Regular article
Oxazolone and diclofenac-induced popliteal lymph node assay reactions are attenuated in mice orally pretreated with the respective compound: potential role for the induction of regulatory mechanisms following enteric administration☆

https://doi.org/10.1016/S0041-008X(03)00091-7Get rights and content

Abstract

The murine popliteal lymph node assay (PLNA) was examined as a preclinical assay with the potential to identify low-molecular-weight compounds (LMWCs) that are likely to be associated with immune-mediated drug hypersensitivity reactions (IDHRs) in humans. We hypothesized that the contact sensitizer oxazolone (OX) would cause a strong PLN reaction in naive mice and that the PLN reaction would be attenuated in mice orally pretreated with OX due to the induction of oral tolerance. In naive mice, OX induced a strong PLN reaction and caused dose-dependent increases in PLN size, weight, cellularity, percentage of CD4+ PLN T cells, and percentage of PLN B cells, with a concomitant decrease in the percentage of CD8+ PLN T cells. Next, the PLNA was conducted in mice gavaged three times with either OX or vehicle alone (olive oil). Mice pretreated with OX had suppressed PLN reactions following the footpad injection of OX (decrease in PLN size, weight, and cellularity), which was associated with an increase in the percentage of PLN CD8+T cells. In contrast, oral pretreatment with OX had no observable effect on the PLN reaction induced following footpad injection of the irrelevant hapten dinitrochlorobenzene (DNCB). Adoptive transfer studies were conducted to examine the mechanism of PLN hyporesponsiveness. It was found that either (1) unfractionated splenocytes or (2) purified CD8+ splenocytes, but not (3) purified CD4+ splenocytes isolated from mice gavaged with OX adoptively transferred PLN suppression to naive BALB/c mice. Because OX is not a pharmaceutical, we also examined the NSAID diclofenac (DF) (Voltaren). Like OX, DF caused dose-dependent increases in PLN size, weight, and cellularity in naive mice. Furthermore, like OX, the diclofenac-induced PLN reaction was attenuated in mice that had been orally pretreated three times with DF. However, splenocytes from mice orally treated with DF were not able to adoptively transfer PLN hyporesponsiveness. Collectively, these observations demonstrate that both OX and DF are potent immunostimulators in the PLNA. As importantly, these results demonstrate that the immunostimulating potential of OX and DF in the PLNA is significantly decreased in mice orally exposed to the respective drug, possibly due to the presence of a cellular mechanism of oral tolerance. For OX, the mechanism appears to involve, in part, CD8+ T cells, whereas the mechanism(s) associated with PLN hyporesponsiveness using DF remain to be defined.

Introduction

Immune-mediated drug hypersensitivity reactions (IDHRs) to low-molecular weight compounds (LMWCs) are common in modern medicine Craig and Mende 1999, deShazo and Kemp 1997. Based on postmarketing drug surveillance, IDHRs account for approximately one-sixth of all drug-induced adverse reactions and as many as 5% of adults may be allergic to one or more drugs Lebrec et al 2001, DeSwarte 1993, Borda et al 1968. These reactions are a significant cause of patient morbidity and mortality and can preclude effective treatment if the drug has to be withdrawn Park et al 1998, Griem et al 1998. In addition to having a negative impact on patients, IDHRs can significantly hamper future drug development (Adkinson et al., 2002). Unfortunately, with the exception of contact hypersensitivity following topical administration of certain LMWCs, IDHRs are difficult to predict using in vitro or animal models (Hastings, 2001). Furthermore, due to their idiosyncratic nature, IDHRs are often not detected in clinical trials (Hastings, 2001). Thus, in general, IDHRs are only detected with postmarketing drug surveillance where their occurrence often leads to the removal of the drug from the market, which has been described as corresponding to the collapse a few meters away from the end of a marathon run (Pichler, 2001). Clearly, current challenges for immunotoxicologists with respect to IDHRs are (1) to develop nonclinical models that can identify immunotoxicological hazards early in drug development, (2) to use these models to better understand the mechanisms associated with the various types of IDHRs, and (3) to use this information to estimate the risk of an IDHR for a given drug candidate.

