Immunochemical Detection and Identification of Protein Adducts of

164

Chem. Res. Toxicol. 1998, 11, 164-171

Articles Immunochemical Detection and Identification of Protein Adducts of Diclofenac in the Small Intestine of Rats: Possible Role in Allergic Reactions Joseph A. Ware,*,† Mary Louise M. Graf,†,§ Brian M. Martin,‡ Lisa R. Lustberg,† and Lance R. Pohl† Molecular and Cellular Toxicology Section, Laboratory of Molecular Immunology, National Heart, Lung, and Blood Institute, NIH, Bethesda, Maryland 20892, Department of Anesthesiology and Critical Care Medicine, The Johns Hopkins Medical Institutions, Baltimore, Maryland 21287, and Clinical Neurosciences Branch, National Institute of Mental Health, NIH, Bethesda, Maryland 20892 Received October 8, 1997

Idiosyncratic adverse drug reactions are unpredictable, target multiple organ systems, and often become life-threatening events. Although the causes of idiosyncratic adverse drug reactions are not known in most cases, evidence suggests that they may be mediated through immunological mechanisms. It is generally thought that for a drug to lead to an immune response, it must first become covalently bound to a carrier protein. Since most drugs are unreactive, it is usually a reactive metabolite that is expected to form covalent adducts. However, it is not clear why more people do not develop immune reactions against drugprotein adducts. One possible explanation is that orally administered drugs may lead to oral tolerance in most individuals through mechanisms similar to that found with orally administered antigens. However, very little is known regarding the interaction of drugs with gut-associated lymphoid tissue of the small intestine, where oral tolerance can develop. As an initial step to test this hypothesis, we have investigated whether diclofenac, a commonly used nonsteroidal antiinflammatory drug, can lead to protein adducts in rat small intestine. Diclofenac was administered to rats by gastric gavage. Immunoblot analysis of small intestine homogenates and isolated enterocyte subcellular fractions with drug-specific antiserum revealed 142-, 130-, 110-, and 55-kDa protein adducts of diclofenac. The 142- and 130-kDa adducts of diclofenac were identified as aminopeptidase N (CD13) and sucrase-isomaltase, respectively, by amino acid sequence analyses and by their reactions with protein-specific antibodies. The adducts were localized by immunohistochemistry and found primarily in the mid-villus and villus-tip enterocytes and also in the dome overlying Peyer’s patches. Similar adducts were detected immunochemically in villus-tip enterocytes of animals treated with halothane or acetaminophen. These results show that intestinal protein adducts of drugs can be formed in gut-associated lymphoid tissue where they may lead to the down-regulation of drug-induced allergic reactions in many individuals.

Introduction Idiosyncratic adverse drug reactions are unpredictable, target multiple organ systems, and often become lifethreatening events. Although the causes of idiosyncratic drug reactions are not known in most cases, evidence suggests that they may be mediated through immunological mechanisms (1). Generally, it is thought that in order for a drug to lead to an immune response, it must first become covalently bound to a carrier protein. Since * To whom correspondence should be addressed at the Molecular and Cellular Toxicology Section, NHLBI, NIH, Bldg 10, Rm 8N110, Bethesda, MD 20892-1760. Phone: (301) 402-7323. Fax: (301) 4804852. E-mail: [email protected]. † National Heart, Lung, and Blood Institute. § The Johns Hopkins Medical Institutions. ‡ National Institute of Mental Health.

S0893-228x(97)00182-3

most drugs are unreactive, it is usually a reactive metabolite that is expected to form covalent adducts. However, it is not clear why more people do not develop immune reactions against protein adducts of drugs. Oral administration of antigens can induce systemic hyporesponsiveness, termed oral tolerance (2). It is known that oral tolerance can develop when an antigen interacts with the gut-associated lymphoid tissues (GALT),1 which constitutes the largest lymphoid organ of the body (3). Although the exact mechanism(s) re1 Abbreviations: GALT, gut-associated lymphoid tissue; NSAID, nonsteroidal antiinflammatory drug; OG, n-octyl β-D-glucopyranoside; ECL, enhanced chemiluminescence; P450, cytochrome P450; UGT, UDP-glucuronosyltransferase; DTT, dithiothreitol; TFA, trifluoroacetyl; APAP, acetaminophen; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; PMSF, phenylmethanesulfonyl fluoride;

This article not subject to U.S. Copyright. Published 1998 by the American Chemical Society Published on Web 02/18/1998

Drug-Protein Adducts in Rat Small Intestine

sponsible for the formation of tolerance have not been clearly defined, clonal anergy and clonal deletion of antigen-specific T-lymphocytes and/or formation of lymphocytes that produce immunosuppressive cytokines are thought to have a role (4-7). The most commonly studied antigens have been soluble proteins and peptides. However, the oral administration of metals such as nickel (8) and reactive organic compound-haptens such as dinitrochlorobenzene (DNCB) or trinitrobenzenesulfonic acid has also been shown to produce tolerance against the sensitizing effects of these compounds (9-11). Oral tolerance produced by these compounds is presumably mediated by covalent adducts formed in GALT. If covalent adducts of drugs were formed in the GALT, it seemed reasonable that they may also lead to oral tolerance and prevent drug-induced allergic reactions. We have chosen to investigate this idea, using the nonsteroidal antiinflammatory drug (NSAID) diclofenac as a model compound for our studies. Diclofenac has been shown to cause a variety of idiosyncratic adverse reactions such as hepatotoxicity (12), hemolytic anemia (13), agranulocytosis (14), and anaphylaxis (15), all of which may in part be due to immune reactions to protein adducts of diclofenac. Several protein adducts of diclofenac have been detected immunochemically in the livers of mice and rats treated with diclofenac (16-18) and in primary cultures of rat and human hepatocytes that had been incubated with diclofenac (19, 20). Protein adducts can be formed by both cytochrome P450 (P450)dependent and UDP-glucuronosyltransferase (UGT)-dependent pathways of metabolism (17, 18). In this study, we found that when rats were treated orally with diclofenac, the drug formed covalent adducts to aminopeptidase N (CD13) and sucrase-isomaltase (SI) in enterocytes of the small intestine of rats. In addition, protein adducts of acetaminophen and halothane were also detected immunohistochemically in enterocytes of animals treated with these drugs. These results show that covalent adducts of drugs can be formed within GALT, where they may have a role in preventing drug allergies.

