- Email: [email protected]
Coupling of Contact Sensitizers to Thiol Groups is a Key Event for the Activation of Monocytes and Monocyte-Derived Dendritic Cells
ORIGINAL ARTICLE
Coupling of Contact Sensitizers to Thiol Groups is a Key Event for the Activation of Monocytes and Monocyte-Derived Dendritic Cells Detlef Becker, Elke Valk, Sabine Zahn, Pia Brand, and Jˇrgen Knop Department of Dermatology, University of Mainz, Germany
Strong contact sensitizers are able to induce distinct signal transduction mechanisms in antigen-presenting cells by coupling to cell proteins. The predominant target structures of haptens are thought to be thiol and amino groups in cysteine and lysine residues.We studied whether coupling of small reactive chemicals to thiol or amino groups might be responsible for the activation of monocytes and mature monocyte-derived dendritic cells. Human peripheral blood mononuclear cells were stimulated in vitro with subtoxic concentrations of the strong haptens 5-chloro-2-methylisothiazolinone plus 2-methylisothiazolinone and 2, 4, 6-trinitrochlorobenzene, the thiol-reactive reagents N-hydroxymaleimide and N-ethylmaleimide, as well as the amino-reactive compounds sulfosuccinimidyl acetate and 2-iminothiolane. Flow cytometric quanti¢cation of tyrosine phosphorylation in CD14 þ monocytes showed that 5-chloro-2-methylisothiazolinone plus 2-methylisothiazolinone, 2, 4, 6-trinitrochlorobenzene, N-hydroxymaleimide, and N-ethylmaleimide but not sulfosuccinimidyl acetate and 2-iminothiolane strongly induced this process. Tyrosine phosphorylation induced by 5-
chloro-2-methylisothiazolinone plus 2-methylisothiazolinone and 2, 4, 6-trinitrochlorobenzene was completely prevented in the presence of cysteine but not lysine, suggesting a competitive mechanism between cysteine and sulfhydryl groups of cell proteins. Using the mouse ear swelling test N-hydroxymaleimide could be classi¢ed as a signi¢cant contact allergen in comparison to 2, 4, 6-trinitrochlorobenzene, whereas no sensitizing potential became apparent for sulfosuccinimidyl acetate and 2-iminothiolane. Western blot analysis on monocytes and mature monocyte-derived dendritic cells con¢rmed the £ow cytometric data for tyrosine phosphorylation and demonstrated a selective capacity of haptens and thiol-reactive compounds to activate ERK1/2 mitogen-activated protein kinase. Our data show that strong a⁄nity of a small reactive chemical toward thiol groups is important for the activation of monocytes and monocyte-derived dendritic cells and can support the process of sensitization. Key words: thiol groups/amino groups/signal transduction/hapten. J Invest Dermatol 120:233 ^238, 2003
T
(Pichowski et al, 2001). We were able to demonstrate a vigorous phosphorylation of tyrosine residues in di¡erent populations of human MHC class II positive cells following stimulation with structurally nonrelated contact sensitizers (Kˇhn et al, 1998) and con¢rmed the relevance of these ¢ndings for murine Langerhans cells (Neisius et al, 1999). Recently, Arrighi et al (2001) reported on the activation of the p38 mitogen-activated protein kinase (MAPK) in human mo-DC by dinitro£uorobenzene and nickel sulfate. Using human monocytes as model cells we found similar results and especially a remarkable translocation of p38 MAPK from the cytoplasm to membrane-associated compartments (Brand et al, 2002). At present, the structures to which contact sensitizers have to bind for initiation of these signaling cascades are unknown. From a theoretical point of view haptens are thought to bind predominantly to thiol and amino groups in cysteine and lysine residues of cell proteins (Roberts and Lepoittevin, 1998). A good reactivity to amino as well as thiol groups has been demonstrated for 2, 4, 6 -trinitrochlorobenzene (TNCB) and other experimental contact sensitizers in vitro (Parker et al, 1983). For mercury compounds (Rosenspire et al, 1998; Santucci et al, 1998; 1999) and 5-chloro-2methylisothiazolinone plus 2-methylisothiazolinone (MCI/MI) (Gruvberger and Bruze, 1998) a direct interaction between hapten and thiol groups has been shown or at least supposed.
