Influence of diet free of nad-precursors on acetaminophen hepatotoxicity in mice

ISSN 0306-3623/96 $15.00 + .00 SSDI 0306-3623(95)00033-X All rights reserved

Gen. Pharmac. Vol. 27, No. 1, pp. 79-82, 1996 Copyright © 1996 Elsevier Science Inc. Printed in the USA. ELSEVIER

Influence of Diet Free of NAD-Precursors on Acetaminophen Hepatotoxicity in Mice Hans Kroger, *l Monika Klewer, 2 Regina Gr/itz, 2 Annette Dietrich, 2 Wilhelm Ehrlich, 2 Sabine Altrichter, 1 Maciej Kurpisz, 3 and Ralf Miesel I IDEuTSCHES RHEUMAFORsCHUNGSZENTRUM(GERMANRHEUMATOLOOYRESEARCHCENTER), DEPARTMENT OF BIOCHEMISTRY,BERLIN,GERMANY;2ROBERTKOCH-INSTITUTE,DEPARTMENTOF BIOCHEMISTRY, BERLIN, GERMANY;3PoLISH ACADEMYOF SCIENCES,INSTITUTEOF HUMAN GENETICS,POENAN,POLAND A B S T R A C T . Recently, we demonst rated the hepatoprotective effects of nicotinic acid amide, a selective inhibitor of poly(ADP-ribose) polymerase (PARP; EC 2.4.2.30) on mice suffering from acetaminophen (AAP)-hepatitis, suggesting that the AAP-induced liver injury involves a step which depends on adenoribosylation. T h e present study investigates the effects of a diet free of precursors of N A D , the substrate on which P A R P acts, in female N M R I mice with A A P hepatitis and evaluates the influence of simultaneous ethanol consumption in these animals. Liver injuries were quantified as serum activities of glutamateoxaloacetate transaminase (GOT) and glutamate-pyruvate transaminase (GPT). While A A P caused a ll7-fold elevation of serum transaminase activities in mice kept on a standard laboratory diet, which was significantly exacerbated by ethanol and inhibited by nicotinic acid amide (NAA), adverse effects were noted in animals fed a diet free of precursors of N A D . In these animals, only m i n o r increases of serum transaminase activities were measured in the presence of A A P , and unlike the exacerbation caused by ethanol in mice on a standard diet, the liver damage was inhibited by 50% by ethanol. A further 64% reduction of hepatitis was observed, w h e n N A A was given to ethanol/AAP-mice. Our results provide evidence that the AAP-induced hepatitis and its exacerbation by ethanol can either be reduced by end-product inhibition of P A R P by N A A or by dietary depletion of the enzyme's substrate N A D . We see the m a i n application of N A A as for the combinational use in pharmaceutical preparations of acetaminophen in order to avoid hepatic damage in patients treated with this widely used analgesic. pEN PHARMAC27;1:79-82, 1996. KEY WORDS. Poly(ADP-ribose), acetaminophen, hepatotoxicity, mice INTRODUCTION Our previous studies demonstrated that liver damage caused by glutathione-depleting drugs like acetaminophen (AAP) involves a step which can be blocked by inhibitors of adenoribosylation. Thus, nicotinic acide amide (NAA), a selective inhibitor of poly(ADP-ribose) polymerase, suppressed the hepatic release of glutamate-oxaloacetate transaminase (GOT) and glutamate-pyruvate transaminase (GPT) in mice suffering from AAP-induced hepatitis (Kr6ger et al., 1994; Miesel et al., 1994a). The intrahepatic biotransformation of A A P by the cytochrome P-450 system produces a highly reactive quinone-imine which combines to sulfhydryl groups of proteins and causes the rapid depletion ofintracellular glutathione (GSH) (Jollow et al., 1974). GSH is the most important intracellular antioxidant and a powerful consumer of reactive oxygen species (ROS). ROS are produced continuously by mitochondrial respiration, cytochrome P-450 biotransformation of xenobiotics, and intracellular oxidases and dioxygenases (Miesel and Zuber, 1993a). GSH is also a substrate of the H202-removing enzyme glutathione peroxidase and cofactor for several enzymes in widely different metabolic pathways (Halliwell and Gutteridge, 1989a). The major function of GSH, however, is the biotransformation ofxenobiotics to mercapturic acids which are excreted into bile (Forth et al., 1983). The AAP-induced depletion of GSH and subsequent increase in intracellular GSSG levels do not only reduce the liver's capacity to cope with oxidative stress, but inactivate also an array of enzymes, including adenylate cyclase, fatty-acid