At present, the only systematic approach proposed to have any value in identifying LMWCs with the potential to cause IDHRs in humans is the rodent popliteal lymph node assay (PLNA) Pieters 2001, Kimber 2001, Gleichmann 1981. In its basic form, LMWCs are injected subcutaneously without adjuvant into the hind footpad of naive rodents, and at specified time-points thereafter, usually 6–8 days, the size, weight, and/or cellularity of the draining PLN are assessed for both treatment and control animals (Pieters, 2001). A significant increase in any PLN endpoint in drug-treated animals compared to vehicle-treated controls suggests the test article had inherent immunostimulating potential and may be identified as an immunotoxicological hazard. Since 1981, the PLNA has been used to assess the immunostimulating potential of >120 test compounds and, in general, only those compounds that are associated with IDHRs in humans cause positive PLN reactions in rodents Pieters 2001, Goebel et al 1996, Bloksma et al 1995. However, despite this preliminary success, the PLNA does have several significant limitations. First, how LMWCs cause PLN enlargement and the mechanism that leads to LMWC-specific sensitization following a single footpad injection remain unknown (Goebel et al., 1996). Second, the route of administration, i.e., footpad injection, is not physiologically relevant, as most drugs are given orally (Goebel et al., 1996). Third, some LMWCs may be false positives in the PLNA Kimber 2001, Goebel et al 1996.

With respect to the latter limitation, it has been shown that contact sensitizers can induce positive PLN reactions following footpad injection (De Bakker et al., 1990). However, the immunostimulating potential of these compounds in the PLNA may not necessarily translate to a high risk for systemic IDHR because it is widely accepted that contact sensitizers induce oral tolerance following oral administration Desvignes et al 2000, Galliaerde et al 1995, Gautam and Battisto 1985, Lowney 1973. Oral tolerance has been demonstrated using a variety of antigens (protein, peptide, as well as LMWCs such as contact sensitizers and transition metals in models of contact hypersensitivity) and may play a significant role in preventing IDHRs in humans. Indeed, it has long been hypothesized that the failure to induce oral tolerance, as opposed to an inherent primary overimmunoreactivity of drugs, may be responsible for the manifestation of an allergic drug reaction (Asherson et al., 1977). However, the induction of oral tolerance to a low-molecular-weight (LMW) pharmaceutical has yet to be demonstrated, principally due to a lack of appropriate animal models.

The objectives of the present study were to examine the immunostimulating potential of the contact sensitizer oxazolone in the PLNA using either naive BALB/c mice or mice that had been orally pretreated with oxazolone. We hypothesize that oxazolone is a potent immunostimulating LMWC in the PLNA when conducted in naive mice, but its potential to stimulate the immune system following a footpad injection is significantly reduced in mice that have been orally exposed to the compound due to the presence of an oxazolone-specific cell-mediated mechanism of oral tolerance. Oral tolerance to oxazolone has been demonstrated using ear swelling tests (Asherson et al., 1979), but it is unknown whether the PLNA can be used to detect and study the same phenomenon. Although using the PLNA to examine the immunostimulating potential of oxazolone in the context of oral tolerance would further advance our understanding of how LMWCs stimulate the immune system, particularly following a single footpad injection, a limitation of the work is that oxazolone is not a human medicinal product. Therefore, we also conducted these studies using a human LMW pharmaceutical. Here, we chose to study diclofenac (Voltaren), a phenylacetic acid NSAID, for several reasons. First, the toxicities associated with diclofenac are well described; it has been on the market for almost 30 years and has been prescribed in >120 countries (Skoutakis et al., 1988). Second, it is known that diclofenac, like all NSAIDs, is associated with certain IDHRs in humans such as IgE-mediated anaphylaxis, idiosyncratic hepatotoxicity associated with rash and eosinophilia, as well as antibody-dependent hemolytic anemia Salama et al 1996, Van der Klauw et al 1996, Boelsterli et al 1995. Third, diclofenac induces a potent T-cell-mediated reaction in the direct PLNA and activates IL-4-secreting T-helper-2 cells in the TNP-OVA reporter antigen PLNA that function to selectively produce IgE and IgG1 (compared with IgG2a and IgG2b) Gutting et al 1999, Gutting et al 2002a. However, we also hypothesized that an important factor in the development of diclofenac-induced IDHR would be the induction of oral tolerance.

Our experimental design consisted of conducting a direct PLNA in naive BALB/c mice or in mice that had been orally pretreated with the respective LMWC. The endpoints used were PLN size, weight, and cellularity, as well as PLN phenotype analysis using flow cytometry. A final set of studies used adoptive transfer methods to examine potential mechanisms of immunostimulation caused by oxazolone or diclofenac.