Experimental Procedures Caution: Because o-dianisidine, p-nitroaniline, and Protogel are considered hazardous materials, standard operating procedures were followed for the handling of toxic substances. Materials. Chemicals were obtained from the following sources: acetaminophen, sodium salt of diclofenac, Gill No. 1 hematoxylin solution, glucose oxidase kit, Ala-p-nitroanilide, p-nitroaniline, and sucrose SigmaUltra from Sigma (St. Louis, MO); halothane from Halocarbon Labs (Hackensack, NJ), distilled before use; dithiothreitol (DTT) and pepstatin from ICN Biomedicals (Aurora, OH); carboxypeptidase Y, endoglycosidaseH, N-glycosidase-F, leupeptin, phenylmethanesulfonyl fluoride (PMSF), and goat anti-rabbit IgG (peroxidase-conjugated) from Boehringer Mannheim (Indianapolis, IN); n-octyl β-D-glucopyranoside (OG) from Calbiochem-Novabiochem (LaJolla, CA); heparin sodium and Permount from Fisher (Fairlawn, NJ); BCA protein assay reagent kit from Pierce (Rockford, IL); calciumand magnesium-free phosphate-buffered saline (PBS) from Biofluids (Rockville, MD); 112-µm nylon mesh from Tetko (Briarcliff Manor, NY); 0.45-µm poly(vinylidene difluoride) PBS, phosphate-buffered saline; PVDF, poly(vinylidene difluoride); SI, sucrase-isomaltase; AmN, aminopeptidase N (CD13); FAE, follicleassociated epithelium; IEL, intraepithelial lymphocytes; LP, lamina propria; PP, Peyer’s patches; DNCB, dinitrochlorobenzene.

Chem. Res. Toxicol., Vol. 11, No. 3, 1998 165 (PVDF) Immobilon-P membrane from Millipore (Bedford, MA); enhanced chemiluminescence (ECL) reagents from Amersham (Arlington Heights, IL); broad-range molecular mass standards, Tris/Gly/SDS 10× running buffer, Tris/Gly 10 × transfer buffer, 0.45-µm nitrocellulose membrane, and Silver Stain Plus Kit from Bio-Rad (Hercules, CA); Protogel from National Diagnostics (Atlanta, GA); Vectastain anti-rabbit IgG (peroxidaseconjugated) ABC kit from Vector Laboratories (Burlingame, CA); minimum essential medium from Gibco-BRL (Grand Island, NY). Antisera that recognize covalently bound metabolites of diclofenac (16) or the trifluoroacetyl (TFA) chloride metabolite of halothane (21) were prepared as previously described. Antiserum that reacts with covalent metabolites of acetaminophen (APAP) (22) was a gift from Dr. Neil R. Pumford (University of Arkansas, Little Rock), while aminopeptidase N (AmN) (23) and sucrase-isomaltase (SI) (24) antisera were gifts from Dr. Nigel Bunnett (University of California, San Francisco) and Dr. Ward Olsen (University of Wisconsin, Madison), respectively. All antisera were prepared in rabbits. Animals. Female Lewis rats (175-200 g; Charles River, Wilmington, MA) were used for the diclofenac and halothane studies, while male B6C3F1 mice (25 g; NCI, Frederick, MD) were used for the studies with APAP. All animals were acclimated following a 12-h light/dark cycle in a humidity- and temperature-controlled environment, for 1 week before experimentation in accordance to the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Immunoblot Analysis of Diclofenac Adducts. SDSPAGE and immunoblot analysis were performed as previously described (25). The immunoblots were incubated with 1:2500 dilutions of diclofenac antisera and 1:10000 of anti-rabbit IgG (peroxidase-conjugated), followed by enhanced chemiluminescence detection, according to the manufacturer’s instructions (Amersham, Arlington Heights, IL). Preparation of Homogenates of Liver and Small Intestine. Rats were administered diclofenac at doses of 1, 10, 25, and 50 mg/kg (dissolved in water) or water by gastric gavage at 2100 (9 p.m.). The animals were killed at 4, 12, or 24 h after dosing, and the liver and small intestine (90 cm, starting at the pyloric sphincter and ending at the ileal-cecal junction) were removed. The small intestine was segmented into five equal sections (15-18 cm/section). PBS containing 1 mM DTT was used to gently flush out the luminal contents of each section. Tissues were frozen immediately in liquid N2 and stored at -80 °C until used for further analysis. Livers were homogenized with a Tekmar tissuemizer (Tekmar Co., Cincinnati, OH) in icecold 100 mM Tris-acetate (pH 7.5), containing 250 mM sucrose, 2 mM EDTA, and protease inhibitors: 1 mM pepstatin, 40 µg/ mL PMSF, and 1 mM leupeptin (homogenization buffer). The small intestine was homogenized in a similar manner, with 1 mM DTT and heparin (3 units/mL) added to the homogenization buffer to aid in the solubilization of mucus. Homogenates were frozen in liquid N2 and stored at -80 °C until they were subjected to SDS-PAGE and immunoblot analysis with diclofenac antisera. Isolation of Enterocytes. Ten rats were administered diclofenac (50 mg/kg) by gavage. After 12 h, the animals were killed and enterocytes were isolated by the chelation/elution method of Weiser (26) with modifications described by others (27-29). In short, approximately 35-40 cm of the small intestine (beginning 10 cm distal to the pyloric sphincter) was removed and rinsed with 40 mL of PBS, containing 1 mM DTT. The section of intestine was filled with solution A (PBS, containing 27 mM sodium citrate, 0.5 mM DTT, and 40 µg/mL PMSF) followed by the clamping of the proximal and distal lumen with hemostats and incubation in PBS, containing 20% glycerol for 15 min at 37 °C, with gentle mixing in a Dubnoff shaker bath. Elution of enterocytes was done by filling the lumen with solution B (PBS, containing 1.5 mM EDTA, 0.5 mM DTT, and 40 µg/mL PMSF) and gently mixing for 6, 10, and 20 min. Enterocytes from each elution were pooled and passed through a 112-µm nylon mesh screen into a polypropylene tube

