Selective acetaminophen metabolite binding to hepatic and extrahepatic proteins: An in vivo and in vitro analysis

TOXICOLOGY

AND

APPLIED

PHARMACOLOGY

99,240-249

( 1989)

Selective Acetaminophen Metabolite Binding to Hepatic and Extrahepatic Proteins: An in Vivo and in Vitro Analysis JOHNB.BARTOLONE,* WILLIAM P. BEIERSCHMITT,~ RAYMOND B. BIRGE,* SUSAN G. EMEIGH HART,+ STUART WYAND,$ STEVEN D. COHEN,? AND EDWARD A. KHAIRALLAH**’ *Department ofMolecular and Ceil Biology, tDepartment of Pharmacology and Toxicology, and *Department ofPathobiology, The University of Connecticut, Storrs, Connecticut 06268

Received October 7, 1988; accepted February I, 1989 Selective Acetaminophen Metabolite Binding to Hepatic and Extrahepatic Proteins: An in Vivo and in Vitro Analysis. BARTOLONE, J. B.. BEIERSCHMITT, W. P., BIRGE, R. B., EMEIGH HART, S. G., WYAND, S., COHEN. S. D., AND KHAIRALLAH, E. A. (1989). Toxicol. Appl. Pharmacol. 99, 240-249. Acetaminophen (APAP) administration (600 mg/kg, po) to fasted male CD-l mice resulted in cellular damage to liver, lung, and kidney. An affinity purified antibody against covalently bound APAP was used to identify APAP-protein adducts in microsomal and cytosolic extracts from these target organs. The proteins were resolved on SDS-PAGE. transblotted to nitrocellulose membranes, and analyzed immunochemically. Covalent binding of APAP to intracellular proteins was only observed in those organs which exhibited cellular damage: no APAP adducts were detected in tissues which did not undergo necrosis. In all target tissues the arylation of proteins was not random but highly selective with two adducts of 44 and 58 kDa accounting for the majority of the total APAP-bound proteins which were detected immunochemically. In addition, a third major APAP-protein adduct of 33 kDa was also observed in kidney cytosol. The severity of tissue damage and the amount of adducts present in these tissues could be significantly reduced when mice were pretreated with the mixed function oxidase inhibitor, piperonyl butoxide, prior to APAP dosing. Immunochemical analysis of plasma from APAP-treated animals indicated the presence of several protein adducts by 4 hr following drug administration. These adducts did not appear to be of plasma origin. Incubation of cytosolic proteins from liver, lung, kidney, spleen, brain, and heart with an APAP metabolite generating liver microsomal system demonstrated that the cytosolic 5%kDa protein target was native to all tissues tested. By contrast, the 58-kDa protein target did not appear to be endogenous to plasma since it was not detected when plasma was incubated in vitro with the liver microsomal system. These studies indicate that, although the 58-kDa proteins appear to be endogenous to both target and nontarget tissues, the 58-kDa APAP-protein adducts are detectable only in tissues which become damaged by APAP. o 1989 Academic PIES, IIIC.

Acetaminophen (APAP) is a widely used analgesic and antipyretic compound considered safe at therapeutic doses. However, upon an acute overdose in experimental animals and ’ To whom correspondence should be addressed at the Department of Molecular and Cell Biology, U- 125. The University of Connecticut, 75 North Eagleville Road, Storm, CT 06268. 0041-008X/89

$3.00

Copyright 0 1989 by Academic Press, Inc. All rights of reproduction in any form reserved.

240

man, APAP induces lesions in several organ systems, the best characterized of which is hepatic centrilobular necrosis (Mitchell et al., 1973a; Davis et al., 1974). Liver damage results from excessive production of a reactive metabolite of APAP generated by the hepatic cytochrome P450 mixed function oxidase system (Potter et al., 1973). The reactive metabolite, N - acetyl - p - benzoquinoneimine