here is plenty of evidence for the capacity of contact sensitizers to activate epidermal Langerhans cells. Although some events might be induced indirectly by cytokines derived from other epidermal cell populations several studies demonstrated a direct in£uence of haptens on cellular and molecular changes in dendritic cells (DC) and other major histocompatibility complex (MHC) class II positive cell populations. This includes increased receptor-mediated endocytosis of membrane molecules by Langerhans cells (Becker et al, 1992; 1995), cytokine production and upregulation of membrane markers in monocyte-derived dendritic cells (mo-DC) (Aiba et al, 1997; Rougier et al, 2000; Tuschl et al, 2000), and induction of mRNA for interleukin-1b (IL-1b) Manuscript received October 31, 2001; revised October 10, 2002; accepted for publication October 14, 2002 Reprint requests to: Detlef Becker, M.D., UniversitHts-Hautklinik, Langenbeckstr. 1, D-55131 Mainz, Germany; Email: [email protected]. uni-mainz.de Abbreviations: MAPK, mitogen-activated protein kinase; MCI/MI, 5chloro-2-methylisothiazolinone plus 2-methylisothiazolinone; mo-DC, monocyte-derived dendritic cells; NEM, N-ethylmaleimide; NHM, Nhydroxymaleimide; PBMC, peripheral blood mononuclear cells; sulfoNHS, sulfosuccinimidyl acetate; TNCB, 2, 4, 6 -trinitrochlorobenzene.
0022-202X/03/$15.00 . Copyright r 2003 by The Society for Investigative Dermatology, Inc. 233
234
BECKER ET AL
From experiments on the capacity of thiol antioxidants to block the activation of cells by contact sensitizers (Bruchhausen et al, submitted) we concluded that coupling to thiol groups might be very important for this process. In this paper we studied whether de¢ned thiol- or amino-reactive reagents would be able to resemble the activation process induced by contact sensitizers. MATERIALS AND METHODS Culture of mo-DC Human mo-DC were prepared from peripheral blood mononuclear cells (PBMC) by short-term adherence in the modi¢cation described by Jonuleit et al (1997). Brie£y, PBMC were isolated from leukocyte-enriched bu¡y coats (Transfusion Center, Mainz, Germany) by Ficoll-Paque 1.077 (Biochrom, Berlin, Germany) and incubated (45 min, 371C, 5% CO2) in six-well plates (Greiner, Frickenhausen, Karlstahe, Germany) using RPMI 1640 medium (Life Technologies) supplemented with 3% autologous plasma. Non-adherent cells were removed by washing gently with phosphate-bu¡ered saline (PBS). The remaining cells were cultured in RPMI with 5% autologous serum, 1000 U per ml IL- 4 (Strathmann Biotech, Hannover, Germany), and 800 U per ml granulocyte-macrophage colony stimulating factor (GM-CSF) (Leucomax, Sandoz, Basel, Switzerland). Cells were fed every other day with RPMI, 5% autologous serum, IL- 4 (1000 U per ml), and GM-CSF (800 U per ml). On day 7 the medium was supplemented with 10 ng per ml IL-1b (Strathmann Biotech), 10 ng per ml tumor necrosis factor a (Strathmann Biotech), and 1 mg per ml prostaglandin E2 (Minprostin, Pharmacia & Upjohn, Heppenheim, Germany). Mature DC were harvested on day 9. The phenotype of DC was determined by £ow cytometry as 70% CD83 þ , 4 80% CD86 þ , and 4 90% MHCII þ ; viability routinely exceeded 90% as determined by propidium iodide staining. Stimulation of PBMC with chemicals Human PBMC (106 per ml) were stimulated in vitro with subtoxic concentrations of the strong haptens MCI/MI (Hermal, Reinbeck, Germany) and TNCB (ICN, Costa Mesa, CA), the thiol-reactive reagents N-hydroxymaleimide (NHM), Nethylmaleimide (NEM), iodoactetamide, and sodium tetrathionate (all obtained from Sigma, St. Louis, MO), as well as the amino-reactive compounds sulfosuccinimidyl acetate (sulfo-NHS, Pierce, Rockford, IL) and 2-iminothiolane (Pierce) for 15 min at 371C in RPMI medium. If indicated 10 mM cysteine (Sigma) or 10 mM lysine (Roth, Karlsruhe, Germany) was present during incubation. Stock solutions of TNCB were prepared in dimethylsulfoxide; the other chemicals were dissolved in culture medium immediately before use. The ¢nal concentration of dimethylsulfoxide never exceeded 0.1%. Flow cytometric analysis To perform