomal suspension (equivalent to approximately 2 mg protein) and a final concentration of 1.5mM benzphetamine, 15mM magnesium 76

chloride and 50mM potassium phosphate buffer (pH 7.25). The

tubes were preincubated at 37 degrees centigrade for 5minutes

prior to the addition of 40ul of 50mM NADPH to initiate the

reaction. After 10 minutes the reaction was terminated by the

addition of 0.5ml of ice cold 12.5% (w/v) trichloroacetic

acid. Samples were mixed and allowed to stand on ice for 5

minutes following which they were centrifuged at 3500 rpm for

15 minutes in a Beckman J6B centrifuge. The supernatant (1ml)

was then removed for the determination of formaldehyde using

the Nash assay.

Nash Assay

REAGENT

Ammonium acetate (150g) was dissolved in 900ml of distilled

water. Acetylacetone (2ml) was added and completely dissolved.

The pH was adjusted to pH 6 with glacial acetic acid and the volume adjusted to 1L. The reagent was stored at 4 degrees centigrade.

METHOD

The supernatant from the trichloroacetic acid precipitation

(lml) was added to Nash reagent( lml), mixed and heated in a water bath at 58 degrees centigrade for 10 minutes in the dark. After cooling for 5 minutes, the absorbance at 412nm was recorded using a Cecil 292 Ultraviolet spectrophotometer. A standard curve was constructed using lml volumes containing 0,

10, 20, 40, 60, 100 and 140 nmol formaldehyde treated in an 77

Identical manner to the test samples.

2.5.5 Enzyme-Linked Immunosorbent Assay (ELISA)

The technique of ELISA (enzyme linked immunoabsorbent assay)

as described by Voller et al., (1978) and modified for the

quantitation of cytochrome P-452 by Sharma et al., (1988) was

utilised to quantitate antigen within specific tissues.

Microsomes were solubilised by the addition of 1 mg cholate

and 0.2 mg Emulgen 911 per mg of protein with stirring for 20

minutes on ice. The antigen was diluted as appropriate

(0.01-0.05 pmoles cytochrome P-450) with 0.1M sodium

carbonate/ bicarbonate buffer (pH 9.6) and coated onto ELISA

plates (200 ul/well) which were left overnight at 4 degrees

centigrade in a humid environment. The plates were washed with

phosphate buffered saline, pH 7.4, containing 0.1% (w/v)

gelatin and 0.05% (v/v) Tween 20 (PBS-GT). The antisera

(cytochrome P-452 antibody) and pre-immune sera were diluted

as required using PBS-GT buffer and applied to the plates in

200 ul volumes.

The plates were incubated at 37 degrees centigrade for 2 hrs

and rewashed with PBS-GT buffer. A second 2 hour incubation at

37 degrees centigrade was completed using a donkey anti-sheep * enzyme label (200 ul per well) with a subsequent PBS-GT wash.

The substrate (lOOmls of 0.025M citric acid, 0.05M sodium

hydrogen phosphate buffer, pH 5.6, containing 40 mg orthophenylenediamirie and 40 ul hydrogen peroxide) was added (150

* horseradish peroxidase 78

ul per well) and the plates incubated for 30 minutes at 37

degrees centigrade. The colour reaction was stopped by the

addition of 2.5M sulphuric acid (50 ul per well). The plates

were read at 490nm using a Dynatech plate reader. The wells

initially incubated with the pre-immune sera act as blanks. A

standard curve was constructed using purified cytochrome P-452

as the antigen.

2.6 ASSESSMENT OF ENZYME ACTIVITY IN THE MITOCHONDRIAL

FRACTION

2.6.1 Determination of Monoamine Oxidase Activity

Monoamine oxidase is localised mainly within the mitochondrial fraction and it is termed an outer mitochondrial membrane enzyme. Its major role is in the degradation of intracellular biogenic amines. The modified fluorimetric method of Krajl,

(1965) was used with kynuramine as the substrate which undergoes cyclisation to 4-hydroxyquinoline, the reaction product. The assay was carried out using fresh whole homogenate.

The 3ml incubation volume contained 0.5ml of the suitably diluted sample (100-200 dilution) and lOOug kynuramine in 50mM potassium phosphate buffer, pH 7.4. The tubes were incubated for 30 minutes at 37 degrees centigrade in a shaking water bath and the reaction was stopped by the addition of 2ml 6%

(v/v) perchloric acid. The samples were spun at 3,500 rpm for

15 minutes in a Beckman J6B centrifuge. In fluorimeter 79

cuvettes, 2ml 1M sodium hydroxide was mixed with lml of

supernatant and the fluorescence was read in a Perkin-Elmer

MPF-3 fluorimeter, set at an excitation wavelength of 315nm

and emission wavelength of 380nm.

A standard curve using 4-hydroxyquinoline (1-5 ug total) was

carried out in a similar manner.

2.6.2 DETERMINATION OF SUCCINATE DEHYDROGENASE ACTIVITY

Succinate dehydrogenase is an inner mitochondrial membrane

enzyme catalysing an intermediate step in the citric acid cycle. A modification of the method of Prospero, (1974) was used in which 2-(p-iodophenyl)-3-(p-nitrophenyl)-5-phenol tetrazolium chloride (INT) acts as the synthetic electron acceptor. On reduction, INT yields a red, water insoluble formazan which can be extracted with ethyl acetate and the absorbance measured at 490nm.

An aliquot of whole liver homogenate was diluted with 0.2M phosphate buffer (pH 7.4) to give approximately 1-2 mg protein/ml and 0.5ml of this was added to 0.25ml of the above buffer containing 1.25 mg/ml INT. The reaction was initiated with 0.25ml 0.3M sodium succinate in 0.2M phosphate buffer (pH

7.4) and incubated at 37 degrees centigrade for 10 minutes.

The reaction was linear over a period of 20 minutes with a protein concentration of 3-5 mg and to an absorbance at 490nm of 2. The tissue blanks were prepared in a similar manner except 0.25ml of 0.3M sodium malonate replaced 0.3M sodium 80

succinate.

Enzyme activity was stopped by the addition of 4ml reagent (a

mixture of 96% (v/v) ethanol, ethyl acetate and 15% (w/v)

trichloroacetic acid, 13:20:2 by volume). Following mixing,

the tubes were centrifuged at 3500 rpm for 15 minutes in a J6B

Beckman centrifuge to remove the denatured protein and the

extinction was read at 490nm in a Cecil 292 Ultraviolet

spectrophotometer. The activity was calculated as nmoles INT

reduced/minute/mg protein using the molar extinction

coefficient of 20,000.

2.6.3 Determination of alpha-Glycerophosphate Dehydrogenase

Activity

This enzyme is located in the inner mitochondrial membrane and

it functions as one of the intracellular shuttle systems for

reducing equivalents which pass into the mitochondrial matrix.

The analytical method of Lee and Lardy, (1965) was used, based

on the principle that diaphorases catalyse the reoxidation of

NADH produced by alpha-glycerophosphate when the cytochrome

chain is blocked by cyanide.

An aliquot of whole liver homogenate (10-50 ul) was mixed with

50 ul of 0.125M phosphate buffer, pH 7.5, containing 5mM potassium cyanide and 0.15M sodium alpha-glycerophosphate.

Sodium alpha-glycerophosphate was omitted in the blanks. The tubes were incubated for 10 minutes at 37 degrees centigrade.

To each sample, 90 ul of distilled water and 100 ul of a 81

freshly prepared solution of INT (4 mg/ml) and phenazine

methosulphate (2.5 mg/ml) in 50mM potassium phosphate buffer,

(pH 7.5) was added with a further incubation period of 15

minutes. The reaction was stopped with 50 ul of 10% (w/v)

trichloroacetic acid and 2.5ml of 96% (v/v) ethanol. After

centrifugation at 3,500 rpm for 15 minutes in a J6B Beckman

centrifuge to remove precipitated protein, the extinction was measured at 500nm in a Cecil 292 Ultraviolet

spectrophotometer. Results were expressed as change in optical

density/min/mg protein.

2.6.4 Determination of Carnitine Transferase Activity

Carnitine transferases for short-, medium- and long-chain acyl groups, occur in mitochondria and are involved in shuttling acyl groups across the membrane. The assays for carnitine transferases are based on the measurement of free CoA formed during the acyl transfer from acyl-CoA to added carnitine; the free CoA is determined using the sulphydryl reagent

5,5'-dithiobis (2-nitrobenzoic acid) [DNB] according to the methodology of Bieber et al.,(1972) and Bock et al.,(1980).

The liberated coloured anion from this reagent is measured spectrophotometrically at 412nm with acyl-CoA as the substrate. Total carnitine acetyltransferase activity from the peroxisomal and mitochondrial fraction was measured. Carnitine palmitoyltransferase is located exclusively in the mitochondria and is assayed using long chain palmitoyl-CoA as the acyl group. 82

The 2ml incubation volume contained 0.2ml of diluted whole

homogenate (1:5 - 1:10 dilution) and a final concentration of

O.lmM substrate (acetyl-CoA or palmitoyl-CoA) and 99mM

Tris-HCl buffer, pH 8, containing 0.25mM DNB, 2.13mM EDTA and

0.2% (v/v) Triton-X-100. The solution was mixed and

centrifuged at l,000rpm in a Beckman J6B centrifuge for 4

minutes and transferred to cuvettes and the reaction started

with 10 ul 1M carnitine. The change in absorbance was recorded

for 5 minutes at 412nm, 25 degrees centigrade in a Beckman DU7

spectrophotometer. A blank without carnitine was run for each

sample. Values were calculated using the molar extinction

coefficient of the reaction product at 412nm of 13,600.

Results were expressed as nmol CoA formed/min/mg protein.

2.7 ASSESSMENT OF ENZYME ACTIVITY IN THE PEROXISOMAL FRACTION

2.7.1 Determination of Catalase Activity

Catalase is a marker enzyme for peroxisomes and catalyses the

breakdown of hydrogen peroxide to oxygen. Catalase activity was measured in fresh liver homogenates using the method

described by Bock et al., (1980) and Chance and Maehley,

(1955).

The test incubation contained 0.1ml of diluted sample (diluted

1:4 0 in the diluent 2% (w/v) sodium chloride containing 1%

(w/v) Triton-X-100 and 0.33% (w/v) bovine serum albumin) and

1.9ml of 33mM sodium phosphate buffer, pH 7, containing 0.33%

(w/v) bovine serum albumin. The reaction was initiated with 83

lml of the substrate (30mM hydrogen peroxide in 50mM sodium

phosphate buffer, pH 7). All assays were carried out at 25

degrees centigrade at 240nm in quartz cuvettes in a Cary 219

split beam spectrophotometer. The reaction was followed for 1

minute and the change in extinction measured over the first 30

seconds of reaction.

Results were calculated as follows:

k = 2.3 / change in time X log [Al / A2]

where;

k = first-order rate constant^

Change in time = 30 seconds (that is 0.5 minutes).

Al = absorbance of reactants at time zero, and

A2 = absorbance of reactants at time 30 seconds.

Values were expressed as k (min. r1. ) / rng proteins

2.7.2 Determination of Uricase Activity

The conversion of uric acid to allantoin is catalysed by

uricase. Uric acid has an absorption peak at 292.5nm and the

rate of reaction was determined by following the change in

absorbance at this wavelength. The method of Beaufay et al.,

(1959) was used.

The substrate (30mM sodium phosphate buffer, pH 7.4, containing 7.6mg uric acid/lOOml and 0.3% (w/v) Triton-X-100) 84

and diluent (30mM sodium phosphate buffer pH 7.4, 0.3% (w/v)

Triton-X-100) were equilibrated at 37 degrees centigrade. To

test tubes containing lml of 5mM sodium phosphate buffer, pH

7.4, lml of the diluted whole homogenate sample was added.

After equilibration at 37 degrees centigrade the solutions

were transferred to cuvettes and lml of the substrate solution was added to the test cuvette and lml of the diluent to the

blank. The change in absorbance was recorded for 4 minutes at

292.5nm at 37 degrees centigrade in a Cary 219

spectrophotometer. The activity was calculated using the molar extinction coefficient of 12.2 and results were expressed as nmoles uric acid metabolised per minute per mg protein.

2.7.3 Estimation of Cyanide Insensitive Palmitoyl-CoA

Beta-Oxidation

The method of Bronfman et al.,(1979) was adopted. The oxidation of palmitoyl-CoA results in the reduction of FAD and

NAD + . In mitochondria, the latter is' reoxidised by the cytochrome chain which can be blocked by cyanide. In peroxisomes, FADH is reoxidised by molecular oxygen but the

NADH accumulates and as added NAD + penetrates peroxisomes, an assay of NADH production in the presence of cyanide serves as a measure of the peroxisomal fatty acid oxidase system.

A reaction mixture containing CoA (75uM), FAD (180uM), NAD +

(555uM), nicotinamide (141mM), DTT (4.2uM), KCN (3uM) and BSA

(0.255mg/ml) in 60mM Tris-HCl buffer (pH 8.3) was freshly prepared and stored on ice. A 2ml aliquot of the reaction 85

cocktail and 0.96ml of 60mM of Tris-HCl buffer, (pH 8.3), were

equilibrated at 37 degrees centigrade for 5 minutes. The whole

homogenate was diluted 1:1 with 60mM Tris-HCl buffer (pH 8.3)

containing 1% (w/v) Triton-X-100 and equilibrated at 37

degrees centigrade for 2 minutes and 40 ul of sample was then

added to the cocktail buffer mix. This was centrifuged at

lOOOrpm for 3 minutes in a J6B Beckman centrifuge and

transferred to a cuvette and placed in a Beckman DU7

spectrophotometer. The reaction was initiated with 20 ul of 75

uM palmitoyl-CoA and the change in absorbance at 340nm was

recorded for 10 minutes at 37 degrees centigrade. A time lag

was occasionally observed before the reaction commenced.

Results were expressed as nmol NAD+ reduced/minute/mg protein

assuming the molar absorbance coefficient of NADH at 340nm is

6220.

2.7.4 Determination of Carnitine Acetyltransferase Activity

The activity of carnitine acetyltransf erase was determined in

peroxisomes and mitochondria using whole homogenate with

acetyl-CoA as the substrate. (See section 2.6.4 for details of

the methodology). This gives a total measurement of carnitine

acetyltransf erase activity from both subcellular fractions.

2.8 ASSESSMENT OF ENZYME ACTIVITY IN THE CYTOSOLIC FRACTION

2.8.1 Determination of Glutathione Peroxidase Activity

Selenium dependent glutathione peroxidase acts as an enzymatic 86

defence against oxidative damage by catalysing the metabolism

of hydrogen peroxide and lipid peroxides, utilising

glutathione as the reducing agent. Glutathione peroxidase

activity was measured in the presence of sodium azide to

inhibit peroxisomal catalase, using the method described by

Awasthi et al (1975) and Lawrence and Burk,(1976).

The incubation volume of lml contained potassium phosphate buffer (0.1M, pH 7), NADPH (0.2mM), glutathione reductase (1 international unit), glutathione (4mM), EDTA (4mM), sodium azide (4mM) and an appropriate amount of enzyme (cytosolic fraction). The reaction mixture was incubated at 37 degrees centigrade for 10 minutes after which 10 ul t-butylhydro- peroxide was added to the test cuvette but not the blank, to initiate the reaction. The rate of reaction was measured at 37 degrees centigrade by following the decrease in absorbance at

340nm using a Cary 219 spectrophotometer. The enzyme solution was diluted using 0.1M potassium phosphate buffer, pH 7.

Results were expressed as umol NADPH oxidised / minute / mg of protein assuming the molar absorbance coefficient of NADPH at

340nm is 6220.

Hydrogen peroxide (0.25mM) may be substituted for t-butylhydroperoxide as the substrate as both give similar results.

2.8.2 Determination of Glutathione-S-Transferase Activity

Glutathione-S-transferase may be referred to as the selenium 87

independent glutathione peroxidase and this enzyme metabolises

cumene hydroperoxide, but not hydrogen peroxide, using

glutathione as the reducing agent. The selenium dependent

glutathione peroxidase catalyses the reduction of both the

above substrates using glutathione. Using the methodology as

described in section 2.8.1, glutathione S-transf erase activity

was determined as the difference between the total activity

using cumene hydroperoxide and the activity with hydrogen

peroxide.

2.9 METHODOLOGY RELATING TO THE MOLECULAR BIOLOGY OF

CYTOCHROME P-452 AND GLUTATHIONE PEROXIDASE

Genetic information in eukaryotes is stored as linear

sequences of nucleotides in deoxyribonucleic acid (DNA) which

replicates itself and ultimately directs the synthesis of all

proteins. DNA acts as the template for the synthesis of a

corresponding RNA molecule by a process known as DNA

transcription. The RNA molecules undergo subsequent modification to give messenger RNA which is complimentary to

the base sequence of DNA. This directs the synthesis of

proteins in the cytoplasm, mediated by transfer RNA and

ribosomes.

To study the expression of cytochrome P-452 and glutathione peroxidase at the molecular level, total RNA was isolated from untreated and ciprofibrate treated animals. Subsequent hybridisation- of these RNA samples to the cytochrome P-452 cDNA (complimentary DNA) probe and glutathione peroxidase 88

genomic DNA probe was carried out in order to determine the

influence of ciprofibrate on levels of mRNA coding for these

proteins.

2.9.1 Materials

Ultrapure enzyme grade agarose, caesium chloride, guanidine hydrochloride, guanidinium isothiocyanate, NACS Prepac columns and nick translation kits were obtained from Bethesda Research

Laboratories (Maryland, USA). Antifoam A emulsion, dextran sulphate, MOPS (3[N-morpholino]propanesulphonic acid ), salmon sperm DNA (Sigma type 111, sodium salt), and ethidium bromide were purchased from Sigma Chemical Company (Poole, Dorset).

Ficoll (400,000) was obtained from Pharmacia Fine Chemicals AB

(Upsalla, Sweden). Sodium N-lauroyl sarcosine (Sarkosyl),

Amberlite MB1 resin and diethylpyrocarbonate were purchased from British Drug Houses Ltd. (Poole, Dorset). (Alpha 32-P) deoxycytidine triphosphate (specific activity 3000 Ci/mmol) was obtained from Amersham International (Amersham,

Buckinghamshire).

All other chemicals were obtained as Analar grade wherever possible.

2.9.2 Isolation of DNA Probes to Cytochrome P-452 and

Glutathione Peroxidase.

1) Cytochrome P-452

A cDNA library from clofibrate-induced rat liver poly (A) + 89

mRNA was constructed in the bacteriophage expression vector

lambda gtll (Earnshaw et al.,1988). Recombinant phage were

screened for cytochrome P-452 protein expression using a

polyclonal P-452 specific antiserum and a biotin-streptavidin-

horseradish peroxidase linked immunoassay. Positive clones

were shown to contain cDNA inserts ranging from 0.9-2.2 Kbp

(kilo base pair) in length. The DNA sequence of these inserts

has been determined and the longest showed an open reading

frame coding for a 58 kd (kilo dalton) protein (Earnshaw et

al.,1988), corresponding almost exactly to the amino acid

sequence of cytochrome P-450 LAw (Hardwick et al.,1987).

A 3' cDNA insert of 1 Kbp was subsequently subcloned into

plasmid M13mpll, propagated and the insert fragment isolated

using restriction enzyme digestion, agarose gel

electrophoresis and phenol extraction. This cDNA probe was

used in hybridisation assays to study the expression of

cytochrome P-452 mRNA in treated and untreated animals from

several species.

2) Glutathione Peroxidase

Recombinants were isolated from a library of Balb-c mouse

genomic DNA fragments cloned in lambda-Ch4A by screening with cDNA derived from 13 day foetal liver cells or adult reticulocyte poly A+ RNA (Goldfarb et al.,1983). One genomic

DNA recombinant, lambda R68A, was subsequently identified as encoding the entire mouse GPX gene (Chambers et al.,1986). The main coding region of the GPX-1 gene was located in a 0.7 kb

EcoRl fragment and an adjacent Xbal-EcoRl fragment (Goldfarb 90

et al.,1985). The 0.7 kb EcoRl fragment (termed 2D6) was

subcloned into plasmid pAT153 using EcoRl sites. The plasmid

was propagated in E.coli MCW61, cut with EcoRl and the insert

fragment isolated using agarose gel electrophoresis and phenol

extraction.

This mouse genomic DNA probe (containing the 3' end of the

gene) was subsequently used in hybridisation assays to study

the expression of GPX-1 mRNA in treated and untreated animals

from several species. It has >85% homology with sequences from

the rat, cat, dog and human (Goldfarb, personal communication)

and appears to be highly conserved in other species (see

chapter 4).

2.9.3 Treatment of Animals

As described in section 2.2.

2.9.4 Extraction of Total Cellular RNA

The major technical problems in experiments involving the isolation of RNA is one of preventing RNA hydrolysis and degradation by ribonucleases. Stringent precautions were followed to minimise ribonuclease activity and to prevent any subsequent contamination.

INHIBITION OF RIBONUCLEASE ACTIVITY

The primary objective was to denature ribonucleases contained within the biological material used and to avoid contamination 91

by exogenous ribonuclease sources. Diethylpyrocarbonate (DEP)

is an effective inhibitor of ribonuclease, reacting with the

imidazole nitrogens of histidine residues and the free amino

groups, causing subsequent loss of enzyme activity. Where

possible, solutions were treated with 0.1% (w/v)

diethylpyrocarbonate for at least 12 hours. Subsequent

autoclaving converts the DEP to ethanol and carbon dioxide.

All glassware was thoroughly washed, rinsed with deionised

water and autoclaved. Alternatively, it was baked at a

temperature of 250 degrees centigrade for a period of at least

5 hours. All solutions were prepared using sterile water and

baked spatulas. Latex gloves were worn at all times in order

to avoid contact between fingers (a source of exogenous RNAase

contamination) and glassware and solutions. The strong

chaotropic agent guanidinium isothiocyanate was also used in

the extraction procedure to further prevent degradation by

RNAases.

ISOLATION OF TOTAL CELLULAR RNA

The method used was a modification of the procedures of

Chirgwin et al.,(1979).

One gram of liver, fresh or frozen, was homogenised in 16ml of

4M guanidinium isothiocyanate solution, pH 7, containing 0.5%

sodium N-lauroyl sarcosine, 25mM sodium acetate, 0.1M

2-mercaptoethanol and 0.1% antifoam A. When using frozen

homogenate (25-30%) a ratio of 2ml to 14ml tissue to

guanidinium isothiocyanate solution was employed. The tissue was solubilised in this medium using an ultraturrax; for 30-60 sec 92

at maximum speed. The homogenate was subsequently centrifuged

for 10 minutes at 9000 rpm at 10 degrees centigrade using a 14

x 15ml rotor in a Beckman J-21 centrifuge. A total of 3.5ml

5.7M caesium chloride solution, pH 7.4, containing lOmM EDTA,

was added to quickseal polyallomer centrifuge tubes. The

homogenate was gently layered onto the caesium chloride

cushion. The tubes were sealed and centrifuged overnight using

a 65Ti rotor (8 x 13ml) in a Beckman ultracentrifuge at 36,000

rpm at 20 degrees centigrade.

After centrifugation the liquid was removed leaving a pellet

at the base of the tube. Using sterile cotton wool swabs the

sides of the tubes were cleaned to remove traces of

supernatent. The pellet was resuspended in a total volume of

5ml of 7.5M guanidine hydrochloride solution, pH 4.5,

containing 20mM potassium acetate, lOmM dithiothreitol and

transferred to a sterile 10ml Falcon tube. Dissolution of the

RNA pellets was facilitated by vigorous mixing with brief

heating in a 68 degree centigrade water bath. To this 0.25ml

of 0.5M EDTA, 0.25ml of 2M potassium acetate (pH 5) and 2.5ml

of ice-cold ethanol (stored at -20 degrees centigrade) were

added and the samples were left overnight at -20 degrees

centigrade to allow precipitation of the RNA. The tubes were

subsequently spun at 9,000 rpm, 10 degrees centigrade for 10 minutes. The liquid was removed and the tubes dried with sterile swabs. The RNA pellets were dissolved in 2ml sterile water and centrifuged at 9,000 rpm at 10 degrees centigrade for 10 minutes. The aqueous phase was transferred to a sterile

Falcon tube and the pellet resuspended in 2ml sterile water 93

prior to spinning at 9,000 rpm, 10 degrees centigrade for 10

minutes. The aqueous phases were combined and 0.4ml 3M sodium

acetate (pH 4.8) and 8ml ice cold ethanol (-20 degrees

centigrade) were added. The samples were left overnight at -20

degrees centigrade and were then centrifuged at 10,000 rpm, 10

degrees centigrade for 20 minutes. The liquid was removed and

all traces of ethanol wiped from the tubes. The pellets were

dissolved in 0.5ml sterile water and stored at -20 degrees

centigrade.

2.9.5 Determination of RNA Purity and Concentration

RNA aliquots were diluted 1:100 with sterile water and

transferred to 1ml quartz cuvettes. The samples were scanned

between 200 and 320nm. RNA absorbs at 260nm and protein at

280nm. A 260:280 ratio of 2.0 is indicative of pure RNA, and a

ratio of less than 2.0 assumes protein contamination. The

concentration of RNA can be determined using the absorbance at

260nm where 1 O.D unit is equal to 40ug/ml (Maniatis et

al.,1982).

2.9.6 RNA Agarose Gel Electrophoresis

A Biorad horizontal mini-gel system was used with a gel former

giving gel dimensions of 10cm x 7cm. The electrophoresis tank

and perspex gel former were cleaned with absolute alcohol and tape was used to seal the gel former ends. The method of

Lehrach et al.,(1977) was adapted and used. 94

A total of 1.05g agarose was dissolved in a total of 50.4ml sterile water, mixed with 12.6ml 37% (v/v) formaldehyde and

7ml lOx running buffer (60ml 2M MOPS, pH 7, 2.46g sodium acetate and 1.2ml 0.5M EDTA). The mini-gel was poured and an 8 track comb inserted. When the agarose had set, the comb and tape were removed and the gel slab placed in the electrophoresis tank filled with lx running buffer. The gel was immersed to a depth of approximately 1cm and the electrophoresis tank was then placed on ice.

The RNA samples were prepared as follows: 20 ug RNA was mixed with 2 ul lOx running buffer, 3.5 ul formaldehyde and 10 ul deionized formamide (2g Amberlite MB1 resin was mixed with

50ml formamide for 1 hr at 4 degrees centigrade and the resin was subsequently removed by filtration). The samples were vortexed and briefly centrifuged prior to incubation at 65 degrees centigrade for 15 minutes. The samples were cooled on ice and mixed with 2 ul of loading buffer (50% glycerol, ImM

EDTA, 0.4% bromophenol blue and 0.4% xylene cyanol) and centrifuged. Using a Gilson pipette, 20 ul of sample was loaded per well. The gel was run at 150mA (constant current) until the bromophenol blue had run three quarters the length of the gel. The gel pattern was visualised by staining with ethidium bromide (50 ul of 20 mg/ml stock solution in 150ml of sterile water) for 1 hr before viewing under ultra violet illumination. Gels were subsequently photographed. 95

2.9.7 Dot Blots with Nitrocellulose

Dot blotting was carried out routinely using an adaptation to

the method of White and Bancroft,(1982). Total RNA was

prepared in sterile water to give stock solutions of 0.01

ug/ul and 0.2 ug/ul. The samples were diluted to give 0.01,

0.1, 1, 10, and 20 ug/lOOul water and then 150 ul of 6.15M

formaldehyde and 20x SSC (3M sodium chloride and 0.3M sodium

citrate, pH 8) were added. The samples were incubated at 65

degrees centigrade for 15 minutes.

The nitrocellulose filter was soaked twice in lOx SSC, sterile

water for 5 minutes and finally in lOx SSC for 5 minutes. The

dot blot apparatus (BRL) was cleaned with acetone and ethanol

and assembled with the filter according to the manufacturers

instructions. The RNA samples were applied to the wells,

loading being completed under a slow vacuum. The wells were

subsequently washed with 400 ul lOx SSC, and when completed

the filter was removed and dried between Whatman 3MM paper

prior to baking at 80 degrees centigrade for 2-3 hr. The baked

filters were stored at room temperature between sheets of

Whatman 3MM paper, covered in foil prior to hybridisation.

2.9.8 Nick Translation

The process of "nick translation" utilises DNAase 1 to create

single strand nicks in double stranded DNA. The 5' to 3' exonuclease and 5' to 3' polymerase properties of E.Coli

DNAase polymerase 1 are used to remove stretches of single 96

stranded DNA starting at the nicks and to replace them with

new strands made with the incorporation of labelled

deoxyribonucleotides (Rigby et al.,1977). As a result, each

nick moves along the DNA strand being repaired in a 5' to 3'

direction (nick translation) (Kelly et al.,1970).

The nick translation kit was obtained from BRL and the

accompanying instructions followed: -

600 ng rat cytochrome P-452 DNA, or,

400 ng mouse glutathione peroxidase DNA,

5 ul dATP, dGTP, dTTP,

100 uCi 32P dCTP;

and sterile water to give a total volume of 40 ul.

This was vortexed briefly and DNAase polymerase 1 (5 ul) was

added. The solution was mixed and centrifuged rapidly prior to

incubating for 2 hr at 15 degrees centigrade. The reaction was

terminated by the addition of 0.3M EDTA (5 ul) to the solution which was stored on ice. A further 200 ul of TNE buffer (lOmM

Tris, lOOmM sodium chloride and ImM EDTA, pH 8) and 5 ul of

salmon sperm DNA (2 mg/ml) were added.

2.9.9 Removal of Unincorporated Alpha 32-P dCTP

NACS Prepac cartridges (1cm x 0.5cm) contain NACS 52, a derivative of hydroxylapatite and these were used to remove unincorporated alpha-32-P nucleotides (Maniatis et al.,1982).

Under low salt concentrations DNA is bound to the column by the phosphates whereas nucleotides do not bind. In conditions of high salt concentration, DNA is eluted. 97

The column was attached to a 2ml syringe and hydrated with 5 x

lml volumes of 2M NaCl in TE buffer (lOmM Tris, ImM EDTA, pH

8) and equilibrated with 10 x lml washes of 0.5M NaCl in TE buffer. The column was secured in a retort stand and the nick translation volume adjusted to lml with 0.5M NaCl/TE and loaded onto the column via the syringe barrel. The column was washed with 15 x lml volumes of 0.5M NaC'1/TE buffer collected into eppindorfs and eluted with 6 x 0.1ml aliquots of 2M NaCl in TE buffer.

2.9.10 Filter Hybridisation with Radioactive Probes

The application of nucleic acid hybridisation has been studied extensively (Hames and Higgins,1985) allowing for modification to optimise conditions. The baked filters were soaked in 5 x

SSC before being transferred to 250ml prehybridisation buffer, warmed to 42 degrees centigrade in a plastic container.

Prehybridisation was carried out at 42 degrees centigrade in a shaking water bath for 7 hours. The buffer was removed and replaced with hybridisation buffer containing the radioactive probe which had previously been boiled for 5 minutes to denature the double stranded DNA. The filters were hybridised for a minimum of 15 hours at 42 degrees centigrade in a shaking water bath. 98

Prehvbridisation buffer

Deionised formamide 250ml

50 x Denhardts solution 50ml

(2% w/v Ficoll, 2% w/v BSA and 2% w/v polyvinyl-pyrrolidine)

20 x SSC 125ml

Boiled salmon sperm DNA (10 mg/ml) 12.5ml

Sterile water 62.5ml

Poly A (20 mg/ml stock soln.) 0.25ml

Poly C (20 mg/ml stock soln.) 0.25ml

Hybridisation buffer:-

Deionised formamide 250ml

50 x Denhardts solution 10ml

20 x SSC 125ml

50% Dextran sulphate 100ml

Boiled salmon sperm DNA (10 mg/ml) 5ml

Poly A (20 mg/ml stock soln.) 0.2 5ml

Poly C (20 mg/ml stock soln.) 0.25ml

2.9.11 Washing of Filters Following Hybridisation

Filters were washed after hybridisation to remove non-specifically bound probe (Hames and Higgins,1985, Maniatis et al.,1982). The hybridisation buffer was removed and replaced with 2 x SSC, 0.1%(w/v) SDS for 3 consecutive washes at room temperature for 30 minutes. There were 2 successive washes in 0.5 x SSC, 0.1% (w/v) SDS for 45 minutes at 55 99 degrees centigrade. The final wash was completed at room temperature in 5 x SSC for 30 minutes. The filters were dried between sheets of Whatman 3MM paper at 37 degrees centigrade

for 30 minutes and then placed in an X-ray cassette with an intensifying screen with fast X-ray film (Kodak GBX2) for 4 days at -70 degrees centigrade. The X-ray films were developed and qualitative assessment of the RNA dot blots made by visual inspection. 100

Chapter 3

HEPATIC INDUCTION POTENCY OF THREE HYPOLIPIDAEMIC DRUGS

3.1 INTRODUCTION

In this chapter the relative potencies of 3 hypolipidaemic

drugs (ciprofibrate, bezafibrate and clofibric acid) were

examined with reference to changes in hepatic enzyme

activities.

Clofibrate was first marketed in 1962 and has been accepted

worldwide as an effective lip id-lowering drug. The long term

use and knowledge of the therapeutic efficacy of clofibrate

coupled with an increasing understanding of lipoprotein metabolism has made this drug a reference compound for comparison with other antihyperlipidaemic agents (Cayen,

1985). Gemfibrozil, a derivative of clofibrate, causes a marked increase in high density lipoprotein (HDL) cholesterol and persistent reductions in serum levels of total, low density lipoprotein and non-HDL cholesterol and triglycerides

(Frick et al.,1987). Modification of lipoprotein levels with gemfibrozil reduces the incidence of coronary heart disease in men with dyslipidaemia (Frick et al.,1987). Ciprofibrate,

2-[-4-(2,2-dichlorocyclopropyl)phenoxyl]-2-methyl-propanoic acid, is a newly developed potent hypolipidaemic congener of clofibrate which is currently undergoing evaluation in clinical trials (Arnold et al.,1979, Edelson et al.,1979).

Bezafibrate is a structural analogue of clofibrate and was introduced to the German market in 1978 following extensive 101

preclinical and clinical evaluation (Cayen, 1985). At present

bezafibrate sales dominate the fibrate market in Germany,

Italy, Spain and the United Kingdom. Clofibrate is second in

the United Kingdom, United States and Spain. As yet,

ciprofibrate is marketed in France only where it holds second

place to fenofibrate (Sterling Winthrop, personal

commu nication).

An adequate risk/benefit assessment for ciprofibrate must

include the development of an appropriate perspective on the

toxicity profile of this compound in relation to other

phenoxyisobutyrate derivatives. The manifestation of toxicity

of peroxisome proliferators is dependent upon their intrinsic

biological activity (i.e. lipid lowering potency) and is

modified by pharmacokinetic characteristics and the

sensitivity of the species/strain to the effect. These are

major influencing factors and must be considered in terms of

study design and data interpretation.

Of the three peroxisome prolif erators investigated,

ciprofibrate is the most potent with respect to hypolipidaemic

action and it exhibits a prolonged duration of effect in the

rat. This is a reflection of the plasma half life which is 3-4

days, as compared to 6-16 and 5-8 hours for bezafibrate and

clofibrate/clofibric acid respectively (Cayen et al.,197 7,

Davison et al.,1975, Schmidt et al.,1980, Sterling Winthrop,

personal communication and Thorp, 1972). The ultimate consequence of this slow elimination of ciprofibrate is prolonged biological activity and therefore sustained 102

biochemical effects in the liver. This clearly indicates that

ciprofibrate has an enhanced toxicity as compared to other

peroxisome prolif erators. To provide an insight into the

relative potencies (with reference to toxicity) of different

compounds it is necessary to conduct studies where

dose-regimes have been modified to compensate for the marked

differences in the pharmacokinetic characteristics of these

peroxisome prolif erators.

Two studies were designed to investigate the hepatic effects

in the rat resulting from chronic administration. In clinical

practice the recommended frequency of dosing varies for these

agents [ciprofibrate is required once daily whereas

bezafibrate and clofibrate are recommended two to three times

daily] (Cayen,1985, Olsson and Lang,1978 and Rouffy et

al.,1985). In study 1, the dose levels chosen were based on

comparative lipid lowering potency, derived from published

data (Sterling Winthrop, personal communication).

Administration was once daily for a duration of 6 months.

In the second study, the dosing regimens for each agent were adjusted to compensate for their different pharmacokinetic profiles in the rat i.e ciprofibrate once every 2 days, bezafibrate and clofibric acid twice a day. The treatment period was 6 months and twice daily administration was separated by 8 to 12 hours. In the second study, the dose levels for bezafibrate and clofibric acid were increased as a result of the widely differing potencies of these compounds exhibited when administered once daily (i.e as in study 1). 103

The objectives of these two studies, completed consecutively,

were to compare the chronic effects of three hypolipidaemic

drugs on hepatic mitochondrial, peroxisomal and cytosolic

enzyme parameters. In addition, the importance of using a

pharmacokinetic basis to choose dose regimens - with respect

to relative potencies of compounds - was examined.

3.2 Hepatomegaly

Hepatomegaly is expressed as an increased liver/body weight

ratio and appears to be an adaptive response to the

administration of peroxisome prolif erators. In general, the

increased liver weight is due to hyperplasia and hypertrophy.

There was a significant increase in the liver/body weight

ratio in treated animals (table 3.1). With once daily dosing,

a maximal 2-fold induction was seen with ciprofibrate with a

minimal increase in 'the clofibric acid treated group. In the

second study (differential dosing frequencies) the extent of

hepatomegaly with all 3 compounds was very similar. In the

ciprofibrate treated animals the increase in liver/body weight

ratio was unchanged by reduced dosing frequency whereas liver

enlargement in the bezafibrate and clofibric acid treatment

groups was greater with twice daily administration and

increased dose levels.

It has been shown with a 3-day treatment period that hepatomegaly in the Wistar rat was greater in bezafibrate treated animals than with clofibric acid even with dose levels of 200 and 250 mg/kg/day for bezafibrate and clofibric acid, 104

Table 3.1.

Comparison of Hypolipidaemic Drugs under Varying Dosing

Regimens on Liver Size in the Fischer Rat following Chronic . . # Administration.

Study Dose Level Liver/Body weight Number Ratio(%)

1[a] Control 3.10 +/- 0.24[b] (100)[c]

Ciprofibrate 7.19 +/- 0.31[e] (232) (20 mg/kg)

Clofibric acid 3.84 +/- 0.13[d] (124) (50 mg/kg)

Bezafibrate 5.06 +/- 0.26[e] (163) (100 mg/kg)

2[f ] Control 3.27 +/- 0.18 (100)

Ciprofibrate 7.08 +/- 0.33[e] (216) (20 mg/kg)

2[g] Control 2.87 +/- 0.19 (100)

Clofibric acid 5.42 +/- 0.52[e] (189) (75 mg/kg)

Bezafibrate 6.27 +/- 0.42[e] (218) (150 mg/kg)

Jfc 26 week administration [a] Dosing once daily.

[b] Values are means +/- S.D.

[c] Figures in brackets are values expressed as percentage of Corresponding control.

P values for results significantly different (Students T-test) from control data are [d] p<0.01 and [e] p<0.001.

[f] Dosing once every 48 hours.

[g] Dosing twice daily. 105

respectively (Sharma et al.,1988). With the dietary

administration of these two compounds at a concentration of

0.4% (w/w) clofibrate and 0.2% (w/w) bezafibrate, a similar

observation has been reported (Bains et al.,1985). The short

term administration of ciprofibrate resulted in a 2 to

2.3-fold increase in liver weight as reported by Lalwani et

al.,( 1983a,b). Following clofibrate and fenofibrate treatment

for 26 weeks, a 1.5 to 2-fold increase in liver weight was

observed (Price et al.,1986). With DEHP administration for a

period of 3 days to 9 months, hepatomegaly was minimal but j sustained (Mitchell et al.,1985). It would appear that

hepatomegaly is a response characteristic of peroxisome

prolif erators, associated with both acute and chronic

administration.

3.3 THE MICROSOMAL DRUG METABOLISING ENZYME SYSTEM

Hepatic microsomes were prepared from animals pretreated with

ciprofibrate, bezafibrate and clofibric acid (once daily

administration) and enzymes of the microsomal drug

metabolising system were assayed (results presented in chapter

5). In the second study, tissue samples were frozen at -20

degrees centigrade for 12 weeks prior to homogenisation and

preparation of the microsomal and cytosolic fractions.

Mitochondrial, peroxisomal and cytosolic enzymes displayed

stable activities as evidenced by comparison of activities of

the control groups between the two studies. In the microsomes, considerable degradation of cytochrome P-450 to cytochrome

P-420 had occurred and the specific content of the 106

haemoprotein could not be determined accurately. No further

microsomal enzyme assays were completed.

3.4 MITOCHONDRIAL ENZYME ACTIVITIES

Administration of peroxisome prolif erators to rats results in

several observable mitochondrial changes which may include

proliferation, structural changes and induction of certain

enzymes. In this investigation the effect of ciprofibrate,

bezafibrate and clofibric acid on the mitochondrial marker

enzymes succinate dehydrogenase and alpha-glycerophosphate dehydrogenase (both inner membrane enzymes) has been studied.

3.4.1 Succinate Dehydrogenase Activity

With once daily administration succinate dehydrogenase activity was significantly elevated in the ciprofibrate treated animals (table 3.2). When the dosing regimen was altered with respect to pharmacokinetic profiles, enzyme activity was decreased in the ciprofibrate and clofibric acid treatment groups. The cause of this differential effect of ciprofibrate on succinate dehydrogenase activity is unclear at present. In both studies Fischer rats were used, however the starting weight was different - 250 and 150 grammes - prior to a 7 day acclimatisation period for studies 1 and 2 respectively.

It has been reported that the short term administration (14 to

21 days) of MEHP and DEHP to rats results in a significantly 0 _ k , 0 ■—< t _^ 44 O o o o o o O O O 0 44 O in o o o o O O o X *H tH CN ro CN tH in tH LO 04 > — >— «■—» • 0 «4 O 44 r— 1 rH 44 0i 44 CP 44 CP X U UJ UJ 0 •H o tH rH rH tH tH 1—1 r H tH tH o O O O o O o o M U 0 O • • 03 0 0 • • • • • G # U 0 o o O O O o o o O G >1 c X O rH 0 1 1 1 1 1 1 I I I u •H O CP \ \ \ \ s \ \ \ s o 44 1 O 4- 4* 4 - 4- 4- 4 - 4- + 4- 44 0 0 >4 •H H X 03 CN m N 1 O O o s 44 a > 1 O o O O O o 0 rH X • - • • • • * G •H < 0 o o o O O o O G Q •fH 0 g o G 03 u 0 G 4 4 g g •H G O •H u •rH 0 O H cp U <44 0 G 03 X O 0 O 05 .—. .— ■ • > CP — H 0 — — U o oo m ro O -— . o -— > — Qi G 4 4 CP x 0 c tH •H 0 G u G *h o 04 o o o o ro •H 0 0 rH i—I t—i rH tH oo tH VO l UJ 00 U 44 c rH x cp ro VO in VO cn CN in •H rH 0 g ro tH o CN ro in oo o 0 M G • • • » • • • O >1 O Q \ X 0 3 U u Ed G tH tH tH t—I CN tH rH •■a ■P X g i 1 i I 1 1 1 0 0 \ \ \ \ \ \ \ \ O CO u 05 G r—l 4- 4- 4- 4- 4- 4- 4- cn •H O H a? 0 0 44 03 W O g c- vo ID tH vo 00 vo r - VO N 1 tH N 1 CO cn cn ro CU 0 G 03 G 0 U C • • •• • • 0 0 D 45 3 • G U CO ro •^r «iv U E* rH •H X X 0 0 X r—i g G c x 0 M 03 03 •rH -r4 G 0 0 •—1 0 r—. •H 0 u 0 <13 u , 0 U 03 3 •H N •H 0 X — < — 0 tr» 4 4 — < — 0 CP rH IU • •H rH 0 O' cn 4-J 4* 0 01 CP 44 4»C 0 in a ••H o • H -P 0 M J* U (8 \ >4 M U JC 0 \ 0i 0 0 • c • V4 rH •H 43 \ -H X \ •H \ H CP G > > \ M O’ c p o 3 O 0 rH •H cp m O i X g H iji rH U C P X g 0 Q > X g -H • •H •H X O <44 g JQ g O 4 4 £ O X 0 o a V X -P H O •H 44 O U O M 44 o U CO >4 0 4-> >4 O H-t O 0 O -P >4 O 4 4 4 4 m 0 m 0 35 U 0 00 0 C QjCN O in n tH C CU CN G o > N tH C 0 1 0 < • 44 o o •H ^ rH W 1 0 0 Q u o u m O u U u m 1 0 3 rH g 0 03 >i l- 0 rH 0 XL 0 0 •H G Sh +-• •H H 0 0 0 N 0 C U 0 0 U 0 > 03 0 G 0 0 0 0 0 H 0 •H W 1 C 0 0 3 u i *“ >4 g X >4 0 0 E 04 0 OX 0 U 0 O4 03 X U 0 G X 0 U •*H 0 0 G V4 •H 03 c > g O X 0 0 0 o XL 0 0 0 rH u V4 0 0 X CP 0 0 3 o 01 CP X 0 rH c 0 M rH u c G •H 3 X •H •H CN g <: 3 3 0 rH • 3 0 0 CP > c 0 0 ro CO 0 o 0 •H o O o CM 05 Q > fxi Q j u Q Q 0 rH 03 3 JG •rH X LJ A -h 0 X # 0 X U 03 CO CN (N 108

decreased succinate dehydrogenase activity (Lake et al., 1975,

Lake at al.,1984). A non significant decrease in enzyme

activity has also been observed in DEHP treated animals (Mann

et al.,1985). The observed decrease in succinate dehydrogenase

activity in ciprofibrate and clofibric acid treated animals

(study 2) would appear to be an effect associated with chronic

administration. Following a 14 day treatment with ciprofibrate

in 5 rat strains, the specific activity of this enzyme was

unchanged (chapter 4).

3.4.2 Alpha-Glycerophosphate Dehydrogenase Activity

Alpha-glycerophosphate dehydrogenase activity was induced 2 to

3-fold in rats treated once daily with all three

hypolipidaemic agents (table 3.2). There were no clear

differences between the compounds with respect to enzyme

activity. A similar response was seen with different dosing

frequencies, although the increase was in the order of 4 to

5-fold. This induction in the bezafibrate and clofibric acid treated animals would appear to be the result of increased dose levels and frequency of dosing. The cause of the unequal increases observed in the ciprofibrate treated groups is unclear.

It has been reported that alpha-glycerophosphate dehydrogenase activity was elevated 3 to 4-fold following a 14 day administration period with clofibrate and tibric acid

(Holloway and Orton,1979). With DEHP treatment for 3,10 and 21 days increased enzyme activity was observed at each time point 109

(Mann et al.,1985). Induction of alpha-glycerophosphate

dehydrogenase activity in DEHP treated animals was dose

dependent and increased with the duration of administration

[3, 7 14 and 28 days and 9 months] (Mitchell et al.,1985).

Induction of this enzyme would appear to be a common response

in rats exposed to a variety of peroxisome proliferators and

it is sustained throughout acute and chronic treatment.

3.5 MITOCHONDRIAL AND PEROXISOMAL CARNITINE ACETYLTRANS FERASE

ACTIVITY

Mitochondria and peroxisomes possess carnitine transferase

activity directed towards short-chain acyl substrates. A

measurement of total carnitine acetyltransf erase activity was

determined, that is a combined activity from both subcellular

fractions.

In the chronically treated rat there was a marked induction of

carnitine acetyltransferase activity which was significant with each hypolipidaemic drug in both studies (table 3.3).

With once daily dosing, the increase in activity in ciprofibrate and bezafibrate treated animals was similar (60 to 65-fold). In the clofibric acid treatment group, the extent of induction was approximately 16-fold. In study 2, where dosing frequencies were altered, the increase in enzyme activity with ciprofibrate and bezafibrate was similar, as in the first study, however the induction was less marked

(13-fold). The clofibric acid treated animals exhibited higher Table 3.3.

Comparison of Hypolipidaemic Drugs under Varying Dosing

Regimens on Mitochondrial and Peroxisomal Carnitine

Acetyltransf erase Activity in the Rat following Chronic

Administration.

Study Dose Level Carnitine Acetyltransferase Number Activity (nmol/min/mg protein)

l[a] Control 1.53 +/- 0.55[bJ (100) [c]

C ip r of ib r at e 98.11 +/- 8.49[e] (6412) (20 mg/kg)

Clofibric acid 24.40 +/- 6.23[d] (1595) (50 mg/kg)

Bezafibrate 100.73 +/-12.56[e] (6584) (100 mg/kg)

2[f ] Control 2.14 +/- 0.35 (100)

Ciprofibrate 28.84 + /-11.77[d] (1348) (20 mg/kg)

2[gJ Control 2.15 +/- 0.24 (100) Clofibric acid 52.06 +/- 9.35[e] (2421) (75 mg/kg)

Bezafibrate 29.16 +/- 7.17[e] (1356) (150 mg/kg)

# 26 week administration [a] Dosing once daily.

[b] Values are means +/-S .D.

[c] Figures in brackets are values expressed as % of corresponding control.

P values for results significantly different (Students T-test' from control data are [d] p<0.01 and [e] p<0.001.

[f] Dosing once every 48 hours.

[g] Dosing twice daily. Ill

activity (24-fold) in the second study as opposed to the first

study, reflecting the increase in dose level and dosing

frequency.

It would appear that both ciprofibrate and bezafibrate have

equivalent inductive effects on carnitine acetyltransferase

activity at the dose levels and with the dosing regimens used

in these two studies. In the first study with once daily

dosing, the difference in enzyme induction between clofibric

acid and bezafibrate was marked. However by altering the

dosing frequency in study 2, the effects were similar. This

increased enzyme induction with clofibric acid in study 2

indicates that this biochemical end point is closely related

to the duration of pharmacological activity reflecting the

clearance characteristics of clofibric acid in the rat. The

cause of the difference in the level of induction between the

2 studies with bezafibrate and ciprofibrate is not known. The

lower induction of carnitine acetyltransferase with

ciprofibrate in the second study may reflect less frequent

dosing.

It has previously been reported that carnitine acetyl­

transferase activity was increased 58-fold in Fischer rats

following the dietary administration of ciprofibrate (Lalwani

et al.,1983b). This is comparable to the increase in enzyme

activity observed with once daily administration of ciprofibrate in the current study. With clofibrate and bezafibrate administration (250 and 200 mg/kg respectively) to

Wistar rats for 3 days, the increase in enzyme activity was 20 112 to 25-fold (Sharma et al.,1988) as opposed to 13 to 24-fold at

150 and 75 mg/kg twice daily for bezafibrate and clofibric acid respectively. Clofibrate (500 mg/kg/day) and DEHP administration for 14 days to rats resulted in a reported 35- and 26-fold induction in carnitine acetyltransferase activity

(Lake et al.,1984b). At the same dose level - 500 mg/kg clofibrate per day - the increase in enzyme activity was found to be in the order of 18-fold (Lake et al.,1984). Induction of carnitine acetyltransferase activity is characteristic in rats exposed to a wide range of peroxisome prolif erators and is apparent following short and long term treatment. The increase in enzyme activity is marked but variable.

3.6 PEROXISOMAL ENZYME ACTIVITIES

The administration of peroxisome prolif erators to rats results in the characteristic proliferation of peroxisomes with the concomitant induction of enzymes associated with peroxisomal beta-oxidation. Catalase, the peroxisomal marker enzyme, metabolises hydrogen peroxide catalatically or peroxidatically preventing the accumulation of this potentially toxic oxygen species. Uricase, an oxidative enzyme exclusive to peroxisomes, catalyses the conversion of uric acid to allantoin with the formation of hydrogen peroxide. Peroxisomal beta-oxidation is cyanide-insensitive in vitro and the reduction of fatty acyl-CoA by fatty acyl-CoA oxidase is rate limiting and generates hydrogen peroxide. The subsequent 3 steps are the same as mitochondrial bet a-oxidation. 113

3.6.1 Cat alas e

The catalase assay was completed on homogenised fresh tissue

as freezing causes dissociation of the enzymes sub-units

leading to inaccurate measurement of enzyme activity. Catalase

activities were determined for animals in study 1 and these

are, in part presented in chapter 5. In the second study,

tissue was frozen at -20 degrees centigrade prior to

homogenisation and therefore no catalase assays were

completed.

3.6.2. Peroxisomal Beta-Qxidation

In the chronically treated rat there was a marked, significant increase in peroxisomal beta-oxidation (figure 3.1). With once daily dosing, induction was 34-, 10- and 5-fold for ciprofibrate, bezafibrate and clofibric acid, respectively. By altering dosing frequency and dose levels for bezafibrate and clofibric acid, the increase in beta-oxidation for the three compounds was by comparison, similar. The increase in peroxisomal beta-oxidation was 28-, 30- and 29-fold for ciprofibrate, bezafibrate and clofibric acid. The induction with ciprofibrate in the second study was less marked than with once daily dosing indicating that the dosing frequency of this compound influences the induction of peroxisomal beta- oxidation. It is clear that altering the frequency of dosing of these compounds with respect to clearance characteristics, fundamentally influences the induction of peroxisomal beta-oxidation. 114

Figure 3.1.

Comparison of Hypolipidaemic Drugs under Varying Dosing

Regimens on Peroxisomal Beta-Qxidation in the Fischer Rat

Following Chronic Administration ^

LE8EHD

3500 | CONTROL I CLOP33.AC10 3000 ^9999999999^ 3EZAFJ3RATF CIPROFIBRATE 2500 CD cc. E— ZZ 2000 CD CD 1500. ai LU CD ''vSs*N\ irZ 1000. si 500

STUDT STUDY 2 C C0NTR0L=100 )

FISCHER RAT B MONTH TREATMENT

26 week administration Legend;

CLOFIB.ACID Clofibric acid

See Table 3.3 for dose levels

Absolute values (control);

nmol product formed/min/mg protein

Study 1 (1 x 24hr) 2.40 +/- 0.31

Study 2;

(1 x 48hr) 3.63 +/- 0.64

(2 x 2 4 h r) 2.11 +/- 0.30 115

It has been reported that palmitoyl-CoA oxidation was elevated

2.5 to 3.5-fold with fenofibrate and clofibrate after one

day's administration (Price et al.,1986). This was sustained

for 18 months following continuous treatment with clofibrate

and fenofibrate (Mann et al.,1985, Mitchell et al.,1985). With

dietary administration of clofibrate and tibric acid, the

increase in peroxisomal beta-oxidation was in the order of 15-

and 4-fold respectively (Holloway and Orton,1980). The

induction was 14- and 18-fold for orally administered MEHP and

clofibrate (Lake et al.,1984). The increase in the peroxisomal

beta-oxidation system occurs rapidly after the administration

of peroxisome prolif erators and is dependent on the relative

potency of the compound, dose level and dosing frequency.

3.6.3. Uricase Activity

With once daily dosing, the specific activity of uricase was

unchanged (table 3.4). When the dosing frequency was altered,

uricase activity was significantly decreased in the

ciprofibrate treated group. Activity with bezafibrate and clofibric acid was unaffected by administration. The cause of this difference in uricase activity in the 2 ciprofibrate groups is unclear at present. It is possible that with dosing once every 48 hours, proliferation of "coreless" (i.e. uricase deficient) peroxisomes occurs. Increasing the dosing frequency to once daily may allow for sufficiently high levels of ciprofibrate to maintain uricase positive peroxisomes.

