J. Lipid Mediators Cell Signalling 15 (1997) 249-254
Effects of eicosanoids on parameters in isolated rat hepatocytes and isolated rat hepatocyte couplets: protective effects of eicosanoid receptor antagonists Sue M. Hutchinson”,
Tim G. Hammondb, Alan A. Horton”,”
~‘&‘hool of BiochemistrJ~, The Uniwrsity c~f Birmingham, Edgbaston, Birmingham B15 2TT, hAstra Sqfety Assessment, Astra Charmwad, Loughborough, Lrics LEI I ORH. lJK
Received 31 July 1996; accepted 16 September 1996
Abstract A number of eicosanoids caused plasma membrane blebbing in hepatocytes and this could be inhibited in a dose-dependant fashion by the receptor antagonists AH6809 and ICI 192605. The pattern of effectiveness of eicosanoids interfering with canalicular vacuole accumulation in hepatocyte couplets differed from that causing blebbing; the two most effective eicosanoids here were PGD, and PGF2,. Copyright 0 1997 Elsevier Science B.V. Keyword.s: Antagonists; Hepatocytes
1. Introduction The
of animal cells is crucially dependent on the maintenance of a low free Ca2+ concentration ( z 0.2 PM). Disruption of Ca2+ homeostasis in
0929.7855/97:‘$17.00 Copyright 0 1997 Elsevier Science B.V. All rights reserved PII SO929-7855(96)00560-3
S.M. Hutc~hinson et al. / J. Lipid Mediators
15 (1997) 249-254
the liver can occur as a consequence of ischaemia, or due to the administration of any one of a substantial number of toxic drugs and chemicals (eg paracetamol (Jepson et al., 1987), Ccl,, trrt-butylhydroperoxide and some benzoquinones). Some of these substances, in common with Ca*+ ionophores, have been shown to elicit the formation of plasma membrane blebs. Plasma membrane blebs are recognised as a characteristic feature of toxic or ischaemic cell injury (Trump et al., 1980). In addition to bleb formation, an increase in the cytosolic free Ca*+ concentration stimulates the activity of enzymes such as glycogen phosphorylase, phospholipase A, (PLA,) and non-lysosomal proteinases (Popper and Keppler, 1986). Increased PLA, activity releases arachidonic acid from cell membrane phospholipids which then stimulates eicosanoid synthesis. Although enhanced eicosanoid formation does not necessarily imply a role of those compounds in toxicity, many tissues respond to chemical injury by generating eicosanoids and it is only during injury that high levels can be measured. Some eicosanoids may have cytotoxic properties, for example, a carbocyclic thromboxane analogue has been shown to induce the release of lysosomal enzymes from large granule fractions of the liver and several diseases have been related to an imbalance in the prostacyclin/thromboxane A, (TXA,) ratio. Paracetamol-induced hepatotoxicity in both mice and rats causes the generation of TXB, and prostaglandin Ez (PGE,); protection from toxicity is afforded by inhibitors of TX and PG synthesis (Horton and Wood, 1989a) and by a TX receptor antagonist (Sulotroban BM 13 177). The generation of eicosanoids in liver as a consequence of the disruption of Ca* + homeostasis by toxic substances, and the protection afforded by inhibitors of enzymes of the arachidonic acid cascade suggests that some eicosanoids may act as intermediaries in Ca2+ -Induced cell injury and death. Isolated hepatocytes incubated with certain eicosanoids (e.g. TXB,) formed plasma membrane blebs at concentrations similar to those measured in vivo in animals with paracetamol-induced hepatotoxicity (Horton and Wood, 1989b). In preliminary experiments, TX receptor antagonists afforded protection suggesting that exogenous TXB, initiates a sequence of events leading to bleb formation by binding to a putative receptor on hepatocytes (Horton and Wood, 1991). This paper reports on experiments designed to widen this investigation by use of (i) other eicosanoids (e.g. PGF,,, PGD,, PGE, and U46619, a stable TXA, analog), and (iii) by investigation of the effects of (ii) different receptor antagonists, eicosanoids not only on bleb formation, but on a further more physiological parameter, canalicular accumulation, using hepatocyte couplets.
