- Email: [email protected]
Modulation of paracetamol metabolism by Kupffer cells: A study on rat liver slices
PI1 SOO24-3205(99)00554-S
ELSEVIER
MODULATION
Life Sciaoeq Vol. 65, No. 26, pp. 2851-2859, 1999 copy@lr01999Elscvi~sciilnc. RhfcdinUwUSA. Allri&tsmsrmd 0024_3205/‘99/hee fimt maltu
OF PARACETAMOL METABOLISM BY KUPFFER A STUDY ON RAT LIVER SLICES
CELLS
:
Neyrinck A.‘, Eeckhoudt S.L.1, Meunier C.J.1, Pampfer S.2, Taper H.S.1, Verbeeck R. K.1 and Delzenne N.1’ tuniversite Catholique de Louvain, Unite de Pharmacocinetique, Mbtabolisme, Nutrition et Toxicologic, UCL-PMNT 7369, 73 Avenue Mounier, B-1200 Brussels, Belgium 2Universite Catholique de Louvain, Unite de Recherche en Physiologie de la Reproduction, UCL-OBST 5330, 53 Avenue Mounier, B-1200 Brussels, Belgium (Received in final form August 16, 1999)
Summary
Recent studies support the hypothesis that non parenchymal cells (mainly macrophages) may play a role in the metabolism and cellular effects of paracetamol. In order to investigate this hypothesis, male Wistar rats were intravenously injected with either 7.5 mg/kg gadolinium chloride (Gd+) or NaCl 0.9% (Gd-). The treatment with GdCls decreased the number and the function of Kupffer cells in liver tissue, as assessed by the histological examination of the liver after colloidal carbon injection in the portal vein. Precision-cut liver slices (PCLS) were prepared from both groups of rats and cultured for 8h in Waymouth’s medium in the presence and absence of 5 mM paracetamol. Interestingly, PCLS obtained from Gd+ rats exhibited a lower release of tumor necrosis factor (TNF-a) and a better viability than PCLS from control (Gd-) rats. Incubation with paracetamol led to a decreased glycogen level in liver slices from Gd+ or Gd-, without modifying neither liver morphology nor ATP level nor LDH release. A higher proportion of paracetamol glucuronide, was secreted from the slices obtained from Gd+ rats. These data suggest that Kupffer cells could affect the viability of PCLS in culture and are involved in the regulation of phase II metabolism in the adjacent hepatocytes. We propose that PCLS in culture is a suitable model to elucidate the biochemical mechanism underlying the modulation of metabolism occurring through hepatocytes-Kupffer cells interactions. Key Words: gl~nidation,
KupEer cells, glycogen, liver slices
The role of sinusoidal cells in the hepatic metabolism has been greatly underestimated until now. However, Kupffer cells, despite their small size represent 80 to 90% of all fixed macrophages in the body and approximatively 14% of the hepatic cellular mass (1). Kupffer cells have the capacity to express typical macrophage functions such as * To whom correspondance should be addressed
2852
Kupffer Cells and Hepatic Effects of Pamcetamol
Vol. 65, No. 26, 1999
phagocytosis and secretion of a wide array of biologically active molecules, allowing them to quickly clear the circulation from cellular residues and microorganisms. Activated Kupffer cells release different mediators including cytokines such as TNF-a, reactive oxygen intermediates such as O2.-l lysosomal enzymes or eicosanoids, which are involved in their cytotoxic and inflammatory processes (1, 2). Such mediators are also capable to disturb both the metabolism or the integrity of adjacent hepatocytes (3, 4, 5). Moreover some authors have suggested that Kupffer cells could play a role in xenobiotic-induced hepatotoxicity. The hepatotoxicity of xenobiotics is often dependent on their metabolism. Although the hepatocytes are the major site of xenobiotic metabolism, several enzymatic activities such as GSH-transferase, UDPglucuronosyltransferase and cytochrome P450-dependent oxidase have been found in non parenchymal cells (6). The aim of this work was therefore to assess in vitro, by using the original model of precision-cut liver slices (PCLS), the influence of Kupffer cells on cellular integrity and on the liver capacity to metabolize a model compound, paracetamol, a well known mild analgesic and antipyretic agent. As compared to cocultures of hepatocytes and Kupffer cells, the model of PCLS preserves the tissue architecture, the proportion of the different cell types, and the polarity of hepatic cells (7). Moreover, it is highly representative of the metabolic pattern observed in w’vo and of the susceptibility of the liver to paracetamol hepatotoxicity. Paracetamol may be metabolized in the liver by several mixed function oxidase systems, dependent on cytochrome P450 isoforms (CYP2E1, CYPlAl and CYPlA2, CYPPCll in rats and CYP2E1, CYP3A4 and CYPlA2 in man) leading to the production of a highly reactive electrophilic and toxic metabolite : the n-acetylparabenzoquinoneimine (NAPQI); this NAPQI may be detoxified by conjugation to glutathione. Paracetamol metabolism in rats is characterized by the relatively low capacity to metabolically activate paracetamol towards its toxic metabolite as well as a high capacity to clear paracetamol via non toxic glucuronidation and sulphation pathways (8,9). The role of Kupffer cells was examined by comparing the results obtained in PCLS prepared from the liver of rats pretreated with saline (controls) or gadolinium chloride (GdC13) , a specific toxin for Kupffer cells which not only blocks phagocytosis by rat liver macrophages but alS0 selectively eliminates these cells situated in the periportal zone of the liver acinus (10). Material and Methods Animals and treatments: Adult male Wistar rats from lffa Credo (Les Oncins, France) weighing 283 f 8 g were housed in a temperature- and light-controlled room (12 h dark/light cycle). They were provided A04 standard diet (UAR, Villemoisson-sur-Crge, France) and tap-water ad libitum. GdC13 (7.5 mg/kg or 20 mg/kg), dissolved in Saline, was administered intravenously to rats 20 h before the injection of colloidal carbon Or the preparation of liver slices. To test for the phagocytic capacity of Kupffer cells, colloidal carbon (Pelikan@ no17 dissolved in saline 1:lO; 1.5 ml/kg body weight) WaS injected intravenously 20 min before liver sampling for further histological study. PCLS preparation and incubation: Rat surgical procedures were carried out under pentobarbital (60 mg/kg) anaesthesia. PCLS (about 220 pm thickness) were prepared with a Krumdieck slicer according to a procedure previously described (11, 12). After 60 min of preincubation at 4°C in Krebs-Ringer solution, PCLS were rinsed with saline and suspended in Waymouth medium supplemented with 0.3% of BSA, 1% Of glutamine solution, I % of penicilline-streptomycin solution, glucose (28 mM), insulin (100 nM) +/- paracetamol (5 mM). The vials, containing ITMXitWfII 4 dices (2 ml medium per slice), were saturated with a mixture of 95%02-5%COa and placed in a shaking water bath at 37°C for 4 to 8 hours.
