Molecular and Cellular Biochemistry 96: 153-161, 1990. © 1990Kluwer Academic Publishers. Printed in the Netherlands. Original Article

Alterations in the integrity of peroxisomal membranes in livers of mice treated with peroxisome proliferators Denis I. Crane, Jane Zamattia and Colin J. Masters Division o f Science and Technology, Griffith University, Brisbane, Australia Received28 September 1989; accepted10 January 1990

Key words: catalase, membranes, peroxisome proliferators, peroxisomes Abstract

Catalase leakage from its particulate compartment within the light mitochondrial fraction of liver was used as an index of the integrity of peroxisomes in untreated mice and in mice treated with the peroxisome proliferators clofibrate(ethyl-p-chlorophenoxyisobutyrate), Wy-14,643(4-chloro-612,3-xylidino)-2-pyrimidinylthio]acetic acid) and DEHP(di-(2-ethylhexyl)phthalate). Catalase leakage represented about 2% of the total catalase activity when fractions from untreated mice were incubated at 4° C, increasing to about 5% during 60 rain incubation at 37° C. In fractions from livers of mice treated with peroxisome proliferators, catalase leakage was significantly higher, being 7--11% at 4°C and increasing to approximately 20% after 60 rain incubation at 37° C. The pattern of release was similar for all proliferators. Parallel data were obtained for catalase latency in these fractions, i.e. following 60 min incubation at 37 ° C, free (non-latent) catalase activity was 18% in control mice and 65, 67, and 83% in fractions from clofibrate-, Wy-14,643- and DEHP-treated mice, respectively. Differences in catalase leakage from peroxisomes in fractions from untreated mice and clofibrate-treated mice were also apparent following treatments designed to effect membrane permeabilization, as in freeze-thawing, osmotic rupture, and extraction with Triton X-100 and lysophosphatidylcholine. These data are consistent with a significant alteration in the integrity of the membranes of peroxisomes in livers of mice which have been treated with peroxisome proliferators, and furthermore indicate a commonality of effect of these agents.

Introduction

A large number of chemical agents is known which produce a marked induction of peroxisomes in livers of rodents. Such peroxisome proliferators include the hypolipidemic drug, clofibrate, and its structural analogs, as well as hypolipidemic agents structurally unrelated to clofibrate, such as Wy-14,643, and plasticizers like D E H P [1-5]. In addition to increasing peroxisome numbers, these agents induce the synthesis of a number of peroxisomal enzymes, particularly those of the perox-

isomal pathway for the [3-oxidation of fatty acids [4-7]. A variety of data accumulated in recent years has indicated that, as well as showing these enzymatic differences, proliferated peroxisomes display physical properties which distinguish them from peroxisomes in untreated rodents. These differences include density-gradient sedimentation characteristics [8], membrane surface properties [9], integral membrane protein profile [10], membrane fluidity [11], and membrane phospholipid composition [12].

154 In the latter observation [12], in particular, we have demonstrated a decrease in the normal content of phosphatidylcholine and an increase in lysophosphatidylcholine content in hepatic peroxisomes from clofibrate-treated mice. The membrane-destabilizing properties of lysophosphatidyl choline are well recognized [13] and we have therefore been interested in defining the possible effect of this lysophospholipid on peroxisomal membrane stability. In this regard, it is of interest to note that we have previously shown that treatment of mice with peroxisomal proliferators, including clofibrate, results in an apparent shift in the relative localization of catalase from the peroxisomal compartment to the cytoplasmic compartment of liver [14, 15], a phenomenon which may relate to increased peroxisomal membrane permeability in these animals. In the present research we show that the integrity of the peroxisomal membranes from livers of mice treated with peroxisome proliferators is altered, and attention is drawn to the fact that the difference in stability of peroxisomal membranes at 37° C would seem to provide a convenient and improved method for the quantitative evaluation of the effect of peroxisome proliferators on peroxisome membrane characteristics.

