Hepatic adducts, circulating antibodies, and cytokine polymorphisms in patients with diclofenac hepatotoxicity

Hepatic Adducts, Circulating Antibodies,and Cytokine Polymorphisms in Patients

With Diclofenac HepatotoxicityGuruprasad P. Aithal,1,5 Lesley Ramsay,2 Ann K. Daly,1,3 Nhareet Sonchit,2 Julian B. S. Leathart,1,3 Graeme Alexander,4

J. Gerald Kenna,2,6 John Caldwell,2,7 and Christopher P. Day1

Diclofenac is a nonsteroidal anti-inflammatory drug that causes rare but serious hepatotoxicity,the mechanism of which is unclear. The purpose of the present study was to explore the potentialrole played by the immune processes. Antibodies to diclofenac metabolite-modified liver proteinadducts were detected in the sera of seven out of seven patients with diclofenac-induced hepato-toxicity, 12 of 20 subjects on diclofenac without hepatotoxicity, and none of four healthy con-trols. The antibodies recognized adducts expressed in livers from rats treated with multiple dosesof diclofenac, but not in those given single doses. In addition, several potential diclofenac adductswere identified in the liver of a patient with diclofenac-induced hepatic failure, but not from anormal human donor liver, by immunoblotting with an adduct-selective rabbit antiserum. Todetermine whether or not polymorphisms in genes encoding cytokine-related proteins influencesusceptibility to hepatotoxicity, genotyping for the polymorphisms -627 in the interleukin(IL)-10 gene, -590 in the IL-4 gene, and codon 551 in the IL-4 receptor (IL-4R) were performedon DNA from 24 patients on diclofenac with hepatotoxicity, 48 subjects on diclofenac withouthepatotoxicity, and healthy controls. The frequencies of the variant alleles for IL-10 and IL-4were higher in patients (OR [odds ratio]: 2.8 for IL-10; 2.6 for IL-4; 5.3 for IL-10 � IL-4)compared with healthy controls and subjects on diclofenac without hepatotoxicity (OR: 2.8 forIL-10; 1.2 for IL-4; 5.0 for IL-10 � IL-4). In conclusion, the observed polymorphisms, resultingin low IL-10 and high IL-4 gene transcription, could favor a T helper (Th)-2 mediated antibodyresponse to neoantigenic stimulation associated with disease susceptibility. Supplementary ma-terial for this article can be found on the HEPATOLOGY website (http://interscience.wiley.com/jpages/0270-9139/suppmat/index.html). (HEPATOLOGY 2004;39:1430–1440.)

Diclofenac is a widely used nonsteroidal anti-in-flammatory drug that can cause rare but poten-tially serious hepatotoxicity. Approximately 3.6

per 100,000 users of diclofenac develop severe liver in-jury1 with an 8% case fatality rate.2 Because of its com-mon use, diclofenac hepatotoxicity has been one of themost common causes of hepatic adverse drug reactions,with 180 confirmed cases reported to the U.S. Food andDrug Administration during the first 3 years of market-ing.2 Although the precise mechanism of diclofenac-in-duced hepatotoxicity is unknown, both metabolic2 andimmunological3 mechanisms have been thought to con-tribute to liver injury. A common mechanism that mayunderlie either immunomediated liver injury or meta-bolic idiosyncrasy is the formation of drug-modified pro-tein adducts.4 For several drugs, such as halothane anddihydralazine, antibodies directed against specific hepato-cellular proteins have been identified.5 In each of thesecases, covalent adducts of reactive drug metabolites andhepatocellular proteins have been implicated in triggering

Abbreviations: IL, interleukin; IL-4R, interleukin-4 receptor; OR, odds ratio;Th, T helper; KLH, keyhole limpet hemocyanin; Ig, immunoglobulin; IP, intra-peritoneal; RSA, rabbit serum albumin; TBS, tris-buffered saline; GSA, goat serumalbumin.

From the 1School of Clinical Medical Sciences (Hepatology) and the 3School ofClinical and Laboratory Sciences, Medical School, University of Newcastle, New-castle-upon-Tyne, United Kingdom; the 2Division of Biomedical Sciences, ImperialCollege School of Medicine, London, United Kingdom; and 4Addenbrookes Hospi-tal, Cambridge, United Kingdom.

Received May 23, 2003; accepted February 13, 2004.This work was supported by the EUROHEPATOX Biomed 2 Programme of the

European Union (contract number: BMH4-CT96-0658).Present addresses: 5Queens Medical Centre University Hospital, Nottingham,

United Kingdom; 6Astra-Zeneca plc, Alderley Park, Cheshire, United Kingdom;7Faculty of Medicine, University of Liverpool, Liverpool L69 3GA, United King-dom.

Address reprint requests to: Christopher P. Day, School of Clinical MedicalSciences (Hepatology), Medical School, University of Newcastle, Newcastle-upon-Tyne, Floor 4, William Leech Building, The Medical School, Framlington Place,Newcastle-upon-Tyne, NE2 4HH, United Kingdom. E-mail: [email protected];fax: 44 191-222-0723.

Copyright © 2004 by the American Association for the Study of Liver Diseases.Published online in Wiley InterScience (www.interscience.wiley.com).DOI 10.1002/hep.20205

1430

an immunological response. Studies undertaken with spe-cific polyclonal antisera produced by immunization ofrabbits with synthetic diclofenac–protein conjugates haveshown that diclofenac–protein adducts are expressed inthe livers of rats treated with the drug in vivo and also byrat and human hepatocytes incubated with diclofenac invitro.6–9 Adducts appear to be localized on the bile canalic-ular plasma membrane of hepatocytes.9,10 Diclofenac-treated hepatocytes carry antigenic determinants that arerecognized by T cell and non–T cell–enriched splenocytesderivedfromdiclofenac/keyholelimpethemocyanin(KLH)–immunized mice, resulting in immunomediated destructionof target hepatocytes.11 Antibodies directed against 4�-hydroxydiclofenac glucuronide have been detected in theserum of a patient with diclofenac-induced immune hemo-lytic anemia,12 raising the possibility that humoral immunemechanisms might also be involved in diclofenac hepatotox-icity.