Similarly, with DEHP administration for 3, 7, 14 and 28 days, a reduction in uricase activity was observed (Mitchell et 116

Table 3.4.

Comparison of Hypolipidaemic Drugs under Varying Dosing

Regimens on Peroxisomal Uricase Activity in the Fischer Rat

Following Chronic Administration,

Study Dose Uricase Activity Number Level (nmol/min/mg)

1[a] Control 20.04 +/- 3.15[b] (100)[c]

Ciprofibrate 23.43 +/- 2.11 (117) (20 mg/kg)

Clofibric acid 18.66 +/- 0.83 (93) (50 mg/kg)

Bezafibrate 20.55 +/- 2.47 (103) (75 mg/kg)

2[e] Control 19.29 +/- 2.34 (100)

Ciprofibrate 15.31 +/- 0.67[d] (70) (20 mg/kg)

2[f ] Control 18.74 +/- 3.27 (100)

Clofibric acid 15.80 +/- 1.37 (84) (75 mg/kg)

Bezafibrate 19.02 +/- 1.05 (101) (150 mg/kg)

if 26 week administration [a] Dosing once daily.

[b] Values are means +/- S.D.

[c] Figures in brackets are values expressed as % of corresponding control.

P values for results significantly different (Student's T-test) from control data are d p<0.05

[e] Dosing once every 48 hours.

[f] Dosing twice daily. al.,1985). This was significant at the high dose level at 28

days.

In general, the specific activity of uricase remained

unchanged or was decreased with peroxisome proliferator

administration. This would indicate that either uricase

synthesis is increased in line with proliferation or that

uricase deficient peroxisomes develop.

3.7. CYTOSOLIC GLUTATHIONE PEROXIDASE ACTIVITY

Glutathione peroxidase (GPX) contains covalently bound seleno-

cysteine in the active site and metabolises both hydrogen

peroxide and organic peroxides as substrates. The selenium

independent glutathione peroxidase enzymes are termed

glutathione-S-transferases (GST) and these enzymes metabolise

only the organic hydroperoxides. The difference in enzyme

activity observed on treatment, when expressed as percentage

of control using hydrogen peroxide as the substrate (utilised

by GPX only) and t-butylhydroperoxide (utilised by GPX and

GST) is minimal (results described in chapter 4). In the two current studies GPX activity was determined using t-butylhydroperoxide as the substrate.

With once daily dosing, GPX activity in ciprofibrate and bezafibrate treated animals was significantly decreased

(figure 3.2). Activity in the clofibric acid treated group was unchanged. When the dosing frequencies were altered, enzyme activity was decreased significantly with all three 118

Figure 3.2.

Comparison of Hypolipidaemic Drugs under Varying Dosing

Regimens on Cytosolic Glutathione Peroxidase Activity fal in to the Fischer Rat Following Chronic Administration.

LE8END

125 H H CONTROL S CLOFIB.ACID

100 S BEZAFIBRATE ] CIPROFIBRATE CD DZ. E—• 73 2 : CD CD 50 ZD

25

v'-.'NK. w

STUDY STUDY 2 ( C0NTR0L=100 )

FISCHER RAT 6 MONTH TREATMENT

Legend;

CLOFIB.ACID Clofibric acid

See Table 3.4 for dose levels. [a] Assayed with t-butylhydroperoxide as the substrate. to 26 week administration

Absolute Values (control);

umol substrate metabolised/min/mg protein

Study 1; 0.58 +/- 0.02

Study 2;

(1 x 48 hr) 0.74 +/- 0.22

(2 x 24 hr) 0.77 +/- 0.12 119

hypolipidaemic drugs. In both studies, the decrease in enzyme

activity was greater in ciprofibrate treated animals.

3.8. DISCUSSION

The selection of compounds to be tested for hypolipidaemic

activity in man is generally based upon an assessment of their

effects on circulating lipids in laboratory animals (Cayen,

1985). The rat is the most frequently used species for the

detection and evaluation of agents affecting lipid metabolism,

although there are differences in the distribution of lipids

between the rat and human serum lipoproteins. The decision to

test such a compound in man is therefore usually based on

results obtained mainly in the rat.

The primary aim of this chapter was to examine the effects of

ciprofibrate, clofibric acid and bezafibrate, administered

under varying dosing regimens, on selected hepatic enzyme

parameters in the Fischer rat. In the first study the dose

levels chosen were based on comparative lipid lowering potency

derived from published data (Sterling-Winthrop, personal

communication). The hypolipidaemic agents were administered

once daily and the dose levels used were 20, 50 and 75 mg/kg

for ciprofibrate, |clofibric acid and bezafibrate respectively.

With reference to the observed hepatomegaly and hepatic enzyme changes it is apparent that the inductive potency of these compounds varied considerably.

It has been estimated that the ED-33 (the dose effective in 120

suppressing serum lipids by 33% from control values) of

ciprofibrate for cholesterol and triglycerides in

hyperlip id aemic animals is 1.5 mg/kg/day and « 1.5 mg/kg/day,

respectively (Edelson et al.,1979). In normolip id aemic rats,

ciprofibrate at a dose level of 1 mg/kg caused a 52% decrease

in serum triglycerides (Dvornik and Cayen, 1980). At 2 mg/kg,

total serum cholesterol and phospholipids were lowered and

increasing the dose to 5 and 25 mg/kg produced no further

decrease. This would infer that the hypolipidaemic properties

of ciprofibrate are similar at dose levels of 2 mg/kg/day and

20 mg/kg/day. The inductive potency of ciprofibrate at these

latter two dose levels with respect to enzyme activities, is

studied in the following chapter. Clofibrate at a dose level

of 48 mg/kg/day, decreased serum triglycerides and LDL

phospholipids (Dvornik and Cayen,1980). Higher doses (121 and

242 mg/kg/day) caused a reduction in serum cholesterol.

Bezafibrate decreased serum triglycerides, HDL cholesterol and

HDL phospholipids at dose levels of 11 and 36 mg/kg/day

(Dvornik and Cayen, 1980). To effectively use the rat as a model to predict hypolipidaemic activity requires assessment of rodent data with relevant clinical data from studies in man. When the minimal effective doses are compared with clinically effective doses for the rat and man respectively, a parallelism in dose requirements was observed (Dvornik and

Cayen,1980). With clofibrate, bezafibrate and ciprofibrate, dose levels of 48, 11 and 1 mg/kg in the rat corresponds to

25, 7.5 and 1.7-3.3 mg/kg in man, respectively (Dvornik and

Cayen, 1980). 121

The dose levels chosen for these studies could have been more

accurately selected with reference to hypolipidaemic activity.

However, the primary objective was to compare the effects of

these hypolipidaemic drugs on hepatic enzyme activities under

similar and different dosing regimens and not to study

hypolipidaemic activity per se.

With once daily dosing, the order of magnitude of response for

liver enlargement, induction of peroxisomal beta-oxidation and

decreased GPX activity was ciprofibrate > bezafibrate >

clofibric acid. By altering the frequency of dosing, the

differences in response with these compounds became

negligible. The liver enlargement observed in rats treated

with peroxisome proliferators appears to be an adaptive

response and does not represent gross hepatocellular damage.

The hepatomegaly was accompanied by a short-lasting burst of

mitosis, centrilobular loss of glycogen and a fall in

glucose-6-phosphatase activity (Mitchell et al.,1985). A

midzonal to periportal accumulation of fat occurs (Mann et

al.,1985).

The peroxisomal beta-oxidation system oxidises saturated acyl-CoAs with chain lengths from 7 to 22 or more carbons

(Lazarow,1982). It also actively oxidises long chain unsaturated acyl-CoAs. The enzymatic capacity of the uninduced system is approximately 1 umole of acetyl-CoA formed/min/g liver (Lazarow,1982). The marked induction of the peroxisomal beta-oxidation system in rats exposed to peroxisome proliferators is probably an adaptive change associated with 122

altered lipid metabolism (Mitchell et al.,1985). The decrease

in GPX activity was associated with an unchanged or slightly

elevated cytosolic protein content and may reflect decreased

enzyme synthesis, impaired enzyme stability or in vivo

inhibition of GPX by the peroxisome proliferator and/or

metabolite. It has been established that ciprofibrate

irreversibly binds to glutathione-S-transferase in vitro

(Aswathi et al.,1984).

Succinate dehydrogenase is a marker enzyme for mitochondria

and its specific content remained unchanged with clofibric

acid and bezafibrate treatment. With ciprofibrate, enzyme

activity was increased in study 1 and decreased in the second

study. The results suggest that there may be no change in mitochondrial number and structure or that proliferation occurs with an associated proportional synthesis of succinate dehydrogenase. Mitochondrial alpha-glycerophosphate dehydrogenase acts as a specific marker for the metabolic activity of thyroxine in the liver (Lee et al.,1959). The activity of this enzyme was increased in a similar manner by the three hypolipidaemic agents. The induction in study 1 was

2 to 3-fold and in study 2, 4 to 5-fold. This enhanced alpha-glycerophosphate dehydrogenase activity appears to involve the synthesis of new protein (Lee et al.,1959). It has been estimated that the relative amount of circulating thyroxine (T-4) present as "free T-4" is increased by 40% on clofibrate administration (Ruegamer et al.,1965). The displaced T-4 was concentrated in the liver which consequently became hyperthyroid as evidenced by elevated alpha- 123

glycerophosphate dehydrogenase activity. A single dose of

clofibrate to the rat caused radiolabelled thyroxine to move

from the plasma to liver (Harland and Orr, 1975). Acute

administration resulted in an increased flow of thyroxine to

liver whilst serum levels of the hormone were maintained. With

DEHP treatment, serum triiodothyroxine (T-3) levels remain

unchanged and T-4 levels in plasma were reduced by half while

morphological characteristics of the thyroid indicated

hyperactivity at 7 and 14 days and 9 months (Hinton et

al.,1986). Alpha-glycerophosphate dehydrogenase activity was

elevated in animals exposed to DEHP. This marked increase in

enzyme activity appears to be due to synthesis of new protein as a selective response to peroxisome proliferators mediated by an, as yet unclarified treatment-induced effect on the thyroid.

Carnitine acetyltransf erase is an enzyme catalysing the reversible transfer of an acetyl group from acetyl-CoA to carnitine (Kahonen and Ylikahri, 1974). The mitochondrial fatty acid beta-oxidation system is dependent on this enzyme as an intermediate carrier of acyl-substrates into the mitochondrial matrix. Mitochondria possess short-, medium- and long-chain carnitine acyltransf erase activities whereas only short- and medium-chain length carnitine acyltransf erase activities are observed in peroxisomes (Markwell et al.,1973). Carnitine is not involved in the entry of fatty acids into peroxisomes but may play a role in the removal of end products which could be shuttled to the mitochondria for further oxidation or conversion into ketone bodies (Lazarow, 1982). Carnitine 124

acetyltransferase activity was increased to a similar extent

by ciprofibrate and bezafibrate in both studies although the

induction was considerably lower in the second study.

Clofibric acid treatment resulted in increased enzyme activity

and this was higher when the compound was administered twice daily. It should be remembered that total carnitine acetyltransferase activity was measured (i.e. mitochondrial and peroxisomal enzyme activity). It has been reported that clofibrate administration elevated carnitine acetyltransferase activity in both subcellular fractions (Goldenberg et al.,1976) and a similar response may be assumed in these two current studies.

It has previously been demonstrated that in clofibrate treated animals, the specific activity of uricase was appreciably decreased (Goldenberg et al.,1976). This decrease in enzyme activity was associated with a marked proliferation of peroxisomes indicating that the synthesis of uricase did not parallel the increase in organelle number. The specific activity of uricase was unchanged in bezafibrate and clofibric acid treated groups, however a decrease in activity was observed with ciprofibrate administration. The results would suggest, that in the event of peroxisome proliferation, synthesis of uricase occurs either in parallel to or not in proportion to the numerical increase in organelles (Goldenberg et al.,1976). In a similar manner the reduced activity of uricase in DEHP treated animals accompanied proliferation of hepatic peroxisomes lacking the uricase-rich core (Mann et al.,1985, Mitchell et al.,1985). 125

The plasma half-lives of ciprofibrate, bezafibrate and

clofibrate/clofibric acid in the rat are 3-4 days, 6^-16 hours

and 5-8 hours respectively (Baldwin et al.,1980, Cayen et

al.,1977, Edelson et al.,1979, Thorp et al.,1972, Schmidt et

al.,1980 and Sterling-Winthrop,personal communication). In my

second study, ciprofibrate was administered once every 48 hrs

and bezafibrate and clofibric acid twice daily in an attempt

to compensate for the markedly different pharmacokinetic

profiles in the rat and to mimic the recommended dosing

frequency of these hypolipidaemic drugs in clinical practice.

It is apparent that at the dose levels used, with daily

administration, considerable differences exist in the

inductive potencies of these compounds with respect to hepatic

enzymes. In contrast, altering the dosing frequency (as in my

second study) rendered all three compounds equipotent with

respect to hepatic induction.

The level of exposure (measured as area under the curve [AUC])

of ciprofibrate in the rat is clearly greater than in bezafibrate and clofibric acid treated animals. Steady state

levels of the three drugs were achieved within 7 days

following study commencement (Sterling-Winthrop, personal communication). With ciprofibrate administration, once every

48 hrs, the steady state concentrations were reached at 7 days although drug levels were lower than with daily administration. As expected with clofibric acid and bezafibrate, steady state levels increased with increasing dosing frequency. It has been estimated that the AUC of ciprofibrate (10 mg/kg once every 48 hrs) is approximately 126 equal to a bezafibrate dose level of 125 mg/kg, twice daily

(Sterling-Winthrop,personal communication). In this study, dose levels of 20 and 150 mg/kg ciprofibrate and bezafibrate respectively, were used.

It is apparent that when making comparisons of biological activity between different members of a class of compounds, that both dose levels utilised and the frequency of dosing should be a fundamental consideration. Clofibric acid and bezafibrate are rapidly eliminated in the rat whereas ciprofibrate elimination is slow. Ciprofibrate treatment predisposes the animal to inductive processes as a direct result of its pharmacokinetic characteristics of accumulation and elimination.

A more scientifically enlightened outlook, as seen with the completion of these studies at Sterling Winthrop Research

Centre, is required in regulatory and research toxicology to optimise knowledge of pharmacokinetics to rationalise the selection of dose levels when using different members of the same class of compounds. 127

Chapter 4

SPECIES AND STRAIN DIFFERENCES IN THE HEPATIC RESPONSES TO

SHORT TERM ADMINISTRATION OF CIPROFIBRATE

4.1 INTRODUCTION

In the preceding chapter the influence of pharmacokinetic

differences in the metabolism of ciprofibrate , bezafibrate

and clofibrate were examined. Using a chronic multiple dose

regimen, the relative potencies of these compounds, with

respect to hepatic enzyme parameters, were studied and the

metabolism and pharmacokinetic profiles discussed. It is

apparent that this approach to optimise drug administration,

compensates for the characteristic accumulation of

ciprofibrate in the rat and that the potency of these

compounds becomes qualitatively similar.

Ciprofibrate is a recently developed phenoxyisobutyrate

derivative and it is considered one of the most potent

peroxisome prolif erators currently available. It effectively

decreases serum cholesterol and triglyceride levels in normal

laboratory animals and in animals rendered hyperlipid aemic by

dietary perturbations (Arnold et al.,1979). The peroxisome proliferation induced by ciprofibrate has been studied in the

Fischer rat and in several non-rodent species (cats, chickens, pigeons, Rhesus and Cynomolgus monkeys) (Lalwani et al.,1983a and b, Reddy et al.,1984). The aim of the following investigation was to clarify the species and rat strain hepatic responses to the 14-day administration of ciprofibrate 128

at a biochemical and molecular level.

Determination of enzyme levels in the cytosolic, microsomal, mitochondrial and peroxisomal subcellular fractions was undertaken. The number of animals per dose group is shown in table 4.0 (experimental design). Hybridisation probes to cytochrome P-452 and glutathione peroxidase were used to study the expression of mRNA in livers from untreated and treated animals. The results obtained will be discussed at length to elucidate the mechanism of induction as species specific or species common following the short term administration of ciprofibrate.

4.2 HEPATOMEGALY

With ciprofibrate, there was a significant dose-dependent increase in liver/body weight ratio with a maximal 2-fold induction at comparable dose levels in the Sprague Dawley,

Fischer, Wistar and Long Evans rat; a similar increase was observed in the mouse (tables 4.1 and 4.2). However, hepatomegaly was less marked in the treated Gunn rat. A significantly increased liver/body weight ratio was seen in the hamster at the high dose level yet in the guinea pig, rabbit and marmoset there was no significant change in liver weight (table 4.2). It should be remembered that the animals were aged matched as appropriately as possible across rat strains and species. In the untreated animal the liver/body weight ratio in all 5 rat strains, the mouse, guinea pig and marmoset were very similar (5-6%). The lowest basal value was 129

Table 4.0

EXPERIMENTAL DESIGN

SPECIES/STRAIN NUMBER OF ANIMALS PER

DOSE LEVEL

Mouse a 1 pool of 6 animals

Hamster a 2 pools of 3 animals

Gunn rat Control: 4 animals 2 and 20 mg/kg: 3 animals

Sprague Dawley rat Control: 5 animals 2 and 20 mg/kg: 4 animals

Fischer rat Control: 5 animals 2 mg/kg: 4 animals 20 mg/kg: 5 animals

Wistar rat Control: 5 animals 2 and 20 mg/kg: 4 animals

Long Evans rat Control: 5 animals 2 and 20 mg/kg: 4 animals

Guinea pig Control: 4 animals 2 and 20 mg/kg: 5 animals

Rabbit Control: 4 animals 2 and 20 mg/kg: 4 animals

Marmoset Control: 4 animals 20 mg/kg: 4 animals 100 mg/kg: 4 animals

[a] no statistical analysis completed except with liver/

body weight ratio, expressed as mean of 6 animals

+/- S.D. .— . S |PS .— . ,— , * o O ' p » ^ m g O C M — O W O o o o 0 H c n O W O H ^ \ O O H O l O H o c o r o OHO 0 h i n H H 1 r-H H H 00 r4 H r4 H H H H H H H H H O O N P W s - > 0 4 O V / w w - ’>«' U U g 4 4 4 4 w c r-l P S P S f- s C C X — 0 0 u u u u O O o *—i •—< yj U4 U U

O S E h m c n w o W O W O C M NOIH h wo in WO WO H 4 4 X Z C O O H H O N H O H H O H H O H O ' g • • • C O 6 3 • • • • • • • • • • • • c 0 0 O E h o o o o o o o o o OOOOOO • H M -14 E h Z * 0 4 4 C J H O I 1 1 1 1 1 1 1 1 1 1 1 1 1 1 C O u o \ \ \ \\\ w \ \\\ W \ O — o + + + 4 - 4 - 4 - 4 - + 4 - 4-4-4- 4-4-4- 04 44 J u 0 0 o < C w r o t j 1 i n C O 0 0 H W O O C M W O W O C M h w o i n 0 0

m E h 6 4 O ' c n t*- C " C N O ' IDON W O C O > W O 0 0 O ' M 44 O »-4 • • t • t • • • • • • • • I * M 1

l E h O o o o OHOOHOOOO OOO O 4 4 0 4 6 3 U 0 4 0 0 CO 4 4 4 4 g 0 C o 0 H P"S 0 T J . c 0 1 4 3 O ' 3 u C 1 *— i 0 4 4 o •H | •—» /-H /-% <-s /-» r» «—» . . »— » 4 4 C O 4-1 0 1 O O O CO or-in OOO O S * H C w U 1 O N I D O C O C O OHO 0 h c n O H O ' 0 u 4-1 1 E h H H r l H H H H H H H H H H H H U 4 4 • C O 1 X - w - w s / w ■>— <■ v _ < '--' V M C H 'a 1 o 0 0 O r— * c 4-1 1 w r-^ p— i m «— 1 r— 1 r*S r-m 1— 1 f— I Q 4 M O 0 0 1 6 3 0 0 3 0 3 0 3 0 0 0 0 0 0 0 0 • o s i £ <*> U l_l b-l 1—1 L _ 1 laJ 0 4 4 O 0 « X — ® o ' o S * H C M o i n o o i n 0 00 C - 0 0 O ' 0 4 4 N 4-1 1 Q O r o o w o rl in V H H CN C M C M S * h j * c m i n •ft V • • • • • • • • • *H C 1 O *-* • • • • • • 0 3 0 3 c o 0 1 0 3 E h o o o o o o o o o OOO OOO 0 r— 1 > 4 1 \ < 0 > 1 0 M 0 i O S O S 1 1 1 i i i 1 1 1 III 1 1 1 0 r-H <— • \ w 0 4 4 1 6 3 \ s \ \ \ \ \ s \ \\\ 0 4 4 > 4 4 1 > + + + 4-4-4- 4 - 4 - 4 * 4 - 4 - 4 - 4 - 4 - 4 - 4 C O •rl • H I t-4 00 H 0 4 0 C Q 1 w o w o r - H W O 0 0 h i n h 0 0 w o i n H S 1 H X u 0 1 H n ' T (N C M H • H C — C O O ' O ' • 0 0 4 4 H »»*t 0 0 0 V M > X 0 <— 1 •ft Q 0 M * 0 4 4 CO > 4 0 — O CD 0 U O ' O ' O ' O ' O ' O ' O ' O ' O ' O ' > 1 0 •* Q i r 4 A S J4, H J* x H Jt£ JX H M h 4*: 4 * CO N 0 4 4 i n •H J o w 0 w O N . N . 0 w 0 w U + 4 4 H O u 6 1 6 3 M O' O' M O' O' > 4 O ' O ' >4 O' O' M O' O' 0 3 • 4 4 e g 4-1 g g - * 0 O C O > 4J g e + > e g - P g g Ml" W 4 4 O 6 3 c c c c C ,T~ c U 0 O Q J o o 0 o o o 0 0 O O 0 0 w v U C M C M O n n U C M C M O C M C M U C M C M I— 0 >4 4 J O g 3 3 W - r *♦— U 0 u 0 0 0 C 4 4 — 4 4 CD M •ft 4 4 O 0 0 0 6 3 TJ 0 0 M W 0 3 1 0 0 > 0 >4 H H 1 4 3 3 3 0 0 • 1 E h 0 H O ' > 4 4 M 1 1 0 > 1 M T > 0 • H 0 1 Z 3 0 0 U W 0 > 64 G u T D HI 1 1-4 O ' H 4 3 . 0 c • • H ! E h < c 0 £ U 4 4 O ' 0 O) • • J 3 1 < O S c >4 0 IQ 0 c > C 1—« P ^ 0 i O S E h 3 a o •<-4 •P-< O 6 3 0 J 3 0 3 E h 1 CO o CO 64 £ J UJ IN C-t JQ rH *0 rD 0

The Effect of Ciprofibrate on Liver Size and Cytochrome P-450 Content <0 • rH <0 4-> •p-i CO Q i—i 0 0 0 0 G Ck 0 u 0 G Ogo 1 N g LO Cn o w < a £ E-<£ O £ 2 0 o O u E »-3 PQ >4 w I— I u £ E" o > H OS 02 O Q Q J Q 63 O H a. w O >H > w CO M 63 CO 1 h

o \ rH U <-/ a. 63 &4 E £ E 63 HH u < E CO u 1— t a J O c g - r h h h

.— . , \ ro o o H<— i rH . i— ! ro o CO m CN CO o o o'C co m rH in -P u _ + _ (0 o C \ \ O 1 • • • • • • • • • • • • • . • . . ,

nOn in VO CN . VO . o _ KD CO — rH in \ r- CN CN CN 4«;44 £ _ _ + _ 0 g g CP CP CP CP CP O 0 10 0 1 . , _ .

no in ,— \ on _ rH .— . .— . uo rH rH o o o o t— i CO rH on t— ! o -t- _ 0 ! • • • • • . • . , . , O — o Ovo CO o o rH r~ o o \ o CO O t— ! .— or CO ro ro ro CO rH ro 44 U CN CN CN U + + o 0 H O 1 1 I I 1 1 1 i ! 1 l 1 ■ I 1 1 l • • * • • • • • • • • • . O CO CN '— tH rH CO on ro -— <— I o Oo CO r-> CN O o CN OCO CO CO 44 44 44 \ \ \ ro j j + *r + g g CP CP CP CP CP rc g in 0 u i i i • • • • No CN — o o CN 1 N .— o \ \ cn rH tH ro uo —.- —.. -— ~ .— .— . -— . . .— . O 1 N a. i . O rH .. o '— CO o CN \ \ \ \ 44 i— I O vo noO o in O oro ro o o o ro rH '— CN o o rH in o -+ H- + — + o H o C i i i I . oo O on ro '— c-~ tH .— . -—.— . . nC O CO in tH r-. —j i— rH in \ \ 4 I 44 rH on o 02 H- 0 g g Cn CP 03 0

O m ro o rH NrH CN ro — !rH i— ! rH CN —■ ■— CO o rH O rH *=rN1 o rH a. 0 • o o o tH o o oCO ro '— \ r-{ VO 4 g g 44 o cm O CN ’ N \ \ \ ni in in in i— 1 CJ CN CN CN CJ *H- H* + -t. 0 CP CP CP 0 o w O O £ i • • • • • • • '— ' nOn on t— i o on '— ' on no on —•.— . .— • \ CN rH an o o o 1 N o • rH CU -4- 4- P CP CP 0 0 CP 1 No CN m \ m » — o i— i c ■— rH m CN t— I i— I i • O o o -— • Hr rH rH rH in t— ! in '— \ o \ \ ro m '— co o o CN rH 4»2 rH ro o 4 g g 44 U CN CN CN U 4- 4- o o c CP CP 0 o w i ■ • • • • • • NC O CN in CN m ^-s o ro cc ■w l> o o o o in Oo VO o w H•N1 tH n'sr in \ o o rH X) jq •rH 44 02 r+H- + -r 4- CP cn CP 0 i no in vo -— * 03 \ \ —•-— ' -— • HrH rH '— ' *— CN o t"- '— H rH O •^r -u I <

O rH o «— On \ \ \ t— i O —1rH t— 1 oCN ro O o CO CN 1 N VO tH H \ rH 4 g 44 a CN rH CN a + o \ CP CP u \ o o o o o c 1 • • ' Pgg CP co -O c- VO rH 1 N '— t— l O -— - Hro rH VO o \ ^— co ro ro CO O o O 4-J £ 4- -4- 4- cn cn 0 h g o in 0 1 • • • KD O H \ r- 1 1— 1 \ VO vo ro o ro ‘— o rH i— l 4*2 • — + 0 cn cn 1 • • • 03 O rcCO. 03 03* CO CO >* <0 0 C 0 O O) 0 i— i T3 44 44 44 'O 44 •rH rH 44 •H XI 44 •H 43 i— 1 pt4 H o 0 CP 0 O U 0 0 0 04 o c CP U O C >i 0 0 0 04 0 >H U 0 0 o U C V4 > 0 0 X 04 M 0 0 0 0 0 0 0 0 0 0 0 U 0 .. 3 0 0 0 CP 44 0 rH JO i— 1 <44 03 40 G 0 44 1 N 03 Q 0 0 3 0 o g U U .. 0 O 0 04 0 0 0 0 • rH -P 44 03 44 44 <44 <44 03 U4 44 -— • 44 rH • rH CO <0 • •0 • 44 ■0 rH O rH 44 10 • rH rH 04 03 04 O U 0 0 0 0 3 0 C 0 0 0 0 O g O 3 0 U O U O 0 tH G 0 v G 0 03 0 0 0 CP • 3 0 O 0 > 44 > V 0 0 O 03 O 0 3 0 0 0 0 0 0 1 o m <— • 0 0 132

seen in the rabbit and hamster- (3-4%).

The response of the Fischer rat to ciprofibrate was consistent

with the published data (Lalwani et al.,1983b). With DEHP,

ME HP and clofibric acid administration, liver enlargement was

observed in the rat and Syrian hamster, the magnitude being

greater in the rat in each case (Lake et al.,1984, Lake et

al.,1986). However with MEHP administration, hepatomegaly was

more marked in Chinese hamsters as opposed to Syrian hamsters

(Lake et al.,1986a). With fenofibrate treatment a non­

significant increase in liver/body weight ratio in the hamster

was observed (Pourbaix et al.,1984). The extent of

hepatomegaly is probably a product of the relative potency of

the peroxisome proliferator in use. Hepatomegaly has also been

observed in the ferret following DEHP treatment (Lake et

al.,1976). There is no published data with respect to the

measurement of liver/body weight ratios in the guinea pig or

rabbit. No liver enlargement was observed in the marmoset

following a 14 day treatment period with clofibrate,

clobuzarit and DEHP (Jackh et al.,1984, Orton et al.,1982,

Orton et al.,1984). It would appear that the guinea pig,

rabbit and marmoset are refractory in response to ciprofibrate

treatment with reference to hepatomegaly, bearing in mind that

ciprofibrate is a very potent compound. In general terms, the

greatest induction of hepatomegaly in peroxisome proliferator treated rats is 2-fold, irrespective of increased dose levels

(Cohen and Grasso,1981). This would suggest saturation of a metabolic process producing a maximal response. 133

4.3 THE MICROSOMAL DRUG METABOLISING ENZYME SYSTEM

The administration of peroxisome prolif erators to rodents

results in the proliferation of the S.E.R. and the

characteristic induction of a cytochrome P-450 species

involved in lipid metabolism. This haemoprotein has been purified, characterised and described as a novel form, cytochrome P-452, which actively catalyses the 12 hydroxylation of lauric acid ( Tamburini et al., 1984 ).

Cytochrome P-452 is an unusual isoenzyme in that it exhibits specificity for endogenous substrates as opposed to model drug substrates unlike the cytochromes P-450 induced by the classic compounds phenobarbitone and 3-methylcholanthrene. In this investigation the ability of ciprofibrate to induce cytochrome

P-452, as assessed catalytically and immunochemically, has been examined. In addition, the metabolism of benzphetamine and ethoxyresorufin, model drug substrates, has been studied.

4.3.1 Cytochrome P-450 Content

Variable changes in the specific content of cytochrome P-450 were observed in the treated animals. In the Sprague Dawley,

Fischer and Wistar rat there was an increase at both dose levels (table 4.1). However this was not dose-dependent and the specific content was higher at the 2 mg/kg than at 20 mg/kg dose levels. The increases were significant at the low dose level and only at the high dose level in the Fischer rat.

There was a significant dose-dependent increase (p<0.05) in the Long Evans rat with no significant changes detected in the 134

cytochrome P-450 content of the Gunn rat. In the rat, the

greatest induction in cytochrome P-450 levels was displayed in

the low dose group in the Sprague Dawley and Fischer rat (1.5

to 2-fold).

As with hepatomegaly a pattern of induction similar to the

Fischer rat was demonstrated in the mouse (table 4.2). In the

hamster an increased specific content was apparent at the high

dose level with a non-significant dose-dependent elevation in

cytochrome P-450 levels in the guinea pig (table 4.2). A

significant 1.5-fold induction (p<0.05) at 20 mg/kg was seen

in the rabbit. A dose-dependent increase in the cytochrome

P-450 content is observed in the marmoset, significant at the

high dose level(p<0.001). The marmoset was the most responsive

species to the inductive influence of ciprofibrate on

cytochrome P-450 content and displayed a 3-fold induction at

the high dose level. The high dose level in the marmoset was

100 mg/kg whereas the high dose level in the other species was

20 mg/kg. The basal level of cytochrome P-450 in the untreated

animal was widely variable between species and rat strain. The

lowest cytochrome P-450 specific content was exhibited by the

marmoset which was 4 to 8-fold lower than the rabbit, mouse,

Fischer, Wistar and Long Evans rat. In close ascending order

was the Sprague Dawley rat and hamster with the Gunn rat and

guinea pig specific content being 8 to 9-fold greater than the marmoset.

By comparison with the above data, administration of clobuzarit resulted in a significant dose-dependent increase 135

in cytochrome P-450 content in the rat and mouse, a minimal

non-significant increase in the hamster and unchanged levels

in the marmoset (Orton et al.,1984). With DEHP, MEHP and

clofibric acid, a maximal 1.5-fold induction of cytochrome

P-450 was observed in the hamster (Lake et al.,1984).

Induction of cytochrome P-450 by peroxisome prolif erators

appears to be a characteristic response in the rat, mouse and

hamster. Cytochrome P-450 levels in the ferret remained

unchanged following the administration of DEHP (Lake et

al.,1976). The increase observed in the marmoset by

ciprofibrate has not been demonstrated with clobuzarit and may

reflect both the relative potencies and pharmacokinetic characteristics of these two compounds.

4.3.2 Benzphetamine-N-Demethylase Activity

In the rat there was a dose-dependent decrease in benzphetamine-N-demethylase activity which was significant in all strains with the exception of the Gunn rat (table 4.3). In the Fischer and Long Evans rat, the activity at the high dose level was 15-30% of the control value whereas in the remaining strains it was approximately 50%. In the mouse, enzyme activity was halved at both 2 and 20 mg /kg ciprofibrate (table

4.4). There was an increased benzphetamine turnover at the low dose level in the hamster, at the high dose level values were similar to control. In the guinea pig, activity remained unchanged however there was a significantly decreased activity at the high dose in the rabbit. An apparent dose-dependent H I rH -'ri e*i XII mi 01I i

The Effect of Ciprofibrate on the Microsomal Drug Metabolising Enzyme p p P p •H p P co i—i Q OJ »H CQ >1 in 0) e G 0 P (1)C fO p fO 0 c m (3 Z CM X *£ xi m Z E < Oi SC £3 Z j M w w E O C3 co o vh SC X 13 X OS D £3 z 0 OS (3 1 1 h 1 i h OSE ^ ^ rfj < < 03 Q Q J 13 O CO> £3 (3 13 £3 r C rH Q E SC *3 di X — s Ed O\o CO \ sn rH Q E JM J < [3 CO h h

•H CO N •J TO Ou •H inp o \ c £ I 3 U O Or O rH P G o o Q< rH P O S3I U £ - g h

Hjt £ c e

\ r—1 4-> 00 CO 00 o in CM CO rH r— i o ,, o u rH rH O o o o rH rH <9* o o o rH H- + c — o p o ro 1 1 • . • • • • • . w CM CM VO vo ■N* CO o co O MCO CM o 00 o in O O —,.— . „— , in + + 3 c C cp e e tn 1 I I . • tn o vo r** vo o CM rH vo vo o o .— . o o O o o o o HJ1 + + b cp 1 • • . • iH CO o \ -p 0in i f0 r—* \ o H . in o + + QiQ ro P o £ ro CPrH 3 0 0 >i m B S CP CP . — 1 1 • * • • > . • s o u vo —i r— O O N o rH H tH H rH O ■ in CM + + cp 0 t 1 1 • • j rH vo CM •P rH rH C^ cn CO \ \ oo O o in CN CM O rH o o o O O r—» .— . \ o o o rH 'CT rH o CN o + + N N O o C P t 1 * • • sz in vo >— • o >— i cn 00 OrH CO O 0rH 00 CM 13 l _ r \ + + + in u 0 p CP CP e CP CP 0 i

• • • •

o nC ovo vo CO 00 in o r—i w 1 N \\ \ \ s UJ 13 \ in 00 *—* O —i r— O O rH O r~. O o rH + e u I 1 • • • * • • rH VO CO CM rH rH rH o p 00 \ 00 rH o o o o «9« CJ CM CM rH O o rH CN rH H- + + > — o c CP p CP o 1 1 • • •H P w M M co 00 00 o r—« m in o £ O «— » o in in O rH H O O •— «■*— • \ + in ro p £ CP CP u u I 1 • . • « • o cn rH <— < cn cn O Hjr O o tH u— HrH tH \ rH CM + + -+ H- e T3 U u u u 1 I . • • • « CO s «H J* P cn 00 OCM CO MCM CM CM p o \ o Hi CM in rH iH CO o o O O CM CN U o o o o o CP p CP c I I • • • • • • • w CM tH r** o CM J CO \ O r“» O Jk_ CP fO S £ CP CP 1 I • • 1 • o VO o m L_J CM cn O O O O rH rH 1— 1 \ *-* 00 CO + + c in U 1 1 . .

<0 ^ £' T3 ■o' ■a —, r— r-. C 0 03 i *• •• CO o (0 (0 fH \ > 3 Qu f3 t> co a + 0 3 0 in 0 p 0 JQ £ P 0 13 0 c in •r-H rH 1313 •H P p p p rQ rQ •H p p iH CP 3 P > f3 3 0 c in a) (3 o in in 0 X a P 0 ro P U 0 u 0 13P 3 CP U CP 0 0 o p p 0 in rt i! in f G >1 0 0 c c in O 3 U 0 G 0 0 in p o £ U O G i 4 data are [c] < 0.01ov<\cA Cd] < 0.001 Table 4.4 The Effect of Ciprofibrate on the Microsomal Drug Metabolising Enzyme H r • H r • p P P 4-> co a W M 0 >1 c o P (D G Q 0 u W in 0 g j E W E < NQrl Q CN E 63 CU z E u 6 M O O C 3 6 X X S O < - E o z 63 z CQ z D o 3 3 3 6 1 1 h w * « r o G 1 1 CO U C 63 U n 63 E 63 E TJ E X p CO 63 63 < co p a 63 O co 63> 63 e E G H r Q 3 6 3 6 P O w \ 3 6 3 T X < h h pin p ©u \ © •H \ p • \ P CU O rH P O P □ g C n i g O G p O i Q G g P U o g H

n i H r CU U C G g g 1 1 r-i o © H in m ■pee CN h cn O' , o o —.. .— . .— . „— O V O o O U U N IN O V H t _ ro o o c m o w • . rs » O' cn p p in © E in , N C 1 N __ cp cn o p n i 0 . • . n© in CO O C O O o o O C • •

6 g J 4 © \ W rl P rl rl © © r_ w 1—1 © © rl rl O' o rl © o o o J N C N C u H r H O r H t O T N \ \ \ —1i l i— 1 1— + + c P O' O' o O o w 1 1 # • • * • P o o E OCO CO w rl rl CO O n i o CO rl O' o o o n i O o N C H- + O' O' O f g n i 0 P 1 1 • • • • w ,— . o o o H r o n i * N + + + - H 1 1 1 1 • • • • . .

g g p - p : * 4 l r CN in © © \ W O' rH 00 rl o r- CO o L) CN CN rl \ W o o O H r ,w w o o H r © P O' O' o o c . — o w • • • • • 1 P © •— UJ in 1 N o w in P CO r— CO rl O O P i f— O O _ i w O H r O O H N C o + l . • • t .

g g P JbC P rH © \ \ \ CN rl rl rl co CN CN U o rH O © CN CO \ N C H r N C o O O V 1 N 1 N o o H r o + + + c o o P o w 1 1 • • • • ’ • H r • H • O' o U C CN H w O' co co —■» «— CO \ O o o O © H r + ' O p C 0 o r O' O' O' O' 1 1 • • • • O' r% O' co -o r- o © \ O o O o o * r H r + * H 1 1 • • .

CO rH rl CN © r* g g p rl © rl C*“ CN CN U rH O' C""*© rl \ \ \ N C O o —■• > •— ■ •— o N C H r N N C 1rl N rl o + + G P O' o O'w o o - 1 1 • • • • • . ' * XJ Q J •H p - © rH © © 132 © CN in O r. O V » r o o l r O H r H- + + + CO O' O' 1 1 1 • • •

' ' a CO CO © co -— • 1 N o O O co l r r* rl 0 1 • * : • '

P - \ \ \ CO h r» © CN © CN U w © N* H © © H H O N N \ o O o H N C N C N C O l r o rl H- + O G P O i 1 • • • • •

in p O O Nf O'\ P >*10\ E 1 N in in © H r» o o r* © 0 0 © ' O n i H- - H ro p g o in 0 O' g CO O' ^ 3 T g 1 1 • • © O H © © © •— • C~» © © —» <— o 0 0 n i l r H- - H O' 1 1 • • • • '

E o U CO“ ro P .E mmm 75 > Du > ■U 0 O C O) .r— O) ir— r— l O f o O P C 0 <

• • • •• •

in p • 0 p i-i a iH O roP p p O' —l i— ' O l — in 0 to fO > ro • p ro p p 0 Q J 0 X CU ro > > 0 P £ 'O o 0 h g r0 O' in P G ro 0 X Qj 0 Qi ro cn 0 Qi o 0 G C P p o a) u-i in P P

U P rH p - •i~4 r-H TO 4-> P p to P P in 0 in p 0 H G U 0 in 0 0 in ro 0 0 x p 0 U p (0 in Qi c c O' u o c p o in o

in

0 O f

P > P 'O O TJ UN J 4 :Q rH co Qi O O O f O r p 0 cu p in rO to to ro P U 0 O UN UN O P O g 0

TO UN UN p P P - P UN P i—I UN n r H r • o •H • P H • rH P rHO P P P •H CO v p in rO 0 0 P in P 0 c 0 p p O' v G p 0 P 0 G in 0 in P O g U O G P O in o U ro C < 1

• TJ n u o o 0 rO G

138

decrease in benzphetamine-N-demethylase activity was exhibited

by the marmoset but this was not significant due to a marked

inter-animal variation within each dose group (table 4.4). The

basal levels in the mouse, guinea pig, Sprague Dawley and Gunn

rats were very similar whereas the marmoset, rabbit, Wistar,

Long Evans and Fischer rat exhibited an activity approximately

2-fold higher. The turnover of benzphetamine was lowest in the

untreated hamster.

Benzphetamine is used as a marker substrate for the major

phenobarbitone-inducible form of cytochrome P-450 in the rat

liver (designated P-450b) (Ryan et al.,1979). A wide range of

peroxisome proliferators including clobuzarit decreased

benzphetamine-N-demethylase activity in a manner similar to

that seen with ciprofibrate in the present study, in the

Wistar rat (Sharma et al.,1988). Aminopyrine may also be used

as a substrate for cytochrome P-450b. With clobuzarit metabolism treatment a significant increase in aminopyrine|was observed

in the rat and mouse (Orton et al.,1984). However activity was

decreased with clobuzarit treatment in the hamster and

marmoset but this was not significant. The influence of

peroxisome prolif erators on microsomal mediated benzphetamine

metabolism has not been studied in the rabbit and guinea pig.

4.3.3 Ethoxyresorufin-O-Deethylase Activity

Ethoxyresorufin-O-deethylase activity was decreased in a dose-dependent manner in the rat (table 4.3). This was significant in all strains except the Gunn rat. In the 139

Fischer, Wistar, Sprague Dawley and Long Evans rat, activity

was decreased to 20-30% of the control value whereas in the

Gunn rat this value was 40%. A similar pattern was observed in

the mouse with a maximal decrease of 30% at the high dose

level (table 4.4). A dose-dependent decrease was seen in the

rabbit and marmoset but this was not significant due to the

large response variation within each group. The activity of

ethoxyresorufin-O-deethylase in untreated animals was highest

in the rabbit and marmoset. Activity in the guinea pig

remained unaffected by ciprofibrate treatment. The changes in

ethoxyresorufin-O-deethylase activity in the hamster were

similar to that with benzphetamine demethylation; there was

increased activity at the low dose but values were similar to

control at the high dose level. The activity in the untreated

Gunn rat was approximately 2-fold higher than in the remaining

rat strains,mouse, hamster and guinea pig.

Cytochrome P-450c is the major 3-methylcholanthrene-inducible

form in rat liver and preferentially catalyses the metabolism

of 7-ethoxyresorufin to resorufin. Administration of the

phthalate ester plasticisers (DEHP and MEHP), clofibrate,

bezafibrate and clobuzarit decreased ethoxyresorufin-O-

deethylase activity (Sharma et al.,1988). This was comparable

to ciprofibrate in the present study. In an earlier study with

a clobuzarit dose level of 10 mg/kg as compared to 50 mg/kg,

no change in activity (nmol/min/mg protein), was seen in the

rat (Orton et al.,1984). Ethoxyresorufin-O-deethylase levels were unchanged in the clobuzarit treated mouse, hamster and marmoset. The activity in the untreated marmoset (expressed 140

as n mol/min/nmol P-450), was quantitatively similar with ciprofibrate in the present study and with clobuzarit in the published data.

4.3.4 Induction of Cytochrome P-452

The induction of cytochrome P-452 was assessed:

1) catalytically using the marker substrate lauric acid. The

11- and 12- hydroxy metabolites were separated by reverse phase HPLC, and

2) immunochemically using an enzyme linked immunosorbent assay

(ELISA) technique utilising a polyclonal non-cross reacting antibody raised to the pure protein.

In the rat there was a dose-dependent increase in the

11-hydroxylation of lauric acid which was significant at the low dose level in the Sprague Dawley, Fischer and Wistar rat and at the high dose level in every strain (table 4.5). The induction of lauric acid (omega-1)-hydroxylase varied between

1.5 to 2-fold in the Long Evans rat to 5-fold in the Wistar rat. In each case the ratio of 12- to 11-hydroxylation was greater than 1, varying from 1.25 in the Long Evans rat to

2.27 in the Sprague Dawley rat. The 12-hydroxylation of lauric acid was preferentially induced by ciprofibrate in the rat in a dose-dependent manner, significant at both low and high dose levels (pcO.OOl) (table 4.5). The induction of omega-hydroxylation of lauric acid was 10-fold in the Gunn and

Long Evans rat and 15-fold in the remaining strains. The

(omega-1)-hydroxylation was induced 3-fold at 20 mg/kg in the The Effect of Ciprofibrate on Cytochroie P-452 Content and Laoric Acid 4) B O CJ Cb I |X (O & o r M h • » r e M M > 9 t « I ■ O U"> LT> O • >M 9 rowi S9 m

M C M C O M C M C o > r t M C M C M - « > / l M C • • > - u > u • • • • m • wr>^ m c a> 4)

-u» « m c » r c o r M C o w M C ^ * r > / i o w * - r f— o o m c mco m o M w u 9 er> r e 9 u > < 7 r ( c u 3 0 o o o o C3 i i i — c o r no «i m c m c 0 0 » cn0 r-c >M V o r

t m m * J.* ■ si

h*> M C » M o c o w m p m o r c t u o o 1 0 o u*> ro ur; r~*^ i a i i I t I I * o w r o cr* o u O C h u o r

■ -» « n -u» Jtaf M J • *— ■ a M J » * I r— O W M C M O C O O O V o v ■ — r O O C M C o r H * n i O p W » c O CM C> M C M C C_> M C M C J C e m 0 o o o o 0 H I I I I I I m o r r c> er> n c cr> er» a> 09 t* en u n e er> u n e m c M C O C

ro o vo o ro oo: ■o' T3 JD C O) 0 0 O © CO CO 1 > CO

: : Figures in brackets are values expressed as percentage of corresponding control P P values for results are significantly different (SMents t-test) froa control data are [c] < 0.05, (d) < 0.01 and [e] < 0 . 0 0 1 142 mouse and a slight increase was seen in the hamster (figure

4.1). The formation of 11-hydroxylaurate was not significantly

altered in the guinea pig, rabbit and marmoset. With the

omega-hydroxylation of lauric acid there was a dose-dependent

increase in the mouse, hamster and rabbit; the induction at the high dose level was 10-, 5.5- and 3-fold respectively

(figure 4.1). The increase was significant at both dose levels in the rabbit. A non-significant decrease in omega- hydroxylation was observed at the high dose level in the marmoset. In the mouse, hamster and rabbit the ratio of 12- to

11-hydroxy metabolites was greater than 1, i.e. 1.99, 1.44 and

1.31 respectively (figure 4.2). In the guinea pig and marmoset the ratios were reversed, 0.50 in the guinea pig and 0.24 in the marmoset.

Comparison of the above data with published data available showed that the administration of a variety of peroxisome prolif erators to Wistar rats resulted in a maximal 2.5-fold induction of 11-hydroxylation with nafenopin and a 9-fold increase in 12-hydroxylaurate with Wy-14,643 (Sharma et al.,1988). In the present study the influence of ciprofibrate in the Wistar rat was more marked, 5- and 15-fold induction in the (omega-1)- and omega-hydroxylation of lauric acid.

Clobuzarit pretreatment resulted in a 6-fold increase in total lauric acid hydroxylation (11- and 12-hydroxylaurate) in the rat (Orton et al.,1984). In the mouse and hamster, a 9-fold induction of lauric acid omega-hydroxylation was observed with clobuzarit and there was no significant alteration to lauric acid hydroxylation in the marmoset. The "active" metabolite of 143

Figure 4.1

The Effect of Ciprofibrate on Lauric Acid Hydroxylase Activity

in Different Species

LE6END

1500 gfes CONTROL H U L.D. 11-OH 1250 f U L . D . 12-CH. I I H.D. ; ;-CK CD 1000 H.D. 12-QH C£ E— :z: o CJ 750 Lu CD 500

250

£ = ^iUBa MOUSE HAMSTER RAT F-344 GU.PIG RABBIT MARMOSET ( C 0 N T R 0 L = 100 )

SPECIES 2 WEEK TREATMENT

Legend; L.D.11-OH Low Dose 11-hydroxylaurate L.D.12-OH Low Dose 12-hydroxylaurate H.D.11-OH High Dose 11-hydroxylaurate H.D.12-OH High Dose 12-hydroxylaurate

Absolute Values (control);

nmol hydroxyproduct formed/minute/ nmol cyt. P-450 Species 11-hydroxylaurate 12-hydroxylaurate

Mouse 2.63 5.23 Hamster 3.50 +/- 0.94 5.06 +/- 1.25 Rat (F-344) [a] 2.68 +/- 0.84 3.52 +/- 1.20 Guinea pig 5.20 +/- 1.73 2.58 +/- 0.78 Rabbit 5.73 +/- 1.94 7.51 +/- 1.50 Marmoset 13.77 +/- 6.46 3.34 +/- 2.42

[a] Data reproduced from table 4.5.

Dose Levels; Low Dose : Mouse, Hamster, Rat, Guinea pig and Rabbit: 2mg/kg Marmoset: 20mg/kg High Dose : Mouse, Hamster, Rat, Guinea pig and Rabbit 20mg/kg Marmoset:lOOmg/kg

Single daily dose 144

Figure 4.2

The Ratio of Lauric Acid Hydroxylated Metabolites in Untreated

Animals of Different Species

LAURIC ACID HYDROXYLASE LE8END RATIO OF METABOLITES

| 11-HYDROXY. I 12-HYDROXY■ CD CO •'‘f I Q_ r-1 CD

CD sn

MOUSE ‘HAMSTER RAT F-344 GU.PI0 ‘ RABBIT ‘1ARM0SET

SPECIES ( UNTREATED )

Absolute Values in Untreated Animals ; expressed as nmol product product formed/min/mg protein

Numerical ratio of metabolites; see text

Legend:

11-hydroxy. 11-hydroxylaurate

12-hydroxy. 12-hydroxylaurate 145

ME HP was added to cultured guinea pig and marmoset hepatocytes

with little effect upon the microsomal hydroxylation of lauric

acid (Elcombe and Mitchell, 1986).

The specific content of cytochrome P-452 (nmol/nmol total

P-450), determined using an ELISA technique, was increased at

both dose levels in the rat and this was dose-dependent in the

Fischer and Long Evans rat (table 4.5). In the remaining

strains the specific content was higher at the low dose level than the high dose level although the increase was significant at both dose levels. It should be noted that the increase in total cytochrome P-450 content was only dose-dependent in the

Long Evans rat - in the other strains the induction was higher at 2 mg/kg than 20 mg/kg ciprofibrate. Cytochrome P-452 was increased in a dose-dependent manner in the rat when expressed as a percentage of total cytochrome P-450 (table 4.5). This isoenzyme constitutes between 3-6% of the total CO-discernible cytochrome P-450 population in uninduced rat liver microsomes.

At the high dose level, the percentage increased to 48% in the

Long Evans rat, representing an induction of 10-fold. In the other rat strains the cytochrome P-452 content varied between

24 and 36% representing induction in the range of 5 to 8-fold.

A dose-dependent increase in cytochrome P-452 content, expressed as specific content and as percentage of total cytochrome P-450, was observed in the mouse, hamster and rabbit (figure 4.3 and 4.4). This was significant in the rabbit. Cytochrome P-452 has not been detected immunochemically in liver microsomes of the guinea pig and marmoset , (the minimum level of detection was 0.002 pmoles/well, an absorbance of 0.040, double the blank value). 146

Figure 4.3.

The Effect of Ciprofibrate on Cytochrome P-452 Specific

Content (nmol/nmol total P-450).

LESEND

.3000 CONTROL LOf DOSE | j l HI8H DOSE E—■ zz: in .2000 CD CD CD

CD ,1000 LD Cu CO

.0000 HOUSE HAHSTER RAT F-344 SU. PIG RABBIT HARHOSET

SPECIES WEEK TREATMENT

[a] Rat data reproduced from table 4.5.

N.D. Not Detected

Dose Levels;

Single daily dose. Low Dose : 2mg/kg

High Dose : 20mg/kg 147

Figure 4.4.

The Effect of Ciprofibrate on Cytochrome P-452 Content

Expressed as Percentage of Tote.! Cytochrome P~450

LEGEND

CONTROL LOI DOSE CD HIGH DOSE CO'3-' I cu o E—■ >—■ C_D

E— CD E— U, 3918 CD

MOUSE HAMSTER RAT F-344 GU. PIG RABBIT MARMOSET

SPECIES 2 WEEK TREATMENT

[a] Rat data reproduced from table 4.5.

N.D. Not Detected

Dose Levels;

Low Dose : 2mg/kg

High Dose : 20mg/kg Single daily dose

(See figure 4.3 for absolute values) 148

It is apparent that cytochrome P-452 is a constitutive form of

the haemoprotein and that ciprofibrate functions as an inducer

of an endogenous cytochrome P-450 isoenzyme. Constitutive

cytochrome P-450 isoenzymes are forms present in uninduced

animals. Using a single radial immunodiffusion technique,

cytochrome P-452 was previously found to constitute 22% of the

total cytochrome P-450 population in untreated animals. This

was increased to 57% on clofibrate administration (Bains et

al.,1985). These figures have subsequently been revised using

the more sensitive technique of ELISA, to 4% and 22-25% with

clofibrate representing a 6-fold induction (Sharma et

al.,1988). This immunochemical assay was utilised in this

investigation and the basal level in uninduced rat liver

microsomes was comparable i.e. 4-6% of the total cytochrome

P-450 content. The influence of a variety of peroxisome

proliferators on cytochrome P-452 content has been examined in

the Wistar rat (Sharma et al.,1988). The assessment of

cytochrome P-452 levels using immunochemical techniques has

not previously been reported in different rat strains and

species.

4.4 MITOCHONDRIAL ENZYME LEVELS

4.4.1 Monoamine Oxidase Activity

Monoamine oxidase is an outer mitochondrial membrane enzyme.