Materials Mature female mice (Quackenbush strain) were obtained from the Central Animal Breeding House, Moggill, Brisbane. Clofibrate (Atromid-S) was a gift from I.C.I. Australia; Wy-14,643 was a gift of Wyeth-Ayerst Research, Princeton, NJ. DEHP was obtained from Fluka AG, Buchs, Switzerland. Synthetic lysophosphatidylcholine (palmitoyl) was from the Sigma Chemical Company, St. Louis, Missouri, USA. All other chemicals were of reagent grade.

Experimental procedures

Administration of peroxisome proliferators Drugs were dissolved in ethanol and added uniformly to laboratory chow pellets, the ethanol being allowed to evaporate. Mice weighing 35-40 g had access to these chow pellets and water ad libitum. The concentrations of the drugs in the chow pellets, and the periods of administration, were as follows: clofibrate, 0.35% (w/w) for 10 days; Wy-14,643, 0.2% (w/w) for 7 days; DEHP, 2% (w/w) for 14 days.

Liver fractionation Livers were removed from mice, finely minced using a tissue chopper, and homogenized in 7.5 ml of cold 0.25 M sucrose containing 0.1% ethanol using a Potter-Elvehjem homogenizer with a loose-fitting pestle. The pestle was rotated at 1000 rpm through one up-and-down passage. The homogenates were centrifuged at 2800 g for 15min at 4°C in a Sorvall SS34 rotor, and the supernatants taken and further centrifuged at 27000 g for 15 min to pellet light mitochondrial fractions, containing most of the liver peroxisomes. These light mitochondrial pellets were resuspended in a small volume of sucrose/ethanol and homogenized as before. Peroxisomes and peroxisomal membranes were isolated from light mitochondrial fractions using a modification [10, 16] of our earlier published procedure [17].

Analysis of peroxisome membrane properties In these experiments, liver light mitochondrial fractions were subjected to a number of treatments designed to test the physical properties of the membranes of the peroxisomes in these fractions. Following the particular treatment, the fractions were centrifuged at 27000 g for 15 rain and the pellet and supernatant fractions which were generated, were separated, suspended in buffer containing 0.1%

155 25,

25 (A)

g

=-,

(e)

2O

20.

15-

15.

10-

lO.

5'

5,

21o

0

41o

61o

i

o

i

20

40

L

L

610

25 (O)

(c) 20

15

10

1 5

5-I

1 21o

41o

0

60

In©ubmtlon tlmo

I t

0

(rain) mt

i

610

370C

Fig. 1. Effect of temperature on the leakage of catalase from peroxisomes. Liver light mitochondrial fractions from mice were incubated

at 37° C and aliquots taken at the times indicated and centrifuged at 27 000 g for 15 rain. Catalase activity in the resultant supernatant fractions was expressed as a percentage of the total catalase activity (supernatant plus pellet activity). Values represent means + S.D., n = 5. A, control mice; B, clofibrate-treated mice; C, Wy-14,643-treated mice; D, DEHP-treated mice.

Table 1. Latency of catalase in liver light mitochondrial fractions from untreated mice and mice treated with peroxisome proliferators

Treatment

Control (4) Clofibrate (4) Wy-14,643 (3) DEHP (3)

Free catalase activity (% total) after incubation for 30 rain at 4° C

30 rain at 37° C

21.6 + 3.7 46.2 + 6.5 N.D. N.D.

18.1 + 1.2 64.1 + 5.5 65.4_ 11.6 83.6_+ 6.3

Liver light mitochondrial fractions were incubated for 30 min at 4° C or 37° C. Catalase activity in these fractions was then determined in the absence (free catalase activity) and presence (total catalase activity) of 0.1% Triton X-100. Values represent means + S.D. for the number of experiments shown in parenthesis. N.D., not determined.

