Chem. Res. Toxicol. 1993, 6, 609-615
609
Art i d e s Serum Antibodies from Halothane Hepatitis Patients React with the Rat Endoplasmic Reticulum Protein ERp72 Neil R. Pumford,fJ Brian M. Martin,$ David Thomassen,? Jennifer A. Burris,li J. Gerald Kenna,tJ Jackie L. Martin,?,#and Lance R. Pohl'pt Laboratory of Chemical Pharmacology, National Heart, Lung, and Blood Institute, Clinical Neurosciences Branch, National Institute of Mental Health, and Laboratory Sciences Section, National Center for Research Resources, National Institutes of Health, Bethesda, Maryland 20892, and Department of Anesthesiology and Critical Care Medicine, The Johns Hopkins Medical Institutions, Baltimore, Maryland 21287 Received March 18, 1993
Immunoblotting studies have previously shown that serum antibodies from halothane hepatitis patients react with several liver microsomal proteins that have been modified by the trifluoroacetyl halide metabolite of halothane. In this study, an 80-kDa protein recognized by the patients' antibodies has been purified from rat liver microsomes and characterized. When the purified trifluoroacetylated 80-kDa and native 80-kDa proteins were employed as test antigens in an enzyme-linked immunosorbent assay, serum antibodies from halothane hepatitis patients reacted with both of these proteins to a significantly greater extent than did serum antibodies from control patients. Amino acid sequence analyses of several hydrolytic peptide fragments of the 80-kDa protein showed that the protein was 99 % identical to the deduced amino acid sequence of a murine cDNA of the luminal endoplasmic reticulum protein ERp72. These results indicate that trifluoroacetylated ERp72 in the liver of halothane hepatitis patients may induce immune responses against epitopes present on the covalently altered protein and those present on the native protein and may have a role in halothane hepatitis. In addition, immunoblot and immunohistochemical studies revealed that the 80-kDa protein was present in all tissues studied, but was in highest concentration in liver, adipose tissue, ovaries, and testes and was enriched in specific cells of some organs. In the future, these findings should help define the physiological function of ERp72.
Introduction Halothane can cause both a mild and fulminant form (halothane hepatitis) of hepatotoxicity ( I , 2 ) . The mild form of hepatotoxicity occurs in approximately 20% of patients administered halothane and results in mild changes in liver function tests. Halothane hepatitis is rarer, is often fatal, and has been classified as an idiosyncratic drug reaction ( I , 3). Accumulated evidence has suggested that halothane hepatitis may have an immunological basis. For example, patients diagnosed with halothane hepatitis often have one or more of the clinical features of fever, rash, eosinophilia, hepatic
* To whom correspondence should be addressed at the Laboratory of Chemical Pharmacology, NHLBI, NIH, Building 10, Rm. 8N 115, Bethesda, MD 20892. Tel: 301-496-4841; Fax: 301-402-0171. + Laboratory of Chemical Pharmacology, National Heart, Lung, and Blood Institute, NIH. f Present address: Department of Pharmacology and Toxicology, University of Arkansas, Little Rock, AR 72205. 1 Clinical Neurosciences Branch, National Institute of Mental Health, NIH. ILaboratory Sciences Section,National Center for Research Resources, NIH. 1 Present address: Department of Phmmacology and Toxicology, St. Mary's Hospital Medical School, Norfolk Place, London W2 lPG, United Kingdom. # Department of Anesthesiology and Critical Care Medicine, The Johns Hopkins Medical Institutions.
lymphocytic infiltration, a high incidence of multiple exposures to halothane, or liver-kidney microsomal autoantibodies, suggestive of an immunological reaction ( 1, 2). They also have been shown to have peripheral blood lymphocytes and serum antibodies that react with altered liver proteins from rabbits treated with halothane ( I , 2 ) . In addition, immunoblotting studies of liver tissue from halothane-treated rats (41, rabbits (51, and humans (6) have revealed that the serum antibodies of the halothane hepatitis patients are directed against liver microsomal protein fractions of 100 kDa, 80 kDa (formerly designated 76 kDa), 59 kDa, 57 kDa, and 54 kDa that are covalently trifluoroacetylated (TFA)1 by the trifluoroacetyl halide metabolite of halothane (4). In order to understand better how halothane might cause hepatitis by an immune mechanism, we have recently begun to purify and characterize the TFA proteins from rat liver microsomes. A 59-kDa protein has been identified as a carboxylesterase (EC 3.1.1.1) (7), a 63-kDa protein corresponded to the calcium-binding protein calreticulin (8),and a 58-kDa protein, not yet identified, has been purified ( 9 ) . In this Abbreviations: TFA, trifluoroacetylated; ERp72, an endoplasmic reticulum protein of 72 kDa; SDS/PAGE, sodium dodecyl sulfate/ polyacrylamide gel electrophoresis; ELISA, enzyme-linked immunosorbent assay; LPS, lipopolysaccharide; GRP78/Bip, 78-kDa glucoseregulated protein; FCS, fetal calf serum; TBS, Tris-buffered saline.