In the Fischer and Long Evans rat a dose-dependent decrease in monoamine oxidase activity was observed and which was

significant at the high dose level (table 4.6). In the other ) I X) I to EiI I rH I O f I G •i I i I l

The Effect of Ciprofibrate on Mitochondrial Enzyme Activities in Different Rat Strains 2 z in Eh W CU CU 2 O U W Eh in Z tHEd O W > G3 O o o H H O O C < CUw o COCO X Ed Ei h Q C Q t l . g — -h z w

X X E < < c X O w o > CO Ed Ed cp < h Ed a X H > a 2 0 O Ed o -a x < z 3 0 Ed < q E H - l H - l ■ > E H > < OS h h h H - t

CO Eh z H r c X s •H — I c E-< ► tH E-* g o > SM C g o M o 04 ) a 3 u g c g O' h h G

J

j »H Qi H J o

4J g g r-1 -X .X N* U CM CM CM U O C m o t o o — o t o o ■ 1 ) X H r H r o O oo — c H r H r N C N C s o H r ^--* - 4 H r o o » «— CNJ O' cn w o s o o rH O O o o c o o w - 4 o o rH 4- i I I I 1 ■ 1 1 • • • • • • • • « • • • • • • O m N C or- to s * — o H r o o c OO H r to H « H r '— N C N C - 4 O C H r ' O 3 ' C c tH o o r p c - 4 . , , o o s in o rH 4- g I J • • • • • • • • u J O' o ro o o s s H r H r o t o r o c O C —. r— 3 0 4 U _ OO H r H r H r s s in - 4 N C ' O o r s rH o o o in o o o o * 4 J L 1 — r t u 4- H 1 1 • • s

X gX g r-» .X .x to M CM CM O * N O H r o O C * N o > w oro ro H r to H r ' O - C N C o - 4 H r O —»— » «— s rH o o c * V *'N. U O t O - 4 o o rH O -4* 4- I I I H | 1 • • • 1. • O' O' 03 O s H r ’ w O c4 3 0 cri s - 4 H r H r to * J H — u a) 3 a) >i ro M ro CP i-4 H r O O U C H r O' O' o o s o o CN t w o o - 4 w *« —* r » — r » «— —* r o , « • * CN o o H G | | i • • u • • > • »— . cn * N S om co N C N C o t —r u— 3 T H r - r O C —. .— H r H r H r H r H r N C to ' O O s - 4 - 4 H r O' N C ' O o r s rH - 4 w , O • o r O o __ 4 — V 1 •• • ■ • • • • - • • • • _ r ,

g X g in T N J M CN CM CJ O s o H r w O C o c n i o o H r s o o —* «— H r o O O O G M O' o w O' s o o O O CN CN - 4 o rH O O O o w w rH rH rH 4- I I I | 1 • r —*» u »— * r— . • H X Es< to * N s * N H t —*—' — * '— O C « r— . — H r O C - r m s H r in o *—<'*— to 14 oto to - 4 ' O — u

• •

•

g C X g J C i H m CJ cn cn cn CJ O C r~ o o ' O O C n i H r H r o CO H r to --- rH s s N C H r H r * 4 ' O H r o O o CN rH C o O o O s o o M O' o w O' w t + H I I I | 1 • • • ■H -P X* in H r — C in s H r O C n i r N O C o * N o * c * H r ro o O H r O C — - 4 ' O 03 M r—u l - U 3 T rH o O s O' O' - 4 o J U w CN o O o + | 1 "T • • • • • • N C o s ' O O C H r H r co s H r * N - 4 * 4 H t co o O P H l - U o o rH O © O s o * 4 —. r— _ J* H V 4- | i • • • •

H g X g P- s s E— N C N C N C J N CN CN CJ o o O C m o * *— H r oo in tor- in NO N H T-i T-i H r O O T—I O r—I L Cl O o O * •N s c S V4 S O O' O' rH O o o o o «—>» rH *4 4- 4- 4* I I I I I I 1 • • * —' *— ' x Ed to H r m O C * N o c —* *— Irt H CO to o c > - 4 O' CO c to O O s O' O' in rH rH l - U o l - U o ■>—* CN 1 • ■ • • • • r- ro s o H r 4 U j - * N n i * I— w - 4 co s r—. 4 U w in rH O O rH o —» r 1 o 1 N 1 • • •

1

-O' T3 ■O' co: CO CO > CO O o CD CO c O) jO X to •rH X TJ < > -< G X H XI XI •H 4 j j • • G o > CD X O' to to Or g X g Or to U G rtJ to > u to G CP ID to M 0 G to to O ro G a CD M g CU CD O' C 3 a> Or a >4 g ID cu M o 0 c J ) ) U X rH s a 03 4 • • 3 0 oG ro oG to ro M 03 CJ c DM (D 1 - ) P— » •H •iH Ed to X X rH T X X •H 03 X rH .. 3 M 3 C 13 01 to 01 M G 3 to 0 X Or - c M (D to a ID to to 0 U U M u a) c G CP a) U CU O c O M to CU X C CP O o G o c O G M o 3 ) ) > * rH X Oi rH X •—I •H •rH X X t-H 1 •— X X GJ o X X X X •a <— » X X X CO 3 G to G 14 iH O G U to G to CP CP O G c X i > U c 3 a 01 14 0 to 0 to 14 g CJ o 14 1 ) ) no X *“ * 1 •— G o in o —* •— r— • rH GJ O s/ u—I — V rH o O V G G M G G G G G 3 • >> • • • 150

rat strains activity was unaffected at 2 mg/kg ciprofibrate

and was significantly decreased at 20 mg/kg ciprofibrate. The decrease in activity varied between 45-7 3% of the control value. A dose-dependent reduction in enzyme activity was observed in the hamster although this was less marked than in the rat. In the mouse, activity was decreased at the high dose level (table 4.7). A non-significant reduction in monoamine oxidase activity was apparent at 20 mg/kg ciprofibrate in the guinea pig. No significant changes were seen in the rabbit. In the marmoset, activity increased in a dose-dependent manner and which was significant at the high dose level, 100 mg/kg.

In untreated animals, the highest turnover of kynuramine was observed in the Long Evans rat. The remaining rat strains, hamster, guinea pig and rabbit exhibited similar monoamine oxidase activity. The activity in the mouse was approximately

2-fold lower than that in the Fischer rat. The decrease in activity observed in the rat with ciprofibrate in this study was comparable with that resulting from clofibrate pretreatment (Gear et al.,1974)

4.4.2 Succinate Dehydrogenase Activity

In the rat the specific activity of succinate dehydrogenase was unchanged following ciprofibrate treatment (table 4.6). In untreated animals, the mouse and Gunn rat exhibited the highest activity whereas in the guinea pig and rabbit it was

4-fold lower. Activity of this enzyme was unaltered by ciprofibrate treatment in the mouse, hamster and guinea pig

(table 4.7). Succinate dehydrogenase activity was TABLE 4.7 The Effect of Ciprofibrate on the Mitochondrial Enzyme Activities In Different Species a Ed o X 4-4 a < CQ O X E X o a o 04 X X *-4 Q C Ed cu p w d w o o a p o s co h P a < El Ed < X *< O X O X Ed X o 04 X o Cu X < E-* Ed CQ 1 a u u ta o a o O 3 1 < - p O E-< 1— 4 *— X ta O o CQ t> Ed W > c a < CQ Q tad CJ *—l H E a a H e x O Q CJ ■—* S E-* t-t > N E-* tn x Ed CQ Q Ed X Qi < > a O Ed CQ O a c < Ed j i

C -s. C 03 s. iH P I P M rH r— i HH C l-H El U E l - > 3 4 0 C O P U s C s 03 h

p r-* » o C 4 0 a cn 6 0 h

. G 1—4 O E E P P P rH %_* w M M rH CMCM CM cn CM cn

m cm HrH 0 1 a in o X , o CP Cn O 3 W V _ * , cn X X oo o o • - •

'

-use P P rH rH Mr rH rH CM in J M CM CM CJ V0 o N O rH O o o o rH nXvo X in o o o * — » _ _ _ o rH rH r— —r rH s. 4- - 4 o w o o c cn cn p 1 1 • p V0 cn cn o X «H cn ■ 1 M CO — cn in N rH O m cn 1CM M1 . N o o o o , o rH 4* 4- 4* 03 E _ cn O P cn tn _ i 1 1 « • • m • • . _ - _ vo V0 O rH cn rH CO o m HiH rH —w'*•* ' w M •— X S. CO _ , CO V0 \ o O o o O rH - CM , -4- 4- H -4- 4- _ _ 4* _ 1 1 1 - . v .

_ _ - .

JC H JC P rH cn in in S. CM J M CM CM CJ rH rH o X

cn M* o m m o U-l M rH CM cn P V0 in X rH M ao ' M* X S. O in cn O o ■i— rH ■ ■ P o o 4* rH - ■ 4- 4* e — 1 1 1 . • « . . P £ £ £ -P m N H r CM VO tH CO ■M* HrH rH cn in in o o N J M CM CM CJ rH O o VO O •»a* o o \ CM O O Hr rHO rH rH o o o o rH 4- 4- - 4 o o c cn cn p S O N 1 1 1 • • * • • . • . • « « * • ■» u -4* ••U s a nin in in CO O cn \ S 04 •M1 rH o o o cn CM rH cn \ o o r—* . — O in O '— * 4- 3 03 + c 0) cn cn cn 1 1 1 • • p O H N cn o X \ in X V0 cn rH O N rH ro rH r H o o CM O O rH o 4- *—* 4- - 4 \ 1 1 • • . •

p CM \ H JC JC JC JC H tn o M m ■M* cn HOO rH «H CM ao M o rH X rH J M CM CM CJ CM rH rH o \ o rH O O o o O o —'—'*— ' '— '"— ' O rH 4- 4- 4- S ' M O o o G G» ' S P 1 1 1 . • . • • . 6 6 -O XI o CM in s Qi CM in s V0 o _ . ■M* •— . rH CM o r CO o O O O O rH rH s o -4- 4- . — 4- - 4 03 n cn cn cn 1 1 1 • # • • • . . • • • Ho rH CM OOOo O O VO X X ro w X X CM CM HT S N rH CM in rH rH o s o o ■— « rH o o o O rH tH 4* - 4 1 1 1 • • • s ■ \ in CM O £ £ p P P rH — rH o '— ' CM rH o o M rH o \ o cn rH J M rH CM CJ o O r—* «H O O HrH rH rH -4- 4- -4- 4- - 4 o o o o c en tr> p *s»O M, i l 1 • • «» «—* • • o o X X o CM X x rH CM V0 s rH cn cn \ rH rH m rH r n cn o o rH rH O r—* w M O O r~» rH O O 4* p 03 E O (0 03 cn cn 1 1 1 • • • • . . . m \ o m CM U-l r— rH tH X rH in N ,— , •*“» ‘— ' rH rH m CM 09 rH •-4 in *—* o o o o ■— ' iH s m m o o 4- 4- - 4 1 1 1 • • • • * *

-C 4) •H tH P •H P Q 03 JQ •Q h tO•h 10 03 ® « * P P p < < > 03 ^ 32 u re u 1 03 >1 03 03 X 3 e 3 03 03 C G cn 03 03 P G 03 to P P X 04 O U U re 03 > 0 X P 04 03 tQ tQ 10 03 03 O O Ot 03 03 P 4 — g C 3 03 03 04 03 P Cn V) 04 p o 0) to U 03 £ C 03 tQ 03 G P 03 4 03 rH P 'V r—I p P > 03 cn 3 P 03 p V) 03 03 03

H P • rH • x> e •H p rH 4 > &4 03 03 P p o> tn tn 03 C 03 3 h Cn rH 3 3 03 P 03 P 05 03 c cn G p o 03 CQ P > a to 03 X 04 p 0 to to 03 03 to o* p u U O P P 03 04 u o 3 Q 03 03 03 c 03 cn 03 o U] O G ) re 03 > p 03 03 3 P < cn© «

re S . © ■S w £ o Z_, 03 J2 -a _ LO ■v 3 X 03 X P 0 « 05 ■‘Of

o : C T3‘ D) 0

152

significantly decreased at the low dose level in the rabbit

yet values were similar to control at 20 mg/kg ciprofibrate. A

dose-dependent elevation in enzyme activity was observed in

the marmoset and this 1.5-fold induction was significant at

the high dose level.

In comparison to the present study, succinate dehydrogenase activity was decreased in the rat with DEHP and ME HP treatment yet activity was unaffected by clofibrate (Lake et al.,1984, Mangham et al.,1981). It has been reported that administration of ME HP and clofibrate to hamsters significantly decreased succinate dehydrogenase activity (Lake et al.,1984). In the present study a slight dose-dependent decrease in activity was observed in the ciprofibrate treated hamster although the statistical significance of this could not be determined. Succinate dehydrogenase activity was significantly decreased in the ferret following treatment with

DEHP (Lake et al.,1976). This enzyme has not previously been measured in the guinea pig, rabbit and marmoset.

4.4.3 Alpha-Glycerophosphate Dehydrogenase Activity

A significant dose-dependent increase in alpha-glycero- phosphate dehydrogenase activity was observed in the Sprague

Dawley, Wistar and Long Evans rat (table 4.6). Elevation in enzyme activity was apparent in the Gunn and Fischer strains and this was significant at both dose levels in the Gunn rat and at the high dose level in the Fischer rat. The induction in the rat varied from 1.5 to 2-fold in the Gunn and Fischer 153

strains to 4.5-fold in the Long Evans rat. The activity of this enzyme was unchanged following ciprofibrate treatment in the mouse, hamster, guinea pig and rabbit (table 4.7). In the marmoset, a dose-dependent increase in activity was observed and this 3-fold induction was significant at the high dose level. In untreated animals the mouse exhibited the highest enzyme activity which was 1.5 to 2-fold greater than the Long

Evans rat and the hamster. In the other rat strains and species, alpha-glycerophosphate dehydrogenase activity was 3.5 to 8-fold lower than that in the mouse.

It has been reported that the administration of clofibrate, gemfibrozil, tibric acid and fenofibrate results in the induction of alpha-glycerophosphate dehydrogenase in the rat

(Holloway et al.,1980, Kahonen and Ylikahri,1979). This was comparable to the effects observed with ciprofibrate. There is no published data regarding the influence of peroxisome proliferators on alpha-glycerophosphate dehydrogenase activity in the guinea pig, rabbit and marmoset.

4.5 MITOCHONDRIAL AND PEROXISOMAL CARNITINE

ACETYLTRANSFERASE ACTIVITY

In the rat there was a significant dose-dependent increase in carnitine acetyltransf erase activity (table 4.8). The induction ranged from 16-fold in the Gunn rat to 70-fold in the Fischer rat. In the mouse, hamster and rabbit, enzyme activity was elevated in a dose-dependant manner and the induction at 20 mg/kg ciprofibrate was 6-, 3- and 4-fold N1 E JQ r—t 0 CU h

The Effect of Ciprofibrate on Mitochondrial and Peroxisomal Carnitine 1 - 4 •H •H U-l p 4 4 •H -P -P > - 4 •H -P rH 4-> Q fa CO E < 0 p 0 g > >1 0 P 0 c in u d) P ro c to H fa E-» fa fa fa < fa E Q fa O fa CO > fa fa O c CO fa fa Qi co h h h h h < CO E fa h

03 p \ •H P •H P CU O u £ c g cn CU P O C 0 N ( H H o o CD cn H o h ir> UJ r—^ rH C" \ \ \ p P P r-H CN CN U + 0 u 03 0 o o c tn p tn o w J* 1 l 1 l . S S CD VD ■ ■ * - CD . CD O nCN cn Dcn CD 1 N rH rH in 1 N o + _ 3 c Cn m c • • .,— „ in o CN +

. .

i— i O O CN C- CD \ \ \ o o o CN rH p N CN CN U + o o G cn cnp w o 1 1 1 I • * g g I 03 03 in rH co CD o NrH CN o in *r O N* rH CN CO H- m QUO to p tocn £ i—3 i 0 i cn cn • • • ■ 4 CO Nin CN 1 N cn O

» *

O o o \ \ \ O p p p rH cn U CN CN CN U -++ + H- o o c \ tn p tn \ o I 1 I 1 • g g Nin CN Ho rH 03 03 r-^ CN NCO CN •— • •— >-— > rH cn .— » o CO CN •I— sz fa p cn cn to u 0 • • I O in in HP CD CN

■ • •

nO in \ \ \ CO hxa; o o r—* oo rH o 00 P NC o CD CN o N CN CN U H- o o p o c 1 1 1 I • • • • •

cn 00 m CN ncn in t— J 1N rH rH HP 03 03 H p H- g g cn cn cn cn to to p • 00 CO rH rH CM UJ CO C- H-

CO •

f-H p p o nCD in \ \ \ rH CN p o O rH CN CN U H- o o —- c p o 1 1 1 1 cn • Ooo CO *

*— 1o H^p rH O NH CN 1 N -o c- o 03 CN fa H* cn cn cn cn g g o w c > cn to • • CO CN CO co •— 11 03 Hp in H- c to

• •

O ~ CO. .E •O' 3 ^ *o •a O) 0 0 o co 0 CO CO >

.0 0 p 0 ,0 rH P 03 •H 4-1 p 03 rH •H JQ O rH P 0 P •H m a f U P O G P O o P O 0 p p tn CU 4-1O c tn U 0 0 QU C 0 p u 0 P G 030 tn 0 u 0 G X P u Q P 0 0 030 0 0 0 P U 0 3 0 0 P 0 > 0 3 3 CT> 3 rH 3 p 0 G 0 • • 0 • p

rH W p p 4-1 p 4-1 rH •H H•H CQ •iH o P •rH i i— tH P P CU o c p o 0 p o g 0 3 0 c 0 p 0 U o U 0 H C >1 V p • 3 0 G 0 P 0 030 tn v C 0 0 0 u o *-» p 0 0 P > 0 0 0 p 0 1 o O UJ —■ V— 03 03 0 0 • • 155

respectively (figure 4.5). The increase was significant at

both dose levels in the rabbit. Carnitine acetyltransferase

activities were unchanged in the guinea pig and marmoset. The

activity in untreated animals was lowest in the rat;in the

mouse and hamster the activity was 2.5 to 13-fold higher.

Enzyme activity was maximal in the untreated guinea pig and marmoset.

The administration of a variety of hypolipidaemic agents to the Wistar rat resulted in a marked induction in carnitine acetyltransf erase activity (Sharma et al.,1988). The extent of induction ranged between 5-fold with aspirin to 30-fold with nafenopin. However, in the present study a 36-fold induction was observed with ciprofibrate. In the Fischer rat a 70-fold increase in the activity of this enzyme occurred compared to the reported 58-fold increase following the dietary administration of ciprofibrate at a concentration of 0.1%

(w/w) (Lalwani et al.,1983b). It has been reported that MEHP,

DEHP, clofibrate and fenofibrate administration to hamsters caused a significant increase in carnitine acetyltransf erase activity in the order of 1.6-,2- and upwards of 10-fold, respectively ( Lake et al.,1984, Lake et al.,1986, Pourbaix et al.,1984 ). With ciprofibrate the induction, was 3-fold in the hamster. This enzyme has not been measured previously in the rabbit or marmoset. It has been reported that ciprofibrate induced this enzyme in cats, chickens, pigeons, Rhesus and

Cynomolgus monkeys when using extremely high dose levels

(Reddy et al.,1984). Hepatocytes from Duncan Hartley guinea pigs, treated in vitro with MEHP showed unchanged carnitine 156

Figure 4.5

The Effect of Ciprofibrate on Mitochondrial and Peroxisomal Carnitine Acetyltransferase Activity in Different Species.

LE8END

7000, CONTROL LOf DOSE 6000 e pj Hi high dose 5000 CD I r OC, E—■ 2: 4000 CD CD l l 3000 Lu CD si 2000. t t

1000 .

0 j M n r - S HOUSE HAHS7ER RAT F-344 G-PIG RABBIT HARH0SE7 ( C O NTROL = 100

SPECIES 2 WEEK TREATMENT

Absolute Values (control);

Species nmol product formed/min/mg protein

Mouse . 4.30 Hamster 6.47. +/- 0.93 Rat (F-344) [a] 0.49 +/- 0.20 Guinea pig 9.52 +/- 2.65 Rabbit 14.01 +/- 2.15 Marmoset 11.82 +/- 1.49

[a] data reproduced from table 4.8.

Dose Levels:

Low Dose : Mouse, Hamster, Rat, Guinea pig and Rabbit 2mg/kg Marmoset 20mg/kg

High Dose: Mouse, Hamster, Rat, Guinea pig and Rabbit 20mg/kg Marmoset lOOmg/kg

Single daily dose 157

acetyltransf erase activity (Lake et al.,1986).

4.6 PEROXISOMAL ENZYME ACTIVITIES

4.6.1 Catalase Activity.

In the Sprague Dawley and Wistar rats, the specific activity of catalase was significantly increased at the high dose level

(table 4.9). In the Gunn, Fischer and Long Evans rats, mouse hamster the specific activity of catalase was unchanged following ciprofibrate treatment (tables 4.9 and 4.10). Enzyme activity was unaltered in the guinea pig and marmoset. However at 20 mg/kg ciprofibrate (i.e. the high dose) in the rabbit, activity was significantly increased by 1.5-fold. Catalase activity in the untreated marmoset was 2 to 3-fold lower than the basal level in the other species.

By comparison, in the rat and mouse clobuzarit treatment resulted in a 2-fold increase in catalase activity; activities were unchanged in the hamster and marmoset (Orton et al.,1984). A similar increase was seen with ciprofibrate in the rat (Lalwani et al.,1983b). It should be noted that in this study the liver/body weight ratio (tables 4.1 and 4.2) was doubled in the rat and mouse with a significant increase at the high dose level in the hamster. Although the specific activity of catalase was largely unchanged, the total catalase activity would be increased in line with the observed hepatomegaly. According to the published data, ciprofibrate at high dose levels increased catalase activities in cats, ■M*

The Effect of Ciprofibrate on Peroxisomal Enzyme Activities in Different Rat Strains Ed OS o X l-H 03 O 5 3 CQ Ed E < I h E- -n y-* s E* < P 44 P < P u c CJ g < co -P Edto O C r-H O 03> t—l T3 HH >■ - i o i Q ref. ref. EH E l-H £-1 X Z § < • CU — HH rl > V 44 c l-H E" HH Ed M u c o < w 03 i O M O | - h E tn J Q Ed O > 03 < < E OS OS , rO h h in —I I— M. M. s. •Q rH rH 4_» ifH u3 z 03 E CU c cn c ) L s g g g )H O CU C c cn e s n h

44 •rH c g O

CD to s CM rH rH ,_u o O CM s o CO , CM c~- CO CM •M* tn to o , c— o oo rH tp 1 1 MP CM O O to rH o . . cn _ u Mu Mu CM Mu O H- H- 4- _ _ _ cn cn cn 3 c 3 1 1 1 m . _ _ o s ~ r u M to CO s c*- M* r— rH rH CO , Hl-H «4H rH CM nrH cn m rH CO tn O p p p o _ CM CO 4- H- + _ _ ct» I 1 1 • • • . _

o s CO o o rs CM CM o H - r tn s o in C~- to o Mto CM to rH O'! rH * rH rH H o r t rH - r n t n t - - — _ 44 p s rH cn rH o O i-H CJ -4 4- 4- 4- H- 4- _ o w o a 1 I 1 • • • • • • 03 • • * • « » _ H co in co s s ’ w CO rH to s m w •M* U-l U-l to in CO in .— .. c— co in - . UH CM CM CM O Mu Mu P o o *to M* CO o o CM 40 X p p 03 -H- H- H* — — cn g g S» cn cn cn CU 3 >H a) CU Q M ro ns > i I 1 • • • . , 1 to r*- HH rH o r- CM to C~* CM cn o ■— cn - UH U-l i i -— CM u M s _, to s rH cn to _ o rH uu H- _ — _ cu 1 1 1 • • • * • » ' • ' _ _

s CM CO w rH r—* o O <3 P in OO CM •M* o o rH O CO CO C— rH tn rH tO CM r- O Mu o o "" — tn CO O O r-H 44 i-H cn o P CJ 4- H- H- c U o o 1 ■ ■ i ■ 1 i • • * P~ OO rH r-H co s s CM to P s rH rH cn —— rH to . — P r- P rH co rH rH ->H u-i OO CM CM CM J3 Mu S CO 4- H- H- — g g g tn tn cn cn CQ u u c M i 1 I • '— ' - . cn tn oo o ■*-' o . — r OO CM o in CO U-l r—. - r-H P o CM co to Mu s o o o w c~- O0 r-H P P cn P cn P H- 4- -H- H- 1 1 i • • • P rH S cn w rH cn tn u M s o rH o CM CO O O0 o cn tn co in co c— rH cn tn rH » — o O — •u g g •ug .SC rH O HH M C M C J C 4- H- s tn w cn o w o o 1 1 • • • •— ' ' -H s CM cn C'* ■M* in CM 03 r-, HT CO tn ■— 1 | 1 • • • • * • • • '— ' ' * *•— • nr r- rH

to CM H - r “ tor“* rH rH rH M' o c o c o in s o p — s s o co CM o m c rH OO r-H o CM to O s '— o to to p O o o g g g P r-H nto in p M C M C J C H- -H- H- o o S tn S w tn O s 1 1 • • • « ' ‘—' ' o c s cn T3 _ u-l o tn «— » rH CO OCO CO o oo rH L^i cn to HH • — r ■ to to m CM Mu '' CM o rH rH MCM CM . Ed 1-3 - 4 4- a o cn cn s cn > s CQ 10 1 i * • • • • P r— • CO CO o cn cn P S to CM (Q cn to o p in to u M o to P CO P u—* * — P a H- 4- 4- 1 1 • • ' • :

1 CO : r r CU \(/)[ ■O ■O' TJ c O) O o <0 CO >1 CQ

rq U jQ •rH 44 i-H 403 44 •rH HH r-H •o •H T3 < 44 o 03 g -a g 03 *0 > ip nj >*« o o c o cu g o t*l CU 0) CQ to CQ o X CO CQrH roO 0) to 1

Mu Q 4- to M cu

p •rl 44 HH 'O 44 4P 44 —H i-H i ro •iH tu M 3 g cn U u M o U O CQ dl 44 O c O o C CQ 8 M a) to 8 0) M u a) to 3 cn 0) 8 O 0) > 3 0) X 3 M U CQ Q U M CQ CU Oj 'O QJ 44 r-H HH 44 •rH r-H 44 44 44 •H 03 •H 44 HH UH 44 •U—r 4rH U4 UH 44 03 3 o H to 3 U to ><44 CU CQ 3 M iQ tn o 0) CQ o M a) CQ 1 h o w 1—H T3 u-r o o P O r—* V rV rH o o in-r m V v* to 3 0) a) • • • • H

rH JQ E ro 0 h

The Effect of Ciprofibrate on Peroxisomal Enzyme A ctivities In Different Species '63 o H co Cu 63 CO 63 O co 63 63 > J Q c 5 1 u < c Eh < £ j 1 : i i 44 >4 l-H O E E-< > D a u < w < co M H - h h h i £

J • rH T3 ‘H \ — a)

tQ0

CU M CP 04 e w G cu O

— r rH i—I rH cn co oo J N N N CN CN u CN CN CJ JQ N / H H i—I cn O i—I O i—1 H H O i/l O O O (N O N N N* CNCN o o ro ro ro rH rH LO o cn co ^ n o cn o O H n rH cn rH 0- N1 44 .X s O 3 10 QJ £ P CP CP n CP cn l 6 O

' - ' rH - i ' O O 0 LO 00 i - T N N CN CNCN co LO r-~ \ \ N O N CN CNCO H N* O LO CNCO o o o o - r o ro o ro cn uo co CN ro ro ro ro o o n - r cn ro \ \ s + + + - - + 4- 4- I I I i i i f 44 \ \ 44 44 tr cn cn 0 e 10 OJ M s £ P CP CP CP CP H o cn h

4-5 i—1 CN o o LO N NO C C J N CN CN CJ CN CN O CN CN U \ ncn cn o rH ■ o o o . CO _ i—1 i—1 LO H ro O H O CN \ \ \ CN 4- + o cn co H —' " - ' O _ G \ o o u + i 1 1 1 . • • • • • • * . • . . _ ..••• • • . . - . • • • • . • . • . . . +J OS 444 44 H H * r \ o H OHO CO H CO — .— ,—. o C" H H LO H H fO £ CP CP + 1 . ro N1 64 N* 03 ,— , o \ ro CO o \ o cn —, H CO LO H- CO H LO e CP Cn 4- I 1 .

44 H oro ro cn H H \ O N1 t"- O H > O O H H o n co o LO \ LO C"- LO H + o o —' H n ao cn H — c C O O H _ 4- i I 1 1 • • • • • • . • * . *H • O \ Cu 44 0 \ 0 CN cn cn v ,—. oCN ro CN 0 oro ro H 3 + cn G QJ rO H CP g £ CP CP P CP CP + _ 1 » ■w —»• ^

\ O 44 ro ro 0 \ \ 0 0 cn cn O - c <-N CO - r H - r 0 + + + + ■+* 1 44 rH 0 to CO CN O 1 LO O LO LO LO + + n cn cn o o —' o C" H o H co H 0 CP V-l O — I 1 • . 1 43 JQ ►H OS 44 N \ 444 44 ro 0 \ co 0 0 H ro 0 nH cn LO CO r-" cn LO CO \ O 1 N 0rO (0 S PCP CP 1 1 1 O O H ro •*3* \ OO CO cn l ,—■ CP O cn 1 l C" + ro \ CN <—1 1—1 g CP 4- —1 —1 0 0 O l 1 •

44 H NCO CN 0 NH CN O m \ 0 0 ro ro 0 H O 44 44 H- Q QJQ JQ JQ f—I 44 z z z Q C O H ro 0 O O 0 U 0 s S 44 \ 0 0 444 44 cn LO \ 1 0 •—> 44 44 + H 44 M S 0 w Q 0 —i £ P CP CP CP 0 u (0 0 0 1 0 0 co \ 0 00 O H 0 LO cn —^ •—> 44 41 44 + r-H 44 Q s CP 0 U ro 0 O 0 1 • *—* H ■— ■ > TJ 44 44 44 •H rH rH 44 rO • • 3 3 3 0 10 rO ru V4 U Cu G 0 JQ 0 X to 0 0 10 10 rO ro ro c rO > W 0 U X CP 0 rO O 0 TJ ? 0 X Q 0 C £ 0 u 0 u 0 10 3 n j 44 — JQ 1—1 •H •H LXj 44 44 rH 44 TJ •H 44 •• PH44 CPH to 0 M 0 03 G U to w u 0 —l 0 to ro > 3 Q u 0 0 0 to 10 fO Q to u 0 c W 0 H G CP 0 O 0 M W 0 10 Q O G U O c u a H 0 O j j j ■— > \ CQ Q 4* u ••»• »0 >4 >4 10 10 0 £ rO to 0 1 1 v— 03 1 03 CU Q 44 44 Q0 JQ cn rH —> 43 >4 ro Q 10 0 0 M 3 U O >4 0 44 O 0 rO £ • 1 j H 44 H 44 •H •H •H 1 •H 44 44 44 44 TJ n3 44 44 44 44 44 .—• 44 1 CQ 44 —1 0 > to O —1 M 0 to 3 to 0 M 10 G 0 U CP G >1 M 0 G 3 0 0 to w to 0 £ 0 U O G O M I 0 44 r-. c LO O V 0 0 r*“ 0 0 0 : * \ i_ ■o' *♦— r— 03 T5 CO. > O 0 0 0 O W 0 0 > C D) 160

chickens, pigeons and Rhesus and Cynomolgus monkeys (Reddy et

al.,1984). Catalase activity was significantly decreased in

the ferret following DEHP administration (Lake et al.,1976).

The effect of peroxisome proliferators on catalase levels in

the guinea pig and rabbit has not previously been reported.

4.6.2 Peroxisomal Beta-Oxidation

In the rat, ciprofibrate treatment resulted in a significant

dose-dependent increase (p<0.001) in peroxisomal beta-

oxidation (table 4.9 ). This induction was 9 to 15-fold in the

Gunn, Sprague Dawley, Wistar and Fischer strains. In the Long 1 Evans rat, a 34-fold increase was observed. In the mouse, the response was dose-dependent with a 45-fold increase at the high dose level (figure 4.6). Peroxisomal beta-oxidation was elevated approximately 2.5-fold at the 20 mg/kg ciprofibrate dose level in the hamster. There was no change in the beta-oxidation of palmitoyl-CoA in the guinea pig. In the rabbit, beta-oxidation was significantly increased in a dose response manner and the induction was 6-fold at the high dose level. At 20 mg/kg ciprofibrate, the increase in peroxisomal beta-oxidation was 2 fold in the marmoset and at the high dose level (100 mg/kg), a 4-fold induction was observed. This was significant at both dose levels. In the uninduced state, peroxisomal beta-oxidation was lowest in the guinea pig and marmoset. The highest activity was exhibited by the hamster which was 3 to 4-fold greater than the rat, mouse and rabbit.

It has previously been reported that the administration of a 161

Figure 4.6.

The Effect of Ciprofibrate on Peroxisomal beta-Qxidation in

Different Species

PEROXISOMAL B-OXIDATION LEGEND

5 0 0 0 CONTROL LOf DOSE HIGH DOSE 4 0 0 0 —3 CD O' £— 3 0 0 0 CD CD Lu 2000 CD

1000

MICE HAMSTER RAT(F334) G.PI6 RABBIT MARMOSET (CONTROL = 100)

SPECIES 2 WEEK TREATMENT

Absolute Values (control);

Species nmol product formed/ min/mg protein

Mouse 2.89 Hamster 13.86 + /- 3.89 Rat (F-344) [a] 3.54 +/- 1.48 Guinea pig 0.97 + /- 0.32 Rabbit 2.33 +/- 0.85 Marmoset 1.23 + /- 0.53

[a] Data reproduced from table 4.9.

Dose Levels ;

Low Dose: Mouse, Hamster, Rat, Guinea pig and Rabbit 2mg/kg Marmoset: 20mg/kg

High Dose: Mouse, Hamster, Rat, Guinea pig and Rabbit 20mg/kg Marmoset: lOOmg/kg

Single daily dose 162

wide range of peroxisome proliferators to Wistar rats caused a

significant induction of the peroxisomal beta-oxidation

system. The extent of induction varied from 2.2-fold with MEHP

to 5.4-fold with nafenopin (Sharma et al.,1988). In the

current study ciprofibrate administration resulted in an

11-fold increase in peroxisomal beta-oxidation in the Wistar

rat. A 12-fold induction was observed in the Fischer rat which

was comparable to the 14-fold increase reported with the

dietary administration of ciprofibrate at a concentration of

0.10% (w/w)(Lalwani et al.,1983a). With clobuzarit treatment,

a 5-fold increase in peroxisomal beta-oxidation was seen in

the rat and a 9-fold induction in the mouse (Orton et

al.,1984). With clofibric acid administration, a 9 to 14-fold

increase was observed in the mouse and a 7-fold induction with

DEHP (Kawashima et al.,1983, Osumi and Hashimoto,1978). A

significant increase in cyanide-insensitive palmitoyl-CoA beta-oxidation was exhibited by the hamster with clobuzarit treatment; however beta-oxidation was unaffected in the marmoset (Orton et al.,1984). Palmitoyl-CoA oxidation in the hamster was increased 1.5 to 2-fold by MEHP, clofibrate and fenofibrate (Lake et al.,1984, Pourbaix et al.,1984). The induction of peroxisomal beta-oxidation was more marked in

MEHP treated Chinese hamsters than in treated Syrian hamsters

(Lake et al.,1986a). With DEHP and clofibric acid administration, the peroxisomal beta-oxidation system was unchanged in the guinea pig (Kawashima et al.,1983, Osumi and

Hashimoto,1978) The effect of MEHP and one active metabolite of MEHP on peroxisomal beta-oxidation in cultured guinea pig and marmoset hepatocytes was examined. No stimulation of 163 peroxisomal beta-oxidation was observed in the guinea pig hepatocytes with either compound. Marmoset cultures occasionally showed small increases in palmitoyl beta- oxidation but these were not dose-dependent with MEHP or its metabolite (Elcombe and Mitchell, 1986). The culture of hamster and Duncan Hartley guinea pig hepatocytes with MEHP had no effect on peroxisomal beta-oxidation (Lake et al.,1986).

Administration of clofibrate to Rhesus monkeys and marmosets did not alter the activity of peroxisomal fatty acid oxidising enzymes (Orton et al.,1984). With high dose levels of ciprofibrate, increased palmitoyl-beta-oxidation has been reported in cats, chickens, pigeons and Rhesus and Cynomolgus monkeys (Reddy et al.,1984). In this study the induction of the peroxisomal beta-oxidation system by ciprofibrate was observed in the rat, mouse, hamster and rabbit with a minimal induction in the marmoset. Peroxisomal beta-oxidation was unchanged in the guinea pig by ciprofibrate treatment.

4.6.3 Uricase Activity

In the Gunn, Sprague Dawley and Fischer rat strains, uricase activity was unchanged by ciprofibrate treatment (table 4.9).

A significant decrease (p<0.05) in enzyme activity was observed at 2 mg/kg ciprofibrate in the Wistar and Long Evans rat. However, values returned to normal (i.e. control activity) at the high dose level. Uricase activity was unchanged in the mouse and guinea pig (table 4.10). A marked dose-dependent decrease in enzyme activity was seen in the hamster; at the high dose level activity was half of the 164

control value. Ciprofibrate treatment resulted in a

significant increase in enzyme activity at the high dose level

in the rabbit. Uricase activity was determined in the marmoset

homogenate samples and was found to be approximately 100-fold

lower than in the rat. The standard deviation values were very

high and in the control were greater than the mean (values not

shown). This minimal activity probably exceeded the level of detection and limits of sensitivity of this assay. No apparent treatment-dependent changes could be determined. Man and certain non-human primates, including several genera of New

World monkeys are devoid of uricase (Christen et al.,1970).

The marmoset is a New World monkey and on the basis of the values obtained (i.e. the high standard deviations) and the very low activity, uricase is postulated to be absent in the marmoset. Uricase activity was reported to be elevated in the cat and Rhesus monkey (Old World) by ciprofibrate administration (Reddy et al.,1984). This has not been substantiated by other published reports.

4.7 CYTOSOLIC ENZYME ACTIVITIES

Two major classes of glutathione peroxidases (GPX) exist. One type contains selenium in the form of covalently bound selenocysteine in the active site. This selenium-dependent enzyme is active with both organic hydroperoxides and hydrogen peroxide. The second type is not dependent on selenium for catalysis and displays negligible activity with hydrogen peroxide. This class constitutes the glutathione-S- transferases (Mannervik,1985). In this context, hydrogen 165

peroxide is a substrate specific for the selenium dependent

GPX, however t-butylhydroperoxide and cumene hydroperoxide can

also be used by the selenium dependent form and the selenium

independent glutathione-S-transf erases (GST) (Lawrence and

Burk,1976, Lawrence et al.,1978). A measure of GST activity

can be determined as the difference between the enzyme

activity with cumene hydroperoxide and hydrogen peroxide

(Lawrence et al.,1976, Mohandas et al.,1984). In this study,

the major interest in the cytosolic enzymes was to examine the

effect of ciprofibrate on glutathione peroxidase which

functions as a biological antioxidant.

The activity of GPX was measured using two substrates,

hydrogen peroxide and t-butylhydroperoxide. In the rat, a

dose-dependent decrease in enzyme activity was observed with

both substrates (table 4.11). In each instance the activity

with hydrogen peroxide was twice that with

t-butylhydroperoxide. It should be noted that the enzyme

activity, expressed as a percentage of control, was very

similar with both substrates (+/-6%), with the exception of

the Fischer rat in the low dose group. With the rat, the

values were derived from one cytosolic pool per dose group. At

the high dose level activity was 40-65% of the control value.

This decrease in enzyme activity was marked in the Sprague

Dawley and Wistar rat. The Gunn rat was the least responsive

rat strain. GPX activity in the untreated rat shows minimal

strain variation.

GPX/GST activity was measured in the Sprague Dawley and Gunn O p-n o P-H P-H o p-h o P-H P—S <—s P H 0 O If) VO O VO r- o 01 ro o Ol C^ CN rH rH O' vo H lO M1 rH CO VO rllOM< in 0 w w w h»p H»p —'*— P P~N JJ >1 © £ JJ £ © JJ T3 O •H © ©J © © U > p—. O' *H P rH •H c O X JJ o O' -p •rH P O © o £ u SH © 'd P JQ •H < E-< JJ>1 © 3 M o X Gu © © £ © > p w p 0 CO M Qj o in CO in o 01 CNCN CO C" N1 O vo in © Qj © E-« o <71 CO CNC"-in 01 COin ro Ol vo VO > © • •• • • 0 0 *o o O' • • •• •• • © -p P rH p 1 M JQ H © JJ © o JJ 1 s © X jj o Qi CO 1 Ir* JJ rH o © u 1 C © >1 P P © © •H JJ 1 g jj © JJ rH © rH © 1 D 3 Q j © © o Q u 1 ij rH X OX g ©J © 1 O o 1 P 3 H © Q JJ 1 g JJ rn © £ © JJ C 1 3 >1 W © © © >1 © 1 > w ' X >— 1 © u P 1 c** r - <71 o CNc- rH N1<71 m N* m CN M1 CN g P © 1 n * CN VO co CN in r o CN VO CN •MTro cn O Qj c JJ 1 • • • • • • • • • • • • • • P X o JJ 1 o o O o o o o o O o O o o o JJ © © w © JJ rH © © c o 3 P w © rH X o © •H JJ • > JJ o U Qi © p &\ cn O' O' O' O' O' O' O' O' 3 P CO a H J i Jk! H AJ rH M J X J»C J»C © O © > •H o w o s \ o w o w O V\ -£ P CO o ta P O' O' P O' O' P O' O' P O' O' P O' O' JJ O' W ■O w jj g g JJ g g j j g g 4J g g JJ£ £ JJ jj o £ c c c c JJ © © o a o o o o o o o o o o O © X U CN CN U CM CN u CM CM U CN CN CN CN O U -p w • v © O- u © P «4— a) 3 P X <4-1 rH © © (44 © Qj £ w > •H o W H •O rH © © o © H X O' o © E-* © Qj P © 0) >1 P P 3 z a a a) p © © rH O' ■O' 01 t—I O' rH x © £ > ^ •H © iH E* »< c © 15 u JJ O' © < Cx< c (0 CO £ > O) XI < OS P r0 0 0 • « c © a: E-* 3 QiQ •H O H E-* co u CO &J 55 J © X w 167

rat using cumene hydroperoxide and hydrogen peroxide (table

4.12 ). The difference in activity between the two substrates was attributed to GST activity. Enzyme activity was greater with cumene hydroperoxide than with the other two substrates

(hydrogen peroxide and t-butylhydroperoxide). There was little variation (+/-8%) in GPX activity when expressed as percentage of control with these 3 substrates. GST activity was decreased in a dose-dependent manner in both rat strains and at the high dose level this was 35-50% of the control value. In the mouse, a dose-dependent decrease in GPX activity was observed and at

20 mg/kg ciprofibrate activity fell to 55-60% of the control value (table 4.13). In the hamster GPX activity was unchanged on treatment however a slight increase was seen at the high dose level. With these two latter species there were no substrate differences. Activity in the guinea pig was unaffected by ciprofibrate treatment. There were no significant changes in GPX activity in the rabbit. No significant changes were apparent in the marmoset but there appeared to be an increase at both dose levels. In the untreated animal, the guinea pig and marmoset exhibited the lowest GPX activity and the highest hydroperoxide turnover was in the rabbit, mouse and hamster.

On the basis of the minimal substrate differences between hydrogen peroxide and t-butylhydroperoxide and with reference to the substrate instability of the former, t-butylhydro­ peroxide would appear to be a suitable substrate for measuring selenium-dependent GPX. It has previously been reported that clofibrate administration to rats did not decrease GPX Table 4.12 JS 44 44 44 44 •H 44 l r • 4Q 44 rH •H 44 44 •H 44 £- Cd o O i > Cd < 44 •C C5 I—I u i Q 01 01 O M O g i > U to a • CL) o o to o U C N i > e QJ U > >1 3 c G Cl a rH 44 crT Q OS 10 c \. r M 10 CP 3 01 10 01 10 :* OHB- OS < i-3 O o O Cd Cd O > CO pel Cd cd ss Cd Ou to Cd o D 1-4 > CJ >-3 < E-* X 44 h4 >1 E X HH Q S CO i Q CJ E 1-4 i E ds Ed < co O OS OS O U S3 E z z << CO Pd < s h I Cd Cd I i h h < JQ s—1 co 4 r no no \ —4 44 —• i-3 *H CO Cb< Cd a S 01 10 o o s i a 01 e c 14 O CP C

43 -P 44 rH 3Qi Q 33 U c s c O -h C E-* s t-H > CP I H S E-* i Q X 3 >1 >1 M X o CP 01 c 3 £ 01 o c 01 1 01 •H to 43 *H TO -—1 -Q ra H X 44 TO M O i Q 01 14 o X 01 «0 14 o 01 >4 O 14 i Q 01 M X ai O 3 14

4-1 rH o r— . o o r—» o J M CM CM CJ rH o o UJ O «—» rH ■—* -—• —- o rH rH CM OG ID GO GO rH o o O *—• rH O O O O o o rH CM r CM o c • •— • rH UD CM •— O o o o C 14 O u • • • .

J* o o c— HT «—•■ rH i a © nID in i a m r—* iH •«— i O CM 0 CO Q0 s c C P CP CP P CP CP e e • • X -X MID CM i a «—* ID ID o O CM .—. ID ID C— o rH n i <—» v-r o i a . • . -use 4X l - r O no in O O Hin rH CM J M CM CM CJ o o rH O — » *—» rH o O GO n i — r Ho© o «H rH ——r *—• *— o —*r U5 H o o o rH rH CM ro CD CD — r » O rH — O O N N O o o c u • • • • • • • • • cp cp cp cp ro - c o o CO CM o r C^ ID m •—» 9 0) i > ) 0 P CP CP . . . JA CM r- n i i a m - P .—- *—* CM •M* co • • •

43 r—. OEi 44 O -a rH O3 TO 44 44 TO rH rH 3 3 a ro 10 M 3 • • 10 Qi 14 O 10 3 3 3 3 U 3 o 14 e • »H i— JQ 44 rH 44 Cti TO r—1 4rH 44 TO • r4 44 44 u C 14 3 U 3 3 14 > X tP ## 3 14 3 3 to 3 3 Qi 14 3 to 10 o 14 M ai to to i Q C i Q 9 M u 3 c 10 CP 01 to 3 10 u o O O 3 CP o (0 u c 14 o . 43 rH 44 •rH .c T3 44 44 rH rH TO *-«r rH E TO 3 9 U 14 o 6 3 G 3 O 2 3 o >1 £ 3 3 u 3 CP 14 o i Q o o 3 O i Q 14 O 3 CP 3 i Q O i Q » 3 14 3 3 3 i Q O 14 h • * ' co ■o !G O' "O ■o C 0 O) O 3 0 ° 3 > 3 > (0 n i—I JQ E" 0 0 1

The Effect of Ciprofibrate on Cytosolic Glutathione Peroxidase Activity U-l •H 4-1 jj H r • n a q c 0 c p 0 c Qi 0 0 0 U D E <3 X t—Io O X O J E-* S &q CU « Qi w Q < CO o > < E 1— 1 W JJ fH h h w Q la \ HH u CQ Q CO O JQ rH JQ rH r?3 •iH o >JJ !> co ta rH _J \ .— „ 0 0 0 H B J rH JJ 1 £ O O B £ r JJ X rrs Cn 6 p i Q o c 0 j JJ ta Cd 03 . X 3 c cn o P 0

• rH J3 r0 rrl •H O. P p 0 o p tn i Q O 0 0 X o 0 P

JJ JQ JJ JJ ,Q 0 3 0 0 0 0 0 P 3 10 rH jQ co C" N C CN O CN IN n rH i—1 O ill O i— CO cn cn __ o o , 1— 1 _ o c w o 0 p • • , , . , r~~ Or Hr- r rH rH CO CD 2 CO rH _ NCN CN NN1 N CN N1 LO CO o g 0 0 B B cn n cn cn _ O o J VO CN CO rH __ i— 1 ID rH . np cn _ k C „ . . JJ rH \ O 1 rH o 1 N o o cn CN , __ CN cn o 1 N \ \ o o rH rH rH * O LO - J _ _ + G o \ o _ _ u I 1 • • • • • , 1 _ , ^ „ JJ 00 o - r \ X o __ rH CO JJ JJ *«H JJ JJ —1 0 * * 0 3 3 3 0 P 0 0 0 0 0 0 0 cn > 0 0 0 0 CU 0 O P p 0 0 P 0 0 0 i Q B 0 G 0 0 0 •H QUT3 U JQ • rH 1 x G Jxi JQ J+ JJ o > U0 U > fo h rH MH j j •H — JJ 03 • JJ rH cn P 0 0 P u 0 0 G 0 0 0 0 > 0 P u X 0 p 0 3 0 0 Q 0 0 0 p O cn 0 c Q 0 0 u u o 0 0 CU G O o cn P o G P P 0 G 3 • ' —• r— 1 N a CQ —1 >1 0 • * 0 0 p 0 0 G 0 0 B 0 1 r—1 1 T3 •H rH MH rH «JJ rH rH 03 JQ rH —1 * * p B 0 CU Q O 0 3 Q U P o 0 3 o 0 O O B 0 0 0 o 0 O P 0 n -H cn £ 0 0 O 0 tn Q O' o o 0 P O 3 Q ' 0 03 03 JQ JJ JJ cp * N rH rH p 0 0 P O U P Q 0 o B 0 • ’ • ~ T - 3 0 co: 03 ' 3 0 N" — H 0 CO 0 0 CO 0 C o o >1 ) D r > 170

activity (Antonenkov et al.,1987), yet activity was increased

by MCPA, a weak peroxisome proliferator (Hietanen et

al.,1985). It has been reported that both clofibrate and

fenofibrate decreased GPX activity to 75-80% of the control

value ( Ciriolo et al.,1982, 1984). With ciprofibrate in this

investigation, the decrease in enzyme activity was more marked

(40-65%). This was similar to the nafenopin and DEHP mediated

decrease in GPX activity which was 40-60% of the control value

(Tomaszewski et al.,198.6, Furukawa et al.,1985). The influence

of peroxisome proliferators on GPX activity has not previously

been examined in other species. The distribution of

non-selenium and selenium dependent GPX activity has been

studied in several species. Non-selenium GPX activity ranged

from 35%, 43% and 100% in the rat, hamster and guinea pig

(Lawrence and Burk, 1978). This would infer that activity with

hydrogen peroxide would be negligible in the guinea pig which

is in contrast to the present study where similar activity was

detected using either t-butylhydroperoxide or hydrogen

peroxide. The reason for this is unclear. The livers from each

species were perfused prior to homogenisation to remove

residual blood. Red blood cells and presumably plasma, contain

only the selenium-dependent GPX so contamination could occur.

However in the mouse, hamster, rat and guinea pig the tissue was well perfused so that contamination with selenium- dependent GPX from red blood cells was unlikely.

Clofibrate, fenofibrate, nafenopin, tiadenol and tibric acid administration resulted in decreased glutathione-S-transferase activity in the rat (Foliot et al.,1984, 1986). Ciprofibrate 171

noncompetitively inhibited glutathione-S-transferase in rat

liver, in vivo and in vitro (Awasthi et al.,1984). This was

substantiated in part by the observed decrease in GST activity

in the Gunn and Sprague Dawley rat with ciprofibrate in this

investigation. It is apparent that ciprofibrate, in common

with other peroxisome proliferators, decreases the activity of

cytosolic GPX and glutathione-S-transf erase in the rat.

4.8 THE MOLECULAR BIOLOGY OF CYTOCHROME P-452 AND

GLUTATHIONE PEROXIDASE

The administration of a wide range of peroxisome prolif erators

to the rat results in the induction of cytochrome P-452, as

measured catalytically using lauric acid as the substrate and

immunochemically using an ELISA technique (Sharma et

al.,1988). During this investigation it became apparent that

ciprofibrate, in addition to inducing cytochrome P-452 caused

a decrease in cytosolic GPX activity as determined

catalytically. Peroxisome prolif erators cause a marked

increase in the peroxisomal beta-oxidation system with the

concomitant generation of hydrogen peroxide. The decrease in

GPX activity was of interest as this enzyme functions as a biological antioxidant. The possible implications of this reduced defence against "oxygen radicals" in the ultimate toxicity, hepatocarcinogenesis, is unclear at present.

The level of cytochrome P-452 was elevated following ciprofibrate treatment and this may be due to an overall enhanced synthesis of the protein as opposed to alterations to 172

catalytic activity and/or protein degradation. The decrease in

GPX activity may result from in vivo inhibition, protein

degradation or non-expression of this gene product. It has been established that Wy-14,643 administration results in a marked increase in the concentration of mRNA species coding

for 3-ketoacyl-CoA thiolase and enoyl-CoA hydratase

(C hatter jee et al.,1983). Thus, to clarify this ciprofibrate- dependent modulation of GPX and cytochrome P-452 activity, it was decided to adopt a molecular approach to study the expression of these two genes. A cDNA probe to cytochrome

P-452 and a genomic probe to GPX are currently available in our laboratory and these were used in hybridisation experiments with hepatic isolated RNA. It was hoped that by studying the gene expression in various species, differences in susceptibility to peroxisome prolif erators could be rationalised. In preliminary experiments, an attempt to optimise the yield of RNA from the biological samples prior to hybridisation with the radiolabelled probe, was undertaken.

ISOLATION OF TOTAL RNA

In the cell, RNA exists mostly in ribonucleoprotein complexes

(ribosomal subunits,hn RNPs [histone ribonucleoproteins] and mRNPs) which further interact with extraneous proteins subsequent to cell disruption. The essence of RNA isolation is therefore highly dependent on a protein denaturation step in conjunction with extensive ribonuclease inhibition. To achieve this the method of Chirgwin et al.,(1979) was utilised as outlined in section 2.9.. Guanidinium isothiocyanate and 173

guanidine hydrochloride extraction was favoured as both

inhibit ribonuclease and are potent chaotropic agents

(Cox,1968). Reducing agents such as 2-mercaptoethanol and

dithiothreitol, acting as disulphide bound reductants, were

also included in the isolation procedure.

In general, RNA is isolated from tissue, which is excised as rapidly as possible and homogenised in the presence of guanidinium isothiocyanate to inhibit ribonuclease activity.

This results in a high yield and purity as a result of minimal

RNA degradation. The RNA isolation and subsequent hybridisation experiments were completed in the final phase of the research. It was hoped that the limited biological samples available would be sufficient and adequate for RNA extraction to avoid further animal studies. As a result, several experiments were completed using tissue from various sources though this was in instances unsupported by biochemical and immunochemical data. It should be noted that when using frozen tissue and homogenate samples, the initial steps prior to freezing - perfusion, homogenisation and aliquoting of sample

- were not completed using sterile techniques. The extent of

RNA degradation by ribonucleases was a possible concern.

In the first experiment, total RNA was isolated from 6 rat liver samples;

1) fresh liver from 2 untreated Wistar rats, (Nos.1,2).

2) frozen homogenate from untreated and ciprofibrate

(20mg/kg/day, 14 days administration) treated Wistar rats. The tissue was excised, perfused with 0.9% saline and homogenised 174

in 0.25M sucrose prior to freezing at -80 degrees centigrade

for approximately 1.5 years before RNA extraction, (Nos 3,4).

3) frozen liver from ciprofibrate treated and untreated

Fischer rats. Tissues were excised and perfused with 0.9%

saline prior to freezing at -80 degrees centigrade. Treatment

duration :9 months at 20mg/kg/day, storage period of 3 months,

(Nos 5,6).