Genetic factors influencing the development of drughepatotoxicity can be grouped into factors affecting theamount of the reactive metabolite—and therefore proteinadduct formed—and factors affecting the immune re-sponse to the adducts. The possibility that polymor-phisms in the cytochrome P450 gene CYP2C9, whichencodes the enzyme responsible for the 4�-hydroxylationof diclofenac, may be associated with diclofenac hepato-toxicity has recently been investigated.13 Possession ofvariant CYP2C9 alleles was not a risk factor for the devel-opment of hepatotoxicity.13 The pattern and magnitudeof the immune response to drug–protein adducts is likelyto vary among individuals. A major determinant of inter-individual variation in immune reactions is likely to be inthe production of immunoregulatory cytokines, such asinterleukin (IL)-10 and IL-4, both of which are encodedby polymorphic genes. IL-10 has diverse immunomodu-lating effects, including the inhibition of antigen-specificactivation, and facilitates cytokine production throughTh-0, Th-1 (IL-2 and interferon-�), and Th-2 (IL-4,IL-5) CD4� lymphocytes.14 It may also exert anti-in-flammatory and antifibrotic effects in the liver.15,16 Sev-eral polymorphisms have been identified in the promoterregion of the IL-10 gene. A C-to-A substitution at posi-tion -627 is present in 21%–23% of healthy individualsand is in linkage disequilibrium with polymorphisms atpositions -854, -1117, and microsatellite loci IL10.G andIL10.R.17–19 In vitro studies and studies in systemic lupuserythematosis, bronchial asthma, and alcoholic liver dis-ease indicate that the -627*A allele is associated withlower transcriptional activity as well as decreased IL-10secretion.18–21

IL-4 is the signature cytokine of Th-2 CD4� cells,which are primarily responsible for humoral immunity.22

IL-4 induces B cells to differentiate and stimulates pro-duction of immunoglobulin (Ig) E as well as non–com-plement-fixing IgG isotypes such as IgG4.22 IL-4 is alsoimportant for the development of Th-2 cells.22 In con-trast, IL-4 down-regulates cytokine production by Th-1cells and inhibits their effector functions.22 A C-to-T ex-change has been identified at position -590 in the pro-moter region of IL-4, and the variant allele has beenassociated with increased transcriptional activity and en-hanced IgE secretion.23 This polymorphism is associatedwith bronchial asthma, atopic dermatitis, and rheumatoidarthritis.14–26 In addition, the gene encoding the IL-4receptor (IL-4R) � subunit is polymorphic.27 A novelIL-4 receptor � allele in which a G-to-A substitution atposition 1902 results in a change from glutamine to argi-nine at position 551 in the IL-4R� protein has been iden-tified, with the variant allele associated with enhancedsignalling activity28 and more common in patients withatopy.29

The potential role of immune mechanisms in thepathogenesis of diclofenac-induced hepatotoxicity hasbeen explored in this study. We have investigated whetheror not diclofenac-modified hepatic proteins could be de-tected in a patient with diclofenac-induced hepatitis andwhether or not antibodies to diclofenac-modified hepaticproteins occur in serum from patients with diclofenachepatotoxicity. We have also assessed the role of allelicvariants in the genes encoding IL-10, IL-4, and the IL-4Ras potential susceptibility factors.

Patients and Methods

Patients. Sera were collected from seven patients re-cruited from Addenbrookes Hospital, Cambridge, UK;Freeman Hospital, Newcastle-upon-Tyne, UK; and theKarolinska Institute, Stockholm, Sweden who were suf-fering from diclofenac-induced hepatotoxicity. Sera werealso collected from 20 patients on diclofenac without hep-atotoxicity and four healthy control individuals not tak-ing diclofenac. An explanted liver was collected from apatient with diclofenac-induced hepatic failure who un-derwent liver transplantation, and a liver biopsy was alsotaken from the donor liver.

For the genetic association studies, 24 patients (19 fe-male, 79%), aged 24 to 70 (mean: 50.8) years who hadsuffered diclofenac hepatotoxicity between 1990 and1999 were identified from the records of the FreemanHospital and adverse drug reaction reports to the Com-mittee on Safety of Medicines, UK. From the case notes,demography, medical history, physical findings, fullblood count, liver function tests (albumin, bilirubin, ala-nine aminotransferase, alkaline phosphatase, and �-glu-

HEPATOLOGY, Vol. 39, No. 5, 2004 AITHAL ET AL. 1431

tamyl transferase), hepatitis B and C serology, auto-antibody screen, liver ultrasound scans, and liver biopsyreports were extracted. The clinical details of these pa-tients are summarized in Table 1. Six patients presentedwith jaundice, three presented with hepatic failure (num-bers 10, 11, and 20), and the rest had nonspecific symp-toms associated with raised liver enzymes. Two patients(numbers 16 and 17) had peripheral eosinophilia in asso-ciation with their hepatic adverse drug reaction. Thecausal relationship of liver injury to diclofenac was estab-lished using international consensus criteria as described

previously.30,31 When patients were taking more than onedrug, causality assessment was performed with regard toeach drug, and the best fit was determined.

Controls. Forty-eight Caucasian subjects (35 female,75%), aged 22 to 77 (mean: 52.3) years who were takingdiclofenac for 0.3 to 20 (mean: 4) years without develop-ing hepatotoxicity were recruited from the outpatientclinic of the Freeman Hospital. The underlying diagnoseswere rheumatoid arthritis in 34 patients, psoriatic arthri-tis in seven, osteoarthritis in four, ankylosing spodylosis intwo, and juvenile arthritis in one. Healthy Caucasian con-

Table 1. Demographic and Clinical Details of Patients With Hepatotoxicity Recruited Into the Genotyping Study

PatientNo.

Age(y) Sex Time*

UnderlyingDiagnosis

Peak LaboratoryTests†

Liver BiopsyFindings Other Drugs

1 56 F 2 Rheumatoidarthritis

Bil 59, ALT 1800,ALP 1950

None

2 44 F 0.3 Osteoarthritis ALT 457, ALP 287 None3 39 F 1 Arthralgia ALT 196, ALP 235 None4 70 F 1.5 Paget’s disease Bil 112, ALT 1000,

ALP 880Perivenular and

bridging necrosisNone

5 54 F 0.5 Frozen shoulder Bil 98, ALT 412,ALP 366

None

6 65 F 2 Osteoarthritis ALT 109, ALP 142 Portal inflammation,eosiniphilinfiltration, spottynecrosis

None

7 61 M 4 Sciatica ALT 182, ALP 220 None8 46 M 1 Sciatica ALT 99, GGT 157,

ALP 79Portal inflammation,

eosinophilinfiltration, spottynecrosis

None

9 52 F 8 Backache ALT 372, ALP 210 None10 24 F 1 Rheumatoid

arthritisBil 114, ALT 550,

ALP 367Confluent necrosis Phenelzine

11 53 F 10 Sciatica Bil 314, ALT 808,ALP 288

Confluent necrosis Dextropropoxyphene

12 54 F 1 Rheumatoidarthritis

Bil 94, ALT 961,ALP 161

Nebametone,Dextropropoxyphene

13 62 F 0.8 Osteoarthritis ALT 190, ALP 108 None14 35 M 0.8 Trauma ALT 148, ALP 165 Dextropropoxyphene15 54 F 12 Osteoarthritis ALT 97, ALP 105 Portal and lobular

inflammationNorethisterone

Cimetidine16 57 F 24 CREST

SyndromeALT 111, ALP 423 Portal and lobular

inflammationNorethisterone

17 39 F 3 Postoperativepain

ALT 207, ALP 145 Contraceptive pill

18 38 M 2 Trauma ALT 188, ALP 201 Ibuprofen19 45 F 12 Backache ALT 138, ALP 401 None20 52 F 3 Arthralgia Bil 379, ALT 498,