This article not subject to U.S.Copyright. Published 1993 by the American Chemical Society
610 Chem. Res. Toxicol., Vol. 6, No. 5, 1993 study, we have purified a TFA-80-kDa and native unlabeled 80-kDa proteins from r a t liver microsomes and have found that several halothane hepatitis patients have serum antibodies that recognize the TFA-80-kDa protein, t h e unlabeled native protein, or both proteins. Amino acid sequence analysis of several internal tryptic and mild acid cleavage fragments of t h e protein has revealed that the protein corresponds to the luminal endoplasmic reticulum protein, ERp72 (10). Immunoblotting and immunohistochemistry studies have shown that this protein of unknown function is present in most tissues of the body and is localized in specific cells of certain organs.
Experimental Procedures Materials. Anti-TFA sera were prepared by immunizing rabbits with TFA-rabbit serum albumin as previously described (11). Halothane from Halocarbon Laboratories Inc. (Hackensack, NJ) was purified by distillation before use. Purification of the 80-kDa Protein. The TFA-80-kDa and native unlabeled 80-kDa proteins were purified from 3 g of liver microsomes from halothane-treated or untreated male SpragueDawley rats (Taconic Farms, Germantown, NY), respectively, with the use of methods described previously for the purifications of the TFA-63-kDa and native 63-kDa proteins (8). In short, 0.1 % (w/v) sodium deoxycholate extracts of liver microsomes were fractionated into partially purified protein fractions by DEAE-Sepharose (Pharmacia, Piscataway, NJ) anion-exchange chromatography. The TFA-80-kDa or 80-kDa proteins eluted from the column at 0.25-0.28M NaC1. Both proteins were further purified by HPLC anion-exchange chromatography on a BioGel TSK DEAE-5-PW column (Bio-Rad, Richmond, CA); they eluted from this column at 0.25-0.35 M NaC1. The overall yields for the purifications were 1 mg (0.03% of microsomal protein) for the TFA-80-kDa protein and 1.3 mg (0.04% of microsomal protein) for the native 80-kDa protein. Reaction of Patients' Serum Antibodies with the TFA80-kDa and Native 80-kDa Proteins. The reactions of serum antibodies from control and halothane hepatitis patients with the purified 80-kDa proteins were measured by a previously described enzyme-linkedimmunosorbent assay (ELISA)method (12),except that in this assay both the TFA-labeled and native proteins were used as the solid-phase antigen, instead of only TFA-labeled proteins. Sera were obtained from patients with a clinical diagnosis of halothane hepatitis ( n = 40) by J. G. Kenna as described in detail elsewhere (12,13). Sera of control patients (n = 32) were collected by J. G. Kenna from King's College Hospital, London, U.K., and by J. L. Martin from The Johns Hopkins Medical Institutions, Baltimore, MD, and included (1) those with multiple halothane exposures without developing evidence of liver dysfunction (n = 5),(2) normal control subjects (n = 5), (3) subjects exposed to subclinical doses of halothane [anesthesiologists, operating room and recovery room nurses (n = 5)], (4) patients with primary biliary cirrhosis ( n = 5), (5) patients with acute fulminant liver disease ( n = 5), (6) patients with chronic active liver disease ( n = 5), and (7) patients with viral hepatitis, who had been exposed to halothane (n = 2). Two statistical hypotheses were tested. The first was whether control and halothane hepatitis patients differed in response to the TFA-80-kDa 0-native 80-kDa antigens. The Wilcoxon rank sum test was app 1 and two-sided p values less than or equal to 0.05 were deen to be statistically significant. The second hypothesis, wheth ;the response to TFA-80-kDa and native 80kDa proteins differed within control and halothane hepatitis patient groups, was done using the binomialtest on the proportion of patients with a TFA response greater than or equal to a response with native protein. Any two-sided p value less than or equal to 0.05 was considered statistically significant. Preparation of Anti-80-kDa Sera. Antisera were raised against the TFA-80-kDa protein by mixing 150 pg of the TFA80-kDa protein in anion-exchange HPLC eluent with an equal
Pumford et al. volume of Freund's complete adjuvant and injecting two 2.5-kg female New Zealand White rabbits (Dutchland, Denver, PA) with the mixture subcutaneously at 14 sites along the back and intramuscularly at 2 sites in the hindquarters. The rabbits were boosted in a similar manner 57 days later with 100 pg of TFA80-kDa protein in Freund's complete adjuvant. After 14 days, sera were collected weekly for 5 weeks. Amino Acid Sequence Analyses. Purified TFA-80 kDa protein (100 pg) in 100 mM potassium phosphate buffer (pH 8.0), containing 2 M urea and 20 mM methylamine, was digested with 10 pg of trypsin/mL overnight at 37 "C. The TFA-80-kDa protein (100 pg) was also digested with 50% acetic acid at 55 "C for 72 h. Peptides were isolated and sequenced by Edman degradation, following standard procedures (8). Sequence homology with known proteins was determined by searchingEMBL, Genbank, Swiss Protein, and NBRF/PIR data bases using Bionet's search program for sequence homology (14). Tissue Distribution of t h e 80-kDa Protein. Entire organs were removed from 8 male Sprague-Dawley rats (200-250 g), except for skeletal muscle, which was taken from the left hind leg; intestine, which was removed from the first 25 cm of the small intestine; fat, which was collected from the scrotum; and ovaries, which were taken from 3 female Sprague-Dawley rats (200-250g). All tissues were homogenizedwitha polytron, except for the liver, which was homogenized with 5 strokes in a Dounce homogenizer. The homogenizing media consistedof four volumes (w/v) of ice-cold 100 mM Tris-acetate (pH 7.4), containing 250 mM sucrose, 1 mM EDTA, 0.2 mM phenylmethanesulfonyl fluoride, 1 mM leupeptin, and 