Vol. 65, No. 26, 1999
Kupffer Cells and Hepatic FiiTem ofParaaamo1
2853
LDH release: The viability of liver slices was estimated by measuring the activity of lactate dehydrogenase (LDH) in the culture medium, according to the procedure of Wrobleski and Ladue (13). The results are expressed as international units (IU) of LDH per milligram of proteins. The amount of protein was determined by the method of Lowry (14) using BSA as standard. ATP content: Liver slices were taken, washed twice in saline and sonicated in 1 ml of 2% perchloric acid. The intracellular ATP content was measured on neutralized perchloric acid extracts by using an HPLC-LKB Pharmacia instrument equipped with a 4.7 x 125 mm (particle size 5 urn) anion-exchange column (Partisphere SAX, Whatman). Separation was achieved by the use of an isocratic elution mode with 0.45 M NH4H2PO4 (pH 3.7) at a flow rate of 1.5 ml/min. The results are expressed as nmol/mg protein. G/ycogen content: Liver slices were taken, washed twice in saline and sonicated in 1 ml of 1 M KOH. They were further heated at 100°C for 10 min. After neutralization with acetic acid and centrifugation, the supernatant was incubated in the presence of a-amylo glucosidase in pH 5 acetate buffer; the glucose produced was quantified by an enzymatic reaction as previously described (15). ffistological study.’Liver sections obtained from rats injected with colloidal carbon were fixed with 10% buffered neutral formalin and then embedded in paraffin, sliced into 3-4 pm sections, to assess by microscopic analysis of the phagocytic activity of Kupffer cells. Quantification of T/V/=-aproduction: TNF-a level was determined with an ELISA kit for quantification of rat TNF-a (Factor-Test-X” Rat TNF-a Kit, code 80-380501, Genzyme) in frozen aliquots of incubation medium (50 ~1). The results are expressed as pg/mg protein. Paracetamol metabolism: Paracetamol glucuronide, paracetamol sulphate and paracetamol glutathione conjugates were quantified by using reverse-phase HPLC according to the procedure of Lau and Critchley (16). Aliquots of incubation medium were stored at -30°C for further analysis. The proteins were precipitated in HC104 1% which contained the internal standard 2-acetamidophenol (40 ug/ml). After centrifugation, samples were injected and separated on a Nova-Pak Cl8 column by using an isocratic solvent system of 0.1 M KH$O@. 1% acetic acid/0.75% propane-2oi at a flow rate of 1.5 ml/min. The retention times were 4 min for paracetamolglucuronide, 10 min for paracetamol-sulphate, 12.8 min for paracetamol, 20.2 min for paracetamol-glutathione conjugate and 24.8 min for internal standard. The results were expressed as pg metabolitelmg protein. Statistical analysis: Results are expressed as mean+/- standard error of the mean. An unpaired student-t test or a one-way anova followed by a Scheffe test were performed for the statistical comparison of the results obtained in the different groups (Statview@). The actual test chosen is specified at the bottom of each figure or table. Results and Discussion The blockade of phagocytosis and the selective elimination of macrophages by i.v. injection of gadolinium chloride (GdCIs), is a generally accepted procedure for gaining knowledge about the function of macrophages in vivo (10). The injection of GdCls at a dose of 7.5 mgikg (Fig. 1.B) resulted in an appreciable decrease of Kupffer cell number and activity in the whole parenchyma compared to control rats who were
2854
Kupffer Cells and Hepatic ESects of Pama%mol
Vol. 65, No. 26, 1999
injected with 0.9 % NaCl (Fig. l.A). A complete disappearance of Kupffer cells COIJld not be obtained even by increasing the dose of GdCls to 20 mg/kg body weight. At tl7is higher dose, some necrotic foci were present (data not shown). Therefore, the dose of 7.5 mg/kg was used for further experiments. In order to assess the contribution of Kupffer cells to hepatic metabolism, PCLS were prepared from Gd- or Gd+ rats, a nd kept in suspension in an oxygenated medium (02K;O2 95/!5)for 4 or 8 hours.
1 .A.
Fig. 1 Influence ot GdCb InjectIon on the colloldrl carbon uptake by Kupffw cells. Distribution of carbon in liver 20 min after intravenous injection 1 day after GdC13 injection (1 .B.) or saline injection (1 .A.). The carbon is mainly taken up in Kupffer ceils (dark spots). x1200
Vol. 65, No. 26, 1999
Kupffer Cells and H-tic
Effects of Paracetamol
2855
The synthesis of TNF-a has been shown to occur in both hepatocytes and Kupffer cells after exposure to toxins or endotoxins (17, 18). As shown in Fig. 2, the release of TNF-a in the medium occurs without addition of xenobiotics or toxins. PCLS prepared from the liver of both Gd- or Gd+ rats released the same amount of TNF-a in the medium after 4 hours of incubation; however, a further secretion of TNF-a was only seen with slices obtained from Gd- rats. The production of TNF-a occurring from 4 to 8 hours of incubation may thus be attributed to Kupffer cells. The release of TNF-a after 4 hours of incubation, observed even in Gd+ animals, could be explained by the facts that either 1) some Kupffer cells escaped the effect of GdCl9, or 2) TNF-a has been produced by other ceil types present in PCLS, namely hepatocytes and endothelial cells. This later hypothesis is supported by the fact that in the absence of stimulation, endothelial cells were found to spontaneously produce several cytokines, namely IL-l and IL-6 within 1 or 2 hours of incubation; such cytokines may thus directly have an effect on TNF-a production by hepatocytes (17, 19).
Time (hour) Flg. 2 Influence of GdClg InjectIon prior to PCLS preparation on the release of TNFn In the medium after 8h of Incubation. Amountof TNF-a releasedin the incubationmediumof PCLS obtained from rats having received, 20h before to PCLS preparation, an i.v. injectioneither of NaCl 0.9% (Gd-, k4) or of GdC13 7.5 mgkg (Gd+, n=4). One way ANOVA test: p-c 0.05.
