The Role of Haptic Macrophages in Regulation of Idiosyncratic Drug Reactions

Transcription

1 Toxicologic Pathology, 37: 12-17, 2009 Copyright 2009 by Society of Toxicologic Pathology ISSN: print / online DOI: / The Role of Haptic Macrophages in Regulation of Idiosyncratic Drug Reactions CYNTHIA JU Department of Pharmaceutical Sciences and Integrated Department of Immunology, University of Colorado Health Sciences Center, Denver, Colorado, USA ABSTRACT Idiosyncratic drug reactions (IDR) account for approximately 6%-10% of all adverse drug reactions. The unpredictable and serious nature of these reactions makes them a significant economic burden and safety concern to the health care community and the pharmaceutical industry. Clinical and laboratory evidence suggests that adverse immune responses against drug protein adducts play a role in the pathogenesis of IDR. However, it remains unclear why only a small percentage of patients are susceptible to developing these reactions. We hypothesized that most patients develop immunological tolerance against drug protein adducts as a default mechanism, and that IDRs can only occur when this tolerance is deficient or abrogated in susceptible individuals. Using a murine model of 2,4-dinitrochlorobenzene (DNCB) induced delayed type hypersensitivity (DTH) reaction, our previously published data demonstrated that intravenous pretreatment of mice with dinitrophenyl-bovine serum albumin (DNP-BSA) induced immunological tolerance to subsequent DNCB sensitization, and that hepatic macrophages (Kupffer cells, KC) played an important role in mediating such tolerance. Further mechanistic investigation revealed that KC, acting as incompetent antigen-presenting cells, cannot elicit strong T cell reactions, and that they actively suppress T cell activation through production of prostaglandins. These findings suggest that KCs may play a critical role in regulating immune reactions within the liver and contributing to liver-mediated systemic immune tolerance. Keywords: Kupffer cells; immune tolerance; delayed-type hypersensitivity. INTRODUCTION Idiosyncratic drug reactions (IDRs) such as halothaneinduced allergic hepatitis, clozapine-induced agranulocytosis, procainamide-induced lupus, and carbamazepine-induced hypersensitivity syndrome account for 6%-10% of all adverse drug reactions (Adkinson et al. 2002). IDRs do not involve known pharmacologic properties of the drug and do not occur in most patients. Although only a small percentage of patients develop IDR, considering the number of drugs involved and the number of patients treated, these reactions are in fact quite common. Recent reports have estimated that IDRs accounted for 137,000 to 230,000 hospital admissions in the United States in 1998, with approximate costs of $275 to $600 million annually (Adkinson et al. 2002; Lazarou et al. 1998). The unpredictable and serious nature makes IDR a significant problem in clinical practice, public health, and drug development. It is important to gain better understanding of the underlying mechanism of IDR to develop strategies to predict and prevent these reactions. Supported by U.S. National Institutes of Health grant RO1 ES (to C.J.) Address correspondence to: Cynthia Ju, Department of Pharmaceutical Sciences, University of Colorado Health Sciences Center, 4200 East 9 th Avenue, Denver, CO 80262, USA; cynthia.ju@uchsc.edu. Abbreviations: APC, antigen-presenting cells; DC, dendritic cells; DNCB, 2,4-dinitrochlorobenzene; DNP-BSA, dinitrophenyl-bovine serum albumin; DTH, delayed-type hypersensitivity; IDRs, idiosyncratic drug reactions; KC, Kupffer cells; L-NMMA, N-monomethyl-L-arginine; MHC, major histocompatibility complex; NO, nitric oxide; PG, prostaglandin; PM, peritoneal macrophages; TFA, trifloroacetyl chloride; TCR, T-cell receptor. THE ROLE OF THE IMMUNE SYSTEM IN THE PATHOGENESIS OF IDR Certain clinical features of IDR suggest the involvement of the immune system. These reactions usually require prior exposure to the drug, or there is often a delay of more than a week between the start of drug therapy and the onset of toxicity. However, upon rechallenge of the culprit drug, the delay of IDR onset is shortened. In many cases, general symptoms of hypersensitivity reactions, including rash, fever, and