ALP 504Paracetamol

21 58 F 5 Osteoarthritis ALT 155, ALP 103 None22 64 M 12 Osteoarthritis ALT 142, ALP 130 Portal inflammation,

eosinophilinfiltration

Sulfasalazine,Indomethacin

23 70 F 0.5 Osteoarthritis Bil 47, ALT 379,ALP 442

Dextropropoxyphene

24 27 F 0.5 Dysmenorrhoea Bil 108, ALT 130,ALP 633

Mefenamic acid

Abbreviations: Bil, bilirubin; ALT, alanine aminotransferase; ALP, alkaline phosphatase.*Duration of diclofenac treatment in months at the time of hepatotoxicity.†Only abnormal values are listed: Bil, �mol (�17); ALT, U/L (�50); ALP, U/L (40–120).

1432 AITHAL ET AL. HEPATOLOGY, May 2004

trols were recruited from the same area of England. Thestudy was approved by the Newcastle-upon-Tyne JointEthics committee, and all subjects gave informed consent.

Animal Dosing Experiments. In the single dosingstudies, male Sprague-Dawley rats (250 g, n � 4–7 pergroup) received single intraperitoneal (IP) doses of di-clofenac at 200 mg/kg (dissolved in normal saline, 0.5mL/rat). Control rats received equivalent amounts of nor-mal saline. Animals were killed at 6 hours and liversubcellular fractions were prepared by differential cen-trifugation. Some groups of animals (n � 3–6) were pre-treated with inducers or inhibitors of drug metabolizingenzymes before administration of diclofenac at 100 mg/kg. Phenobarbital was given IP at a daily dose of 75 mg/kg/d in saline for 4 days; rats were then treated on day 5.Borneol was administered IP as a single dose of 750 mg/kgin tricaprylin, while galactosamine-HCl was administeredas a single IP dose of 600 mg/kg in saline, following whichrats received an IP dose of diclofenac after 30 minutes. Inmultiple dosing studies, male Sprague-Dawley rats (250g, n � 4–7 per group) received daily doses of diclofenac at30 mg/kg (dissolved in normal saline, 0.5 mL/rat) for 5days. Animals were then killed 6 hours after the final doseand liver subcellular fractions were prepared.

Subcellular Fractionation. Human liver fractionswere prepared from an explanted liver from a patient withdiclofenac-induced hepatic failure and a liver biopsy froma normal donor liver. Both animal and human liver frac-tions were prepared in the same manner; all steps tookplace at 4°C. Livers were minced in three volumes ofice-cold sucrose buffer (0.25 M sucrose, 15 mM Tris-HCl, 0.1 mM ethylenediaminetetraacetic acid pH 6.8)and subsequently homogenized using five strokes of a Pot-ter homogenizer (Wheaton Science Products, Milleville,NJ) at 800 rpm. Homogenates were then filtered throughtwo layers of muslin and nuclear (600 � gav pellet), mi-tochondrial (10,000 � gav pellet), microsomal(100,000 � gav pellet) and cytosolic fractions (100,000 xgav supernatant) were prepared by sequential centrifuga-tion, with each pellet washed twice and resuspended insucrose buffer using a Dounce homogenizer (WheatonScience Products) followed by further centrifugation.42

The resulting fractions were then aliquoted and stored at�70°C.

Immunoblotting. Immunoblotting was carried outusing rabbit polyclonal anti-diclofenac KLH antisera10

and sera from patients with diclofenac hepatotoxicity,subjects on diclofenac without hepatotoxicity, and con-trols. The sera (dilution 1:1000 for both diclofenac anti-body and patients sera) were used to probe nuclearfractions from diclofenac-treated and control rats (50 �g/lane) as described by Wade et al.10 Anti–rabbit and anti–

human IgG horseradish peroxidase secondary antibodieswere used at 1:10,000 dilution followed by enhanced che-moluminescence development.

Immunoblotting was also employed to detect the pres-ence of diclofenac-modified proteins in hepatic subcellu-lar fractions from the patient with diclofenac-inducedhepatic failure. Subcellular fractions prepared from a nor-mal donor liver were used as controls. Methods outlinedby Wade et al. were employed.10 Subcellular fractionswere run at a concentration of 10 �g/lane and screenedwith a primary diclofenac antibody dilution of 1:1000and a secondary antibody dilution of 1:10,000, followedby development using enhanced chemoluminescence.Desitometric scanning was performed using a CS-930dual wavelength scanner coupled to a DR-2 data recorder(Shimadzu, Kyoto, Japan).

Preabsorption of Patients’ Sera. Patients’ sera at1/20 dilution were incubated at 4°C for 30 minutes, withcontinuous shaking, with control rat liver nuclear frac-tions. Samples were then centrifuged at 10,000g for 20minutes, and the pellets were discarded. This procedurewas repeated twice using either fresh nuclear fractionsfrom rats that had received multiple doses of diclofenac ornuclear fractions from control rats (negative control).

Blocking Studies. Anti-diclofenac KLH antiserumwas incubated for 3 hours at 1:1000 dilution with eitherdiclofenac–rabbit serum albumin (RSA) conjugate orRSA—both of which were at 25 ng/mL—before immu-noblots were probed.10 Patients’ sera were incubated over-night at 4°C at 1:1000 dilution with 50 mM glycine ordiclofenac and were then used to probe immunoblots.

Immunohistochemistry. Sections from the left lobe ofthe explanted liver were fixed in formol saline for 24hours, embedded in paraffin, and then 5-�m sectionswere cut. Sections were dewaxed by sequential incuba-tions in xylene (2 � 15 minutes), 100% ethanol (2 � 10minutes), 90% ethanol (1 � 3 minutes), 75% ethanol(1 � 3 minutes), and 50% ethanol (1 � 3 minutes) andrinsed with tris-buffered saline (TBS: 0.2 M sodium chlo-ride, 50 mM Tris-HCl pH 7.4). The slides were incu-bated for 30 minutes in 3% hydrogen peroxide to inhibitendogenous peroxidase activity and rinsed again withTBS. Nonspecific binding was blocked by incubationwith 10% goat serum albumin (GSA) in TBS (10% GSA/TBS) for 1 hour. The sections were incubated for 2 hoursat room temperature with anti-diclofenac KLH antiserumat 1:750 dilution in 10% GSA/TBS and washed withTBS (2 � 5 min). Subsequent incubations were under-taken using reagents from Vector Laboratories (Peterbor-ough, UK). Slides were incubated for 30 minutes at roomtemperature with biotinylated goat anti–rabbit IgG at1:500 dilution in 5% GSA/TBS, washed with TBS (2 �


5 min), and incubated for 30 minutes with Streptavidin–horse radish peroxidase at 1:200 dilution in 5% GSA/TBS. After two further 5-minute washes with TBS, slideswere developed using 3,3�-diaminobenzidine reagent andcounterstained for 10 seconds with hemotoxylin (Gill’sformula). Slides were air dried and glass cover slips weremounted using DPX mounting medium.