SELECTIVE

APAP BINDING

TO HEPATIC

(NAPQI; Dahlin et al., 1984), can be detoxified through conjugation to glutathione, but following an APAP overdose, glutathione is depleted allowing NAPQI to bind more readily to hepatic proteins and initiate processes which lead to cytotoxicity (Mitchell et al., 1973b). Agents which either deplete glutathione or alter mixed function oxidase (MFO) activity subsequently affect the amount of covalent binding to protein and ultimately produce corresponding changes in the degree of hepatotoxicity (Jollow et al., 1973). Although the mechanistic linkages between covalent binding of APAP to macromolecules and the resultant liver toxicity are unknown, there have been no reports of APAP-induced hepatic damage in the absence of such binding. Lung and kidney also possess cytochrome P450 enzymes which are able to metabolize xenobiotics (Devereux et al., 1979; Mohandas et al., 198 1) and we have previously demonstrated necrosis in both organs subsequent to hepatotoxic doses of APAP to CD- 1 mice (Placke et al., 1987). In our studies of APAP hepatotoxicity in the CD-l mouse we have demonstrated that covalent binding is not random but is quite selective with respect to the targetted proteins (Bartolone et al., 1987). Moreover, the selectivity does not appear to be uniquely dependent upon the abundance or the sulphydryl content of the arylated proteins (Bartolone et al., 1988). Using our affinity purified antiAPAP antibody we have demonstrated that the major APAP-binding macromolecules, when resolved electrophoretically with SDS under reducing conditions, are proteins of 44 and 58 kDa. The present study was undertaken to determine if similar targetting of APAP electrophile to proteins occurs in extrahepatic tissues which are damaged after an APAP overdose. Comparisons were also made with tissues which do not exhibit APAP-induced necrosis in our animal model and attempts were made to discern whether the targetted proteins were ubiquitous or present only in susceptible tissues. The data clearly demon-

AND EXTRAHEPATIC

PROTEINS

241

strated that APAP-protein adducts of similar molecular weight were detected in lung and kidney but not in nontarget tissues. Furthermore, the data also indicated that proteins similar in molecular weight to those targetted in liver, lung, and kidney are also evident in the cytosol of nontarget tissues, suggesting that inadequate exposure to APAP electrophile is key to the prevention of toxicity in the nontarget organs. METHODS Animals and treatment protocol. Fasted (18-20 hr), 3month-old male CD- 1 mice (Charles River, Wilmington, MA) were dosed by gavage with APAP (600 me/kg, po) using a 6% stock drug solution of propylene glycol in water (50:50). This treatment has previously been reported to elevate levels of plasma sorbitol dehydrogenase, to produce significant hepatic protein covalent binding, and to elicit necrotic changes in the liver, lung, and kidney (Ginsberg et al., 1982; Placke et al., 1987). In experiments where MFO inhibition was desired, piperonyl butoxide (PB, 600 mg/kg, 5 ml/kg. ip) was administered to mice in corn oil 1 hr prior to APAP. Control animals were given vehicle(s) only. For immunochemical and histopathological analysis, mice were killed by cervical dislocation 4 and 24 hr following drug or vehicle, respectively. Assessment of tissue injury. Livers and kidneys were immersed in 10% buffered formalin. The tracheas were cannulated and the lungs were inflated with the buffered formalin solution. AI1 fixed tissues were embedded in paraffin, sectioned, routinely stained with hematoxylin and eosin, and then examined using light microscopy. Lesions were graded based on the following scale: 0, no lesion; If, minimal; 2+, mild; 3+, moderate; 4+, severe. Blind screening of tissue samples was conducted to prevent biased conclusions. Scores greater than 2+ were considered significant evidence of damage. Immunochemical detection ofAPAP-protein adducts. Tissues of interest (livers, lungs, kidneys, spleens, brains, and hearts) were excised and rinsed in ice-cold 10 mM phosphate buffer (pH 7.4) containing 0.8% NaCl (PBS). Only cortical tissue samples were analyzed for kidney. Tissues from control and treated animals were homogenized in 9 vol of ice-cold 0.25 M sucrose, 10 mM TrisHCl, 1 mM MgClz at pH 7.4 using a Polytron PT-10 (Brinkman Instruments, Westbury, NY) and fractionated by centrifugation at 9000g for 20 min followed by 105,OOOg for 1 hr to obtain microsomal and cytosolic fractions. For plasma analysis. blood samples from control and treated mice were obtained immediately following decapitation. Blood was collected into heparinized

242

BARTOLONE

tubes and centrifuged at 12,OOOgfor 5 min at 4°C. The plasma was removed and frozen at -70°C until analyzed. Samples ofplasma and all six tissues were immunochemically analyzed for APAP-protein adducts using an affinity purified anti-APAP antibody developed, purified, and specificity characterized as previously described (Bartolone et al., 1988). Proteins were resolved (30 rg/ lane) according to molecular weight using the procedure of Laemmli (1970) on a discontinuous 10% SDS-polyacrylamide gel electrophoresis system with a 3% stacking gel layer. The gels were run at a constant current of 20 mA/slab. Electrophoretically separated proteins were electroblotted onto nitrocellulose membranes (Schleicher & Scheull, Keene, NH) at 80 V for 6 hr in 25 mM Tris/l92 mM glycine buffer (pH 8.3) containing 20% methanol. Membranes were rinsed in T&buffered saline (TBS) at pH 7.4 containing 0.05% Tween 20 (TBSTween) for 1 hr, then blocked overnight in 3% bovine serum albumin (BSA) in TBS at 4’C, and the membranes were incubated with affinity purified anti-APAP antibodies diluted 1:400 in TBS-BSA for 3 hr at room temperature. After washing five times for 10 min each in TBSTween, the membranes were then incubated for 90 min with ‘2SI-conjugated goat anti-rabbit IgG (10 &i/l00 ml TBS-BSA). Following immunostaining, the nitrocellulose was again washed extensively, air dried, and exposed at -70°C to Kodak XAR-5 film. Duplicate gels were stained with 0.1% Coomassie brilliant blue in 50% methanol and 3% acetic acid and then destained with 25% methanol and 10% acetic acid. Gels used for electroblotting were also stained to determine the quality of transfer; only those gels with complete transfer of proteins were utilized for immunostaining. Formation of APAP-protein microsomal-cytosolic systems.