osp put? ‘Sap!UOJnOnlfj lOwela%Jed 40 lunowe ~a46!q a : ~Ow~~~k?Jed 40 se~!~oqe~ew 11 aseyd $lJOJ@44!p941 0OnpOJd 01 Sl3d 40 &OedleO aql pe!4!pOw “13pg hq SISJ 40 lUew~B3JlaJd au1 ‘E ‘614u! UMOys Sv ‘lOwe$3)?J13d 40 -440 Je(nll80 pu0 wS!loqlgew au1 uo luewl&‘eJleJd-SI~pg 40 eOu8n(4U! pa!pnlS OSle eAt?y aM ‘r(pn$3 1uaseJdeyj ul
SA +loUW=SJsd 100’0 > d :lSSi %Wi%~ ‘(P = U) +PD Sl3d PUS (9 = U) -PE) S-lad 10 (u!aiOJd fk1J/6tf) NJe1UO9 uei!bOo&) ‘-ps SA +PE) 90.0 > d :&Sol queq3Ssq ‘(p = U) +PD Sl3d :t~at
PUe (9 = U) -P!3 S-lad P uJn!P’JJ U! (u!mJd 6w-1) hl~W= Hal ‘-Pg SA +PE> 10’0 > d ?uaw& .(P = U) +Pg-S-I&I Pm (9 = U) -P’w313d 10 (U!eloJd Bu~/IOUJU) aUWJO3 dlV
38’c9 39’96
+
r B’OZ r
1
8’LL
2’8!Gl
f
1’8W
960’0 r L9’0 81’0 =FOZ.1
qSZ’0 * 08’0 PC’0 =F60’1
(srnoq
8)
+
9361 5: 9’SSB
eP’1 r 2’61 Z’l r E’PC
eL’1 r L’81 8’0 * L’&l -
lOltlS~S3SJSd suo(~(puoo
(wd
6w6rl)
ua6oOh16-anss!l +
+
(loJd Win) Ha-www (IoJd 6w/lowu) dlV-=ss!l
lOlIlSg338JSd uo~~Sqn3u~ 1
PD
SJSgSllJSJSd
318Vl
IJO!lleqnOU! S13d 6U!Jnp 6u!JJnOOO SJOlD34 O!xOlOtiO 40 aSeale~ aql SSOSS)?01 JO/PUB UO!leA!lO~ SllRI Jaydn)l 40 WS!UeqOaW ayl alep!On(a 01 h??SSQOau aJB s&Jaw!Jsdxa JaylJn4 ‘sesaylotiy asaql alE!&Jelsqns 01 slauueq0 wnplle0 pue aseue6r(xoo~3k1 BU!A~OAU! ws!ut?gOaw e r(q IClq!ssod ‘slla0 IewhqOuaJed u! uogdwnsuo0 ua6hxo Sa$l?Aala 43!4M ‘z=Jgd ey!l SJO#!paW t3S~C3laJ10 40 GOueseJd ayl U! SllaO JaUdn)l Ik?ql UMOUS OSle GARY (Sz) /t? Ja Ilo ‘(pz ‘cz) 3 aWOJl&)OflO jQJpUOl@Ol!W 40 suo!1eO!4!pow pue ayeadn ua6hxo JBMO( e hq paleO!pu! ‘J~A!I au1 u! wsyoqetaw ua6ho 40 uo!ge~n6aJ-umop e 01 spaal ~11%) Jalldn)l 40 uo!i!q!qu! S13pg ~leql BU!MOLJS ‘suogeglqnd IUKBJ hq pauoddns s! s!saqlodhy s!ql .uo!aeqnOu! aql 6u!Jnp uaMx0 pue slIJa!JvIu L(loq aJeds plnoM slla0 Ja44dn)l 40 p!o~ap Slgd
‘z
%?Z ‘lz) aSI3p!XO-&jaVN 40 UO!@A!138 Sql 01 0np aq pInO uouawouaqd S!yl ‘(C)z:) aJnaln0 u! slla0 Je44dn)l ~0~4 _,Q 40 UOgOnpOJd v~e0~4yti!s e 03 speel uo~~eus6rotoa~ 01 e!xodh./ WOJ4 UO!l!SUEJ$ aql $1?y1u~oys uaeq SBL/ $! ‘1324 ul ssflOo)eday SpJeMOl O!XolO~O aq 01 UMOUy se!%ds aA!lOtBJ 40 Slla:, J944dn)l hq aSt3laJ ayl ’1 : e@p GU!MOIIO~alla r(q pauoddns s! s!saylodr(y s!41 ‘(i wn!paw au1 u! UO!l~J~lJ~3UO3 “0 y6!4 L) hq palowoJd) a6owep C3A!aI?p!XO 01 Way1 aZ!l!SUaS pInO Sl&/ u! SIlKI JaUdn] 40 aOuas0Jd aqa days esodoJd aM 3JnllnO u! S13d U! saaholedaq 40 &pedeO O!lOqle~aw JO/plJe r(t!l!qo!A ~StBJO0p pin00 slle0 Jalldny 40 aOuasaJd ayl ‘(sap!JeqOOedlodod!( .a.!) s~o~r34 Gu!gqnw!gs u~ouy 40 aOuesqe acli u! ue~e ‘iaqa lse66ns sNnsaJ Jng -1ue0!4y6!s IOU SAM ~ua~uo~ ue600~~6 uo lU8wleaJiaJd 3l# 40 33U@nl4U! all1 UO!~~qnOlJ! 40 SJnOq 8 JaUe ‘Wn!&leW ayl u! @3B(aJ HQl JaMol B pU)3 3U0luOO dlV JE?lnllaO J6L46!4 e U!e~U!leW 01 p&ViOllB uo!~ti%?deJd Sl3d 01 JO!Jd Sl3pg 40 UO!P&J! al41 ’1 a(@?1 U! UMOI#3 Sbj ‘S#?J -pg WOJ4 Sl3d au1 UeLil UO!leqnOU! eql 6u!Jnp panraseJd J~UCI~ 81E?sII?J +pg ~0~4 S13d ‘I~I~u!~s~~J@~uI
6461 ‘9Z “‘N ‘$9 ‘PA
9282
Vol. 65, No. 26, 1999
Kupffer tills and Hepatic Effects of Pawetamol
2857
a high production of glutathione conjugates (reflecting both activation by cytochrome P450 isoforms and conjugation to glutathione), was secreted in the medium of Gd+ PCLS. This observation suggests that Kupffer cells could play a role as regulators (and in the present case, as inhibitors) of xenobiotic metabolism in the hepatocytes. In vitro, it has been shown that stimulation of Kupffer cells by Clostridium perfringens decreases cytochrome P450 dependent activity, thus decreasing paracetamol hepatotoxicity in vivo (26). TNF-a and interleukin 9 have been proposed as mediators, as they decrease CYP gene expression. It may be important to note that in viva, GdCls injection paradoxically also decreases the total hepatic content of cytochrome P450 (27). No studies until now have reported an effect of GdCl3 injection on the activity and on the content of the specific CYP isoforms responsible for paracetamol metabolism; this could be interesting to measure, as a “paradoxical” induction of several CYP isoforms despite a lower total CYP content has already been described in some situations (28).