T r i t o n X-100, a n d assayed for catalase activity. T h e increase in soluble catalase, as a p r o p o r t i o n of total catalase, was t a k e n as a m e a s u r e of the p e r m e a b i l ity of the p e r o x i s o m a l m e m b r a n e in r e l a t i o n to this enzyme. I n some e x p e r i m e n t s , catalase l a t e n c y in the fractions was also d e t e r m i n e d i m m e d i a t e l y following i n c u b a t i o n by m e a s u r i n g the catalase activity m e a s u r a b l e in the a b s e n c e of T r i t o n X-100 (i.e. with the p e r o x i s o m e s intact) a n d expressing this as a p e r c e n t a g e of the total activity, o b t a i n e d in the p r e s e n c e of this d e t e r g e n t [18]. T h e t r e a t m e n t s of the light m i t o c h o n d r i a l fractions w e r e as follows: (a) H e a t stability: i n c u b a t i o n for 6 0 r a i n at 37 ° C, with aliquots b e i n g t a k e n at suitable intervals d u r i n g this time.

156 2O

40

.9,o

30

_=

20

o

15

10

e

.m,_

g

lO

5

1 J

i 20

10 Number

of

freeze-thaw

0

from peroxisomes of mouse liver. Liver light mitochondrial fractions were frozen in liquid nitrogen and then thawed at 37° C for the number of cycles indicated. The fractions were then centrifuged and assayed for catalase as in Fig. 1. Values represent means _+S.D., n= 4. * P < 0.02. [], untreated mice; 0, clofibrate-treated mice.

(b) F r e e z e - t h a w experiments: freezing in liquid nitrogen and then thawing (minimal time) in a 37 ° C water bath, this p r o c e d u r e being repeated up to 20 times. (c) Effect of osmotic pressure: addition to sucrose solutions to give a final sucrose c o n c e n t r a t i o n of 0 . 2 5 M , 0 . 5 0 M , 0 . 7 5 M , or 1 M , incubation in these solutions for 15 min at 4 ° C and then dilution back to 0.25 M sucrose. (d) Sonication: sonication for 30 sec at varying amplitudes of an M S E sonicator, with the fractions being maintained on ice during these procedures. (e) Triton X-100 extraction: incubation for 5 rain at 4 ° C in different concentrations of Triton X-100. (f) Extraction with lysophosphatidylcholine: incubation for 10 rain at 37°C with concentrations of lysophosphatidylcholine (dissolved in 0.25 M sucrose/0.1% ethanol) up to 1 raM.

Analytical procedures Catalase activity was d e t e r m i n e d as described pre-

0.4 Sucrose

cycles

Fig. 2. Effect of freezing and thawing on the leakage of catalase

i

0.2

i

0,6

i

i0

0.8

concentration

1.

(M)

Fig. 3. Effect of sucrose concentration on leakage of catalase from peroxisomes of mouse liver. Liver light mitochondrial fractions (in 0.25 M sucrose) were suspended to the sucrose concentrations indicated for 15 rain and then brought back to 0.25 M sucrose. The fractions were then centrifuged and assayed for catalase as in Fig. 1. Values represent means + S.D., n = 8. rq, untreated mice; O, clofibrate-treated mice. Regression slopes were statistically different, P < 0.01. viously [19]. Protein was m e a s u r e d using the method of Peterson [20].

Statistical analyses Significance testing of data means was p e r f o r m e d using Student's t-test, and of slopes using regression analysis. D a t a are expressed as means _ S.D.

Results

In regard to the release of catalase f r o m its particulate-bound f o r m in the light mitochondrial fraction of liver, fractions f r o m u n t r e a t e d mice, w h e n maintained at 4 ° C, released only about 2% of their total catalase into the supernatant fraction. This release increased gradually to about 4% over a 60 rain incubation period w h e n the t e m p e r a t u r e of incubation was 37 ° C (Fig. 1A). I n c u b a t i o n at 25 ° C yielded results which were intermediate b e t w e e n those at 4 ° C and 37°C (data not shown). In contrast, fractions f r o m mice treated with the p e r o x i s o m e proliferators exhibited m o r e m a r k e d

157 70

80,

6O o

60'