adducts

in reconstituted

In an effort to establish the endogenous localization of the major proteins targetted by APAP in cytosol from extrahepatic tissues, a reconstituted liver microsomal system was used to activate APAP to NAPQI as we found liver to be optimum in generating APAP metabolites. Cytosol derived from either target or nontarget tissues was added to the liver microsome system which has been shown to produce an APAP binding profile in liver cytosol similar to that observed in vivo (Bartolone et al., 1988). All tissues were obtained from fasted untreated CD- 1 mice, and were homogenized ( 1:4) using a Dounce in a 10 mM phosphate buffer (pH 7.4) containing 0.25 M sucrose. Homogenates were separated by centrifugation into three subcellular fractions as described above. To remove contaminating cytosol, liver microsomal pellets were washed twice by suspension in 0.1 M phosphate buffer and recentrifuged at l05,OOOgfor 1 hr. To enhance APAP metabolite binding to proteins, glutathione and other low molecular weight constituents less than 5000 Da were removed from plasma and the 105,OOOgsupematant fractions of each tissue by passage over a Sephadex G-25 column. Cytosolic constituents larger than 5000 Da were col-

ET AL. lected in the void volume of the G-25 column and utilized for these experiments. APAP was activated in vitro as previously described (Buckpitt et al., 1977). The activation system contained 10 mg of washed liver microsomes; 4 ml of a cofactor mixture consisting of 0.1 M phosphate buffer (pH 7.4) containing 3.3 mM NADP, 80 mM glucose 6-phosphate, 60 mM MgC12; 2 ml of 4 mM APAP and 10 IU of glucose-6-phosphate dehydrogenase and incubated for 45 min at 37°C with 2 ml of cytosol from one of the six tissues utilized or with plasma. After incubation, the activated systems were then recentrifuged at 105,OOOg for 1 hr to separate the intact microsomes from the soluble supematants for immunochemical analysis of the high speed supematant. As a control, washed liver microsomes were incubated in the absence of added cytosol and centrifuged and both the supematant and pellet fractions were also analyzed. A similar protocol was followed using plasma from control mice to substitute for tissue cytosol in an effort to determine which plasma proteins might be preferentially targetted by NAPQI generated from liver microsomes incubated in vitro.

RESULTS APAP-Induced Histopathological of Tissues from CD-l Mice

Alterations

The results of histopathological examination of tissue from control mice and mice killed 24 hr after receiving APAP (600 mg/ kg, po) with or without piperonyl butoxide pretreatment are presented in Table 1. Cellular damage in liver, lung, and kidney from APAP-treated mice was evident with 57.5, 85.7, and 42.9% of each tissue type, respectively, graded either moderate or severe (i.e., > +2). Briefly, the hepatocellular necrosis, as expected, was primarily centrilobular in nature. Lesions of the kidney were confined to the cortical regions of the proximal tubules and consisted of single tubular cell necrosis. The pulmonary alterations consisted of degeneration and coagulation necrosis of the bronchiolar epithelial cells. Piperonyl butoxide reduced the incidence of moderate to severe damage from 57.5 to 0% in liver, from 85.7 to 25% in lung, and from 42.9 to 12.5% in kidney.