synthetase, phosphofructokinase, and phosphorylase phosphatase, thereby blocking major metabolic pathway s and diminishing both intracellular NADH and ATP levels (Halliwell and Gutteridge, 1989a). The breakdown of the GSH-dependent antioxidant defense system increases the intracellular flux of oxygen free radicals (Miesel et al., 1994b). Oxygen free radicals are potent inducers of PARP which catalyzes the posttranslational modification of proteins by covalent attachment of ADP-ribosyl-moieties derived from the splitting of NAD to nicotinamide and ADP-ribose (Cerutti, 1985). Excessive activation of PARP depletes the cell from its intracellular NAD-pools and initiates apoptosis (Hoshino et al., 1993). The present study investigates whether a diet free of NAD-precursors protects mice from AAP-induced hepatitis by diminishing PARP's substrate availability for adenoribosylation reactions which are considered the final step of liver damage in this experimental model of drug-induced hepatitis. Synergistic effects of ethanol are evaluated and compared to mice fed a standard laboratory diet. MATERIALS AND METHODS Chemicals Unless otherwise indicated, all reagents were purchased from SigmaAldrich, Deisenhofen, Germany. Animals

*Address all correspondence to: Prof. Hans Kr6ger, MD, PhD, Deutsches RheumaForschungsZentrum, Department of Biochemistry, Nordufer 20, D-13353 Berlin, Germany. Tel: 0049-30-454 2060; Fax: 0049-30-454 2090. Received 23 March 1995.

Female NMRI mice (n = 200), weighing 25-30 g, were housed under specific pathogen-free conditions in groups of 10 in wire-topped polycarbonate cages with a layer of sawdust as bedding. The facility had

80

A

H. Kr6ger et al.

10000

._I

B

10000

.J

1000

(.9

(.9 O'J o

o)

o

100

.v

10

"x ~ .~x ..~Ox" xw

¢° ,,o+ #,+.# +¢ ~.~0 "q'x~ x

F I G U R E 1. Acetaminophen-hepatitis in mice fed a standard laboratory diet: exacerbation by ethanol and inhibition by nicotinic acid amide. (A) GOT; (B) GPT. Female N M R I (n = 100) mice were fed an Altromin standard laboratory diet (containing 50 mg/kg of nicotinic acid and 2.3 g/kg of L-tryptophan) for 25 days. The drinking water of 60 mice contained ethanol (EtOH; 20% v/v). Prior to the application of drugs, the animals were starved for 12 h. Acetaminophen (AAP; 500 mg/kg) was given by gavage, and nicotinic acid amide (NAA; 300 mg/kg) was administered intraperitoneally. Sixteen hours later, the animals were sacrificed and the activity of G O T and G P T determined spectrophotometricaUy in serum. T h e data are presented as means _+standard deviations. Statistical analysis was performed using M a n n - W h i t n e y ' s nonparametric test. P < 0.05 was considered significant.

controlled lighting (light: 7.00-19.00 h), temperature (22°C), and relative humidity (50%). One hundred mice received a standard laboratory diet supplemented with all essential amino acids, vitamins, minerals, and trace elements (Altromin C 1000, Lage, Germany); this standard diet contained 2.3 g/kg of L-tryptophan and 50 mg/kg of nicotinic acid. Another group of 100 mice received a modified diet free of nicotinic acid, nicotinic acid amide, and L-tryptophan (Altromin, Lage, Germany). The drinking water of 60 mice of both the standard diet and modified diet group contained 20% of ethanol (v/v). Twenty-five days later the animals were starved for 12 h. Twenty untreated mice each of both diet groups received 10 ml/kg of saline (i.p.). Acetaminophen (500 mg/kg) was given by gavage to 20 untreated and 40 ethanol pretreated mice of both groups. Nicotinic acid amide (300 mg/kg; i.p.) was administered to 20 ethanol/AAP-mice in both diet groups. Sixteen hours after the application of drugs, the animals were killed by cervical dislocation and the activity of GOT and GPT determined in serum.