1 mM pepstatin. The concentration of immunoreactive 80-kDa protein in tissue homogenate5 was determined by comparing the intensity of staining of immunoreactive proteins on immunoblots, scanned with a 2202 Ultroscan Laser densitometer (LKB Instruments, Bromma, Sweden), to that of a standard curve of purified 80-kDa protein. Immunohistochemical Detection of 80-kDa Protein in Tissues. Samples of tissue (approximately 2 mm by 2 cm) were collected from stomach, small intestine, spleen, kidney, brain, liver, lung, heart, testes, and thymus of male Sprague-Dawley rats (200-250 9). The tissues were placed into plastic cassettes, microwave-fixedin saline at 60 "C for 2 min, and stored overnight in70% ethanol(l5). Thefixedtissueswereembeddedinparaffii, and 4-6-pm sections were placed on poly(L-lysine)-treatedglass slides by American Histolabs (Gaithersburg, MD). The slides were deparaffinized and rehydrated using standard techniques and counterstained with Mayer's hematoxylin solution. All the following steps were done at room temperature. Nonspecific binding to the tissues was blocked by incubation for 30 min with 5% (v/v) fetal calf serum (FCS) in 50 mM Tris-HC1 (pH 7.6), containing 150 mM NaCl (Tris-buffered saline, TBS). Anti80-kDa serum, diluted 1:loOO or 1:3oOO in 2% (v/v) FCS-TBS, was added to the tissue and incubated for 2 h. After washing with TBS, alkaline phosphatase-conjugated goat anti-rabbit IgG (Life Technologies, Inc., Bethesda, MD) diluted 1500 in 2% (v/v) FCS-TBS was added, and the tissues were incubated for 1 h. After washing with TBS, AS/AP alkaline phosphatase substrate (BIO/CAN Scientific,distributed by Accurate Chemical & Scientific Corp., Westbury, NY) was added, and staining was developed for 20 min. The reactions were stopped by washing the tissues with TBS. The tissues were mounted under glass cover slips in an aqueous mounting medium (Glycergel, Dako Corp., Carpinteria, CA). No staining was found when preimmune rabbit serum was used under similar conditions. Tissues were examined under the light microscope for the distribution and relative intensity of staining. Other Methods. Sodium dodecyl sulfate/polyacrylamide gel electrophoresis (SDS/PAGE) and immunoblotting were performed as reported elsewhere, except that standard-length gels were used instead of minigels (4). Protein was determined according to the method of Lowry et al. (16)with bovine serum albumin as a standard.
Chem. Res. Toxicol., Vol. 6, No. 5, 1993 611
Anti-ERp72 Antibodies in Halothane Hepatitis
C
A kDa 200
-
117
-
97% 66.
43
2
3
kDa
200
200
117
117
97
97
66
66
43
43
---
~-
1
kDa
4
5
6
7
8
2
3
4
5
6
1
2
3
4
5
6
7 8
Figure 1. Purification and immunoreactivity of TFA-80-kDa and native 80-kDa proteins. The figure represents the progressive purification of the TFA-80-kDa and 80-kDa proteins, as followed by SDS/PAGE and immunoblotting with anti-TFA and anti-80-kDa sera. The individual lanes were liver microsomal fractions from untreated (1)and halothane-treated rats (2); residue (3) and extract (4) from 0.1 7% sodium deoxycholate extraction of liver microsomes from halothane-treated rats; partially purified TFA-80-kDa protein by anion-exchange DEAE-Sepharose chromatography (5); purified TFA-80-kDa (6) and native 80-kDa proteins (7) after anionexchange Bio-Gel TSK DEAE-5-PW chromatography; and human liver microsomes (8). (A) SDS/PAGE gel stained for protein with Coomassie Blue: lanes 1-5 and 8,25 pg; lanes 6 and 7,2 pg. (B) Immunoblot blot developed with anti-TFA sera: lanes 2-4,30 pg; lane 5,5 pg; lane 6,2 pg. (C) Immunoblot developed with anti-80-kDa sera: lanes 1-3,6 pg; lane 4,4 pg; lane 5,0.6 pg; lanes 6 and 7, 0.2 pg; lane 8, 24 pg. TFA-80 kDa
Results Purification of the 80-kDa Protein. As previously reported (4,7-9,17), when halothane was administered to rats, immunoblottingof liver microsomes with anti-TFA sera showed that several proteins were trifluoroacetylated by the reactive trifluoroacetyl halide metabolite of halothane (Figure lB, lane 2). A TFA-80-kDa protein was purified from this mixture by a method recently described for the purification of a 63-kDa protein target of the trifluoroacetyl halide metabolite (8). The first step was the extraction of liver microsomes with 0.1% sodium deoxycholate. This procedure extracted a 80-kDafraction (Figure lA, lane 4) that contained the TFA moiety (Figure lB, lane 4). The next step involved partial purification of the TFA-80-kDa protein from the extract by DEAESepharose anion-exchange chromatography (Figure 1A, lane 5, and Figure lB, lane 5). The TFA-80-kDa was further purified by the use of anion-exchange HPLC (Figure lA, lane 6, and Figure lB, lane 6). The same procedure was used to purify the native unlabeled 80-kDa protein from liver microsomes of untreated rats (Figure lA, lane 7). Reaction of Halothane Hepatitis Patients' Antibodies with the Purified TFA-80-kDaand Native 80kDa Proteins. It was found in an ELISA that serum antibodies of the group of halothane hepatitis patients reacted to a significantly greater extent with either the purified TFA-80-kDa protein (p < 0.001) or native 80kDa protein (p < 0.001) than did serum antibodies from the group of control patients (Figure 2). As a group, there was no significant difference in the response of the halothane hepatitis patients (p = 0.64) with the two antigens. However, there were 3 of 40 of the halothane hepatitis patients, but none of the control patients, who had serum antibodies that reacted appreciably greater with
1.2 0
36
1.o
1.0 -
0.81
0.8
0.6-
0.6 -
0.4 -
0
0 0
0.2 -
0.0
'
0
Halothane Hepatitis
Control
Figure 2. ELISA determination of antibodies in sera of halothane hepatitis patients (n = 40) and control subjects ( n = 32) that react with TFA-80-kDa and native 80-kDa proteins. Purified TFA-80-kDa or native 80-kDa proteins were applied to the wells (1pg/well) of microtiter plates.