SELECTIVE

APAP

BINDING

TO

HEPATIC

AND

EXTRAHEPATIC

PROTEINS

243

1

TABLE

ASSESSMENTOF TISSUE NECROSISIN CD- 1 MICE 24 hr AFTER ACETAMINOPHEN ADMINISTRATION 0

I+

2+

3+

4+

1 1

3 3

2

5

0 0 1

0 0 0

0

1

2

3

0 0 0 1

57.0

Lung Control

1

PB PB/APAP APAP

1 0 0

2 1 3 0

2 2 3 1

0 0 1 4

0 0 1 2

0 0 25.0 85.7

4 4 4 1

1

0

0

0

0

0 2 1

0

0

0

0

Liver Control PB PB/APAP APAP

Kidney Control PB PB/APAP APAP

?6>2+

0 0 0

1

1

0

12.5

2

3

0

42.9

Note. Fasted CD- 1 mice were dosed as indicated, killed 24 hr later and the extent of hepatic, pulmonary, and renal necrosis was scored in survivors by the criteria given under Methods. The Table denotes the number of mice exhibiting histology scores indicated. Scores greater than 2+ were considered significant and were calculated as a percentage of the total number of animals graded. PB, piperonyl butoxide (600 mg/kg, ip); APAP, acetaminophen (600 mg/kg, po). PB was administered in corn oil; APAP was administered in propylene glycol.

Detection of Specific APAP-Bound Target Tissues

Proteins in

Immunochemical analysis of electrophoretically resolved tissue proteins revealed selective arylation in liver, lung, and kidney 4 hr after APAP dosing (Fig. 1). Liver binding was in agreement with our previous findings (Bartolone et al., 1987) with protein bands of apparent M, of 44 and 5 8 kDa accounting for the majority of the APAP-bound proteins detected in both cytosol and microsomes. The 58-kDa band was also prominent in lung but not in kidney microsomes whereas it was prominent in cytosol from both tissues. By contrast, the 44-kDa adduct was notable in kidney microsomes but in lung microsomes was much less significant than the 58-kDa band. Similarly, the 44-kDa band was also present in kidney cytosol but was not clearly evident in lung cytosol. Of interest is an additional intense band in kidney cytosol with an apparent M, of 33 kDa. In addition to these

predominant APAP-bound protein bands, several other minor APAP-protein adducts of various M, were also detected especially in liver and kidney; fewer adducts were detected in the lung. No binding of APAP to proteins was observed in heart, brain, or spleen (data not shown). Pretreatment of mice with piperonyl butoxide greatly diminished the amount of immunochemically detectable APAP-protein adducts in both microsomes and cytosol of liver, lung, and kidney with almost complete loss of detectable binding in lung and kidney microsomes (lanes PB). Comparison of the distribution of proteins of different molecular weight among the fractions (lanes CB) from the three tissues of control mice does not reveal any clear explanation of the distribution of arylated proteins on the basis of protein content, with perhaps one exception. In kidney cytosol from control mice there is a high quantity of protein at 33 kDa which corresponds to that identified by Western blot analysis (compare lanes CB and AP).

244

BARTOLONE LIVER

LUNG

ET AL. LIVER

KIDNEY

66.000

-

45.000

-

LUNG

-PY 15

35,000

-

-

24.000

--u,

-

16.000

-

KIDNEY

(L

u -

I ” ,“”

-

CB c PB AP

-”

CB c PB AP

-9

CB c PB AP

06

i

IMi

c PB AP

CB c PB AP

CB c PB AP

Si

FIG. 1. Immunochemical detection of APAP-bound proteins in liver, lung, and kidney from mice dosed in vivo. Fasted CD- 1 mice were administered APAP (600 mg/kg, po) in 50% propylene glycol 1 hr following either corn oil vehicle (lanes AP) or the mixed function oxidase inhibitor, piperonyl butoxide (lanes PB). Animals were killed 4 hr after APAP and the tissues were homogenized and fractionated by centrifugation. Control animals received vehicles only (lanes C). The microsomal (M) and cytosolic (S) fractions were analyzed for APAP-bound proteins using an affinity purified anti-APAP antibody following SDS-PAGE and Western blotting as previously described (Bartolone et al., 1988). Lanes CB show Coomassie blue stained gels from representative fractions from control mice. All other lanes are Western blot profiles. The relative migration of A4,, standards are indicated in the center.

Analysis ofAPAP-Protein

Adducts in Plasma

To determine if APAP-protein adducts may be released into blood following hepatotoxic exposure, plasma collected from mice 4 hr after APAP treatment was also analyzed. Figure 2 demonstrates that the three major adducts (i.e., 58, 44, and 33 kDa) were detected in plasma. The 5%kDa band was the most intense. Several other adducts of various M, were also present in the plasma from treated mice but interestingly, there was no significant binding to albumin, the most abundant plasma protein (IM, = 67,000), above the level of nonspecific binding observed in the control. Comparison of the protein distribution as evidenced by Coomassie blue staining following electrophoresis indicates the presence of a number of proteins in plasma from treated mice which were not detectable in control plasma. By contrast, plasma from mice pretreated with piperonyl butoxide, 1 hr prior to APAP administration, did not have any immunochemically detect-

able APAP-protein adducts nor was there any alteration in Coomassie blue staining patterns (data not shown). APAP-Protein Adducts Produced in NAPQI Generating Systems Figure 3 shows the distribution of APAPderived protein adducts which were produced when cytosolic extracts from target and nontarget tissues were incubated for 45 min with a liver microsomal fraction fortified with an NADPH regenerating system and APAP. For all tissues tested the most prominant cytosolic bands detected immunochemically were of 44 and 58 kDa. In addition, a band of approximately 33 kDa was detected in kidney but not in lung cytosol consistent with our in vivo findings. Many other lighter bands were also observed in liver cytosol. The lane labeled MICRO represents the supernatant from an incubation of APAP with fortified liver microsomes without added cytosol.