166 Chem. Res. Toxicol., Vol. 11, No. 3, 1998 and sedimented at 50g for 5 min at 4 °C. The cells were washed with solution B twice, suspended in homogenization buffer (1: 3, v/v), pulse-sonicated, and homogenized with a Potter/Elvehjm homogenizer. Subcellular fractions were obtained by differential centrifugation with the use of a Beckman Optima TLX ultracentrifuge. The homogenate was transferred to 1.5-mL polyallomer tubes and centrifuged at 10000g for 15 min to yield a 10000g pellet and supernatant. The 10000g supernatant was then centrifuged at 100000g for 20 min to yield a 100000g pellet (microsome) and a 100000g supernatant (cytosol). The fractions were stored at -80 °C until subjected to SDS-PAGE and immunoblot analysis with diclofenac antisera. N-Terminal Amino Acid Sequence Determination and Glycoprotein Nature of 142- and 130-kDa Adducts in Enterocytes. The 10000g pellet fraction of enterocytes (5 mg/ mL) was homogenized in PBS, containing 25 mM OG, 1 mM leupeptin, 1 mM EDTA, 1 mM pepstatin, and 40 µg/mL PMSF, and then gently rotated for 4 h at 4 °C. After centrifugation at 100000g for 20 min, the supernatant extract (S1) was retained, and the pellet was reextracted overnight (approximately 10 h), to yield a 100000g pellet and a second supernatant extract (S2). The fractions were subjected to SDS-PAGE and immunoblot analysis with diclofenac antisera or were transferred from SDSPAGE gels onto PVDF membrane for N-terminal amino acid sequence analysis by automated Edman degradation (30). The glycoprotein nature of the 142- and 130-kDa proteins labeled by diclofenac in the S2 extract of enterocytes was evaluated by the treatment of sample with either N-glycosidase-F or endoglycosidase-H. Briefly, 10 µg of S2 extract was incubated at 37 °C for 20 h with 0.4 unit of N-glycosidase-F in 50 mM sodium phosphate (pH 7.0), containing 0.5% (w/v) OG, 0.02% (v/v) SDS, and 1% (v/v) mercaptoethanol, or with 2.5 mU endoglycosidase-H in 20 mM sodium phosphate (pH 5.5), containing 0.002% (v/v) SDS and 1% (v/v) mercaptoethanol. The reaction mixtures were subjected to SDS-PAGE and silver staining (per manufacturer’s guidelines, Bio-Rad, Hercules, CA) or to immunoblot analysis with diclofenac antisera. Sucrase-isomaltase and Aminopeptidase N Assays. Intestines were removed from rats 12 h after treatment with diclofenac (50 mg/kg diclofenac) by gavage. Three 5-cm segments were taken from the distal jejunum (50 cm post-pyloric sphincter), snap-frozen in liquid N2, and stored at -80 °C until analyzed. The tissues (150-200 mg) were homogenized in PBS, containing 1 mM pepstatin, 40 µg/mL PMSF, and 1 mM leupeptin (80 mg/mL, w/v), with a Potter/Elvehjm homogenizer. SI activity was determined by a previously described method for use in 96-well flat-bottom microtiter plates (31). The amount of glucose formed was determined from a standard curve, prepared from glucose, with SI activity units calculated as originally described by Dahlqvist (32). AmN activity was determined by an adapted method of van Hal et al. (33) for use in 96-well microtiter plates. In short, 50 µL of 10 mM Ala-pnitroanilide substrate was incubated with 50 µL of tissue homogenate (diluted 1:25 or 1:50, v/v) at 37 °C for 30 min. The amount of p-nitroanilide formed was determined from a standard curve of p-nitroanilide. SI and AmN activities were normalized per mg of protein or to relative amounts of SI or AmN in samples as determined by laser densitometric scans of immunoblots developed with SI or AmN antisera using Imagequant software (Molecular Dynamics personal densitometer, Sunnyvale, CA). Immunohistochemistry of Small Intestine Sections. Rats were treated with diclofenac (50 mg/kg) by gavage or with halothane (10 mmol/kg, dissolved in sesame oil) by intraperitoneal injection and killed after 12 or 16 h, respectively. Mice were treated with APAP (400 mg/kg, dissolved in 40% propylene glycol) by gavage and were killed after 4 h. Isolation of sections of small intestines was as described above. The tissues were further cross-sectioned (2 mm), fixed in neutral buffered formalin (3.7%, v/v, formaldehyde in 0.029 M NaH2PO4‚H2O and 0.046 M NaHPO4‚7H2O, pH 7.0), and then embedded in paraffin and mounted on poly(L-lysine)-treated glass slides by American

Ware et al. Histolabs (Gaithersburg, MD). After deparaffination, sections were stained immunohistochemically for diclofenac-, halothane-, or APAP-labeled proteins. All procedures were done at room temperature. Endogenous peroxidase activity was inhibited by incubations of the sections with 0.3% (v/v) H2O2 in methanol for 30 min. After blocking nonspecific binding with 1.5% (v/v) normal goat serum in PBS for 30 min, the sections were treated for 60 min with rabbit anti-diclofenac (1:1000), anti-halothane (1:500), or anti-APAP (1:1000) sera. Labeled proteins were visualized with a Vectastain immunoperoxidase ABC kit, following manufacture’s instructions (Vector Laboratories Inc., Burlingame, CA). The tissue was then counterstained with Gill No. 1 hematoxylin solution. Diclofenac Incubation with Isolated Enterocytes or Enterocyte Microsomes. Enterocytes were isolated from the small intestine of three female Lewis rats by chelation/elution method as described above. The cells were washed with Hanks’ balanced saline solution and suspended in minimum essential medium that was supplemented with the sodium salt of penicillin G (500 units/mL) and streptomycin (500 µg/mL); 1 mL of the enterocyte suspension (3.2 mg/mL protein) was incubated with 1 mM diclofenac in a Dubnoff shaker bath at 37 °C. After 4 h, the cells were pelleted, sonicated, and homogenized as described above. Diclofenac (1 mM) was also incubated with enterocyte microsomes (1 mg/mL) in 0.1 M potassium phosphate (pH 7.4), containing NADPH (2 mM) and EDTA (1 mM) at 37 °C for 12 h. Both mixtures were stored at -80 °C until subjected to SDS-PAGE and immunoblot analysis with diclofenac antisera. Bile Duct Cannulation. A female Lewis rat was anesthetized by an intramuscular injection of ketamine (45 mg/kg) and xylazine (15 mg/kg), and the bile duct was cannulated and exteriorized as described by Wayneforth and Flecknell (34). Diclofenac (25 mg/kg) was administered by tail vein injection. After 8 h, the animal was killed, and small intestinal homogenates were prepared for SDS-PAGE and immunoblotting with diclofenac antisera as described above. Other Methods. Protein concentrations were determined using BCA protein reagent kit with bovine serum albumin as standard. Data are reported as mean ( SD. Statistical analysis was done by employing the Student’s t-test for unpaired samples. A probability level of p < 0.05 was considered significant.