Paracetamol sulphate
Paracetamol glucuronide
GSH-conjugates
Fig. 3
Influence of GdCls injection on the excretion phaae
II-metabolites
in the medium
by PCLS of peracetamol alter 8h of incubation. Measurementof
glucorono-.sulfo- and glutathionconjugated-metabolites of paracetamolreleasedin the mediumaftersh of incubationof PCLS obtainedfrom rats havingreceived,20h beforeto PCLS preparation,an i.v. injectioneitherof saline0.9%(Gd-) or of GdCiS 7.5 mgkg (Gd+). Results are presented as mean f 8.s.m. (n ~6). Unpaired t-test Student: “p c 0.05 vs Gd- and ““p < 0.01 vs Gd-
We have thus shown that paracetamol metabolism in rat PCLS is characterized by a high sulphation and glucuronidation and a low production of cytochrome P450 and glutathione conjugates : as suggested earlier, it makes rat liver more resistant to paracetamol hepatotoxicity (6). Even after 8 hours of incubation in the presence of
2858
Kupffer Cells and Hepatic Effects of Ftaracetamol
Vol. 65, No. 26, 1999
paracetamol at a relatively high dose (5 mM), no sign of morphological alterations could be detected by histological examination (data not shown). ATP levels and LDH leakage by PCLS were not modified by the addition of paracetamol to the medium. However, the glycogen content dropped dramatically, both in Gd+ and Gd- PCLS An activation of glycogenolysis by paracetamol was described by several authors, both in vivo (29, 30) and in isolated hepatocytes. The most probable explanation would be the activation of glycogen phosphorylase a (31, 32). The decrease in glycogen in the presence of paracetamol was even more pronounced in PCLS from Gd+ than from Gd- rats. This difference could be explained by the fact that part of glucose moieties of glycogen is deviated towards the production of UDP-glucuronic acid, the co-substrate for glucuronidation of xenobiotics (33). In fact, Billiard et al (1) consider Kupffer cells and the adjacent hepatocytes as a functional unit capable of complex and tightly regulated interaction. The present study offers support for a relevant role of Kupffer cells in the control of hepatocyte metabolism under physiological conditions. The real implication of the modulation by Kupffer cells of drug metabolism could be confirmed by using another Kupffer cell inhibitor such as methyl palmitate (34). Some authors have already reported that unstimulated Kupffer cells in culture released enough prostaglandins to alter the phosphorylation of specific proteins inside the hepatocytes (35). We propose that PCLS in culture would constitute a useful model to study the physiological roles of Kupffer cells in xenobiotic metabolism. Acknowledgements This work was supported by a grant from the “Region Wallonne” in Belgium; References 4: 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.
BILLIAR T.R. and CURRAN R.D. J. Parent. Enter. Nutri. 1 4 175S-180s (1990) DECKER K. Eur. J.Biochem. 192 245261 (1990) LASKIN D.L. and PENDINO K.J. Ann. Rev. Pharmacol. Toxicol. 35, 655-677 (1995) LASKIN D. and PILARO A. Toxicol. Appl. Pharmacol. 86 204-215 (1986) LASKIN D., PILARO A. and JI S. Toxicol. Appl. Pharmacol. 86 216226 (1986) LAFRANCONI W.M., GtAlT H. and OESCH F. Toxicol. Appl. Pharmacol. 8 4 500511, (1986) BACH P.M., VICKERS A.E.M., FISHER R., BAUMANN A., BRETTEBO E., CARLILE D.J., LAKE BG., SALMON F., SAVOYER T.W. and SKILINSKI G. ATLA, 4 893-923 (1996) MILLER M.G., BEYER J., ADAMS P.E. Toxicol. Appl. Pharmacol. 122 108-116, (1993) PATTEN J.C., THOMAS P.E., GUY R.L., LEE M., YANG C.S. Chem. Res. Toxicol. 6 51 I-518 (1993) HARDONK M.J., DIJKHUIS F.W.J., HULSTAERT C.E. and KOUDSTAAL J. J. Leuk. Biol. 5 2 296-302 (1992) GOETHALS F., DEBOYSER D., LEFEBVRE V., DECOSTER I. and ROBERFROID M. (1990) Toxicol. in Vitro 4 435-438 (1990) GOETHALS F., DEBOYSER D., CAILLAU E. and ROBERFROID M. In vitro methods in toxico/ogy G. Jolles and A. Cordier (eds), 196-20, Academic Press, London, San Diego, New York, Boston, Sydney, Tokyo, Toronto (1992) WROBLESKY F. and LADUE J. Proc. Sot. Exp. Biol. Med. 9 0 21O-213 (1955) LOWRY O.H., ROSEBROUGH N.J., FARR A.L. and RANDALL R.J. J. Biol. Chem. 193 265-273 (1951)
Vol. 65, No. 26, 1999
15. KRACK G., GOETHALS 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35.
KupfferCells and HepaticEEects of Pametamol
2859
F., DEBOYSER D. and ROBERFROID M. Toxicology 18 213-223 (1980) LAU G. and CRITCHLEY J.A.J.H. J. Pharm. Biomed. Anal. 2 1563-1572 (1994) SAAD B., FREI K., SCHOOL F., FONTANA A. and MAIER P. Eur. J. Biochem. 229 349-355 (1995) BLAZKA M.E., WILMES J.L., HOLLADAY S.D., WILIAM R.E. and LUTER M.I. Toxicol. Appl. Pharmacol. 133 43-52 (1995) FEDER L.S., TODARO J.A. and LASKIN D.L. J.Leukoc.Biol. 53 126-132 (1993) RYMSA B., WANG J. and DE GROOT H. Am. J. Physiol. 281 G602-G607, (1991) GYENES M. and DE GROOT H. Am. J. Physiol. 284 G535-G540 (1993) SAMARASINGHE D.A. and FARRELL G.C. Hepatology 2 4 1230-I 237 (1996) CUTRIN J.C., LLESUX S. and BOVERSES A. Cell Biochem. Funct. 1 6 65-72 (1998) FERREIRA J., TAPIA G. and VIDELA L.A. FEBS Lett. 426 2638 (1998) QU WEI, ZHONG ZHI, ARTEES G.I. and THURMAN R.G. Am. J. Physiol. 275 G542-G549 (1998) RAIFORD D.S. and THIGPEN M.C. Toxicol. App. Pharmacol. 12 9 36-45 (1994) BADGER D.A., KUESTER R.K., SAUER J.M. and SIPES I.G. Toxicology 121 143153 (1997) WELTMAN M.D., FARRELL G.C., LIDDLE C. Gastroenterology 111 1645-1653 (1996) HINSON J.A., MAYS J.B. and CAMERON A.M. Biochem. Pharmacol. 32 19791988 (1983) ITINOSE A.M.. LEIKO M. and BRACHT A. Cell Biochemi. Funct. 7 263-273 (1989). ’ BURCHAM P.C. and HARMAN A.W. Biochem. Pharmacol. 3 8 2357-2362 (1989) BRAUN L., CSALA M., POUSSU A. and MANDL J. Fed. Eur. Biochem. Sot. 388 173-176 (i996) MANDL J., WALL C., LERANT I., FALUS A. and THURMAN R.G. Fed. Europ.Biochem. Sot. 376 65-66 (1995) DI LUZIO, N.R. and BLICKENS, D.A. J. Reticuloendothel. Sot. 3 250-270 (1966) CASTELEIGN E., KUPPER J., VAN ROOIJ H.C.J. et al. Biochem. J. 252 601 (1988)
ELSEVIER
MODULATION
Life Sciaoeq Vol. 65, No. 26, pp. 2851-2859, 1999 copy@lr01999Elscvi~sciilnc. RhfcdinUwUSA. Allri&tsmsrmd 0024_3205/‘99/hee fimt maltu
OF PARACETAMOL METABOLISM BY KUPFFER A STUDY ON RAT LIVER SLICES
CELLS
:
Neyrinck A.‘, Eeckhoudt S.L.1, Meunier C.J.1, Pampfer S.2, Taper H.S.1, Verbeeck R. K.1 and Delzenne N.1’ tuniversite Catholique de Louvain, Unite de Pharmacocinetique, Mbtabolisme, Nutrition et Toxicologic, UCL-PMNT 7369, 73 Avenue Mounier, B-1200 Brussels, Belgium 2Universite Catholique de Louvain, Unite de Recherche en Physiologie de la Reproduction, UCL-OBST 5330, 53 Avenue Mounier, B-1200 Brussels, Belgium (Received in final form August 16, 1999)
Summary
Recent studies support the hypothesis that non parenchymal cells (mainly macrophages) may play a role in the metabolism and cellular effects of paracetamol. In order to investigate this hypothesis, male Wistar rats were intravenously injected with either 7.5 mg/kg gadolinium chloride (Gd+) or NaCl 0.9% (Gd-). The treatment with GdCls decreased the number and the function of Kupffer cells in liver tissue, as assessed by the histological examination of the liver after colloidal carbon injection in the portal vein. Precision-cut liver slices (PCLS) were prepared from both groups of rats and cultured for 8h in Waymouth’s medium in the presence and absence of 5 mM paracetamol. Interestingly, PCLS obtained from Gd+ rats exhibited a lower release of tumor necrosis factor (TNF-a) and a better viability than PCLS from control (Gd-) rats. Incubation with paracetamol led to a decreased glycogen level in liver slices from Gd+ or Gd-, without modifying neither liver morphology nor ATP level nor LDH release. A higher proportion of paracetamol glucuronide, was secreted from the slices obtained from Gd+ rats. These data suggest that Kupffer cells could affect the viability of PCLS in culture and are involved in the regulation of phase II metabolism in the adjacent hepatocytes. We propose that PCLS in culture is a suitable model to elucidate the biochemical mechanism underlying the modulation of metabolism occurring through hepatocytes-Kupffer cells interactions. Key Words: gl~nidation,
KupEer cells, glycogen, liver slices
The role of sinusoidal cells in the hepatic metabolism has been greatly underestimated until now. However, Kupffer cells, despite their small size represent 80 to 90% of all fixed macrophages in the body and approximatively 14% of the hepatic cellular mass (1). Kupffer cells have the capacity to express typical macrophage functions such as * To whom correspondance should be addressed
2852
Kupffer Cells and Hepatic Effects of Pamcetamol
Vol. 65, No. 26, 1999
phagocytosis and secretion of a wide array of biologically active molecules, allowing them to quickly clear the circulation from cellular residues and microorganisms. Activated Kupffer cells release different mediators including cytokines such as TNF-a, reactive oxygen intermediates such as O2.-l lysosomal enzymes or eicosanoids, which are involved in their cytotoxic and inflammatory processes (1, 2). Such mediators are also capable to disturb both the metabolism or the integrity of adjacent hepatocytes (3, 4, 5). Moreover some authors have suggested that Kupffer cells could play a role in xenobiotic-induced hepatotoxicity. The hepatotoxicity of xenobiotics is often dependent on their metabolism. Although the hepatocytes are the major site of xenobiotic metabolism, several enzymatic activities such as GSH-transferase, UDPglucuronosyltransferase and cytochrome P450-dependent oxidase have been found in non parenchymal cells (6). The aim of this work was therefore to assess in vitro, by using the original model of precision-cut liver slices (PCLS), the influence of Kupffer cells on cellular integrity and on the liver capacity to metabolize a model compound, paracetamol, a well known mild analgesic and antipyretic agent. As compared to cocultures of hepatocytes and Kupffer cells, the model of PCLS preserves the tissue architecture, the proportion of the different cell types, and the polarity of hepatic cells (7). Moreover, it is highly representative of the metabolic pattern observed in w’vo and of the susceptibility of the liver to paracetamol hepatotoxicity. Paracetamol may be metabolized in the liver by several mixed function oxidase systems, dependent on cytochrome P450 isoforms (CYP2E1, CYPlAl and CYPlA2, CYPPCll in rats and CYP2E1, CYP3A4 and CYPlA2 in man) leading to the production of a highly reactive electrophilic and toxic metabolite : the n-acetylparabenzoquinoneimine (NAPQI); this NAPQI may be detoxified by conjugation to glutathione. Paracetamol metabolism in rats is characterized by the relatively low capacity to metabolically activate paracetamol towards its toxic metabolite as well as a high capacity to clear paracetamol via non toxic glucuronidation and sulphation pathways (8,9). The role of Kupffer cells was examined by comparing the results obtained in PCLS prepared from the liver of rats pretreated with saline (controls) or gadolinium chloride (GdC13) , a specific toxin for Kupffer cells which not only blocks phagocytosis by rat liver macrophages but alS0 selectively eliminates these cells situated in the periportal zone of the liver acinus (10). Material and Methods Animals and treatments: Adult male Wistar rats from lffa Credo (Les Oncins, France) weighing 283 f 8 g were housed in a temperature- and light-controlled room (12 h dark/light cycle). They were provided A04 standard diet (UAR, Villemoisson-sur-Crge, France) and tap-water ad libitum. GdC13 (7.5 mg/kg or 20 mg/kg), dissolved in Saline, was administered intravenously to rats 20 h before the injection of colloidal carbon Or the preparation of liver slices. To test for the phagocytic capacity of Kupffer cells, colloidal carbon (Pelikan@ no17 dissolved in saline 1:lO; 1.5 ml/kg body weight) WaS injected intravenously 20 min before liver sampling for further histological study. PCLS preparation and incubation: Rat surgical procedures were carried out under pentobarbital (60 mg/kg) anaesthesia. PCLS (about 220 pm thickness) were prepared with a Krumdieck slicer according to a procedure previously described (11, 12). After 60 min of preincubation at 4°C in Krebs-Ringer solution, PCLS were rinsed with saline and suspended in Waymouth medium supplemented with 0.3% of BSA, 1% Of glutamine solution, I % of penicilline-streptomycin solution, glucose (28 mM), insulin (100 nM) +/- paracetamol (5 mM). The vials, containing ITMXitWfII 4 dices (2 ml medium per slice), were saturated with a mixture of 95%02-5%COa and placed in a shaking water bath at 37°C for 4 to 8 hours.