50

g 40 ~Q0

40,

30 20

20

2o

qa ¢=

10

0 ~ 1~0

Sonlcatlon

2~0

amplitude

30

(microns)

~

~

f(mg/mE/ raP/mRTrefl:br:i: membrane protein I

0.00

0.10

Triton

X-lO0

0.20 (g/lOOml)

Fig. 4. Effect of sonication of peroxisomes on the leakage of

Fig. 5. Leakage of catalase from peroxisomes exposed to Triton

catalase. Liver light mitochondrial fractions were sonicated for 30 sec on ice at the amplitudes indicated, and then centrifuged and assayed for catalase as in Fig. 1. Values represent means + S.D., n = 4. [2, untreated mice; O, clofibrate-treated mice.

X-100. Liver light mitochondrial fractions were incubated for 5 rain at 4° C in the concentrations of Triton X-100 indicated, and then centrifuged and assayed for catalase as in Fig. 1. Values represent means _+S.D., n = 6. * P < 0.02. [3, untreated mice; O, clofibrate-treated mice. Insert: Peroxisomes were isolated from livers of mice and incubated for 5 rain at 4° C with Triton X-100 at the indicated ratios of Triton X-100 to peroxisomal membrane protein. Peroxisomal membranes were prepared for protein determination by carbonate extraction of peroxisomal fractions [17]. Values represent the means of 2 experiments. [2, untreated mice; O, clofibrate-treated mice.

release of p a r t i c u l a t e - b o u n d catalase u n d e r all conditions. Following clofibrate t r e a t m e n t , for example, catalase release f r o m liver light mitochondrial fractions m a i n t a i n e d at 4 ° C was 7 % , and incubation at 3 7 ° C resulted in a dramatic increase in the a m o u n t of catalase release such that within 10 min almost 18% o f the e n z y m e h a d b e e n released. L o n g e r incubation at 37 ° C p r o d u c e d only slightly greater release a b o v e the 10 min level (Fig. 1B). A d d i t i o n of 1 m M glutaraldehyde to the light mitochondrial fractions to stabilize peroxisomal m e m b r a n e s [17] did not alter the results o b t a i n e d for either u n t r e a t e d mice or clofibrate-treated mice (data not shown). Similar results to those seen for clofibrate were o b t a i n e d with the fractions f r o m livers of mice treated with Wy-14,643 (Fig. 1C) and D E H P (Fig. 1D) in that the release of catalase was initially a b o u t 8 - 1 0 % at 4 ° C and increased to a b o u t 20% after 60 min at 37 ° C. Catalase latency was also m e a s u r e d in these light mitochondrial fractions (Table 1). F o r control mice, free catalase activity was a b o u t 2 0 % , both at 4 ° C and after 30 rain incubation at 37 ° C. Fractions f r o m clofibrate-treated mice exhibited m o r e than double the free catalase activity of u n t r e a t e d mice w h e n i n c u b a t e d at 4 ° C, and this level was further

increased u p o n heating the fractions at 37 ° C. V e r y high free catalase activities were also seen w h e n fractions f r o m livers of Wy-14,643- and D E H P treated mice were incubated at 37 ° C.

Effect of membrane permeabilization procedures R e p e a t e d freezing and thawing of liver light mitochondrial fractions resulted in a progressive release of catalase f r o m its particulate-bound f o r m (Fig. 2). This release was significantly m o r e pron o u n c e d in the case of clofibrate-treated mice than for u n t r e a t e d mice, with a highly significant statistical difference being evident after only four freezethaw cycles. T h e m e m b r a n e s of peroxisomes f r o m clofibratetreated mice and untreated mice also exhibited different susceptibilities to osmotic pressure. A s can be seen f r o m the results presented in Fig. 3, exposure of fractions to increasing concentrations of sucrose, followed by dilution back to the initial

158 11111.

m

80.

g 60.

i >,

m

40-

20"

el

d

0 0.0

i

O.~2

0.4

Lysophosphatidylchollne

i

0.6

i

0.8

i

1.0

(mmol/I)