Results Detection of Diclofenac-Labeled Proteins in Rat Small Intestine. Immunoblot analyses with diclofenac antisera of homogenates of liver and small intestine 12 h after diclofenac treatment revealed a dose-dependent increase in the formation of protein adducts (Figure 1). As previously shown, the liver had major adducts of 110 (17) and 51 (18) kDa. The adducts at approximately 140 and 200 kDa were present in lower levels than what had been reported in previous studies (17). This may be due to differences in dose and route of diclofenac administration and/or to differences in rat strain and sex of animals utilized. The adduct at approximately 66 kDa may correspond to labeled albumin that was present in the sample as a result of blood contaminating the samples (17). In contrast, adducts of 142 and 130 kDa were most prominent in the small intestine. Minor adducts were detected at 110 and 66 kDa (Figure 1). A high-molecularmass protein of approximately 200 kDa was also found to be present. The specificity of the immunochemical staining was demonstrated by the absence of immunoreactivity with preimmune sera and the blocking of the immunochemical staining of most of the protein fractions, when diclofenac antiserum was preincubated with di-

Drug-Protein Adducts in Rat Small Intestine

Figure 1. Dose-dependent covalent modification of hepatic and small intestine proteins in diclofenac-treated rats. Rats were given 0 (lanes 1 and 2, water control) or 1 (lanes 3 and 4), 10 (lanes 5 and 6), 25 (lanes 7 and 8), and 50 (lanes 9 and 10) mg/ kg diclofenac by gavage. After 12 h, proteins from liver (odd number lanes) and small intestine (even number lanes) homogenates were separated by SDS-PAGE (150 µg/lane), transferred to nitrocellulose, and immunoblotted with diclofenac antisera.

Figure 2. Subcellular localization of diclofenac-labeled proteins in isolated enterocytes. Rats were given 50 mg/kg diclofenac by gavage, and after 12 h small intestines were removed and enterocytes were isolated: lane 1, enterocyte protein homogenate from vehicle-treated rats; lanes 2, 3, 4, and 5, enterocyte homogenate, 10000g pellet, microsomes, and cytosolic fractions, respectively, from diclofenac-treated rats. Proteins were separated by SDS-PAGE (25 µg/lane), transferred to nitrocellulose, and immunoblotted with diclofenac antisera.

clofenac (data not shown). While greater than 90% of the immunoreactivity of the 142-, 130-, 110-, and 66-kDa fractions could be inhibited by preincubation of diclofenac antisera with 10 µM diclofenac, the 200-kDa adduct could not be inhibited even when 1 mM diclofenac was preincubated with diclofenac antisera (data not shown). This finding suggests that the immunochemical staining of the 200-kDa fraction was not due to the binding of diclofenac antisera to covalently bound diclofenac. The time to peak intensity of gut diclofenac-protein adducts occurred between 4 and 12 h postdosing of diclofenac. While the longitudinal distribution of diclofenac-protein adducts indicated that the highest concentration of adducts occurred in 40-60 cm of post-pyloric sphincter, however, all segments showed diclofenac labeling (data not shown). Identification of the 142- and 130-kDa Diclofenac Adducts in Enterocytes. Immunoblot analysis showed that the 142-, 130-, and 110-kDa adducts were associated predominately with the membrane-bound fractions of isolated enterocytes, 12 h after treatment with diclofenac (Figure 2, lanes 2-4). Another adduct of approximately 55 kDa, not previously detected in whole tissue homo-

Chem. Res. Toxicol., Vol. 11, No. 3, 1998 167

Figure 3. Extraction of 142- and 130-kDa diclofenac-labeled proteins from the 10000g pellet of enterocytes. The 10000g pellet of enterocytes from diclofenac-treated rats was homogenized and extracted with 25 mM OG: lane 1, homogenate; lane 2, 100000g supernatant after extraction (S1); lane 3, 100000g supernatant after second extraction (S2); lane 4, 100000g pellet after second extraction. Proteins in the various fractions were separated by SDS-PAGE and stained with Coomassie blue dye (panel A) or transferred to nitrocellulose and immunoblotted with diclofenac antisera (panel B).

genates, was observed in the enterocyte lysate and was concentrated in the 10000g pellet fraction of the enterocytes (Figure 2, lane 3). The 142-, 130-, and 110-kDa adducts of diclofenac in the 10000g membrane fractions of enterocytes were selectively extracted out with 25 mM OG, particularly in the second extraction (S2) (Figure 3A,B, lanes 2 and 3). In contrast, the 55-kDa adduct was not soluble in 25 mM OG (Figure 3A,B, lane 4), 100 mM OG, or 2.2% (w/ v) sodium deoxycholic acid (data not shown). The S2 extract was immunoblotted onto PVDF membrane and the N-terminal amino acid sequences of the fractions corresponding to the 142- and 130-kDa adducts were determined by automated Edman degradation. The N-terminal amino acid sequence of the 142-kDa adduct, A K G F Y I S K T L G I, showed 100% identity to the glycoprotein AmN (35), also known as CD13, while the 130-kDa adduct, I K L P S N P I S T L R, showed 100% identity with the sucrase subunit of the glycoprotein SI (36). No other proteins in databases showed identity or near identity to these sequences. These assignments were confirmed by comparing the N-linked oligosaccharide characteristics of the 142- and 130-kDa adducts with that of AmN or SI, respectively (Figure 4). Treatment of the S2 extract of enterocytes from diclofenac-treated rats with N-glycosidase-F led to losses of both adducts, as determined by immunoblot analysis of the reaction mixtures with diclofenac antisera (Figure 4B, lane 5). Similar losses of immunoreactive AmN (Figure 4C, lane 8) and SI (Figure 4D, lane 11) were found to occur with this treatment. A major protein adduct at 115 kDa appeared after N-glycosidase-F treatment (Figure 4B, lane 5), which corresponded to a peptide fragment of AmN that was produced by N-glycosidase-F treatment (Figure 4C, lane 8). No change in mass of the 142-kDa adduct (Figure 4B, lane 6) or AmN (Figure 4C, lane 9) was found when the S2 extract was incubated with endoglycosidase-H, indicating that the mature form of this glycoprotein had been covalently altered by diclofenac (37). In contrast, endoglycosidase-H treatment resulted in a change in the mass of the 130-kDa diclofenac adduct to 127 kDa (Figure 4B, lane 6). Similar changes in the mass of the SI protein were found by this treatment (Figure 4D, lane 12) and are consistent with that reported by other researchers (38).

168 Chem. Res. Toxicol., Vol. 11, No. 3, 1998

Ware et al.