Vol. 65, No. 26, 1999
Kupffer Cells and Hepatic FiiTem ofParaaamo1
2853
LDH release: The viability of liver slices was estimated by measuring the activity of lactate dehydrogenase (LDH) in the culture medium, according to the procedure of Wrobleski and Ladue (13). The results are expressed as international units (IU) of LDH per milligram of proteins. The amount of protein was determined by the method of Lowry (14) using BSA as standard. ATP content: Liver slices were taken, washed twice in saline and sonicated in 1 ml of 2% perchloric acid. The intracellular ATP content was measured on neutralized perchloric acid extracts by using an HPLC-LKB Pharmacia instrument equipped with a 4.7 x 125 mm (particle size 5 urn) anion-exchange column (Partisphere SAX, Whatman). Separation was achieved by the use of an isocratic elution mode with 0.45 M NH4H2PO4 (pH 3.7) at a flow rate of 1.5 ml/min. The results are expressed as nmol/mg protein. G/ycogen content: Liver slices were taken, washed twice in saline and sonicated in 1 ml of 1 M KOH. They were further heated at 100°C for 10 min. After neutralization with acetic acid and centrifugation, the supernatant was incubated in the presence of a-amylo glucosidase in pH 5 acetate buffer; the glucose produced was quantified by an enzymatic reaction as previously described (15). ffistological study.’Liver sections obtained from rats injected with colloidal carbon were fixed with 10% buffered neutral formalin and then embedded in paraffin, sliced into 3-4 pm sections, to assess by microscopic analysis of the phagocytic activity of Kupffer cells. Quantification of T/V/=-aproduction: TNF-a level was determined with an ELISA kit for quantification of rat TNF-a (Factor-Test-X” Rat TNF-a Kit, code 80-380501, Genzyme) in frozen aliquots of incubation medium (50 ~1). The results are expressed as pg/mg protein. Paracetamol metabolism: Paracetamol glucuronide, paracetamol sulphate and paracetamol glutathione conjugates were quantified by using reverse-phase HPLC according to the procedure of Lau and Critchley (16). Aliquots of incubation medium were stored at -30°C for further analysis. The proteins were precipitated in HC104 1% which contained the internal standard 2-acetamidophenol (40 ug/ml). After centrifugation, samples were injected and separated on a Nova-Pak Cl8 column by using an isocratic solvent system of 0.1 M KH$O@. 1% acetic acid/0.75% propane-2oi at a flow rate of 1.5 ml/min. The retention times were 4 min for paracetamolglucuronide, 10 min for paracetamol-sulphate, 12.8 min for paracetamol, 20.2 min for paracetamol-glutathione conjugate and 24.8 min for internal standard. The results were expressed as pg metabolitelmg protein. Statistical analysis: Results are expressed as mean+/- standard error of the mean. An unpaired student-t test or a one-way anova followed by a Scheffe test were performed for the statistical comparison of the results obtained in the different groups (Statview@). The actual test chosen is specified at the bottom of each figure or table. Results and Discussion The blockade of phagocytosis and the selective elimination of macrophages by i.v. injection of gadolinium chloride (GdCIs), is a generally accepted procedure for gaining knowledge about the function of macrophages in vivo (10). The injection of GdCls at a dose of 7.5 mgikg (Fig. 1.B) resulted in an appreciable decrease of Kupffer cell number and activity in the whole parenchyma compared to control rats who were
2854
Kupffer Cells and Hepatic ESects of Pama%mol
Vol. 65, No. 26, 1999
injected with 0.9 % NaCl (Fig. l.A). A complete disappearance of Kupffer cells COIJld not be obtained even by increasing the dose of GdCls to 20 mg/kg body weight. At tl7is higher dose, some necrotic foci were present (data not shown). Therefore, the dose of 7.5 mg/kg was used for further experiments. In order to assess the contribution of Kupffer cells to hepatic metabolism, PCLS were prepared from Gd- or Gd+ rats, a nd kept in suspension in an oxygenated medium (02K;O2 95/!5)for 4 or 8 hours.
1 .A.
Fig. 1 Influence ot GdCb InjectIon on the colloldrl carbon uptake by Kupffw cells. Distribution of carbon in liver 20 min after intravenous injection 1 day after GdC13 injection (1 .B.) or saline injection (1 .A.). The carbon is mainly taken up in Kupffer ceils (dark spots). x1200
Vol. 65, No. 26, 1999
Kupffer Cells and H-tic
Effects of Paracetamol
2855
The synthesis of TNF-a has been shown to occur in both hepatocytes and Kupffer cells after exposure to toxins or endotoxins (17, 18). As shown in Fig. 2, the release of TNF-a in the medium occurs without addition of xenobiotics or toxins. PCLS prepared from the liver of both Gd- or Gd+ rats released the same amount of TNF-a in the medium after 4 hours of incubation; however, a further secretion of TNF-a was only seen with slices obtained from Gd- rats. The production of TNF-a occurring from 4 to 8 hours of incubation may thus be attributed to Kupffer cells. The release of TNF-a after 4 hours of incubation, observed even in Gd+ animals, could be explained by the facts that either 1) some Kupffer cells escaped the effect of GdCl9, or 2) TNF-a has been produced by other ceil types present in PCLS, namely hepatocytes and endothelial cells. This later hypothesis is supported by the fact that in the absence of stimulation, endothelial cells were found to spontaneously produce several cytokines, namely IL-l and IL-6 within 1 or 2 hours of incubation; such cytokines may thus directly have an effect on TNF-a production by hepatocytes (17, 19).
Time (hour) Flg. 2 Influence of GdClg InjectIon prior to PCLS preparation on the release of TNFn In the medium after 8h of Incubation. Amountof TNF-a releasedin the incubationmediumof PCLS obtained from rats having received, 20h before to PCLS preparation, an i.v. injectioneither of NaCl 0.9% (Gd-, k4) or of GdC13 7.5 mgkg (Gd+, n=4). One way ANOVA test: p-c 0.05.