TFA-80-kDa protein than with the native 80-kDa protein (absorbances of 1.019 vs 0.331,0.755 vs 0.229, and 0.613 vs 0.137). No obvious differences were observed in the reactivity of the sera of the different groups of control patients with either of the test antigens. Characterizationof the 80-kDa Protein. Immunoblotting studies with sera raised against the TFA-80-kDa protein showed that most of the TFA-80-kDa protein was extracted from liver microsomes with 0.1 % sodium deoxycholate (Figure lC, lanes 3 and 4), a solubility property consistent with the 80-kDa protein being either a peripheral or luminal protein of the endoplasmicreticulum (18). The antibody reacted comparablywith the purified TFA80-kDa and native 80-kDa proteins, confirming that the two proteins were highly similar (Figure lC, lanes 6 and
612 Chem. Res. Toxicol., Vol. 6, No. 5, 1993
Pumford et al.
tacular cells of Sertoli did not appear to stain with anti80-kDa sera. In the lung, immunoreactivity was stronger VKEENGVWVLNDGNFDNFVADKDTVLLEFYAPWCGHCKQFAPEYEKIAST 1 0 0 in the bronchial epithelium than in the bronchiolar epithelium or peribronchial lymphoid nodular cells, while LKDNDPPIAVAKIDATSASMLASKFDVSGYPTIKILKKGQAVDYDGSRTQ 1 5 0 1 2 --3 very little immunoreactivity was seen in the alveolar cells EEIVAKVREVSQPDWTPPPEVTLSLTKDNFDDVVNNADIILVEFYAPWCG 2 0 0 (Figure 5C). Cells displaying intense immunoreactivity, in other organs that were studied, included the transitional HCKKLAPEYEKAAKELSKRSPPIPLAKVDATEQTDLAVFDVSGYPTLKI 2 5 0 epithelium of the renal pelvis, a small percentage of the 4 5 FRKGRPFDYNGPREKYGIVDYMIEQSGPPSKEILTLKQVQEFLKDGDDW 300 lymphocytes in splenic corpuscles and mucosa/submucosa 6’ - 1 of the small intestine, large cerebral neurons, and large IIGLFQGDGDPAYLQYQDAANNLREDYKFHHTFSPEIAKFLKVSLGKLVL 350 neurons in the peripheral ganglia (results not shown). MKLRKAWLLVLLLALTQLLAAASAGDAQEDTSDTENATEEEEEEDDDDLE
50
THPEKFQSKYEPRFHVMDVQGSTEASAIKDYWKHALPLVGHRKTSNDAK 4 0 0
Discussion RYSKRPLVVVYYSVDFSFDYRAATQFWRNKVLEVAKDFPEYTFAIADEED 450 8 YATEVKDLGLSESGEDVNAAILDESGKKFAMEPEEFDSDTLREFVTAFKK 5 0 0
A unique characteristic of most patients diagnosed with halothane hepatitis has been the presence of serum antibodies that react with TFA-labeled liver microsomal GKLKPVIKSQPVPKNNKGPVKVWGKTFDAIVMDPKKDVLIEFYAPWCGH 550 protein antigens. Immunoblotting of liver microsomes from halothane-treated animals or humans has revealed CKQLEPIYTSLGKKYKGQKDLVIAKMDATANDITNDQYKVEGFPTIYFAP 600 10 9 that these proteins have apparent monomeric molecular SGDKKNPIKFEGGNRDLEHLSKFIDEHATKRSRTKEEL 638 masses of 100 kDa, 80 kDa, 59 kDa, 57 kDa, and 54 kDa -- 1 1 (4-6) and has indicated that the patients’ serum antibodies Figure 3. Sequencehomology of polypeptidesderived from the are directed against epitopes that consist of the TFA TFA-80-kDaprotein with the deduced amino acid sequence of hapten and undefined specific structural features of the ERp72 murine cDNA. The amino acid sequences of peptides individual carrier proteins (4,7).Not all patients’ serum derived from trypsin hydrolysis (peptides 2-11) and mild acid hydrolysis (peptide 1) of the TFA-80-kDaprotein are indicated antibodies react with the same TFA proteins, and the vast by the underlines. The dashes (- - -) represent amino acids of majority of patients receiving halothane do not form these the sequenced peptides whose assignments could not be made antibodies or develop hepatitis (4,12). The reasons why unambiguously. A vertical bar separates two tryptic peptides patients are sensitized to a particular TFA-labeled enthat were sequenced. doplasmic reticulum protein are unknown. Several of the microsomal antigens have been purified and identified. A 7). An immunochemically related protein was also deTFA-59-kDarat liver microsomalprotein has been purified tected in human liver microsomes (Figure lC, lane 8). and identified as a carboxylesterase (EC 3.1.1.1) (7) and Edman sequence analysis of 11internal peptides of the found to cross-react in an ELISA with antibodies in the 80-kDaprotein produced by trypsin or mild acid hydrolysis sera of 2 of 10 patients with halothane hepatitis (21). revealed that the protein was 99% percent identical to Recently, a TFA-63-kDa protein has been purified from the luminal endoplasmic reticulum protein ERp72, enrat liver microsomes and identified as calreticulin (8). coded by a mouse cDNA (10) (Figure 3). Serum antibodies of only 1of 40 patients with halothane Since the function of ERp72 is unknown (10,19,20) its hepatitis cross-reacted in an ELISA with this protein (8). tissue and cellular distribution was investigated by immunoblotting