One gram of fresh or frozen tissue was homogenised in 16ml

guanidinium isothiocyanate whereas 2ml whole homogenate and

14ml of guanidinium isothiocyanate was utilised. The RNA was

isolated as previously described and assessed for purity by means of an A260/280 ratio reading and the concentration

determined using the absorbance at 260 nm (table 4.14).

The major disadvantage of this isolation procedure is the presence - although in the minority - of the RNA species of the hnRNAs and also the nuclear precursors of the cytochrome

P-450 mRNA sequences (Tay lor,1979). With this extraction protocol there is little variation in the purity of the isolated RNA (1.69-1.74) although the RNA concentration (mg

RNA / g liver ) is considerably higher in samples derived from fresh tissue. Horizontal slab gels of the RNA fractions, denatured by formamide (Maniatis et al.,1982) were completed to determine the structural integrity of the isolated RNA

(figure 4.7).

From the gel it can be seen that there was no appreciable degradation of RNA. The two distinct bands correspond to the Table 4 .14

RNA Purity and Concentration in Samples from Fresh and

Frozen Livers and Frozen Whole Homoqenates.

Treatment Animal RNA RNA concentration

Number (260/280 ratio)[a] (mg/g liver)

Control 1.69 1.03

Control 1.71 0.91

Control 3 1.74 0.30

Induced[b] 4 1.72 0.29

Control 5 1.71 0.45

Induced 6 1.71 0.53

[a] Ultrapure RNA is indicated by an A260/280 ratio of 2.0.

[b] Treatment with ciprofibrate. Figure 4.7.

Gel Electrophoresis of RNA Samples Derived from Fresh and

Frozen Liver Tissue and Frozen Whole Homogenate.

18 s

28 s

Track 1 ; Fresh untreated tissue.

Track 2 ; Fresh untreated tissue.

Track 3 ; Untreated frozen whole homogenate.

Track 4 ; Ciprofibrate treated frozen whole homogenate.

Track 5 ; Frozen untreated tissue.

Track 6 ; Ciprofibrate treated frozen tissue. 177

28s and 18s subunits of ribosomal RNA. The background smear is mRNA, the minimum size being approximately 0.5 kb. If appreciable degradation had occurred the majority of the RNA would run at

This the best method of quantitation when the amount of radioactivity is low in the hybrids. Unfortunately this technique was not available at the time and the intensity of hybridisation signal on the autoradiographs between series of dots was compared visually i.e. semi-quantitatively.

A positive hybridisation with the cytochrome P-452 probe was observed with all RNA samples indicating that this haemoprotein is a constitutive form expressed in untreated rats (Nos.1,2,3,5) (figure 4.8). Induction of the mRNA species coding for cytochrome P-452 was apparent in the ciprofibrate treated animals with a hybridisation signal visible at 1 ug/dot. In animal no.4 - derived from frozen homogenate - this was associated with a 5-fold induction of lauric acid omega-hydroxylase activity and a 8-fold increase in cytochrome

P-452 protein content by comparison with the control. One would assume a similar response in animal no.6, although biochemical and immunochemical analyses have not been completed. The duplicate filter was hybridised with the GPX probe. Filter washing was insufficient however to remove the 178

Figure 4.8,

Hybridisation of Control and Ciprofibrate Induced Liver RNA,

Derived from Fresh and Frozen Tissue and Frozen Whole

Homogenate, with the Cytochrome P-452 cDNA Probe

No. 1

Control No.2

Control No. 3

Ciprofibrate • No.4

Control No' 5

No.6 C iprofibrate

1 10 20 pg RNA

Cytochrome P-452 Dot Blot

Track 1 Fresh untreated tissue.

Track 2 Fresh untreated tissue.

Track 3 Untreated frozen whole homogenate.

Track 4 Ciprofibrate treated frozen whole homogenate.

Track 5 Untreated frozen tissue.

Track 6 Ciprofibrate treated frozen tissue. 179

unhybridised probe and to dissociate unstable hybrids with the

resultant autoradiograph being cross^contaminated with a high

background count (not shown). All samples hybridised with the

genomic probe but any differences in the intensity of the

signal could not be detected.

Total liver poly (A)+ enriched from clofibrate treated D2 mice

has previously been probed with mouse cytochrome P-450b cDNA

and mouse P-l-450 cDNA clones (PB and 3-MC induced cytochrome

P-450 isoenzymes respectively) (Stupans et al.,1984). No cross

hybridisation was detected indicating that the clofibrate

inducible cytochrome P-450 process involves P-450 genes

distinct from those involved in the induction by PB and 3-MC

(Stupans et al.,1984). Laurie acid omega-hydroxylase

(cytochrome P-450 LAw), thought to be identical to cytochrome

P-452, has been purified and its cDNA clone isolated and

sequenced (Earnshaw et al.,1988, Hardwick et al.,1987,

Tamburini et al.,1984). Total RNA was isolated from untreated

and clofibrate treated rats and mRNA levels quantitated using

the cytochrome P-450 LAw cDNA probe. A mRNA of approximately

2200 bases was detected in control and clofibrate-induced rat

RNA (Hardwick et al.,1987). In the present study, the presence

of mRNA coding for cytochrome P-452 was apparent in

ciprofibrate and untreated rats. Cytochrome P-450 LAw mRNA

levels in liver increased approximately 3-fold above control

levels within 3 hr of clofibrate administration and continued to increase *5 to 7-fold above control levels at 24 hr after treatment. These increases in mRNA closely paralleled the increase in immunodetectable enzyme after clofibrate 180

administration. In contrast, no significant increases in

cytochrome P-450 PCN or P-450e mRNAs were detected in induced

animals. These results verify that cytochrome P-450 LAw and

its mRNA are specifically induced by clofibrate and it is

assumed other peroxisome prolif erators. This haemoprotein

would appear to be a member of a new cytochrome P-450 gene

family because it shares little overall sequence similarity

with other cytochromes P-450 characterised to date.

Prostaglandin omega-hydroxylase (cytochrome P-450 p2) shares

74% amino acid similarity with cytochrome P-450 LAw suggesting

that this haemoprotein is also a member of the P-450 IV gene

family (Matsubara et al.,1987, Nebert et al.,1987).

The cDNA probe to cytochrome P-452 was isolated from a lambda

gtll expression library. Northern blot analysis revealed a

single mRNA species in male Wistar rat liver of approximately

2.4 Kbp hybridising to the radiolabelled probe. No significant

hybridisation of P-452 cDNA to poly (A)- RNA was observed in

either control or clofibrate treated liver; poly (A)+ RNA

levels were barely detectable in control liver but a very

marked increase in cytochrome P-452 mRNA was observed in

clofibrate treated liver (Earnshaw et al.,1988). It is

apparent that in the rat, both clofibrate and ciprofibrate

pretreatment results in the induction of cytochrome P-452 as measured immunochemically and catalytically. This correlated to an increase in the respective mRNA species as apparently observed in this study.

In the second experiment, RNA was isolated from frozen 181

homogenates from untreated and ciprofibrate treated mice,

hamsters and guinea pigs. In this instance the specific

activity of the labelled cytochrome P-452 probe was very low.

When the autoradiograph was developed the level of

radioactivity in the hybrids was insufficient to visually

detect dots (not shown).

The results obtained with the GPX probe were inconclusive (not

shown). This probe cross-hybridised with all species as expected but the hybridisation of the probe to filter-bound

RNA was not linear with dilution and any treatment-induced changes to the expression of this gene could not be detected.

At this stage it was decided not to continue investigation into the expression of GPX but to concentrate on cytochrome

P-452 mRNA expression. Total hepatic RNA was isolated from frozen homogenates prepared from untreated and treated (20 mg/kg ciprofibrate for 14 days) mice, hamsters, rats, guinea pigs and rabbits. Biological samples from marmosets acutely exposed to ciprofibrate were not available but equivalent studies were completed on chronically treated marmosets

(chapter 5).

A clear induction of mRNA species hybridising to the cDNA probe for cytochrome P-452 was observed in the mouse and rat

(figure 4.9). An increase in the level of radioactivity in the hybrids from treated hamsters, guinea pigs and rabbits was detected. In the hamster and rabbit this was associated with a

7 to 8- and 2.5-fold increase in lauric acid hydroxylation and 182

Figure 4.9.

Hybridisation of Control and Ciprofibrate Induced Mouse, Hamster/ Ratf Guinea pig and Rabbit Liver RNA with the Cytochrome P-452 cDNA Probe* 1 10 20 |jg RNA C Mouse

C Hamster I • • Rat • %

Guinea pig

Rabbit

Table 4.15. Correlation of Cytochrome P-452 cDNA Probe Analysis with Lauric Acid Hydroxylation and Cytochrome P-452 Content in Treated and Untreated Animals.

Lauric Acid Species Treatment Cytochrome P-452 Hydroxylation (% of total) 11-hydroxy. 12-hydroxy (nmol/min/nmol P-450)

Mouse Control 3.30 2.63 5.23 Induced 33.26 7.91 52.68 Rat Control 2.53 2.85 6.14 Induced 24.41 5.47 28.92 Hamster Control 2.04 2.84 4.18 Induced 18.72 6.29 31.27 Guinea Control ND 3.29 1.87 pig Induced ND 4.28 2.92 Rabbit Control 2.71 4.51 8.82 Induced 7.31 4.71 23.37 183

a 9- and 2.5-fold induction of cytochrome P-452 content,

respectively (table 4.15).

In the guinea pig, lauric acid hydroxylation was unchanged on

treatment and cytochrome P-452 could not be detected

immunoc hemic ally in hepatic microsomes. The increase in mRNA

hybridisation to the cytochrome P-452 probe in the guinea pig

(figure 4.9) was not associated with an increase in

catalytically active protein. The reason for this apparent

discrepancy is unclear at present, but may represent a

post-transcriptional processing event. Accuracy in determining

RNA concentration (mg/ml) is essential when using visual

quantitation of autoradiographs where a series of

concentrations are applied to the filter. On this basis, a

duplicate filter should be hybridised with a probe to a gene

product unlikely to be affected by treatment e.g. albumin,

myosin. This would allow for a more objective analysis of mRNA

expression to be made and would negate any possible

inaccuracies in the determined RNA concentrations. An added

complication in these experiments is the use of a single probe

to cross-hybridise between species. The assumption is made that the rat cytochrome P-452 probe hybridises with equal efficiency to mRNA(s) coding for this protein in all species - a concept analogous to antigen-antibody affinity. In untreated animals the signal was most intense in the rat and as the cDNA probe was originally derived from rat liver this probably represents an accurate measure of the amount of cytochrome

P-452 mRNA present. However caution must be exercised when making assumptions using low stringency to detect the apparent 184

homology between species since the signal intensity here may

be a combination of the absolute amount of homologous

sequences in the other species plus a contribution from

cross-hybridisation of the rat probe with closely related, yet

heterologous cytochrome P-452 sequences.

It is also possible that under the experimental conditions

used the cytochrome P-452 cDNA probe hybridised to some extent

to conserved regions in other cytochrome P-450 mRNAs. Although

experiments to determine the degree of non-cytochrome

P-452/P-450 LAw cross-hybridisation have yet to be performed, results from other cytochrome P-450 gene family studies would indicate that it will make a relatively small contribution to the total hybridisation. In the case of the GPX gene this is strongly conserved between species, e.g. mouse, rat, cat, human and bovine forms and cross-species hybridisation is specific. The problems of using cDNA probes in the cytochrome

P-450 system is due to the well documented existence of isoenzymes. The cytochrome P-450 superfamily comprises of at least 10 gene families, the amino acid sequence of a protein from any one family being <36% similar to that from the other

9 families. Those cytochromes P-450 genes considered to be in the same family have >36% homology, and those in the same subfamily have >68% homology (Nebert and Gonzalez,1987). The proposed evolution of the cytochrome P-450 gene family is shown in figure 4.10. 185 ,

Figure 4.10

Evolution of the cytochrome P-450 family 2,000 1,500

2in 1,000 1

UJ Q_ LLI

UJ

U_ 250 UJ

UJ

UL

Present

CL

based on rat for families I-IV, cow and human for

families XI, XVII, XIX and XXI; the LI family exists in yeast

and the Cl family in Pseudomonas (Nebert and Gonzalez, 1987).

The different isoenzymes are placed in families or subfamilies

according to the degree of deduced amino acid homology.

The observed homology between different families is primarily

limited to the active site, a region displaying considerable amino acid sequence conservation. The amino acid sequence deduced from the full-length cDNA of cytochrome P-450 LAw is

<33% similar to that encoded by genes in any other family

(Earnshaw et al.,1988, Hardwick et al.,1987). 186

In this study identical procedures were utilised for both

cytochrome P-452 and GPX hybridisations. If there had been

sufficient time to optimise hybridisation conditions for both

probes, several protocols are available for hybridising a

probe in solution to nucleic acid immobilised on filters. The

conditions used depend on the purpose of the experiment and in

general are governed by whether DNA-DNA or DNA-RNA

hybridisation is involved and whether closely related or

distantly related sequences are reacting. Reaction conditions

that permit formation of hybrids which have a high degree of mismatching are said to be low stringency while those allowing

only well matched hybrids to form are said to be high

stringency (Hames and Higgins, 1985). Conditions favouring the detection of well matched hybrids involve high temperatures of hybridisation (>42 degrees centigrade in 50% formamide and 6 x

SSC) combined with washing at higher temperatures and at lower salt concentrations. The temperature and salt concentration of the washing solution determines which hybrids will remain on the filter. In general, washing should be completed under as stringent conditions as possible and in practice 60-65 degrees centigrade at 0.1 x SSC is usually chosen for hybrids having a high degree of homology (75% +) and 42-56 degrees centigrade at 0.5-0.2 x SSC for poorly matched hybrids (40-60 % homology). The washing conditions used in these experiments were moderately stringent (0.5 x SSC at 55 degrees centigrade) to allow for detection of cross species, reasonably well matched hybrids.

The expression of mRNAs coding for cytochrome P-452 and GPX 187

has not been previously investigated across species. In the

present study an increase in mRNA hybridising to the

radiolabelled cytochrome P-452 cDNA probe was apparent in the rat, mouse, hamster and rabbit with a parallel increase in cytochrome P-452 levels and activity. A slight increase was seen in the guinea pig and this accompanied unchanged lauric acid hydroxylation with no immunodetectable protein. The absence of increased catalytic activity would suggest that the increase in mRNA does not reflect an increased translation of the protein.

The mouse genomic GPX probe cross reacted with all species.

However, the hybridisation procedures have not been optimised and as such any changes in mRNA expression resulting from treatment could not be discerned. It should be noted that only preliminary experiments have been completed and that further investigation is required before the precise inter-relationships between changes in enzyme levels and gene expression can be determined and a casual or causal association defined. 188

4.9 DISCUSSION

The primary aim of this research was to examine and describe

the short term effects resulting from ciprofibrate treatment

in different rat strains and different species at a

biochemical and molecular level.

The first objective was to characterise the biological

response in the rat to this hypolipidaemic drug and secondly,

to clarify the species differences using the rat as the

reference model. In the rat the induced hepatic responses can be sub-divided into;

1) hepatomegaly ; a 1.5 to 2-fold increase in the liver/body weight ratio,

2) induction of cytochrome P-452 ; a 6 to 10-fold increase in immunodetectable protein with a minimal increase in (omega-1)- hydroxylation and a marked induction in the formation of

12-hydroxylaurate (10 to 16-fold). This correlated to an increase in the mRNA species hybridising to the cytochrome

P-452 cDNA probe. The induction of lauric acid omega-hydroxylase was associated with a marked depression in the metabolism of model drug substrates (benzphetamine and ethoxyresoruf in).

3) peroxisome proliferation (Lalwani et al.,1983a,b) and induction of peroxisomal enzymes; peroxisomal beta-oxidation was increased 9 to 34-fold and the specific activity of catalase and uricase was unchanged. Total carnitine acetyltransferase activity (mitochondrial and peroxisomal) was induced 16 to 70-fold. 189

4) monoamine oxidase activity was decreased and the specific

activity of succinate dehydrogenase unchanged. Alpha-glycero­

phosphate dehydrogenase activity was elevated.

5) decreased cytosolic glutathione peroxidase activity,

Gunn rats are unable to esterify bilirubin as this rat strain

lacks hepatic bilirubin UDPglucuronyltransf erase activity

(Burchell and Blanckaert,1984). These animals were used in

this study to examine whether modulation of this phase II

enzyme would effect the hepatic toxicity of ciprofibrate.

Clofibrate and its structural analogues, are hydrolysed to the

free acid (i.e. active compound) prior to conjugation which

gives rise to the acylglucuronide which is rapidly eliminated

in the urine (Emudianughe et al.,1983). Impairment of this

metabolic pathway could result in an in vivo enhanced response

to ciprofibrate as this compound would be present only as the

pharmacologically active parent acid. The Gunn rat was the

least responsive rat strain to peroxisome prolif erators

although the variation in the enzyme parameters tested was minimal and the differences were not significant using analysis of variation. This would indicate that the enzymic capacity for glucuronidation of phenoxyisobutyrates is unchanged in the Gunn rat. This has been substantiated by recent studies demonstrating that Gunn rats conjugate clofibric acid (Burchell and Blanckaert,1984) and that this compound specifically induces bilirubin glucuronidation

(Fournel et al.,1986) but not the conjugation of itself (Odum and Orton,1983). 190

Cytochrome P-450 monooxygenases catalyse the omega- and

(omega-1)-hydroxylation of fatty acids, arachidonic acid and prostaglandins. In rat liver, the isoenzyme cytochrome P-452 which is induced by peroxisome prolif erators, hydroxylates

lauric acid and arachidonic acid preferentially at the omega- position (Capdevila et al.,1985, Tamburini et al.,1984).

Hepatic hydroxylation of lauric acid has been reported in the rat, rabbit and mouse; the rabbit exhibiting the highest activity (Ichihara et al.,1969). Oxidation of laurate at the omega-position was observed in guinea pig liver microsomes

(Kupfer and Orrenius,1970). The administration of PB to rats had no effect on omega-hydroxylation of laurate but (omega-1)- hydroxylation was elevated 2 to 3-fold (Bjorkhem and

Danielsson,1970). BNF and 3-MC had no effect on omega- or

(omega-1)-hydroxylation of lauric acid suggesting that distinct cytochrome( s) P-450 mediate the two fatty acid hydroxylases in liver microsomes (Okita and Masters, 1980). A slight increase in 11-hydroxylation of laurate was seen in the mouse and hamster and omega hydroxylation was induced in the mouse, hamster and rabbit. In these species it correlated to an increase in cytochrome P-452 content and an apparent elevation in the mRNA species hybridising to the cytochrome

P-452 cDNA probe.

Given the high degrees of nucleotide homology within the same cytochrome P-450 family, cross hybridisation of cDNA probes with more than one homologous mRNA species of similar size presents a problem. The cDNA probe to cytochrome P-452 cross reacted with all samples indicating a high degree of homology 191

between rat mRNA and that of the different species. With the

immunochemical ELISA technique, low levels of protein can be

detected however accessibility of the epitopes to the antibody

may be affected. Elucidation of the number of related

cytochrome P-452 mRNA species within a given tissue and

construction of specific cDNA probes is necessary before the

induction role of ciprofibrate at the transcriptional and mRNA

levels can be deduced. DNA sequences of genomic and cDNA

clones will enable the relationship between cytochromes P-452

in different tissues, strains and species to be elucidated. It

should be emphasized that species differences in the

expression of cytochrome P-452 may additionally be

rationalised by differences in the non-coding, upstream 5'

regulatory site of the gene. It has been reported that

transcription of the cytochrome P-450 LAw (i.e.cytochrome

P-452) gene is elevated as early as 1 hr after clofibrate

treatment and remains elevated for up to 24 hr ( Hardwick et

al.,1987). This accompanied a marked increase in the mRNA

species coding for the protein and immunodetectable protein.

The specific increase in cytochrome P-450 LAw was due to transcriptional activation. Transcriptional activation of cytochrome P-450 genes has been observed with PB (Pike et al.,1985) and 3-MC (Gonzalez et al.,1984). The GPX probe cross hybridised with all 6 species. It was not possible however to correlate the decrease in enzyme activity in the rat to alterations in mRNA levels and quite clearly further experimentation is required.

In the guinea pig and marmoset lauric acid hydroxylation was 192 unaffected by ciprofibrate treatment and there was no immuno- detectable protein. Immu.)rioquantitation of cytochrome P-452 was completed using an antibody raised to the pure protein. This antibody did not cross react with the major PB and 3-MC inducible cytochromes P-450 and was therefore described as monospecific for the clofibrate induced isoenzyme although the antibody was polyclonal in origin. When a pure protein is injected into an animal a series of antibodies are produced which recognise certain components in the antigen. These resulting polyclonal antibodies bind to a wide range of different epitopes. Monomeric proteins like serum albumin carry 10 or more antigenic determinants. Monoclonal antibodies

(Mabs) are characterised by high specificity for a single epitope - often a sequence of 4-7 amino acids - on a protein.

With cytochrome P-452, the antibody cross reacted with the rat, mouse, hamster and rabbit microsomes. The catalytic activity of the fatty acid hydroxylase system in the guinea pig and marmoset and the lack of cross reactivity with the rat cytochrome P-452 antibody using ELISA would indicate a fundamental difference between species. Preliminary experiments in our laboratory using Western blotting and ELISA techniques indicates the presence of a cytochrome P-450 isoenzyme in human liver biopsy samples. The extent of cross reactivity with the cytochrome P-452 antibody suggested the existence of a closely related but distinct gene product. In these human samples antibody inhibition of lauric acid hydroxylation was less extensive than in the corresponding rat sample which was commensurate with the above results. 193

A Mab to rat liver epoxide hydrolase did not recognise this

enzyme in hamster, rabbit, monkey or human microsomes

(Telakowski-Hopkins et al.,1983). However, rabbit polyclonal

antibodies produced a strong cross reaction with purified

human liver epoxide hydrolase and a weak to strong reaction

with the other species. This would suggest that the Mab was

specific for an epitope present only in the rat protein

whereas the raised polyclonal antibodies recognised "common"

antigenic determinants. A series of 9 Mabs raised against

purified rat liver cytochrome P-450c reacted with 6 different

epitopes and one of these was shared by cytochrome P-450d

(Thomas et al.,1984). Two proteins immunorelated to cytochrome

P-450c and P-450d were observed in the rabbit, hamster, guinea

pig and C57BL/6J mouse. The conservation of the number of rat

cytochrome P-450c epitopes among these species varied from 2

(guinea pig) to 5 (C57BL/6J mouse and rabbit). It is apparent

that cytochrome P-450 proteins carry several epitopes. These

may be specific to one isoenzyme or one or more epitopes may

be shared between haemoproteins forming a family of

immunochemically related forms (Thomas et al.,1987). Antigenic

determinants may be conserved across species reflecting the

extent of sequence homology between cytochrome P-450

isoenzymes. An advantage of using polyclonal antibodies as

opposed to Mabs is that antibodies are raised to a variety of

epitopes thereby increasing the probability of cross reaction between species although the specificity is affected. In this study the ELISA results suggested that rat cytochrome P-452 epitopes were shared by the mouse, hamster and rabbit and that these were absent in the guinea pig and marmoset. However 194

there may be antigenic determinant(s) common to all 6 species

to which there was no antibody raised. Positive antibody

binding indicates that the epitopes are present, located on

the antigen surface and easily accessible to the antibody. Low

or no binding may be due to either the absence of a cytochrome

P-450 with an appropriate epitope or inaccessibility of the

epitope owing to steric hindrance by other microsomal

components such as reductase or membrane. This latter

explanation is unlikely as the microsomes were solubilised

prior to ELISA.

It would be useful to raise a series of Mabs to cytochrome

P-452 to examine the distribution of the antigenic

determinants. However, the cDNA probe hybridised to mRNA from

all 6 species and this would infer that there is considerable

homology in this cytochrome P-450 isoenzyme between species.

This is a more sensitive technique than using a polyclonal

antibody which may be raised to a single epitope on the

protein which may or may not be conserved across species.

The ratio of 12- to 11-hydroxylaurate was less than 1 in the

guinea pig and marmoset but this ratio was reversed in the

species exhibiting cytochrome P-452 induction, measured

catalytically and immunochemically. It is possible that in the

guinea pig and marmoset the complement of monooxygenases

responsible for fatty acid oxidation are different to those in

the rat. In these latter species, the cytochrome( s) P-450 display a regioselectivity for (omega-1)-hydroxy lation of

lauric acid. The incubation of PGA-1 and PGE-1 with guinea pig 195

liver microsomes resulted in the corresponding omega- and

(omega-l)-hydroxylation products; the latter was the major

route (Kupfer et al.,1.97.8). A possible explanation is that the

hydroxylation at the two positions are catalysed by different

enzymes and that there is a higher affinity of PGA-1 for the

omega-hydroxylation enzyme. The classic inducers, DDT,

benzapyrene and Aroclor 1254, induced (omega-1)-hydroxylation

but omega-hydroxylation was unaffected (Kupfer et al.,1979).

In the rabbit omega- and (omega-1)-hydroxylations are

catalysed by different cytochrome P-450 isoenzymes

(Theoharides and Kupfer,1981). The km values for

omega-hydroxylation are lower than for (omega-1)-

hydroxylation. At low substrate concentrations therefore

omega-hydroxylation will predominate but at saturating levels

of substrate, (omega-1)-hydroxylation will be greater

(Kupfer, 1982). A similar system is possible in the guinea pig

and marmoset where the concentration of laurate could be saturating with respect to (omega-1)-hydroxylation giving this apparent reversal in regioselectivity. However, in the rabbit it appears that the monooxygenases catalysing the hydroxylation of PGs are different to those catalysing laurate hydroxylation (Theoharides and Kupfer,1981). The administration of clofibrate and DEHP to rabbits resulted in a significant and paralleled increase in the PGA-1 omega- hydroxylase and laurate hydroxylase systems in liver microsomes (Yamamoto et al.,1986). Hamster liver microsomes convert PGF 2-alpha to 19- and 20 hydroxy products in the ratio of 1:2 whereas in the rat the ratio is 10:1 (Powell,

1978). The hydroxylation of various PGs is evident in 196

mammalian species, however the relative ability to form omega-

vs (omega-1)-hydroxy products varies extensively between the

species examined. It would seem that the oxidation of fatty

acids is widely variant, with the guinea pig and marmoset

displaying regioselectivity towards (omega-1)-hydroxylation.

No induction of cytochrome P-452, as measured catalytic ally,

was observed indicating that the omega-hydroxylase system in

the guinea pig and marmoset is controlled differently to the

rat at the gene level. The cytochrome P-450 system is

inducible in these species for example, Aroclor 1254 with the

guinea pig (Kupfer et al.,1979) and PB in the marmoset

(Challiner et al.,1980). There was no response to

ciprof ibrate.

In the mouse a doubling in liver/body weight ratio was

observed, however in the hamster hepatomegaly was significant

but minimal. No change was seen in the other species.

Benzphetamine-N-demethylase activity was decreased in the

mouse and rabbit and ethoxyresorufin-deethylase activity in

the mouse alone. Monoamine oxidase activity was decreased in

the mouse and hamster and increased in the marmoset. The

specific activity of succinate dehydrogenase and

alpha-glycerophosphate dehydrogenase was elevated in the marmoset. Carnitine acetyltransf erase was induced in the mouse, hamster, and rabbit yet was unchanged in the guinea pig and marmoset. Catalase was unaffected by treatment except at the high dose level in the Sprague Dawley and Wistar strains and rabbit, where it was elevated. Uricase was induced at 20 mg/kg ciprofibrate in the rabbit and activity was decreased in 197

the hamster. This enzyme appeared to be absent in the

marmoset. Peroxisomal beta-oxidation was induced in the mouse,

hamster and rabbit and was unchanged in the guinea pig. A

minimal increase was seen in the marmoset and at the high dose

level activity was comparable to the basal levels in the

untreated mouse, rat and rabbit. Glutathione peroxidase was

decreased in the mouse.

The major response in the rat to peroxisome proliferators is

the coordinate induction of cytochrome P-452, carnitine

acetyltransferase and peroxisomal beta-oxidation enzymes with

unchanged/slight increased catalase activity and decreased GPX

activity. It is apparent that ciprofibrate administration

results in similar biochemical responses in the rat and mouse

liver. In the hamster and rabbit, induction of lauric acid

omega-hydroxylation, peroxisomal beta-oxidation and carnitine

acetyltransferase was observed. These parameters were

unchanged in the guinea pig. In the marmoset a slight increase

in only peroxisomal beta-oxidation was seen. This would

suggest that the mechanism of induction of peroxisomal

beta-oxidation in the absence of other enzymatic changes in

the marmoset is different to the rat which exhibits coordinate

induction. Hepatomegaly was observed in the mouse, rat and

hamster but not in the rabbit. However, in the rat induction

of peroxisomal beta-oxidation dissociated from hepatomegaly

and peroxisome proliferation has been reported (Lazarow et al.,1982). The induction of a novel long chain acyl-CoA

/ hydrolase in rat liver by clofibrate, DEHP and acetylsalicylie acid has been demonstrated (Miyazawa et al.,1981). The 198

induction of two cytosolic hydrolases was later found to occur

with clofibrate and DEHP administration (Kawashima et

al.,1982) with the activity being greater in male rats than

females with both peroxisome prolif erators. The induction of

long chain acyl-CoA hydrolase by clofibrate was studied in

rats, mice and guinea pigs (Kawashima et al.,1983). Two

hydrolases were detected in rats and one hydrolase was

demonstrated in mice corresponding to the low weight species

(hydrolase 2). No hydrolase induction was exhibited in the

guinea pig. DEHP, DEHA, 2,4,5-T, clofibric acid and tiadenol

induced both hydrolases whereas acetylsalicylic acid and 2,4-D

induced hydrolase 2 only (Katoh et al.,1984). Induction of

acyl-CoA hydrolase 2 appeared to be associated with enhanced

peroxisomal beta-oxidation activity. A selective increase in

microsomal 1-acyl-glycerophosphoryl choline acy1-transferase

has been observed with DEHP, clofibric acid, tiadenol and

acetylsalicylic acid (Kawashima et al.,1984). Treatment with

clofibric acid increased the activity in mice and rats but not

in guinea pigs, suggesting a common induction with the long

chain acyl-CoA hydrolases. Although the role of these enzymes

in the toxicology of peroxisome prolif erators requires

elucidation, it is apparent that induction occurs in the rat

and mouse but not the guinea pig. In the rat, mouse, hamster

and rabbit induction of cytochrome P-452 is associated with a

parallel increase in peroxisomal beta-oxidation and carnitine

acetyltransf erase. It would be of interest to examine the

induction of long chain acyl-CoA hydrolase in the hamster, rabbit and marmoset. It would be tempting to predict enzyme induction in the hamster and rabbit but not in the marmoset as 199

the guinea pig clearly does not respond in this manner.

The lack of response in the marmoset to ciprofibrate cannot be

attributed to non/poor absorption of the drug. In the marmoset

there was a linear relationship between plasma ciprofibrate

levels and dose. At 50 mg/kg ciprofibrate for a duration of

fourteen days the concentration of drug in the plasma was

approximately 230 ug/ml ( Sterling Winthrop, personal

communication). In the rat, following repeated oral doses of

30 mg/kg ciprofibrate every 24 hours the plasma level of drug

was in the range of 300-350 ug/ml (Edelson et al.,1979).

Assuming a linear relationship with dose, this would

correspond to 200-235 ug ciprofibrate/ml at a dose level of 20

mg/kg per day in the rat. It is possible to make an assumption

that the initial plasma levels (measured 4hr post dose) in the

marmoset at the high dose level (lOOmg/kg) are 1.5 to 2 fold

higher than that in the rat. The pharmacokinetics of

ciprofibrate in the other species investigated in this study has not been examined. The marmoset and guinea pig do not exhibit enzyme induction in vivo and in vitro. The species differences in the metabolic conjugation of clofibrate and clofibric acid have been studied (Emudianughe et al.,1983).

The rat, guinea pig, rabbit and man form only the ester glucuronide. In the rat ciprofibrate is excreted as the free drug and glucuronide and it is probable that a similar metabolic pathway is seen with ciprofibrate in the guinea pig and rabbit. It is also assumed that the ester glucuronide of ciprofibrate is formed in the hamster and marmoset. The lack of induction in the guinea pig and presumably the marmoset is not due to reduced bioavailability of the drug i.e lack of 200 absorption or to a different route of metabolism.

On the basis of biochemical and immunochemical data, a species dependent response occurs with ciprofibrate administration, the order of susceptibility being the rabbit < hamster < mouse

< rat. The marmoset and guinea pig are non-susceptible, bearing in mind that the high dose level is 100 mg/kg in the marmoset and 20 mg/kg in the other species. 201

Chapter 5

COMPARATIVE RESPONSES IN THE RAT AND MARMOSET FOLLOWING

CHRONIC ADMINISTRATION OF CIPROFIBRATE

5.1 INTRODUCTION

In chapter 4, the species and strain differences in various

hepatic responses to the short term administration of

ciprofibrate were studied. The results clearly indicated

differences in response between species, the order of

susceptibility being the rat> mouse> hamster> rabbit. The

marmoset and guinea pig were relatively non-responsive.

Consequently the study was expanded to investigate the chronic

effects of ciprofibrate in the Fischer rat and marmoset, in an

attempt to identify the fundamental differences which may

exist between rodents and primates. In addition the effect of

the chronic administration of clofibric acid and bezafibrate

on microsomal enzyme parameters was investigated in the

Fischer rat.

An additional group of animals was included in the primate

study in order to assess the reversibility of the induced

hepatic responses. Animals in this latter treatment group were

administered ciprofibrate for 26 weeks and received no further

treatment for a 4 week recovery period prior to sacrifice at

30 weeks. The low dose level in the marmoset was 20 mg/kg/day and was 80 mg/kg in the high dose level and recovery group.

Animals in the high dose group i.e. 26 weeks continuous 202

treatment (i.e. the non-recovery group) received the maximum

amount of drug which could be tolerated by the marmoset on a

daily basis. Initially, a top dose of 100 mg/kg was chosen,

however during the course of the study this was reduced to 80

mg/kg at week 21 to maintain the health status of the animals.

This was based on clinical examination of the animals and was

indicated by deterioration of general condition, body weight

loss and lethargy.

The dose levels chosen for the three hypolipidaemic drugs were

based on comparative lipid lowering potencies derived from

published data (Sterling-Winthrop, personal communication).The

objectives of this study were primarily to characterise the biochemical, molecular and morphological changes associated with the liver in both species and to assess the importance of a recovery period following treatment.

5.2 HEPATOMEGALY

In the treated rat, there was a sustained increase in the liver/body weight ratio characterised by a maximal 2-fold induction with ciprofibrate and a minimal but significant increase with clofibric acid (table 5.1). The liver/body weight ratio in the marmoset was not affected by ciprofibrate treatment at either time point (26 or 30 weeks) (table 5.2).

The positive response in the rat to ciprofibrate was consistent with the published data (Lalwani et al.,1983a,b). A similar increase in liver/body weight ratio was seen the

Wistar rat as opposed to the Fischer rat with bezafibrate and Table 5.1.

Comparison of Different Hypolipidaemic Agents on Liver Size

and Cytochrome P-450 Content in the Fischer Rat Following ... Chronic Administration.

Dose Level Liver/Body weight Total Cytochrome

Ratio (%) Tal P-450

Specific Content

(nmol/mg)

Control 3.10+/-0.24[b] (100)[c] 0.51+/-0.10 (100)

Ciprofibrate 7.19+/-0.31[e] (232) 0.41+/-0.03 (82)

(20 mg/kg)

Clofibric acid 3.84+/-0.13[d] (124) 0.60+/-0.13 (118)

(50 mg/kg)

Bezafibrate 5.06+/-0.26[e] (163) 0.96+/-0.11[d] (188)

(100 mg/kg)

5|c Single daily dose for 26 weeks [a] Data reproduced from table 3.1.

[b] Values are means +/- S.D.

[c] Figures in brackets are values expressed as a % of

corresponding control.

P values for results significantly different (Students T-Test) from control data are [d] p<0.01 and [e] p<0.001. Table 5.2 The Effect of Ciprofibrate on Liver size and Cytochrome P-450 Content x: - 03 02 •H •H 4= 43 43 Z &U O < p a 03 03 C 05 P £ O to 03 CT> P o o C o C P 03 o e £ to to 04 Ed CO =) E-* »-i Cd Z E* w u Q 02 OS Cd < E* < Z O Cx< E-*O J> CQ cd w o a O Cd CQ o E* a >•* Cz3 <*> o z a *-•o 3e w —* HH o •M* Lf> E E3 02 o z r 04 Cd O a >4 E-< »r* o u 1 h 02 Z o <3 CO i O Cd E O >H Cl4 t-H O U E o Cd Z h h

• rH •H 43 — £ c o i O p £ cn 3\ 03 3 cu p o 03 ' 43 43 p to co in W N CMO CO O HT ~ f UO W W W O omno to m o © o m p CM O O O O MCM CM o O «H O p o HT H- H- H- O o o cno p H- Oo 1 C O O 03 1 1 1 1 C 3 H 1 H H • . . . • * . nr m 43 U5 CM Z O pH VO p pH HT Pt o o s n o ncn cn m o +*** O 3 p O to £ 03 3 £ 03 to + cn £ £ £ H 1 • # * • . \ p 30 p o p cn .— p O o o cn 5* H- H H t t H H • • . to CM 43 43 43 43 p f H m p o o u p H9* * p p O p O no \ O o HJ« **-» p o + 03 3 3 03 P V 3 £ C o c p 03 — H- H 1 H H H H H H • . . 43 H U p 03 XI X Z 03 p CO £ 03 o O 3P 03 5 5e C3 03 to H m CO o \ x C— *3* P H o CM .—. cn CM to \ o O CM CM ro H- P > u O 03 03 5 H- cn cn £ 1 I m • . • 43 p p tjO CM 30 02 43£ i—i N cm w w o ro r— p o © ro o CM O p— pH HP CM o Lf> Hp o —• o o \ s o o —If— O pH m o H p o o o 03 H* H- u 03 03 to •4- H- o o C cn p o \ 1 1 t I H • • • • • • • • » ro p j \ p *—p 03 P-, cn p-% CM p ao cn * H H H

a? •D C ) D (3 O <0 CQ Q H* 03 *• rp 9 03 to 03 P 03 03 to £ 03 c 1 H • • X3U ffQ H • H • X5 9 W 0 CO 30 43 43 P I H r u cu o Cu 03 43 44 73 H • 43 43 Cn 03 03 3 43 > P 3 P 0) a) 03to C O «P O C M M 0) 03 (D T3 U 43 •a u 3U 9 03UJ O 10 t0 M > - > n> ® h 3 to o> X CU M v to to a 03 o to cu a M u c OS O' a O U O M

in h in

3 O 73 JQ O' h cu to cu P U CO £ 03 p p h

"H c: H r pH 43 43 -p pH 03 •H 44 4H -P -p 43 03 43 «p E-< CO rtJ e: >4 <3 P 03 c 9 03 C to 0 CO I )

control data are [dl < 0 .0 0 1 . 205 clofibrate following a 3 day treatment period (Sharma et al.,1988). There is no published data with respect to the assessment of liver/body weight ratio in the marmoset following chronic administration of peroxisome prolif erators.

In the Rhesus monkey (Old World) a slight non-significant increase was seen in liver weight with DL-040, a newly developed peroxisome prolif erator [figure 5.1] (Lalwani et al.,1985).

Figure 5.1.

Chemical Structure of DL-040 T4-((( 1,3-benzodioxol)-5-yl)

methyl)aminobenzoic acid]

0

C H o - N H

II 0 DL - 040

5.3 THE MICROSOMAL DRUG METABOLISING ENZYME SYSTEM

5.3.1 Cytochrome P-450 Content

In the bezafibrate treated rat, an approximate 2-fold induction in cytochrome P-450 content was seen (p<0.01) (table

5.1). In the ciprofibrate treated group a non-significant decrease in the specific content of cytochrome P-450 was

observed (table 5.1 and 5.2). No significant changes were seen

at either time point in the marmoset (table 5.2). The basal

level in the untreated marmoset was lower than that in the

rat.

In the rat, a dose-dependent increase in cytochrome P-450

content was seen with clobuzarit treatment, however levels were unchanged in the marmoset (Orton et al.,1984). With a treatment period of 3 days in the Wistar rat, similar

increases to that in the Fischer rat with bezafibrate and clofibrate, were observed (Sharma et al.,1988).

5.3.2 Benzphetamine -N- Demethylase Activity

A marked decrease in benzphetamine-N-demethylase activity was observed in the treated rat (table 5.3). With ciprofibrate, activity was 25% of the control value and with clofibric acid and bezafibrate administration activity was halved. In the marmoset the metabolism of benzphetamine was not significantly altered on ciprofibrate treatment (table 5.4). The basal activity was greater in the recovery group and there was a slight decrease in activity on treatment which was not statistically significant due to a large inter-group variation in response.

With bezafibrate and clofibrate, a decrease in activity was seen in the Fischer rat similar to that in the Wistar rat following a 3 day treatment period (Sharma et al.,1988). In Table 5.3.

Comparison of Different Hypolipidaemic Drugs on the Microsomal

Drug Metabolising Enzyme System in the Fischer Rat * Following Chronic Administration.

Dose Level Benzphetamine-N- Ethoxyresorufin-

Demethylase O-Deethylase

(nmol product/min/ (nmol product/min

nmol P-450) /nmol P-450)

Control 25.93+/-2.76[a] (100)[b] 0.02+/-0.01 (100

Ciprofibrate 6.13+/-1.36[e] (24) 0.01+/-0.01[d] (50)

(20 mg/kg)

Clofibric acid 14.31+/-3.91[c] (55) 0.01+/-0.01 (50)

(50mg/kg)

Bezafibrate 12.43+/-1.47[d] (48) 0.01+/-0.01[d] (50)

(100 mg/kg)

& Single daily dose for 26 weeks a Values are means +/- S.D.

b Figures in brackets are values expressed as a percentage of

corresponding control.

P values for results significantly different (Students T-Test)

from control data are [c] p<0.05, [d] p<0.01 and [e] p<0.001. Table 5.4 The Effect of Ciprofibrate on the Microsomal Drug Metabolising Enzyme •P JC T3 JC •P *H H f • s •P •P h s a X (X4 o 0 ro C fO ITJu £ O to o o c cn c u fO M o U £ 10 o c H 5 2 C H X r d C a M C i O d C X d C E HH Z d C Z < X d C 3 5 2 E x H > d C a o X O OS 4 * £ H H O C OS o 1 1 h 1 1 h Cd CO O O Cd CO Cd < O 3 Eh 02 OS E hh Eh Cd Z o o x E-< O Dtl >-3 Q O Cd Cd O > CO W Cd d C X E X d i < O C d C 1 3 X E < X 3 w d C d ( X d C O C h h h j < Z V—r U C 3 T p \ o S H * O N H r 3 * > - 4 H » s G O H r M O — S E O U C U c G U 3 e s O + O i Q M G

n i HT i n i * M i O e H c £ o 1 H 6 6 H H JkJrH H r n c M C HIT o n c —r «— » •— w —J h— CJ CM CO CM CJ o o O C O V n c —rH r c— H r o O M C O O o r — c o c H r o M C + o c o t o o o c h o o — to i 1 . . • • • • . . Cn o P M C KO P X n c M C N O VD . - r m I H H r H r M C \ CO 0 £ o 0 > - H . — 0 0 0 - H cn g 1 1 1 • . \ p ­ ' e H r n i n c O H n i O C M C H r O C n i o o o r H r O C O l — r o . .— \ . s i T < + + cn cn i . • • • • + . voo M C 0 * J P T H P 0 P p rH n e \ fH H » H r u o M C M C H r o * C * o o t o o > O 0 H r ^ £ 0 Q J I H H - X 0 0 * i 0 C Hi c o O H i 1 1 • « . P - m c H r o • C A X CO 0 ro M o £ . o t CO o •AC \ O l \ n i M C n c X n i n c n c - O l o n c \ O M C M C H r CO ? a o t o o o C ' O I H -4- ■+■ ■+■ -4- o \ 1 f I • • • • • • . T3 £ cn 0

■o TJ 0 CO. >0 u c <0 O CO ' 5 ®' •• c O) r nj n •rH •H jQ3 At tu P H r T3 *0 P «H •P I H O G 0 M 0 ( cn ro 3 I H 0 to J U 0 too s 03 0 ( 0 3 U T3 P I H rH > rO 0 toro 0 X cu M to 01 0 to 0 ro to 0 M 0 cn 0 I H G u C 0 0 U O M to CU O cn O 0 c u o c W o • in p •O jO 0 Oi i H U i h 0

P values for results significantly different (Students T-test) from control data are Cd] < 0.01 and (el < 0.001. the marmoset this enzyme has not previously been studied using

benzphetamine as the substrate. An alternative substrate,

aminopyrine has been used in past studies. With clobuzarit treatment, aminopyrine demethylase activity was not

significantly changed however at the high dose level a slight decrease was observed (Orton et al.,1984). Using the classic cytochrome P-450b inducer, phenobarbitone, a marked 3-fold increase in activity was observed in the marmoset (Challiner et al.,1980). It is interesting to note that this phenobarbitone induction was not accompanied by an increase in liver mass as has been observed in other species.

5.3.3 Ethoxyresorufin-O-Deethylase Activity

In the rat, ethoxy resorufin-O-deethylase activity was halved

(table 5.3). In the marmoset, activity was unaffected by ciprofibrate treatment (table 5.4), although the inter-group variation in response was marked and no apparent trends could be detected.

In the Wistar rat the decrease in ethoxyresorufin-O- deethylase activity resulting from a 3 day clofibrate and bezafibrate treatment, was greater than that in the Fischer rat following chronic administration with these compounds

(Sharma et al.,1988). Activity was unchanged in the marmoset following the administration of clobuzarit (Orton et al.,1984). 210

5.3.4. Induction of Cytochrome P-452

The induction of cytochrome P-452 was assessed catalytically

and immunochemically in liver microsomes. In the rat there was

a slight increase in the 11-hydroxylation of lauric acid in

treated animals which was significant with bezafibrate and

clofibric acid (table 5.5). The 12-hydroxylation of lauric

acid was induced, 8-, 12- and 20-fold with clofibric acid,

bezafibrate and ciprofibrate, respectively (table 5.5). In the

marmoset, the 11-hydroxylation of lauric acid was

significantly decreased to 60% of the control value at the

high dose level at 26 weeks and in the recovery group (figure

5.2). At the high dose level at 26 weeks lauric acid

12-hydroxylase activity was significantly decreased to 60% of

the control value.

The increase in the formation of 11-hydroxylaurate in the

Fischer rat was in the order of 2.25 to 2.5-fold with

clofibric acid and bezafibrate. This was less marked in the

Wistar rat following a 3 day treatment period (Sharma et

al.,1988). A similar differential induction was seen with the

12-hydroxylation of lauric acid. With clobuzarit treatment,

total lauric acid hydroxylation (11- and 12-) was unchanged in

the marmoset (Orton et al.,1984). The (omega-1)-oxidation

product of ME HP, the metabolite VI active in the rat, had no

effect (in vitro) on marmoset hepatocytes (Elcombe and

Mitchell,1986). In these latter 2 studies, total lauric acid

hydroxylation was measured. In the current study the omega-

and (omega-l)-hydroxy products were separated and quantitated Table 5.5.

Comparison of Hypolipidaemic Drugs On Lauric Acid Hydroxylase ft Activity in the Fischer Rat Following Chronic Administration.

DOSE LEVEL LAURIC ACID HYDROXYLASE ACTIVITY (nmol product per min per nmol P-450)

11-hydroxy laurate 12-hydroxy laurate

Control 2.13+/-0.51[a] (100)[b] 2.80+/-0.24 (100)

Ciprofibrate 6.74+/-3.80 (316) 57.7+/-11.3[d] (2060) (20 mg/kg)

Clofibric Acid 4.74+/-0.98[c] (223) 23.2+/-5.17[d] (829) (50 mg/kg)

Bezafibrate 5 .55+/-0.27[e] (261 32.7+/-2.43[e ] (1167 (75 mg/kg)

[a] Values are means +/- S.D.

[b] Figures in brackets are values expressed as percentage of corresponding control.

P values for results significantly different (Students T-Test) from control data are [c] < 0.05, [d] < 0.01 and [e] < 0.001.

# Single daily dose for 26 weeks 212

Figure 5.2.

The Effect of Ciprofibrate on Lauric Acid Hydroxylase Activity

in the Rat and Marmoset Following Chronic Administration.

LEGEND

2500 CONTROL L.D. 11-OH L.D. 12-OH 2000 JH.D. 11-OH CD 1 H.D. 12-OH DC £— 1500 rz: CD CD U, 1000 CD

500

RAT.F-344 MARM. 26 WKS HARM. REC ( COHTR0L=100 )

SPECIES 6 MONTH TREATMENT

Legend: L.D.11-OH Low Dose 11-hydroxylaurate

L.D.12-OH Low Dose 12-hydroxylaurate

H.D.ll-OH High Dose 11-hydroxylaurate

H.D.12-OH High Dose 12-hydroxylaurate

See following page for the absolute values and dose levels. Figure 5.2.(cont.)

Absolute Values (control);

Species/ nmol hydroxyproduct formed/min/nmol P-450

Dose Group 11-hydroxylaurate 12-hydroxylaurate

Rat [a] 2.13 +/- 0.51 2.80 +/.- 0.24

Marmoset (26 wks) 12.31 +/- 2.00 2.50 +/- 0.32

Marmoset (4 wk rec.) 11.57 +/- 1.50 3.07 +/- 1.84

Dose Levels;

Low Dose Marmoset (26 wks): 20mg/kg

High Dose Rat: 20mg/kg

Marmoset(26 wks): 80-100mg/kg

Marmoset(4 wks rec.): 80mg/kg

Single daily dose

[a] Data reproduced from table 5.5. 214

individually. In the untreated rat, the ratio of 12- to

11-hydroxylaurate was greater than 1 (1.30) however in the

marmoset this was reversed (0.20-0.30). The omega-1

hydroxylation of lauric acid predominated in the marmoset.

In the rat, cytochrome P-452 content, expressed as the

specific content (nmol/nmol P-450), was induced on

hypolipidaemic treatment and this was in the order of 18-, 11-

and 8-fold for bezafibrate, ciprofibrate and clofibric acid

(figure 5.3). When cytochrome P-452 content was expressed as

percentage of total cytochrome P-450, it was markedly

increased on treatment. The induction was 13-, 9- and 6.5-fold

with ciprofibrate, bezafibrate and clofibric acid respectively

(figure 5.4). It should be noted that the specific content of

cytochrome P-450 was increased with clofibric acid and

bezafibrate yet was non-significantly reduced with

ciprofibrate treatment. This would account for the cytochrome

P-452 content, expressed as percentage of total, being highest

in the ciprofibrate treated animals yet the specific content

being lower than the bezafibrate treated animals. Cytochrome

P-452 was not detected immunoc hemic ally in marmoset liver

microsomes (the minimal level of detection was an absorbance

change of double the blank reading corresponding to 0.002

pmoles/well).

In the untreated rat, cytochrome P-452 constitutes 3-4% of the

total CO-discernible cytochrome P-450 population. With a 3 day clofibrate treatment period this was increased to 22-25% in the Wistar rat (Sharma et al.,1988).This is comparable to the 215

Figure 5.3.

Comparison of Different Hypolipidaemic Agents on Cytochrome

P-452 Specific Content in the Rat Following Chronic

Administration.

CYTOCHROME P-452 CONTENT LE8END

.4000 I CONTROL | CLOFIBRATE | BEZAFIBRATE .3000 ia I CIPROFIBRATE E— z: CD CD .2000 CD

CD DJ Q, • 1000 CO

.0000 R A T F — 3 4 4

8 MONTH TREATMENT

Cytochrome P-452 Specific Content: nmol cyt. P-452/nmol P-450

Dose Levels;

Single daily dose Ciprofibrate 20mg/kg

Clofibric Acid 50mg/kg

Bezafibrate 75mg/kg 216

Figure 5.4.

Comparison of Different Hy.poJLipidaemic Agents on Cytochrome

P-452 Content in the Rat Following Chronic Administration

CYTOCHROME P -4 5 2 CONTENT LEGEND

CONTROL CLOFIB.ACIO BEZAFIBRATE MM CIPROFIBRATt

R A T F — 3 4 4

6 tlONTH TREATMENT

Cytochrome P-452 Content; nmol cyt. P-452 as percentage of

total cytochrome P-450.

( See figure 5.3 for absolute values)

Dose Levels;

Single daily dose

Ciprofibrate 20mg/kg

Clofibric Acid 50mg/kg

Bezafibrate 75mg/kg 217

present study where chronic clofibric acid administration caused an increase to 24%. In bezafibrate treated animals cytochrome P-452 content constituted 33-35% of the total and ciprofibrate 45-50%.

5.4. MITOCHONDRIAL ENZYME LEVELS

5.4.1 Monoamine Oxidase Activity.

In the rat, monoamine oxidase activity was significantly decreased by ciprofibrate treatment to 55-60% of the control value (figure 5.5). Activity in the marmoset was unchanged by treatment although a slight non-significant increase was seen at both dose levels with 26 weeks ciprofibrate administration

(figure 5.5).

5.4.2 Succinate Dehydrogenase Activity.

In the rat, succinate dehydrogenase activity was elevated on ciprofibrate treatment (figure 5.6). In the marmoset there was a dose-dependent increase in enzyme activity and which was significant at both dose levels (figure 5.6). The increase in enzyme activity was significant in the recovery group although induction was less extensive than in the 26 week dose groups. Figure 5.5.

The Effect of Ciprofibrate on Mitochondrial Monoamine Oxidase

Activity in the Rat and Marmoset Following Chronic

Administration.

MONOAMINE OXIDASE A C TIV IT Y LE6END

200 yiiliiiControl Low Dose H H High Dose 150 CD C£ E— :z: CD CJ 100 Cx, CD

50

RAT F-344 MARMo 26 WKS MARM« RECo ( C0NTR0L=100 )

SPECIES 6 MONTH TREATMENT

Absolute Values (control);

Species/Dose Group nmol product formed/min/mg protein

Rat 5.17 +/- 0.08

Marmoset (26 wks) 0.24 +/- 0.08

Marmoset (4 wk recovery) 0.47 +/- 0.18

Single daily dose

Dose Levels;

Low Dose Marmoset (26 wks) 20mg/kg

High Dose Rat 20mg/kg

Marmoset (26 wks) 80-100mg/kg

Marmoset (4 wk rec.) 80mg/kg £ . J . ?

Figure 5.6.

The Effect of Ciprofibrate on Mitochondrial Succinate

Dehydrogenase Activity in the Rat and Marmoset Following

Chronic Administration. LE8END

250 WM Control O il Low Dose

200 I ® High Dose

'NV-nV-.- co NN\\n>\ oc 150 1 CD C'PPP CD p p p 100 I! CD P i I n I t • '•$« 50 PM si

M SMS RAT F-344 MARM» 26 WKS MARMo REC C C0NTR0L=100 )

SPECIES 6 MONTH TREATMENT

Absolute Values (control);

Species/Dose Group nmol product formed/min/mg protein

Rat [a] 13.77 +/- 1.33

Marmoset (26 wks) 10.65 +/- 1.63

Marmoset (4 wk recovery) 10.32 +/- 1.92

[a] Data reproduced from table 3.2

Dose Levels;

Single daily dose

Low Dose: Marmoset (26 wks) 20mg/kg

High Dose Rat 20mg/kg

Marmoset (26 wks) 80-100mg/kg

Marmoset (4 wk rec. 80mg/kg 220

5.4.3 Alpha-glycerophosphate Dehydrogenase Activity

Alpha-glycerophosphate dehydrogenase activity was

significantly induced 2-fold in the rat by ciprofibrate

(figure 5.7). With a 26 week ciprofibrate administration in

the marmoset, enzyme activity was increased in a

dose-dependent manner and the 2-fold increase at the high dose

level was significant (figure 5.7). In the recovery group

where treatment was discontinued for 4 weeks prior to

sacrifice, alpha-glycerophosphate dehydrogenase activity was

unchanged.