Collagenase (Clostridiurn lzistolyticum) was obtained from Boehringer Mannheim (Lewes, Sussex). Ketamine hydrochloride (Vetalar) was supplied by Parke Davis (Pontypool, Gwent) and medetomidine hydrochloride (Domitor) by Smith Kline Beecham (Tadworth, Surrey). ICI 192605, originally synthesised as a TP receptor
S.M. Hutchinson et cd. / J. Lipid Mediators Cell Signalling 1.5 (1997) 249-254
antagonist, was a gift from Zeneca (Macclesfield, Cheshire) and AH6809 (an EP,, DP, TP receptor antagonist) was a gift from Glaxo Group Research (Ware, Hertfordshire). All biochemicals were purchased from Sigma (Poole, Dorset) and other reagents, of the highest grade available, were obtained from Fisons (Loughborough). Male Wistar rats (230-280 g) from Biomedical Services of this University were used under deep anaesthesia for the isolation of hepatocytes as described previously (Horton and Wood, 1989b) except that cells were resuspended in Leibovitz’s medium (L15). Hepatocyte couplets were obtained by the method of Wilton et al. (1991). Viability was determined by Trypan blue exclusion; mean viability was 85% (singlets) and greater than 90% (couplets). The cells were quantified using a Neubauer haemocytometer. Hepatocytes (approx 1 x 10’ cells/ml of L15) were plated 2 ml/30 mm plastic petri dish (Corning, sterile, tissue culture coated dishes) and incubated under atmospheric conditions at 37°C for 4 h. This incubation period allowed the couplets to attach to the dishes and reform their canalicular structure as a vacuole. Eicosanoids and ICI 192605 were dissolved separately in ethanol in a volume which ensured that the final ethanol concentration in the incubation did not exceed 0.1% AH6809 was dissolved in 0.025% NaHCO, solution (final concentration). The solutions were stored at - 20°C for a maximum period of 6 weeks. ICI 192605 or AH6809 was added to the singlet hepatocyte incubations after 4 h and eicosanoids 15 min later at concentrations given in the figure legends. After 30 min, plasma membrane bleb formation was assessed by light microscopy using an inverted microscope (Zeiss Axiovert 10). For the determination of effects on canalicular vacuolar accumulation, test substances were added to the dishes after a 4 h initial incubation and then incubated for the times indicated in the figure legends. Fifteen min before the end of the incubation 2 ~11of cholyl-lysyl-fluorescein (CLF) (Mills et al., 1991) (2 mg/ml normal saline) or vehicle was added to each dish. At the end of the incubation, excess CLF-containing medium was removed and replaced with L15 (2 ml). The couplets were washed with a further 2 ml of L15 and then examined by fluorescence microscopy (Zeiss Axiovert 10). The number of couplets showing canalicular vacuole accumulation was recorded as a percentage of the total number of couplets (normally 40 to 50) observed. All values in the text and figures are the mean + S.D. Statistical analysis of the data was performed using Student’s t-test. Differences between values were considered significant if P -c 0.05.
3. Results and discussion Fig. 1 shows the effect of different concentrations of ICI 192605 and AH6809 (receptor antagonists) on the formation of plasma membrane blebs in plated isolated hepatocytes treated with either U46619, PGD,, PGF,,, or PGE, at the concentrations shown. Each of the eicosanoids caused bleb formation in the range
et al. /(/J. Lipid Mrdiators
Cell Sigrzalling 15 (1997) 249-254
of 38-76X of cells, significantly greater than the controls (usually 14422X, not shown in the diagram). Eicosanoids were studied at concentrations up to 40 ,LIM. These higher concentrations are probably non-physiological and blebbing therefore represents a pathological or toxicological response to elevated eicosanoid levels. Concentrations at least as high as 4 ,LIM may exist in the liver and concentrations in the higher range have been used to study the effect of eicosanoids on carbohydrate metabolism in isolated hepatocytes and perfused livers. In Fig. 1 the concentrations of eicosanoids used were selected to yield a similar degree of plasma membrane bleb formation with each compound. The TP receptor antagonist ICI 192605 (Brewster et al., 1988) gave very strong protection when used at 100 ,LLMin all cases but there were variations in the extent of protection provided at lower concentrations. For example, at 10 ,LLM ICI 192605, bleb formation induced by differed according to the eicosanoid used, but in all U46619, PGE, and PGF,,, cases the numbers of blebs were significantly less than in the absence of receptor antagonist.
Fig. I. Inhibition of eicosanoid-induced blebbing by AH6809 (upper graphs) or ICI 192605 (lower graphs). (a) Comparison of the effect of AH6809 (designated as A on s-axis) on the blebbing induced by II46619 (40 /LM) and PGD, (40 /tM, 30 min incubation on cells plated for z 4 h). Results expressed as a percentage of the blebs induced by the eicosanoid alone (?I = 3 5, conditions as described in Section 2). Statistics: Student’s t-test relative to eicosanoid = 100%. *P < 0.05, **P < 0.01 and ***P < 0.001. (b) Comparison of the effect of AH6809 (designated as A on s-axis) on the blebbing induced by PGEz (20 p M) and PGF,, (20 /i M). Results expressed as a percentage of the blebs induced by the eicosanoid alone (n = 3). (Conditions as for (a)). (c) Comparison of the effect of ICI 192605 (designated as ICI on r-axis) on the blebbing induced by U46619 (40 /IM) and PGD, (40 PM). Results expressed as a percentage of the blebs induced by the eicosandid alonc (n = 4, conditions as described in Section 2). (d) Comparison of the effect of ICI 192605 (designated as ICI on _r-axis) on the blebbing induced by PGE, (20 ,uM) and PGF,, (20 /AM). Results expressed as a percentage of the blebs induced by the eicosanoid alone (n = 3, conditions as for (c)).
S.M. Hutc~lzinsotzet al.
/)J. Lipid Mediators
I5 (1997) 249-254
20 Eicosanoid concentration (PM) Fig. 2. The effect of a range of eicosanoids on the ability of hepatocyte couplets to accumulate, into their canalicular vacuole, the fluorescent bile salt analogue cholyl-lysyl-fiuorescein (CLF). Statistics: Student’s t-test relative to control accumulation 100% (n = 3-12, *P < 0.05, **P < 0.01 and ***Pi 0.001).
