T-cell-mediated cytotoxicity against keratinocytes in sulfamethoxazol-induced skin reaction.

BACKGROUND: The incidence of skin rashes or erythema multiforme to sulfamethoxazole in exposed patients is about 3%. Among patients with acquired immunodeficiency syndrome the risk is approximately 10 times higher. The pathogenesis of these reactions and the reason for the increased frequency in HIV infections are not understood. OBJECTIVE: To investigate drug specific T-cell-mediated cytotoxicity in sulfamethoxazole- induced skin reactions. METHODS: Specific T-cell lines and T-cell clones generated from a donor who developed a skin rash to sulfamethoxazole were assessed with a standard 4 h 51Cr cytotoxicity assay in the presence or absence of soluble sulfamethoxazole. B lymphoblasts and keratinocytes with and without interferon gamma pretreatment were used as target cells. Selective blockers of FasL/Fas and perforin-mediated killing and immunostaining for perforin were used to evaluate the involvement of the different cytolytic pathways. RESULTS: CD4+ and CD8+ sulfamethoxazole specific T-cell clones showed a drug-specific and MHC-restricted cytotoxicity against autologous B lymphoblasts in the presence of soluble sulfamethoxazole. Keratinocytes, if pretreated with interferon gamma, were specifically killed predominantly by CD4+ T-cell clones. Specific T-cell clones of both CD4+ and CD8+ phenotype showed a strong immunoreactivity for perforin and the cytotoxicity was blocked by concanamycin A which suggests a perforin-mediated killing. CONCLUSION: Perforin-mediated killing of autologous keratinocytes in the presence of soluble sulfamethoxazole by drug-specific CD4+ lymphocytes may be a pathway for generalized drug-induced delayed skin reactions. The requirement of interferon gamma pretreatment of keratinocytes for efficient specific killing might explain the increased frequency of drug allergies in generalized viral infections like HIV, when interferon gamma levels are elevated.

Direct, MHC-dependent presentation of the drug sulfamethoxazole to human alphabeta T cell clones.

T cells can recognize small molecular compounds like drugs. It is thought that covalent binding to MHC bound peptides is required for such a hapten stimulation. Sulfamethoxazole, like most drugs, is not chemically reactive per se, but is thought to gain the ability to covalently bind to proteins after intracellular drug metabolism. The purpose of this study was to investigate how sulfamethoxazole is presented in an immunogenic form to sulfamethoxazole-specific T cell clones. The stimulation of four CD4(+) and two CD8(+) sulfamethoxazole-specific T cell clones by different antigen-presenting cells (APC) was measured both by proliferation and cytolytic assays. The MHC restriction was evaluated, first, by inhibition using anti-class I and anti-class II mAb, and second, by the degree of sulfamethoxazole-induced stimulation by partially matched APC. Fixation of APC was performed with glutaraldehyde 0.05%. The clones were specific for sulfamethoxazole without cross-reaction to other sulfonamides. The continuous presence of sulfamethoxazole was required during the assay period since pulsing of the APC was not sufficient to induce proliferation or cytotoxicity. Stimulation of clones required the addition of MHC compatible APC. The APC could be fixed without impairing their ability to present sulfamethoxazole. Sulfamethoxazole can be presented in an unstable, but MHC-restricted fashion, which is independent of processing. These features are best explained by a direct, noncovalent binding of sulfamethoxazole to the MHC-peptide complex.

High IL-5 production by human drug-specific T cell clones.

To analyze whether and how T cells are involved in drug allergies, we analyzed the drug-induced activation of T cell subsets, T cell receptor V-beta usage and cytokine secretion of T cells from the peripheral blood of drug-allergic individuals. The specificity of the T cells was demonstrated by specific restimulation of drug specific clones. We found that drugs which do not need to be metabolized to become immunogenic (haptens like penicillin G) can stimulate CD4+ and CD8+ T cells in vitro. The T cell response to penicillin can be oligoclonal (use of a certain T cell receptor Vbeta only) or polyclonal. Only polyclonal T cell lines were cross-reactive with other beta-lactam antibiotics. Sulfamethoxazole and lidocaine are thought to gain their ability to bind to proteins by intracellular drug metabolism. They were found to stimulate CD4+ and CD8+ T cells in vitro, and some reactive T cell lines were oligoclonal. The majority of lidocaine-specific clones secreted rather high amounts of IL-5 and IL-4 after PMA/ionomycin stimulations (Th2-like), but some CD4+ and all CD8+ clones had a Th1-like phenotype (high INF-gamma and TNF-alpha). The data clearly demonstrate the existence of drug-specific alphabeta+ T cells in the circulation of drug-allergic individuals and reveal a great heterogeneity of T-cell-mediated responses. Further studies are needed to correlate the type of T cell response to the clinical picture, which can be quite heterogeneous.

Characterization of lidocaine-specific T cells.