Figure 4. Glycoprotein nature of the 142- and 130-kDa diclofenac-labeled proteins in the S2 extract of the enterocytes. The S2 extract of the 10000g fractions of enterocytes from diclofenac-treated rats was incubated without (lanes 1, 4, 7, and 10) or with N-glycosidase-F (lanes 2, 5, 8, and 11) or endoglycosidase-H (lanes 3, 6, 9, and 12). After 20 h, proteins in the reaction mixtures were separated by SDS-PAGE (10 µg/lane) and silver-stained (panel A) or transferred to nitrocellulose and immunoblotted with diclofenac (panel B), AmN (panel C), or SI (panel D) antisera.

Twelve hours after treatment with diclofenac (50 mg/ kg), no statistically significant changes (p > 0.05) in SI (238 ( 129.4 versus control 137.3 ( 37.4 activity units/ mg of protein) or AmN (9.3 ( 1.8 versus 8.4 ( 2.0 µM/ min/mg of protein) activity were observed. Moreover, no significant differences were observed when activities were normalized per gram of tissue or to the relative levels of SI or AmN in the tissues as determined by laser densitometry determination of the respective immunoblot (data not shown). Immunohistochemical Localization of Protein Adducts of Diclofenac, Acetaminophen, and Halothane in the Small Intestine. Immunohistochemical analysis of small intestine sections 12 h after diclofenac treatment revealed that protein adducts were primarily localized in the villus enterocytes of the small intestine (Figure 5A), where they were in juxtaposition to intraepithelial lymphocytes (IEL) and lymphocytes of the lamina propria (LP) (Figure 5D). Protein adducts of diclofenac were also found in the follicle-associated epithelium (FAE) of the dome of Peyer’s patch (PP) (Figure 5D). No immunochemical staining was observed when preimmune serum was substituted for the immune serum (Figure 5B) or when the immunochemical staining was done with sections from the control animals (data not shown). The specificity of the immunochemical detection of diclofenac adducts was confirmed by the finding that preincubation of the diclofenac antisera with diclofenac inhibited the immunochemical staining of enterocytes (Figure 5C). Similarly, covalently bound metabolites of halothane and APAP were also detected immunochemically in villus-tip enterocytes in the small intestine of rat 16 h after treatment with halothane (Figure 5E) or in that of the mouse 4 h after the administration of APAP by gavage (Figure 5F). Preincubation of the antisera with N--(trifluoroacetyl)-L-lysine (21) or APAP (22) inhibited the immunoreactivity against the covalent adducts of halothane or APAP, respectively, confirming the specificity of these reactions (data not shown). Mechanism of Diclofenac Adduct Formation in Rat Small Intestine. To determine whether diclofenac could be bioactivated by enzymes in enterocytes, diclofenac was incubated with enterocytes or enterocyte microsomes. When the reaction mixtures were subjected to immunoblot analysis, no protein adducts of diclofenac were detected (data not shown). This finding suggested that the metabolite(s) of diclofenac responsible for the formation of the protein adducts in enterocytes in vivo was likely formed at another site, possibly in the liver. To test this idea, diclofenac was administered intrave-

nously to two rats, one of which had its bile duct cannulated, and after 8 h the level of protein adduct formation in homogenates of small intestine was evaluated. The intravenous administration of diclofenac resulted in a similar pattern of protein adduct formation to that found 12 h after oral administration (Figure 6, lanes 1 and 2). In contrast, cannulation of the bile duct extensively inhibited adduct formation following the intravenous administration of diclofenac (Figure 6, lane 3).

Discussion Oral tolerance is an active arm of the immune system that is known to produce systemic immunological hyporesponsiveness to antigen through the oral feeding of soluble antigen. This has been shown to affect both humoral and cell-mediated immune responses (3). The dose and type of antigen have been shown to be the primary determinants in the mechanism(s) through which oral tolerance may occur. In particular, high doses of antigen have been shown to produce anergy and clonal deletion (5-7), while low doses of antigen can lead to the production of regulatory cytokines that have a selective suppressor effect on the immune response (4). These findings have recently led researchers to evaluate the efficacy of the oral administration of autoantigens associated with a variety of autoimmune diseases such as multiple sclerosis, rheumatoid arthritis, uveoretinitis, and type I diabetes as a therapeutic approach for the treatment of these diseases (39). Although the mechanism(s) of oral tolerance remains to be clearly defined, it is known that the interaction of soluble antigens with GALT is a requirement for both the induction and the maintenance of this response (40). It appears that several different types of cells within GALT may have a role in the development of oral tolerance. Enterocytes in their constitutive state express low levels of major histocompatibility complex class II antigens and may function as weak antigen-presenting cells (41). Recent studies suggest that enterocytes may participate directly in oral tolerance to the hapten DNCB by inhibiting T-cell-mediated immune responses through the secretion of the immunosuppressive cytokine, transforming growth factor-β (42). Immediately beneath the monolayer of epithelium is the LP. The LP contains a heterogeneous population of cells that include T- and B-cells, plasma cells, macrophages, mast cells, dendritic cells, and eosinophils (40). The finding that a delayedtype hypersensitivity reaction could be inhibited when

Drug-Protein Adducts in Rat Small Intestine

Chem. Res. Toxicol., Vol. 11, No. 3, 1998 169

Figure 5. Immunohistochemical detection of protein adducts in enterocytes of the small intestine of animals treated with diclofenac, halothane, or APAP. Panels A-D show small intestine sections 12 h after treatment of rats with diclofenac (50 mg/kg) by gavage: panel A, immunochemical staining of protein adducts with diclofenac antisera (200× magnification); panel B, no immunochemical staining with preimmune serum (200× magnification); panel C, immunochemical staining inhibited by preincubation of diclofenac antiserum with 10 µM diclofenac (200× magnification); panel D, same as panel A (630× magnification); VT, villus tips; PP, Peyer’s patch; IEL, intraepithelial lymphocyte; LP, lamina propria. Diclofenac-protein adducts were detected in enterocytes of villus, juxtaposition to IEL and LP lymphocytes, and in the follicle-associated epithelium of PP. Immunochemical staining of protein adducts: panel E, with halothane antisera of villus-tip enterocytes in the small intestine 16 h after treatment of a rat with halothane (10 mmol/kg) by intraperitoneal injection (630× magnification); panel F, with APAP antisera in villus-tip enterocytes in the small intestine 4 h after treatment of a mouse with APAP (400 mg/kg) by gavage (630× magnification).