Genotyping. Blood (10 mL) was collected from eachsubject and DNA was extracted as described by Daly etal.33 Genotyping for the -627 IL-10 polymorphism wasperformed as described by Grove et al.19 Genotyping forthe -590 IL-4 polymorphism was performed by minormodification of the procedure described by Rosenwasseret al.23 Primers IL4P590F (5� -GTTGTAATGCAGTC-CTCCTG-3�) and IL-4P590R (5� -ACTAGGCCT-CACCTGATACG-3�) were used for polymerase chainreaction amplification. The temperature conditions con-sisted of 30 cycles of 94°C for 1-minute denaturation,55°C for 1-minute annealing, and 72°C for 1-minuteextension followed by 1 cycle at 70°C for 10 minutes. A280-bp polymerase chain reaction product was digestedwith BsmF1 at 37°C and analyzed by electrophoresis onan agarose gel. Genotyping for the IL-4R Q576R poly-morphism was performed as described by Aithal et al.34

Statistical Analysis. The significance of the differ-ences between groups was assessed using either the �2-testor Fisher’s exact test depending on the sample size.

Results

Detection of Diclofenac–Protein Adducts in Ratand Human Livers. Diclofenac–protein adducts weredetected immunochemically using a polyclonal antiserumraised by immunizing rabbits with a synthetic diclofenac–KLH conjugate. The production and characterization ofthe antiserum has been described previously.10 Rabbitpolyclonal antiserum recognized a major 110 kDa proteinadduct expressed in the livers from rats receiving singledoses of diclofenac, but not in the control rat liver (Fig.1A). In addition, the antiserum recognized several less-abundant protein adducts (30, 144, and 200 kDa). Thevarious diclofenac adducts exhibited dose-dependant ex-pression in livers of diclofenac-treated rats, but not con-trol rats, and therefore could be distinguished from non–adduct-modified liver proteins that were recognized bythe antiserum in control rat liver (Fig. 1A). Recognitionof the adducts by the antiserum was inhibited in the pres-ence of diclofenac (Fig. 1B). Previously it has been shownthat adduct recognition is not inhibited in the presence ofequivalent concentrations of indomethacin, tolmetin,fenoprofen, naproxen, sulindac, or ibuprofen.10

Immunoblotting analysis of subcellular fractions re-vealed that the diclofenac adducts were present in thenuclear fraction (Supplemental Fig. 1). Immunohisto-chemical studies further demonstrated that diclofenac ad-ducts were localized to the bile canalicular domain ofhepatocytes (Fig. 2A).

To investigate the role played by metabolism in adductformation, rats were pre-treated with modulators of drugmetabolizing enzymes before administration of diclofe-nac. A marked decrease in adduct expression was observedwhen rats were pretreated with the glucuronidation inhib-itors borneol (63% decrease) or galactosamine (67% de-crease).32 Adduct formation was also decreased (by 46%,as assessed by densitometry) following pretreatment ofrats with phenobarbitone (Fig. 3) which is a potent in-ducer of various cytochrome P450 isoforms. These resultsindicate that the formation of the diclofenac–protein ad-ducts recognized by the antiserum requires glucuronida-tion of diclofenac into reactive acyl glucuronidemetabolites.

The antidiclofenac adduct antiserum was used toprobe subcellular fractions obtained from the liver of apatient with diclofenac-induced hepatic failure and froma normal (control) donor liver. The antiserum recognizedseveral proteins that were expressed in the liver from thepatient but not in the control liver. These proteins wereconcentrated in the nuclear fraction (Fig. 4A). The recog-nition of proteins of molecular mass 130, 216, and 240

Fig. 1. Detection by immunoblotting of protein adducts expressed inlivers of rats treated with single doses of diclofenac. (A) Livers from ratstreated IP with single doses of diclofenac were homogenized and nuclearfractions were prepared by differential centrifugation. Immunoblots wereprobed using antidiclofenac adduct rabbit antiserum. The positions of themajor protein adducts (200, 144, 110, and 30 kd) are indicated. (B)Inhibition of the recognition of the diclofenac adducts by addition of 100�M of diclofenac to the antidiclofenac adduct rabbit antiserum. body wt,body weight.


kDa was abolished in the presence of RSA–diclofenacconjugate, but not by RSA alone (Fig. 4B and C), imply-ing that they correspond to diclofenac adducts. Recogni-tion of these putative human liver adducts was notobserved when the immunoblots were probed in the ab-sence of primary antiserum or with pre-immune serum(data not shown).

Protein adducts derived from diclofenac were also de-tected by immunoblotting in livers from rats treated dailywith the compound for 5 days at 30 mg/kg. However, themajor adduct had a molecular mass of 97 kDa (not 110kDa, as in rats that were given a single dose), and severalless-abundant adducts were also seen, including an 85kDa adduct (Fig. 5A). Recognition of 97- kDa and 85kDa adducts was inhibited in the presence of diclofenac(Fig. 5B).

Detection of Human Antibodies to Diclofenac Ad-ducts. Sera from seven patients with diclofenac hepato-toxicity, 20 subjects who received diclofenac withoutdeveloping hepatotoxicity, and four healthy controls werescreened against nuclear fractions from rats treated chron-ically with diclofenac. Details of clinical presentation and

laboratory tests of the seven patients with hepatotoxicityare shown in Table 2. Three of these patients presentedwith jaundice, two presented with hepatic failure (num-bers 25 and 27), and two presented with raised liver en-zymes. Four of these patients (numbers 21–24) alsoparticipated in the genotyping study. All seven patientswith the drug-associated liver injury had serum antibodiesthat recognized protein antigens expressed in nuclear frac-tions from livers of rats treated with multiple doses ofdiclofenac, but not in fractions from control rat liver orfrom livers of rats treated with a single dose of diclofenac(Fig. 6). All of the patients’ sera recognized a diclofenac-induced protein antigen of 85 kDa and all but three alsorecognized a 97 kDa diclofenac-induced antigen. Severalsera recognized additional diclofenac-induced antigens(Fig. 6, Supplemental Fig. 2, and Table 2). Interestingly,the 85 kDa and 97 kDa antigens comigrated on sodiumdodecyl sulfate polyacrylamide gels with diclofenac ad-ducts recognized by the rabbit antiserum described previ-ously (Fig. 5). Antibodies that recognized the 85 kDaand/or 97 kDa diclofenac-induced antigen were alsopresent in sera from subjects treated with diclofenac with-out hepatotoxicity, while two of these sera also recognizeda 55 kDa diclofenac-induced antigen (Table 3). The in-cidence of antibodies to diclofenac-induced antigens waslower in subjects treated with diclofenac without hepato-

Fig. 2. Immunohistochemical detection of protein adducts expressedin livers of rats treated with diclofenac. Liver sections from rats treated IPwith single doses of (A) diclofenac at 200 mg/kg body weight or (B)vehicle alone were probed using antidiclofenac adduct rabbit antiserum.The antiserum recognized adducts expressed on the canalicular surfaceof hepatocytes and in hepatocytes immediately adjacent to the centralvein (A). (Original magnification �400.)