To investigate the cellular immune response to the drug lidocaine, we generated T cell lines and clones from the peripheral blood of four patients with proven allergy to lidocaine. The patients had contact dermatitis after topical application of lidocaine, and local swelling or generalized erythema exudativum multiforme after submucosal/subcutaneous injection of lidocaine. Two of three lidocaine-specific T cell lines were oligoclonal and one even became monoclonal, while the simultaneously analyzed immune response to tetanus toxoid was polyclonal. The lidocaine-specific T cell lines cross-reacted to mepivacaine, but not to other local anesthetics (bupivacaine, procaine, oxybuprocaine, and tetracaine). The majority of reactive T cells belonged to the CD4 cell lineage and were MHC class II restricted, but cloning also revealed some MHC class I-restricted CD8+ clones. A total of 2 of 56 lidocaine-specific T cell clones were CD4-CD8- and expressed TCR-gammadelta. The majority of 13 analyzed CD4 clones produced a rather polarized cytokine pattern, with a dominance of Th2-like cytokines showing a high IL-5 production. In addition, three CD4+ and all CD8+ (n = 7) clones secreted high IFN-gamma and low levels of IL-5/IL-4 (Th1-like). The data illustrate that a drug that sensitizes via the skin elicits a heterogeneous T cell response. The high IL-5 production and the participation of specific CD4+CD8+ and even gammadelta+ T cells appear to be distinguishing features of this hapten-specific immune response.

T cell reactions in patients showing adverse immune reactions to drugs.

OBJECTIVE AND DESIGN: To better understand how T cells react to small compounds, we investigated the in vitro T cell reactivity to drugs from drug allergic patients. MATERIAL AND SUBJECTS: Peripheral blood mononuclear cells (PBMC) of three drug allergic individuals were stimulated in vitro by different drugs. METHODS: Proliferation was assayed by 3H-thymidine incorporation. Upregulation of activation parameter on T cells was done by immunofluorescence and cytokine release determined via standard ELISA. RESULTS: Drugs can stimulate both CD4 and CD8 T-cell subsets. PenG-stimulated PBMC showed a heterogenous cytokine pattern and clones secreted high amounts of INF gamma. In contrast, sulfamethoxazole and lidocaine-stimulated PBMC secreted high levels of IL-5 and lidocaine-specific clones can be Th1 or Th2-like. CONCLUSION: Drug specific T cells play a pivotal role in drug hypersensitivity reactions, both by regulating the immune response and probably also as specific effector cells with different patterns of cytokine release.

Cross-reactivity of T cell lines and clones to beta-lactam antibiotics.

To clarify on a molecular level the specific T cell response to haptens like penicillin G, we generated T cell lines and clones from penicillin-allergic patients. Two types of beta-lactam reactivity of T cells could be delineated: one group of patients showed a rather restricted specificity, as the penicillin-elicited T cell lines generated from such donors proliferated only to the stimulating penicillin, but not to other beta-lactam antibiotics nor to cephalosporines, even if the side chain was identical. This indicates that the penicilloyl structure together with the side chain was recognized by these T cells. The second group comprised patients with more broadly reactive T cells, as they were restimulated by penicillin G as well as by related penicillins like amoxicillin or ampicillin, but not cephalosporines. This indicates that the penicilloyl structure, a common motif of penicillins, was important for T cell recognition. Clones generated from a broadly reactive patient confirmed this heterogeneity, as either monospecific or broadly specific T cell clones could be identified. This broad or very restricted pattern of T cell reactivity was reflected in the use of TCR Vbeta-chains: while the broadly reactive T cell lines showed a heterogenous TCR usage, the highly restricted T cell lines showed an up-regulation of one TCR Vbeta-chain. Thus, our data suggest that the outgrowth of T cells bearing a certain TCR Vbeta may be a sign of a limited cross-reactivity.

T cell recognition of penicillin G: structural features determining antigenic specificity.

Penicillin G (Pen G) and other beta-lactam antibiotics frequently induce allergic reactions constituting typical examples of human immune responses to haptens. In fact, penicillins represent a unique set of haptens with outstanding structural variability on the basis of an identical protein-reactive beta-lactam containing backbone. Although both cellular and humoral responses are involved in drug-induced allergies, little is known about the T cell reactivity to penicillins. To understand which structural features determine antigenic specificity, we isolated a panel of MHC-restricted, Pen G-reactive T cell clones from different penicillin-allergic patients and tested them for their capacity to proliferate in the presence of other penicillin derivatives. We found that the antigenic epitope consists of both the amide-linked side chain, which is different in every member of the penicillin family, as well as the thiazolidine ring common to all penicillin derivatives. We also demonstrated the presence of two different types of penicillin-specific T cells, one dependent, and the other independent of antigen processing by autologous antigen-presenting cells. Our data strongly suggest that penicillins form part of the epitopes contacting the antigen receptors of T cells.

Heterogeneous T cell responses to beta-lactam-modified self-structures are observed in penicillin-allergic individuals.