SELECTIVE protein

stain

-

Lr

c

APAP BINDING

4

TO HEPATIC

western

-

67

kd

-

-

45

kd

-

-

35

kd

-

-

24

k,,

-

AND EXTRAHEPATIC

PROTEINS

245

tic contrast to the binding pattern obtained after in vivo dosing (Fig. 2). DISCUSSION

.,

V:.,

4

c

FIG. 2. Analysis of APAP-bound proteins in blood plasma. Blood was collected in heparinized tubes from control mice (lanes C) or mice receiving APAP (600 mg/ kg, po) and killed 4 hr following dosing (lanes 4). The plasma was immunochemically analyzed by Western blotting following protein separation using SDS-PAGE (Western lanes). Coomassie Blue stained gels are shown in “protein stain” lanes.

This lane indicates that the 44-kDa adduct forms in microsomes and can be released into cytosol. Since no 58kDa adduct was detected when plasma was substituted for tissue cytosol one can conclude that the 58-kDa adducts observed in the tissue cytosolic fractions reflects the interaction of NAPQI generated by the liver microsomes with proteins which were native to the cytosol of those tissues. By contrast, it is not possible to ascertain the origin of the 44-kDa adduct detected in these experiments since it may have been derived from the liver microsomal system utilized for APAP activation. Also shown in Fig. 3 is the binding profile obtained when plasma from control mice was added to fortified microsomes and APAP in place of tissue cytosol. Two APAP-derived adducts are evident, one of 44 kDa which corresponds with the microsomal adduct discussed above and one of approximately 67 kDa which corresponds to the M, of serum albumin. There was no binding detected in either the 33 or 58 kDa ranges. This is a dras-

The potential for fatal liver necrosis following an acute overdose with APAP is widely recognized (Rumack and Matthew, 1975). Experimental studies by ourselves and others have indicated that organs other than the liver may also be damaged by hepatotoxic doses of APAP (Placke et al., 1987; Boyer and Rouff, 1971). The hepatotoxicity is thought to result from the binding of the P450-mediated reactive metabolite of APAP, NAPQI, to critical cell macromolecules (Potter et al., 1974). Although the exact mechanisms by which the reactive metabolite exerts its cytotoxic effects are not completely understood, good correlations are noted between the extent of APAP covalent binding to protein and the severity of hepatic necrosis. In the present investigation, pretreatment with the MFO inhibitor, piperonyl butoxide, significantly alleviated the severity of the APAP-induced necrosis not only in the liver but also in lung and kidney. This suggests that the toxicity in lung and kidney at least in part results from MFO-mediated activation of APAP. In addition, previous studies with hepatectomized animals have demonstrated that radiolabeled APAP can also become covalently bound to proteins in the lung and kidney (Breen et al.. 1982). The present study more definitively addressed this issue by examining the selectivity of tissue protein arylation during APAP overdose. We have recently reported the development of an immunoassay for detecting APAP bound to proteins and demonstrated that the binding of APAP metabolites to hepatic protein was highly selective in vivo (Bartolone et al., 1987). Analysis of the liver extracts from treated mice using SDS-PAGE/Western blotting revealed that APAP-protein adducts with M, of 44 and 58 kDa accounted for approximately 85% of the total immunochemi-

246

BARlOLONE

_)

-+

g

-+-+--f--f-+

ET

AL.

i+e .

-

67

kD

-

58

kD

-

44kD

-

33

kD

-++

FIG. 3. Protein-adducts formed in reconstituted NAPQI generating systems. Cytosol from liver. lung, kidney cortex, spleen. brain, or heart was incubated with washed liver microsomes, 1 mM APAP and an NADPH regenerating system as described under Methods. Each ofthe lanes represent Western blot profiles of the soluble 105,OOOg fraction retreived from the reconstituted systems. In place of tissue cytosol, blood plasma was added to the reconstituted system. Lanes (-) and (+) indicate the absence or addition, respectively, of NADP to the reconstituted systems. As an additional control. APAP was activated by washed liver microsomes in the absence ofany added cytosol and the supernatant from the 105,OOOgcentrifugation following incubation was analyzed immunochemically (lane MICRO). The molecular weights of the predominant APAP-bound proteins are indicated at right.