and immunohistochemistry, respectively, In the present study, we have purified TFA-80-kDa and native 80-kDa microsomal proteins from rat liver miwith the use of anti-80-kDa sera. Such information ultimately should help define its physiologic role. Imcrosomes and have found that as a group the halothane hepatitis patients’ serum antibodies reacted to a statismunoblotting of organ homogenates revealed that the tically greater extent with both of these of proteins than highest concentration of immunoreactive 80-kDa protein did serum antibodies from control patients (Figure 2). Of was in the fat, liver, ovaries, and testes (Figure 4). In the intestines and stomach, the major immunoreactive protein the individual halothane hepatitis patients studied, 8 of fraction detected had an apparent molecular mass of 73 40 and 6 of 40 responded noticeably stronger with the kDa (Figure 4B, lanes 5 and 12). Several other tissues purified TFA-80-kDa and native 80-kDa proteins, respecappeared to contain small amounts of this fraction. Even tively, than did any of the control patients. These findings though all of the tissue homogenates were prepared in the are in contrast to the results of the immunoblotting studies presence of a mixture of several protease inhibitors, it is with liver microsomes, which showed very little, if any, reaction of halothane patients’ serum antibodies with possible that the 73-kDa protein fraction is a proteolytic degradation fragment derived from the 80-kDa protein. native proteins (4-6). The reason for this discrepancy is In addition, the testes contained aprotein of approximately not known at this time. It might be due to the patients’ sera containing antibodies that reacted with conforma146 kDa that was immunochemically related to the 80tional epitopes (22)of the native unmodified 80-kDa kDa protein (Figure 4B, lane 13). Immunohistochemical analysis of tissues showed that protein, which were destroyed by the denaturing conditions the periveinous parenchymal cells of the liver had a higher of immunoblotting, but were not altered appreciably by concentration of the 80-kDaprotein than did the periportal the relatively mild conditions of the ELISA (23-25). In and midzonal hepatocytes (Figure 5A). The bile duct contrast, antibodies to the TFA-induced epitopes present epithelial cells also appeared to contain the 80-kDaprotein. only on the TFA-80-kDa protein might have recognized Within the testes, the highest concentration of the 80sequential epitopes that were not as susceptible to kDa protein was found in the spermatogonia and primary denaturation and so were detectable by immunoblotting. Although these possibilities cannot be proven until epitope spermatocytes, while the spermatozoa, maturing spermatids, and interstitial cells of Leydig contained lower mapping of both the TFA-80-kDa and native 80-kDa concentrations of the protein (Figure 5B). The sustenproteins has been done, our findings are consistent with
-
Chem. Res. Toxicol., Vol. 6, No. 5, 1993 613
Anti-ERp72 Antibodies in Halothane Hepatitis
A
5 ,
1 2 3 4
5
B
kDa 200
-
117
-
6 7 8 9 10 11 12 13 14
Adrenals Brain Fat Heart Intestines Kidney Liver Lung Muscle Ovaries Spleen Stomach Testes Thymus
-
97 80 + Q
66
1 2 3 4 5 6 7 8 9 1011121314
Figure 4. Distribution of immunoreactive 80-kDa protein (ERp72) in tissues from untreated Sprague-Dawleyrats as determined by immunoblotting. (A) The concentration of immunoreactive 80-kDa protein in tissues was determined by comparing the intensity of staining:of the immunoblots of tissue homogenates scanned bv laser densitometer to that of a standard curve of purified 80-kDa protein. (B) Imkunoblots of the tissue homogenaces (10 pg each). "
immunofluorescence studies, which have shown that 19of 76 (25% ) (26) and 2 of 6 (33% ) (27) of halothane hepatitis patients had serum antibodies that reacted with native unmodified proteins present in frozen liver sections. We have shown that peptides derived from the 80-kDa protein were 99% identical to several distinct regions of the deduced amino acid sequence of mouse ERp72 (Figure 3). The only difference found in the sequences of these two proteins was a conservative change of one amino acid from Ile to Val, which may be due to a species difference. The subcellular localization of ERp72 and the 80-kDa protein also appear to be similar. ERp72 resides within the lumen of the endoplasmic reticulum (28) and studies of the topography of TFA-modified proteins in liver microsomes from halothane-treated rats indicate that the TFA-modified form of the 80-kDa protein exhibits this same subcellular localization (29). On the