Enzyme Determinations Serum glutamate-oxaloacetate transaminase (GOT; EC 2.6.1.1) and glutamate-pyruvate transaminase (GPT; EC 2.6.1.2) activities were determined photometrically according to Bergmeyer (1974).

Statistical Analysis All data were analyzed with Mann-Whimey's nonparametric test using the Instat 2.01 statistics program (GraphPad, San Diego, CA) for Apple/

Macintosh and are presented as means _+standard deviations; P< 0.05 was considered significant. RESULTS

Intluence of Ethanol and Nicotinic Acid A m i de on G O T and G P T Activity in Serum o# Mice With AAP.Hepatitis Fed a Standard Laboratory Diet When compared to saline-treated controls, which showed basic serum GOT activities of 50 _+ 8 U/L, a single peroral application of 500 mg/ kg of acetaminophen to female NMRI mice fed a standard laboratory chow increased the serum GOT activity to 3353 _+ 2778 U/L (P < 0.0001) (Fig. 1A). A further 63% increase of GOT activity (5292 _+ 3241 U/L; P = 0.254) was measured in mice which had received additionally to AAP 20% of ethanol in their drinking water for 25 days. In the absence of AAP, ethanol did not significantly increase the basic levels of GOT observed in saline controls (48 -+ 7 U/L). The ethanol/AAP-induced hepatitis was significantly inhibited by the biogenic PARP-inhibitor nicotinic acid amide (910 _+ 2056 U/L; P = 0.0271). The powerful inhibitory effects of nicotinic acid amide on the ethanol/AAP-induced hepatitis were confirmed when GPT was used for the quantification of liver injury (Fig. 1B): the disease was inhibited by 89% in ethanol/AAP-mice additionally treated i.p. with 300 mg/kg of nicotinic acid amide (P = 0.0325). Ethanol did not significantlyincrease the basic levels of GPT of healthy animals (P = 0.341) but exacerbated strongly the AAP hepatitis by 68% (P < 0.001).

Acetaminophen Hepatotoxicity in Mice A

.J

81

10000

B 10000

1000

1000

I0

(.9 o

o

100

10

..oo

100

10

#J

,-°+ # , J e" #

,.o

X

FIGURE 2. Acetaminophen-hepatitis in mice fed a standard laboratory diet free of nicotinic acid, nicotinic acid amide, and L-tryptophan: inhibition by ethanol and inhibition by nicotinic acid amide. T h e application of drugs to female N M R I mice (n = 100), determination of serum transaminase activity, and statistical analysis followed the modus described in the legend to Fig. 1. (A) G O T ; (B) G P T .

Influence o f Ethanol a n d Nicotinic A c i d A m i d e on G O T a n d G P T A c t i v i t y in Serum o f M i c e W i t h A A P . H e p a t i t i s Fed a Diet Free o f Nicotinic Acid, Nicotinic A c i d A m i d e , and L . T r y p t o p h a n

amide given intraperitoneally to these animals pretreated with ethanol/ AAP (GOT: 217 +_ 452 U/L, P = 0.0055; GPT: 133 + 300 U/L, P = 0.1117).

Female NMRI mice (n = 100) were kept on a diet free of nicotinic acid, nicotinic acid amide, and L-tryptophan for 25 days. Sixty mice received additional ethanol (20% v/v) in the drinking water ad libitum. After 12 h of starvation, hepatitis was induced by AAP (500 mg/kg; p.o.). Three hundred mg/kg of nicotinic acid amide were given intraperitoneally to 20 mice each suffering from ethanol AAP-hepatitis. When compared to animals fed a standard laboratory chow, the basic GOT and GPT levels of saline-treated or ethanol-treated controls did not differ significantly from mice fed the NA/NAA/L-Trp-free diet. A highly significant difference between both diet groups was observed in both GOT and GPT levels of AAP-treated mice: while transaminaselevels of 3353 and 3730 U / L were measured in mice on standard diet (see above), acetaminophen caused a 79-84% diminished release of G O T (703 _+ 1201 U/L; P = 0.0003) and GPT (600 _+ 1112 U/L; P < 0.0001) in mice kept on NA/NAA/L-Trp-free diet (Figs. 2A,B). Much to our surprise, a further 38% reduction of G O T levels was observed in mice which had been treated additionally to AAP with ethanol (439 _+ 831 U / L versus 5292 _+ 3241 U / L in mice with standard diet; P < 0.0001). The same adverse effect was noted, when GPT was used to judge the hepatic damage: while ethanol exacerbated strongly the AAP-hepatitis in mice kept on a standard diet (Fig. 1B), the N A / NAA/L-Trp-free diet reduced the GPT levels in ethanol/AAP mice by 49% (303 -+ 625 U/L). A further reduction of both G O T and GPT levels by 51 and 64%, respectively, was obtained with nicotinic acid