Similarly, the receptor antagonist AH6809 (Coleman et al., 1985) showed potent protection at 100 PM for U46619, PGD, and PGE,, but again there were variations in the extent of the significant protection afforded at lower concentrations. As a general rule for both receptor antagonists, their protective effects were concentration dependent. These data confirm and extend previous observations made in this laboratory on the cytotoxicity of some prostaglandins and thromboxanes (Horton and Wood, 1989a) and on the protective effects of certain TX receptor antagonists. The protective effects of receptor antagonists on cells exposed to prostaglandins and a thromboxane analogue suggest that receptors are involved in mediating the effects of the eicosanoids used but their molecular mechanism of toxicity remains obscure. Fig. 2 shows the effects of a range of different eicosanoids on the ability of hepatocyte couplets to accumulate the fluorescent bile salt analogue, cholyl-lysylfluorescein (CLF) into their canalicular vacuole. PGD, and PGF,, produced a marked decrease in canalicular accumulation at 20 and 40 /IM whereas U46619, PGE, and TXB, caused only a small decrease at 40 PM. Canalicular vacuolar accumulation (CVA) is the end result of a number of processes: uptake of the bile acid analogue, its transport across the cell, its secretion across the canalicular membrane and its retention in the vacuole due to the integrity of the tight junctions; it also requires an intact, fully functioning cytoskeleton (Coleman et al., 1995). Clearly the various eicosanoids/analogues could be affecting any component part of the overall process. In two other instances of interference with CVA (by menadione and by cyclosporin A) disorganization of the cytoskeleton with possible influence on tight junctional integrity has been implicated
S. M. Hutchinson
I J. Lipid Mediators Ceil SignaIling I.5 (1997) 249 -254
(Coleman et al., 1995). Since membrane blebbing also involves cytoskeletal malfunction a cytoskeletal/tight junctional target for eicosanoid-induced CVA reduction is an interesting possibility.
Acknowledgements We thank MRC/Astra for a grant (to RC).
for a CASE
(to SH) and the Wellcome
References Brewster, A.G., Brown, G.R., Foubister, A.J., Jessup, R. and Smithers, M.J. (1988) The synthesis of a novel thromboxane receptor antagonst 4(z)-6-2-0-chlorophenyl-4-0-hydroxyphenyl-l-l,3-dioxan-cis5-yl hexenoic acid ICI 192605. Prostaglandins 35, l73- 178. Coleman, R.A., Kennedy, L. and Scheldrick, R.L.G. (1985) AH6809. A prostanoid EP, receptor blocking drug. Br. J. Pharmacol. 85 273P. Coleman, R., Wilton, J.C., Stone, V. and Chipman, J.K. (1995) Hepatobiliary function and toxicity in vitro using isolated hepatocyte couplets. Gen. Pharmacol. 26, 1445-1453. Horton. A.A. and Wood, J.M. (1989a) Effects of inhibitors of phospholipase AZ, cyclooxygenase and thromboxane synthetase on paracetamol hepatotoxicity in the rat. Eicosanoids 2. l23- 129. Horton, A.A. and Wood, J.M. (1989b) Prevention of Ca’+ -Induced hepatocyte plasma membrane bleb formation by inhibitors of eicosanoids synthesis. J. Lipid Mediators 1, 231-242. Horton, A.A. and Wood, J.M. (1991) Prevention of Ca” -induced or thromboxane Bz-induced hepatocyte plasma membrane bleb formation by thromboxane receptor antagonists. Biochim. Biophys. Acta 1133, 31-37. Jepson, M.A., Davis, M.J.H., Horton, A.A. and Walker, D.G. (1987) Histochemical and biochemical observations on the cytotoxicity of paracetamol and its effects on glycogen metabolism in rat liver. Toxicology 47, 325-337. Mills, C.O., Rahman, K., Coleman, R. and Elias. E. (1991) Cholyl-lysyl-fluorescein (CLF) synthesis, biliary excretion in vivo and during single pass perfusion of isolated perfused rat liver. Biochim. Biophys. Acta 1 I 15, I51 156. Popper, H. and Keppler, D. (1986) Networks of interacting mechanisms of hepatocellular degeneration and death. In: H. Popper and I. Schaffner (Eds.), Progress in Liver Diseases. Vol. 8, Grune and Stratton, Orlando, FL. pp. 209-235. Trump, B.F., Berezesky, I.K., Laiho, K.U., Osornio. A.R., Mergner, W.J. and Smith, M.W. (1980) The role of calcium in cell injury; a review. In: A.M.F. O’Hare (Ed.), Scanning Electron Microscopy, Vol. 2, SEM Inc, Chicago, IL, pp. 437-462. Wilton, J.C., Williams, D.E., Strain, A.J., Parslow, R.A., Chipman, J.K. and Coleman, R. (1991) Purification of hepatocyte couplets by centrifugal elutriation. Hepatology 14, 180- 183.