5.5. MITOCHONDRIAL AND PEROXISOMAL CARNITINE

ACETYLTRANSFERASE ACTIVITY.

In the rat, a 64-fold induction in total carnitine acetyltransferase activity was observed with ciprofibrate pretreatment (figure 5.8). In the marmoset, enzyme activity was unchanged by treatment (figure 5.8) and in the recovery group values were similar to control. The observed increase in the rat was consistent with the published report of a 58-fold increase following the dietary administration of ciprofibrate for 14 days (Lalwani et al.,1983b). This enzyme has not previously being measured in the marmoset. However, a newly developed peroxisome proliferator DL-040 (as shown in figure

5.1) induced this enzyme in the Rhesus monkeys (Lalwani et al.1985) and ciprofibrate administration was reported to increase enzyme activity in the Rhesus and Cynomolgus monkey

(Reddy et al.1984). Figure 5.7.

The Effect of Ciprofibrate on Mitochondrial

Alpha-Glycerophosphate Dehydrogenase Activity in the Rat and

Marmoset Following Chronic Administration.

enxnns 250 nr™ Control Low Dose MS High Dose 200 i H IM

C D ilk i n i«t E— ■ 150 sa: CD Mf C_D II i H 1 Cju 100 I*. CO M i i t ill 50 i n H ■S-jJsS n ii

RAT F-344 MARMo 2G WKS MARMo REC ( C0NTR0L=100 )

SPECIES S MONTH TREATMENT Absolute Values (control).

Species/Dose Group Change in O.D./min/mg protein

Rat [a] 0.02 + / - 0.01

Marmoset (26 wks) 0.01 + / - 0.01

Marmoset (4 wk recovery ) 0.01 + / - 0.01

[a] Data reproduced from table 3.2.

Dose Levels:

Single daily dose

Low Dose Marmoset (26 wks) 20mg/kg

High Dose Rat 20mg/kg

Marmoset (26 wks) 80-lOOmg/kg

Marmoset (4 wk rec. 80mg/kg Figure 5.8.

The Effect of Ciprofibrate on Total Carnitine

Acetyltransferase Activity in the Rat and Marmoset Following

Chronic Administration. LE8END

7000 CONTROL LOI DOSE GO 00 il HIGH DOSE

5000 CD

£— ■ rz: 4000 CD CD 3000 LU CD ir? 2000

1000

RAT F-344 HARM.26 WKS MARM.RECo ( C0NTR0L-100 )

SPECIES 6 MONTH TREATMENT Absolute Values (control);

Species/Dose Group nmol product formed/min/mg protein

Rat [a] 1.53 +/- 0.55

Marmoset (26 wks) 17.51 +/- 1.87

Marmoset (4 wk (recovery ) 13.40 +/- 3.48

[a] Data reproduced from table 3.3.

Dose Levels;

Single daily dose Low Dose Marmoset (26 wks) 20mg/kg

High Dose Rat 20mg/kg

Marmoset (26 wks) 80-100mg/kg

Marmoset (4 wk rec.) 80mg/kg 223

5.6 PEROXISOMAL ENZYME ACTIVITIES

5.6.1 Catalase Activity.

Catalase activity was significantly decreased to 50% of the

control activity in the rat following the chronic

administration of ciprofibrate (table 5.6). In the marmoset,

catalase activity was unaffected by treatment at both time

points. A 2-fold increase in total liver catalase activity has

been reported in the Fischer rat following the dietary

administration of ciprofibrate (0.10% w/w) (Lalwani et

al.,1983a,b). With the short term administration of

ciprofibrate it was found that the specific activity of

catalase was unchanged in the Fischer rat but that the

doubling in liver/body weight ratio accounted for the reported

2-fold increase in total liver catalase activity. However,

with the chronic administration of ciprofibrate, hepatomegaly

was maintained and the specific activity of catalase decreased

indicating that synthesis of catalase was not maintained.

Catalase activity was unaffected by clobuzarit administration

in the marmoset (Orton et al.,1984), but in the DL-040 treated

Rhesus monkey catalase activity was reported to be doubled

(Lalwani et al.,1985).

5.6.2 Peroxisomal Beta-Oxidation.

In the rat, ciprofibrate treatment resulted in a significant

35-fold induction in peroxisomal beta-oxidation (figure 5.9).

With a 26 week ciprofibrate treatment in the marmoset, a o c— W B O rH 0; . H Qj V 44 r— . r a o 0 0 34 0 ) 03 •—' 0 Qi 0 CO •H 44 44 i n i— i W H O' U u IH rH 0 • o 6 a) V JQ •P 44 CO CN O 0 34 a) 0 M i ) 01 a Q I I 02 V Qt N \ z) e 44 44 4- 4- V o O rH CO H 44 z z O N* O 3 £ B O CO c * d CN CN 03 W S JB 44

B HH o 34 44 in c 03 u b o •H B 0 u 44 O : 4 M •H •H o O —' O —» 44 o in O CO CP > r4 co rH E •H ro a ♦H 44 34 44 *0 £ O D) B O < m •H E‘ O 34 Qi U4 03 B 6 3 e CO ^ in in CN c— cn in rH rH O CO O rH a E 44 N 'V • • • • 34 o O o o M in B < o o o $ z o 0 6 3 t I u u 5 *E i I I t I I 44 rH •»H w s \ N. 1 u E 4-4-4- 4- 4- M-4 E-* 0 B 0 £ O o >4 in * >4 CP o M CP 34 CP 44 34 M 10 O 63 44 £ < H 44 £ £ to 44 Qi Oh Oi s Q m3 B 1 B c O o o o X rH •H O o o 44 o •a O CN CO u co O CN £ 0 o B B 34 0 0 44 ro 44 O 0 U 44 3 • r H • iH 44 in 44 (0 * a © 0 —H o U Q£ in • • 03 40 u > B 44 44 03 3 a CP o B 03 0 3 « 0 •H 44 V 63 03 IS O CO 34 0 £ 34 0 r- H 03 44 -Sfi Q i i 0 i ro 0 N tn 44 HD 43 6 44 r—1 M l B-i O E-i 03 > H 34 4 - Z 34 JQ 0 3 Z 63 10 44 in to 10 40 0 0 o z 40 03 40 44 B U 0 34 0 40 03 34 03 0 0 0 34 0 E-* <3 a) 03 IS 03 03 0 0 J 0 34 £ JQ 34 0 < 63 I* 03 O > IS ( 0 IS rH O 44 O 44 02 OS o B> Em lO rH U VjO 0 0 B 44 0 d CN <4> O 03 CN ■D' 34 34 • H 0 3 CN 44 34 0 0 > 0 0 0 rH in in 0 3 o 0 0 34 i— ( 34 U 3 3 0 44 >x> • * a 44 rH CP > B 03 34 0 •H O in CO 0 © in in u r> > 6 i Qi U 63 UH 0 o O ! o ) £ £ • • 0 0 rH o ; C 0— + r— i-^ XI 63 u >4 44 0 Qj 0 ro ro 0 X) U EH CO S Z 0 2 w 225

Figure 5.9.

The Effect of Ciprofibrate on Peroxisomal Beta-Oxidation in

the Rat and Marmoset Following Chronic Administration.

LE8END

3500 CONTROL [ H LO T DGSr! 3000 i H H HIGH DOSE

2500 o 111 C£ E—" 2: 2000 O CO 1500. Cjl is CD i f 1000.

500

KZCZZSZS .I;,. .ia . _ E Z 5 3 — — tsssssa. RAT F-344 MARMo 26 WKS MARMo REC ( CONTROL=100 )

SPECIES S MONTH TREATMENT

[a] Assayed as beta-oxidation of palmitoyl-CoA.

Absolute Values (control);

Species/Dose Group nmol product formed/min/mg protein

Rat [a] 2.40 +/- 0.31

Marinos et (26 wks ) 3.18 +/- 0.87

Marmoset (4 wk recovery) 0.92 +/- 0.25

[a] Data reproduced from figure 3.1.

Dose Levels;

Low Dose Marmoset (26 wks) 20mg/kg

High Dose Rat 20mg/kg

Marmoset (26 wks) 80-100mg/kg

Marmoset (4 wk rec.) 80mg/kg

Single daily dose 226

significant dose-dependent increase in peroxisomal

beta-oxidation was observed (3.5-fold) although this

represented a response considerably lower than that in the rat

(figure 5.9). In the marmoset recovery group activity was

similar to control levels.

Comparing the above data to that previously reported, a

14-fold increase in peroxisomal beta-oxidation in the Fischer

rat was observed following the short term dietary

administration of ciprofibrate (0.10% w/w) (Lalwani et

al.,1983b). In contrast, clobuzarit (45 mg/kg) treatment in

the marmoset did not induce peroxisomal beta-oxidation (Orton

et al.,1984). Consistent with the above, is the report that

the (omega-1)-oxidation product of ME HP induced peroxisomal

beta-oxidation in rat hepatocyte cultures yet had little

effect on marmoset hepatocytes (Elcombe and Mitchell, 1986). A

slight increase was observed but this was not dose-dependent.

Using excessively high doses of ciprofibrate, an increase in

peroxisomal beta-oxidation was demonstrated in the Rhesus and

Cynomolgus monkey (Reddy et al.,1984). In a similar manner

DL-040 induced peroxisomal beta-oxidation in the Rhesus monkey

(Lalwani et al.,1985). The administration of LY-171883,

(tetrazole-substituted alkoxyacetophenone, figure 5.10), a

rodent peroxisome proliferator, to Rhesus monkeys at the

maximum tolerated dose revealed no changes in peroxisomal

beta-oxidation or catalase levels (Eacho et al.,1986). In the

current study the maximal 3.5-fold increase in peroxisomal beta-oxidation in the marmoset was clearly of much less

significance when compared to the 34-fold induction in the 227

Figure 5.10.

Chemical Structure of LY- 171883, a Tetrazole-Substituted

Alkoxyacetophenone

0 II

H

;scher rat considering the dose levels were 20 and 80-100

/kg for the rat and marmoset, respectively.

5.6.3 Uricase Activity

Uricase activity in the rat was unaffected by the chronic

administration of ciprofibrate (table 5.6). In these studies

in the marmoset no uricase activity could be detected in

contrast to the Rhesus monkey (Old World genus) where an

increase in uricase activity was observed following

ciprofibrate and DL-040 administration (Reddy et al.,1984,

Lalwani et al.,1985).

5.7 CYTOSOLIC GLUTATHIONE PEROXIDASE ACTIVITY.

Glutathione peroxidase activity was measured using t-butylhydroperoxide as the substrate. In the rat, treatment 228

with ciprofibrate and bezafibrate resulted in a significant

decrease in enzyme activity (65 and 78% of the control value

respectively) (table 5.7 and figure 5.11). By comparison,

glutathione peroxidase activity was unchanged in the rat

following clofibric acid administration. In the marmoset

enzyme activity was unchanged by ciprofibrate treatment (table

5.7).

5.8 THE MOLECULAR BIOLOGY OF CYTOCHROME P-452 AND

GLUTATHIONE PEROXIDASE

As shown previously (chapter 4), ciprofibrate administration

to the rat resulted in the induction of cytochrome P-452, as

measured catalytically and immunochemically. In addition,

following a 14 day treatment period this induction correlated

to a marked increase in the amount of mRNA hybridising to the

cytochrome P-452 cDNA probe. It was not possible to correlate the decrease in glutathione peroxidase activity with gene expression as technical difficulties were encountered using this glutathione peroxidase probe (chapter 4).

Tissue from chronically treated rats and marmosets was limited and in the following studies was unsupported by biochemical and immunochemical data in the rat. It should be remembered that the objective at the start of the research work was to examine species and strain differences at a biochemical level only and that the decision to implement a molecular approach to study species susceptibility was a late development. 3

>-t 3 E-* 4 4 r — « ► H 3 u >1 > 3 O IH •H 4-9 o u • H o r — » <—* o c— E-» g o r - X) > o o cn r - o r o rH IX c— rH wo a •H < M C o 4-» 3 •pH U Ed Or 3 u < C CO 4 4 O O' O < 33 09 c 09 •H a 3 M X •^4 to 4-9 H-r CO Or *— 1 -H ro iH CO ro m CN T3 3 3 X C r H o O o o o O O 33 M O cn o ♦H 44 0 2 O g • • • o o o o o O O 04 X to Ed X to o •H Or 3 M 4 4 i 1 1 I I I I 09 M c 3 M x » 09 •H Ed 3 Or N "N \ \ N X + X -+ • + M to Or . g Z g O 09 'S o U 4 - 9 09 IH r H < CO rH C— CO 1 e X o ro g rH O »H u • iH •H C 3 • • o o o o o to X c E-* 09 4 - 9 44 o 09 cn c 3 M X ro 09 X 44 o x> 3 3 3 u 09 c 3 rH X> 09 4 - 9 C3 cn 3 u CO c M u U x > 09 •H > CO 0 4 rH o rH XI O 3 to 3 rH tn O O to to \ 44 Etr 09 T3 >1 S' cn cn X 09 U 44 cn g r H AS r H AS X) to 3 rH AS Ed XI N. \ to CO 0 ^ 0 o o C M cn to 09 o O M 0>0 M cn 44 g rH 44 g 44 g rO M g Q Ed 0 4 C I C C 3 H X3 09 X 44 3 o o <=> o o o O N 09 o CO o ra 0 9 • 3 Z •H CN M x 10 * X 33 0 9 r o •*H 0 C M 3 XI 3 t o 09 rH 09 o t0 H •H O w -P AS 01 0 • > 3 c • Or 3 44 3 3 m a cn Cn o • H 02 c Qj *H •H 3 3 e • CJ g VX CO W U-l to v 3 44 x XI X rH I to r— O 44 3 18 e 3 3h S \ to o -P 09 Esr rH * M X x > X 4 - 9 M 44 C O E-1 09 X I 3 H Z 4 4 3 U CO X I to 10 09 3 Z Ed (0 'O > 1 -X 3 M A£ 1 c U HD 09 U XI O Z AS X9 (Q i h E-< 0) 3 > 3 3 (0 0) M rO XI 09 M E-* << 3 O > 3 0 U Ed 09 X I 3 M ro < Ed £ £ r H O > X g |H CO x > 33 O 44 3 02 02 U WO WO o 3 w o O •h 09 C O x i ro JZ S E-< •h u Q cn CN XI U CN TJ > H •3 E-* ro Or to 1*0 CO 09 09 rH 09 CO 09 M 3 O 09 rH W 3 M ro 3 3 ro tO 4-9 ■o' 44 x> tO rH t n X 9 > C 3 0) to ro •H tO O CO 0 9 EX4 O Ed to to 39 ™ < > 04 U o o w D) 0 g g C Ed M w J4 01 (0 ro ro C/3 *5 X C9 *3 CO z z 02 r- I • I m i oi 1i rH I X I I E-*3 I Comparison of Different Hypolipidaemic Agents on Cytosolic

Glutathione Peroxidase Activity in the Fischer Rat Following

Chronic Administration.

LE8END

1 5 0 gg| CONTROL m CLOFIB.ACID | j BEZAFIBRATE -3 WM CIPROFIBRATE CD loo O' E— 2; CD CD Lu CD 50 y e

R A T F - 3 4 4 ( C0NTR0L=100 )

6 MONTH TREATMENT

Legend;

Clofib. Acid Clofibric Acid

Data reproduced from figure 3.2.

Absolute Values (control);

0.57 umol substrate metabolised/min/mg protein

Substrate : t-butylhydroperoxide

Single daily dose

Dose Levels;

Ciprofibrate 20mg/kg

Clofibric Acid 50mg/kg

Bezafibrate 75mg/kg 231

Total hepatic RNA was isolated from;

1) frozen tissue from untreated Fischer rats and animals from

the following treatment protocols; ciprofibrate - 20 mg/kg by

gavage once every 48 hr., clofibric acid - 75 mg/kg by gavage

twice a day and bezafibrate - 150 mg/kg by gavage twice a day.

The twice daily administration was separated by 8-12 hr and

the duration of treatment was 9 months. With a similar dosing

regimen for 6 months it was found that the relative potencies

of the compounds were similar with respect to mitochondrial,

peroxisomal and cytosolic enzyme parameters (chapter 3).

Unfortunately there was no data at this 9 month time point

however the effects at 6 months with daily dosing were very

similar to that at 2 weeks and it was therefore assumed that

lauric acid hydroxylase activity and cytochrome P-452 were

probably elevated in the treated groups and glutathione

peroxidase activity decreased.

2) untreated and treated marmoset frozen tissue. The treatment

period was 26 weeks with daily dosing at 80-100 mg/kg (100 mg/kg for 21 weeks with the final 5 weeks at 80 mg/kg).No

samples were available from the recovery group.

Duplicate RNA dot filters were prepared using the isolated RNA with the following concentrations: 0.01, 0.10. 1, 10 and 20 ug

RNA/dot. The filters were hybridised with the cytochrome P-452 cDNA and genomic GPX probes which hybridised to all the RNA samples (figures 5.12 and 5.13). In the rat, the levels of mRNA coding for cytochrome P-452 were markedly increased in treated animals (figure 5.12). By inspection, it appeared that 232

Figure 5.12.

The Effect of Ciprofibrate, Bezafibrate and Clofibric Acid

Treatment on Hepatic Cytochrome P-452 mRNA Concentration in

the Rat and Marmoset After A 26 Week Administration Period.

o CD •+— » CD -4— ' -4— ' 03 CD 03 03 -4— ' v _ 03 — -Q JD JQ i_ o H— H— M— n o 03 •4— » V+— oc N c ' C Q. o Q. o CD o O) o o b GQ o o o CO

-I—* o 0 CVJ • •• Q CN • • LO 1 CL

0 E o _c o H o LL) > CO o o oc

the intensity of the hybridisation to RNA dots from the

clofibric acid and bezafibrate treated rats appeared to be

slightly higher than that in the ciprofibrate sample but as

mentioned previously more accurate comparisons could not be

made. It should be noted that in acutely treated rats the

increase in the concentration of mRNA hybridising to the

cytochrome P-452 cDNA probe accompanied a marked induction of

lauric acid omega-hydroxylase enzyme activity and an increased

cytochrome P-452 specific content (chapter 4). In the marmoset

there was a minimal increase in the hybrid intensity in the

treated animal indicating a slight increase in the mRNA coding

for cytochrome P-452 (figure 5.12). However, this accompanied

no immunodetectable cytochrome P-452 protein and a 4 0 and 60%

decrease in (omega-1)- and omega-hydroxylation of lauric acid,

respectively. Accordingly, this apparent increase in mRNA

hybridising to the probe did not appear to correspond to

increased translation of catalytically active protein.

The genomic GPX probe cross-hybridised with both the rat and

marmoset (figure 5.13). In the ciprofibrate and clofibric acid

treated rats the intensity of hybridisation was very similar

or slightly less intense than the control sample suggesting

treatment-induced differences in glutathione peroxidase mRNA

synthesis. By comparison in the bezafibrate treated animal,

the hybridisation response did not appear to be a linear

function of the RNA loaded and was clearly more intense than

the control sample. In both control and ciprofibrate treated with the marmosets a non-linear hybridisation was probe

(figure 5.13). 234

Figure 5.13.

The Effect of Ciprofibrate, Bezafibrate and Clofibric Acid on

Hepatic Glutathione Peroxidase mRNA Concentration in the Rat

and Marmoset After A 26 Week Administration Period.

0 CD 0 -t—» 03 03 _Q cc _ -Q _ -Q o c: < O ^ ^ -Q _h o z o ^ O 03 — ■t «- DC Q P «- N ^ E Q- o £■ 0 — o — 0 O) C/3 o o CO o o o 03 T3 • • 2 0 >< i—O 0 o CL o 0 c CD o !c 1- LU 03 -*— > CO D o o DC < < dc 235

The evaluation of changes in hybrid signal intensity should be

made with caution especially when low amounts of radioactivity

are involved. Accuracy in determining RNA concentrations is of

paramount importance. It should be emphasized that these

experiments with the GPX probe were preliminary and that

further experimentation is required to;

1) optimise hybridisation conditions,

2) complete biochemical and immunochemical analyses, and,

3) use more animals instead of making comparisons between one

untreated and one treated animal.

As described in chapter 4, the GPX gene is strongly conserved

between species whereas the cytochrome P-450 system comprises

of at least 10 gene families.

With respect to hybridisation with the cytochrome P-452 cDNA

probe, two assumptions are made;

1) the cytochrome P-452 probe hybridises equally to mRNA(s)

coding for this protein in the rat and marmoset, and

2) the cDNA probe does not hybridise to other cytochrome P-450 mRNA(s) to any great extent. It has been established that in the rat, clofibrate administration resulted in a 5 to 7-fold increase in cytochrome P-450 LAw (P-452) mRNA levels which closely paralleled the increase in immunodetectable protein

(Hardwick et al.,1987). This accompanied a marked elevation in the transcription of the cytochrome P-450 LAw gene. One would assume a similar transcriptional activation in the rat with ciprofibrate treatment as a marked increase in mRNA hybridising to the probe with an associated induction of cytochrome P-452, measured catalytically and immunochemically, 236

was observed. The small increase in cytochrome P-452 mRNA in

the marmoset was accompanied by a reduction in catalytic

activity and no immunodetectable protein, indicating that

there was no increase in mRNA translation. This implies that a

distinct species difference in gene expression and regulation

exists.

5.9 DISCUSSION

The main objective of these studies was to examine and

describe the chronic effects resulting from ciprofibrate

treatment in the rat and marmoset. The biological response in

the rat following short term administration has been

documented in the previous chapter. In summary, the chronic

hepatic responses in the rat are;

1) hepatomegaly,

2) an induction of cytochrome P-452 with an associated

decrease in the metabolism of model drug substrates,

3) increased peroxisomal beta-oxidation, unchanged catalase

activity and induction of total carnitine acetyltransferase

activity,

4) mitochondrial enzyme changes, and

5) decreased glutathione peroxidase activity.

Essentially, the short term responses are duplicated in the

chronically treated rat. Hepatomegaly and the observed

cytochrome P-450 changes persisted in animals treated with all

three hypolipidaemic agents. An increase in the

11-hydroxylation of lauric acid was apparent but this was not significant with ciprofibrate. The specific activity of 237

catalase was halved in the ciprofibrate and bezafibrate

treated animals. With a 14 day treatment period, monoamine

oxidase activity was decreased, succinate dehydrogenase

activity unchanged and alpha-glycerophosphate dehydrogenase

activity increased. A similar series of enzyme changes were

seen in chronically treated animals except that succinate

dehydrogenase activity was increased in the ciprofibrate

treated group. Glutathione peroxidase activity was decreased

with ciprofibrate and bezafibrate treatment but not with

clofibric acid.

It is instructive for the purposes of this discussion to

reiterate the acute hepatic changes occurring in the marmoset;

1) an increase in monoamine oxidase, succinate dehydrogenase

and alpha-glycerophosphate dehydrogenase activities,

2) a 3.5 fold increase in peroxisomal beta-oxidation and no

apparent uricase activity suggesting that the marmoset is a

member of a New World genus devoid of this enzyme, and,

3) an increase in total cytochrome P-450 specific content

associated with a decrease in benzphetamine-N-demethylase

activity and unchanged lauric acid hydroxylation and ethoxyresorufin-O-deethylase activity. Cytochrome P-452 was not detected immunochemically.

It should be noted that the regioselectivity in lauric acid hydroxylation was different in the rat and marmoset with 11 - hydroxylation predominating in the marmoset. With 2 6 weeks continuous ciprofibrate treatment, the increases in succinate dehydrogenase and alpha- glycerophosphate dehydrogenase 238

activities were maintained. A slight non-significant elevation

in monoamine oxidase activity was observed. Similarly a 3.5

fold increase in peroxisomal beta-oxidation was apparent but

the specific content of cytochrome P-450 was unchanged on

treatment. The hydroxylation of lauric acid was significantly

decreased at both positions in the high dose group. In the

recovery group (26 weeks treatment at 80-100 mg/kg/day

followed by 4 weeks with no treatment), a minimal but

significant increase in succinate dehydrogenase activity

persisted and the 11-hydroxylation of lauric acid was

similarly decreased 60% of the control value.

It is clear that marked differences exist between the rat and

marmoset. In both species the hepatic changes resulting from

chronic administration were very similar to the responses

observed with short term treatment. In the rat the fundamental

change was in catalase activity. The specific activity was

unchanged at 2 weeks but this was associated with a doubling

in liver/body weight ratio which would account for the

published 2-fold induction in total catalase activity (Lalwani

et al.,1983b). At 26 weeks, the specific activity was halved

although hepatomegaly was maintained suggesting that with chronic administration, synthesis of new protein does not occur or that the proliferated peroxisomes contain very low enzyme levels. The toxicological implications arising from the observed decrease in catalase activity are discussed in chapter 7. The difference in regioselectivity of lauric acid hydroxylation in the marmoset may be the result of enzyme kinetics dependent on substrate concentration for the 239

competing omega- and (omega-1)-fatty acid hydroxylases.

A subchronic (3 month) toxicity study was completed in the

Fischer rat with dose levels of 0.4, 1.6, 6.4, 25 and 100

mg/kg/day ciprofibrate (Sterling-Winthrop, personal

communication). In terms of liver weight, a no-effect dose

level was not achieved as hepatomegaly was observed in all

groups. No biochemical analyses were completed but increased

lauric acid hydroxylation and peroxisomal beta-oxidation was

assumed to occur as hepatomegaly has always been associated

with enzyme changes but not vice versa (Lazarow et al.,1982).

In the 2 week ciprofibrate study, a marked increase in

peroxisomal beta-oxidation, carnitine acetyltransferase

activity and cytochrome P-452 content was observed at 2 mg/kg.

The pharmacokinetic dose response relationship in the rat for

2 and 20 mg/kg was not linear (figure 5.14). This would

indicate that the 2 mg/kg dose level was on the plateau region

of the dose-response curve. In figure 5.15, it is apparent

that following a 3 month continuous ciprofibrate treatment,

the plasma concentration in rats administered 25 and 100 mg/kg/day ciprofibrate was very similar 24 hr post dose

indicating that as the dose increases, absorption decreases

(Sterling-Winthrop personal communication). Steady state levels were achieved within 7 days with once daily dosing

(Sterling-Winthrop, personal communication). This would suggest that the drug absorption capacity was saturated at high dose levels. In the marmoset, the high dose level of

80-100 mg/kg ciprofibrate would likewise appear to correspond to a maximal dose/response relationship (figure 5.16) and X lttin peroxidase Glutathione PX Acetyltransferase. Carnitine AT

J arc cd omega-hydroxylase. acid Lauric UJ % % OF CONTROL P Peroxisomal beta-oxidation. Peroxisomal P oe epne uv i h FshrRt olwn te Short the Following Rat Fischer the in Curve Response Dose iue 5.14. Figure ige al ds fr 4 days 14 for dose daily Single 5000 2000 3000 4000 6000 7D00 1000 IRFBAE DS RSOS ) RESPONSE ( DOSE CIPROFIBRATE emAmnsrto o Ciprofibrate. of Administration Term ICE RAT FISCHER L E V E L E S O D 8

.. G K / G M AT

12 240

18 0 CONTROL®100 ) t20 ^ 241

Figure 5.15.

Plasma Concentrations of Ciprofibrate in Albino Rats Medicated

Orally Once a Day For 3 Months.

iooo

o»

CO

0 3

10 4

TIME AFTER LAST OOSE(hours) 242

Figure 5*16.

Dose Response Curve in the Marmoset Following the Short Term

Administration of Ciprofibrate.

CIPROFIBRATE ( DOSE RESPONSE ) MARMOSET

400

300 o

E— o o 200 cx CO PX 100

20 40 eo B0 100 C C0NTR0L= 100 ) MG/KG DOSE L E V E L

^ Single daily dose for 14 days AT Carnitine acetyltransferase, PX Glutathione peroxidase, p Peroxisomal beta-oxidation, u3 Laurie acid omega hydroxylase.

it should be emphasized that the dose level was the maximal tolerated dose on a daily basis.

A 13 week recovery period was incorporated into the 3 month rat study (Sterling-Winthrop, personal communication). Slight histopathological liver changes were seen in the 25 and 100 mg/kg dose levels, 13 weeks following cessation of treatment.

The hepatic enlargement and eosinophilic change in the 243

cytoplasm of hepatic parenchymal cells seen in this study were

similar to those described by Paget,(1959). In another study,

animals were treated with ciprofibrate (0.2, 1, 5 and 25

mg/kg) for 6 months followed by a 3 month recovery period

(Sterling Winthrop, personal communication). With respect to

liver weight, a no-effect dose was not achieved and

significant liver enlargement was seen at 25 mg/kg 13 weeks

following the cessation of treatment. It is unfortunate that

no biochemical analyses were completed to ascertain the

relationship between hepatomegaly and enzyme levels following

an extensive recovery period. However, reversal of biochemical

parameters has been reported in rats administered clofibrate

for 14 days with a 6 day recovery period prior to sacrifice

(Orton et al.,1982). Cytochrome P-450 content and

ethoxycoumarin deethylation were still significantly increased

and the inhibition of ethoxyresorufin O-deethylation was

reversed. The liver enlargement was maintained unchanged for 6

days after treatment (Orton et al.,1982). Peroxisomal

beta-oxidation, carnitine acetyltransf erase and lauric acid

hydroxylase assays were not completed. The increases in liver

weight and peroxisomal beta-oxidation in the rat resulting

from a 14 day treatment period with LY-171883 were reversed

within 16 days of discontinuing treatment (Eacho et al.,1986).

In this study the increase in peroxisomal beta-oxidation in

the marmoset was readily reversible following a 4 week

recovery period although mitochondrial succinate dehydrogenase

and microsomal (omega-1)-lauric acid hydroxylase activities were still Effected (p<0.05). 244

Total hepatic RNA was isolated from rats treated with

clofibric acid (2 x 24 hr), bezafibrate (2 x 24 hr) and

ciprofibrate (1 x 48 hr) for a period of 9 months and from

untreated and ciprofibrate treated marmosets. The samples were

hybridised with the cytochrome P-452 cDNA and GPX genomic

probes. A marked increase in the mRNA hybridising to the

cytochrome P-452 cDNA probe was observed in the treated rat

samples. A slight increase in the treated marmoset was

apparent but this did not correlate to increased catalytic

activity and was associated with no immunodetectable protein.

No clear treatment differences were visible with hybridisation

to the GPX genomic probe. The inability to detect cytochrome

P-452 in marmoset liver microsomes using an immunochemical

technique (ELISA) was discussed in the previous chapter.

In the rat, transcription of the cytochrome P-450 LAw (P-452)

gene was elevated as early as 1 hr after clofibrate treatment

and this accompanied an increase in the mRNA species coding

for the protein and increased levels of immunodetectable protein (Hardwick et al.,1987). A similar transcriptional activation of the gene is assumed in the rat treated with ciprofibrate and bezafibrate. The apparent, albeit small increase in mRNA-cDNA hybridisation for cytochrome P-452 in the treated marmoset was not accompanied by an increase in catalytically active protein. It is possible that either the level at which cytochrome P-452 gene expression is controlled is different in the two species, e.g. ribosome selection among mRNAs, mRNA degradation control or, more likely, that the small increase in mRNA synthesis was insufficient to give a 245

marked increase in active enzyme. It should be remembered that

the control of gene expression is multifaceted with complex

regulatory mechanisms.

The peroxisome prolif erator-induced cytochrome P-452 is a

member of the cytochrome P-450IV gene family and the amino

acid sequence is <33% similar to that encoded by genes in any

of the other 7 mammalian P-450 gene families (Hardwick et

al.,1987). The rat cytochrome P-452 cDNA probe hybridised with

mRNA from 6 different species (treated and untreated animals)

indicating that cytochrome P-452 is a constitutive form with

strong gene conservation across species. By comparison, liver

microsomes prepared from rabbit, hamster, gerbil and mouse

contain inducible cytochrome(s) P-450 that resemble cytochrome

P-450p (rat liver cytochrome P-450 induced by steroids), both

catalytically and immunoc hemic ally (Wrighton et al.,1985).

These proteins were encoded for by mRNAs of sufficient

homology to hybridise specifically to cloned cDNAs to P-450p mRNA. However in this case, there were striking qualitative

and quantitative interspecies differences in the responses to

inducers (Wrighton et al.,1985). In all species examined there was a notable disparity in the magnitude of induction of immunoreactive protein as compared to catalytic activities with immunoreactive protein exceeding enzyme activity. In the current studies the reverse was true for the rat chronically treated with ciprofibrate, bezafibrate and clofibric acid and in the hamster following a 14 day treatment period. With short term ciprofibrate administration, the increase in immunodetectable protein matched the induction of omega-hydroxylation in the mouse, rabbit and Long Evans rat.

In the remaining rat strains the induction of cytochrome P-452 measured catalytically exceeded that determined immunoc hemic ally. These strain and species differences may be explained by unequal synthesis of the apoprotein and holocytochrome.

The conservation of highly homologous gene sequences across species is well documented . in the cytochrome P-450 monooxygenase system (Nebert et al.,1987). The existence of at least two cytochrome P-450 genes in humans that are highly homologous to the TCDD-inducible cytochromes P-450 in rabbits and mice has been demonstrated (Quattrochi et al.,1985).

Regions known to be highly conserved in cytochrome P-450 isoenzymes isolated from rat, rabbit and mouse were found to be conserved in the amino acid sequence derived from pHP450(l)

[cDNA clone to PB-inducible cytochrome P-450e] (Philips et al.,1985). TCDD induced P-4504 and P-4506 mRNAs in the rabbit and the sizes of these mRNAs differed from their analogues in the mouse as a result of divergence in the 3' untranslated portions of the mRNAs (Okino et al.,1985).

In general the control of protein synthesis may be due to transcription, processing, transport, translation or degradation of mRNA. The control of gene expression seems to operate at the level of DNA transcription i.e. the synthesis of the primary pre-mRNA transcript. This is thought to involve a relatively small set of gene regulatory proteins e.g. growth factors, differentiation factors, receptors, DNA-binding 247

proteins etc., that modulate mRNA synthesis in order to

produce and maintain a large number of different cell and

tissue types. The actual mechanisms by which most eukaryotic

gene regulatory proteins act are not clearly defined. However

chromatin must be selectively decondensed prior to

transcription, and this is probably the primary role of gene

regulatory proteins. Other further levels of control of gene

expression have now been documented for the cytochrome P-450

superfamily. For example, PB and 3-MC elevate the level of rat

liver glutathione S-transferases primarily by augmenting the

transcriptional rates of their respective genes (Ding and

Pickett, 1985), possibly by enhancing RNA polymerase binding.

On the other hand, nuclear RNA transport has been implicated

in the regulation of cytochrome P-450c induction (Foldes et

al.,1985). In diabetic rats cytochrome P-450j mRNA increased

10-fold but this was not due to transcriptional activation but

rather to specific stabilisation of the cytochrome P-450j mRNA

(Song et al.,1987). To assess the conservation of the

cytochrome P-452 gene between species requires further

investigation, and the precise molecular mechanisms governing

gene expression, the influence of peroxisomal proliferators on

DNA binding proteins, transcriptional activation and mRNA

accumulation need to be clarified. The functional control of

the peroxisomal beta-oxidation enzymes and glutathione

peroxidase genes are clearly areas that need to be examined.

The administration of Wy-14,643 to the rat causes a marked

increase in the concentration of mRNA species coding for the bifunctional protein - 3-ketoacyl-CoA thiolase and enoyl-CoA 248

hydratase (Chatter jee et al.,1983). In DEHP treated animals,

the hepatic levels of translatable mRNAs coding for acyl-CoA

oxidase and the bifunctional protein were elevated (Furuta et

al.,1982). The cDNA probe to the latter protein has been

sequenced and the enzyme has no terminal peptide extension as

a signal for translocation into peroxisomes (Osumi et

al.,1985). With ciprofibrate and clofibrate administration,

mRNA levels coding for catalase, fatty acyl-CoA oxidase and

the bifunctional protein were elevated and the rate of

transcription of the beta-oxidation enzyme genes were

increased (Reddy et al.,1986). The rate of transcription was

nearly maximal by 1 hr post dose and was maintained for at

least 16 hr. In W y - 14,643 treated animals there was no

increase in the transcriptional rate of the bifunctional protein at 5 hr post dose and a 2- and 11-fold increase was seen at 10 and 20 hr post dose (Chatterjee et al.,1987). The increased activity of this protein was attributed to an enhancement of the rate of transcription of the bifunctional protein gene. However, the relatively long lag period suggested that the induction of the bifunctional protein mRNA involves an indirect effect (binding of these peroxisome proliferators to DNA has not been demonstrated) on the transcription of this gene (Chatterjee et al.,1987). The cause of this apparent discrepancy between the rate of transcription in these two studies is unclear at present. Wy-14,64 3 is considered a more potent peroxisome proliferator than clofibrate so it cannot be attributed to relative potencies of the compounds. With clofibrate, the transcription rate of the cytochrome P-450 LAw gene was increased dramatically within 1 249

hr of administration which would tend to support the rapid

increase in the transcription of the bifunctional protein as

demonstrated by Reddy et al.,(1986), (Hardwick et al.,1987).

Chlorophenoxyacetic acids structurally related to clofibric

acid and arylacetic acids induce bilirubin UDPglucuronosyl-

transferase activity in a structure-dependent fashion. The

nature of the chemical group fixed in the alpha-position of

the carboxyl moiety and the number of methyl substituents

appears to be of great significance in the inductive process

(Fournel et al.,1985). A close linkage was established between

bilirubin UDPglucuronosyltransferase induction and that of

cytochrome P-452 (Fournel et al.,1985). It has been suggested

that both processes could be under coordinate regulation,

mediated by a molecular interaction dependent on the

physiochemical/sterospecific properties of the inducers

(Fournel et al.,1986).

The major responses in the rat to peroxisome proliferators is

the coordinate induction of cytochrome P-452, carnitine

acetyltransferase and peroxisomal beta-oxidation enzymes. In

addition, the induction of bilirubin UDPglucuronosyl­ transferase and long chain acyl-CoA hydrolases occurs in treated rats (Fournel et al.,1985, Katoh et al.,1984). It would be of interest to examine the activity of the latter two enzymes across a wide range of species to clarify the response. No induction of peroxisomal beta-oxidation enzymes, carnitine acetyltransferase, cytochrome P-452 and long chain acyl-CoA hydrolases occurred in the guinea pig (chapter 4, 250

Kawashima et al.,1983). In the marmoset, a minimal induction

of peroxisomal beta-oxidation was observed in the absence of

other enzyme changes. This would suggest in the rat and

presumably other responsive species (mouse, hamster and

rabbit) that these enzymes are members of a gene battery

regulated by peroxisome proliferators (Nebert et al.,1987).

With respect to the hybridisation experiments with the

cytochrome P-452 cDNA probe it would be useful to perform a

control experiment, hybridising a duplicate filter to a rat

albumin cDNA probe. This gene was transcribed at a rate

approximately equal to that seen in the nuclei of normal rat

liver following ciprofibrate and clofibrate administration

(Reddy et al.,1986). The translatable mRNA level for albumin was not significantly changed by DEHP treatment (Osumi et al.,1985). Using the rat albumin probe would allow for a more comparative analysis in the guinea pig and marmoset where a slight increase in mRNA-cDNA hybridisation occurred with no associated increase in catalytic activity or immunodetectable protein.

The species responses to ciprofibrate have been documented at a biochemical level in this work. To delineate which changes are a prerequiste for the ultimate expression of toxicity with this class of compounds requires further investigation using the cloned genes to cytochrome P-452, peroxisomal beta-oxidation enzymes and glutathione peroxidase, a logical extension of the work presented in this thesis. 251

Liver sections for electron microscopy were prepared from all

animals treated with ciprofibrate for 14 days and from marmosets in the 6 month treatment and recovery groups.

Unfortunately technical problems were encountered and no

sections could be used except the primate samples. In the marmoset at each time point, 2 week, 6 month and recovery groups, there were no obvious treatment-related ultrastructural changes (figures 5.17 to 5.22). It was not possible to make any definite conclusions, however there was a marked decrease in glycogen content in treated animals making organelle distribution comparisons difficult. There was no proliferation of the SER and peroxisomes and no apparent structural changes to these organelles. Mitochondria appeared elongated with a slight increase in number although this would not be considered significant. In the recovery group the glycogen content was increased with levels similar to the control group. In the ciprofibrate treated rat (administration for 14 days) there was an obvious increase in peroxisome number has been observed in hepatocytes (Lalwani et al.,1983b,

Sterling Winthrop, personal communication). The R.E.R appeared more diffusely distributed and the endoplasmic cisternae were dilated in the treated groups. From the photographs it is apparent that the reported hepatic ultrastructural changes in the treated rat were not reproduced in the marmoset by ciprofibrate treatment. In the marmoset study, full autopsies were completed which included tissue pathology, haematology and blood biochemistry analyses (appendix 1). No significant treatment-related effects were observed. ELECTRON MICROGRAPHS

Glossary of Terms ;

E.R. Endoplasmic Reticului G Golgi Body

L Lipofuscin

M Mitochondrion

N Nucleus

P Peroxisome 253

14 Day Study With Ciprofibrate in the Marmoset. Figure 5.17. ; Micrograph from the control group.

Figure 5.18. ; Micrograph from the 100 mg/kg dose group.

x 5. 4k M a g 254 26 Week Study With Ciprofibrate in the Marmoset. Figure 5.19. Micrograph from the control group.

Figure 5.20. Micrograph from the 80—100 mg/kg dose group.

F-J V - 255

26 Week Treatment Period with Ciprofibrate in the Marmoset Followed by a 4 week Recovery period. Figure 5.21. Micrograph from the control group.

r | p ,

Figure 5.22. Micrograph from the treated recovery group.

x 5 4 k M a g By comparison, clobuzarit treatment resulted in a proliferation of peroxisomes in the mouse, rat and to a lesser extent in the hamster (Orton et al.,1984). No such changes were observed in the marmoset by these latter authors. The in

vitro exposure of rat hepatocytes to ME HP resulted in a marked

proliferation of peroxisomes and a concomitant increase in beta-oxidation (Elcombe and Mitchell, 1986). No significant peroxisome proliferation was induced in cultured marmoset hepatocytes. In the Rhesus monkey, gemfibrozil administration

caused an increase in peroxisome number per hepatocyte with an

increased volume fraction in aged animals (Gray and De La

Iglesia,1984). In young Rhesus monkeys peroxisome numbers were unchanged. Ciprofibrate treatment was reported to cause peroxisome proliferation in Rhesus and Cynomolgus monkeys and DL-040 administration in Rhesus monkeys (Reddy et al.,1984,

Lalwani et al.,1985). The marmoset is considered a

non-responsive species to peroxisome prolif erators with respect to both catalytic and morphological analyses. It is possible that there is a fundamental difference between the marmoset and Rhesus monkey (New World versus Old World) and that the latter may be considered responsive. However, no measurement of lauric acid hydroxylase activity has ever been made in these Old World genera and peroxisomal proliferation remains questionable at the very high dose levels used.

The importance of metabolism and drug bioavailability with respect to species differences in the response to peroxisome proliferators must be considered. The major problem encountered in anticipating the fate of a xenobiotic is not in 257

predicting the possible routes of metabolism but rather in predicting which of these potential reactions will occur in

particular animal species (Caldwell, 1981). Interspecies

differences in metabolism may be attributed to;

1) inability of a species to carry out a metabolic pathway, 2) occurrence of a unique reaction in a particular species, 3) variations in relative extents of two or more competing reactions which a compound may undergo.

On the basis of available data it would appear likely that Old

World monkeys, notably the Rhesus monkey, resembles man most closely in metabolic terms (Caldwell, 1981), and therefore that the marmoset resembles the Rhesus monkey more closely than rodents. Clofibric acid, the hydrolysis product of its ethyl ester clofibrate, is the active metabolite and in the rat,

rabbit, guinea pig and man is excreted as the free acid and ester glucuronide (Emudianughe et al.,1983). In this thesis, the term ciprofibrate refers to 2-[-4-(2,2- dichloro- cyclopropyl) phenoxyl]-2-methyl-propanoic acid, the pharmacologically active compound. Therefore, in the short term and chronic studies using ciprofibrate, any compounding influence of possible species differences in initial hydrolytic biotransformation are eliminated by using the active compound.

On a similar theme, DEHP is hydrolysed in the intestine to ME HP and it is absorbed in this form (Lake et al.,1977). By using MEHP in in vivo and in vitro studies negates species differences in absorption and bio avail ability (Elcombe and 258 Mitchell, 1986). Peroxisome proliferation and enzyme induction

was observed in vivo and in vitro in the rat with ME HP yet no

induction was demonstrated in guinea pig and marmoset cultured

hepatocytes (Lake et al.,1986, Elcombe and Mitchell, 1986). To

avoid possible species divergence in metabolism, the active metabolite of ME HP was isolated. This proximate prolif erator

(metabolite VI) was added to rat, guinea pig, marmoset and human hepatocyte cultures (Elcombe and Mitchell, 1986).

Induction of peroxisomal beta-oxidation and lauric acid

hydroxylation occurred only in rat hepatocytes. Cinnamyl

anthranilate, a synthetic food flavouring, is a peroxisome prolif erator in the mouse but not the rat. This has been attributed to a species difference in metabolism with the active metabolite being formed by hydrolysis in the mouse only (Caldwell et al.,1985, Caldwell et al.,1987).

Pharmacokinetic parameters have been established for clobuzarit in the rat, mouse, hamster, dog and marmoset after oral administration (Orton et al.,1984). These parameters were used to select dose levels which resulted in peak or steady

state serum levels of 100-200 ug/ml in a 14 day oral

administration study. Peroxisomal proliferation and enzyme induction were observed in the rat, mouse and hamster (descending order of response magnitude) only. The interspecies differences in hepatic response were probably not due to differences in exposure since the systemic levels of the compound in responsive (rat) and non-responsive (dog) species were similar (Orton et al.,1984). 259 The half life of ciprofibrate in the rat is 3-4 days whereas

with bezafibrate it is 6-16 hr and with clofibrate 5-8 hr

(Baldwin et al.,1980, Cayen et al.,1977, Edelson et al.,1979,

Schmidt et al.,1980, Sterling-Winthrop, personal communication and Thorp,1972). This increased exposure of ciprofibrate in the rat, resulting from the sustained high drug concentration in the plasma would account for its high potency when compared

to clofibrate and bezafibrate, assuming that plasma levels are

directly proportional to active tissue levels. The maximal

plasma ciprofibrate concentration was attained at 2 hr after a single oral dose to rats, monkeys and man (Davison et al., 1975). The drug was slowly eliminated in all three species and in the Rhesus monkey and man elimination appeared to follow

the kinetics of a two-compartment model. The long half lives

of ciprofibrate may be attributed to;

1) enterohepatic circulation, 2) accumulation of drug in fat, and, 3) extensive plasma binding of ciprofibrate which may inhibit clearance from the plasma compartment (Davison et al.,1975).

Ciprofibrate was excreted as the free acid and glucuronide in the rat, monkey and human. The ratio of ciprofibrate excreted as unchanged drug to glucuronide is unknown. In the marmoset and rat steady state levels of ciprofibrate were achieved within 7 days (Sterling-Winthrop,personal communication). Assuming a linear relationship of ciprofibrate dose with plasma concentrations, plasma levels in the marmoset at the high dose level (100 mg/kg) were 1.5 to 2-fold higher than that in the rat at the 20 mg/kg dose level (chapter 4). It has been estimated that in the marmoset, elimination in the high 260

dose group is biphasic [tl/2 a 4,5 hr, tl/2 b 8.2 hr]

(Sterling Winthrop,personal communication). The half life is clearly shorter in the marmoset than in the rat. Assuming (by extrapolation) plasma concentrations (measured 4 hr post dose), at the high dose level in the marmoset were 500 ug/ml and 45-50 ug/ml at the low dose level (2 mg/kg) in the rat the

plasma concentrations at 24 hr post dose were not markedly

different in the two species i.e. 50-80 ug/ml (calculated using a minimal and extensive duration of the a tl/2) and 37-44 ug/ml (1/3-1/4 of a half life) in the marmoset and rat respectively. At the high dose level in the rat (20 mg/kg), the plasma concentration of ciprofibrate was approximately 250 ug/ml and this would correlate to between 215 and 225 ug/ml 24 hr post dose (1/3-1/4 of a half life). Although the values are derived from a theoretical basis and as such should be treated with caution, some indication of ciprofibrate exposure in the rat and marmoset can be made. Although there are clear differences in the bioavailability of ciprofibrate in these two species it should be noted that in the marmoset that the initial plasma concentration was considerably higher than in the rat. Enzyme induction was apparent at both dose levels in the marmoset following 2 and 26 weeks ciprofibrate treatment although it should be remembered that with peroxisomal beta-oxidation, induction was considerably less extensive in the marmoset in comparison to the rat. The difference in the pattern of enzyme changes in the marmoset - induction of peroxisomal beta-oxidation accompanied by increases in the activities of mitochondrial enzymes and in the absence of carnitine acetyltransf erase and lauric acid hydroxylase 261

induction, is unlikely to be due to differential drug exposure but more probably to a fundamental species difference in gene

expression.

An additional factor to consider in induction potency is the

extent of binding to plasma proteins and this has been examined in the rat, human and Rhesus monkey with ciprofibrate

(Davison et al.,1975). A greater proportion of ciprofibrate was unbound at the higher concentration of available drug in

the three species. There was less ciprofibrate bound to rat plasma at the higher concentration than to plasma from the

other two species studied (Davison et al.,1975). A similar trend was observed with clofibric acid in the rat, dog and man

(Cayen et al.,1977). Unfortunately there is no data with respect to ciprofibrate plasma binding in the marmoset although a pattern similar to the Rhesus monkey might be expected (Davison et al.,1975). In figure 5.23, the percentage of unbound ciprofibrate versus plasma drug concentration

(ug/ml) for the rat and Rhesus monkey is shown. Assessment of the percentage of unbound ciprofibrate at both dose levels in the rat and the high dose level in the marmoset can be made from the graph if it is assumed that:- 1) the relationship between increasing dose and decreasing plasma binding is linear, and

2) the pattern observed in the Rhesus monkey is similar to that in the marmoset.

In the rat the percentage of free ciprofibrate was 2.1 and 8.6% for 2 and 20 mg/kg dose levels respectively and at 100 262 mg/kg in the marmoset it was 1.7% (figure 5.23). There was no difference in the percentage of free drug in the low dose

level (2 mg/kg) in the rat and the high dose level (100 mg/kg)

in the marmoset. Therefore, the difference in hepatic enzyme induction between both dose levels in the rat is clearly not a product of the percentage of free drug as both 2 and 20 mg/kg dose levels are on the plateau region of the dose response curve (figure 5.14). Similarly, the difference in hepatic response between the rat (low dose level) and marmoset (high dose level) is due to differences in the inherent response of these two species and cannot be attributed to tissue drug availability. Therefore the extent of ciprofibrate plasma binding is not of significance in the induction phenomenon and is not the reason for species differences in response to peroxisome prolif erators.

In recent years, growing concern has arisen with respect to the long term exposure of man to hypolipidaemic drugs and industrial plasticisers and the possible effects on human health. This is based largely on the ever increasing evidence that peroxisome prolif erators are, as a class, hepatic carcinogens in rodents (Reddy et al.,1980). Extrapolation from rodent carcinogenesis studies to other species is rarely satisfactory as both qualitative and quantitative differences exist between the respective responses. It should be noted that predictions are based on a direct interspecies comparison and that it is assumed that the mechanism of action in the animal model is both identical and reproducible in man. 263

Figure 5.23.

Ciprofibrate Plasma Protein Binding In The Rat and Rhesus Monkey.

20_

RAT

16_

U) 3 1—

Q

o 0 LL

■ ■ *♦— O 8 _

MARMOSET

100 200 300 400 500

ug ClPROFIBRATE/ml

The percentage of "free ciprofibrate” at different dose levels for both species was determined graphically; Hat : 2 mg/kg - 2.1%, 20 mg/kg - 8.6%. Marmoset : 100 mg/kg - 1.7%.

Percentage Of Free Ciprofibrate (14-C) Tal

5ug ciprofibrate/ml 500ug ciprofibrate/ml Rat 0.90 18.90 Rhesus Monkey 1.00 5.90 [a] Derived from Davison et al.,1975 264 Ciprofibrate is negative in a wide range of mutagenicity

tests;

1) no induction of unscheduled DNA synthesis in rat

hepatocytes (Glauert et al. 1984), 2) negative in strains TA 1535, 1537, 1538, 98 and 100 of Salmonella typhimurium in the Ames test both with and without metabolic activation,

3) inactive in mouse Lymphoma Forward Mutation Assay without

metabolic activation, active with metabolic activation, and 4) inactive in Balb/3T3 in vitro transformation assay and Balb/3T3 rat hepatocyte mediated transformation assay (Sterling Winthrop,personal communication).

However, in vivo long term ciprofibrate treatment in the

Fischer rat results in the development of hepatocellular adenomas and carcinomas in the high dose group [10 mg/kg/day] (Sterling Winthrop,personal communication). Histologically hepatic lesions were classified as altered areas, nodules or

hepatocellular carcinomas and on the basis of this criteria a 100% incidence of hepatocellular carcinomas was observed in

the treated animals (10 mg/kg/day) (Rao et al.,1984). With

short term and chronic ciprofibrate administration in the rat, hepatomegaly and induction of peroxisomal beta-oxidation, cytochrome P-452 and carnitine acetyltransf erase has been demonstrated in this work. These hepatic changes induced by ciprofibrate are typical of those obtained with other peroxisome prolif erators such as bezafibrate and clofibric acid. 265

The magnitude of the hepatic responses appears to correlate

with the relative lipid lowering potency of the compound and

is markedly influenced by pharmacokinetic characteristics, particularly biological half life. It is suggested that these liver changes serve as early indicators of the ultimate toxicity, hepatocarcinogenesis, in rodents. It is also

possible that hepatomegaly and enzyme induction are a

prerequiste for the final expression of toxicity in this class

of compounds as will be mechanistically enunciated in chapter 7.

There is an obvious need to clarify the rodent response as species specific or species common in order to make an assessment of the potential risk to man. With reference to the short term and chronic ciprofibrate studies, a clear species difference in susceptibility exists. The marmoset and guinea pig may be considered as refractory species whereas the rat, mouse, hamster and rabbit are susceptible. These latter 4 species exhibited marked differences in the magnitude of response indicating that both a species specific and dose dependent induction may occur. The relative advantages and disadvantages of using the marmoset and rat as models for extrapolation purposes to assess potential toxicity in man is discussed in chapter 7. 266 CHAPTER 6 SPECIES AND RAT STRAIN DIFFERENCES IN RENAL RESPONSES TO

CIPROFIBRATE ADMINISTRATION.