Figure 6. Effect of bile duct cannulation on diclofenac adduct formation in the rat small intestine. The bile duct of a rat was cannulated before the intravenous administration of diclofenac (25 mg/kg, lane 3). Other rats received the same amount of diclofenac either by gavage (lane 1) or intravenously (lane 2); 8 h after the intravenous administration and 12 h after the gavage, the animals were killed, small intestines were homogenized, and proteins were separated by SDS-PAGE (100 µg/ lane), transferred to nitrocellulose, and immunoblotted with diclofenac antisera.

antigen-pulsed LP cells were adoptively transferred to syngeneic animals prior to antigenic challenge indicates that LP cells have a role in oral tolerance (43). Intraepithelial lymphocytes primarily express CD8+ and either

R/β or γ/δ T-cell receptor repertoires (44). Of particular interest are recent studies that indicate that γ/δ T-cell receptor knock-out mice are unable to respond to traditional oral tolerance induction protocols (45), indicating that IELs can have an important role in the development of oral tolerance. Follicle-associated epithelium is a unique epithelium associated with the covering of PP and has been shown to contain specialized M-cells that are thought to be involved in particulate antigen uptake and delivery of intact antigen to lymphoid cells within PP (46). Transforming growth factor-β-producing suppressor cells from this region induced by the oral administration of antigen have been shown to adoptively transfer T-cell tolerance (47). However, oral tolerance has been shown to occur in animals that have had PPs surgically removed from the lumen (48). Clearly, further work is needed before the complexities behind the development of oral tolerance are unraveled. The results of the present study show that protein adducts of diclofenac, APAP, and halothane can be formed in the enterocytes of the small intestine. In the case of diclofenac, protein adducts of 142, 130, 110, and 55 kDa were detected by immunoblot analysis and were formed independent of the route of administration of

170 Chem. Res. Toxicol., Vol. 11, No. 3, 1998

diclofenac (Figure 6, lanes 1 and 2). This finding is important because it shows that covalent binding of drugs in GALT can be formed when drugs are administered by routes other than oral administration. Although the proximate reactive intermediate of diclofenac in the small intestine is unknown, the results of the bile cannulation experiment (Figure 6, lanes 2 and 3) and studies with isolated enterocytes indicate that the metabolite(s) responsible for the covalent labeling of proteins in enterocytes in vivo originated in the liver and were not formed directly from the metabolism of diclofenac within the enterocytes. A likely metabolite of diclofenac that might label enterocyte proteins is its acyl glucuronide, which is formed in the liver and secreted into the bile of rats and to a lesser extent in humans (49, 50). In this regard, the acyl glucuronide metabolite of the NSAID diflunisal appears to be responsible for the covalent binding of this drug to intestinal proteins of rats (51). It is known that acyl glucuronide metabolites are chemically reactive and can covalently bind to proteins by transacylation and/or glycosylation after acyl migration within the glucuronide acid molecule (52). These metabolites are thought to be involved in idiosyncratic adverse drug reactions associated with several NSAIDs (52, 53). However, since P450s, UGTs, and other xenobiotic-metabolizing enzymes have been found in enterocytes of animals and humans, it is possible that NSAIDs or other xenobiotics can be bioactivated at this site if proper substrate specificities and levels of these enzymes are present (27, 28, 54-57). It is likely that the trifluoroacetyl chloride metabolite of halothane and the N-acetyl-p-benzoquinone imine metabolite of APAP were formed locally in the enterocytes, because these reactive species would not be expected to escape from the liver (58). The 142- and 130-kDa protein adducts of diclofenac have been identified as AmN and SI on the basis of N-terminal amino acid sequence analysis and immunochemical properties of these proteins. While neither of the enzymatic activities was significantly altered by diclofenac treatment, a trend toward an increase in SI activity was noted. No attempt was made to evaluate the effect of diclofenac on the other activities associated with these proteins. For example, SI has been recently shown to function as a receptor for Clostridium difficile toxin A, a major cause of antibiotic-associated colitis (59). While SI is only expressed in mature enterocytes (26), AmN can be found in monocytes, neutrophils, endothelial cells, smooth muscle cells, enterocytes, renal tubular epithelium, synaptic membranes within the CNS, tumor cells, and intra- and extrahepatic epithelial cells (60). Cytomegalovirus (CMV) has been shown to enter cells by binding CD13 (61). Moreover, CMV contains a CD13related component leading to CD13 autoantibodies, which may be responsible for chronic graft-versus-host reactions, graft rejection, and other autoimmune reactions (62). Therefore, it is possible that the covalent alteration of SI and AmN by diclofenac may be responsible for previously unrecognized pharmacological activities of diclofenac and possibly that of other NSAIDs, which also form acyl glucuronide metabolites. Although it remains to be determined whether the formation of protein adducts of drugs in GALT can prevent drug-allergic reactions, the previous findings with chemically reactive haptens suggest that this is a reasonable hypothesis that should be directly tested in the future (8, 9, 63, 64).

Ware et al.