Fig. 3. Decreased expression of diclofenac adducts in rats treatedwith modulators of metabolism. Rats received no pretreatment (none) orwere treated with phenobarbitone (PB), galactosamine (Gal), or borneol(Born). Animals then received a single IP dose of diclofenac at 200mg/kg body weight (D) or vehicle control (C), and immunoblots of livernuclear fractions were probed using antidiclofenac adduct rabbit anti-serum.


toxicity (12/20) than in the patients with diclofenac hep-atotoxicity (7/7; P � .07). Antibodies that recognizeddiclofenac-induced liver antigens were not detected in thefour healthy control sera (P � .003) (Fig. 6 and Supple-mental Fig. 2).

To test the specificity of interaction between the pa-tients’ sera and the diclofenac-induced liver protein anti-gens, competitive inhibition studies were undertaken inthe presence of 50 mM of diclofenac. As a control fornonspecific ionic inhibition, 50 mM of glycine was alsoincubated with patient sera. Diclofenac inhibited anti-body binding, while glycine failed to show any inhibition(Supplemental Fig. 3).

Genotyping for Cytokine Polymorphisms. Therewas a higher frequency of patients with diclofenac hepa-totoxicity with one or two A alleles at position -627 in theIL-10 gene (AA/AC genotype, n � 14/24) when com-pared with healthy controls (n � 75/227; P � .02; OR2.84 [range: 1.20–6.69]) as well as subjects on diclofenacwithout hepatotoxicity (n � 16/48; P � .07; OR 2.80[range: 1.02–7.68]) (Table 4). Similarly, there was anincreased frequency of subjects with one or more T allelesat position -590 in the IL-4 gene (TT/CT genotype, n �8/24) when compared with healthy controls (51/321;P � .04; OR 2.6 [range: 1.1–6.5]), but this was notsignificant compared with subjects on diclofenac withouthepatotoxicity (n � 14/48; P � .78; OR 1.2 [range:0.42–3.47]) (Table 4). Six out of 24 patients had at least

one variant allele for both the IL-10 -627 polymorphismand the IL-4 -590 polymorphism when compared with10/167 in the healthy control group (P � .007; OR 5.3[range: 1.7–16.3]) and 3/48 in subjects on diclofenacwithout hepatotoxicity (P � .05; OR 5.0 [range: 1.12–22.19]). There were no significant differences in the fre-quency of variant genotypes between those presentingwith jaundice or liver failure and those with raised liverenzymes (IL-10: 3/9 vs. 11/15, P � .09; IL-4: 2/9 vs.6/15, P � .66; and IL-10 � IL-4: 2/9 vs. 5/15, P � .66)or between those developing hepatotoxicity after less than3 months of diclofenac intake and those presenting after alonger period (IL-10: 9/16 vs. 5/8, P � .10; IL-4: 4/16 vs.4/8, P � .36; and IL-10 � IL-4: 4/16 vs. 3/8, P � .65).

There was no significant difference in the number of pa-tients with one or more Q551R alleles of the IL-4R� gene(n � 9/24) when compared with healthy controls (51/162;P � .64) and the group on diclofenac without hepatotoxicity(15/48; P � .61) (Table 4). Four patients had genotypingand serum identification of antidiclofenac antibodies (num-bers 21–24 in Tables 1 and 2), and three out of four pos-sessed at least one variant IL-10 or IL-4 allele.

Fig. 4. Expression of diclofenac adducts in human liver. Nuclear(Nuc), mitochondrial (Mito), microsomal (Mic), and cytosolic (Cyt) frac-tions were prepared from livers of a patient with diclofenac hepatitis (D)and from a normal liver donor (C). (A) Immunoblots were probed usingantidiclofenac adduct antiserum. (B) Nuclear fractions were probed usingthe antiserum in the presence of 25 ng/mL RSA. (C) Nuclear fractionswere probed using the antiserum in the presence of 25 ng/mL RSA–diclofenac conjugate (RSA-Dic). The positions of the major diclofenacadducts (130, 216, and 240 kd) are indicated. RSA, rabbit serumalbumin.

Fig. 5. Altered expression of protein adducts in rats treated withmultiple doses of diclofenac. (A) Nuclear fractions were prepared fromlivers of rats treated daily for 5 days with diclofenac (Multiple) at 0 or 30mg/kg body weight or from livers of rats given a single dose of diclofenac(Single) at 0 or 200 mg/kg. Immunoblots were probed using antidiclofe-nac adduct rabbit antiserum. (B) Nuclear fractions from livers of ratstreated daily for 5 days with diclofenac at 0 or 30 mg/kg body weightwere probed using antidiclofenac adduct rabbit antiserum in the pres-ence of 50 mM of diclofenac. body wt, body weight.


DiscussionThese results have provided evidence that diclofenac-

modified proteins are formed in human liver in associa-tion with hepatotoxicity and that these adducts elicit aselective antibody response. The immunoblotting studiesperformed using an adduct-selective rabbit antiserumidentified several proteins expressed in the explanted liverfrom a patient with diclofenac-induced hepatic failurethat could comprise diclofenac adducts (as indicated by

blocking studies using free drug). In addition, serum an-tibodies that recognized rat liver diclofenac–protein ad-ducts were detected in all of the patients with diclofenachepatotoxicity tested. The major adducts recognized bythe patients’ antibodies (97 kDa and 85 kDa adducts)were also recognized by the diclofenac adduct–selectiverabbit antiserum. Although the evidence for diclofenacadduct formation in vivo in humans was derived from asingle patient, it is reasonable to infer that adduct forma-tion is a frequent event, because adduct specific antibodieswere commonly detected in patients’ sera. The nature ofthe diclofenac-modified target proteins recognized by thepatients’ antibodies remains to be determined. However,

Table 2. Proteins Identified by Sera From Seven Patients on Diclofenac With Hepatotoxicity

PatientNo.