basis of these findings, it is concluded that the 80-kDa protein is ERp72. Moreover, since the deduced amino acid sequences of murine and human cDNAs of ERp72 are 89% identical (30)and the 80-kDa protein appears to be expressed in human liver (Figure l),it is possible that human TFAERp72 is the immunogen responsible for induction of the antibody response seen in patients with halothane hepatitis. If so, then it is feasible that antibodies reacting with native unmodified ERp72 might have been formed because of a loss of T-cell tolerance due to the modification of the 80-kDa protein by the TFA hapten (31-35). If TFA-ERp72 indeed acts as an immunogen in patients developing halothane hepatitis, it must be explained how this endoplasmic reticulum protein comes in contact with the immune system. ERp72 is retained in the endoplasmic reticulum to a major extent by its COOH-terminalKEEL retention signal (IO, 19). Therefore, under normal con-
ditions, it would not be expected to be secreted from hepatocytes or be incorporated into the plasma membrane. However, experiments performed with a recombinant expression vector have shown that when ERp72 is overexpressed at levels in cells, it does get secreted, possibly because of saturation of the retention processes (20). Perhaps, patients that develop halothane hepatitis overexpress ERp72, which then reacts with the trifluoroacetyl halide metabolite of halothane to produce high levels of TFA-ERp72. In this regard, it is known that the expression of ERp72 can be induced severalfold,when cells are treated with calcium ionophores (20), tunicamycin (20), or lipopolysaccharide (LPS)(36). Alternatively, patients that develop halothane hepatitis may express aberrant forms of ERp72 that do not contain the KEEL endoplasmic reticulum retention signal. Similarly, the KEEL retention signal might become ineffective, if the lysine residue was covalently modified by the TFA moiety. Another way the immune system might come in contact with TFA-ERp72 or other TFA-microsomal proteins is by leakage from damaged hepatocytes, since approximately 20 % of patients administered halothane develop mild liver damage (1,2). This latter possibility could explain, at least in part, how immune reactions against TFA-ERp72and native ERp72 might cause exclusively hepatotoxicity, even though ERp72 is present in most tissues. The physiological function of ERp72 is presently not known. Although its sequence has three copies of an 11 amino acid repeat containing the proposed CGHC active sites of protein disulfide isomerase (EC 5.3.4.1), it has not yet been shown to have a role in formation of disulfide bonds in proteins (IO, 19,20). Another function proposed for ERp72, but not yet proven, is that of a endoplasmic stress protein, like GRP78/Bip (37). It is thought that
614
Pumford et al.
Chem. Res. Toxicol., Vol. 6, No. 5, 1993
GRP78/Bip has a role in preventing the aggregation of incompletely assembled or aberrantly folded proteins in the endoplasmic reticulum and/or preventing their exit from the endoplasmic reticulum (37). Whatever its function, ERp72 appears to be essential because it is expressed in all organs that have been studied (Figures 4 and 5). Why it is concentrated in the periveinous hepatocytes and biliary epithelial cells of the liver, the bronchial epithelium of the lung, spermatogonia and primary spermatocytesof the testes, selected lymphocytes of the spleen, and specific central and peripheral neurons is not known, but should help define in the future the physiological role of this protein. For example, the specifically stained lymphocytes in the spleen might represent activated lymphocytes, since the activation of splenic lymphocytes by LPS has been shown to increase the level of ERp72 (36). Once a function has been assigned to ERp72, it will be possible to determine whether this activity is altered when ERp72 is covalently modified by the TFA moiety. If so, this may have a role to play in the development of halothane hepatitis. Furthermore, in the future, it may be found that ERp72 is the target of toxic reactive metabolites of other drugs or environmental chemicals. In this regard, it has recently been found that HCFC-123 (CF&HClB), which has been developed as a replacement for ozone-depleting chlorofluorocarbons, is metabolized in rats to form nearly identical patterns and levels of TFA liver proteins as that produced by halothane (38) Acknowledgment. We thank Dr.James R. Gillette for valuable suggestions in editing the manuscript and John W. George for doing the data base searches for us. D.T. was supported by research funds from Anaquest.