To investigate the role of T cells in drug allergy, we stimulated PBMC from penicillin-allergic patients with reactive penicillin G itself or penicillin G coupled with human serum albumin (BPO-HSA). T cell clones specific for penicillin G or BPO-HSA were established and their phenotype and reactivity to both forms of the beta-lactam were analyzed. T cell clones stimulated by penicillin G were CD4 and CD8 positive, whereas BPO-HSA stimulated the growth of CD4+ T cells. The penicillin G-specific clones were HLA class I or class II restricted and processing was not required as fixed APC could still present penicillin G. In contrast, BPO-HSA has to undergo processing to stimulate BPO-HSA-specific T cell clones. In addition to classical APC, activated MHC class II expressing T cells could also restimulate the penicillin G-specific clones, indicating that various cell types might serve as APC. Penicillin G and BPO-HSA-specific T cell clones produced a heterogeneous cytokine pattern as most clones produced high amounts of IL-2, IFN-gamma, TFN-alpha, and rather variable levels of IL-4 and IL-5. Since no Ag processing was required, penicillin G may stimulate T cells by binding directly to MHC molecules on the cell surface or to their embedded peptide. Alternatively, it may bind to soluble proteins like HSA, which are processed and subsequently presented in an immunogenic form. These different modes of presentation, which elicit a variety of immunological reactivities, may explain the great heterogeneity of the clinical pictures seen in penicillin allergy.

Activation of drug-specific CD4+ and CD8+ T cells in individuals allergic to sulfonamides, phenytoin, and carbamazepine.

To investigate how T cells are involved in hypersensitivity reactions to drugs that become immunogenic after metabolization, e.g., sulfonamides and antiepileptics, we analyzed in vitro the drug-induced activation of CD4+ and CD8+ T cell subsets, cytokine secretion, TCR V beta distribution, and proliferation of T cells from four drug-allergic individuals. In addition, the activation parameters CD25 and HLA-DR were analyzed in vivo on CD4+ and CD8+ T cells from five patients with acute drug allergies, some of them with anticonvulsant hypersensitivity syndrome with hepatitis. Our results show that, in vitro, drug-induced proliferation of PBMC from patients with allergy to sulfamethoxazole, phenytoin, or carbamazepine was specific and dose dependent. CD4+ as well as CD8+ T cells expressed elevated levels of CD25 and HLA-DR molecules after drug stimulation. Drug-activated lymphocytes secreted high amounts of IL-5 and normal or low levels of IL-2, IFN-gamma, IL-4, and TNF-alpha. An enhanced expansion of TCR V beta 17+ T cells 9 days after in vitro stimulation with sulfamethoxazole was observed in one patient with sulfamethoxazole allergy. The drug specificity of the in vitro-activated T cells was confirmed by generation of different sulfamethoxazole specific T cell lines and CD4+ and CD8+ T cell clones. T cell analysis of patients with acute drug allergy to carbamazepine, phenytoin, allopurinol, or paracetamol confirms the in vitro data, because all patients had activated CD4+ or CD8+ T cells in the circulation. Our data clearly show the involvement of drug-specific T cells in drug allergies.

The B7 adhesion molecule is expressed on activated human T cells: functional involvement in T-T cell interactions.

The B cell antigen B7 delivers a strong co-stimulatory signal for the activation of T cells by binding to its ligands CD28 and CTLA4. Here we demonstrate the surface expression of the B7 molecule on activated human T cells in vitro and under certain conditions in vivo and its functional importance in T-T cell interactions. B7 was detected by flow cytometry on antigen-specific CD4+ and allospecific CD8+ cloned T cells from different donors with anti-B7 monoclonal antibody (mAb) or a soluble CTLA4-C gamma 1 chimera molecule and by reverse transcription-polymerase chain reactions. The expression of B7 was up-regulated following restimulation of the T cell clones and peaked after 7-9 days. Moreover, we show that the B7 molecule on T cells is functionally involved in T-T cell interactions: mAb to CD28 and the CTLA4-Ig fusion protein could inhibit the proliferation of specific T cell clones in response to T cells as antigen-presenting cells (APC) or the proliferation of peripheral blood mononuclear cells in a primary allostimulation with activated T cells as stimulator cells. Finally, we found that B7 can be expressed on freshly isolated circulating T cells since in a preliminary study with a limited number of patients, B7 was present on a subset of CD3+ cells. B7 was expressed on activated T cells (CD4+ and CD8+) of certain human immunodeficiency virus (HIV)-infected individuals (0.5-20% B7+CD8+ cells) or some patients with autoimmune diseases whereas CD3+ cells of healthy individuals did not express B7. The coexpression of major histocompatibility complex class II molecules and B7 may be relevant for the capacity of activated T cells to function as APC. The expression of B7 on T cells in vivo in autoimmune diseases and in HIV infection may be important for a better understanding of these diseases.