Age(y) Sex Time*

UnderlyingDiagnosis

Peak LaboratoryTests† Liver Biopsy

ProteinsRecognized (kd)

21 58 F 5 Osteoarthritis ALT 155, ALP 103 97, 8522 64 M 12 Osteoarthritis ALT 142, ALP 130 Portal inflammation,

eosinophil infiltration110, 97, 85

23 70 F 0.5 Osteoarthritis Bil 47, ALT 379,ALP 442

85

24 27 F 0.5 Dysmenorrhoea Bil 108, ALT 130,ALP 633

85

25 41 M 0.7 Arthralgia Bil 219, ALT 113,ALP 1260

Confluent necrosis 144, 85

26 51 M 4 Gout Bil 119, ALT 1400 Portal, lobular inflammation,eosinophil infiltration,portal fibrosis

97, 85, 55

27 18 F 0.3 Arthralgia Bil 300, ALT 635 Submassive necrosis 200, 144, 110, 97,85

Abbreviations: ALT, alanine aminotransferase; ALP, alkaline phosphatase; Bil, Bilirubin.*Duration of diclofenac treatment in months at the time of hepatotoxicity.†Only abnormal values are listed: ALT, U/L (�50); ALP, U/L (40–120); Bil, �mol (�17).

Fig. 6. Antibodies to diclofenac adducts in patients with diclofenac-associated liver injury. Nuclear fractions were prepared from livers of ratstreated daily for 5 days with diclofenac (Multiple) at 0 or 30 mg/kg bodyweight or from livers of rats given a single dose of diclofenac (Single) at0 or 200 mg/kg. Immunoblots were probed using (A, B) sera frompatients with diclofenac-associated liver injury (patients 23 and 22 inTable 2) and (C) serum from a normal control individual. The position ofdiclofenac adducts, which were expressed only in livers from rats givenmultiple doses of diclofenac, are indicated. body wt, body weight.

Table 3. Proteins Identified by Sera From 20 Subjects onDiclofenac Without Hepatotoxicity

Patient No. Proteins Recognized (kd)

1 97, 852 973 None4 None5 None6 977 97, 85, 558 None9 97, 85

10 97, 8511 97, 85, 5512 97, 8513 97, 8514 None15 97, 8516 97, 8517 None18 None19 97, 85, 5520 None


our in vivo experiments in rats indicate that this requiresformation of acyl glucuronides. The latter has been pre-viously shown to mediate protein adduct formation inhepatocytes in vivo.7 Hence acyl glucuronide formationmay be responsible for protein adduct formation and en-suing antibody response in humans.

Circulating antibody directed against 4�-hydroxydi-clofenac glucuronide has also been detected in diclofenac-induced immune haemolytic anemia.12 This raises thepossibility that antibodies to diclofenac adducts couldplay a role in the mechanism of hepatotoxicity. The ob-servation that similar antibodies were also present in serafrom 60% of subjects who had not developed hepatotox-icity on diclofenac does, however, suggest that antibodyproduction alone may not result in a clinically significanthepatotoxicity. One explanation for the appearance ofantibodies in the absence of obvious hepatotoxicity is thata mild direct toxicity of the drug metabolite may be aprerequisite for immunomediated liver injury to developin some of the affected individuals. It has been hypothe-sized that a mild toxic effect of the drug metabolite, whichleads to the release of drug adducts, would increase thelikelihood of an immunization reaction.35 Indeed, drugssuch as halothane that produce immunoallergic hepatitisin a few patients also lead to a mild increase in serumtransaminase activity in a much larger proportion of re-cipients.36 Similarly, up to 15% of patients taking diclofe-

nac develop mild increases in serum transaminases37 andasymptomatic threefold increases of transaminases wereseen in 5% of subjects during premarketing testing of thedrug,2 although clinically significant hepatotoxicity israre. This implies that multiple steps are involved in thedevelopment of diclofenac hepatotoxicity.

In the case–control study, we have shown an associa-tion between variant alleles of the IL-10 and IL-4 genesand diclofenac hepatotoxicity, providing further evidencefor immune mechanisms in the pathogenesis of liver dis-ease. The A allele at position -627 in the IL-10 gene pro-moter region and the T allele at position -590 in the IL-4promoter region were more common in patients withhepatotoxicity than healthy controls as well as age- andgender-matched subjects on diclofenac without hepato-toxicity. Although there was no significant difference be-tween the frequencies of variant IL-4 allele in patientswith hepatotoxicity compared with those on diclofenacwithout hepatotoxicity, the combination of variant IL-10and IL-4 alleles was more frequent in patients with hepa-totoxicity compared with both groups of controls. Thevariant genotype would predict the production of lowerlevels of IL-10 and higher levels of IL-4 by stimulatedcells. Low IL-10 could increase the antigen presentationof diclofenac-related neoantigens by monocytes and leadto the subsequent activation of T cells and immunomedi-ated liver injury. High levels of IL-4 promote a Th-2–mediated immune response and induce B celldifferentiation.22 The resulting enhanced production ofantibodies against diclofenac-related neoantigen may in-crease the likelihood of liver injury. It is possible, there-fore, that mild direct hepatotoxicity and the subsequentrelease of low levels of diclofenac–protein adducts occursquite commonly, but that severe hepatotoxicity only oc-curs when these adducts elicit a strong immune responsedue to an individual’s “immune” genotype.

Direct correlation between specific genotypes and an-tidiclofenac antibody formation could not be establishedin our study. The majority of patients recruited for thegenetic association study were identified retrospectivelyafter the hepatic adverse reaction had resolved and weretherefore not suitable for the study of antidiclofenac an-tibodies. Only sera collected during an episode of diclofe-nac-induced hepatotoxicity were included in theimmunoblotting assays, and many of these subjects didnot participate in the genotyping studies. Hence only fourpatients (numbers 21–24 in Tables 1 and 2) had bothgenotyping and serum identification of antidiclofenac an-tibodies performed, and three of these possessed variantalleles for IL-10 and/or IL-4 genes. However, this sub-group is too small for statistical analysis. The immuno-blotting methods used to identify diclofenac antibodies

Table 4. Genotype Frequencies in Patients With DiclofenacHepatotoxicity Compared With Controls

GenotypeWild-type/Wild-type

Wild-type/Mutant

Mutant/Mutant

IL-10-627Healthy controls (n � 227) 152 (67%) 70 (31%) 5 (2%)Subjects on diclofenac (n � 48) 32 (67%) 13 (27%) 3 (6%)Cases (n � 24) 10 (42%) 12 (50%) 2 (8%)

IL-4-590Controls (n � 321) 270 (84%) 46 (14%) 5 (2%)Subjects on diclofenac (n � 48) 34 (71%) 14 (29%) 0Cases (n � 24) 16 (67%) 8 (33%) 0

IL-4R codon 551Controls (n � 162) 111 (69%) 34 (21%) 17 (10%)Subjects on diclofenac (n � 48) 33 (69%) 14 (29%) 1 (2%)Cases (n � 24) 15 (62%) 9 (38%) 0

NOTE: Odds ratio for possession of IL-10-627A: 2.8 (95% CI 1.2–6.7; P � .02[compared with healthy controls]; 2.8 (95% CI 1.02–7.68; P � .07 [comparedwith subjects on diclofenac]).