osp put? ‘Sap!UOJnOnlfj lOwela%Jed 40 lunowe ~a46!q a : ~Ow~~~k?Jed 40 se~!~oqe~ew 11 aseyd $lJOJ@44!p941 0OnpOJd 01 Sl3d 40 &OedleO aql pe!4!pOw “13pg hq SISJ 40 lUew~B3JlaJd au1 ‘E ‘614u! UMOys Sv ‘lOwe$3)?J13d 40 -440 Je(nll80 pu0 wS!loqlgew au1 uo luewl&‘eJleJd-SI~pg 40 eOu8n(4U! pa!pnlS OSle eAt?y aM ‘r(pn$3 1uaseJdeyj ul
SA +loUW=SJsd 100’0 > d :lSSi %Wi%~ ‘(P = U) +PD Sl3d PUS (9 = U) -PE) S-lad 10 (u!aiOJd fk1J/6tf) NJe1UO9 uei!bOo&) ‘-ps SA +PE) 90.0 > d :&Sol queq3Ssq ‘(p = U) +PD Sl3d :t~at
PUe (9 = U) -P!3 S-lad P uJn!P’JJ U! (u!mJd 6w-1) hl~W= Hal ‘-Pg SA +PE> 10’0 > d ?uaw& .(P = U) +Pg-S-I&I Pm (9 = U) -P’w313d 10 (U!eloJd Bu~/IOUJU) aUWJO3 dlV
38’c9 39’96
+
r B’OZ r
1
8’LL
2’8!Gl
f
1’8W
960’0 r L9’0 81’0 =FOZ.1
qSZ’0 * 08’0 PC’0 =F60’1
(srnoq
8)
+
9361 5: 9’SSB
eP’1 r 2’61 Z’l r E’PC
eL’1 r L’81 8’0 * L’&l -
lOltlS~S3SJSd suo(~(puoo
(wd
6w6rl)
ua6oOh16-anss!l +
+
(loJd Win) Ha-www (IoJd 6w/lowu) dlV-=ss!l
lOlIlSg338JSd uo~~Sqn3u~ 1
PD
SJSgSllJSJSd
318Vl
IJO!lleqnOU! S13d 6U!Jnp 6u!JJnOOO SJOlD34 O!xOlOtiO 40 aSeale~ aql SSOSS)?01 JO/PUB UO!leA!lO~ SllRI Jaydn)l 40 WS!UeqOaW ayl alep!On(a 01 h??SSQOau aJB s&Jaw!Jsdxa JaylJn4 ‘sesaylotiy asaql alE!&Jelsqns 01 slauueq0 wnplle0 pue aseue6r(xoo~3k1 BU!A~OAU! ws!ut?gOaw e r(q IClq!ssod ‘slla0 IewhqOuaJed u! uogdwnsuo0 ua6hxo Sa$l?Aala 43!4M ‘z=Jgd ey!l SJO#!paW t3S~C3laJ10 40 GOueseJd ayl U! SllaO JaUdn)l Ik?ql UMOUS OSle GARY (Sz) /t? Ja Ilo ‘(pz ‘cz) 3 aWOJl&)OflO jQJpUOl@Ol!W 40 suo!1eO!4!pow pue ayeadn ua6hxo JBMO( e hq paleO!pu! ‘J~A!I au1 u! wsyoqetaw ua6ho 40 uo!ge~n6aJ-umop e 01 spaal ~11%) Jalldn)l 40 uo!i!q!qu! S13pg ~leql BU!MOLJS ‘suogeglqnd IUKBJ hq pauoddns s! s!saqlodhy s!ql .uo!aeqnOu! aql 6u!Jnp uaMx0 pue slIJa!JvIu L(loq aJeds plnoM slla0 Ja44dn)l 40 p!o~ap Slgd
‘z
%?Z ‘lz) aSI3p!XO-&jaVN 40 UO!@A!138 Sql 01 0np aq pInO uouawouaqd S!yl ‘(C)z:) aJnaln0 u! slla0 Je44dn)l ~0~4 _,Q 40 UOgOnpOJd v~e0~4yti!s e 03 speel uo~~eus6rotoa~ 01 e!xodh./ WOJ4 UO!l!SUEJ$ aql $1?y1u~oys uaeq SBL/ $! ‘1324 ul ssflOo)eday SpJeMOl O!XolO~O aq 01 UMOUy se!%ds aA!lOtBJ 40 Slla:, J944dn)l hq aSt3laJ ayl ’1 : e@p GU!MOIIO~alla r(q pauoddns s! s!saylodr(y s!41 ‘(i wn!paw au1 u! UO!l~J~lJ~3UO3 “0 y6!4 L) hq palowoJd) a6owep C3A!aI?p!XO 01 Way1 aZ!l!SUaS pInO Sl&/ u! SIlKI JaUdn] 40 aOuas0Jd aqa days esodoJd aM 3JnllnO u! S13d U! saaholedaq 40 &pedeO O!lOqle~aw JO/plJe r(t!l!qo!A ~StBJO0p pin00 slle0 Jalldny 40 aOuasaJd ayl ‘(sap!JeqOOedlodod!( .a.!) s~o~r34 Gu!gqnw!gs u~ouy 40 aOuesqe acli u! ue~e ‘iaqa lse66ns sNnsaJ Jng -1ue0!4y6!s IOU SAM ~ua~uo~ ue600~~6 uo lU8wleaJiaJd 3l# 40 33U@nl4U! all1 UO!~~qnOlJ! 40 SJnOq 8 JaUe ‘Wn!&leW ayl u! @3B(aJ HQl JaMol B pU)3 3U0luOO dlV JE?lnllaO J6L46!4 e U!e~U!leW 01 p&ViOllB uo!~ti%?deJd Sl3d 01 JO!Jd Sl3pg 40 UO!P&J! al41 ’1 a(@?1 U! UMOI#3 Sbj ‘S#?J -pg WOJ4 Sl3d au1 UeLil UO!leqnOU! eql 6u!Jnp panraseJd J~UCI~ 81E?sII?J +pg ~0~4 S13d ‘I~I~u!~s~~J@~uI
6461 ‘9Z “‘N ‘$9 ‘PA
9282
Vol. 65, No. 26, 1999
Kupffer tills and Hepatic Effects of Pawetamol
2857
a high production of glutathione conjugates (reflecting both activation by cytochrome P450 isoforms and conjugation to glutathione), was secreted in the medium of Gd+ PCLS. This observation suggests that Kupffer cells could play a role as regulators (and in the present case, as inhibitors) of xenobiotic metabolism in the hepatocytes. In vitro, it has been shown that stimulation of Kupffer cells by Clostridium perfringens decreases cytochrome P450 dependent activity, thus decreasing paracetamol hepatotoxicity in vivo (26). TNF-a and interleukin 9 have been proposed as mediators, as they decrease CYP gene expression. It may be important to note that in viva, GdCls injection paradoxically also decreases the total hepatic content of cytochrome P450 (27). No studies until now have reported an effect of GdCl3 injection on the activity and on the content of the specific CYP isoforms responsible for paracetamol metabolism; this could be interesting to measure, as a “paradoxical” induction of several CYP isoforms despite a lower total CYP content has already been described in some situations (28).