eosinophilia, are observed in patients afflicted with IDR (Uetrecht 1999). Although it has been difficult to establish the causal link between immune reactions and tissue damage, the detections of drug-specific antibodies and T cells in patients afflicted with IDR provide laboratory evidence to support the involvement of the immune system in IDR (Bourdi et al. 1994; Bowen et al. 2004; Fibbe et al. 1986; Satoh et al. 1989; Vergani et al. 1980; von Greyerz et al. 1998; Weiss and Adkinson 1988; Zanni et al. 1998). IMMUNE RESPONSES TO DRUG PROTEIN ADDUCTS The hapten hypothesis provides an explanation for how drugs, as low-molecular-weight molecules, could elicit immune reactions. This hypothesis proposes that chemically reactive drugs, or more likely reactive metabolites of drugs, act as haptens and bind to endogenous proteins. These drug protein adducts are perceived as foreign and induce immune responses. A large amount of circumstantial evidence supports that reactive metabolite formation and covalent modification of cellular 12

2 Vol. 37, No. 1, January 2009 TOLEROGENIC FUNCTION OF KUPFFER CELLS 13 proteins are important in causing IDR. Among structurally similar anesthetics, which include halothane, isoflurane, and desflurane, the likelihood of causing IDR appears to correlate with the degree of reactive metabolite formation. Twenty percent of halothane is metabolized to trifloroacetyl chloride (TFA), and it is associated with the highest incidence of allergic hepatitis compared with isoflurane and desflurane, which are metabolized to TFA to much lesser degrees (Njoku et al. 1997; Satoh et al. 1989; Vergani et al. 1980). The sites of reactive metabolite formation often correlate with the major tissue targets of IDR. Halothane is predominantly metabolized in the liver, and the major target of halothane-induced IDR is the liver. In contrast, clozapine is oxidized to a reactive metabolite by cytochrome P450 within the liver and by myeloperoxidase in activated neutrophils, corresponding to its association with hepatotoxicity and agranulocytosis. The hapten hypothesis cannot explain how drug protein adduct formation results in pathogenic immune reactions. Although most drugs form reactive metabolites, many of them, such as acetaminophen, do not cause overt immune toxicity (Nelson and Pearson 1990). Moreover, reactive metabolites are generated in most patients, but only a small proportion will develop symptoms of IDR. Understanding why most individuals do not develop IDR may be the key to understanding the mechanism of these reactions and to identifying risk factors. One hypothesis is that immunological tolerance against drug protein adducts, rather than adverse immune responses, develops in most patients as a default mechanism, and that IDR can occur only when the tolerance is deficient or abrogated in susceptible individuals. THE ROLE OF HEPATIC KUPFFER CELLS IN IMMUNOLOGICAL TOLERANCE AGAINST HAPTEN PROTEIN ADDUCTS The hypothesis that drug-protein adducts induce immune tolerance was examined in a previous study using a mouse model of T cell mediated, delayed-type hypersensitivity (DTH) reaction against a chemically reactive hapten, 2,4- dinitrochlorobenzene (DNCB) (Ju et al. 2003). The study demonstrated that intravenous (i.v.) pretreatment of female C57BL/6J mice with a bovine serum albumin conjugate of DNCB (DNP- BSA) resulted in a significant inhibition of the DTH response to subsequent DNCB sensitization. In contrast, pretreatment of mice with BSA did not reduce DNCB-induced DTH reaction, suggesting an antigen-specific tolerance (Figure 1). Immunohistochemical analysis revealed that, after i.v. injection to mice, DNP-BSA accumulates in sinusoidal cells, identified to be Kupffer cells (KC) (Figure 2). The tolerance to DNCB-induced DTH reaction could be inhibited when KC were depleted by i.v. injection of mice with liposome-entrapped clodronate prior to DNP-BSA pretreatment (data not shown). Furthermore, the tolerance could be induced in naïve mice by adoptive transfer with a KC-enriched fraction of liver NPC obtained from mice tolerized by DNP-BSA pretreatment (Figure 3). In contrast, a KC-depleted fraction of liver NPC could not transfer tolerance (Figure 4). FIGURE 1. Induction of tolerance against 2,4-dinitrochlorobenzene (DNCB) sensitization by intravenous pretreatment with DNP-BSA (adapted from Ju et al. 2003). Female C57BL/6J mice were i.v. injected with DNP-BSA, BSA, or saline one week prior to sensitization with 5% DNCB. After five days, the mice were challenged by applying 2.5% DNCB to the ear, and DTH responses were determined twentyfour hours post-challenge by measuring ear swelling using a caliper micrometer. The results are expressed as the mean ± SEM of ten mice per group. (*) p <.05 relative to BSA or saline-treated controls. FIGURE 2. Immunohistochemical detection of DNP-labeled cells in the liver of Kupffer cell (KC)-depleted and -nondepleted mice (adapted from Ju et al. 2003). C57BL/6J female mice were i.v. injected with empty liposomes as a control (A) or liposome / clodronate to deplete KC (B). After two days the mice were i.v. injected with DNP-BSA, and two hours later, the animals were sacrificed and paraffin-embedded liver sections were stained for DNP-labeled cells with rabbit anti-dnp antisera (1 to 1000 dilution, original magnification 400x). MECHANISM OF KC-MEDIATED T CELL TOLERANCE Collectively, these data indicate that KC are a primary inducer of immunological tolerance against a hapten-induced DTH response, which is a T cell mediated immune reaction. The mechanism by which KC induce T cell tolerance was investigated in a recently published study (You et al. 2008). The expression of antigen presentation related molecules, such as

3 14 JU TOXICOLOGIC PATHOLOGY FIGURE 3. Adoptive transfer of DNP-BSA induced tolerance against 2,4-dinitrochlorobenzene (DNCB) sensitization to naïve mice with nonfractionated, T cell depleted subpopulation of liver NPC (adapted from Ju et al. 2003). Donor mice were i.v. injected with 5 mg of DNP- BSA or BSA. Four weeks later, liver NPC were isolated from both groups and then either injected directly to recipient mice (3 x 10 6 liver NPC / mouse) or used for the purification and injection into naïve mice with T cell depleted CD45 + CD90 - (non-t) subpopulation of liver NPC (8 x 10 5 cells / mouse, > 99% pure). Two hours later, recipient mice were sensitized with 5% DNCB. After six days, the DTH responses were determined by measuring ear swelling twenty-four hours post-challenge with 2.5% DNCB. The results are expressed as the mean ± SEM of ten mice per group. (*) p <.05 relative to recipients of liver NPC obtained from BSA-treated donor mice. Major Histocompatibility Complex (MHC) class II, B7-1, B7-2, and CD40, was analyzed by flow cytometry. The expression levels were compared among various antigen-presenting cells (APC), including KC, peritoneal macrophages (PM) and splenic dendritic cells (DC, known as potent professional APC). The data revealed that KC expressed similar levels of MHC II and co-stimulatory molecules, including B7-1, B7-2, and CD40, when compared with PM, but lower when compared with splenic DC (Figure 5). The abilities of KC, PM, and splenic DC to induce antigen-specific T cell activation was examined using CD4 + ovalbumin (OVA)-T cell receptor (TCR) transgenic T cells, which respond to OVA T cell activation and proliferation were determined measuring 3 H-thymidine uptake. In contrast to a strong proliferative response induced by splenic DC, T cells were nearly completely refractory to KCinduced activation (Figure 6). PM-induced T cell activation decreased with the increase of antigen concentration. This finding appeared to be a result of the increased nitric oxide (NO) production by PM at higher concentrations of OVA , as the use of a NO inhibitor, N-monomethyl-L-arginine (L-NMMA), drastically increased PM-induced T cell proliferation (Figure 6). However, NO was not detectable in cultures containing KC and T cells, and the addition of L-NMMA did not affect KCmediated T cell proliferation (Figure 6). These data indicated that the mechanisms for the lack of T cell proliferation by the two types of macrophages were different. The lack of T cell proliferation in response to KC stimulation could be owing to insufficient antigen presentation (signal 1) or inadequate co-stimulation (signal 2). The addition of anti-cd28 FIGURE 4. Adoptive transfer of tolerance against DNCB sensitization to naïve mice with liver NPC obtained from Kupffer cell (KC)-depleted or -nondepleted mice