Fig. 6. Catalase leakage from peroxisomes incubated with ly-

sophosphatidylcholine.Liver lightmitochondrialfractionsfrom mice were incubated for 10 rain at 37° C with the concentrations of lysophosphatidylcholineindicated, and then centrifugedand assayed for catalase as in Fig. 1. Values represent means + S.D., n = 4. [3, untreated mice; 0, clofibrate-treated mice. Values from clofibrate-treated mice were statisticallydifferent to the correspondingvaluesfrom untreated mice at all lysophosphatidylcholine concentrations, P < 0.05. sucrose concentration of 0.25 M, resulted in different tendencies with respect to catalase leakage (regression slopes, P < 0.01), with the peroxisomes from clofibrate-treated mice showing the lesser tendency: Sonication of light mitochondrial fractions resulted in substantial release of particulate catalase at all sonicator amplitudes tested (Fig. 4); this method of membrane disruption released 60-80% of the particulate catalase in fractions from both untreated mice and clofibrate-treated mice. No statistical differences in the level of catalase release between fractions from untreated and clofibratetreated mice were apparent following any of the sonication conditions used. The effect of exposure of light mitochondrial fractions to increasing concentrations of Triton X-100 is shown in Fig. 5. A distinct difference in the pattern of catalase release was evident between fractions from untreated mice and those from clofibrate-treated mice. In particular, at a Triton X-100 concentration of 0.075%, catalase release

from untreated mice was 39%, but only 17% for clofibrate-treated mice. This type of experiment was repeated using isolated peroxisomes, so that the ratio of Triton X-100 to peroxisomal membrane protein could be more effectively controlled (Fig. 5, insert). Similar trends to those obtained with the liver light mitochondrial fractions were again observed, indicating that the differential effects seen are not attributable to disparate amounts of detergent available for interaction with peroxisomal membranes. Lysophosphatidylcholine itself was tested for its effect on the permeability of peroxisomes from untreated mice. Incubation of liver light mitochondrial fractions at 37°C with increasing concentrations of lysophosphatidylcholine resulted in a progressive increase in the leakage of catalase (Fig. 6). The data obtained closely resembled those seen with Triton X-100 extraction, in that there was a distinct difference in the pattern of catalase release with increasing concentration of lysophosphatidylcholine between fractions from normal and clofibrate-treated mice. Moreover, as with Triton X-100 extraction, fractions from clofibrate-treated mice displayed a greater tendency to release catalase at low lysophosphatidylcholine concentration, but a lesser tendency at higher concentrations of this lysolipid.

Discussion

In these experiments we have made use of catalase latency and catalase leakage from peroxisomes as indices of the permeability of the peroxisomal membrane. Catalase is the most abundant protein in peroxisomes and has been reported to be localized within the matrix compartment of these organelles, from which it is able to leak [21]. Thus measurement of the leakage of this enzyme from peroxisomes in liver fractions presents a ready index of the permeability of the peroxisomal membrane in relation to this enzyme. Furthermore, in these experiments use has been made of liver light mitochondrial fractions as our main source of peroxisomes, and it may be noted in this regard that this fraction is rendered amenable to such studies