tally detectable APAP binding. Additional minor protein adducts including those of 25, 33, 78, and 130 kDa were also noted. In the present study, the immunochemical analysis conducted with kidney cortex and lung from mice dosed in vivo revealed that APAP-protein conjugates of 44 and 58 kDa were also detected in microsomes and cytosol of these extrahepatic target organs. In addition the kidney exhibited a prominant cytosolic 33kDa adduct. Other subcellular fractions were not examined in this study and may also contain selectively targetted proteins. Consistent with the histopathological observations, prior treatment of mice with piperonyl butoxide greatly diminished the detectable protein adducts in all target tissues. No arylated proteins were detected in nontarget tissues such as the brain and spleen (data not shown). Hence it can be argued that the tissue necrosis was coincident with the selective formation and/or accumulation of specific APAP-protein adducts.

We and others have studied APAP nephrotoxicity (Placke et al.. 1987: Newton et al., 1983; Mudge et al., 1978; McMurtry et al.. 1978). The nature of the arylating metabolite in the mouse kidney may differ from that reported in the Fischer rat (Newton et al., 1982). In the rat necrosis has been reported to result from the deacetylation of APAP topaminophenol (PAP) with subsequent protein binding of a PAP-derived, P450 independent metabolite (Newton et al., 1983). However, since the MFO inhibitor, piperonyl butoxide, reduced the nephrotoxicity this pathway may not be as important in the CD- 1 mouse. Furthermore, our antibody has a much weaker affinity for PAP than for APAP (Bartolone et al., 1988). If the CD- 1 mouse were similar to the Fischer rat in that the primary APAP-derived adduct in kidney was secondary to APAP deacetylation then there would have been minimal immunochemically detectable binding in the kidney. On the contrary, since the antibody detected adducts in kidney as

SELECTIVE

APAP

BINDING

TO

HEPATIC

AND

EXTRAHEPATIC

PROTEINS

247

readily as in liver, then MFO activation and electrophile. Nonetheless, this observation subsequent binding of the intact APAP derivindicates that monitoring of the plasma for ative is likely to be occurring in the mouse. APAP-protein adducts should provide a senThe present study also addressed the ques- sitive, early, and noninvasive means to assess tion of whether the proteins which become APAP toxicity in humans. The most prominent APAP-protein adarylated upon overdose are normal constituents of target cells only. Addition of cytosol duct observed in plasma after in vivo treatderived from target and nontarget tissues to a ment was of 58 kDa. This is in contrast to the fortified liver microsomal incubate containprofile of adducts formed when plasma was ing APAP resulted in the formation of APAPincubated with APAP and fortified liver miderived protein adducts similar to those ob- crosomes. In the latter case the major adduct served in target organs after oral dosing. This was approximately 67 kDa and would be indicates that, in the susceptible organs, the consistent with arylation of albumin, the identified APAP-protein conjugates of 33 most abundant plasma protein (Brown, and 58 kDa can result from APAP arylation 1976). In contrast, no 67-kDa adduct was deof endogenous normal cell constituents. The tected after in vivo dosing suggesting that alsame conclusion cannot be definitively stated bumin is not targetted to any significant extent in vivo. Moreover, in vitro incubation of for the 44-kDa adduct since in experiments plasma with fortified liver microsomes and conducted without added cytosol this adduct APAP did not result in the formation of the was shown to be formed and released from 58-kDa protein adduct normally observed in microsomes. Therefore any 44-kDa adduct identified in cytosol could have resulted from tissues. Therefore, it is unlikely that this proarylation of either microsomal liver proteins tein target is endogenous to plasma. These or of native cytosolic proteins or both. It is of data imply that NAPQI probably does not esinterest that arylated proteins of approxicape the target cell in unconjugated form and mately 58 kDa were also detected in cytosol suggest that the adducts detected in plasma from heart, spleen, and brain after the in vitro from APAP-treated mice entered through leakage from the targetted tissues. Furtherincubation even though no such arylation was evident 4 hr after APAP in vivo. This sug- more, it is unlikely that the majority of the adducts detected in extracts from treated gests that these tissues, which were not damaged by APAP, may lack adequate APAP ac- mice were the result of adduct transport in tivating capacity. This is consistent with the the blood since no adducts were detected in relative distribution of mixed function oxidaextrahepatic tissues from either spleen, brain, tion among organ systems (Bend and Hook, or heart. This is further supported by the ob1974; Dees et al., 1982). servation that the 33- and 58-kDa proteins Examination of APAP-bound proteins which bind APAP metabolites appear to be present in plasma revealed the presence of native to the cytosol of the tissues in which several adducts by 4 hr following dosing. This these adducts were detected and moreover, is earlier than was reported for elevation of the profile of APAP-bound proteins immusorbitol dehydrogenase activity which did not nochemically detected differ for each target become statistically significant until 8 to 12 tissue. However, these findings do not completely preclude the possibility that the adhr following dosing in the same animal model ducts detected in lung or kidney may have re(Brady et al., 1988). It is not possible to distinguish from the present study whether the sulted from the selective uptake of some arylpresence of the adducts in plasma represents ated liver proteins which had been released as a sensitive indicator of plasma membrane re- a result of the hepatotoxicity. lease or whether it represents the resultant of The prevalence of the 58-kDa binding prothe cells attempt to clear covalently bound tein in target and nontarget tissues suggests