pretreated with DNP-BSA (adapted from Ju et al. 2003). Donor mice were i.v. injected with 5 mg of DNP-BSA or BSA, and after four weeks, the mice were treated with liposome/ clodronate to deplete KC or with empty liposomes as a control. After two days, liver NPC were isolated and injected to three groups of recipient mice (3 x 10 6 cells/mouse). Two groups of naïve recipient mice were injected with liver NPC isolated from KC-nondepleted donor mice that were pretreated with BSA or DNP-BSA. The third group was injected with liver NPC isolated from KC-depleted mice that were pretreated with DNP-BSA (NPC-KC). Two hours later, mice were sensitized with 5% DNCB. After six days, DTH responses were determined by measuring ear swelling twenty-four hours post-challenge with 2.5% DNCB. The results are expressed as the mean ± SEM of ten mice per group. (*) p <.05 relative to recipients of liver NPC from BSA-treated donor mice. antibody in some KC/T cell co-cultures caused only a one-fold increase in T cell proliferation (data not shown), suggesting that enhancing signal 2 could not restore T cell activation to the level that was induced by splenic DC. However, KC were found to be as potent as DC in eliciting T cell proliferation when anti-cd3 antibody induced T cell activation was examined (Figure 7). These data suggested that KC are poor APC owing to their inability to present antigens. Interestingly, when KC were included in the co-cultures of DC and T cells, they could suppress DC-induced T cell activation (Figure 8). The studies ruled out the possible involvement of interleukin-10, NO, 2,3-dioxygenase, and transforming growth factor (TGF)-β in KC-mediated T cell suppression (data not shown). However, the data revealed that KC produce significant amounts of prostaglandin (PG)E 2 and 15d-PGJ 2, much higher than the levels detected in the culture supernatants of T cell alone or DC/T cell co-cultures (Figure 9A). Addition of indomethacin inhibited KC-mediated T cell suppression, and this inhibition was reversed by the addition of exogenous PGE 2 and 15d-PGJ 2 (Figure 9B). Collectively, these results suggested that PGE 2 and/or 15d-PGJ 2 produced by KC play an

4 Vol. 37, No. 1, January 2009 TOLEROGENIC FUNCTION OF KUPFFER CELLS 15 FIGURE 5. Phenotypic analysis of Kupffer cells (KC) in comparison with peritoneal macrophages (PM) and splenic dendritic cells (DC) (adapted from You et al. 2008). Liver NPCs, splenocytes, and lavaged peritoneal cells were stained with various antibodies and analyzed by flow cytometry. The cells were gated on KC (CD45 + F4/80 + ), PM (CD11b + F4/80 + ), or splenic DC (CD45 + CD11c + ), and further analyzed for the expression of various surface molecules. The results are shown as histograms with fluoresce intensity on the x-axis and cell number on the y-axis. Solid black lines represent isotype control antibodies, and shaded histograms, specific staining. The data depicted represent three independent experiments producing similar results. FIGURE 6. Comparing the abilities of Kupffer cells (KC), peritoneal macrophages (PM), and dendritic cells (DC) to activate OVA TCR transgenic T cells (adapted from You et al. 2008). CD4 + OVA-TCR transgenic T cells (5 x 10 4 / well) were incubated with DC, PM, or KC (2.5 x 10 4 / well) for three days in the presence of various concentrations of OVA N-monomethyl-L-arginine (L-NMMA, 0.1 mm) was included in some KC/T cell or PM/T cell co-cultures. T cell proliferation was determined by [ 3 H]-thymidine incorporation during the last sixteen hours of incubation. The experiments were carried out in triplicate, and the results represent mean ± SEM. (*) p <.05 compared with T cells stimulated by KC. (**) p <.05 compared with T cells stimulated by PM. FIGURE 7. Kupffer cells (KC) and dendritic cells (DC) are equally potent in mediating anti-cd3 antibody induced T cell activation (adapted from You et al. 2008). CD4 + T cells (2 x 10 5 / well) isolated from C57BL/6J mice were co-cultured with KC or splenic DC in the wells of a ninety-six-well plate precoated with anti-cd3 antibody (5 μg/ml). T cell proliferation was determined by [ 3 H]-thymidine incorporation during the last sixteen hours of incubation. The experiments were carried out in triplicate, and the results represent mean ± SEM.