159 because catalase is a peroxisome-specific enzyme, not being found within any of the other organelles in liver light mitochondrial fractions [22]. In addition however, the use of light mitochondrial fractions precludes the necessity of having to use purified peroxisomal fractions in such experiments. Purified peroxisomal fractions are, by the nature of the isolation procedures, modified in terms of membrane permeability properties [17, 21] and are therefore not particularly suitable for the assessments of membrane properties of the type carried out in the present studies. The second method of analysis of peroxisomal membrane permeability examines the latency of catalase in peroxisomes of light mitochondrial fractions. Catalase latency refers to the phenomenon where most of the catalase activity within intact peroxisomes is not normally accessible to the assay procedure; only after permeabilization of the membrane is determination of total activity possible. This latency of catalase activity in intact peroxisomes has been attributed to a permeability barrier to hydrogen peroxide, the substrate for this enzyme, at the peroxisomal membrane [23]. Both of these measurements, viz. catalase leakage and catalase latency, therefore represent parameters which reflect the permeability properties of the peroxisomal membrane; catalase latency in terms of permeability of a small metabolite, hydrogen peroxide, and catalase leakage in terms of permeability of a large molecular weight protein. Overall, the results of these experiments indicate pronounced changes in the permeability of the peroxisome membrane as judged by the degree of catalase leakage and free (non-latent) catalase activity in peroxisomes from livers of mice treated with peroxisome proliferators. These changes were particularly apparent in fractions exposed to incubation at 37 ° C. The data from the freezing and thawing treatments support the above indications in that they are consistent with increased susceptibility of peroxisomes from clofibrate-treated mice to physical disruption. The sonication experiments served to indicate that a large pool of liver peroxisomal catalase is readily released by sonication, and that the size of this pool is similar in both untreated mice and clofibrate-treated mice. The

smaller catalase pools released by heat treatment therefore presumably represent part of a larger pool which is easily leaked through peroxisome membranes when these become damaged. The molecular basis of the change in the peroxisome membrane resulting from treatment with peroxisome proliferators which would account for these observations has not been defined, but the previous data of ours relating to a change in the membrane phospholipid composition would seem to be of relevance [12]. Indeed, the increased content of lysophosphatidylcholine and the decreased levels of phosphatidylcholine in these membranes may explain many of the resulted obtained. As an ampholytic detergent, lysophosphatidylcholine would be expected to readily permeabilize the peroxisome membrane, or make it more susceptible to permeabilization [13, 24] and this action would be more rapid at higher temperatures [13], conclusions born out by the experiments described in Fig. 6. The effects seen with Triton X-100 titration are of interest in this regard. Triton X-100 is a nonionic detergent which destroys membrane lipid bilayers by binding hydrophobic proteins and displacing membrane phospholipids. An altered membrane phospholipid composition in peroxisomes of drug-treated animals [12] may lead to a decreased binding of Triton X-100 to the membrane with the result that a higher concentration of this detergent would be necessary for membrane disruption, as observed in these studies. The altered protein profile of the peroxisome membrane in proliferator-treated mice [10] may also account in part for this difference in susceptibility to detergent extraction of peroxisomes from control and clofibrate-treated mice. It should be noted in this regard that the ratio of phospholipid to membrane protein is the same in peroxisomes from control and clofibrate-treated mice [12]. The experiments assessing the effect of lysophosphatidylcholine added to liver light mitochondrial fractions are consistent with the data seen for Triton X-100 extraction in demonstrating a significant difference in response of peroxisomes from control mice with those from clofibrate-treated mice, to the lytic action of this lipid. Again the sensitivity of

160 membranes to lysis by this lysolipid is dependent on both the phospholipid composition of the membrane and the membrane protein composition [13], both of which are modified following treatment of mice with clofibrate [10, 12]. In this regard, it should be appreciated that it is not possible to equate the permeabilizing effects of lysophosphatidylcholine distributed within membrane bilayers via normal cellular metabolic pathways (as seen, for example, following clofibrate-treatment) with that arising from exogenously added lysophosphatidylcholine as in the in vitro experiments described here, since in the latter case there may be restriction of incorporation and distribution of lysophosphatidylcholine to specific regions of the membrane lipid bilayer [13]. One further observation that may be drawn from this data however, is that, in contrast to the situation with peroxisomes from normal mice, the fractions from clofibrate-treated animals exhibited a less acute pattern of catalase release with increasing lysophosphatidylcholine concentration, a finding which may be construed as indicating greater heterogeneity of peroxisomes in these animals, at least in respect of membrane properties. The results associated with osmotic rupture of peroxisomes in fractions from control and clofibrate-treated animals, indicating as they do a tendency of the proliferated peroxisomes to be less susceptible to damage in hypertonic sucrose, may also be interpreted in terms of modified membrane permeability properties. Previous data, for example, have indicated that peroxisomes are entirely accessible to sucrose, and these organelles display an absence of osmotic space in sucrose media [18]. The membrane rupture resulting when these organelles are shifted from hypertonic sucrose to a solution of lower sucrose concentration has nevertheless been suggested to be due to water penetrating the sucrose-loaded peroxisomes much faster than sucrose can leave them [18]. If, as has been suggested above, the membranes of peroxisomes from clofibrate-treated mice are more permeable, then sucrose efflux may be marginally faster than with control peroxisomes, with a resultant decrease in the relative rate of water infux, and consequently the degree of membrane rupture.