References (1) Pohl, L. R., Satoh, H., Christ, D. D., and Kenna, J. G. (1988) The immunologic and metabolic basis of drug hypersensitivities. Annu. Rev. Pharmacol. Toxicol. 28, 367-387. (2) Weiner, H. L., and Mayer, L. F. (1996) Oral tolerance: Mechanisms and applications. Introduction. Ann. N. Y. Acad. Sci. 778, xiii-xviii. (3) Mowat, A. M. (1987) The regulation of immune responses to dietary protein antigens. Immunol. Today 8, 93-98. (4) Miller, A., Lider, O., and Weiner, H. L. (1991) Antigen-driven bystander suppression after oral administration of antigens. J. Exp. Med. 174, 791-798. (5) Whitacre, C. C., Gienapp, I. E., Orosz, C. G., and Bitar, D. M. (1991) Oral tolerance in experimental autoimmune encephalomyelitis. III. Evidence for clonal anergy. J. Immunol. 147, 21552163. (6) Friedman, A., and Weiner, H. L. (1994) Induction of anergy or active suppression following oral tolerance is determined by antigen dosage. Proc. Natl. Acad. Sci. U.S.A. 91, 6688-6692. (7) Chen, Y., Inobe, J., Marks, R., Gonnella, P., Kuchroo, V. K., and Weiner, H. L. (1995) Peripheral deletion of antigen-reactive T cells in oral tolerance. Nature 376, 177-180. (8) Van Hoogstraten, I. M., Boos, C., Boden, D., Von Blomberg, M. E., Scheper, R. J., and Kraal, G. (1993) Oral induction of tolerance to nickel sensitization in mice. J. Invest. Dermatol. 101, 26-31. (9) Chase, M. W. (1946) Inhibition of experimental drug allergy by prior feeding of the sensitizing agent. Proc. Soc. Exp. Biol. Med. 61, 257-259. (10) Reese, R. T., and Cebra, J. J. (1975) Anti-dinitrophenyl antibody production in strain 13 guinea pigs fed or sensitized with dinitrochlorobenzene. J. Immunol. 114, 863-871. (11) Gautam, S. C., and Battisto, J. R. (1983) Suppression of contact sensitivity and cell-mediated lympholysis by oral administration of hapten is caused by different mechanisms. Cell Immunol. 78, 295-304. (12) Banks, A. T., Zimmerman, H. J., Ishak, K. G., and Harter, J. G. (1995) Diclofenac-associated hepatotoxicity: Analysis of 180 cases reported to the Food and Drug Administration as adverse reactions. Hepatology 22, 820-827. (13) Bougie, D., Johnson, S. T., Weitekamp, L. A., and Aster, R. H. (1997) Sensitivity to a metabolite of diclofenac as a cause of acute immune hemolytic anemia. Blood 90, 407-413. (14) Salama, A., Schutz, B., Kiefel, V., Breithaupt, H., and MuellerEckhardt, C. (1989) Immune-mediated agranulocytosis related to drugs and their metabolites: Mode of sensitization and heterogeneity of antibodies. Br. J. Haematol. 72, 127-132. (15) van der Klauw, M. M., Wilson, J. H., and Stricker, B. H. (1996) Drug-associated anaphylaxis: 20 years of reporting in The Netherlands (1974-1994) and review of the literature. Clin. Exp. Allergy 26, 1355-1363. (16) Pumford, N. R., Myers, T. G., Davila, J. C., Highet, R. J., and Pohl, L. R. (1993) Immunochemical detection of liver protein adducts of the nonsteroidal antiinflammatory drug diclofenac. Chem. Res. Toxicol. 6, 147-150. (17) Hargus, S. J., Amouzedeh, H. R., Pumford, N. R., Myers, T. G., McCoy, S. C., and Pohl, L. R. (1994) Metabolic activation and immunochemical localization of liver protein adducts of the nonsteroidal antiinflammatory drug diclofenac. Chem. Res. Toxicol. 7, 575-582. (18) Shen, S., Hargus, S. J., Martin, B. M., and Pohl, L. R. (1997) Cytochrome P4502C11 is a target of diclofenac covalent binding in rats. Chem. Res. Toxicol. 10, 420-423. (19) Kretz-Rommel, A., and Boelsterli, U. A. (1994) Selective protein adducts to membrane proteins in cultured rat hepatocytes exposed to diclofenac: Radiochemical and immunochemical analysis. Mol. Pharmacol. 45, 237-244. (20) Gil, M. L., Ramirez, M. C., Terencio, M. C., and Castell, J. V. (1995) Immunochemical detection of protein adducts in cultured human hepatocytes exposed to diclofenac. Biochim. Biophys. Acta 1272, 140-146. (21) 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. (22) Matthews, A. M., Roberts, D. W., Hinson, J. A., and Pumford, N. R. (1996) Acetaminophen-induced hepatotoxicity. Analysis of total covalent binding vs specific binding to cysteine. Drug Metab. Dispos. 24, 1192-1196.

Drug-Protein Adducts in Rat Small Intestine (23) Terashima, H., Wong, H., Kobayashi, R., and Bunnett, N. W. (1992) Immunochemical localization of aminopeptidase M in the alimentary tract of the guinea pig and rat. Gastroenterology 102, 1867-1876. (24) Lorenzsonn, V., Korsmo, H., and Olsen, W. A. (1987) Localization of sucrase-isomaltase in the rat enterocyte. Gastroenterology 92, 98-105. (25) Amouzadeh, H. R., and Pohl, L. R. (1995) Processing of endoplasmic reticulum luminal antigens associated with halothane hepatitis in rat hepatocytes. Hepatology 22, 936-943. (26) Weiser, M. M. (1973) Intestinal epithelial cell surface membrane glycoprotein synthesis. I. An indicator of cellular differentiation. J. Biol. Chem. 248, 2536-2541. (27) Bonkovsky, H. L., Hauri, H. P., Marti, U., Gasser, R., and Meyer, U. A. (1985) Cytochrome P450 of small intestinal epithelial cells. Immunochemical characterization of the increase in cytochrome P450 caused by phenobarbital. Gastroenterology 88, 458-467. (28) Watkins, P. B., Wrighton, S. A., Schuetz, E. G., Molowa, D. T., and Guzelian, P. S. (1987) Identification of glucocorticoid-inducible cytochromes P-450 in the intestinal mucosa of rats and man. J. Clin. Invest. 80, 1029-1036. (29) Traber, P. G., Gumucio, D. L., and Wang, W. (1991) Isolation of intestinal epithelial cells for the study of differential gene expression along the crypt-villus axis. Am. J. Physiol. 260, G895G903. (30) Martin, B. M., and Eliason, W. K. (1991) Glycopeptidase F treatment and amino acid sequence determination of glycoproteins immobilized on PVDF and its derivatives. In Techniques in Protein Chemistry (Villafranca, J. J., Ed.) pp 191-196, Academic Press, San Diego. (31) Quezada-Calvillo, R., Markowitz, A. J., Traber, P. G., and Underdown, B. J. (1993) Murine intestinal disaccharidases: Identification of structural variants of sucrase-isomaltase complex. Am. J. Physiol. 265, G1141-G1149. (32) Dahlqvist, A. (1964) Assay of intestinal disaccharidases. Anal. Biochem. 7, 18-25. (33) van Hal, P. T., Hopstaken-Broos, J. P., Prins, A., Favaloro, E. J., Huijbens, R. J., Hilvering, C., Figdor, C. G., and Hoogsteden, H. C. (1994) Potential indirect antiinflammatory effects of IL-4. Stimulation of human monocytes, macrophages, and endothelial cells by IL-4 increases aminopeptidase-N activity (CD13; EC 3.4.11.2). J. Immunol. 153, 2718-2728.