6.1. INTRODUCTION

In the three previous chapters the results from several

different investigations were discussed. In essence, these

studies examined ; a) the hepatic response in the rat to the short term administration of ciprofibrate, and, b) compared the treatment-induced changes in five rat strains

and six species.

The chronic effects of ciprofibrate on hepatic enzyme parameters in the marmoset and rat were studied. In the rat, the relative potencies of three peroxisome proliferators were investigated following chronic administration.

The hepatic changes in the rat included hepatomegaly,

differential effects on the microsomal drug metabolising enzyme system, induction of peroxisomal beta-oxidation, mitochondrial enzyme changes and a decrease in glutathione peroxidase activity. A considerable variation in species response was apparent and the order of induction susceptibility was shown to be the rat > mouse > hamster > and rabbit. The marmoset and guinea pig were relatively non-responsive.

f 267 The objectives of the research work presented and discussed in this chapter were two-fold;

1)to ascertain if the rat response is tissue specific or tissue common, and

2) to study rat strain and species variation in the kidney.

The effects of a wide range of peroxisome prolif erators on renal peroxisomal and mitochondrial enzymes has been examined at length in our laboratory (Sharma et al.,1988a). In the current study, the influence of the short term administration of ciprofibrate on the drug metabolising enzyme system was examined in five rat strains and five species. In addition, renal glutathione peroxidase activity was measured in the rat. The chronic effects of ciprofibrate, clofibric acid and bezafibrate on the microsomal drug metabolising enzyme system were also studied in the rat. No statistical analysis could be completed on the data due to an insufficient number of kidney samples. In the rat and mouse there was one microsomal pool per dose level whereas in the hamster, guinea pig and rabbit there were 2 microsomal pools per dose level. No results are reported for the marmoset as the specific content of cytochrome P-450 was too low to be detected spectrophotometrically and no further assays were completed. 268

6.2. THE MICROSOMAL DRUG METABOLISING ENZYME SYSTEM.

6.2.1. Cytochrome P-450 Content.

In the Gunn and Sprague Dawley strains there was a dose-dependent increase in the specific content of cytochrome

P-450 (table 6.1). At the high dose level in the Fischer rat the cytochrome P-450 content was induced. In both the Long Evans and Wistar rat, the specific content of cytochrome P-450 was increased and the extent of induction was similar at both dose levels. In the treated rat, the increase in cytochrome P-450 content varied between 1.35- and 1.7-fold. In the mouse and hamster the cytochrome P-450 content was unchanged by ciprofibrate administration (table 6.2). The specific content of cytochrome P-450 was decreased at the low dose level in the guinea pig. The induction of cytochrome P-450 was dose dependent in the rabbit. In the marmoset the specific content of cytochrome P-450 was too low to be spectrophotometrically detected. Chronic treatment of Fischer rats with bezafibrate resulted in a 1.5-fold increase in the specific content of cytochrome P-450 (table 6.3). For ciprofibrate and clofibric acid the cytochrome P-450 specific content was 50 and 70% of the control value, respectively. There was little strain or species variation in the basal level of cytochrome P-450 content i.e. in untreated animals.

Treatment of Wistar rats for 3 days with W y - 14,64 3 and clobuzarit caused an increase in cytochrome P-450 content.

Induction with W y - 14,643 was minimal whereas a 1.5-fold Table 6.1 The Effect of Ciprofibrate on Renal Cytochrome P-450 Content In Different Rat Strains X < E CO E X < M z o > x w x X Q Q J E O E-* “*3*in CP O J E o c o g a 0 x x o X r-lo cu£ X x co P X o X 0 O «h u C w M O X X o Z E Z E h h h h 1 g h h h 0 u u

rH P O O O O O O O H H H in ^ o ' m •'T o HH H H O O O o C H O 0 o D z z 4 0 \ s X X g g CP O' ' O P C

4 0

4 0 4 0 0 l r—1 H r r O O O O *H rH rH rH *H O o» m o CO X c D X X X X O \ \ 4 0 g P g C P C P C P C co

p 4 0 4 0 0 O O O t p CO rl H 04 CO O —s — CD O o c u o H O r-l O

in o X X X CO X X X \ g ' g O ' O ' O P C

0 O O O O H H O O ' O HH H H ^ O CO CO O

04 CO E % \ s \ X X g P g C P C P C P C h

04

X X H r p gp g CM O H H O o vo o o o o H H ID VOH O * 0 - 0 0 c o o U \ \ o • • • • CP ' O Z Z > J O X O ' O P C 04 Z < CO

-'P; TJ •O' v) 0 > to CO- 0 c o O) 0 0 a H - P r-l U4rH r Q» Q •rH P > p 3 0 0 1 0 M 3 > 0 0 u (0 0 0 I X 0 M X M U X 0 0 O' x 0 0 Qi ro r 3 x to p g 0 U C to u o (0 04 10 0 < ) a U 0 to 0 £ C O' a) O •H • P •a a f *0 •H P rH 3 U 0 0 c CP p 0 10 10 W 0 > 0 a a M a) to 0 10 o u M 0 to o C c O' m 0 0 Qt u o w e M O H

Table 6.2 The Effect of Ciprofibrate on Renal Cytochrome P-450 Content In Different Species ta M ta CO CO 4 0 U j Q ta o > CO w ta o in 0 3 o X w H c o U o z o a z Cu CtSr-l 0 ta CO o cu a) ta M o Z ta U -H U C w >•t*4 — O O E-* z £ £ 1 p B p tr> o* P

p M H r U CM CM CM U O H H H tX> r—n —'• — o O'-'io cn o h o o o H0 H i—rH 1 00 H r O C H r rH 00 rH o o G O' O' P w o 0 • • • z o cn o 3 10 0 B B O' O' O'

o IX

n o n n o

rH 4* P 4* rH 4-> O H rH rH rH H \ \ \ CM O U CM CM CM U O rH CM + O o o c P O 1 I 1 1 . •

44 s o (X o o o o 0 £ p + + 0 0 £ O' O' O' O' O' • • o O MrH CM - ft B ft

p £ -p O U CM CM CM U rH \ \ \ O rH O H* P O' O' O' P o G w o 1 1 1 t ft ft o • m OS o o o O rH p —' 0 -H- H- O' O' O' . • ta n cn 1 M *—* 03 O O in O rH O rH B I • ft

£ £ P .X 44 rH O CM CM CM O O rH \ \ \ O O HT O H CM rH rH cno cn D ^CD<** ^ + G O' O' P o o w o 1 1 1 1 tft ft tft ft •H •H O O O O o o rH rH rH n 00 in H- 3 G 0 0 O' 04 O' O' O' H- ft ft

P £ £ -P rH X 4«C X rH o IX O CM CM CM O o o o O S N \ H o i O c- o r- H rH CM rH H H* O O C O' O' P w o 1 I 1 1 ft 43 n Qi •H P o O rH O rH rH rH 0 *4- O' O' O' • ft o rH m o + ft •

•u •a •O' fn 0 43 U 0> U 43 0 fn o c CO CO 0 O c 0 CO O) - u l-P J U l U rH I> O T ip rH p rH P p •rH •rH a p MH rH a 0 0 0 p 0 0 X 4 0 P 0 0 0 0 0 0 c 0 > 0 p 0 O' > u c 0 o o ) ro a) X 4 0 0 p 03 £ P 0) 0 O' P a)0 O 3 u 0 1

•H •H P X •H 3 3 0 43 a > ta 03 P P *0 3 0 •H P rH 0 P O' O' P rH 3 3 O Qi C 0 0 o « 0 0 U £ 0 P 0 0 U 0 P 0 0 P 0 P 0 P 0 0 0 • 0 Q X 04• 0 G G 0 o p p a) 0 1 Or O P 0 0 0 0 G c O' u o C P 0 0 0 «p

w ro 0 0 P 03 P 0 0 0 0 3 43 3 rH 0

o P

a IX) 0 04 0 £ 0 0

271

Table 6.3.

The Effect of Hypolipidaemic Drugs on Renal Cytochrome P-450

Content and Ethoxyresorufin-O-Deethylase Activity in the # Fischer Rat Following Chronic Administration.

Treatment Cytochrome P-450 Ethoxyresorufin-O- Specific Content Deethylase Activity

(nmol/mg protein) (pmol product/

min/nmol P-450)

Control 0.23[a] (100)[b] 4.01 (100

Clofibric Acid 0.16 (70) 3.09 (77) (50 mg/kg/day)

Bezafibrate 0.34 (148) 2.90 (72) (75 mg/kg/day)

Ciprofibrate 0.22 (48) 5.11 (127) (20 mg/kg/day)

K Single daily dose for 26 weeks

[a] Values are expressed as an average of two experimental results .f

[b] Figures in brackets are values expressed as a percentage of corresponding control. 272 increase was apparent following clobuzarit treatment (Sharma et al.,1988a). With clofibrate administration a slight

decrease in the specific content of cytochrome P-450 occurred

in the rat. In contrast, a slight increase in cytochrome P-450 content has been described with a single dose of clofibrate, 72 hours after administration (Makowski, 1985). A 2.5-fold induction of cytochrome P-450 has been reported with

clobuzarit in the rat following a 14-day treatment period

(Parker and Orton, 1980). In the present study an increase in cytochrome P-450 content was observed with ciprofibrate in the rat with minimal strain differences. A similar increase was seen in the hamster with a marked dose-dependent increase in the rabbit. However, following chronic administration of ciprofibrate a marked decrease in cytochrome P-450 specific

content occurred. This would appear to be a characteristic of

ciprofibrate alone as a decrease with clofibric acid occurred at both time points and a 1.5-fold induction with bezafibrate was maintained following chronic administration.

6.2.2. Benzphetamine - N - Demethylase Activity.

Benzphetamine-N-demethylase activity was measured in the acutely treated rat only. In the other species - mouse, hamster, guinea pig, rabbit and marmoset there were insufficient microsomal samples to determine the activity of this enzyme. In the chronically treated rat no enzyme activity was detected when using similar conditions to assaying samples from animals treated with ciprofibrate for 14 days. 273 A slight increase in benzphetamine-N-demethylase activity was

observed in the Sprague Dawley rat at the high dose level

(table 6.4). A dose-dependent decrease in activity occurred in the Wistar rat. With both the Fischer and Long Evans rat, activity was decreased at the high dose level. These treatment-induced changes to enzyme activity did not appear to

correlate with changes observed in the specific content of

total cytochrome P-450, with the exception of the Sprague Dawley rat. In this latter rat strain, a dose-dependent increase in cytochrome P-450 content and benzphetamine-N-demethylase activity occurred. However at the low dose level the extent of induction between these two parameters was not equivalent. The variation in enzyme

activity in untreated animals between rat strains was minimal.

Following a single dose of clofibrate, a slight increase in

benzphetamine turnover was observed at 6 hr post dose yet by 12 hr, activity had returned to control levels (Makowski, 1985).

6.2.3. Ethoxyresoruf in - 0 - Deethylase Activity.

Ethoxyresorufin-O-deethylase activity was decreased in a dose-dependent manner in the Wistar and Long Evans rats (table 6.4). Activity was decreased at the high dose level in the

Gunn and Fischer rats. Enzyme activity was decreased to approximately 50% of the control value in the Sprague Dawley rat at both dose levels. It should be noted that the treatment-induced changes occurring in ethoxyresorufin-O- deethylase and benzphetamine-N-demethylase activities were rs -"■■s o —* O ' O O O / ■ > o 1 o O T o vo vo O CM CM o h o o o rH N4 in rH rH CN rH cn p- rH i •iH NZ £ —* C M ta \ o ta &4 co -P in D < u n4 CP os J 3 1 c o > 1 3 Hi •H CO z o in Cxi V4 rH •H OS a a o rH >4 ta g 0 X O H C XI O o o X? 00 in io o in cn O C^- -H O'-'— * o in — •H 4-> 1 1 \ o CO CM © O N 4 O CM H o m n o n > s C 1 Z 0 rH P- rH rH rH i—1 rH rH LO rH LO CN rH rH P" 0 1 1 •H rH M 1 ta g ~ id 0 1 ta c o \ o c 44 1 z < 4-> in 0 144 1 t—Ij u n 4 os •H I s X 3 1 Q 1 < k 3 a 0 1 EH E-4 O x: C 1 ta ta M rH 4-1 M 1 s s 04 o 1 04 ta g r—1 c g 1 CM O H C 3 o 0 1 z o >—> 4-> « ta g o 00 o n * cn in c m cn c m cn o cm 0 0 1 PQ c rH rr p~ (N rH 00 rH ro oi in ^c4 co co 4-1 >1 1 — ■ • ■ • • • • • • • • • • * « • m CO 1 CO m in o IX P- CO cn cm p* cn in u JQ •r| <44 0 M Cu •H u •44 CP CP CP CP CP CP CP CP Cp CP (/) O J rH r-H 4* J* rH XS M rH A- M I-H .* > ta ta O \ \ o w o w o w o w 4-» CO > U CP CP W CP CP U CP CP >4 CP CP M CP CP -o U o ta 4J g g 4J g g -p g g 4-> g g 4J g g 0 Q XI C c 3 c c 44 O o o o o o o o O O T - 4H o CM CN U CN CN U CN CN u CN CM U CM CM H 0 0 42 ( 0 &4 O ■ u ' z 0 « I TJ UD I OS ta >4 OS I E-* d ta ta Z W Q) 0 1 CO CD J z z < & u z os c CO co z > .E D 04 o hH m o ta co: [b 3: [b Figures 3: in brackets are values expressed as percentage of corresponding control o CO (X4 £ J [a]: Values are expressed as an average of two experimental results 275

very similar in the Fischer rat. The decrease in ethoxyresorufin-O-deethylase activity in the mouse was

dose-dependent (table 6.5). Enzyme activity in the hamster and guinea pig was unchanged and in the treated rabbit activity was 20% of the control value. Considerable strain variation in the basal levels of ethoxyresorufin-O-deethylase activity

(i.e. in untreated animals) was evident. The Wistar and Gunn

rats exhibited high activity whereas basal levels in the Long Evans, Sprague Dawley and Fischer strains, mouse, hamster and rabbit were 3 to 12-fold lower. Activity in the untreated guinea pig was barely detectable.

The chronic administration of ciprofibrate, bezafibrate and clofibric acid had no effect on the activity of ethoxyresorufin-O-deethylase (table 6.3). Activity in the control group was lower than that in the Fischer rat at 2 weeks. This may represent an aging process.

A decrease in ethoxyresorufin-O-deethylase activity was observed with clofibrate and bezafibrate in the Wistar rat following a 3 day treatment period (Sharma et al.,1988a). Several other peroxisome prolif erators caused an equivalent decrease in activity although ciprofibrate was not included in the same study (Sharma et al.,1988a). Following a single dose of clofibrate in the rat, a decrease in enzyme activity was demonstrated at 6, 12, 24 and 72 hr after administration (Makowski,1985). The decrease in ethoxyresorufin-O-deethylase activity was observed in the current study at the high dose level in all rat strains as well as in the mouse and rabbit. to 1 —* r—* 0) 1 O •H I X m . 0 UJ U 1 E h N* <1) 1 M I O 4 1 > 04 '—• •—■ ■—> CO 1 1—1 O'-'-' O CN CN O O O O O'-'—' 1 Eh rH O i n rH O CN O O CN CN O H LO O LO LO • •P 1 O O HHH rH rH rH rH rH CN r-l CN CN H r-l H 10 WWW www www WW Q C 1 < £ — <— — CD 1 c P P 1 M -P CD 1 CO P 4-1 1 < (D c m l i-3 Ot o • r t | X u a i X CD cn 1 Eh -P c c 1 W 3 •H M 1 W C 1 Q «H * 0 c >1 1 1 e -P 1 0 o •H 1 1 p 0 4 (0 > 1 2 CO CN C— CO 00 rH 00 CN P u 1 D -P

0 1 •H O 4 C 1 •rH | p 4-1 1 3 \ LO p O ' Cn P O ' O ' P O ' O ' p O ' O ' ro 0 ta -P £ £ . p s s -P £ £ - p £ £ -P £ £ ro CD rH c 1 Q J G c G c G £ > to 0 1 O 0 0 0 O O 0 0 O O rO C J3 1 U C N CN U CN CN O CN CN U CN CN 0 CN CN (D 10 ro CD 1 to P CD -P -P 1 ro ro £ OT £ f6 l i to to 0 p 1 ! TO ; CD - p ro p J0 1 "D •iH | to CD 4H 4H 1 * ! u 1 cn 1X4 > Q •H u/ 0 1 1 t-H 0) -P CO CD jQ • • •• • • • • rH 1 CD 1 0 10 tO 1 c ■ 5 •H JQ c r— 1 r-^ i—* *—• JO 1 XS 1 ta 3 £ - p ta JQ U T3 rO l Eh 1 Oi O ro UJ 3 ro «0 CO UJUJ E h I CO S X os 0 X 1—1 277 It should be noted that in liver microsomes, ethoxyresorufin-O-deethylase activity was expressed as nmol product/min/nmol P-450 as opposed to pmol product/min/nmol P-450 in the kidney. A point of concern is that these low

levels of activity in the kidney may exceed the levels of detection and sensitivity that can be accurately determined with this assay.

6.2.4. Induction of Cytochrome P-452.

Laurie acid hydroxylation at the 11-position was increased at both dose levels in the Gunn and Wistar rat and induction was 1.9 to 2.5-fold (table 6.6). A slight increase in 11-hydroxy- laurate was observed at the high dose level in the Sprague

Dawley rat. Activity was unchanged in the Long Evans and

Fischer rats. The 12-hydroxylation of lauric acid was unaffected by treatment in both Wistar and Long Evans strains.

Dose-dependent increases in 12-hydroxylaurate were demonstrated in the Gunn rat and activity was increased at the low dose level in the Sprague Dawley rat.

The 11-hydroxylation of lauric acid was decreased at the high dose level in the mouse and increased at both dose levels in the hamster (table 6.7). The increase in 11-hydroxylaurate was observed at the low dose level in the guinea pig whereas activity was decreased at the high dose level in the rabbit. The 12-hydroxylation of lauric acid was increased at both dose levels in the mouse. The increase in 12-hydroxylaurate was dose-dependent in the hamster and a decrease in activity was Table 6.6 The Effect of Ciprofibrate on Renal Cytochrome P-452 Content and Lauric •rH 3 H 1 44 V i-4 44 4-> 1 44 4 1 44 | •rl < tc S 103 O 1 CO o o ro C H x >4 w a) >1 0 1 1 CD C l ro 1 M 1 H 1 G to o I ro h 1 1 1 O C < » N* LO CM J i O OS W O > O W C W E-4 o tc OS g fa fa 1-4 o tH o O 32 fa CO fa o O O O fa 53 E-4 fa 1-4 S3 E ij as 1-4 < P o Q O a tc os CD X CO fa >H 1-4 SH 04 o < 1 1 h r— 1 i-4 1< 44 rH rH .— . s a ' d? 44 44 T3 44 LO o 1—4 CL4 — g >4 O 04 N* 0 G o 3 o O tc G g 04 U O to to 3 O 33 O 04 CD M g >n g O U C G 1 • » o LO N* 44 LO o rH T—1 04 —1 r— •a i-H 04 CM i—1 S3 U >1 M >1 O X 3 a M o >n 3 0 X to 1 1 1 1 • rH CP CP rH o £ P t 4 4 CM c- O Oin CO o l - r cn J U M C M C O rH rH CM rH rH O rH i—ICM O o O JQ i— I N* V o o 4* O \ C g C 3 4 • . * . . CP \ O O N4 O in CM LO t—1«—l c*~ CM ro LO rH 1 M CM rH on CM 1 N CO 3 G G • • • • • •

o 4 *LT o r— LO CM CM ro r~ LO rH co ur> ao M* r—1 P C * .

o g P C P ■ CPJ*: rH O CM r~ M C M C O CM Mto CM r- tn CM o ro CM LO nCM on rH rH rH CO O O LO O o o o P C g c \ M 4* o \ • • • • • » • • o o o ro CO CM oLO ro ro r- r— rH co OO CM LO T—| LQCL Q ro M ro£ P C CL) 3 ) a • •

1-4 O t— 1 N LO N* ro CM C— rH on CM ro CM l CO LO O O N* >1 P C — . . • 1

r-l CP.!* CP.!* r-l O O O O O O - S’ g ’ S 4-) o O o C o M o C O C M C M C J C r—I — r 1 M O C -sr rH M C M C M C I 1— t— O L LO on rH rH LO O O Ocn LO cn o g o G CP \ M O • • •rl G . 4 rH fa rH rH CO rH 1 N rH cn ro o cm 10 u a; M * t— O L CP O O O C M O C H r S O O C O H r H • r O L • ' t M C H H M C M C H r on if) 1cn N co co to M C M C O 1 M O C N C If) O O O L O M L C O O L C N C O C cn O co L O L ocn o o CM LO O L M C . . . . • • • * 4-) s \ \ * 4 cm rH LO (0 ro U g g P C CP J* CP J* co o lo CP CP

on i—l lo O L O L O L 4* * 4 P C H r H r o rH cn cm O - C g g CP 4-> O O O C M O C H r M C M C H r n o H r O c— L o O c L o o - r — r co rH O ljo M C M C O oco o o O O P C . V " M g G * 4 O on on m ~ r J ■ O Ct) Ct) O > G tn ro 1 lo o l n m o l m G P C 10 * 1 1

U ' o ■ 3 T CO 05 CO v > « CD O CO 0 ^ ) O ro X) 44 r—I O 4 to44 u •rH ro44 I—I > fa m O d n £ a) to X 04 04 > M ro a) tn 0 3 3 3 ro w to ro > 04 ro >4 U to 0) 10 44 • •• • o 44 G (0 CD to a) 04 ro *n CD M (0 CD rO ro G M CD G CD M JQ X CD 4* a) co h a) ) a . a) o w 4 4 T3 ■rH i-H •-* O G G VH to to >4 0 ( 04 G CP u o o M 0 a) 3 x rO to 0 CP )

Table 6.7 The Effect of Ciprofibrate on Renal Cytochrome P-452 Content and Lauric •rH P H r P •rH 4-> < fa Q W o >1 P X X C 0) G fa OS u o ro 10 OS >*1 OS P as (0 4 N in N C fa u l l— fa CO CO fa a f a f Q O > i o c fa fa o o fa fa o O X E z fa fa CO fa <*>w a l-HO O H o 5 S E w Z &H H U a f fa P O N4 P O fa rtj fa OS I-H o X OS a o X 0) o I-H Q fa X fa £ < n c fa I h h — rH rH pi rH O 4 N - n i i o •rH fa — G £ o fa -cr as P G £ O (Bin P p I O G £ o fa O G o fa as fa p £ G P C o 1 • 1

n i O * P rH rH 0 rH fa rH rH CN o fa fa . — a X X X X p o ro in X X ro in as o OS p X 1 1 1

4J o o O C O o C 4 N P P P C H r 1 N N C cn in o O LO t— NCN CN O N C N C O rH NN4 N CN i— i HrH rH O n c n c O JQ o r 4 N rH rH O O o o ' S * V P c c X X . O r0 . . z n c N C n i CN CN rH n i WO rH HrH o r rH o r G 10 as e cr>6 o . . . rH n i 4 N t— X CN o r n i C— O 4 N CP •

N N N £ P C P - o r o r o r * . P C H r o o o o o o o o O o O O CN r—1 o —I i— H r H r 4 N N C N C o r o r N C N C O o w o r o c m m o r o c— o c N l— l o o o r—» o o i o o o o o C— 4 N o o n c i— O rH o o r m o \ - H* H- + + H- O O £ P c C X X * P . O — fO u I I I I I 1 1 1 • • • • . . • • • • . . . . 1 l P fa 4 N N n c CN ON4 N CO i C rH o o r n i O rH o rH o r 4 N \ rH o o - H rO £ 10 as M H- 1 1 • » . * n i N C 4 N \ o r n i n c o r o w CN i— 1 n i 4 N WO rH X o a CN \ o w rH n c CN - H H- P C — I 1 . . • l

■ P P C H r o o o o o o o r n c N C N C O N C N C N C N CO H r C— n i wo r- r— rH rH 4 N WO \ 4 N O C— rH o 4 N rH o n i O WO o o \ n c rH rH n i O p H- o £ o C p t X P H- N C J O 1 1 1 1 1 • • . e > o -P fa -P n c OS X CO \ rH r-H rH rH 1 N n c o r 1 CN o w n c rH CN 4 N o w n c O \ H- O—* fO— H- — . . « 1 • I o r rH rH ro 4 N rH o w \ o o X CN n c rH o r m n i rH n c X OO N - 4 H- P C CP C l l . . • . P££ CP P CPJX rH \ \ \ o w o w o r WO rH WO 4O N CN o r o H r N C N C U o r o * r r o r o c N C c-~ O o O o c n c • WO rH in ^ ^ \ \ \ O C H r n O i o N C rH rH o r O W H r n I C i— o O * H - H - H - - + H- H- o o £ G tP '— X P P O I I I I I I I I • • • • . . . • CN • • • • • .rH.rH P P O p T3 p TO p P Z O n c o n c z a G fa G as to O OS OS u as u co o a) ai

o w h o r CP

O O O WO N* CN u £ > g ■u * . P C rH w \ o o O o O O n i o r n c 4 n N o o r c ^ • • rH o wo t- r« \ o r w H H i—I rHrH f ^ ^ o r n c n c o w n i N CN CN O n c H r O C o o r H r H O CN CN o r p** «-H o r - r o in wo in wo cn wo O O WO rH WO O o o r r— o o r r o c n i ■N* o CO c rH o o n n i c o o r c o r •+• * H - H ■+■ - H * H o £ o C X CP X X P P O I I I I I t I l I I I I I I I I • • • • O CN O • ... n i o p X3 P OS o r o c to n c rH CP

■u •o'cn •o ■-» GO > n c s

^

_ •rH £> p I i— O -p P •rH P .. M u a tn ro a) tn ro 3 OS tn OS x fa p tn OS fO tn u as P P a) w fa c o c P O fO fa (U M c ro P C OS u o o c P C u

Cbl: Values are expressed as an average of two experimental results, [cl: Values expressed as means +/- S.D. [d]: Data reproduced from table 6.5. 280 observed in the rabbit at the high dose level.

With the chronic administration of clofibric acid and

ciprofibrate, a 2-fold induction in both 11- and 12-hydroxylaurate was observed (table 6.8). A slight increase in both hydroxyproducts was demonstrated in the bezafibrate treated group. The basal levels of 11- and 12-hydroxylation of

lauric acid were lowest in the untreated guinea pig. Activity

for both products was similar in the remaining species and in

all rat strains.

It should be noted that in each species and rat strain the extent of 12-hydroxylation was greater than the 11-

hydroxylation of lauric acid. In the rat the ratio

(12-hydroxylaurate to 11-hydroxylaurate) varied from 3.0 in the Sprague Dawley rat to 8.0 in the Wistar rat. In the mouse, guinea pig and rabbit the ratio of metabolites was in the order of 2 to 4. Hamster renal microsomes displayed a 9-fold higher omega-hydroxylation than (omega-1)- hydroxylation.

It has been reported that Wy-14,643, DEHP, bezafibrate, nafenopin, clofibrate and clobuzarit administration resulted in a 1.5 to 3-fold increase in renal 11- and 12-hydroxylation

of lauric acid in the Wistar rat (Sharma et al.,1988a). The

extent of clofibrate induction following this 3 day treatment

period was similar to my results for the chronically treated

group as described above. Higher activities have been reported with bezafibrate at a dose level of 200 mg/kg (Sharma et al.,1988a) as opposed to a dosing level of 75 mg/kg in the Table 6.8.

The Effect of Hypolipidaemic Drugs on Renal Lauric Acid Hydroxylase Activity in the Fischer Rat following Chronic Administration.

Treatment Lauric Acid Hydroxylase Activity

(nmol product/min/nmol P-450) 11-hydroxylaurate 12-hydroxylaurate

Control 10.04[a] (100)[b] 30.37 (100)

Clofibric Acid 23.05 (230) 75.46 (248) (50 mg/kg/day)

Bezafibrate 11.28 (112) 37.40 (123) (75 mg/kg/day)

Ciprofibrate 21.68 (216) 74.39 (245 (20 mg/kg/day)

^ Single daily dose for 26 weeks

[a] Values expressed as an average of two experimental results. [b] Figures in brackets are values expressed as a percentage of corresponding control. 282 in the present study. A 2 to 3-fold increase in 11- and

12-hydroxylauric acid formation (expressed as nmol product/min/mg protein) was observed in the rat following a 14 day clobuzarit treatment (Parker and Orton, 1980). However, when expressed as nmol product/min/nmol P-450, there were no treatment induced changes. The omega- and (omega-1)-lauric

acid hydroxylase activities in the rabbit were enhanced

approximately 2-fold by clofibrate and DEHP, unlike the effects seen in this study with ciprofibrate (Yamamoto et al.,1986).

Cytochrome P-452 content was also determined immunochemically

using an ELISA technique. The specific content of cytochrome P-452, (expressed as nmol P-452/nmol P-450), was increased in a dose-dependent manner in the Sprague Dawley, Wistar and Long Evans rat (table 6.6). This represented a 1.5 to 3-fold

induction. In the Gunn rat the specific content was increased 2-fold at both dose levels yet was unchanged in the Fischer

rat. When cytochrome P-452 content was expressed as percentage

of total cytochrome P-450 content, a dose-dependent increase was observed with all rat strains except the Fischer rat which exhibited a dose-dependent decrease in cytochrome P-452

content. The 1.5 to 3-fold induction (nmol P-452/nmol P-450) did not reflect a similar induction when expressed as

percentage of total cytochrome P-450 content and did not

correlate to treatment induced changes in total cytochrome P-450 content and lauric acid hydroxylase activity.

An increase in cytochrome P-452 specific content (nmol 283 P-452/nmol P-450) was observed at both dose levels in the mouse and at the high dose level in the hamster (table 6.7). This induction was in the order of 1.5 to 2-fold. Cytochrome P-452 was not detected immunochemically in guinea pig microsomes. A 3-fold increase in cytochrome P-452 content was

observed at the high dose level in the rabbit. Induction of

cytochrome P-452 content (expressed as percentage of total cytochrome P-450) was observed at the high dose levels in the hamster and rabbit. In the . mouse an increase in cytochrome P-452 content occurred at both dose levels. The extent of cytochrome P-452 induction (nmol P-452/nmol P-450) in the mouse and hamster correlated to the increase in cytochrome P-452 content when expressed as percentage of total cytochrome P-450 content and to the increased 12-hydroxylation of lauric

acid. These treatment induced changes did not reflect the extent of total cytochrome P-450 content. The 2 to 3-fold increase in cytochrome P-452 (nmol P-452/nmol P-450) in the

rabbit, accompanied decreased omega- and (omega-1)-

hydroxylation of lauric acid.

The basal levels of cytochrome P-452 (i.e. in untreated animals) when expressed either as specific content or as percentage of total cytochrome P-450, exhibited little strain

or species variation. Lauric acid hydroxylase activity could

not be detected at 3 mg protein/ml incubate for renal marmoset microsomes. In the rat and other species, activity was detected between 0.5 and 1.5 mg protein/ml incubate. It was not possible to immunochemically detect cytochrome P-452 in these marmoset samples. Cytochrome P-452 levels were not 284

quantitated in chronically treated rats following

ciprofibrate, bezafibrate and clofibric acid administration.

The immunochemical assessment of cytochrome P-452 has not previously been carried out in renal microsomes of different species. In addition there is little published data concerning renal cytochrome P-452 quantitation in the rat. This ELISA technique was developed in our laboratory and early results indicate that a wide range of peroxisome proliferators, ciprofibrate not examined, caused a 1 to 2-fold increase in cytochrome P-452 content (Sharma et al.,1988a). In the present study, a 1 to 1.6-fold increase was observed in different rat f strains. After treatment of rats with clofibrate, kidney microsomes showed an increased amount of cytochrome P-450 LAw [cytochrome P-452] (Hardwick et al.,1987). A slight increase in immunodetectable cytochrome P-452 in rat renal microsomes would appear to be a treatment-induced change associated with peroxisome proliferators.

6.3. GLUTATHIONE PEROXIDASE ACTIVITY.

The selenium dependent glutathione peroxidase (GPX) utilises hydrogen peroxide and organic hydroperoxides as substrates whereas glutathione-S-transferase (selenium independent) metabolises only organic hydroperoxides. As described previously (chapter 4) the hepatic GPX activity (expressed as percentage of control), with the substrates hydrogen peroxide and t-butylhydroperoxide, was very similar. As a result, t-butylhydroperoxide was used as a substrate for the selenium 285 dependent glutathione peroxidase.

There appeared to be little change in renal glutathione peroxidase activity in the rat following ciprofibrate administration (table 6.9). Activity was unchanged in the

Gunn, Sprague Dawley, Wistar and Long Evans rats. However in the Fischer rat, glutathione peroxidase activity was decreased at both dose levels to 50-60% of the control value. Cytosolic samples from the remaining species and chronically treated

rats were not available for enzyme determination. The effect

of peroxisome proliferators on renal glutathione peroxidase

activity has not previously been determined.

6.4. DISCUSSION.

The renal metabolism of fatty acids occurs either by

beta-oxidation through the mitochondrial pathway or by omega-

and (omega-1)-oxidation in the microsomal subcellular fraction. In addition, an extra-mitochondrial beta-oxidation pathway has also been demonstrated with this being located in peroxisomes in the medulla and medullary rays (Le Hir and Dubach,1982).

The distribution of omega-fatty acid hydroxylase activity was studied in various rat tissues and in the extra hepatic tissues examined, only the kidney was highly active (Ichihara et al.,1969). Marked omega-hydroxylase activity was observed in rabbit, mouse and guinea pig renal microsomes. Rat kidney & rH E 0 0 h JZ E 0 h Effect of Ciprofibrate on Renal Glutathione Peroxidase Activity Different In Rat Strains 05 CO < Eh Eh Q Q J Pd O W W CO > 3 O g J Eh Eh0) Q X D OD 3 CU w 4J o x 0 M P Q U E > HT f-(If) Eh O J as W I W h h a Oa

Qi < J J 1 —• 0

N N N N N N M N CN CN U CM CN U CN N I U CN CN U CN CN o O O O O O O rH rH rH rH rH lo 05 in Xi t— O CN o O o 0 I u \ \ Xi Xi r-1 r-1 3 c c g g O' O' O' O' rH

rH rH rH rH rH O If) r* CO — — O rH CTlrH O nr O CO \ N p 0 > O' rH 3 a) >h o a g g O' O' O' O' O' ai 0

•P O O O rH rH CN CO CTlLO H kO • rH ■— « CN CN >— O o O £ M o •H j Cu Xi N \ in 0 u P e e e O' O' O' O' O' O' C J4

O O O CN rH CN CN CDrH co

rH -P O O O CN CN rH ncn n o O If) — O O H If) O rl rl H O c O o P • • • • J \ i \ X M c > 0 O' o w g g O' O' O' O' (0 c

•o’ TJ ■o' OT co CO —>c t—' 0 CO O CO CD ) . .. O) .. rH rH 03 03 rH «P rH UH 0 tn •p •H MH rH •P > > a ■P 0 * rH 0 3 > a u P 0 in p 0 0 0 -H P 0 P 0 0 0 0 0 O' P O £ o 03 0 in x a 0 p g C ro O' P in 3 0 P 0 0 u tn

x

Xi rH 01 •H •P O' 0 0 a o p 0 C O P 3 0 0 in 0 0) P 0 a O C c u 0 C in 0 a in 0 m o P 0 in O' 0 0

287 cortex microsomes catalysed the omega-oxidation of laurate at

a similar rate to liver microsomes (Jakobsson et al.,1970).

However renal metabolism of the drug substrate aminopyrine and

testosterone was negligible. The dietary administration of

laurate enhanced renal cytochrome P-454 content and stimulated omega-hydroxylation of lauric acid 1.5-fold. In the kidney

laurate hydroxylation occurs at both positions yielding the

12- and 11-hydroxyproducts in the ratio of 2:1 (Ellin et

al.,1972). The ratio of omega- to (omega-1)-hydroxylation of

fatty acids decreases with increasing chain length of the substrate (Ellin et al.,1973). The kidney cytochrome P-450 system is specifically involved in omega- and (omega-1)- hydroxylation of fatty acids and in contrast to the liver, metabolism of model drug substrates is minimal in the kidney.

COMPARISON OF HEPATIC AND RENAL RESULTS.

Ciprofibrate administration caused an increase in total renal and hepatic cytochrome P-450 content in the Fischer, Wistar,

Sprague Dawley and Long Evans rat, hamster and rabbit (tables 4.1, 4.2, 6.1 and 6.2). Cytochrome P-450 content was elevated in the Gunn rat renal microsomes and hepatic cytochrome P-450 was increased in the mouse (table 4.2 and 6.1). Guinea pig cytochrome P-450 content was unchanged by treatment in both organs (tables 4.2 and 6.2). Following the chronic administration of clofibric acid and bezafibrate the hepatic cytochrome P-450 content was increased yet decreased with ciprofibrate administration (table 5.1). In the kidney, both clofibric acid and ciprofibrate caused a decrease in 288

cytochrome P-450 content whereas bezafibrate administration increased the renal cytochrome P-450 content (table 6.3). The specific content of cytochrome P-450 in untreated rat liver microsomes was 5 to 10-fold higher than in the kidney. In the mouse, hamster, guinea pig and rabbit the renal cytochrome P-450 content was 3 to 7-fold lower than in the hepatic samples.

With benzphetamine-N-demethylase activity in the kidney there were differential treatment changes: activity was increased at the high dose level in the Sprague Dawley rat and unchanged in the Gunn rat. Activity was decreased in the other rat strains (table 6.4). Hepatic metabolism of benzphetamine was decreased in the rat to approximately 20-50% of the control value at the high dose level in all strains (table 4.3). The rat hepatic metabolism of benzphetamine was 2 to 3-fold higher than in renal microsomes (comparing basal levels in untreated animals).

Hepatic and renal ethoxyresorufin-O-deethylase activity was decreased in the rat and mouse (table 4.3, 4.4, 6.4 and 6.5). Enzyme activity was reduced in rabbit kidney microsomes and was unchanged in liver microsomes (table 4.4 and 6.5). Ethoxyresorufin-O-deethylase activity was unchanged in both organs in the hamster and guinea pig (table 4.4 and 6.5). With the chronic administration of peroxisome proliferators, renal ethoxyresorufin-O-deethylase activity was unchanged yet was markedly decreased in liver microsomes (table 5.5 and 6.3). Ethoxyresorufin-O-deethylase activity in liver microsomes in 289

the rat and all species examined was expressed as nmol product/min/nmol P-450, as opposed to pmol product/min/nmol P-450 in renal metabolism.

It is clear that the hepatic cytochrome P-450 system is directed towards the metabolism of model drug substrates and xenobiotics whereas it would appear that the renal

contribution to xenobiotic metabolism is minimal.

In the rat, hepatic (omega-1)- and omega-hydroxylase activity was induced 2 to 5-fold and 10 to 15-fold respectively, whereas in the kidney, the increase was 1.5 to 2.5-fold for

omega- and (omega-1)-hydroxylation (table 4.5 and 6.6). There was considerable strain variation in response. With chronic administration of peroxisome proliferators the increase in lauric acid hydroxylation was 1 to 2.5-fold at both positions (table 6.8). However in the liver, omega-hydroxylation was induced 8 to 20-fold (table 5.5). In the mouse and hamster hepatic (omega-1)- hydroxylation was induced between 1.5 to

3-fold whereas the increase in 12-hydroxylase activity was 3-, 5- and 10-fold in the rabbit, hamster and mouse respectively (figure 4.1). In the rabbit, renal omega- and (omega-1)- hydroxylase activity was decreased (table 6.7). Kidney

(omega-1)-hydroxylation was induced 2-fold in the guinea pig and 1.5-fold in the hamster (table 6.7). The increase in omega-hydroxylase activity was between 1.5 and 2-fold in the mouse and hamster (table 6.7). In the rat hepatic cytochrome P-452 content (% of total P-450) was induced 5 to 10-fold

(figure 4.3). In the kidney cytochrome P-452 induction was 290 less than 2-fold (table 6.6). Likewise marked hepatic

induction was observed in the mouse, hamster and rabbit with minimal renal treatment-induced changes.

Kidney microsomal (omega-1)- and omega-hydroxylation of lauric

acid was 3 to 14-fold and 9 to 22-fold higher than in liver

microsomes. In the mouse, renal hydroxylation of lauric acid at both positions was 7-fold higher than the liver, whereas in the hamster and guinea pig only renal omega-hydroxylation was greater, 7- and 3.5-fold, respectively. (Omega-1)- and

omega-hydroxylase activity was 2- and 5-fold higher in rabbit

kidney as opposed to the liver. It is interesting to note that

in guinea pig liver microsomes (omega-1)-hydroxylation predominated whereas in the kidney the reverse was true.

DISCUSSION

This information reiterates the belief that renal cytochrome P-450 metabolism is directed towards omega- and (omega-1)- hydroxylation of fatty acids as opposed to the liver where xenobiotic metabolism predominates in the untreated animal. It

is interesting to note that in the rat, the basal levels of cytochrome P-452 (nmol P-452/nmol P-450) in the liver and kidney were similar. However when these were expressed as a percentage of total P-450, they were markedly different i.e.

3-6% in the liver as opposed to 18-30% in the kidney. A similar observation was demonstrated in the mouse, hamster and rabbit. 291 It is apparent that in the untreated animal there is considerable variation in enzyme activity between the liver and kidney and that with the cytochrome P-450 system, this

represents fundamental differences in the metabolic role of

these two organs. It has been demonstrated that species and rat strain variation in response occurs. Treatment-induced changes (whether observed as increased or decreased enzyme activities) were exhibited in the kidney, reflecting a similar

but more marked response in the liver. In general the renal responses are an order of magnitude lower than in the liver. The induction of cytochrome P-452, measured immunochemically and catalytically, was considerably lower in the kidney of the Gunn and Sprague Dawley rat, hamster and mouse. Renal omega- and (omega-1)-hydroxylase activity was unchanged in the Fischer and Long Evans rat and induction of only the

(omega-1)-hydroxylase occurred in the Wistar rat. Similarly renal (omega-1)-hydroxylase activity was induced in the guinea pig yet was unchanged in the liver. The increase in cytochrome P-452 content in the rabbit was not associated with an increase in catalytically active protein but with decreased omega- and (omega-l)-hydroxylation of lauric acid. The reason for this is unclear. It is possible that in the rabbit, liver cytochrome P-452 is responsible for hydroxylation of fatty acids. However in the kidney a haemoprotein with strong homology to cytochrome P-452 but with minimal activity towards omega- hydroxylation, may be recognised by the antibody under the ELISA conditions used. The chronic administration of peroxisome proliferators would appear to result in more marked renal enzyme changes. In comparison with other peroxisome 292 prolif erators, ciprofibrate although a highly potent compound in the liver, has minimal effects in the kidney suggesting that the hepatic response is tissue specific as opposed to tissue common.

It has been suggested that the renal omega- and (omega-1)- hydroxylation of lauric acid is catalysed by a single monooxygenase enzyme (Ellin and Orrenius,1975). Two kidney

cortex haemoproteins, cytochrome P-450k-l and P-450k-2, were

purified from untreated rats (Yoshimoto et al.,1986). Both forms catalysed omega- and (omega-1)-hydroxylation of fatty acids. In addition P-450k-l metabolised prostaglandin A-l and prostaglandin A-2 at the omega-position. Cytochrome P-450k-5 was similarly isolated from untreated rats and this isoenzyme, with an absorption maximum at 452nm, catalysed the omega- and (omega-1)-hydroxylation of lauric acid (Imaoka and Funae, 1986). Guinea pig renal cytochrome P-450 monooxygenases have also been found to metabolise prostaglandin E-l at the omega- and (omega-1) positions (Navarro et al.,1978). Cytochrome

P-450ka, isolated from untreated rabbit kidney cortex microsomes catalysed omega-hydroxylation of prostaglandin A-l and prostaglandin A-2 and the omega- and (omega-1)- hydroxylation of fatty acids (Kusunose et al.,1985). This cytochrome P-450 isoenzyme was induced by clofibrate and DEHP (Yamamoto et al.,1986).

It has been reported that rabbit kidney cortex microsomes metabolise arachidonic acid to the 19- and 20-hydroxyproducts, presumably via a cytochrome P-450 monooxygenase (Morrison and 293

Pascoe,1981, and Oliw et al.,1981). The formation of the omega- and (omega-1) products of arachidonic acid occurred preferentially (75%) in the cortex whereas the formation of prostaglandins predominated in the medulla (60%) (Chacos et al.,1983). The omega- and (omega-1) hydroxylation of arachidonic acid can occur prior to or after epoxidation via

the epoxygenase pathway (Needleham et al.,1986). Epoxy-

eicosatrienoic acids have been isolated and purified from female rabbit kidneys (Falck et al.,1987). In the rabbit, cytochrome P-450 isoenzyme(s) specific for arachidonic acid metabolism are primarily localised in the outer medulla (Schwartzman et al.,1986).

It is apparent that in the kidney considerable omega- and (omega-1)-hydroxylation of fatty acids (such as lauric acid, arachidonic acid and prostaglandins) occurs. It is equally

clear that there is more than one renal cytochrome P-450

isoenzyme capable of omega- and ( o m e g a - 1)-hydroxylation of

fatty acids. Cytochrome P-450 LAw was purified from livers of

clofibrate treated rats (Hardwick et al.,1987). Immunoblot analysis with anti cytochrome P-450 LAw antibody revealed the

presence of two clofibrate-induced cytochrome P-450 proteins in both the liver and kidney. The second protein of higher

molecular weight was thought to be a closely related isoenzyme

in the same cytochrome P-450 family or a modified form of

cytochrome P-450 LAw. Similarly, western blot analysis of control and clofibrate liver microsomes with anti cytochrome

P-452 antibody revealed two protein bands (Earnshaw et al.,1988). Total RNA samples prepared from control and 294

clofibrate induced hepatic and renal microsomes were hybridised with the cytochrome P-452 cDNA probe (Earnshaw et al.,1988). Comparable levels of hybridisation to a mRNA species of approximately 2 Kbp were observed with control liver and kidney RNA preparations. A marked induction in

cytochrome P-452 mRNA levels occurred in rat liver following

clofibrate administration; this induction was less marked in the kidney (Earnshaw et al.,1988). It should be noted that rabbit pulmonary prostaglandin omega-hydroxylase (cytochrome P-450 p-2) shares 74% amino acid similarity with rat hepatic cytochrome P-450 LAw suggesting that both forms are members of

the same cytochrome P-450 gene family (Hardwick et al.,1987, and Matsubara et al.,1987).

There may be a tissue specificity for the presence or inducibility of these cytochrome P-452 enzymes. Further investigation is required to clarify the renal cytochrome

P-450 isoenzymes responsible for omega- and (omega-1)- hydroxylation. Correlation using catalytic and immunochemical techniques and analysis of RNA levels and gene expression of these renal hydroxylase forms with hepatic cytochrome P-452, should be attempted. DNA sequence analysis of genomic and cDNA clones will enable the relationship between cytochromes P-452 in different tissues, strains and species to be elucidated.

Rat hepatic glutathione peroxidase activity was 1.75 to 4-fold higher than in the kidney. Renal GPX activity was unchanged following ciprofibrate administration with the exception of the Fischer rat where activity was shown to be decreased. In 295 the hepatic cytosolic fractions, GPX activity was decreased to 50% of the control value.

The effects of a wide range of peroxisome prolif erators on renal mitochondrial and peroxisomal enzymes has been studied

in the Wistar rat (Sharma et al.,1988a). A 1.5 to 3-fold induction in peroxisomal beta-oxidation was observed with DEHP, bezafibrate, nafenopin and clofibrate. A maximal 2-fold induction in carnitine acetyltransferase, carnitine

palmitoyltransferase (mitochondrial enzyme), total enoyl-CoA

hydratase (mitochondrial and peroxisomal) and peroxisomal

enoyl-CoA hydratase activities resulted from treatment (Sharma et al.,1988a). The relative potencies of the compounds tested varied widely. The induction of renal enzymes was considerably lower than in the liver. Peroxisome proliferation has been

demonstrated in clofibrate treated kidney cells (Svoboda et

al.,1969). This proliferation was sex-related with male rats

displaying more extensive peroxisome proliferation than females at the same dose levels (Henry and De Morrow, 1985). Induction of peroxisomal beta-oxidation, enoyl-CoA hydratase and PPA 80 with an associated peroxisome proliferation was

reported in mouse kidney following methyl clofenapate, BR-931

and Wy-14,643 treatment (Lalwani et al.,1981).

Induction of peroxisomal enzymes occurs in the kidney of treated rats and mice. In each instance, renal response was minimal compared to the enzyme changes demonstrated in the liver. Renal peroxisomal enzyme parameters have not been studied in the hamster, guinea pig and rabbit although with 296

reference to the hepatic response, species susceptibility varied with the rat > mouse > hamster > rabbit. The guinea pig was not responsive to ciprofibrate treatment.

The absorption and pharmacokinetic properties of ciprofibrate

in the rat have been discussed in the previous chapters. The treatment-induced changes in the kidney, in comparison to the liver, were minimal indicating that the hepatic response was probably tissue specific. This may be caused by unequal distribution of ciprofibrate in vivo i.e. low levels of the drug in the kidney versus high levels in the liver. However the tissue distribution of radiolabelled ciprofibrate has been studied in rats (Davison et al.,1975). Autoradiographs were developed at 24 hours and 16 days following administration (3 mg/kg). At the 24 hour interval, ciprofibrate levels were similar in both organs: 11.2 to 11.4 ug ciprofibrate/g tissue for the liver and kidney respectively. At 16 days the ciprofibrate distribution was not markedly different in these tissues (1.0 to 0.3 ug ciprofibrate/g tissue for the liver and kidney). Assuming that in both instances all the detected radiolabel was ciprofibrate, the apparent tissue specific response of the liver as opposed to the kidney, was not due to a tissue selective distribution of the compound. The lack of response in the kidney following administration of peroxisome prolif erators is likely to be due to inherent tissue differences in response. 297

Chapter 7

DISCUSSION

7.1. INTRODUCTION

The aims of the work presented in this thesis were fourfold; i) to utilise the pharmacokinetic properties of peroxisome proliferators to rationalise the selection of both dose levels and frequency of dosing for subsequent induction studies, ii) to examine the species variability in hepatic response to ciprofibrate, iii) to compare and contrast hepatic and renal responses to ciprofibrate, and iv) to consider the validity of using laboratory animal studies to predict the biological response to ciprofibrate in a non-human primate, and ultimately in man.

7.1.1. Influence of pharmacokinetics on the inductive potency

of three hypolipidaemic drugs in the Fischer rat.

It has been established that the pharmacokinetic properties of clofibric acid, bezafibrate and ciprofibrate are markedly different with ciprofibrate exhibiting a long half life (3-4 days) and a prolonged pharmacological action (Edelson et al.,1979). With once daily dosing, the inductive potencies of these three compounds were dissimilar and the order of response was ciprofibrate (20mg/kg/day) > bezafibrate

(75mg/kg/day) > clofibric acid (50mg/kg/day). My results have shown that by altering the dosing regimen with respect to 298 pharmacokinetic parameters (i.e. clofibric acid and bezafibrate administration twice daily and ciprofibrate once every 48 hours) the different inductive potencies of the three

compounds were ablated. Accordingly, the importance of utilising pharmacokinetic properties to select dose regimens

in comparative studies cannot be ignored or underestimated.

7.1.2 Species Variability in the Hepatic Response to Ciprofibrate Administration

In all rat strains, the mouse, hamster and rabbit, peroxisomal beta-oxidation and carnitine acetyltransf erase activity was increased and the elevation in lauric acid omega-hydiroxylase activity was associated with an increase in the specific content of cytochrome P-452. No treatment related enzyme changes were observed in the guinea pig. In the riiarmoset peroxisomal beta-oxidation was increased only slightly with unchanged carnitine acetyltransf erase and lauric acid omega-hydroxylase activities. Cytochrome P-452 could not be detected immunochemically in marmoset or guinea pig liver microsomes. In the marmoset, mitochondrial enzyme activities were elevated. The responses in the rat and marmoset to the chronic administration of ciprofibrate were similar to those observed following acute exposure. In the marmoset the increase in succinate dehydrogenase activity and the decrease in (omega-1)-lauric acid hydroxylase activity was maintained after a one month recovery period. The induction of the peroxisomal beta-oxidation system was found to be readily reversible. In the marmoset, peroxisomal beta-oxidation was 299 increased and additionally succinate dehydrogenase and alpha-glycerophosphate dehydrogenase activities were elevated

whereas in the rat only mitochondrial alpha-glycerophosphate dehydrogenase activity was stimulated.

There is a marked species variability in the hepatic response to ciprofibrate. The rat, mouse, hamster and rabbit are termed

responsive due to induction of a similar pattern of enzymes

(cytochrome P-452 and lauric acid omega-hydroxylase, carnitine acetyltransferase and peroxisomal beta-oxidation). The guinea pig and marmoset were designated non-responsive.

7.1.3. The Rat and Species Renal Response to the

Administration of Ciprofibrate

The effect of ciprofibrate on the renal drug metabolising

enzyme system was examined in different rat strains and

species. Renal enzyme changes occurred but in each case the response was less marked than that in the liver. The basal levels (i.e. in the untreated animal) of cytochrome P-450

(specific content) and benzphetamine-N-demethylase and ethoxyresorufin-O-deethylase activities were considerably lower in the kidney in comparison to the liver whereas renal lauric acid omega- and (omega-1)-hydroxylation was more extensive. The increase in omega- and (omega-1)-hydroxylation of lauric acid was minimal in the rat and induction did not occur in every strain or species. Cytochrome P-452 content was slightly increased in responsive species yet did not correlate 300 with catalytic activity. In the rat, renal glutathione peroxidase activity was unchanged. The comparison of liver and kidney responses in the rat, mouse, hamster and rabbit, < treated with ciprofibrate, indicates that the tissue response is largely specific to the liver.

7.1.4. Extrapolation of Animal Data to Predict the Response

in Non-human Primates.

The rat, mouse, hamster and rabbit are considered responsive species whereas the guinea pig and marmoset are non-responsive. It is clear that using rodent data to predict

the response in the non-human primate (i.e. marmoset) in this

instance, is questionable as the pattern of hepatic enzyme changes in these two species were different. All the peroxisome prolif erators that have been tested for carcinogenic potential, cause hepatocarcinogenesis in rats and mice and whether or not this can be considered as a possible

outcome in man is of obvious concern. It is important to

consider the mechanisms of induction of enzyme changes and hepatocarcinogenesis in the rat. This may be a useful consideration when comparing the species response to ciprofibrate and may reveal a greater insight into the I apparent species specificity of the hepatic response.

7.2. MECHANISMS OF TOXICITY IN THE RAT

A hypothesis is presented in figure 7.1 to explain the induction of enzyme parameters and hepatocarcinogenesis in the 301 rat. This is based largely on the data described in the previous chapters with ciprofibrate in the rat, although a

similar response by all peroxisome prolif erators would be expected. Induction of the hepatic response in different species will be examined in a later section. The liver changes observed in the rat on acute exposure :- hepatomegaly,

increased omega-lauric acid hydroxylase, carnitine

acetyltransf erase and peroxisomal beta-oxidation activities - are sustained at 26 weeks and in long-term bioassays accompany hepatocarcinogenesis. With acute and chronic treatment these enzyme changes are not associated with liver toxicity per se

(i.e. cell necrosis) and may be used as an index of toxicity to predict the probability of ultimate liver damage, hepatocarcinogenesis.