Overall, these data provide indications of significant modification of the permeability and stability properties of membranes of proliferated peroxisomes. The proliferated organelles are indicated as being more susceptible to damage in vivo, a conclusion which accords with the many reports of damage to proliferated peroxisomes during isolation and manipulation [10, 14, 15, 17, 18, 21]. The long-term consequences of such membrane changes in the living animal are not yet fully defined, although some studies have impinged on this question. In particular, work by Reddy and coworkers [25, 26] has led to the postulate that long-term exposure of rodents to peroxisome proliferators leads to the development of hepatocellular carcinoma, and this effect has been ascribed to oxidative damage resulting from hydrogen peroxide escaping from induced peroxisomes. Such a postulate is not inconsistent with the membrane permeability changes described in this paper. In conclusion, a comment on the nature of the peroxisomal compartment in liver, and the mechanism of proliferator-induced peroxisome biogenesis, is warranted. The current model of peroxisome biogenesis which has gained wide acceptance predicts that new peroxisomes form from pre-existing peroxisomes into a reticulum [27]. The catalase 'pools' of differing magnitude which are seen to be released following the treatments employed in the present studies, however, are not readily explained in terms of a single homogenous peroxisomal reticulum. These observations may, of course, be related to variability in the structure of the reticulumper se, and it should be noted in this regard that this structure is as yet ill-defined. The results described in our study, which indicate differences in the membrane stability properties of peroxisomes from drug-treated animals, provide a means of examining the temporal changes in the peroxisomal compartment in livers undergoing peroxisomal proliferation, and would seem to allow the derivation of further relevant information on the reticulum model. Such studies are presently in progress.

161

Acknowledgement The receipt of financial support from the National

13.

Health and Medical Research Council of Australia is g r a t e f u l l y a c k n o w l e d g e d .

14.

References

15.

1. Hess R, Staubli W, Riess W: Nature of the hepatomegalic effect produced by ethyl-chlorophenoxyisobutyrate in the rat. Nature 208: 856-858, 1965 2. Reddy JK, Krishnakantha TP: Hepatic peroxisome proliferation: induction by two novel compounds structurally unrelated to clofibrate. Science 190: 787-789, 1975 3. Leighton F, Coloma L, Koenig C: Structure, composition, physical properties and turnover of proliferated peroxisomes. A study of the trophic effects of Su-13437 on rat liver. J Cell Biol 67: 281-309, 1975 4. MoodyDE, Reddy JK: The hepatic effects ofhypolipidemic drugs (clofibrate, nafenopin, tibric acid, and Wy-14,643) on hepatic peroxisomes and peroxisome-associated enzymes. Am J Pathol 90: 435-445, 1978 5. Osumi T, Hashimoto T: Enhancement of fatty acyl-CoA oxidising activity in rat liver peroxisomes by di-(2-ethylhexyl)phthalate. J Biochem 83: 1361-1365, 1978 6. Lazarow PB, de Duve C: A fatty acyl-CoA oxidising system in rat liver peroxisomes; enhancement by clofibrate, a hypolipidemic drug. Proc Natl Acad Sci USA 73: 2043-2046, 1976 7. Ganning AE, Brunk U, Dallner G: Effects of dietary di(2ethylhexyl)phthalate on the structure and function of rat hepatocytes. Biochim Biophys Acta 763: 72-82, 1983 8. Flatmark T, Christiansen EN, Kryvi H: Polydispersity of rat liver peroxisomes induced by the hypolipidemic and carcinogenic agent clofibrate. Eur J Cell Biol 24: 62-69, 1981 9. Horie S, Ishii H, Orii H, Suga T: Changes in membrane surface properties of hepatic peroxisomes of rats under several conditions as determined by partition in aqueous polymer two-phase systems. Biochim Biophys Acta 690: 74-80, 1982 10. Crane DI, Chen N, Masters CJ: Changes to the integral membrane protein composition of mouse liver peroxisomes in response to the peroxisome proliferators clofibrate, Wy-14,643 and di(2-ethylhexyl)phthalate. Mol Cell Biochem 81: 29-36, 1988 11. Hayashi H, Nakata K, Hashimoto F: Study on membrane fluidity of liver peroxisomes. In: HD Fahimi, H Sies (eds) Peroxisomes in Biology and Medicine. Springer-Verlag Berlin, Heidelberg, 1987, pp 205-209. 12. Crane DI, Masters CJ: The effect of clofibrate on the