248

BARTOLONE

that this protein may have an important biochemical function. Xenobiotic metabolite binding to proteins can be envisioned to be protective, innocuous, or deleterious to cell viability. We have recently reported that the 58-kDa proteins contain abundant N-ethylmaleimide reactive protein thiols at pH 7.5 and are therefore probably very nucleophilic under physiological conditions. In addition, NAPQI can be shown to bind with high affinity to 58-kDa proteins in vitro using either liver extracts which enzymatically activate APAP or to which chemically synthesized NAPQI was added (Bartolone et al., 1988). These molecular characteristics provide for a highly susceptible cellular target which would bind electrophiles, such as NAPQI, most avidly. The presence of the protein which forms the 58-kDa APAP-protein adduct in both target and nontarget tissues, the detection of this adduct only in tissues undergoing necrosis and the good correlation between the extent of arylation and the ensuing toxicity all strongly suggest that this protein when bound by APAP may play a pivotal role in modulating the cytotoxic process in these target tissues. Future studies will focus on identifying this major APAP-binding protein to better understand the importance of xenobiotic covalent binding to the ensuing cytotoxicity. ACKNOWLEDGMENTS This research is supported by grants from NIH (GM 31460 and ES-07163) the University of Connecticut Research Foundation, and Center for Biochemical Toxicology.

REFERENCES BARTOLONE. J. B., BIRGE, R. B.. SPARKS, K.. COHEN, S. D., AND KHAIRALLAH, E. A. (I 988). Immunochemical analysis of acetaminophen covalent binding to proteins: partial characterization of the major acetaminophen-binding liver proteins. Biochem. Pharmacol. 37,4763-4774. BARTOLONE,

J. B., SPARKS, K.. COHEN, S. D., AND KHAIRALLAH. E. A. (1987). Immunochemical detec-

ET AL. tion ofacetaminophen-bound liver proteins. Biochem. Pharmacol. 36,1193- 1196. BEND, J. R., AND HOOK, G. E. R. (1974). Hepatic and extrahepatic mixed function oxidases. Han&. Physiol. React. Environ. Agents 26,4 19-440. BOYER, T. D., AND ROUFF, S. L. (1971). Acetaminophen-induced hepatic necrosis and renal failure. J. Amer. Med. Assoc. 218,440-44 I. BRADY,J. T., MONTELIUS, D.A., BEIERSCHMITT,~.~., KHAIRALLAH,E.A..ANDCOHEN.S. D.(l988). Effect of piperonyl butoxide post-treatment on acetaminophen hepatotoxicity. Biochem. Pharmacol. 37,2097. BREEN. K.. WANDSCHEER, J., PEIGNOUX, M.. AND PESSAYRE, D. (I 982). In situ formation ofacetaminophen metabolite covalently bound in kidney and lung. Supportive evidence provided by total hepatectomy. Biothem. Pharmacol. 31, 115- 116. BROWN. J. R. (1976). Structural origins of mammalian albumin. Fed. Proc. 35,2 14 1. BUCKPITT. A. R., ROLLINS, D. E.. NELSON, S. D., FRANKLIN, R. B., AND MITCHELL, J. R. ( 1977). Quantitative determination of glutathione. cysteine, N-acetylcysteine conjugates of acetaminophen by HPLC. Anal. Biochem. 83, 168- 177. DAHLIN, D. C.. MIWA, G. T., Lu, A. Y.. AND NELSON, S. D. (1984). N-Acetyl-p-benzoquinoneimine: A cytochrome p-450 mediated oxidation product of acetaminophen. Proc. Natl. .dcad. Sci. USA 81, 13271331. DAVIS, D. C.. POTTER, W. Z.. JOLLOW, D. J., AND MITCHELL, J. R. (1974). Species differences in hepatic glutathione, covalent binding and hepatic necrosis after acetaminophen. Life Sci. 14,2099-2 109. DEES, J. H., MASTERS. B. S., MULLER-EBERHARD, U.. AND JOHNSON, E. F. (1982). Effect of 2.3,7,8tetrachlorodibenzo-p-dioxin and phenobarbital on the occurrence and distribution of four cytochrome p-450 isozymes in rabbit kidney, lung and liver. Cancer Res. 42, 1423-1432. DEVEREUX, T. D.. HOOK, G. E. R., AND Fours. J. R. (1979). Foreign compound metabolism by isolated cells from rabbit lung. Drug Metab. Dispos. 7,70-75. GINSBERG. G. L., PLACKE, M. E., WYAND, D. S., AND COHEN, S. D. (1982). Protection against acetaminophen-induced hepatotoxicity by prior treatment with Fenitrothion. Toxicol. Appl. Pharmucol. 66, 383. JOLLOW, D. J., MITCHELL. J. R.. POTTER. W. Z., DAVIS, D. C., GILLETTE, J. R., AND BRODIE, B. B. (1973). Acetaminophen-induced hepatic necrosis. II. Role of covalent binding in Vivo. J. Pharmacol. Exp. Thera. 187,195-202.