Paracetamol sulphate
Paracetamol glucuronide
GSH-conjugates
Fig. 3
Influence of GdCls injection on the excretion phaae
II-metabolites
in the medium
by PCLS of peracetamol alter 8h of incubation. Measurementof
glucorono-.sulfo- and glutathionconjugated-metabolites of paracetamolreleasedin the mediumaftersh of incubationof PCLS obtainedfrom rats havingreceived,20h beforeto PCLS preparation,an i.v. injectioneitherof saline0.9%(Gd-) or of GdCiS 7.5 mgkg (Gd+). Results are presented as mean f 8.s.m. (n ~6). Unpaired t-test Student: “p c 0.05 vs Gd- and ““p < 0.01 vs Gd-
We have thus shown that paracetamol metabolism in rat PCLS is characterized by a high sulphation and glucuronidation and a low production of cytochrome P450 and glutathione conjugates : as suggested earlier, it makes rat liver more resistant to paracetamol hepatotoxicity (6). Even after 8 hours of incubation in the presence of
2858
Kupffer Cells and Hepatic Effects of Ftaracetamol
Vol. 65, No. 26, 1999
paracetamol at a relatively high dose (5 mM), no sign of morphological alterations could be detected by histological examination (data not shown). ATP levels and LDH leakage by PCLS were not modified by the addition of paracetamol to the medium. However, the glycogen content dropped dramatically, both in Gd+ and Gd- PCLS An activation of glycogenolysis by paracetamol was described by several authors, both in vivo (29, 30) and in isolated hepatocytes. The most probable explanation would be the activation of glycogen phosphorylase a (31, 32). The decrease in glycogen in the presence of paracetamol was even more pronounced in PCLS from Gd+ than from Gd- rats. This difference could be explained by the fact that part of glucose moieties of glycogen is deviated towards the production of UDP-glucuronic acid, the co-substrate for glucuronidation of xenobiotics (33). In fact, Billiard et al (1) consider Kupffer cells and the adjacent hepatocytes as a functional unit capable of complex and tightly regulated interaction. The present study offers support for a relevant role of Kupffer cells in the control of hepatocyte metabolism under physiological conditions. The real implication of the modulation by Kupffer cells of drug metabolism could be confirmed by using another Kupffer cell inhibitor such as methyl palmitate (34). Some authors have already reported that unstimulated Kupffer cells in culture released enough prostaglandins to alter the phosphorylation of specific proteins inside the hepatocytes (35). We propose that PCLS in culture would constitute a useful model to study the physiological roles of Kupffer cells in xenobiotic metabolism. Acknowledgements This work was supported by a grant from the “Region Wallonne” in Belgium; References 4: 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.
BILLIAR T.R. and CURRAN R.D. J. Parent. Enter. Nutri. 1 4 175S-180s (1990) DECKER K. Eur. J.Biochem. 192 245261 (1990) LASKIN D.L. and PENDINO K.J. Ann. Rev. Pharmacol. Toxicol. 35, 655-677 (1995) LASKIN D. and PILARO A. Toxicol. Appl. Pharmacol. 86 204-215 (1986) LASKIN D., PILARO A. and JI S. Toxicol. Appl. Pharmacol. 86 216226 (1986) LAFRANCONI W.M., GtAlT H. and OESCH F. Toxicol. Appl. Pharmacol. 8 4 500511, (1986) BACH P.M., VICKERS A.E.M., FISHER R., BAUMANN A., BRETTEBO E., CARLILE D.J., LAKE BG., SALMON F., SAVOYER T.W. and SKILINSKI G. ATLA, 4 893-923 (1996) MILLER M.G., BEYER J., ADAMS P.E. Toxicol. Appl. Pharmacol. 122 108-116, (1993) PATTEN J.C., THOMAS P.E., GUY R.L., LEE M., YANG C.S. Chem. Res. Toxicol. 6 51 I-518 (1993) HARDONK M.J., DIJKHUIS F.W.J., HULSTAERT C.E. and KOUDSTAAL J. J. Leuk. Biol. 5 2 296-302 (1992) GOETHALS F., DEBOYSER D., LEFEBVRE V., DECOSTER I. and ROBERFROID M. (1990) Toxicol. in Vitro 4 435-438 (1990) GOETHALS F., DEBOYSER D., CAILLAU E. and ROBERFROID M. In vitro methods in toxico/ogy G. Jolles and A. Cordier (eds), 196-20, Academic Press, London, San Diego, New York, Boston, Sydney, Tokyo, Toronto (1992) WROBLESKY F. and LADUE J. Proc. Sot. Exp. Biol. Med. 9 0 21O-213 (1955) LOWRY O.H., ROSEBROUGH N.J., FARR A.L. and RANDALL R.J. J. Biol. Chem. 193 265-273 (1951)
Vol. 65, No. 26, 1999
15. KRACK G., GOETHALS 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35.
KupfferCells and HepaticEEects of Pametamol
2859
F., DEBOYSER D. and ROBERFROID M. Toxicology 18 213-223 (1980) LAU G. and CRITCHLEY J.A.J.H. J. Pharm. Biomed. Anal. 2 1563-1572 (1994) SAAD B., FREI K., SCHOOL F., FONTANA A. and MAIER P. Eur. J. Biochem. 229 349-355 (1995) BLAZKA M.E., WILMES J.L., HOLLADAY S.D., WILIAM R.E. and LUTER M.I. Toxicol. Appl. Pharmacol. 133 43-52 (1995) FEDER L.S., TODARO J.A. and LASKIN D.L. J.Leukoc.Biol. 53 126-132 (1993) RYMSA B., WANG J. and DE GROOT H. Am. J. Physiol. 281 G602-G607, (1991) GYENES M. and DE GROOT H. Am. J. Physiol. 284 G535-G540 (1993) SAMARASINGHE D.A. and FARRELL G.C. Hepatology 2 4 1230-I 237 (1996) CUTRIN J.C., LLESUX S. and BOVERSES A. Cell Biochem. Funct. 1 6 65-72 (1998) FERREIRA J., TAPIA G. and VIDELA L.A. FEBS Lett. 426 2638 (1998) QU WEI, ZHONG ZHI, ARTEES G.I. and THURMAN R.G. Am. J. Physiol. 275 G542-G549 (1998) RAIFORD D.S. and THIGPEN M.C. Toxicol. App. Pharmacol. 12 9 36-45 (1994) BADGER D.A., KUESTER R.K., SAUER J.M. and SIPES I.G. Toxicology 121 143153 (1997) WELTMAN M.D., FARRELL G.C., LIDDLE C. Gastroenterology 111 1645-1653 (1996) HINSON J.A., MAYS J.B. and CAMERON A.M. Biochem. Pharmacol. 32 19791988 (1983) ITINOSE A.M.. LEIKO M. and BRACHT A. Cell Biochemi. Funct. 7 263-273 (1989). ’ BURCHAM P.C. and HARMAN A.W. Biochem. Pharmacol. 3 8 2357-2362 (1989) BRAUN L., CSALA M., POUSSU A. and MANDL J. Fed. Eur. Biochem. Sot. 388 173-176 (i996) MANDL J., WALL C., LERANT I., FALUS A. and THURMAN R.G. Fed. Europ.Biochem. Sot. 376 65-66 (1995) DI LUZIO, N.R. and BLICKENS, D.A. J. Reticuloendothel. Sot. 3 250-270 (1966) CASTELEIGN E., KUPPER J., VAN ROOIJ H.C.J. et al. Biochem. J. 252 601 (1988)