Peak plasma levels of ciprofibrate were achieved two hours post dose (Edelson et al.,1979). Metabolism of the peroxisome prolif erator to the active compound may be necessary, for example clofibrate (ethyl ester) is converted to clofibric acid. On treatment an increase in liver/body weight ratio was observed and this was probably due to hyperplasia and hypertrophy (Mann et al.,19.85). A midzonal and periportal accumulation of fat was observed with treatment and this was similar to the fatty change produced by an excess intake of lipid (Mann et al.,1985). The composition of the accumulated lipid has not been defined. A shift in the composition of free fatty acids in the liver was observed in rats fed a partially hydrogenated fish oil diet with a significant increase in the levels of 19; 1 and 18:1 fatty acids and a decrease in the 302

Figure 7.-1. Possible Mechanisms of Induction of Enzyme Changes and Hepatocarcinogenesis in the Rat. Peroxisome.Proliferator ▼ Liver i Hepatomegaly

Increased Proliferation Increased Unchanged Decreased Carnitine of Cytochrome Catalase Glutathione Acetyl- Peroxisomes P-452 Peroxidase Transferase ( + )I Accumulation of Medium and Long Chain Fatty Acids

Omega-/(omega-1)- Arachidonic ( + ) Hydroxy Medium or Acid Long Chain Fatty Metabolism ( + ) Acids (-)

Medium ior Long Chain Dicarboxylic (-) Acids j Peroxisomes

— _ Peroxisome Proliferation and Induction of Enzymes

Short Chain Increased Hydrogen (Epoxides) Fatty Acids Peroxide Lipid 1 Peroxidation Mitochondria Active Oxygen Hepatomegaly) I Increased beta- Oxidation in (DNA Damag Mitochondria

HEPATOCARCINOGENESIS

(+) Activation of Enzyme Activity (-) Inhibition of Enzyme Activity 303 levels of 18:2 and 20:4 fatty acids (Nilsson et al.,1987). In clofibrate treated mice, thecomposition of individual free fatty acids of the liver triacylglycerols showed a shift

towards shorter fatty acids and nutritionally essential

alpha-linolenic acid (Jones and Neill,1982). A substrate

overload linked induction ofcytochrome P-452 may occur as this isoenzyme utilises medium and long chain saturated and

unsaturated fatty acids [e.g. lauric and arachidonic acids] (Bains et al.,1985, Tamburini et al.,1984). Cytochrome P-452 was increased several fold on treatment. Microsomal oxidation

of fatty acids results in omega- and (omega-l)-hydroxy medium and long chain fatty acids and subsequent oxidation by cytosolic alcohol and aldehyde dehydrogenases to dicarboxylic fatty acids may occur (Mitz and Heinrikson,1961). The short chain fatty acids are probably metabolised by mitochondria prior to microsomal oxidation. According to their chain length

substrate specificity, medium and long chain monocarboxylic-

and omega- hydroxymonocarboxylic fatty acids can be metabolised by mitochondria but the dicarboxylic fatty acids are substrates for peroxisomal beta-oxidation only (Vamecq and Draye, 1987). The end products of omega- and (omega-1)- hydroxylation may ultimately initiate peroxisomal proliferation and induction of peroxisomal beta-oxidation enzymes. Long chain dicarboxylic fatty acids would undergo several cycles of the peroxisomal beta-oxidation pathway until the chain length was sufficiently shortened to provide suitable substrates for the mitochondrial beta-oxidation system. 304 Induction may similarly occur by substrate overload. Induction of microsomal, peroxisomal and mitochondrial oxidising

pathways have been observed in rats fed a partially

hydrogenated fish oil / high fat diets (Ishii et al.,1980, Krahling and Tolbert,1982 and Nilsson et al.,1987). The hepatic peroxisomal beta-oxidising system and fatty acyl-CoA

oxidase activity were significantly increased with a high cholesterol diet (Hayashi et al.,1986). Cholesterol may be a

physiological substrate for liver peroxisomes leading to a

substrate overload induction, however as yet, there is no evidence for peroxisomal metabolism of cholesterol. The effect of this diet on the microsomal omega-oxidation system was not studied. It has recently been demonstrated that 3-hydroxy-3- methylglutary1-coenzyme A reductase, the key regulatory enzyme

in cholesterol biosynthesis, is localised in peroxisomes in

addition to the endoplasmic reticulum (Keller et al.,1985). A diet rich in (n-3) fatty acids hasbeen associated with increased peroxisomal beta-oxidation activity and lowered plasma triacylglycerol levels (Yamazaki et al.,1987).

Chlorpromazine administration causes a slow accumulation of lipid in large droplets in centrilobular hepatocytes (Price et al.,1985). Peroxisomal beta-oxidation and microsomal omega-oxidation of fatty acids was unchanged, and these latter authors concluded that hepatic lipid accumulation does not automatically induce fatty acid oxidising enzymes. However, accumulation of a specific fatty acid or lipid may be necessary to trigger cytochrome P-452 induction and the pattern of lipid accumulation observed in peroxisome 305 proliferator treated animals was different to that in animals administered chlorpromazine (Price et al.,1985).

If the hypothesis accounting for induction of the microsomal, peroxisomal and mitochondrial oxidising systems in hepatocytes is attributed to substrate overload as outlined above, the chronological sequence of induction must be established. By

increasing the microsomal omega-oxidation pathway, there is a

shift in fatty acid substrates from the mitochondria to peroxisomes as evidenced by an increase in long chain dicarboxylic fatty acids (Vamecq and Draye,1987). Peroxisomal beta-oxidation of long chain dicarboxylic acids would give rise to short and medium chain fatty acids, substrates for the

mitochondrial system. However it has been reported that in

treated rats, mitochondrial beta-oxidation proceeds at a maximum rate before enhancement of peroxisomal beta-oxidation occurs (Berge and Aarsland,1985). The mechanism of induction was attributed to the cellular level of long chain acyl-CoA. A

time-course study examining the induction of peroxisomal

beta-oxidation and microsomal omega-oxidation following a

single dose of clofibrate has been examined (Milton et al.,1988a). There was no apparent time lag between induction of fatty acid oxidation in the separate subcellular fractions.

However this experiment was conducted in vivo and may have been insufficiently sensitive to differentiate induction in the two subcellular organelles. The use of primary hepatocytes at frequent time periods after ciprofibrate challenge may be a useful experimental system to clarify the precise temporal relationships between induction of fatty acid oxidation in the 306

endoplasmic reticulum and peroxisomes.

Administration of medium to long chain dicarboxylic acids to rats had no effect on peroxisomal beta-oxidation indicating

that the products of microsomal omega-oxidation do not cause peroxisomal proliferation and induce peroxisomal enzymes

(Nilsson et al.,1987). However these dicarboxylic acids were administered by oral gavage and considerable metabolism may have occurred prior to entry into the liver causing an inadequate intra-hepatic lipid accumulation. The hypothesis that long chain dicarboxylic acids may cause peroxisome proliferation and induce peroxisomal enzymes is substantiated by a series of experiments recently completed with tiadenol.

Tiadenol, (bis( hydroxyethylthio)- 1.10-decane, is a rodent peroxisome prolif erator (Berge et al.,1984). The tiadenol molecule is characterised by two primary alcohol groups

(figure 7.2). The compound, (bis( carboxy ethylthio)-1.10- decane, a possible dicarboxylic metabolite of tiadenol, was synthesised [figure 7.3] (Berge et al.,1987). Normal beta-oxidation of this derivative would be blocked because of the sulphur atoms in the beta positions of the carbon chain. The administration of this dicarboxylic derivative to rats caused a decrease in serum cholesterol and triglyceride levels, hepatomegaly and induction of peroxisomal beta-oxidation, responses similar to that seen with the parent compound, tiadenol (Berge et al.,1987). Cytosolic acyl-CoA hydrolase and enzyme activities involved in the formation and breakdown of long chain acyl-CoA/xenobiotic acyl-CoAs were increased in treated animals. It was suggested that the 307

Ficure 7.1. Chemical Structure of Tiadenol (bis( hydroxyethvlthio) - 1.10' Decane

HOCH2-CH2-S-(CH2),o-S-CH2-CH2OH

Figure 7.2. Chemical Structure of The Dicarboxylic Derivative of Tiadenol, (bis (carboxy-ethylthio) -1.10-Decane

HOOC-CH2-S-(CH2),o-S-CH2-COOH 308 CoA-ester of this dicarboxylic derivative of tiadenol may initiate peroxisome proliferation and induction of peroxisomal enzymes (Aarsland et al.,1987, Berge et al.,1987). The in vivo formation of this derivative has yet to be established.

The mitochondrial fatty acid beta-oxidation system is

dependent on carnitine acetyltransf erase as an intermediate carrier of acyl-substrates into the mitochondrial matrix. Carnitine is not involved in the entry of fatty acids into peroxisomes but may play a role in the removal of end products

which could be shuttled to the mitochondria for further

oxidation or conversion into ketone bodies (Lazarow, 1982). Carnitine acetyltransf erase activity was induced on treatment and this represented an increase in enzyme activity in mitochondria and peroxisomes. Long chain acyltransferase

activity is observed only in mitochondria and this was

increased 2 to 4-fold on treatment (Sharma et al.,1988). This marked increase in enzyme activity would facilitate the transfer of fatty acyl substrates out of the peroxisome and into the mitochondria. Induction could be associated with a substrate overload or altered lipid metabolism as indicated in figure 7.1. The stimulation of mitochondrial beta-oxidation following clofibrate treatment reflects increased activity of non-latent CoA-carnitine acyltransferase and increased activity of beta-oxidation enzymes (Mackerer,1977). Hepatic synthesis of carnitine and its esterification to acylcarnitines was increased (Paul et al.,1986). The induction of cytochrome P-452 by peroxisome proliferators has been shown to correlate well with the increase in peroxisomal and 309 mitochondrial parameters (Gibson and Sharma, 1987).

Induction of fatty acid metabolism could also result from an interaction of the peroxisome proliferator itself and/or a metabolite with a receptor in the cytoplasm of hepatocytes (Reddy and Lalwani, 1983), in a similar fashion to the glucocorticoid receptor (Heuman et al.,1982) and the Ah receptor (Poland and Knutson,1982). Tritiated nafenopin was shown to bind to a specific saturable pool of binding sites in cytosols from rat liver and tissue levels of this protein correlated with the ability of nafenopin to induce hepatic peroxisomal enzymes (Lalwani et al.,1983). This peroxisome-prolif eration binding protein has been purified and may play an important role in the regulation of the induced response (Lalwani et al.,1987). It is both plausible and tempting to speculate that a cytosolic receptor binding protein is responsible for the induction of a "battery" of enzymes. However the existence of this putative receptor has not been universally accepted with the original binding studies still to be verified in other laboratories. Our laboratory has found no evidence of specific binding of peroxisome prolif erators to hepatic cytosolic proteins or receptor material (Milton et al.,1988).

Peroxisome prolif erators and/or their metabolites could act as substrates for the different oxidation pathways thereby initiating enzyme induction. The limited formation of hydroxybezafibrate by a cytochrome P-450 mediated process has been demonstrated in vitro (Facino et al.,1981). Tiadenol is 310

metabolised by an S-oxidative hepatic microsomal cytochrome

P-450 dependent monooxygenase (Facino et al.,1986). Similarly

in vitro metabolism of ME HP to omega-, (omega-1)- and

(omega-2)-hydroxy products has been observed and this was

thought to be by cytochrome P-450 isoenzymes associated with

hydroxylation of fatty acids (Albro et al.,1984). Phthalate

ester plasticisers may undergo mitochondrial or peroxisomal

beta-oxidation. This mechanism lacks evidence and would not

account for the marked variation in enzyme induction observed

with this class of structurally diverse compounds.

Acute toxicity in rats following exposure to a variety of

xenobiotics may result in carcinogenicity. With the

administration of peroxisome prolif erators, enzyme induction occurs but this is not related to gross toxicity, however hepatocarcinogenesis develops in treated rats between 1 and 2 years (Reddy and Lalwani, 1983). These compounds are uniformly negative in a wide range of mutagenicity tests indicating a class of non-genotoxic carcinogens. No binding of clofibrate or fenofibrate to hepatic nuclear DNA has been detected (Von

Daniken et al.,1981) and nafenopin and Wy-14,643 exhibited no significant binding to DNA in vivo or in vitro (Goel et al.,1985). No hepatic DNA damage has been observed in rats treated with DEHP and DEHA (Ruddick et al.,1981, Von Daniken et al.,1984) and single and sub-chronic dosing of DEHP to rats was not associated with initiating potential (Garvey et al.,1987). Both DEHP and its metabolite MEHP were non- genotoxic in a range of short term in vitro mutagenicity assays (Schmezer et al., 1988). No peroxisome proliferator DNA 311

adducts were detected in hepatocytes under in vivo and in

vitro conditions with clofibrate, ciprofibrate, W y - 14,643 or

DEHP (Gupta et al.,1985).

How then does one explain the rodent carcinogenicity of

peroxisome prolif erators? It has been suggested that the

marked increase in peroxisomal beta-oxidation leads to an

excessive production of hydrogen peroxide by fatty acyl-CoA

oxidase. The specific activity of catalase was unchanged by

treatment (total activity was increased in parallel with

hepatomegaly) and this may be insufficient to remove excessive

hydrogen peroxide produced in stimulated peroxisomes. In

animals treated for 26 weeks with bezafibrate or ciprofibrate

the specific activity of catalase was significantly decreased.

It has been suggested that stimulated peroxisomes may be

"leaky" and that hydrogen peroxide diffuses into the cytoplasm

of the cell. The hydrogen peroxide may interact with a non-DNA

target producing active oxygen species. The decrease in

hepatic glutathione peroxidase activity (to approximately 50%

of control value) observed with ciprofibrate in my studies may potentiate oxidative stress by inadequate removal of cytosolic hydrogen peroxide. The active oxygen species may cause lipid peroxidation, DNA attack and hepatocarcinogenesis. Consistent with this lipid peroxidation proposal of neoplastic transformation is the observed accumulation of lipofuscin in the livers of treated rats (Reddy et al.,1982).

Hydrogen peroxide may also be produced in the endoplasmic reticulum by the uncoupling of cytochrome P-450 mixed function 312 oxidase activity. Metabolic pathways involving cytochrome

P-450 enzymes may initiate or modulate oxidative damage due to

oxygen radicals (Gonder et al.,1985). In mice, oxygen toxicity

has been shown to parallel the genetic control of cytochrome

P-450 enzyme induction and toxicity developed at a time when

the cytochrome P-450 enzyme system was induced by oxygen. With

clofibrate administration microsomal and cytosolic epoxide

hydrolases were induced and this may be a protective

mechanism (Moody et al.,1985). Under physiological conditions

the hydrogen peroxide produced in peroxisomes is largely

disposed of by catalase whereas hydrogen peroxide arising from

mitochondria, endoplasmic reticulum or cytosolic enzymes such

as superoxide dismutase, is acted upon by glutathione

peroxidase (Halliwell and Gutteridge,1985).

Evidence for the involvement of hydrogen peroxide and or free

radicals in hepatocarcinogenesis has been indicated by the

inhibitory influence of the antioxidants, ethoxyguin and

2( 3)-tert-butyl-4-hydroxyanisole (BHA) on ciprofibrate-induced

hepatic tumourigenesis (Rao et al.,1984). This was believed to

be due to the ability of antioxidants to scavenge hydrogen

peroxide and/or free oxygen radicals. DNA damage occurred when

purified tritiated SV40 DNA was co-incubated with a system

containing purified peroxisomes, palmitate and the cofactors

required for beta-oxidation (Fahl et al.,1984). The extent of

DNA damage correlated with the level of hydrogen peroxide

generated by the control and Wy-14,643 treated peroxisomes.

Catalase has a relatively minor role in hydrogen peroxide catabolism at low rates of hydrogen peroxide generation in the 313

endoplasmic reticulum however enzyme function increases as the

rate of production is enhanced (Jones et al.,1981). Catalase

readily leaks from peroxisomes (Alexson et al.,1985). It has

been shown that liver peroxisome-enriched fractions from

animals treated with peroxisome prolif erators can produce

increased levels of hydrogen peroxide and hydroxyl radicals by

the fatty acyl-CoA oxidase system, the contribution of such

reactive oxygen species to any genotoxicity in vivo remains to be firmly established (Elliot et al.,1986).

Hepatic lipid peroxidation was observed in rats treated with ciprofibrate and Wy-14,643 and this was associated with increased hydrogen peroxide generation and decreased glutathione peroxidase and glutathione reductase activities, indicative of oxidative stress (Goel et al.,1986). Hepatic glutathione peroxidase and superoxide dismutase activities were decreased with fenofibrate administration with a concomitant increase in malonyl dialdehyde production (Ciriolo et al.,1984). Lipid peroxidation is an efficient source of peroxyl radicals (Marnett,1987). The extent of hepatic lipid peroxidation in DEHP treated rats displaying peroxisome proliferation, did not differ from that in controls (Ruddick et al.,1981). The role of lipid peroxidation in preneoplastic change has yet to be clarified. The extent of radical production in the livers of rats exposed to peroxisome proliferators is estimated to be insufficient to result in a biologically significant degree of DNA damage (Elliot and

Elcombe,1987). However the hepatocarcinogenic potency of ciprofibrate, DEHP and DEHA correlated with their ability to 314

induce peroxisome proliferation, peroxisomal beta-oxidation

and PPA80 (Reddy et al.,1986a).

Reactive fatty acid epoxides could also be formed in

hepatocytes and these may lead to a direct covalent

interaction with cellular macromolecules and DNA. The liver

microsomal cytochrome P-450 system metabolises polyunsaturated

fatty acids, notably arachidonic acid, to a variety of

oxygenated products such as hydroxyeicosatetraenoic acids

(HETEs), epoxyeicosatrienoic acids (EETs) and omega- and

(omega-l)-hydroxylated acids (Capdevila et al.,1981, 1982,

Chacos et al.,1982, 1983, Manna et al.,1983). It has been

reported that ciprofibrate administration stimulated the

omega- and (omega-1)-oxidation of arachidonic acid in rat

liver microsomes, concomitant with a net decrease in the

formation of both HETEs and EETs (Capdevila et al.,1985),

whereas phenobarbitone administration enhanced the metabolism

of arachidonic acid to the 11-/12- and 14-/15-epoxides (Oliw

et al.,1982).

Peroxisome prolif erators are potent promoting agents. The

altered hepatic areas, neoplastic nodules and hepatocellular

carcinomas induced by peroxisome prolif erators are negative

for gamma-glutamyltransf erase (G-GT) and placental

glutathione-S-transferase unlike the hepatic lesions induced

by classical liver carcinogens (Rao et al.,1986a). Peroxisome prolif erators do not appear to derepress the activity of the

G-GT gene during hepatocarcinogenesis. 315

The specific content of cytochrome P-450 was decreased in rats

treated with ciprofibrate for 65-70 weeks and aryl hydrocarbon

hydroxylase activity was similarly reduced (Rao et al.,1987).

With a 26 week ciprofibrate treatment period, the activities

of benzphetamine-N-demethylase and ethoxyresoruf in-0-

deethylase were decreased to approximately 50% of the control

value. The specific content of cytochrome P-452 was increased

11-fold and lauric acid omega-hydroxylase activity elevated

20-fold. Induction of cytochrome P-452 by peroxisome

prolif erators showed a negative correlation when compared to benzphetamine-N-demethylase and ethoxyresorufin-O-deethylase activities indicating a specific gene switch on for cytochrome

P-452 concomitant with a gene switch off for these two activities (Gibson and Sharma, 1987).

These chemicals may promote tumour formation from cells altered through effects other than induced initiation eg. spontaneous mutation or inherited genetic abnormalities

(Williams,1981). It has been demonstrated that livers of untreated aged Wistar rats of both sexes contain putative preneoplastic cells and rates of DNA synthesis in these foci were higher than in normal liver and were stimulated by a single dose of nafenopin (Schulte-Hermann et al.,1983).

Non-genotoxic carcinogens could produce liver tumours if they promote development from preneoplastic foci susceptible to endogenous and exogenous growth stimuli. Depending on the strain and the conditions, liver tumours are not particularly uncommon in elderly, untreated SPF (specified pathogen free) rats fed ad libitum. The percentage of liver tumours in 316

untreated 2 year old rats of different strains varied between

0.4 to 22% (ECETOC, 1982).

In the rat hepatomegaly is a characteristic response to

administration of peroxisome prolif erators and is due to

hyperplasia and hypertrophy. With prolonged drug

administration continued stimulation of non-reparative cell

division could eventually lead to tumour formation. In DEHP

and methyl clofenapate induced hepatic hyperplasia, a

reduction in the frequency of binucleated cells was

demonstrated and an increase in the tetraploid: diploid ratio

was observed due to the continued conversion of newly formed

binucleates to tetraploid mononucleates (Styles et al.,1987).

In animals treated with genotoxic carcinogens there was a

permanently depressed tetraploid: diploid hepatocyte ratio

(Styles et al.,1987). The role of increased mitotic activity as a determinant of tumour development has undergone much speculation. The occurrence of somatic mutation with hyperplasia has been suggested with increased hepatic ploidy acting as a promotional event (ECETOC, 1982). However it must be emphasised that the ploidy of tumours resulting from peroxisome proliferator administration has not been determined.

7.3. TISSUE SPECIFICITY OF THE RESPONSE TO PEROXISOME

PROLIFERATORS IN THE RAT.

A considerable rat strain and species variation was demonstrated in the kidney response to ciprofibrate. These 317

enzyme changes were minimal compared to that observed in the

liver and the statistical significance could not be

determined. This suggested that the hepatic response in the

rat and other species was tissue specific, not associated with

a similar response in the kidney. With ciprofibrate this was

attributed to an inherent tissue difference in induction

responsiveness between the liver and kidney as distribution

studies revealed an identical ciprofibrate burden (ug/g

tissue) in these two organs (Davison et al.,1975). No kidney

enlargement has been observed in treated rats and similarly

renal lipid accumulation is absent or has not been assessed. A

1.5 to 3-fold increase in the 11- and 12-hydroxylation of

lauric acid has been observed in the Wistar rat on

administration of peroxisome prolif erators (Sharma et

al.,1988a). The induction was 1.5 to 2.5-fold with

ciprofibrate but this did not occur in every rat strain. A

slight increase in cytochrome P-452 content was observed. The minimal increase in cytochrome P-452 protein and catalytic

activity would indicate a slightly increased fatty acid

substrate overload. A diet supplemented with 10% lauric acid has been reported to increase the level and activity of a cytochrome P-450 isoenzyme catalysing lauric acid omega-hydroxylation (Jakobsson et al.,1970). A lipid overload could induce renal cytochrome P-452, increasing renal levels of dicarboxylic fatty acids. It should be remembered that basal levels of microsomal fatty acid oxidation are considerably higher in the kidney than in the liver.

A 1.5 to 3-fold increase in renal peroxisomal beta-oxidation 318

has been observed in treated animals (Sharma et al.,1988a). A

2-fold induction in carnitine acetyltransferase, carnitine

palmitoyltransferase, total enoyl-CoA hydratase and

peroxisomal enoyl-CoA hydratase activities has been observed

with treatment (Sharma et al.,1988a). Analysis of various

tissues of ciprofibrate treated rats revealed only subtle

increases in renal peroxisome number, peroxisomal

beta-oxidation activity and the mRNA levels of the peroxisomal

fatty acyl-CoA oxidase and enoyl-CoA hydratase (Reddy et

al.,1985). These increases in mitochondrial, microsomal and

peroxisomal fatty acid metabolism would appear to be linked to

an excess of substrate. An intermediate of lipid metabolism

such as a fatty acid or fatty acyl-CoA derivative may be

responsible for induction. An increase in hepatic and renal

long chain acyl-CoA hydrolase has been reported with

clofibrate and tiadenol administration (Katoh et al.,1987).

There is no reported occurrence of renal carcinomas in

response to peroxisome proliferators. The increase in hepatic

peroxisomal beta-oxidation on treatment has been estimated to

be up to 25 and 30-fold (Reddy and Rao,1986). Clearly the

2-fold increase in renal peroxisomal beta-oxidation is less

likely to result in oxidative stress and lipid peroxidation.

In addition, unlike the liver, renal glutathione peroxidase

activity was unchanged by treatment.

It is possible that these compounds exert their effect by a ligand-receptor mediated mechanism. Nafenopin has been found to bind to a specific saturable pool of binding sites in 319

kidney cortex cytosol and that tissue levels of this protein

correlated with the ability of nafenopin to induce renal peroxisome enzymes (Lalwani et al.,1987). A cytosolic receptor binding protein would provide a plausible explanation for tissue differences in responsiveness to peroxisome prolif erators (Rao and Reddy, 1987). As stated previously the existence of such a protein has not been verified in other laboratories and has not been accepted universally.

It is apparent that in "responsive" species a pattern of enzyme changes occurs in the liver of treated animals.

Considerable rat strain and species differences were (observed in the renal response to ciprofibrate and of particular interest were changes in the level and activity of cytochrome

P-452. In the rat the response to peroxisome prolif erators is largely tissue specific for the liver and one would assume the same for responsive species. j

7.4. MECHANISM OF INDUCTION OF TOXICITY IN DIFFERENT SPECIES

The different species investigated in my study have been classified as responsive (rat, mouse, hamster and rabbit) and non-responsive (guinea pig and marmoset). An hypothesis was presented (figure 7.1) to explain the induction of toxicity in the rat and this will now be discussed with reference to the other species examined.

Rapid absorption of ciprofibrate following administration was assumed in all species; in the marmoset peak plasma levels 320

occurred between 2 and 4 hours post dose (Sterling-Winthrop,

personal communication). Ciprofibrate (i.e. ciprofibric acid)

is the active compound therefore possible species variation in

metabolism are not relevant. Hepatomegaly was observed in the

rat, mouse and hamster. The specific content of cytochrome

P-452 was increased in the rat, mouse, hamster and rabbit but was not detected immunochemically in guinea pig or marmoset microsomes. The 11-hydroxy lation of lauric acid was slightly

increased in the mouse and hamster however omega- hydroxylase activity was markedly induced in all responsive species. As described previously, an intra-hepatic accumulation of lipid could initiate the induction of cytochrome P-452 resulting in increased levels of medium and long chain dicarboxylic acids and a substrate influx into peroxisomes. An induction of peroxisomal beta-oxidation was demonstrated in the rat, mouse, hamster and rabbit. Carnitine acetyltransferase activity was increased in all four species presumably to facilitate the transfer of fatty acyl substrates out of peroxisomes and into mitochondria. The specific activity of catalase was unchanged on treatment and glutathione peroxidase activity was decreased only in the rat and mouse.

If hepatocarcinogenesis in the rat is attributed to oxidative stress and lipid peroxidation arising indirectly from increased peroxisomal beta-oxidation, it is conceivable that species exhibiting a similar peroxisomal induction could develop liver tumours. Induction of microsomal and peroxisomal parameters in the hamster and rabbit are considerably lower than in the rat and mouse. Several peroxisome prolif erators 321

including DEHP and DEHA induce liver tumours in rats and mice

(Rao and Reddy,1987). The latency period and the incidence of

tumours appeared to correlate well with the effectiveness of

the compound to induce peroxisome proliferation. Decreased

glutathione peroxidase activity may potentiate oxidative

stress in the rat and mouse. Similarly sustained hepatomegaly

of a hyperplastic nature may cause a mitogenic burden on the

liver in these species. Two long term bioassays with DEHP were

completed in Syrian golden hamsters using i.p. and inhalation

routes of administration (Schmezer et al.,1988). There was no

treatment related increase in tumour incidence indicating that

hepatocarcinogenesis is restricted to the rat and mouse. If

the above enzyme changes are relevant to the postulated

mechanism of toxicity of peroxisome proliferation (figure

7.1), then the balance of "activation" and "detoxication"

enzymes are critically important. Thus an examination of the

activity ratio of;

Omega-hydroxylation + Peroxisomal Beta-Oxidation + Carnitine

Acetyltransferase Activities, versus

Catalase + Glutathione Peroxidase Activities

should provide an overall assessment of the validity of my

hypothesis. As shown in table 7.1, there is a clear

dose-dependent increase in this ratio in responsive species whereas the ratios are essentially unchanged in non-responsive

species. The extent of activation is highest in the rat and mouse, the two species exhibiting hepatocarcinogenesis with treatment, whereas the non-responsive species (particularly the guinea pig and marmoset) are refractory to induction to their response to ciprofibrate. This is patently apparent in 322

Table 7.1.

Species Differences in the Ratio of Activation to

Detoxification in Control and Ciprofibrate Treated Animals

Following Acute Exposure.

Species Dose Level Activation / Detoxication

Ratio

Fischer Control 3.01

Rat 2 mg/kg 43.67

20 mg/kg 60.12

Mouse Control 3.34

2 mg/kg 26.70

20 mg/kg 66.88

Rabbit Control 9.63

2 mg/kg 15.93

20 mg/kg 32.66

Hamster Control 5.21

2 mg/kg 7.75

20 mg/kg 15.29

Guinea Control 4.09

pig 2 mg/kg 3.78 20 mg/kg 4.41

Marmoset Control 15.80

20 mg/kg 17.59

100 mg/kg 19.78 323

the marmoset where the highest ciprofibrate dose was five

times that in all other species.

In the guinea pig/ no enzyme changes were detected with

ciprofibrate in my study. Similarly, nafenopin has been found

to cause only comparatively small changes in palmitoyl-CoA

oxidation and lauric acid 12-hydroxylation in the guinea pig

as determined elsewhere (Lake et al.,1987). In contrast a

4-week treatment period with tiadenol increased peroxisomal

beta-oxidation 3.4-fold (Oesch et al.,1987), indicating that

the guinea pig may be responsive to certain compounds, but as

discussed previously this may be related to the structural

analogy of this compound to fatty acids. Omega-lauric acid

hydroxylation and carnitine acetyltransf erase activities were

not determined in this latter study. In the marmoset a 4-fold

increase in peroxisomal beta-oxidation was observed with both

short term and chronic ciprofibrate administration. In these

two species (omega-1)-hydroxylation of lauric acid was more

extensive than the 12-hydroxylation of lauric acid. This

indicates that in responsive species there is a regioselective

difference in hepatic microsomal fatty acid metabolism in

comparison to the non-responsive species. In responsive

species, induction of cytochrome P-452 was observed, with it

is assumed, a subsequent increase in long chain dicarboxylic

acids initiating peroxisomal beta-oxidation and ultimately mitochondrial beta-oxidation. In the marmoset and guinea pig there would be no increase in long chain dicarboxylic acids and no substrate overload linked induction of peroxisomal and mitochondrial beta-oxidation enzymes. In the marmoset the 324

slight increase in peroxisomal beta-oxidation is not initiated

by products of microsomal oxidation of fatty acids. This

apparent anomaly in the marmoset response may be explained by

the different ability of the marmoset to handle lipids. In

addition it must be emphasised that the increase in

peroxisomal beta-oxidation in the marmoset was minimal

compared to other responsive species commensurate with

non-inhibited complement of the protective glutathione

peroxidase.

The lack of response in the guinea pig and marmoset may be due

to a dose-dependent sensitivity phenomenon, absence \ of the

putative cytosolic receptor or to a fundamental difference in

lipid metabolism between species. With reference to

peroxisomal beta-oxidation, the high dose level in the marmoset appears to correlate to the plateau on tfre dose

response curve indicating that increasing the dose level would

have little effect on this parameter (figure 5.16). It should be emphasised that this high dose level represented the maximum tolerated dose in the marmoset. Induction of peroxisomal beta-oxidation in the marmoset is not due to a dose dependent phenomenon. In the rat, sex differences in the hepatic response to peroxisome prolif erators have been observed (Reddy and Lalwani, 1983). In general the response was greater in male animals than females at the same dose level.

By increasing the dose level the difference in response between male and female animals is abolished indicating a dose-dependent induction. With ciprofibrate the extent of hepatomegaly in the Fischer rat and treatment-induced changes 325

in glutathione peroxidase and omega- and (omega-1)-hydroxylase

activities were the same in both sexes at the same dose level

(Makowska et al., unpublished observations,1987).

In the rat, most of the very low density lipoproteins and

chylomicra remnants are taken up into liver cells whilst in

man and primates these are remodelled on the cell surface and

re-exported into blood as low density lipoproteins (Cohen and

Grasso,1981). Therefore in the guinea pig and marmoset there

is no intra-hepatic accumulation of lipid associated with

treatment, in accordance with the hypothesis presented in

figure 7.1.

The lack of response in the marmoset is not due to low drug

absorption (Sterling-Winthrop, personal communication), or to

a difference in metabolic transformation as ciprofibrate is

the active compound. On a theoretical basis, drug plasma

levels and the extent of plasma protein binding at the low

dose level in the rat (2mg/kg) and high dose level in the marmoset (80-100 mg/kg) were similar and would not therefore

account for species differences in response. The lack of response in the guinea pig and marmoset would appear to be due to inherent species differences in induction probably related to species differences in hepatic lipid metabolism or genetic responsiveness. 32 6

7.5. EXTRAPOLATION OF LABORATORY ANIMAL STUDIES TO PREDICT

THE RESPONSE IN NON-HUMAN PRIMATES AND MAN.

The evaluation of the safety of chemicals in man is based

largely upon the extrapolation of animal data to predict the

likelihood of toxicity. A marked variability in response

between species may exist and this may be attributed to both

qualitative and quantitative differences. This raises

questions concerning the validity of using animal models for

the purposes of extrapolation. Detailed knowledge of the

pharmacodynamic and pharmacokinetic profiles of the compound

both in man and the comparative animal model is required.

Evidence of toxicity in the animal model may be absent in man

or alleviated by reducing both the dose and frequency of

administration. In the rat, the ultimate expression of early

toxicity is frequently carcinogenesis and the relevance of

using rodent data in interpreting the possible carcinogenic

risk to man is a major concern.

It is clear that the hepatic response to ciprofibrate in the

rat is different to the response in the marmoset. In the

previous sections the mechanisms of induction were described

in an attempt to elucidate the cause of the apparent species

specificity of ciprofibrate-induced liver changes. The differences in hepatic response between these two species cannot be attributed to the pharmacokinetic phase of drug action. The long half life of ciprofibrate in the rat has been attributed to extensive enterohepatic recirculation which has also been demonstrated with fenofibrate (Davison et al.,1975, 327

Weil et al.,1987). This may explain the sensitivity of the rat

to these compounds. In the marmoset elimination of

ciprofibrate was biphasic (t 1/2 alpha is 4.5 hours, t 1/2

beta is 8.2 hours). The half life of ciprofibrate in the

marmoset is clearly shorter than in the rat.

In the rat peroxisome prolif erators produce a predictably

similar pleiotropic response which includes hepatomegaly and

induction of cytochrome P-452, carnitine acetyltransferase and

mitochondrial and peroxisomal beta-oxidation enzymes. In

addition long chain acyl-CoA hydrolase and microsomal

1-acyl-glycerophosphorylcholine acy 1-transferase activities

were induced in clofibrate treated rats and mice but not in

treated guinea pigs (Kawashima et al.,1983 and 1984). A close

link was established between bilirubin UDP-glucuronosyl-

transferase induction and increased catalytic activity of

cytochrome P-452 (Fournel et al.,1985). Measurement of these

latter enzymes in the marmoset would be of interest. These three enzymes would appear to be under coordinate regulation with microsomal, peroxisomal and mitochondrial parameters.

The administration of peroxisome prolif erators to rats is associated with a decrease in serum triglyceride levels giving rise to the term "hypolipidaemic agent". At high dose levels a reduction in serum cholesterol has been observed for example with clofibrate, bezafibrate and ciprofibrate (Dvornik and

Cayen,1980). However hypocholesterolaemic drugs such as probucol reduce cholesterol levels but have no triglyceride- lowering effects (Cohen and Grasso,1981). The early liver 328

changes observed with peroxisome prolif erators are absent in

animals treated with hypocholesterolaemic compounds. This

would correlate with the hypothesis that intra-hepatic lipid

accumulation initiates induction. Probucol acts by enhancing

faecal excretion of bile acids while reducing cholesterol

synthesis and slightly decreasing absorption of dietary

cholesterol (Miettinen,1972). It has been speculated that

increased microsomal, peroxisomal and mitochondrial oxidation

of fatty acids may be responsible for the hypolipidaemic

action of peroxisome prolif erators however the relative

contributions have not been established. An increase in

hepatic fatty acid binding protein (hFABP) has been observed

with clofibrate treatment (Bass et al.,1985). This correlated

with an increased rate of uptake and utilisation of free fatty

acids by perfused liver following clofibrate treatment. It has

been suggested that the early hypolipidaemic effect of

clofibrate in rats and primates may be related to an enhanced

hepatic oxidation of long chain fatty acids but this was not

merely due to a reduction in esterification to complex lipids

(Capuzzi et al.,1983). Bezafibrate administration may cause

inhibition of hepatic fatty acid synthesis (Ayala et al.,1987)

and inhibition of beta-hydroxymethylglutaryl CoA reductase, the rate limiting step in the formation of cholesterol from acetate in the liver (Monk and Todd,1987). The decrease in serum triglycerides in patients during bezafibrate treatment was reported to be due to an increased activity of skeletal muscle tissue lipoprotein lipase activity (Vessby et al.,1982). The mechanisms accounting for the hypolipidaemic action need to be elucidated to provide a possible insight 329

into species differences.

In figure 7.2. a scheme is presented to explain induction of

microsomal, peroxisomal and mitochondrial enzymes in the rat

under states of altered lipid metabolism. Diabetes and

starvation cause increased lipase activity giving rise to

medium and long chain fatty acids (Mortensen et al.,1986). An

influx of fatty acid substrates into hepatocytes is predicted,

initiating cytochrome P-452 induction. Omega-oxidation of

sterate was stimulated 3 to 7-fold in starved and diabetic

rats (Bjorkhem,1973). A 2-fold increase in the level and

catalytic activity of cytochrome P-452 has been observed in

streptozotocin-diabetic rats (Sharma et al.,1988b). The

omega-hydroxylase products of cytochrome P-452 could induce

peroxisomal beta-oxidation enzymes. Increased peroxisomal

enoyl-CoA hydratase activity and peroxisomal palmitoyl-CoA

oxidation has been observed in diabetic and starved rats

(Hertz and Bar-Tanna,1987). Peroxisomal proliferator acyl-CoA derivatives, such as clofibroyl-CoA, may act as the reactive intermediate initiating induction (Berge et al.,1987, and

Caldwell, 1984). However it should be pointed out that elevation of long chain acyl-CoA seems not to be involved in the mechanism of induction of the beta-oxidation enzymes since induction of mRNA for enoyl-CoA hydratase : beta-hydroxyacyl-CoA dehydrogenase preceded the rise in long chain acyl-CoA (Fukami et al.,1986).

DNA-damage may arise from acylglucuronide metabolites of peroxisome prolif erators such as clofibrate (Faed and 330

Figure 7.2.

Induction of Hepatic Fatty Acid Oxidising Systems In States of

Altered Lipid Metabolism.

Peroxisome High Fat Diets Diabetes and Proliferators (Long Chain Fatty Starvation Acids) I Increased I Lipolysis Increased Medium and Long Chain Fatty Acids

Increased CytochromeI P-452 Omega-/(omega-1)-hydroxy Medium and Long Chain Monocarboxylic- Fatty Acids I (Cytosolic Oxidation)

Dicarboxylic Medium and Long Chain Fatty1 Acids Peroxisome

Peroxisome Proliferation and Induction1 of Enzymes Chain Shortened1 Fatty Acids Mitochondria1 Further Beta-OxidationI Excretion of Short Chain Fatty Acids 331

McQueen,1978). Acyl or ester glucuronides are distinguished

from the more common "ether-type" by their susceptibility to nucleophile substitution which is due to the presence of the unsaturated carbonyl group in the molecule (Faed,1984). Acyl migration may take place resulting in intra-molecular rearrangement of the conjugate with covalent binding capacity

(Wells et al.,1987). The 1-0-acyl-glucuronide of clofibrate is formed in vivo in the rat, mouse, guinea pig, rabbit and man

(Caldwell and Emudianughe,1979, Caldwell et al.,1979, Sinclair and Caldwell,1982). This metabolite was shown to be an active chemical electrophile acylating albumin in vitro via trans-esterification and may represent a molecular mechanism of toxicity (Van Breeman and Fenselau, 1985). An acyl-linked mercapturic acid metabolite of clofibrate has been identified in human urine and the electrophilic properties may result in covalent interactions (Stogniew and Fenselau,1982). Aldehydes, epoxides and other lipid peroxidation products may be implicated in the development of hepatocarcinogenesis.

Peroxisomal aldehyde dehydrogenase may function as a detoxification mechanism metabolising aldehydes into carboxylic acids (Turteltaub and Murphy,1987).

Tumour promoters such as the peroxisome prolif erators, may act directly on cell membranes. The phorbol ester, 12-0-tetra- decanoyl-phorbol-13-acetate (TPA), in rat cell culture, increased peroxisomal beta-oxidation, carnitine acetyltransf erase, palmitoyl-CoA hydrolase and catalase activities (Lillehaug and Berge,1986). Phorbol esters interact with a high affinity saturable membrane receptor which appears 332 to be a calcium and lipid binding protein kinase C.

Phosphorylation of proteins by this kinase may modulate many cell regulatory functions including gene activation (Kensler and Trush,1984). An excellent correlation between tumour promoting activity of various compounds and their ability to induce ornithine decarboxylase (ODC) has been reported

(O'Brien et al.,1975). The induction of this enzyme by peroxisome prolif erators does not appear to be a specific event associated with the development of preneoplastic foci and hepatocarcinogenesis but with sustained maintenance of hepatomegaly and increased liver cell proliferation (Izumi et al.,1981). ODC induction was not accompanied by an elevation in cAMP or increased activation of cytoplasmic cAMP-dependent protein kinases [types I and II] (Fukami et al.,1986).

The carcinogenic process in laboratory animals can be modified by genetics, age, sex and endocrine status and extrinsic factors such as mode and regimen of exposure, housing, nutrition and xenobiotic administration (Williams,1984). A high fat and low selenium diet has been shown to potentiate the initiation and promotion of aflatoxin B[l]-induced preneoplastic foci in rat liver (Baldwin and Parker, 1987).

Hepatic glutathione peroxidase activity was decreased in patients with hepatocellular carcinomas (Corrocher et al.,1986). Chemical modulation of this enzyme may increase the carcinogenic risk in laboratory animals and man. However in neoplastic human adult lung tissues, glutathione peroxidase was unchanged or increased (Di Ilio et al.,1987). 333

Treatment-induced changes may arise from modulation of growth factors/hormones. It has been proposed that TPA induces synthesis of a protein[s] similar in function to growth factors (Levine,1982). This protein reacts with a cell membrane receptor leading to a growth factor stimulated arachidonic acid metabolic cascade causally related to promotion. Arachidonic acid transformation may mediate tumour promotion by altering membrane fluidity (Levine, 1982). In mouse liver, the expression and repression of sex-dependent testosterone hydroxylases are regulated pretranslationally by growth hormone (Noshiro and Negishi, 1986). The processes by which serum growth hormone mediates transcriptional regulation of hydroxylase genes through the cell membrane are unclear.

Steroids can serve as highly specific substrates for hepatic cytochrome P-450 enzymes and as hormone regulators of isoenzyme expression (Waxman,1988). Growth hormone may act directly to activate cytochrome P-450 gene expression or indirectly, mediated by other endocrine factors such as somatomedians. The hyperplastic effect induced by peroxisome prolif erators while transient in the acute phase, may after chronic administration, induce constitutive cell cycling in some mononucleated cells by changing the expression of genes controlling differentiation in particular, those responsible for receptors to growth factors such as epidermal growth factor (Styles et al.,1987).

Hepatocarcinogenesis in laboratory animals may arise from activation of cellular oncogenes. An active cellular oncogene 334

was demonstrated in hepatocellular neoplasms arising

spontaneously in untreated 24-month old B6C3F1 mice (Fox and

Watanabe,1985). Preliminary investigations with nafenopin

indicated that treatment was associated with prolonged

activation of cellular oncogenes (Bentley et al.,1987). A

cellular gene termed "hap" has been isolated from human

hepatocellular carcinoma and the sequence was similar to DNA

binding hormone receptors (De The et al.,1987). This "hap"

gene product may be a novel ligand-responsive regulatory

protein whose inappropriate expression in liver may relate to

hepatocarcinogenesis.

The hepatic response to ciprofibrate in the rat is

characteristic of all peroxisome prolif erators. This suggests

that there is a set of gene products or a "peroxisome-

proliferator domain" which is under coordinate regulation.

Increased catalytic activity correlates to transcriptional

activation of the genes. The species difference in response

probably reflects a different mechanism of induction and gene

regulation. Gene dosage may be a critical factor.

The occurrence of peroxisome proliferation in non-human

primates and man remains questionable. This study has shown

the guinea pig and marmoset to be non-responsive. Peroxisome proliferation and induction of peroxisomal beta-oxidation enzymes does not occur with clofibrate in human hepatoma cell lines PLC/PRF/5 and SK-HEP-1 (Hertz et al.,1987). The assessment of chemical safety in man depends often upon laboratory animal studies. An added complication in human risk 335

prediction is the use of excessively high doses in animal

studies. This may affect the disposition of the compound as

metabolic pathways can be dose-related or saturable.

Understanding the potential role that cellular oncogenes may

play in tumour development in animals used for long term

studies is important for proper interpretation of data and

assessment of potential risk. In cases where there is evidence

that the test compound is acting by non-genotoxic mechanisms

to enhance liver tumour incidence under conditions of high

exposure it seems likely that low exposure to humans does not

represent a serious risk (ECETOC,1982).

The research presented in this thesis has shown that marked

species differences in hepatic responses to ciprofibrate

exist. It is equally clear that with ciprofibrate,

extrapolation from rat studies to predict the response in the

marmoset is not scientifically accurate as the response in the

rat is dissimilar to that in the marmoset. The response

between the marmoset and man is assumed to be more similar

than the relationship with rodents, with the inference that in man peroxisome prolif erators do not pose as a toxicological hazard.

SUGGESTIONS FOR FURTHER WORK

In the last few years considerable progress has been made in the study of peroxisome prolif erators. The pleiotropic response to administration appears to be tissue specific for the liver and induction is not associated with all species. 336

Various mechanisms of induction (enzyme changes and

hepatocarcinogenesis) have been proposed but further

supportive experimental evidence is required.

The mechanism(s) of induced changes must be established in

order to rationalise species differences in response. The chronological sequence of induction in different subcellular

fractions must be ascertained and similarly the mechanism leading to the observed hypolipidaemic effect should be clarified. A greater understanding of gene structure and gene expression has been achieved recently with insight into gene regulation. It is anticipated that progress will continue and that techniques of molecular biology will be increasingly used in toxicology. At present cDNA probes to cytochrome P-452, glutathione peroxidase, fatty acyl-CoA oxidase and the peroxisomal bifunctional protein, are available in our laboratory and these should be used to study gene expression.

The influence of peroxisome prolif erators on mRNA accumulation, transcriptional activation and DNA binding proteins of the genes coding for the above enzymes should be determined. The temporal aspects of gene expression in responsive species and tissues must be assessed. The ultimate goal of research into these compounds must be to delineate which changes in gene expression are a prerequiste for the final manifestation of toxicity.

A greater understanding of genetic regulation upon xenobiotic pretreatment would be attained and in addition recombinant DNA technology could play a role in clinical / toxicological 337 diagnosis. Isolation and characterisation of human genomic cytochrome P-452, glutathione peroxidase and peroxisomal enzyme DNA will allow for more accurate risk assessment of these compounds in man.

f 338 REFERENCES

Aarsland,A-, Stensland,E., Berge,R.K., Aarsaether,N. and Bremer,J., Poster Abst., 28th Congress of European Society of Toxicology, Strasbourg, (1987).

Adesnik,M. and Atchison,M., C.R.C. Critical Reviews in Biochemistry, 19, pp. 247-305, (1986).

Affara,N., Fleming,J., Goldfarb,P.S., Black,E., Thiele, B. and Harrison,P.R., Nucleic Acids Research, 13r pp. 5629-5644, (1985).

Alberts,B., Bray,D., Lewis,J., Raff,M., Roberts, K. and Watson,J.D., In "Molecular Biology of the Cell", Garland Publishing, (1983).

Albro,P.W., Chae,K., Philpot,R., Corbett,J.T., Schroeder,J. and Jordan,S., Drug Metab. Dispos., 1_2, pp. 742-748, (1984).

Alexson,S.E.H., Fujiki,Y., Shio,H. and Lazarow,P.B., J. Cell Biol., 101, pp. 294-305, (1985).

Angermuller,S. and Fahimi,H.D., J. Histochem. Cytochem., 34, pp. 159-165, (1986).

Antonenkov,V.D., Pirozhkov,S.V. and Panchenko,L.F., Biochem., 48, pp. 1429-1436, (1983).

Antonenkov,V.D., Gusev,V.A. and Panchenko,L.F., Int. J. Biochem., 19, pp. 187-192, (1987).

Armato,U., Ranano,F. and Andreis,P., Prostaglandins Leukotrienes and Medicine, JL3, pp. 237-247, (1984).

Arnold,A., McAuliff,J.P., Powers,L.G., Phillips,D.K. and Beyler,A.L., Athero., 32, pp. 155-163, (1979).

Awasthi,Y.C., Beutler, E. and Srivastava,S.K., J. Biol. Chem., 250, pp. 5144-5149, (1975).

Awasthi,Y.C., Singh,S.V., Goel,S.K. and Reddy,J.K., Biochem. Biophys. Res. Commun., 123, pp. 1012-1018, (1984).

Ayala,A. and Machado,A., J. Pharm. Pharmacol., 39, pp. 551-552, (1987).

Azamoff,D.L., Tucker,D.R. and Barr,G.A., Metab., 14, pp. 959-965, (1965).

Bains,S.K., Gardiner,S.M., Mannweiler,K., Gillet,D. and Gibson,G.G., Biochem. Pharmacol., 3_4, pp. 3221-3229, (1985).

Bakke,0.M. and Berge,R.K., Biochem. Pharmacol., 33, pp. 3077-3080, (1984).

Baldwin,J.R., Witiak,D.T. and Feller,D.R., Biochem. Pharmacol., 29, pp. 3143-3154, (1980). 339 Bald win,S. and Parker, R.S., Carcin., 8, pp. 101-107, (1987).

Balias,L.M., Lazarow,P.B. and Bell,R.M., Biochim. Biophys. Acta, 195, pp. 297-300, (1984).

Bass,N.M., Manning, J. A., Ockner,R.K., Gordon,J.I., Seetharam,S. and Alpers,D.H., J. Biol. Chem., 260, 1432-1436, (1985).

Beaufay,H., Bendall,D.S., Baudhuin,P., Wattiaux,R. and De Duve,C., Biochem. J., 7_3, pp. 628-637, (1959).

Beckett,R.B., Weiss,R., Stitzel,R.E. and Cenedella,R.J., Toxicol. Appl. Pharmacol., 13, PP* 42-53, (1972).

Bentley,P., Bieri,F., Kelly,S., Staubli,W. and Waechter,F., Poster Abst., 28th Congress of European Society of Toxicology, Strasbourg, (1987).

Berenblum,I., In "Carcinogenesis as a Biological Problem"., (eds. Neuberger,A. and Tatum,E.L.) North Holland Publishing Company, J4, pp. 67-167, (1974).

Berge,R.K., Flatmark,T. and Osmundsen,H., Eur. J. Biochem., 141, pp. 637-644, (1984).

Berge,R.K., Hosoy,L.H., Aarsland,A., Bakke,0.M. and Farstad,M., Toxicol. Appl. Pharmacol., 73, pp. 35-41, (1984)a.

Berge,R.K. and Aarsland,A., Biochim. Biophys. Acta., 837, pp. 141-151, (1985).

Berge,R.K., Aarsland,A., Aarsaether,N. and Bremer,J., Poster Abst., 28th Congress of European Society of Toxicology, Strasbourg, (1987).

Bieber,L.L., Abraham,T. and Helmrath,T., Anal. Biochem., 50, 509-518, (1972).

Bieri,F., Bentley,P., Waechter,F. and Staubli,W., Carcin., 5, pp. 1033-1039, (1984).

Bjorkhem,I. and Danielsson,H., Eur. J. Biochem., 17, pp. 450-459, (1970).

Bjorkhem,I., Eur. J. Biochem., 40, pp. 415-422, (1973).

Bjorkhem,I., J. Biol. Chem., 251, 5259-5266, (1976).

Bjorkhem,I., J. Lipid Res., _19, 585-590, (1978).

Blumcke,S., Schwartzkopff,W., Lobeck,H., Edmondson, N.A., Prentice,D.E. and Blane,G.F., Athero., 46, pp. 105-116, (1983).

Bock,PI’., Kramar,R. and Pavelka,M., Cell Biol. Monographs, 7, Springer-Verlag, pp. 44-74, (1980). 340 Bosterling,B. and Trudell,J.R., Biochem. Biophys. Res. Commun., 114, pp. 850-854, (1983).

Bremer,J. and Osmundsen,H., In "Fatty acid Metabolism and its Regulation" (ed. Numa,S.), Chapter 5, Elsevier Press, (1984).

Brcnfman,M., Inestroea,N.C. and Leighton,F., Biochem. Biophys. Res. Commun., £8,pp. 1030-1036, (1979).

BurcheIl,B. and Blanckaert,N., Biochem. J., 223, pp. 461-465, (1984).

Burke,M., Prough,R. and Mayer,R., Drug Meta. Dispos., 5, pp. 1-8, (1977).

Caldwell,J. and Emudianughe,T.S., Biochem. Soc. Trans., 7, pp. 521, (1979).

Caldwell,J., Emudianughe,T.S. and Smith,R.L., Proc. Brit. Pharmacol. Soc., pp. 421P, (1979)a.

Caldwell,J., Drug Meta. Revs., 12, pp. 221-237, (1981).

Caldwell,J., Drug Meta. Revs., 13, pp. 745-777, (1982).

Caldwell,J., Biochem. Soc. Trans., _12, pp.9, (1984).

Caldwell,J., Anthony,A., Cotgreave,I.A., Sangster,S.A., Sutton,J.D., Bernard,B.K. and Ford,R.A., Fd. Cosmet. Toxicol., 23, pp. 559-566, (1985).

Caldwell,J., Viswalingam,A., Gibson,G.G. and Fournel,S., Abst., Clearwater Symposium, Florida, (1987).

Capdevila,J., Chacos,N., Werringloer,J., Prough,R.A. and Estabrook,R.W., Proc. Natl. Acad. Sci. (U.S.A.), 78, pp. 5362-5366, (1981).

Capdevila,J., Marnett,L.J., Chacos,N., Prough,R.A. and Estabrook,R.W., Proc. Natl. Acad. Sci. (U.S.A.), 79, pp. 767-770, (1982).

Capdevila,J., Kim,Y.R., Martin-Wixtrom,C., Falck,J.R., Manna,S. and Estabrock,R.W., Arch. Biochem. Biophys., 243, pp. 8-19, (1985).