DISCUSSION The present study reports the inhibition of ethanol/acetaminophen hepatotoxicity in female NMRI mice kept on a diet free of nicotinic acid, nicotinic acid amide, and L-tryptophan. These results add to our earlier findings of hepatoprotection by PARP inhibitors in mice with liver damage induced by various anti-inflammatory and anti-rheumatic drugs (Kr6ger et al., 1989). Moreover, our results support the hypothesis that the hepatotoxic effects caused by these drugs involve an adenoribosylation step, which can be blocked by PARP inhibitors or, as described here, by depletion of the precursors of the substrate for the PARPreaction. Acetaminophen causes the collapse of the glutathione-dependent antioxidant barrier within hepatocytes (Jollow et al., 1974). Diminished antioxidant defenses enhance the intracellular flux of oxygen-free radicals and induce the expression ofPARP (Cerutti, 1985). Adenoribosylation can cause apoptosis which is prevented by specific PARP inhibitors (Hoshino et al., 1993). The acetaminophen hepatoxicity is exacerbated by simultaneous consumption of ethanol in animals fed a standard laboratory diet (Seeff et al., 1986). Ethanol, which is either oxidized by alcohol dehydrogenase (ADH) to acetaldehyde or metabolized by the microsomal ethanol oxidizing system (MEOS), produces oxygen radicals during its metabolisation (Halliwell and Gutteridge, 1989b). In the presence of diminished intracellular GSH levels (as is the case in AAP hepatitis), acetaldehyde is further oxidized by xanthine oxidase to acetic acid, superoxide, hydrogen peroxide, hydroxyl radicals, and

82

H. Krdger

possibly singlet oxygen which are powerful oxidants

and may signifi-

cantly increase liver damage (Miesel and Zuber, 199313). Subsequently, poly(ADP-ribose) cient substrate

polymerase

is activated,

(NAD) or precursors

to NAD

reactions on proteins.

The latter

the enzyme starts its adenoribosylation step in ethanol/AAP tionofPARP

hepatotoxicity

by nicotinic

is prevented by end-product

acid amide. ThePARP-dependent

of AAP hepatotoxicity

was confirmed

of precursors

the substrate

of NAD,

mice, significantly

and in the presence of suffi-

which can be converted

diminished

in animals

the AAP hepatotoxicity

from AAP-induced

liver injury.

but protected

of NAD

(Ritchie,

substrate preventing

NAD+/NADH

redox potential consumption

of precursors

for the PARP reaction

adenoribosylation

did not

the animals partially

by excessive ADH-dependent

1985). In the absence

availability

diet, ethanol

This adverse effect may be explained

by ethanol’s effects on the intracellular which is in favour ofNADH

In these

effects were noted (see Figs.

1 and 2), and unlike in mice kept on a standard exacerbate

kept on a diet free

for the PARP reaction.

hepatotoxic

inhibi-

mechanism

and exerting

of NAD’,

is limited, additonal

the

subsequently

protection

from

liver injury. Our data provide unequivocal

evidence that liver damage caused by

acetaminophen

and its exacerbation

by end-product

inhibition

by ethanol can either be prevented

of poly(ADP-ribose)

polymerase

tinic acid amide or by a diet free of precursors

giving nico-

of the substrate

for the

enzyme. However, it should be kept in mind that adenoribosylation the ultimate oxygen

step in AAP

radicals

dependent

hepatitis,

excessively

antioxidant

while the primary

produced

barrier

represents inducers

upon breakdown

are

of the GSH-

caused by AAP biotransformation.