Chem. Res. Toxicol., Vol. 11, No. 3, 1998 171 (45) Ke, Y., Pearce, K., Lake, J. P., Ziegler, H. K., and Kapp, J. A. (1997) Gamma delta T lymphocytes regulate the induction and maintenance of oral tolerance. J. Immunol. 158, 3610-3618. (46) Owen, R. L., and Ermak, T. H. (1990) Structural specializations for antigen uptake and processing in the digestive tract. Springer Semin. Immunopathol. 12, 139-152. (47) Santos, L. M., al-Sabbagh, A., Londono, A., and Weiner, H. L. (1994) Oral tolerance to myelin basic protein induces regulatory TGF-beta-secreting T cells in Peyer’s patches of SJL mice. Cell Immunol. 157, 439-447. (48) Enders, G., Gottwald, T., and Brendel, W. (1986) Induction of oral tolerance in rats without Peyer’s patches. Immunology 58, 311314. (49) Stierlin, H., Faigle, J. W., Sallmann, A., Kung, W., Richter, W. J., Kriemler, H. P., Alt, K. O., and Winkler, T. (1979) Biotransformation of diclofenac sodium (Voltaren) in animals and in man. I. Isolation and identification of principal metabolites. Xenobiotica 9, 601-610. (50) Stierlin, H., and Faigle, J. W. (1979) Biotransformation of diclofenac sodium (Voltaren) in animals and in man. II. Quantitative determination of the unchanged drug and principal phenolic metabolites, in urine and bile. Xenobiotica 9, 611-621. (51) King, A. R., and Dickinson, R. G. (1993) Studies on the reactivity of acyl glucuronides-IV. Covalent binding of diflunisal to tissues of the rat. Biochem. Pharmacol. 45, 1043-1047. (52) Spahn-Langguth, H., and Benet, L. Z. (1992) Acyl glucuronides revisited: Is the glucuronidation process a toxification as well as a detoxification mechanism? Drug Metab. Rev. 24, 5-47. (53) Zia-Amirhosseini, P. Z., Harris, R. Z., Brodsky, F. M., and Benet, L. Z. (1995) Hypersensitivity to nonsteroidal antiinflammatory drugs [letter]. Nature Med. 1, 2-4. (54) Ilett, K. F., Tee, L. B., Reeves, P. T., and Minchin, R. F. (1990) Metabolism of drugs and other xenobiotics in the gut lumen and wall. Pharmacol. Ther. 46, 67-93. (55) Kaminsky, L. S., and Fasco, M. J. (1991) Small intestinal cytochromes P450. Crit. Rev. Toxicol. 21, 407-422. (56) Kolars, J. C., Benedict, P., Schmiedlin-Ren, P., and Watkins, P. B. (1994) Aflatoxin B1-adduct formation in rat and human small bowel enterocytes. Gastroenterology 106, 433-439.

(34) Waynforth, H. B., and Flecknell, P. A. (1992) Experimental and Surgical Technique in the Rat, pp 206-209, Academic Press, San Diego.

(57) Murray, G. I., and Burke, M. D. (1995) Immunohistochemistry of drug-metabolizing enzymes. Biochem. Pharmacol. 50, 895-903.

(35) Watt, V. M., and Yip, C. C. (1989) Amino acid sequence deduced from a rat kidney cDNA suggests it encodes the Zn-peptidase aminopeptidase N. J. Biol. Chem. 264, 5480-5487.

(58) Gillette, J. R. (1974) Commentary. A perspective on the role of chemically reactive metabolites of foreign compounds in toxicity. I. Correlation of changes in covalent binding of reactivity metabolites with changes in the incidence and severity of toxicity. Biochem. Pharmacol. 23, 2785-2794.

(36) Chandrasena, G., Osterholm, D. E., Sunitha, I., and Henning, S. J. (1994) Cloning and sequencing of a full-length rat sucraseisomaltase-encoding cDNA. Gene 150, 355-360. (37) Semenza, G. (1986) Anchoring and biosynthesis of stalked brush border membrane proteins: Glycosidases and peptidases of enterocytes and renal tubuli. Annu. Rev. Cell Biol. 2, 255-313. (38) Naim, H. Y., Sterchi, E. E., and Lentze, M. J. (1988) Biosynthesis of the human sucrase-isomaltase complex. Differential O-glycosylation of the sucrase subunit correlates with its position within the enzyme complex. J. Biol. Chem. 263, 7242-7253. (39) Weiner, H. L. (1997) Oral tolerance for the treatment of autoimmune diseases. Annu. Rev. Med. 48, 341-351. (40) Mowat, A. M., and Viney, J. L. (1997) The anatomical basis of intestinal immunity. Immunol. Rev. 156, 145-166. (41) Bland, P. W., and Kambarage, D. M. (1991) Antigen handling by the epithelium and lamina propria macrophages. Gastroenterol. Clin. North Am. 20, 577-596. (42) Galliaerde, V., Desvignes, C., Peyron, E., and Kaiserlian, D. (1995) Oral tolerance to haptens: Intestinal epithelial cells from 2,4dinitrochlorobenzene-fed mice inhibit hapten-specific T cell activation in vitro. Eur. J. Immunol. 25, 1385-1390. (43) Harper, H. M., Cochrane, L., and Williams, N. A. (1996) The role of small intestinal antigen-presenting cells in the induction of T-cell reactivity to soluble protein antigens: Association between aberrant presentation in the lamina propria and oral tolerance. Immunology 89, 449-456. (44) Mowat, A. M. (1990) Human intraepithelial lymphocytes. Springer Semin. Immunopathol. 12, 165-190.

(59) Pothoulakis, C., Gilbert, R. J., Cladaras, C., Castagliuolo, I., Semenza, G., Hitti, Y., Montcrief, J. S., Linevsky, J., Kelly, C. P., Nikulasson, S., Desai, H. P., Wilkins, T. D., and LaMont, J. T. (1996) Rabbit sucrase-isomaltase contains a functional intestinal receptor for Clostridium difficile toxin A. J. Clin. Invest. 98, 641649. (60) Shipp, M. A., and Look, A. T. (1993) Hematopoietic differentiation antigens that are membrane-associated enzymes: Cutting is the key! Blood 82, 1052-1070. (61) Giugni, T. D., Soderberg, C., Ham, D. J., Bautista, R. M., Hedlund, K. O., Moller, E., and Zaia, J. A. (1996) Neutralization of human cytomegalovirus by human CD13-specific antibodies. J. Infect. Dis. 173, 1062-1071. (62) Soderberg, C., Larsson, S., Rozell, B. L., Sumitran-Karuppan, S., Ljungman, P., and Moller, E. (1996) Cytomegalovirus-induced CD13-specific autoimmunitysa possible cause of chronic graftvs-host disease. Transplantation 61, 600-609. (63) Lowney, E. D. (1968) Tolerance of a contact sensitizer in man. Lancet 1, 1377. (64) Van Hoogstraten, I. M., Andersen, K. E., Von Blomberg, B. M., Boden, D., Bruynzeel, D. P., Burrows, D., Camarasa, J. G., DoomsGoossens, A., Kraal, G., and Lahti, A. (1991) Reduced frequency of nickel allergy upon oral nickel contact at an early age. Clin. Exp. Immunol. 85, 441-445.

TX970182J