Capuzzi,D.M., Intenzo,C.M., Lackman,R.D., Whereat,A.F. and Scott,D.M., Biochem. Pharmacol., 32, pp. 2195-2203, (1983).

Cayen,M.N., Ferdinandi,E.S., Greselin,E., Robinson,W.T. and Dvomik,D., J. Pharmacol. Exp. Ther., 200, pp. 33-43, (1977).

Cayen,M.N., Pharmac. Ther., 29, pp. 157-204, (1985).

Cederbaum,A.I., Madhavan,T.V. and Rubin,E., Biochem. Pharmacol., 25, pp. 1285-1292, (1976). 341 Chacos,N., Falck,J.R., Wixtrom,C. and Capdevila,J., Biochem. Biophys. Res. Commun., 104/ pp. 916-922, (1982).

Chacos,N., Capdevilla,J., Falck,J.R. and Jacobson,H., Fed. Proc., 42, pp. 1778 (Abst.) 127, (1983).

Chacos,N., Capdevila, J., Falck,J.R., Manna,S., Martin-WixtronvC., GiU^S.S., Hammock,B.D. and Estabrook,R.W., Arch. Biochem. Biophys., 223, pp. 639-648, (1983)a.

Challiner,M.R., Park,B.K., Odum,J., Orton,T.C. and Parker,G.L., Biochem. Pharmacol., 29, pp. 3319-3324, (1980).

Chambers,I., Frampton,J., Goldfarb,P., Affara,N., McBain,W. and Harrison,P., EMBO Journal, 5, pp. 1221-1227, (1986).

Chance,B. and Maehley,A.C., Methods In Enzymology (Eds. Colowick,S.P. and Kaplan,N.O.), Academic Press, 2, pp. 764-775, (1955).

Chatterjee,B., Demyan,W.F., Lalwani,N.D., Reddy,J.K. and Roy,A.K., Biochem. J., 214, pp. 879-883, (1983).

Chatterjee,B., Murty,C.V., Olson,M. and Roy,A., Eur. J. Biochem., 166, pp. 273-278, (1987).

Chirgvdn,J.M., Przybyla,A.E., MacDonald,R.J. and Rutter,W.J., Biochem., 1_8, pp. 5294-5299, (1979).

Christen,P., Peacock,C., Christen,A. and Wacker,W., Eur. J. Biochem., 12, pp. 3-5, (1970).

Christiansen,R.Z., Osmundsen,H., Borrebaek,B. and Bremer,J., Lipids, JL3, pp. 487-491, (1978).

Ciriolo,M.R., Mavelli,I., Rotilio,G., Borzatta,V., Cristofari,M. and Stanzani,L., Febs Lett., 144, pp. 264-268, (1982).

Ciriolo,M.R., Rossi,L., Mavelli,I.,Rotilio,G.,Borzatta,V., Cristofari,M. and Barbanti,M., Drug Res., 34, pp. 465-467, (1984).

Cohen,A.J. and Grasso,P., Fd. Cosmet. Toxicol., JL9, pp. 585-605, (1981).

Conney,A.H., Pharmacol. Rev., 19, pp. 317-366, (1967).

Coronary Drug Project Research Group, J. Am. Med. Assn., 231, pp. 360-381, (1975).

Corrocher,R., Casaril,M., Bellisola,G., Gabrielli,G., Hu]pe,M., Garofoli,E. and Nicoli,N., Tumori, 72, pp. 617-619, (1986).

Cox,R.A., Methods Enzymology, 12B, pp. 120-129, (1968). 342 Crane, DJ. and Masters,C.J., Biochim. Biophys. Acta., 876, pp. 256-263, (1986).

Dalton,C., Hope,W.C., Hope,H.R. and Sheppard,H., Biochem. Pharmacol., 2_3, pp. 685-696, (1974).

Davison,C., Benziger,D., Fritz,A. and Edelson,J., Drug Metab. Dispos., 3, pp. 520-524, (1975).

De Duve,C. and Baudhuin,P., Physiol. Rev., 46, pp. 323-357, ( 1966).

De Duve,C., Proc. Royal Soc. B., 173, pp. 71-83, (1969).

De Duve,C., Ann. N. Y. Acad. Sci., 168, pp. 369-381, (1970).

De La Iglesia,F.A., Pinn,S.M., Lucas,J. and McGuire,E.J., Micron., 12, pp. 97-98, (1981).

De La Ig]esia,FJL, Lewis,J.E., Buchanan,R.A., Marcus,E.L. and McMahon,G., Athero., 4_3, pp. 19-37, (1982).

De The,H., Marchio,A., Tiollais,P. and Dejean,A., Nature, 330, pp. 667-670, (1987).

Di Ilio,C., Del Boccio,G., Casaccia,R., Aceto,A., Di Giacomo,F. and Federici,G., Carcin., 8, pp. 281-284, (1987).

Ding,V.and Pickett,C., Arch. Biochem. Biophys., 240, pp. 553-559, (1985).

Dommes,V., Baumgart,C. and Kunau,W.H., J. Biol. Chem., 256, pp. 8259-8262, (1981).

Dvornik,D. and Cayen,M.N., In "Drugs Affecting Lipid Metabolism", (Eds. Fumagalli,R., Kritchevsky,D. and Paoletti,R.), Elsevier/ North Holland Press, pp. 263-272, (1980).

Eacho,P.I., Foxworthy,P.S., Johnson,W.D., Hoover,D.M. and White,S.L., Toxicol. Appl. Pharmacol., 83, pp. 430-437, (1986).

Earnshaw,D., Dale,J. and Gibson,G.G., Unpublished observations, ( 1988).

ECETOC, Eur. Chem. Ind. Ecol. Toxicol. Centre, Monograph 4, (1982)..

Edelscn,J., Benziger,D.P., Arnold,A. and Beyler,A.L., Athero., 33, pp. 351-357, (1979).

Elcombe,C.R. and Mitchell,A.M., Enviro. Hlth. Pers., 70, pp. 211-219, (1986).

Ellin,A., Jakobsson,S.V., Schenkman,J.B. and Orrenius,S., Arch. Biochem. Biophys., 150, 64-71, (1972). 343 Ellin.A., Orrenius,S., Pilotti,A. and Swahn,C.G., Arch. Biochem. Biophys., 158, pp. 597-604, (1973).

Ellin,A. and Orrenius,S., Molec. Cell. Biochem., 8, pp. 69-79, (1975).

Elliot,B.M., Dodd,N.J.F. and Elcombe,C.R., Carcin., 7, pp. 795-799, (1986).

Elliot,B.M. and Elcombe,C.R., Carcin., 8, pp. 1213-1218, (1987).

Emudianughe,T.S., Caldwell,J., Sinclair,K.A. and Smith,R.L., Drug Meta. Dispos., 11, pp. 97-102, (1983).

Facino,R.M., Carini,M., Bertuletti,R. and Tofanetti,0., Pharmacol. Res. Commun., 13, pp. 121-131, (1981).

Facino,R.M., Carini,M. and Tofanetti,0., Drug Res., 36, pp. 722-728, (1986).

Faed,E.M. and McQueen,E.G., Clin. Exper. Pharmacol. Physiol., 5, pp. 195-198, (1978).

Faed,E.M., Drug Meta. Revs., 15, pp. 1213-1249, (1984).

Fahimi,D.H., Reinicke,A., Sujatta,M., Yokota,S. and Ozel,M., Ann. N. Y. Acad. Sci., 386, pp. 111-135, (1982).

Fahl,W.E., Lalwani,N.D., Watanabe,T., Goel,S.K. and Reddy,J.K., Proc. Natl. Acad. Sci. (U.S.A.), 8.1, pp. 7827-7830, (1984).

Falck,J.R., Siddhanta,A.K., Estabrook,R.W. and Capdevila,J., Tetr. Letts., 25, pp. 1457-1458, (1984).

Falck,J.R., Schueler,V.J., Jacobson,H.R., Siddhanta,A.K., Pramanik,B. and Capdevila,J., J. Lipid Res., 28, pp. 840-846, (1987).

Fitzgerald,J.E., Sanyer,J.L., Schardein,J.L., Lake,R.S., McGuire,E.J. and De La Iglesia,F.A., J. Natl. Cancer Inst., 67, pp. 1105-1116, (1981).

Foldes,R., Hines,R., Ho,K-L., Shen,M-L., Nagel,K. and Bresnick,E., Arch. Biochem. Biophys., 239, pp. 137-146, (1985).

Foil lot,A., Tcuchard,D. and Celier,C., Biochem. Pharmacol., 33, pp. 2829-2834, (1984).

Fallot,A., Tcuchard,D. and Mallet,L., Biochem. Pharmacol., 35, pp. 1685-1690, (1986).

Fournel,S., Magdalou,J., Thomassin,J., Villoutreix,J., Siest,G., CaldwellyJ. and Andre,J-C., Biochim. Biophys. Acta., 842, pp. 202-213, (1985). 344 Foumel,S., CaldwellyJ., Magdalou,J. and Siest,G., Biochim. Biophys. Acta., 882, pp. 469-472, ( 1986).

Fox,T.R. and Watanabe,P.G., Science, 228, pp. 596-597, (1985).

Frick,M.H., Elo,0., Haapa,K., Heinonen,O.P., Hein salmi, P., Helo,P., Huttunen,J.K., Kaitaniemi,P., Koskinen,P., Manninen,V., Maenpaa,H., Malkonen,M., Manttari,M., Norola,S., Pasternack,A., Pikkarainen,J., Romo,M., Sjoblom,T. and Nikkila,E.A., New Eng. J. Med., 317, pp. 1237-1245, (1987).

Fringes,B. and Reith,A., Lab. Invest., 47, pp. 19-26, (1982).

FujikiyY., Fowler,S., Shio,H., Hubbard,A.L. and Lazarow,P.B., J. Cell Biol., 93, pp. 103-110, (1982).

Fujiki,Y., RachubinskiyR.A. and Lazarow,P.B., Proc. Natl. Acad. Sci. (U.S.A.), 81, pp. 7127-7131, (1984).

Fujiki,Y. and Lazarow,P.B., J. Biol. Chem., 260, pp. 5603-5609, (1985).

FujikiyY., RachubinskiyR.A., Mortensen,R.M. and Lazarow,P.B., Biochem. J., 226, pp. 697-704, (1985).

Fukami,M.H., Eikhom,T.S., Ekanger,R., Flatmark,T. and Nilsson,A., Carcin., 7, pp. 1441-1446, (1986).

Furukawa,K.,Numoto,S., Furuya.K., Furukawa,N.T. and Williams,G.M., Cancer Res., 45, pp. 5011-5019, (1985).

Furuta,S., Hashimoto,T., Miura,S., MoriyM. and Tatibana,M., Biochem. Biophys. Res. Commun., 105, pp. 639-646, (1982).

Gariot,P., Barrat,E., Mejean,L., Pointel,J., Drouin,P. and Debry,G., Arch. Toxicol., 53, pp. 151-163, (1983).

Gariot,P., Pointel,J.P., Barrat,E., Drouin,D. and Debry,G., Biomed. Pharmacother., 38, pp. 101-106, (1984).

Garvey,L.K., Swenberg,J.A., Hamm,T.E. and Popp,J.A., Carcin., 8, pp. 285-290, (1987).

Gear,A.R.L., Albert,A.D. and Bednarek,J.M., J. Biol. Chem., 249, pp. 6495-6504, (1974).

Ghis,M.P., Verrando,P., Negrel,R. and Ailhaud,G., J. Pharmacol., JU5, pp. 17-25, (1984).

Gibson,G.G., Orton,T.C. and Tamburini,P.P., Biochem. J., 203, pp. 161-168, (1982).

Gibson,G.G. and Sharma,R., In "Cells, Membranes, and Disease, Including Renal (Ed. Reid,E., Luzio,J.P. and Cook,G.M.W.), Plenum Publishing Coroporation, pp. 79-88, (1987).

Glauert,H.P., Reddy,J.K., Kennan,W.S., Sattler,G.L., Rao,V.S. and Pitot,H.C., Cancer Letts., 24, pp. 147-156, (1984). 345 Goel,S.K., Lalwani,N.D., Fahl,W.E. and Reddy,J.K., Toxicol. Letts., 24, pp. 37-43, (1985).

Goel,S.K., Lalwani,N.D. and Reddy,J.K., Cancer Res., 46, 1324-1330, (1986).

Goldberg,D.A., Proc. Natl. Acad. Sci. (U.S.A.)., 77, pp. 5794-5798, (1980).

Goldenberg,H., Huttinger,M., Kampfer,P., Kramar,R. and Pavelka,M., Histochem., 4_6, pp. 189-196, (1976).

Goldfarb,P.S., 0'Prey,J., Affara,N., Yang,Q-S. and Harrison,P.R., Nucleic Acids Research, 11, pp. 3517-3530, (1983).

Galdfarb,P.S., MaBain,W., Chambers,I. and Harrison,P., Eur. J. Cell Biol., 36, pp. Abst. 33, (1985).

Goldman,B.M. and Blobel,G., Proc. Natl. Acad. Sci., 75* pp. 5066-5070, (1978).

Gonder,J.C., Proctor,R.A. and Will,J.A., Proc. Natl. Acad. Sci. (U.S.A.), 82, pp. 6315-6319, (1985).

Gonzalez,F.J., Tukey,R.H. and Nebert,D.W., Molec. Pharmacol., 26, pp. 117-121, (1984).

Gray,R.H. and De La Iglesia,F.A., Hepatology, 4, pp. 520-530, (1984).

Gray,T.J.B., Beamand,J.A., Lake, B.G., Foster,J.R. and Gangolli,S.D., Tox. Letts., 1JD, pp. 273-279, (1982).

Gray,T.J.B., Lake,B.G., Beamand,J.A., Foster,J.R. and Gangolli,S.D., Toxicol. Appl. Pharmacol., 67, pp. 15-25, (1983).

Guengerich,F.P., Prog. Drug Meta., JL0, pp. 1-55, (1987).

Gupta,R.C., Goel,S.D., Earley,K., Singh,B. and Reddy,J.K., Carcin., 6, pp. 933-936, (1985).

HalliwellyB. and Gutteridge,J.M.C., Free Radicals in Biology and Medicine, Clarendon Press, Oxford, (1985).

Halvorsen,0., Biochem. Pharmacol., 32, pp. 1126-1128, (1983).

Hames,B.D. and Higgins,S.J., Nucleic Acid Hybridisation. A Practical Approach., IRL Press, 1985.

Hanefeld,M., Kemmer,C., Leonhardt,W., Kunze,K.D., Jaross,W. and Haller.H., Athero., 36, pp. 159-172, (1980).

Hanefeld.,M., Kemmer,C. and Kadner,E., Athero., 46, pp. 239-246, (1983). 346 Hardwick,J.P., Song,B-J., Huberman,E. and Gonzalez,F.J., J. Biol. Chem., 262, pp. 801-810, (1987).

Harland,W.A. and Orr,J.S., In "Thyroid Hormone Metabolism" (Ed. Harland,W.A. and Orr,J.S.), Academic Press, Chapter 5, (1975).

Harrison,E.H., Biochim. Biophys. Acta., 796, pp. 163-168, (1984).

Hartig,F., Stegmeier,K., Heboid,G., Ozel,M. and Fahimi,H.D., Ann. N. Y. Acad. Sci., 386, pp. 464-467, (1982).

Hashimoto,T., Ann. N. Y. Acad. Sci., 386, pp. 5-12, (1982).

Hayashi,H., Hashimoto,F. and Nakata,K., Chem. Pharm. Bull., 14, pp. 3964-3967, (1986).

Heimberg,M., 01ubadewo,J.0. and Wilcox,H.G., Endocrine Rev., 6, pp. 590-607, (1985).

Henry,E.W. and DeMorrow,J.M., Cytobios, pp. 279-293, (1985).

Hertz,R., Amon,J., Hoter,A., Shouval,D. and Bar-Tana,J., Cancer Letts, 34, pp. 263-272, (1987).

Hertz,R. and Bar-Tana,J., Biochim. Biophys. Acta., 917, pp. 57-61, (1987).

Hess,R., Staubli,W. and Riess,W., Nature, 208, pp. 856-858, (1965).

Heuman,D.M., Gallagher,E.J., Barwick,J.L., Elshourbagy,N.A. and Guzelian,P.S., Mol. Pharmacol., 21, pp. 753-760, (1982).

Hietanen,E., Ahotupa,M., Heinonen,T., Hamalainen,H., Kunnas,T., Linnainmaa,K., Mantyla,E. and Vainio,H., Toxicol., 34, pp. 103-111, (1985).

Hinton,R.H., Mitchell,F.E., Mann,A., Chescoe,D., Price,S.C., Nunn,A., Grasso,P. and Bridges,J.W., Envim. Hlth. Pers., 70, pp. 195-210, (1986).

Holloway,B.R. and Thorp,J.M., Toxicol. Appl. Pharmacol., 48, pp. A125, Abst.250, (1979).

Holloway,B.R. and Orton,T.C., Br. J. Pharmacol., 69, pp. 318P, (1980).

Hruban,Z., Swift,H. and Slesers,A., Lab. Invest., JL5, pp. 1884-1901, (1966).

Ichihara,K., Kusunose,E. and Kusunose,M., Biochim. Biophys. Acta., 176, pp. 704-712, (1969).

Ide/F. and Sugano,M., Biochem. Pharmacol., 32, pp. 2583-2588, (1983). 347 Imacka/3. and Funae,Y., Biochem. Biophys. Res. Commun., 141, pp. 711-717, (1986).

Inestrosa,N.C., B:rcnfman,M. and Leighton,F., Biochem. J., 182, pp. 779-788, (1979).

IshiiyH. and Suga,T., Biochem. Pharmacol., 28, pp. 2829-2833, (1979).

Ishii,H., FukumcOyN., Horie,S. and Suga,T., Biochim. Biophys. Acta., 617, pp. 1-11, (1980).

Izumi,K., Reddy,J.K. and Oyasu,R., Carcin., 2, pp. 623-627, (1981).

Jackh,R., Rhodes,C., Grasso,P. and Carter,J.T., Fd. Cosmet. Toxicol., 2 2, pp. 151-155, (1984).

Jakcfosscn,S., Thor,H. and Orrenius,S., Biochem. Biophys. Res. Commun., 39, pp. 1073-1080, (1970).

Jones,D.P., Eklow,L., Thor,H. and Orrenius,S., Arch. Biochem. Biophys., 210, pp. 505-516, (1981).

Jones,G.L. and Neill,A.R., Biochim. Biophys. Acta., 712, pp. 420-426, (1982).

Just,W.W., and Hartl,F.U., Hoppe-Seyler's Z. Physiol. Chem. Bd., 364, pp. 1541-1547, (1983).

Kahcnen,M.T. and YHkahri,R.H., Febs Letts., 43, pp. 297-299, (1974).

Kahonen,M.T. and Ylikahri,R.H., Athero., 32, pp. 47-56, (1979).

Karki,T., Hakkola,E., Hassinen,I.E. and Hiltunen,J.K., Febs Letts., 215, pp. 228-232, (1987).

Katoh,H., Nakajima,S., Kawashima,Y., Kozuka,H. and Uchiyama,M., Biochem. Pharmacol., 33, pp. 1081-1085, (1984).

Katoh,H., Kawashima,Y., WatanukiyH., Kozuka,H. and Isono,H., Biochim. Biophys. Acta., 920, 171-179, (1987).

Kawashima,Y., Katoh,H. and Kozuka,H., Biochim. Biophys. Acta., 712, pp. 48-56, (1982).

Kawashima,Y., Katoh,H., Nakajima,N. and Kozuka,H., Biochim. Biophys. Acta., 752, pp. 182-185, (1983).

Kawashima,Y., Hirose,A. and Kozuka,H., Biochim. Biophys. Acta., 793, pp. 232-237, (1984).

Kawashima,Y., Katoh,H., Nakajima,S., Kozuka,H. and Uchiyama,M., Biochem. Pharmacol., 33, pp. 241-245, (1984)a. 348 Keller,G. A., Barton,M.C., Shapiro,D.J. and Singer,S.J., Proc. Natl. Acad. Sci. (U.S.A.), 82, pp. 770-774, (1985).

Kelly,R.B., Cozzarelli,N.R., Deutscher,M.P., Lehman,I.R. and Kornberg,A., J. Biol. Chem., 245, pp. 39-45, (1970).

Kensler,T.W. and Trush,M.A., Enviro. Mutagen., 6, pp. 593-616, (1984).

Kondrup,J and Lazarow,P.B., Biochim. Biophys. Acta., 835, pp. 147-153, (1985).

Krahling,J.B. and Tolbert,N.E., Ann. N. Y. Acad. Sci., 386, pp. 433-435, (1982).

Krajl,M., Biochem Pharmacol., 14, pp. 1683-1685, (1965).

Rrisans,S.K., Mortensen,R.M. and Lazarow,P.B., J. Biol. Chem., 255, pp. 9599-9607, (1980).

Krishnakantha,T.P. and Kurup,C.K.R., Environ. Physiol. Biochem., 4, pp. 80-88, (1974).

Kupfer,D. and Orrenius,S., Molec. Pharmacol., 6, pp. 221-230, (1970).

Kupfer,D. and Navarro,J., Life Sci., JL8, pp. 507-514, (1976).

Kupfer,D., Navarxo,J. and Piccolo,D.E., J. Biol. Chem., 253, pp. 2804-2811, (1978).

Kupfer,D., Miranda,G.K., Navarro,J., Piccolo,D.E. and Theoharides,A.D., J. Biol. Chem., 254, 10405-10414, (1979).

Kupfer,D., In "Hepatic Cytochrome P-450 Monooxygenase System" (ed. Schenkman,J.B. and Kupfer,D.), Pergamon Press, Oxford, Chapter 4, pp. 157-187, (1982).

Kurup,C.K.R., Aithal,H.N. and Ramasarma,T., Biochem. J., 116, pp. 773-779, (1970).

Kusunose,M., Kusunose,E., Ichihara,K., Ogita,K., Kaku,M. and Yamamoto,S., Advances in Prostaglandin, Thromboxane and Leukotriene Research, L5, pp. 155-158, (1985).

Lake,B.G., Gangolli,S.D., Grasso,P. and Lloyd,A.G., Toxicol. Appl. Pharmacol., 32, pp. 355-367, (1975).

Lake,B.G., Brantom,P.G., Gangolli,S.D., Butterworth,K.R. and Grasso,P., Toxicol., 6, pp. 341-356, (1976).

Lake,B.G., Phillips,J.C., Linnell,J.C. and Gangolli,S.D., Toxicol. Appl. Pharmacol., 39, pp. 239-248, (1977).

Lake,B.G., Gray,T.J.B., Stubberfield,C.R., Beamand,J.A. and Gangolli,S.D., Life Sci., 33, pp. 249-254, (1983). 349 Lake,B.G., Gray,T.J.B., Foster, J.R., Stubberfield,C.R. and Gangolli,S.D., Toxicol. Appl. Pharmacol., 72, pp. 46-60, (1984).

Lake,B.G., Gray,T.J.B., Pels Rijcken,W.R., Beamand,J.A. and Gangolli,S.D., Xeno., _14, pp. 269-276, (1984)a.

Lake,B.G., Pels Rijcken,W.R., Gray,T.J.B., Foster,J.R. and Gangolli,S.D., Acta Pharmacol, et Toxicol., 54, pp. 167-176, (1984)b.

Lake,B.G., Tredger,J.M., Gray,T.J.B., Stubberfield,C.R., Hodder,K.D., Gangolli,S.D. and Williams,R., Life Sci., 35, pp. 2621-2626, (1984)c.

Lake,B.G., Gray,T.J.B. and Gangolli,S.D., Environ. Hlth. Pers., 67, pp. 283-290, (1986).

Lake,B.G., Gray,T.J.B., Sjoberg,P., Hodder, K.D., Beamand,J.A., Stubberfield,C.R. and Gangol1i,S.D., Fd. Chem. Toxicol., 24, pp. 573-575, (1986)a.

Lake,B.G., Gray,T.J.B., Evans,J.G., North,C.J. and GangolliyS.D., Poster Abst., 28th Congress Of European Society of Toxicology, Strasbourg, (1987).

Lalwani,N.D., Reddy,M.K., Mangkomkanok-Mark,M. and Reddy, J.K., Biochem. J., 198, pp. 177-186, (1981).

Lalwani,N.D., Reddy,M.K., Qureshi,S.A. and Reddy,J.K., Carcin., 2, pp. 645-650, (1981)a.

LalwaniyN.D., FahlyW.E. and Reddy,J.K., Biochem. Biophys. Res. Commun., 116, pp. 388-393, (1983).

Lalwani,N.D., Reddy,M.K., Qureshi,S.A., Moehle,C.M., Hay as hi, H. and Reddy ,J.K., Cancer Res., 43, pp. 1680-1687, (1983)a.

Lalwani,N.D., Reddy,M.K., Qureshi,S.A., Sirtori,C,R., Abiko,Y. and Reddy,J.K., Human Toxicol., 2, pp. 27-48, (1983)b.

Lalwani,N.D., Reddy,M.K., Ghosh,S., Barnard,S.D., Molello,J.A. and Reddy,J.K., Biochem. Pharmacol., 34, pp. 3473-3482, (1985).

Lalwani,N.D., Alvares,K., Reddy,M.K., Reddy,M.N., Parikh,I. and Reddy,J.K., Proc. Natl. Acad. Sci. (U.S.A.), 84, pp. 5242-5246, (1987).

Landriscina,C., Gnoni,G.V. and Quagliariello,E., Biochem. Med., 12, pp. 356-364, (1975).

Lawrence,R.A. and Burk,R.F., Biochem. Biophys. Res. Commun., 71, pp. 952-958, (1976).

Lawrence,R.A. and Burk,R.F., J. Nutr., 108, pp. 211-215, (1978). 350 Lawrence,R.A, Parkhill,L.K. and Burk,R.F., J. Nutr., 108, pp. 981-987, (1978).

Lazarow,P.B. and De Duve,C., Proc. Natl. Acad. Sci. (U.S.A.), 73, pp. 2043-2046, (1976).

Lazarow,P.B., J. Biol. Chem., 253, pp. 1522-1528, (1978).

Lazarow,P.B., In "Metabolic Compartmentation" (ed. Sies,H.), Academic Press, pp. 317-329, (1982).

Lazarow,P.B., Shio,H. and Leroy-Houyet,M.A., J. Lipid Res., 23, pp. 317-326, (1982).

Lee,YJ?., Takemori,A.E. and Lardy,H., J. Biol. Chem., 234, pp. 3051-3054, (1959).

Lee,Y.P. and Lardy,H., J. Biol. Chem., 240, pp. 1427- 1436, (1965).

Lehinger,A.L., Biochemistry, Worth Publishers, 2nd edition, (1977)

Le Hir,M. and Dubach,U.C., J. Histochem. Cytochem., 30, pp. 441-444, (1982).

Lehrach,H., Diamond,D., Wozney,J.M. and Boedtker,H., Biochem., 16, pp. 4743-4751, (1977).

Leighton,F., Brandan,E., Lazo,0. and Bronfman,M., Ann. N. Y. Acad. Sci., 386, pp. 62-80, (1982).

Levine,W.G., Ord,M.G. and Stocken,L.A., Biochem. Pharmacol., 26, pp. 939-942, (1977).

Levine,L., In "Carcinogenesis", 7, (Ed. Hecker,E.), Raven Press, pp. 477-494, (1982).

Lewis,B., The Hyperlipidaemias: Clinical and Laboratory Practice, Blackwell Scientific Publications, Chapters 9, 10 and 15, (1976).

Lewis,P.R. and Knight,DI5., In "Practical Methods for Electron Microscopy, (ed. Glauert,A.), Elsevier / North Holland Publishing, 5, pp. 43-46, (1977).

Lewis,D.F.V., Drug Meta. Revs., 1/7, pp. 1-66, (1986).

Lillehaug,J.R. and Berge,R.K., Br. J. Cancer, 53, pp. 121-127, (1986).

Linnainmaa,K., Carcin., 5, pp. 703-707, (1984).

Lipsky,N.G. and Pedersen,P.L., J. Biol. Chem., 257, pp. 1473-1481, (1982).

Lowry,O.H., Rosebrough,N.J., Farr,A.L. and Randall,R.J., J. Biol. Chem., 1_93, pp. 265-275, (1951). 351 Lu,A.Y.H. and Coon,M.J., J. Biol. Chem., 243, pp. 1331-1332, ( 1968).

Lund,H. and Bremer^J., Biochim. Biophys. Acta., 750, 164-170, (1983).

Mackerer,C.R., Biochem. Pharmacol., 26, pp. 2225-2230, (1977).

Makowska,J.M., Bonner,FiW. and Gibson,G.G., Unpublished Observations, (1988).

Makowski,R., Davies,R. and Gibson,G.G., Rev. Drug Metab. Drug Inter., IV, pp. 123-170, (1982).

Makowski,R.J., Doctoral Thesis, (1985).

Mangham,B.A., Foster,J.R. and Lake,B.G., Toxicol. Appl. Pharmacol., 61, pp. 205-214, (1981).

Maniatis,T., Fritsch,E.F., Sambrook,J., Molecular Cloning, A Laboratory Manualy Ccild Spring Harbour Laboratory, Chapters 5 and 6, (1982).

Mann,A.H., Price,S.C., Mitchell,F.E., Grasso,P., Hinton,R.H. and BridgeSjJ.W., Toxicol. Appl. Pharmacol., 77, pp. 116-132, (1985).

Manna,S., Falck,J.R., Chacos,N. and Capdevila,J., Tet. Lett., 24, pp. 33-36, (1983).

Mannaerts,G.P., Debeer,L.J., Thomas,J and De Schepper,P.J., J. Biol. Chem., 254, pp. 4585-4595, (1979).

Mannervik,B., Methods in Enzymology, 113, pp. 490-496, (1985).

MarkwellyMJLK., McGroarty,E.J., Bieber,L.L. and Tolbert,N.E., J. Biol. Chem., 24 8, pp. 3426-3432, (1973).

Marnett,L.J., Carcin., 8, pp. 1365-1373, (1987).

Masters,C. and Crane,D., TIBS, 9, pp. 314-317, (1984).

Matsubara,S., Yamamoto,S., Sogawa,K., Yokotani,N., Fujii-Kuriyama,Y., Haniu,M., Shively,J.E., Gotoh,0., Kusunose,E. and Kusunose,M., J. Biol. Chem., 262, pp. 13366-13371, (1987).

Miettinen,T., Athero., 15, pp. 163-176, (1972).

Milton,M., Elcombe,C., Kass,G.E. and Gibson,G., Biochem. Pharmacol., 37, pp. 793-798, (1988).

Miltcn,M., Elccmbe,C. and Gibscn,G., Unpublished Observations, (1988)a.

Mitch ell, F.E., Price,S.C., Hinton,R.H., Grasso,P. and Bridges,J.W., Toxicol. Appl. Pharmacol., 01, pp. 371-392, (1985). 352 MittalyB. and Kurup,C.K.R., Biochim. Biophys. Acta., 619, pp. 90-97, (1980).

Mittal,B. and Kurup,C.K.R., Biochem. J., 194, pp. 249-255, (1981).

Mitz,M.A. and Heinrikson,R.L., Biochim. Biophys. Acta., 46, pp. 45-50, (1961). Miyazawa,S., Furuta,S. and Hashimoto,T., Eur. J. Biochem., 117/ pp. 425-430, (1981).

MizugakiyM., Nishimaki,T., Yamamota,H.,Sagi,M. and Yamanaka, H., J. Biochem., 92, pp. 2051-2054, (1982).

Mohandas,J., Marshall,J.J., Duggin,G.G., Horvath,J.S. and Tiller,D.J., Biochem. Pharmacol., 33, pp. 1801-1807, (1984).

Monk,J.P. and Todd,P.A., Drugs, 33, pp. 539-576, (1987).

Moody,D.E. and Reddy,J.K., Res. Commun. Chem. Pathol. Pharmacol., 9, pp. 501-510, (1974).

Moody,D.E. and Reddy,J.K., J. Cell Biol., 11, pp. 768-780, (1976).

Moody,D.E. and Reddy,J.K., Am. J. Pathol., 90, pp. 435-446, (1978).

Moody,D.E. and Reddy,J.K., Toxicol. Letts., 10, pp. 379-383, (1982).

Moody,D.E., Reddy,J.K. and Azamoff,D.L. Biochem. Pharmacol., 33, pp. 2591-2597, (1984).

Moody,D.E., Loury,D.N. and Hammock,B.D., Toxicol. Appl. Pharmacol., 1_Q, pp. 351-362, (1985).

Morrison,A.R. and Pascoe,N. Proc. Natl. Acad. Sci. (U.S.A.), 78, pp. 7375-7378, (1981).

Mortensen,P.B., Kolvraa,S., Gregersen,N. and Rasmussen,K. Biochim. Biophys. Acta., 713/ pp. 393-397, (1982).

Mortensen,P.B., Biochim. Biophys. Acta, 878, pp. 14-19, (1986).

Nair,N. and Kurup,CJK. Biochem. Pharmacol., 35, pp. 3441-3447, (1986).

Nair,N. and Kurup,C.K.R., Ind. J. Biochem. Biophys., 23/ 76-79, (1986)a.

Narbonne,J.F., Pelissier,M.A., Bonnamour,D., Borin,C. and Albrecht,R., Toxicol. Letts., 23, pp. 73-77, (1984).

Nash,T., Biochem. J., 55, pp. 416-421, (1953). 353 Navarro,J., Piccolo,D.E. and Kupfer,D., Arch. Biochem. Biophys., 19JL, pp. 125-133, (1978).

Neat,C.Ev Thomassen,S. and Osmundsen,H., Biochem. J., 186, pp. 369-371, (1980).

Nebert,D.W., Adesnik,M., Coon,M.J., Estabrook,R.W., Gonzalez,F.J., Guengerich,F.P., Gunsalus,I.C., Johnson,E.F., Kemper,B., Levin,W., Phillips,I.R., Sato,R. and Waterman,M.R., DNA, 6, pp. 1-11, (1987).

Nebert,D.W. and Gonzalez,F.J., Ann. Rev. Biochem., 56, pp. 945-993, (1987).

Needleman,P., Turk,J., Jakschik,B.A., Morrison,A.R. and Lefkowith,J.B., Ann. Rev. Biochem., 55, pp. 69-102, (1986).

Nilsson,A., Prydz,K., Rortveit,T. and Christiansen,E.N., Biochim. Biophys. Acta., 920, pp. 114-119, (1987).

Noshiro,M. and Negishi,M., J. Biol. Chem., 261, pp. 15923- 15927, (1986).

Novika£f,A.B. and Galdfischer,S., J. Histochem. Cytochem., 17, pp. 675-680, (1969).

O'Brien,T.G., Simsiman,R.C. and Boutwell,R.K., Cancer Res., 35, pp. 2426-2433, (1975).

Odum,J. and Orton,T.C., Br. J. Pharmacol., 69, 317P Abst., (1980).

Odum,J. and Orton,T.C., Biochem. Pharmacol., _32/ PP« 3565-3569, (1983).

Oesch,F. and Schladt,L., Poster Abst., 28th Congress of European Society of Toxicology, Strasbourg, (1987).

Okino,S.T., Quattrochi,L.C., Barnes,H.J., Osanto,S., Griffin,K.J., Johnson,E.F. and Tukey,R.H., Proc. Natl. Acad. Sci. (U.S.A.)., 82, pp. 5310-5314, (1985).

Okita,R.T. and Masters,B.S.S., Drug Meta. Dispos., 8, pp. 147-151, (1980).

Okita,R.T., Parkhill,L.K., Yasukochi,Y., Masters,B.S.S., Theoharidies,A.D. and Kupfer,D., J. Biol. Chem., 256, pp. 5961-5964, (1981).

Oliver,M., New England J. Med., 299, pp. 1360-1362, (1978).

01iw,E.H., Lawson,J.A., Brash,A.R. and Oates,J.A., J. Biol. Chem., 256, pp. 9924-9931, (1981).

01iw,E.H. and Oates,J.A., Biochim. Biophys. Acta., 666, pp. 327-340, (1981). 354 01iw,E.H., Guengerich,F.P. and Oates,J.A., J. Biol. Chem./ 257, pp. 3771-3781, (1982).

01sson,A.G. and Lang,P.D., Athero., 31, pp. 421-428, (1978).

Omura,T. and Sato,R., J. Biol. Chem., 239, pp. 2370-2378, ( 1964).

Orton,T.C. and Higgins,J.E., Toxicol. Appl. Pharmacol., 48, A126, Abst. 251, (1979). Orton,T.C. and Parker,G.L., Drug Metab. Dispos., JLO, pp. 110-115, (1982).

Orton,T.C., Adam,H.K., Bentley,M., Holloway,B. and Tucker,M.J., Toxicol. Appl. Pharmacol., 73, pp. 138-151, (1984).

Osmundsen,H., Ann. N. Y. Acad. Sci., 386, pp. 13-29, (1982).

Osumi,T. and Hashimoto,T., J. Biochem., 83, pp. 1361-1365, (1978).

OsumiyT. and Hashimoto,T., J. Biochem., 85, pp. 131-139, (1979).

Osumi,T., Ishii,N., Hijikata,M., Kamijo,K., Ozasa, H., Furuta^5., Miyazawa^., Kondo,K., Inoue, K., Kagamiyama,H. and Hashimoto,T., J. Biol. Chem., 260, pp. 8905-8910, (1985).

Paget,G.E., J. Athero. Res., 3, pp. 729-736, (1963).

Pande,S.V. and Parvin,R., Biochim. Biophys. Acta., 617, pp. 363-370, (1980).

Parker,G. and Orton,T., In "Biochemistry, Biophysics and Regulation of Cytochrome P-450" (eds. Gustafsson.J, Duke.J, ModeJ^ and RafterJ), Elsevier/North Holland Biomedical Press, pp 373-377, (1980).

Paul,H.S., Gleditsch,C.E. and Adibi,S.A., Am. J. Physiol., 251, pp. E311-E315, (1986).

Phi Hips,I.R., Shephard, E. A., Ashworth,A. and Rabin,B.R., Proc. Natl. Acad. Sci. (U.S.A.)., 82, pp. 938-987, (1985).

Pike,S.F., Shephard,E.A., Rabin,B.R. and Phillips,I.R., Biochem. Pharmacol., 34, pp. 2489-2494, (1985).

Platt,D.S. and Thorp,J.M., Biochem. Pharmacol., JL5 pp. 915-925, (1966).

Platt,D.S. and Cockrill,B.L., Biochem. Pharmacol., 18, pp. 429-444, (1969).

Poland,A. and Knutscn,J.C., Ann. Rev. Pharmacol. Toxicol., 22, pp. 517-554, (1982). 355 Pollard,A.D. and Brindley,D.N., Biochem. Pharmacol., 31, pp. 1650-1652, (1982).

Powell,W.S., J. Biol. Chem., 251, pp. 6711-6716, (1978).

Pourbaix,S., Heller,F. and Harvengt,C., Biochem. Pharmacol., 33, pp. 3661-3666, (1984).

Price,S.C., Hall,D.E. and Hinton,R.H., Toxicol. Letts., 25, pp. 11-18, (1985).

Price,S.C., Hinton,R.H., Mitch ell,F.E., Hall,D.E., Grasso,P., Blane,G.F. and Bridges,J.W., Toxicol., 41, pp. 169-191, (1986).

Prospero,T.D., Method. Developments in Biochemistry, 4, 411-412, (1974).

Quattrochi,L.C., Okino,S.T., Pendurthi,U.R. and Tukey,R.H., DNA, 4, pp. 395-400, (1985).

Rao,M.S., Lalwani,N.D., Scarpelli,D.G. and Reddy,J.K., Carcin., 3, pp. 1231-1233, (1982).

Rao,M.S., Lalwani,N.D., Watanabe,T.K. and Reddy,J.K., Cancer Res., 44, pp. 1072-1076, (1984).

Rao,M.S., Tatematsu,M., Subbarao,V., Ito,N. and Reddy,J.K., Cancer Res., 4_6, pp. 5287-5290, (1986).

Rao,M.S., Kokkinakis,D.M., Subbarao,V. and Reddy,J.K., Carcin., 8, pp. 19-23, (1987).

Rao,M.S. and Reddy,J.K., Carcin., 8, pp. 631-636, (1987).

Reddy^J.K., Bunyaratvej,S. and Svoboda,D., J. Cell Biol., 42, pp. 587-596, (1969).

Reddy^JK., Svcboda,D. and AzamoffyD., Biochem. Biophys. Res. Commun., 52, pp. 537-543, (1973).

Reddy^JK. and Krishnakantha,T.P., Science, 190, pp. 787-789, (1975).

Reddy,J.K., Krishnakantha,T.P., Azamoff,D.L. and Moody,D.E., Res. Commun. Chem. Pathol. Pharmacol., JL0, pp. 589-592, (1975).

Reddy,J.K., Rao,M.S. and Moody,D.E., Cancer Res., 36, pp. 1211-1217, (1976).

Reddy,J.K. and Kumar,N.S., Biochem. Biophys. Res. Commun., 77, pp. 824-829, (1977).

Reddy,J.K. and Rao,M.S., Am. J. Pathol., 86, 25a Abst. 13, (1977). 356 Reddy,J.K. and Rao,M.S., J. Natl. Cancer Inst., 59, pp. 1645-1647, (1977)a.

Reddy,J.K. and Rao,M.S., Br. J. Cancer, 38, pp. 537-543, ( 1978).

Reddy,J.K. and Qureshi,S.A., Br. J. Cancer, 40, pp. 476-482, (1979).

Reddy,J.K., Rao,M.S., Azamoff,D.L. and Sell,S., Cancer Res., 39, pp. 152-161, (1979).

’ Reddy,J.K., Fed. Proc., 39, ]284, Abst.98, (1980).

Reddy,J.K., Azamoff,D.L. and Hignite,C.E., Nature, 283, pp. 397-398, (1980).

Reddy,J.K., Lalwani,N.D., Dabholkar,A.S., Reddy,M.K. and Qureshi,S.A., Biochem. Int., 3, pp. 41-49, (1981).

Reddy^JJC., Lalwani,N.D., Reddy,M.K. and Qureshi,S.A., Cancer Res., 42, pp. 259-266, (1982).

Reddy,J.K. and Lalwani,N.D., C.R.C. Critical Reviews in Toxicology, JL2, pp. 1-58, (1983).

Reddy,J.K., Lalwani,N.D., Qureshi,S.A., Reddy,M.K. and Moehle,C.M., Am. J. Pathol., 114, pp. 171-183, (1984).

Reddy^JK., Osumi,T., Hahimoto,T., Nemali,M., Lalwani,N.D. and Reddy,M.K., J. Cell Biol., 101, pp. 66, Abst. 248, (1985).

Reddy,J.K. and Rao,M.S., TIPS, pp. 438-443, (Nov. 1986).

Reddy,J.K., Goel,S.K., Nemali,M.R., Carrino,J.J., Laffler, T.G., Reddy,M.K., Sperbeck,S.J., Osumi,T., Hashimoto,T,, Lalwani,N.D. and Rao,M.S., Proc. Natl. Acad. Sci. (U.S.A.), 83,- 1747-1751, (1986).

Reddy^J.K., Reddy,M.K., Usman,M.I., Lalwani,N.D. and Rao,M.S., Enviro. Hlth. Pers., 65, pp. 317-327, (1986)a.

Reddy,M.K., Hollenberg,P.F. and Reddy,J.K., Biochem. J., 188, pp. 731-740, (1980).

Reddy,M.K., Lalwani,N.D., Qureshi,S.A. and Reddy,J.K., Human Toxicol., 1, pp. 135-147, (1982).

Reich,N.O. and Ortiz De Montellano,P.R., Biochem. Pharmacol., 35, pp. 1227-1233, (1986).

Rigby,P.W.J., Dieckmann,M., Rhodes,C. and Berg,P., J. Mol. Biol., 113, pp. 237-251, (1977).

Robbi,M. and Lazarow,P.B., Proc. Natl. Acad. Sci. (U.S.A.), 75, pp. 4344-4348, (1978). 357 Rogers,H. and Spector,R., Aids to Pharmacology, Churchill Livingstone, pp. 186-187, (1983).

Romer,A.S., In "Man and the Vertebrates:2", Penguin Publishing, ( 1966).

Rouffy,J., Chanu,B., Bakir,R., Djian,F. and Goy-Loeper,J., Athero., 54, pp. 273-281, (1985).

Ruddick,J.A., Villeneuve,D.C., Chu,I., Nestmann,E. and Miles,D., Bull. Environ. Contain. Toxicol., 27, pp. 181-186, (1981).

Ruegamer,W.R., Ryan,N.T., Richert,D.A. and Westerfeld,W.W., Biochem. Pharmacol., 1_8, pp. 613-624, (1969).

Ryan,D., Thomas,P., Korzeniowski,D. and Levin,W., J. Biol. Chem., 254, pp. 1365-1374, (1979).

Sasame,H.A., Thorgeirsson,S.S., Mitchell, J.R. and Gillette,J.R., Life Sci., 14, pp. 35-46, (1974).

Schmezer,P., Pool,B.L., Klein,R.G., Komitowski,D. and Schmahl,D., Carcin., 9, pp. 37-43, (1988).

Schmidt,F., Stegmeier,K., Stork,H., Endele,R. and Koch.K., In "Lipoproteins and Coronary Heart Disease” (Eds. Gretan,H., Lang,P. and Schettler,G.), Gerhard Witzstrock Publishing house, pp. 102-105, (1980).

Schulte-Hermann,R., Timmermann-Trosiener,I. and Schuppler,J., Cancer Res., 43, pp. 839-844, (1983).

Schwartzman,MX., Abraham,N.G., Carroll,M.A., Levere,R.D. and McGiff,J.C., Biochem. J., 238, pp. 283-290, (1986).

Sharma,R., Lake,B.G., Foster,J. and Gibson,G., In Press, (1988)

Sharma,R., Lake,B.G. and Gibson,G., In Press, (1988)a

Sharma,R., Skett,P. and Gibscn,G.G., Unpublished Observations, (1988)b.

Sinclair,K.A. and C aid well, J., Biochem. Pharmacol., 31, 953-957, (1982).

Singh,H., Derwas,N. and Poulos,A., Arch. Biochem. Biophys., 254, pp. 526-533, (1987).

Small,G.M., Burdett,K. and Connock,N.J., Ann. N. Y. Acad. Sci., 386, pp. 460-463, (1982).

ScQberg,H.E., Aas,M. and Daae,L.N.W., Biochim. Biophys. Acta., 280, pp. 434-439, (1972). 358 Song,B.J., Matsunaga,T., Hardwick, J.P., Park,S.S., Veech,R.L., Yang,C.S., Gelboin,H.V. and Gonzalez,F.J., Molec. Endocrin., 1, pp. 542-547, (1987).

Stakkestad,J.A. and Bremer,J., Biochim. Biophys. Acta., 750, 244-252, (1983). Stegmeier,K., Schmidt,F.H., Reinicke,A. and Fahimi,H.D., Ann. N. Y. Acad. Sci., 386, pp. 449-452, (1982).

Stogniew,M. and Fenselau,C., Drug Meta. Dispos., 10, pp. 609-613, (1982).

Stupans,I., Ikeda,T., Kessler,D.J. arid Nebeirt,D.W., DNA, 3, pp. 129-137, (1984).

Styles,J.A., Kelly,M. and Elcombe,C.R., Carcin., 8, pp. 391-399, (1987).

Suzuki,Y., Orii,T., Takiguchi,M., Mori,M., Hijikata,M. and Hashimoto,T., J. Biochem., 101, pp. 491-496, (1987).

Svoboda,D.J. and Azarnoff,D.L., J. Cell Biol., 30, pp. 442-450, (1966).

Svoboda,D., Grady,H. and Azamoff,D., J. Cell Biol., 35, pp. 127-152, (1967).

Svcboda,D., Azamoff,D. and Reddy,J.K., J. Cell Biol., 40, pp. 734-746, (1969).

Tamburini,P.P., Masson,H.A., Bains,S.K., Makowski,R.J., Morris,B. and Gibscn,G.G., Eur. J. Biochem., 139, pp. 235-246, (1984).

Taussig,M.J., In "Processes in Pathology", Blackwell Scientific Publications, pp. 689-809, (1984).

Taylor,J., Ann. Rev. Biochem., 48, pp. 681-717, (1979).

Telakowski-Hopkins,C.A., Lu,A.Y.H. and Pickett,C.B., Arch. Biochem. Biophys., 221, pp. 79-88, (1983).

Theoharides,A.D. and Kupfer,D., J. Biol. Chem., 256, pp. 2168-2175, (1981).

Thomas,P.E., Reidy,J., Reik,L.M., Ryan,D.E., Koop,D.R. and Levin,W., Arch. Biochem. Biophys., 235, pp. 239-253, (1984).

Thomas,P.E., Bandiera,S., Reik,L.M., Maines,S.L., Ryan,D.E. and Levin,W., Fed. Proc. 46, pp. 2563-2566, (1987).

Thorp,J.M., Proc. Eur. Soc. Study Drug Toxicol., 13, pp. 98-109, (1972).

Tolbert,N.E., Ann. Rev. Biochem., 50, pp. 133-157, (1981).

Tomaszewski,K.E., Agarwal,D.K. and Melnick,R.L., Carcin., 7, pp. 1871-1876, (1986). 359 Toto,R., Siddhanta,A., Manna,S., Pramanik,B., Falck,J.R. and Capdevila,J., Biochim. Biophys. Acta., 919, pp. 132-139, (1987).

TurteJtanb/K.W. and Murphy,P.A., Arch. Biochem. Biophys., 255, pp. 120-126, (1987).

Vainio,H., Linnainmaa,K., Kahonen,M., Nickels,J., Hietanen,E., Mamiemi,J and Peltonen,P., Biochem. Pharmacol., 32, pp. 2775-2779, (1983).

Vamecq,J. and Draye,J-P., J. Biochem., 102, pp. 225-234, (1987).

Van Breemen,R.B. and Fenselau,C., Drug Meta. Dispos., 13, pp. 318-320, (1985).

VanRollins,M., Baker,R.C., Sprecher,H.W. and Murphy,R.C., J. Biol. Chem., 259, pp. 5776-5783, (1984).

Van Veldhoven,P.P., Declercq,P.E., Debeer,L.J. and Mannaerts, G.P., Biochem. Pharmacol., 33, pp. 1153-1155, (1984).

Van Veldhoven,P.P., Just,W.W. and Mannaerts,G.P., J. Biol. Chem., 262, pp. 4310-4318, (1987).

Vessby,B., Lithell,H. and Ledermann,H., Athero., 44, pp. 113-118, (1982).

Voller,A., Bartlett,A. and Bidwell,D.E., J. Clin. Pathol., 31, pp. 507-520, (1978).

Vcn Daniken,A., Lutz,W.K. and Schlatter,C., Tox. Letts.,7, pp. 311-319, (1981).

Vcn Daniken,A., Lutz,W.K., Jackh,R. and Schlatter,C., Toxicol. Appl. Pharmacol., T3, pp. 373-387, (1984).

Wada,F., Usami,M., Goto,M. and Hayashi,T., J. Biochem., 70, pp. 1065-1067, (1971).

Warren^J.R., Simmon,V.F. and Reddy,J.K., Cancer Res., 40, pp. 36-41, (1980).

Waxman,D.J., Biochem. Pharmacol., 31, pp. 71-84, (1988).

Weil,A., Caldwell,J., Guirhard,J.P. and Blane,G.F., Poster Abst., 28th Congress of European Society of Toxicology, Strasbourg, (1987).

Wells,D.S., Janssen,F.W. and Ruelius,H.W., Xeno., _17, pp. 1437-1449, (1987).

Wendel,A., Enymatic Basis of Detoxication (ed. Jacoby,W.B.), Academic Press., 1, pp. 333-335, (1980).

Westerfeld,W.W., Richert,D.A. and Ruegamer,W.R., Biochem. Pharm., JL7, pp. 1003-1016, (1968). 360 White,B.A. and Bancroft,F.C., J. Biol. Chem., 257, pp. 8569-8572, (1982).

Whitlcck,J.P., Ann. Rev. Pharmacol. Toxicol., 26, pp. 333-369, (1986).

Williams,G.M., Fd. Cosmet. Toxicol., 19, pp. 577-583, (1981).

Williams,G.M., Fund. Appl. Toxicol., 4, pp. 325-344, (1984).

Wrighton,S.A., Schuetz,E.G., Watkins,P.B., Maurel, P., Barwick,J., Bailey,B.S., Hartle,H.T., Young,B. and Guzelian,P., Molec. Pharmacol., 28, pp. 312-321, (1985).

Yamada,J., Horie,S., Watanabe,T. and Suga,T., Biochem. Biophys. Res. Commun., 125, pp. 123-128, (1984).

Yamamoto,S., Kusunose,E., Kaku,M., Ichihara,K. and Kusunose,M., J. Biochem., 100, pp. 1449-1455, (1986). Yamazaki,R.K., Shen,T. and Schade,G.B., Biochim. Biophys. Acta., 920, pp. 62-67, (1987).

Yang,S-Y, Cuebas,D. and Schulz,H., J. Biol. Chem., 261, pp. 15390-15395, (1986).

Yokota,S. and Fahimi,H.D., Ann. N. Y. Acad. Sci., 386, pp. 491-494, (1982).

Yoshimoto,M., Kusunose,E., Yamamoto,S., Maekawa,M. and Kusunose,M., Biochem. Int., _13, pp. 749-755, (1986). 361 APPENDIX 1.

HAEMATOLOGY, CLINICAL BIOCHEMISTRY AND AUTOPSY PROCEDURES

COMPLETED IN THE MARMOSET STUDIES.

Haematology ;

Red cell count, white cell count (including differential),

haemoglobin, haematocrit and platelet count.

Clinical biochemistry ;

Aspartate aminotransferase, alanine aminotransferase, alkaline

phosphatase, lactate dehydrogenase, gamma-glutamyl

transpeptidase, creatine kinase, total protein, urea,

creatinine, cholesterol, triglyceride and bilirubin. Plasma

hormones determined by radioimmunoassay; insulin and gastrin.

Autopsy procedures ;

Gross observations; examination of all external surfaces and

orifice, macroscopic observation of organs in situ.

Tissue for histological examination;

Adipose, adrenals, aorta, bone marrow (sternum), bladder,

brain (medulla/pens; cerebellar cortex; cerebrum), eyes, gall

bladder, heart, intestine (duodenum, jejunum, ileum, colon),

kidneys, liver, lungs, lymph node (mesenteric and cervical),

oesophagus, pancreas, pituitary, prostate, salivary gland

(submaxillary), skeletal muscle (thigh), skin, spinal cord

(cervical; mid-thoracic; lumbar), spleen, stomach, testes

(with epididymes), thymus, thyroid and parathyroid, tongue, trachea and any gross changes c£ uncertain nature of tissue masses. 362 APPENDIX 2.

TAXONOMY OF THE DIFFERENT SPECIES*

Terrestrial Vertebrates;

Class : Mammalia

Order : Rodentia

Sub-order : Myomorpha

for example - mouse and rat.

Class : Mammalia

Order : Rodentia

Sub-order : Hystricomorpha

for example - guinea pig and hamster.

Class : Mammalia

Order : Lagomorpha Duplicidentata

for example rabbit.

Class : Mammalia

Order : Primates

Sub-order : Anthropoidea

Infra-order : Platyrrhini

Family : Hapalidae

Species : Callithrix jacchus

Derived from Romer, (1966).