Section snippets

Mice

BALB/c mice were obtained from Charles River Laboratories (Wilmington, MA). Upon arrival wild-type mice were weighed (≅20 g), randomly assigned to their respective treatment groups, and housed approximately 1 week in polycarbonate cages maintained at a temperature of 70 ± 2°F, relative humidity of 50 ± 5%, and under a 12-hour light/dark cycle. They had free access to drinking water that was purified by a reverse osmosis system and to a commercial laboratory animal food. All experiments were

Oxazolone is immunostimulating in the PLNA in naive BALB/c mice

To determine the immunostimulating potential of oxazolone in the PLNA, naive BALB/c mice received an injection of oxazolone (0.50 mg in 10 μl DMSO) or vehicle alone (10 μl DMSO), followed by an assessment of PLN size, weight, and cellularity 7 days later. Oxazolone was immunostimulating (Fig. 1), as it caused significant increases in PLN size (diameter, mm), weight (mg), and cellularity (×106) (Fig. 1B). PLNs from a representative treatment and control mice are shown in Fig. 1A. The dose of

Discussion

The rodent PLNA was developed in 1981 as a tool to study the immunostimulating potential of diphenylhydantoin (Gleichmann, 1981). To date the PLNA has been used to study the immunostimulating potential of >120 compounds and the results suggest the PLNA may be useful as a preclinical test system for hazard identification Hastings 2001, Pieters 2001. However, despite the preliminary success of the rodent PLNA, the assay does have significant limitations. For example, the PLNA examines the

References (50)

  • K.L Hastings

    Pre-clinical methods for detecting the hypersensitivity potential of pharmaceuticalsregulatory considerations

    Toxicology

    (2001)
  • I Kimber

    The local lymph node assay and potential application to the identification of drug allergens

    Toxicology

    (2001)
  • H Lebrec et al.

    Th(1)/Th(2) responses to drugs

    Toxicology

    (2001)
  • E.D Lowney

    A single step procedure for inducing partial tolerance of DNCB in human subjects

    J. Invest. Dermatol.

    (1974)
  • W.J Pichler

    Predictive drug allergy testingan alternative view point

    Toxicology

    (2001)
  • R Pieters

    The popliteal lymph node assaya tool for predicting drug allergies

    Toxicology

    (2001)
  • G.L Asherson et al.

    Contact sensitivity and the DNA response in mice to high and low doses of oxazolonelow dose unresponsiveness following painting and feeding and its prevention by pretreatment with cyclophosphamide

    Immunology

    (1979)
  • A Barbaud et al.

    Bufexamac and diclofenacfrequency of contact sensitization in the absence of cross-reactivity

    Contact Dermatitis

    (1998)
  • N Bloksma et al.

    Predictive immunological test systemssuitability of the popliteal lymph node assay in mice and rats

    Crit. Rev. Toxicol.

    (1995)
  • U.A Boelsterli et al.

    Idiosyncratic liver toxicity of nonsteroidal antiinflammatory drugsmolecular mechanisms and pathology

    Crit. Rev. Toxicol.

    (1995)
  • I.T Borda et al.

    Assessment of adverse reactions within a drug surveillance program

    JAMA

    (1968)
  • Y Chen et al.

    Induction of oral tolerance to myelin basic protein in Cd8-depleted miceboth CD4+ and CD8+ cells mediate active suppression

    J. Immunol.

    (1995)
  • T.J Craig et al.

    Common allergic and allergic-like reactions to medicationswhen the cure becomes the curse

    Drug Allergies

    (1999)
  • R.J Dearman et al.

    Inducible interleukin-4-secreting cells provoked in mice during chemical sensitization

    Immunology

    (1994)
  • J.M De Bakker et al.

    Kinetics and morphology of chemically induced popliteal lymph node reactions compared with antigen-, mitogen-, and graft-versus-host-reaction-induced responses

    Virchows Arch. B. Cell Pathol. Incl. Mol. Pathol.

    (1990)
  • Cited by (0)

    This work was supported in part by National Institutes of Health AR40072 and AR44076, the Arthritis Foundation and the Lupus Foundation of America and their Connecticut chapters, and the S.L.E. Foundation, Inc. and a Kirkland Scholar’s Award (J.C.).

    1

    B.W.G. is a Post-Doctoral Assistant supported by financial contributions from Pfizer Global Research and Development, Groton, Connecticut, 06340.

    View full text