LAEMMLI, U. K. (1970). Cleavage of structural proteins during the assembly ofthe head bacteriophage T4. Nature (London) 227,680-685. MCMURTRY, R. J., SNODGRASS,W. R., AND MITCHELL, J. R. (1978). Renal necrosis, glutathione deple-

SELECTIVE

APAP

BINDING

TO

HEPATIC

tion, and covalent binding after acetaminophen. Toxicol. Appl. Pharmacol. 46,87100. MITCHELL, J. R., JOLLOW, D. J., POTTER, W. Z.. DAVIS, D. C., GILLETTE, J. R., AND BRODIE, B. B. (1973). Acetaminophen-induced hepatic necrosis. I. Role of drug metabolism. J. Pharmacol. E.xp. Thu. 187, 185194.

MITCHELL. J. R., JOLLOW, D. J.. POTTER, W. Z., GILLETTE, J. R., AND BRODIE, B. B. (1973). Acetaminophen-induced hepatic necrosis. IV. Protective role of glutathione. J. Pharmacol. Exp. Ther. 187,2 1l-2 17. MOHANDAS, J.. DUGGIN, G. G., HORVATH, J. S., AND TILLER, D. J. (1981). Metabolic oxidation of acetaminophen mediated by cytocrome P450 mixed-function oxidase and prostaglandin endoperoxidase synthetase in rabbit kidney. Toxicol. Appl. Pharmacol. 61, 252-259.

MUDGE, G. H., GEMBORYS, M. W.. AND DUGGIN, G. (1978). Covalent binding of metabolites of acetaminophen to kidney protein and depletion of renal glutathione. J. Pharmacol. Exp. Thera. 206,2 18-226. NEWTON. J. F., Kuo. C. H., GEMBORYS, M. W., MUDGE, G. H., AND HOOK, J. B. (1982). Nephrotoxicity of paminophenol, a metabolite of acetaminophen

AND

EXTRAHEPATIC

249

PROTEINS

in the Fischer 344 rat. Toxicol.

Appl.

Pharmacol.

65,

336-344.

NEWTON, J. F., YOSHIMOTO, M., BERNSTEIN, J., RUSH, G. F., AND HOOK, J. B. (1983). Acetaminophen nephrotoxicity in the rat. Il. Strain differences in nephrotoxicity and metabolism ofp-aminophenol, a metabolite of acetaminophen. Toxicol. Appl. Pharmacol. 69, 307-318. PETERS, T.. AND REED, R. G. (1980). The biosynthesis of rat serum albumin. J. Biol. Chem. 2553 156-3 163. PLACKE, M. E., WYAND. D. S.. AND COHEN, S. D. (1987). Extrahepatic lesions induced by acetaminophen in the mouse. Toxicol. Pathol. 15,38 l-387. POTTER, W. Z., DAVIS, D. C., MITCHELL, J. R., JOLLOW, D. J., GILLETTE, J. R., AND BRODIE, B. B. (1973). Acetaminophen-induced hepatic necrosis. III. Cytochrome p-450 mediated covalent binding in vitro. J. Pharmacol.

Esp. Ther. 187,203-210.

POTTER, W. Z., THORGEIRRSON, S. S., JOLLOW, D. J., AND MITCHELL, J. R. (1974). Acetaminophen-induced hepatic necrosis. V. Correlation of hepatic necrosis, covalent binding and glutathione depletion in hamsters. Pharmacology 12, 129-143. RUMACK, B. H., AND MATTHEW, I-I. (1975). Acetaminophen poisoning and toxicity. Pedafrics 55,87 l-876.