5 16 JU TOXICOLOGIC PATHOLOGY FIGURE 8. Kupffer cells (KC) inhibit splenic dendritic cell (DC) induced expansion of OVA-TCR transgenic T cells (adapted from You et al. 2008). (A) CD4 + OVA-TCR transgenic T cells (5 x 10 4 / well) were incubated with 2.5 x 10 4 / well of DC, KC, or DC plus KC (1:1 ratio) for three days in the presence of 10 μg/ml of OVA T cell proliferation was determined by [ 3 H]-thymidine incorporation during the last sixteen hours of incubation. The experiments were carried out in triplicate, and the results represent mean ± SEM. (*) p <.05 compared with T cells stimulated by DC alone. (B) CD4 + OVA-TCR transgenic T cells were labeled with carboxy-fluorescein diacetate succinimidyl ester (CFSE), followed by incubation with DC alone, or in the presence of KC with or without separation by a transwell membrane. OVA (10 μg/ml) were included in all cultures, and after three days the cells were analyzed for CFSE staining on CD4 + T cells. The data shown in panels A and B are representative of four independent experiments producing similar results. essential role in their suppression of antigen-specific T cell activation induced by DC. SUMMARY KC are a tolerogenic population of APC within the liver, and they play an important role in the induction of tolerance to hapten protein adducts. These cells may have a similar regulatory effect on immune reactions to drug protein adducts, because the liver is the predominant site of drug metabolism, hence, the major site of drug protein adduct formation. Further investigation of the underlying mechanism revealed the role of PGE 2 and 15d-PGJ 2 in mediating the immunosuppressive effect of KC. The findings suggest that impairment of the tolerogenic functions of KC, such as decreased prostaglandin production, owing to genetic or environmental factors may result in increased susceptibility to developing IDR in certain individuals. REFERENCES Adkinson, N.F. Jr., Essayan, D., Gruchalla, R., Haggerty, H., Kawabata, T., Sandler, J.D., Updyke, L., Shear, N.H. and Wierda, D. (2002). Task force report: future research needs for the prevention and management of immune-mediated drug hypersensitivity reactions. J Allergy Clin Immunol 109, S461 S478. FIGURE 9. Kupffer cell (KC) induced T cell suppression was mediated by prostaglandis (PGs) (adapted from You et al. 2008). (A) CD4 + OVA-TCR transgenic T cells were cultured alone, or in the presence of dendritic cells (DC), KC, or DC and KC. After forty-eight hours, the supernatants were collected and the levels of PGE 2 and 15d-PGJ 2 were determined. The measurements were carried out in triplicate, and the results represent mean ± SEM. (*) p <.05 compared with T cell alone. (B) DC (2.5 x 10 4 /well) and KC (2.5 x 10 4 /well) were cultured overnight prior to the addition of CD4 + OVA-TCR transgenic T cells (5 x 10 4 /well). Indomthacin (INDO, 1μM) was included in some DC/KC overnight co-cultures. Further, a combination of exogenous PGE 2 (3 ng/ml) and PGJ 2 (3 ng/ml) was added in DC/KC overnight cultures. T cell proliferation was determined after three