References
Figure 5. Immunohistochemicaldetection of the 80-kDaprotein (ERp72) in sections of liver, testes, and lung from untreated Sprague-Dawley rats. (A) Liver: multiple lobules with central veins (c), portal veins (p), and bile ducts (b); 35X (reduction of original photograph at 50X). The periveinous hepatocytes were the most intensely stained parenchymalcells. Bile duct epithelial cells were also stained relativelydark. (B) Testes: multiple cross sections of seminiferous tubules, 140X (reduction of original photograph at 20029. The spermatogonia (a) and primary spermatocytes (b) were stained darker than the secondary spermatocytesand young spermatids (brackets). The interstitial cells of Leydig (c) were less intensely stained, while the sustentacular cells of Sertoli (d) did not appear to be stained. (C) Lung: secondary bronchus (B),pulmonary artery (pa),lymphoid nodule (L),bronciole (b),and alveoli (a);3% (reductionof original photograph at 50x1. The cytoplasm of the columnar epithelial cells lining the bronchus was stained intensely. Cells stained less intensely were the bronchiolar epithelium and lymphocytes of the peribronchial lymph node.
(1)Neuberger,J., and Kenna, J. G. (1987)Halothane hepatitis: A model of immune mediated drug hepatotoxicity. Clin. Sci. 72,263-270. (2) Satoh, H.,Davies, H. W., Takemura, T., Gillette, J. R., Maeda, K., and Pohl, L.R. (1987)in h g m g in h g Metiitbokm {Biidgea,J. W., Chasseaud, L. F., and Gibson, G. G., Eds.) pp 187-206,Taylor & Francis, Philadelphia. (3) Pohl, L. R. (1990)Drug-induced allergic hepatitis. Semin. Liver Dis. 10,305-315. (4)Kenna,J. G., Satoh,H., Chriit, D. D., and Pohl, L. R. (1988)Metabolic basis for a drug hypersensitivity: antibodies in sera from patients with halothane hepatitis recognize liver neoantigens that contain the trifluoroacetyl group derived from halothane. J. Pharmacol. Exp. Ther. 245,1103-1109. (5) Kenna, J. G., Neuberger,J., and Williams, R. (1987)Identification by immunoblotting of three halothane-induced liver microsomal polypeptide antigens recognized by antibodies in sera from patients with halothane-associatedhepatitis. J.Pharmacol.Exp. Ther.242, 733-740. (6) Kenna, J. G., Neuberger, J. M., and Williams, R. (1988)Evidence for expression in human liver of halothane induced neoantigens recognized by antibodies in sera from patients with halothane hepatitis. Hepatology 8, 1635-1641. (7) Satoh, H., Martin, B. M., Schulick, A. H., Christ, D. D., Kenna, J. G., and Pohl, L. R. (1989)Human anti-endoplasmic reticulum antibodiesin sera of halothane hepatitis patients are directed against a trifluoroacetylatedcarboxylesterase. Proc. Natl. Acad. Sci. U.S.A. 86,322-326. (8) Butler, L.E.,Thomassen, D., Martin, J. L., Martin, B. M., Kenna, J. G., and Pohl, L. R. (1992)The calcium-binding protein calreticulin is covalently modified in rat liver by a reactive metabolite of the inhalation anesthetic halothane. Chem. Res. Toxicol. 6,406-410. (9) Martin, J. L.,Pumford, N. R.,LaRosa, A. C.,Martin, B. M., Gonzaga, H. M. S., Beaven, M. A., and Pohl, L. R. (1991)A metabolite of halothane covalently binds to an endoplasmic reticulum protein that is highly homologous to phosphatidylinositol-specific phospholipase C-alpha but has no activity. Biochem. Biophys. Res. Commun. 178,679-685.
’
Anti-ERp72 Antibodies in Halothane Hepatitis (10) Mazzarella, R. A., Srinivasan, M., Haugejorden, S. M., and Green, M. (1990) ERp72, an abundant luminal endoplasmic reticulum protein, contains three copies of the active site sequences of protein disulfide isomerase. J. Biol. Chem. 266,1094-1101. (11) Satoh, H., Fukuda, Y.,Anderson, D. K., Ferrans, V. J., Gillette, J. R., and Pohl, L. R. (1985) Immunological studies on the mechanism of halothane-induced hepatotoxicity: Immunohistochemical evidence of trifluoroacetylated hepatocytes. J.Pharmacol. Exp. Ther. 233,857-862. (12) Martin, J. L., Kenna, J. G., and Pohl, L. R. (1990) Antibody assays for the detection of patienta sensitized to halothane. Anesth. Analg. 70,154-159. (13) Kenna, J. G., Neuberger, J., and Williams, R. (1987) Specific antibodies to halothane-induced liver antigens in halothane-associated hepatitis. Br. J.Anuesth. 69, 1286-1290. (14) Pearson,W.R.,andLipman,D. J. (1988)Improvedtoolsforbiological sequence comparison. Roc. Natl. Acad. Sci. U.S.A. 85,2444-2448. (151 Roberta,D. W., Bucci,T. J.,Benson,R. W., Warbritton, A. R., McRae, T.A., Pumford, N. R., and Hinson, J. A. (1991) Immunohistochemical localizationand quantification of the 3-(cystein-S-yl)-acetaminophen protein adduct in acetaminophen hepatotoxicity. Am. J.Pathol. 138,359-371. (16) Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) Protein measurement with the folin phenol reagent. J.Biol. Chem. 193, 265-275. (17) Kenna, J. G., Martin, J. L., Satoh, H., and Pohl, L. R. (1990) Factors affecting the expression of trifluoroacetylated liver microsomal protein neoantigens in rata treated with halothane. Drug. Metab. Dispos. 18, 788-793. (18) Weihina, R. R., Manaaniello, V. C., Chiu, R., and Phillips, A. H. (1972) Purification of 6epatic microsomalmembranes. Biockemistry 11,3128-3135. (19) Haugejorden, S. M., Srinivasan, M., and Green, M. (1991) Analysis of the retention signalsof two resident luminal endoplasmicreticulum proteins by in vitro mutagenesis. J. Biol. Chem. 266, 6015-6018. (20) Dorner, A. J., Wasley, L. C., Raney, P., Haugejorden, S., Green, M., and Kaufman, R. J. (1990) The stress response in Chinese hamster ovary cells. Regulation of ERp72 and protein disulfide isomerase expression and secretion. J. Biol. Chem. 266, 22029-22034. (21) Pohl, L. R., Thomassen, D., Pumford, N. R., Butler, L. E., Satoh, H., Ferrans, V. J., Perrone, A., Martin, B. M., and Martin, J. L. (1991) Haptencarrier conjugatesassociatedwith halothane hepatitis. Adu. Exp. Biol. Med. 283, 111-120. (22) Laver, W. G., Air, G. M., Webster, R. G., and Smith-Gill, S. J. (1990) Epitopes on protein antigens: misconceptions and realities. Cell 61,553-556. (23) Waxman,D. J.,Lapenson,D.P.,Krishnan, M.,Bemard, O.,Kreibich, G., and Alvarez, F. (1988) Antibodies to livedkidney microsome in chronic active hepatitis recognize specific forms of hepatic cytochrome P-450. Gastroenterology 96, 1326-1331.