16.

17.

18. 19.

20.

21. 22.

23.

24. 25.

26.

27.

phospholipid composition of the peroxisomal membranes in mouse liver. Biochim Biophys Acta 876: 256-263, 1986 WeitzienHU: Cytolyticandmembrane-perturbingproperties of lysophosphatidylcholine. Biochim Biophys Acta 559: 259-287, 1979 Klucis E, Crane D, Masters C: Sequential alterations in the micro-localization of catalase in mouse liver after treatment with hypolipidemic drugs. Mol Cell Biochem 65: 73-82, 1985 Hemsley A, Pegg M, Crane D, Masters C: On the compartmentalization of catalase, fatty acyl-CoA oxidase and urate oxidase in mammalian livers, and the influence of clofibrate treatment on this microlocalization. Mol Cell Biochem 83: 187-194, 1988 Chen N, Crane DI, Masters CJ: Analysis of the major integral membrane proteins of peroxisomes from mouse liver. Biochim Biophys Acta 945: 135-144, 1988 Crane DI, Hemsley AC, Masters CJ: Purification of peroxisomes from livers of normal and clofibrate-treated mice. Anal Biochem 148: 436-445, 1985 Baudhuin P: Liver peroxisomes, cytology and function. Ann NY Acad Sci 168: 214-228, 1969 Holmes RS, Masters CJ: Epigenetic interconversions of the multiple forms of catalase in mouse liver. FEBS Lett 11: 45-48, 1970 Peterson GL: A simplification of the protein assay of Lowry et al which is more generally applicable. Anal Biochem 83: 346--356, 1977 Alexsoo SEH, Fujiki Y, Shio H, Lazarow PB: Partial disassembly of peroxisomes. J Cell Biol 101: 294-305, 1985 Masters0CJ, Holmes RS: Peroxisomes: New aspects of cell physiology and biochemistry. Physiol Rev 57: 816-882, 1977 Poole B: Diffusion effects in the metabolism of hydrogen peroxide by rat liver peroxisomes. J Theor Bio151: 149-167, 1975 Helenius A, Simons K: Solubilization of membranes by detergents. Biochim Biophys Acta 415: 29-79, 1975 Reddy JK, Rao MS, Moody DE: Hepatocellular carcinomas in acatalasemic mice treated with nafenopin, a hypolipidemic peroxisome proliferator. Cancer Res 36: 12111217, 1976 Reddy JK, Lalwani ND, Reddy MK, Qureshi SA: Excessive accumulation of autofluorescent lipofuscin in the liver during hepatocarcinogenesis by methyl clofenapate and other hypolipidemic peroxisome proliferators. Cancer Res 42: 259-266, 1982 Lazarow PB, Fujiki Y: Biogenesis of peroxisomes. Ann Rev Cell Biol 1: 489-530, 1985

Address for offprints: D.I. Crane, Division of Science and Technology, Griffith University, Brisbane, Australia 4111