days as described above. The measurements were carried out in triplicate and the results represent mean ± SEM. (*) p <.05 compared with T cells stimulated by DC. Bourdi, M., Tinel, M., Beaune, P.H. and Pessayre, D. (1994). Interactions of dihydralazine with cytochromes P4501A: a possible explanation for the appearance of anti-cytochrome P4501A2 autoantibodies. Mol Pharmacol 45, Bowen, D.G., Zen, M., Holz, L., Davis, T., McCaughan, G.W. and Bertolino, P. (2004). The site of primary T cell activation is a determinant of the balance between intrahepatic tolerance and immunity. J Clin Invest 114, Fibbe, W.E., Claas, F.H., Van der Star-Dijkstra, W., Schaafsma, M.R., Meyboom, R.H. and Falkenburg, J.H. (1986). Agranulocytosis induced by propylthiouracil: evidence of a drug dependent antibody reacting with granulocytes, monocytes and haematopoietic progenitor cells. Br J Hematol 64, Ju, C., McCoy, J.P., Chung, C.J., Graf, M.L. and Pohl, L.R. (2003). Tolerogenic role of Kupffer cells in allergic reactions. Chem Res Toxicol 16, Lazarou, J., Pomeranz, B.H. and Corey, P.N. (1998). Incidence of adverse drug reactions in hospitalized patients: a meta-analysis of prospective studies. JAMA 279, Nelson, S.D. and Pearson, P.G. (1990). Covalent and noncovalent interactions in acute lethal cell injury caused by chemicals. Annu Rev Pharmacol Toxicol 30, Njoku, D., Laster, M.J., Gong, D.H., Eger, E.I. 2nd, Reed, G.F. and Martin, J.L. (1997). Biotransformation of halothane, enflurane, isoflurane, and desflurane to trifluoroacetylated liver proteins: association between protein acylation and hepatic injury. Anesth Analg 84,

6 Vol. 37, No. 1, January 2009 TOLEROGENIC FUNCTION OF KUPFFER CELLS 17 Satoh, H., Martin, B.M., Schulick, A.H., Christ, D.D., Kenna, J.G. and Pohl, L.R. (1989). Human anti-endoplasmic reticulum antibodies in sera of patients with halothane-induced hepatitis are directed against a trifluoroacetylated carboxylesterase. Proc Natl Acad Sci U S A 86, Uetrecht, J.P. (1999). New concepts in immunology relevant to idiosyncratic drug reactions: the danger hypothesis and innate immune system. Chem Res Toxicol 12, Vergani, D., Mieli-Vergani, G., Alberti, A., Neuberger, J., Eddleston, A.L., Davis, M. and Williams, R. (1980). Antibodies to the surface of halothanealtered rabbit hepatocytes in patients with severe halothane-associated hepatitis. N Engl J Med 303, von Greyerz, S., Zanni, M., Schnyder, B., Pichler, W.J. (1998). Presentation of non-peptide antigens, in particular drugs, to specific T cells. Clin Exp Allergy 28 (Suppl 4), Weiss, M.E. and Adkinson, N.F. (1988). Immediate hypersensitivity reactions to penicillin and related antibiotics. Clin Allergy 18, You, Q., Cheng, L., Kedl, R.M. and Ju, C. (2008). Mechanism of T cell tolerance induction by murine hepatic Kupffer cells. Hepatology 48, Zanni, M.P., von Greyerz, S., Schnyder, B., Brander, K.A., Frutig, K., Hari, Y., Valitutti, S. and Pichler, W.J. (1998). HLA-restricted, processing- and metabolism-independent pathway of drug recognition by human alpha beta T lymphocytes. J Clin Invest 102,