Odds ratio for possession of IL-4-590T: 2.6 (95% CI 1.1–6.5; P � .04[compared with healthy controls]; 1.2 (95% CI 0.42–3.47; P � .78 [comparedwith subjects on diclofenac]).

Odds ratio for possession of IL-4R R variant: 1.3 (95% CI 0.54–3.18; P � .64[compared with healthy controls]; 1.32 (95% CI 0.47–3.69; P � .61 [comparedwith subjects on diclofenac]).

Odds ratio for possession of IL-10-627A and IL-4-590T: 5.3 (95% CI 1.7–16.3;P � .007 [compared with healthy controls]; 5.0 (95% CI 1.12–22.19; P � .05[compared with subjects on diclofenac]).


do not allow any estimation of the magnitude of the im-mune response. Hence we have not been able to comparethe titres of antibody formation in subjects taking diclofe-nac without hepatotoxicity with those who had an adversereaction.

It could be argued that the high frequency of variantalleles in the patient group might represent an associationof cytokine polymorphism with the underlying disease(rather than hepatotoxicity) for which patients were pre-scribed diclofenac. The hepatotoxicity patients (Table 1)were taking diclofenac for a variety of indications. Osteo-arthritis accounted for 7/24 (29%) of the cases. Immunemechanisms have not been implicated in the pathogenesisof osteoarthritis, and the disease has not been associatedwith any cytokine polymorphisms. Moreover, the fre-quency of variant alleles among patients who had osteo-arthritis was similar to that in the rest of the group (IL-10:6/7 vs. 8/17, P � .2; IL-4: 2/7 vs. 6/17, P � 1.0). Inaddition, the association of diclofenac hepatotoxicity withthe variant IL-10 allele and the combination of the variantIL-10 and IL-4 alleles persisted when compared with agroup that took diclofenac without developing hepato-toxicity. The strengths of these associations with hepato-toxicity (i.e., the odds ratios) were similar when comparedwith the two control groups. It should be noted that theunderlying diagnoses in the group on diclofenac withouthepatotoxicity were different than those in patients withdiclofenac hepatotoxicity. This reflects the fact that theformer group was recruited from hospital-based practice,where patients with osteoarthritis are not often referredunless it is for complications such as hepatotoxicity. In thediclofenac control group, 70% of patients were sufferingfrom rheumatoid arthritis compared with 12% of thehepatotoxicity patients. It is possible that cytokine poly-morphisms might be risk factors for rheumatoid arthritisas well as hepatotoxicity, offering an explanation for thesmaller differences seen for some genotypes between casesand diclofenac controls compared with those betweencases and healthy controls. In addition, the anti-inflam-matory (rather than the immunoregulatory) properties ofIL-10 might be important in drug-induced liver injury15

as demonstrated by the susceptibility of IL-10 “knockout”mice to acetaminophen-induced hepatotoxicity.38 There-fore, the association of cytokine polymorphisms with di-clofenac hepatotoxicity demonstrated in this study maybe independent of the immune mechanisms described.

The association of cytokine polymorphisms with idio-syncratic drug hepatotoxicity is a novel finding. Polymor-phisms in the major histocompatibility complexmolecules have previously been shown to modify suscep-tibility to drug-induced hepatotoxicity as they determinethe efficient presentation of alkylated immunogenic pep-

tides.38 Associations have been reported between HLA11and diclofenac hepatotoxicity, which is consistent with animportant role for the immune system in determiningsusceptibility to this disease.39 The current observationsare based on relatively small groups and need to be con-firmed in a larger study, but they point to the possibility ofidentifying individuals who may be at increased risk ofadverse drug reactions by genotyping for immune systempolymorphisms.

Acknowledgment: We are grateful to Dr. Eric Elias-son, Karolinska Institute, Stockholm, Sweden, and theRegional Centre of the Swedish Medical Products Agencyin the Division of Clinical Pharmacology at HuddingeUniversity for providing the serum samples used in thisstudy.

References1. Rodriguez LAG, Williams R, Derby LE, Dean AD, Jick H. Acute liver

injury associated with non-steroidal anti-inflammatory drugs and the roleof risk factors. Arch Intern Med 1994;154:311–316.

2. Banks AT, Zimmerman HJ, Ishak KG, Harter JG. Diclofenac-associatedhepatotoxicity: analysis of 180 cases reported to the Food and Drug Ad-ministration as adverse reactions. HEPATOLOGY 1995;22:820–827.

3. Scully LJ, Clarke D, Barr RJ. Diclofenac-induced hepatitis in 3 cases withfeatures of autoimmune chronic active hepatitis. Dig Dis Sci 1993;8:744–751.

4. Boelsterli UA, Zimmerman HJ, Kretz-Rommel A. idiosyncratic liver tox-icity of non-steroidal anti-inflammatory drugs: molecular mechanisms andpathology. Crit Rev Toxicol 1995;25:207–235.

5. Beaune P, Pessayre D, Dansette PM, Mansuy D, Manns M. Autoantibod-ies against cytochrome P450: role in human diseases. Adv Pharmacol 1994;30:199–245.

6. Pumford NR, Myers TG, Davila JC, Highet RJ, Pohl LR. Immunochemi-cal detection of liver protein adducts of the non-steroidal anti-inflamma-tory drug diclofenac. Chem Res Toxicol 1993:6;147–150.

7. Kretz-Rommel A, Boelsterli UA. Diclofenac covalent protein binding isdependent on acyl glucuronide formation and is inversely related to P450-mediated acute cell injury in cultured rat hepatocytes. Toxicol Appl Phar-macol 1993;120:155–161.

8. Kretz-Rommel A, Boelsterli UA. Selective protein adducts to membraneproteins in cultured rat hepatocytes exposed to diclofenac: radiochemicaland immunochemical analysis. Mol Pharmacol 1994;45:237–244.

9. Hargus SJ, Amouzedeh HR, Pumford NR, Myers TG, McCoy SC, PohlLR. Metabolic activation and immunochemical localisation of liver proteinadducts of the non-steroidal anti-inflammatory drug diclofenac. Chem ResToxicol 1994;7:575–582.