Chem. Res. Toxicol., Vol. 6, No. 5, 1993 615 (24) Finke, R., Seto, P., and Rapoport, B. (1990) Evidence for the highly conformationalnature of the epitope(s) on human thyroid peroxidase that are recognized by sera from patienta with Hashimoto's thyroiditis. J. Clin. Endocrinol. Metab. 71, 53-59. (25) Leedman, P. J., Harrison, P. J., and Harrison, L. C. (1991) Immunoblotting for the detection of TSH receptor autoantibodies. J. Autoimmun. 4, 529-542. (26) Walton, B., Simpson, B. R., Strunin, L., Doniach, D., Perrin, J., and Appleyard, A. J. (1976) Unexplained hepatitis followinghalothane. Br. Med. J. 1, 1171-1176. (27) Homberg, J. C., Abuaf, N., Helmy-Khalil, S., Biour, M., Poupon, R., Islam, S., Darnis, F., Levy, V. G., Opolon, P., Beaugrand, M., et al. (1985) Drug-induced hepatitis associated with anticytoplasmic organelle autoantibodies. Hepatology 6,722-727. (28) Lewis, M. J., Turco, S. J., and Green, M. (1985) Structure and assembly of the endoplasmic reticulum. Biosynthetic sorting of endoplasmic reticulum proteins. J. Biol. Chem. 260, 6926-6931. (29) Kenna, J. G., Martin, J. L., and Pohl, L. R. (1992) The topography of trifluoroacetylated protein antigens in liver microsomalfractions from halothane treated rata. Biochem. Pharmacol. 44,621-629. (30) Huang, S. H., Tomich, J. M., Wu, H., Jong, A., and Holcenberg, J. (1991) Human deoxycytidine kinase. Sequence of cDNA clones and analysis of expression in cell lines with and without enzyme activity. J. Biol. Chem. 266, 5353. (31) Allison, A. C., Denman, A. M., and Barnes, R. D. (1971) Cooperating and controlling functions of thymus-derived lymphocytes in relation to autoimmunity. Lancet 2, 135-140. (32) Weigle, W. 0. (1965) The induction of autoimmunity in rabbits following injection of heterologous or altered homologous thyroglobulin. J. Exp. Med. 121, 289-309. (33) Cox, K. O., and Keast, D. (1973) Erythrocyte autoantibodies induced in mice immunized with rat erythrocytes. Immunology 26, 531539. (34) Haba, S., and Nisonoff,A. (1987) Induction of high titers of anti-IgE by immunization of inbred mice with syngeneic IgE. Roc. Natl. Acad. Sci. U.S.A. 84,5009-5013. (35) Adelstein, S., Pritchard-Briscoe, H., Anderson, T. A,, Crmbie, J., Gammon, G., Loblay, R. H., Basten, A., and Goodnow, C. C. (1991) Induction of self-tolerance in T cells but not B cella of transgenic mice expressing little self antigen. Science 261, 1223-1225. (36) Lewis, M. J., Mazzarella, R. A,, and Green, M. (1985) Structure and assemblyof the endoplasmicreticulum. The synthesis of three major endoplasmic reticulum proteins during lipopolysaccharide-induced differentiation of murine lymphocytes. J. Biol. Chem. 260, 30503057. (37) Lenny, N., and Green, M. (1991) Regulation of endoplasmicreticulum stress proteins in COS cells transfected with immunoglobulin mu heavy chain cDNA. J.Biol. Chem. 266, 20532-20537. (38) Harris, J. W., Pohl, L. R., Martin, J. L., and Anders, M. W. (1991) Tissue acylation by the chlorofluorocarbon substitute 2,2-dichlorol,l,l-trifluoroethane. Roc. Natl. Acad. Sci. U.S.A. 88,1407-1410.