It

may therefore be advisable to block both the induction

and the ultimate

step in AAP

by antioxidants

hepatitis

like N-acetylcysteine adenoribosylation

via the synergistic

like nicotinic

of pharmacologically

and antioxidants

use in pharaceutical

and by inhibitors

preparations

its in order to avoid complications

Professor

Ockenfels

safe inhibitors

as for the combinational ofGSH-depleting

derived from excessive

induced by these drugs. We dunk

of

acid amide.

We see the main application of polyadenoribosylation therapeutic

inhibition

(a powerful GSH protector)

for reading

the manuscript.

analgetliver injury

et al.

References Bergmeyer H. U. (1974) In: Methoden der enzymatischen Analyse I (Bergmeyer H. U., ed.), p. 491. Verlag Chemie, Weinheim. Cerutti P. A. (1985) Prooxidant states and tumor promotion. Science 227,375381. Forth W., Henschler D., Rummel W. (1983) Allgemeine und spezielle Pharmakologie und Toxikologie, B. I., pp. l-85. Wissenschaftsverlag, Mannheim. Halliwell B. and Gutteridge J. M. C. (1989a) Protection against oxidants in biological systems: the superoxide theory of oxygen toxicity. In Free Radicals in Biology and Medicine II (Halliwell B. and Gutteridge J. M. C., eds.), pp. 86187. Clarendon, Oxford. Halliwell B. and Gutteridge J. M. C. (1989b) Ethanol metabolism. In Free Radicals in Biology and Medicine II (Halliwell B. and Gutteridge J. M. C., eds.), p. 47. Clarendon, Oxford. Hoshino J., Beckmann G. and KrBger H. (1993) 3-Aminobenzamide protects the mouse thymocytes in vitro from dexamethasone-medicated apoptotic cell death and cytolysis without changing DNA strand breakage. J. Steroid. Mol. Biol. 44, 113-l 19. Jollow D. J., Thorgeirsson S. S., Potter W. Z., Hashimoto M. and Mitchell J. R. (1974) Acetaminophen-induced hepatic necrosis. Metabolic disposition of toxic and nontoxic doses of acetaminophen. Phurmacol. 12, 251-271. Kriiger H., Klewer M., Grgtz R., Dietrich A., Ockenfels H. and Miesel R. (1994) Exacerbation of acetaminophen hepatoxicity by thalidomide and protection by nicotinic acid amide. Gen. Phmnacol., submitted. Kroger H., Sohst S., G&z R., Dietrich A. and Grahn H. (1989) Short time action of antirheumatic substance on the liver of rats. Protective effects of tryptophan + methionine. Gen. Pharmacol. 20, 211-212. Miesel R., Ehrlich W., Klewer M., Gritz R., Dietrich A. and KrBger H. (1994a) Theinfluenceofantagonistsofpoly-ADP-metabolismonacetaminophen hepatotoxicity. Gen. Pharmucol., in press. Miesel R. and Zuber M. (1993a) Copper-dependent antioxidast defenses in inflammatory and autoimmune rheumatic diseases. Inflammation 17, 283-294. Miesel R. and Zuber M. (1993b) Elevated levels of xanthine oxidase in serum of patients with inflammatory and autoimmune rheumatic diseases. Inflammation 17, 551-561. Miesel R., Zuber M., Sanocka D., Gratz R. and Krbger H. (199413) The effects of allopurinol on the in viva suppression of arthritis in mice and the ex viva modulation of the phagocytic production of oxygen radicals in whole human blood. inflammation 18, 597-611. RitchieJ. M. (1985)Thealiphatic alcohols. InGoodman Gilmnn’sThePhannacological B&s ofTherapeutics VII (Goodman-Gilman A., Goodman L. S., Rail T. W. and Murad F., eds.), pp. 372-386. Macmillan, New York. Se&L. B., Cuccherini B. A., Zimmerman H. J., Adler E. and Benjamin S. B. (1986) Acetaminophen hepatotoxicity in alcoholics: a therapeutic misadventure. Ann. Intern. Med. 104, 399-404.