10. Wade LT, Kenna JG, Caldwell J. Immunochemical detection of mouse he-patic protein adducts derived from the non-steroidal anti-inflammatory drugsdiclofenac, sulindac and ibuprofen. Chem Res Toxicol 1997;10:546–555.

11. Kretz-Rommel A, Boelsterli UA. Cytotoxic activity of T cells and non-Tcells from diclofenac-immunised mice against cultured syngeneic hepato-cytes exposed to diclofenac. HEPATOLOGY 1995;22:213–222.

12. Bougie D, Johnson ST, Weitekamp LA, Aster RH. Sensitivity to a metab-olite of diclofenac as a cause of acute immune haemolytic anaemia. Blood1997;90:407–413.

13. Aithal GP, Day CP, Leathart J, Daly AK. Relationship of polymorphism inCYP2C9 to genetic susceptibility to diclofenac-induced hepatitis. Pharma-cogenetics 2000;10:511–518.

14. de Waal Malefyt R, Yssel H, Rocarolo M-G, Spits H, de Vries JE. Inter-leukin 10. Curr Opin Immunol 1992;4:314–322.

15. Louis H, Van Laetherm JL, Wu W, Quertinmont E, Degraef C, Berg KV,et al. Interleukin-10 controls neutrophilic infiltration, hepatocyte prolifer-


ation, and liver fibrosis induced by carbon tetrachloride in mice. HEPATOL-OGY 1998;28:1607–1615.

16. Nelson DR, Lauwers GY, Lau JYN, Davis GL. Interleukin 10 treatmentreduces fibrosis in patients with chronic hepatitis C: a pilot trial of inter-feron nonresponders. Gastroenterology 2000;118:655–660.

17. Eskdale J, Kube D, Tesch H, Gallagher G. Mapping of the IL-10 gene andfurther characterisation of the 5� flanking sequence. Immunogenetics1997;46:120–128.

18. Turner DM, Williams DM, Sankaran D, Lazarus M, Sinnott PJ, Hutchin-son IV. An investigation of polymorphism in the interleukin-10 gene pro-moter. Eur J Immunogenet 1997;24:1–8.

19. Grove J, Daly AK, Bassendine MF, Gilvarry E, Day CP. Interleukin 10promoter region polymorphisms and susceptibility to advanced alcoholicliver disease. Gut 2000;46:540–545.

20. Lazarus M, Hajeer AH, Turner D, Sinnott P, Worthington J, Ollier WE,et al. Genetic variation in the interleukin 10 gene promoter and systemiclupus erythematosus. J Rheumatol 1997;24:2314–2317.

21. Lim S, Crawley E, Woo P, Barnes PJ. Haplotype associated with lowinterleukin-10 production in patients with severe asthma. Lancet 1998;152:113.

22. Abbas AK, Murphy KM, Sher A. Functional diversity of helper T lympho-cytes. Nature 1996;383:787–793.

23. Rosenwasser LJ, Klemm DJ, Dreback JK, Inamma H, Masca JJ, KlinnertM, et al. Promoter polymorphism in the chromosome 5 gene cluster inasthma and atopy. Clin Exp Allergy 1995;25(Suppl 2):74–78.

24. Kawashima T, Noguchi E, Arinami T, Kobayahi KY, Nakagawa H, Ot-suka F, et al. Linkage and association of an interleukin 4 gene polymor-phism with atopic dermatitis in Japanese families. J Med Genet 1998;35:229–239.

25. Noguchi E, Shibasaki M, Arinami T, Takeda K, Yokouchi Y, KiwashimaT, et al. Association of asthma and interleukin-4 promoter gene in Japa-nese. Clin Exp Allergy 1998;28:449–453.

26. Cantagrel A, Navaux F, Loubet-Lescoulie P, Nourhashemi F, Enault G, AbbalM, et al. Interleukin-1b, interleukin-1 receptor antagonist, interleukin-4, andinterleukin-10 gene polymorphisms: relationship to occurrence and severity ofrheumatoid arthritis. Arthritis Rheum 1999;42:1093–1100.

27. Nelms K, Keegan AD, Zamorano J, Ryan JJ, Paul WE. The IL-4 receptor:signaling mechanisms and biologic functions. Ann Rev Immunol 1999;17:701–738.

28. Shirakawa I, Deichmann KA, Izuhara I, Mao I, Adra CN, Hopkin JM.Atopy and asthma: genetic variants of IL-4 and IL-13 signalling. ImmunolToday 2000;21:60–64.

29. Hershey GK, Fredrich MF, Esswein LA, Thomas ML, Chatila TA. Theassociation of atopy with a gain of function mutation in the alpha subunitof the interleukin-4 receptor. N Engl J Med 1997;337:1720–1725.

30. Benichou C. Criteria of drug-induced liver disorders. Report of an Inter-national Consensus Meeting. J Hepatol 1990;11:272–276.

31. Aithal PG, Day CP. The natural history of histologically proven drug-induced liver disease. Gut 1999;44:731–735.

32. Gardener I, Bergin P, Stening P, Kenna JG, Caldwell J. Immunochemicaldetection of covalently modified protein adducts in livers of rats treatedwith methyleuginol. Chem Res Toxicol 1996;9:713–721.

33. Daly AK, Steen VM, Fairbrother KS, Idle JR. CYP2D6 multiallelism.Methods Enzymol 1996;272:199–210.

34. Aithal GP, Day CP, Leathart JBS, Daly AK, Hudson M. Association ofsingle nucleotide polymorphism in the interleukin-4 and interleukin-4receptor gene with Crohn’s disease in a British population. Genes andImmunity 2001;2:44–47.

35. Pessayre D. Role of reactive metabolites in drug-induced hepatitis. J Hepa-tol 1995;23(Suppl 1):16–24.

36. Pessayre D, Larrey D, Biour M. Drug-induced liver injury. In: Bircher J,Benhamou JP, McIntyre N, Rizzetto M, Rodes J, eds. Oxford Textbook ofClinical Hepatology. Volume II. Oxford: Oxford University Press, 1999:1261–1315.

37. Package insert. Voltaren (diclofenac sodium). Ardsley, New York: GeigyPharmaceuticals of Ciba-Geigy Corporation, October 1988.

38. Bourdi M, Masubuchi Y, Reilly TP, Amouzadeh HR, Martin JL, GeorgeJW, et al. Protection against acetaminophen-induced liver injury and le-thality by interleukin 10: role of inducible nitric oxide synthase. HEPATOL-OGY 2002;35:289–298

39. Berson A, Freneaux E, Larrey D, Lepage V, Douay C, Mallet C, et al.Possible role of HLA in hepatotoxicity. An exploratory study in 71 patientswith drug-induced idiosyncratic hepatitis. J Hepatol 1994;20:336–42.


Hepatic adducts, circulating antibodies, and cytokine polymorphisms in patients with diclofenac hepatotoxicity

Documents