intracellular £uorescenceactivated cell sorter (FACS) analysis, PBMC were permeabilized immediately after stimulation with PBS/1% fetal bovine serum (FBS)/ 0.25% saponine (Roth) and stained by phycoerythrin-conjugated antiCD14 (10 mg per ml; Immunotech, Marseille, France) in combination with £uorescein isothiocyanate conjugated antiphosphotyrosine (PY20FITC; 4 mg per ml; Leinco Technologies, St. Louis, MO) dissolved in PBS/1% FBS/0.25% saponine for 20 min at 41C. Finally, samples were washed twice in cold washing bu¡er (PBS/1% FBS/0.25% saponine). Fluorescence intensities in channel 1 (phosphotyrosine) were recorded for cells gated in channel 2 (CD14 þ ) using a FACScalibur £ow cytometer (Becton Dickinson, Heidelberg, Germany) equipped with the CellQuest software. Standard calibration procedures were used following the recommendation of Becton Dickinson. The level of signi¢cance for a di¡erence between medium treated and stimulated cells was calculated using a paired Student t test. Measurement of IL-1b production by enzyme-linked immunosorbent assay (ELISA) PBMC were resuspended in Iscove’s medium/ 1% autologous plasma and seeded onto a 12-well plate (8 106 cells per well). After 45 min the nonadherent cells were removed by washing with warm PBS. After stimulation with MCI/MI (0.4 mg per ml), NHM (25 mg per ml), 2-iminothiolane (400 mg per ml), and sulfo-NHS (2 mg per ml) for 15 min, the medium was aspirated and replaced by 1.5 ml fresh medium containing 1% autologous plasma. Supernatants were collected after 24 h and assessed for IL-1b production by ELISA according to the recommendations of the distributor of the pair of antibodies employed (R&D Systems, Wiesbaden, Germany). Brie£y, Maxisorp Nunc Immunoplates (Nunc, Wiesbaden, Germany) were coated overnight at 41C with 2 mg per ml capture antibody diluted in 0.1 M NaHCO3, pH 8.2. After washing with PBS/0.05% Tween-20, the plates
THE JOURNAL OF INVESTIGATIVE DERMATOLOGY
were blocked for 2 h at 371C with PBS/1% bovine serum albumin. Serial dilutions of the provided standards (R&D Systems) and collected supernatants were incubated overnight at 41C. The next day the plates were washed and incubated at 371C for 1 h with 10 ng per ml biotinylated detection antibody, diluted in PBS/0.1% bovine serum albumin. After washing, 2.5 mg per ml peroxidase-conjugated avidin (Sigma) was added for 30 min. The plates were then developed with ABTS and read in a microplate reader model 450 (Bio-Rad) at 405 nm. IL-1b concentration was calculated according to the standard curves with a detection limit of 16 pg per ml. IL-1b values after stimulation were calculated as percentage of the corresponding medium control. The level of signi¢cance for a positive di¡erence between medium treated and stimulated cells was calculated using a paired Student t test. Preparation of cell extracts, sodium dodecyl sulfate (SDS) polyacrylamide gel electrophoresis, and Western blot 5 106 enriched monocytes or 1 106 mature mo-DC were stimulated as indicated in culture medium or PBS. Cells were resuspended in 50 ml icecold lysis bu¡er (50 mM HEPES, pH 7.4, 150 mM NaCl, 1.5 mM MgCl2, 1 mM ethylenediamine tetraacetic acid, 10% glycerol, 1% Triton X-100, 1 mM phenylmethylsulfonyl £uoride, 10 mg per ml aprotinin, and 1 mM sodium orthovanadate) and incubated on ice for 30 min. Insoluble material was spun down (10 min, 14,000 rpm, 41C) and clear supernatants were stored at ^701C. Protein content of the lysates was determined by BCA protein assay (Pierce). Cell lysates were mixed with loading bu¡er (Roti Load, Roth), heated for 5 min at 961C, and resolved on 8% SDS polyacrylamide gels (100 mg per lane) using standard procedures. Proteins were transferred to PVDF membranes (Millipore, Bedford, MA) by semidry blotting (Bio-Rad, Hercules, CA) in blotting bu¡er (Roti-Blot A, Roti-Blot K, Roth). Membranes were blocked for 1 h in blocking bu¡er (1 PBS, 0,1% Tween-20, 5% nonfat dry milk powder) and incubated with the following primary antibodies: mouse antiphosphotyrosine (clone 4G10; 0.5 mg per ml; Upstate Biotechnology, Lake Placid, NY), rabbit anti-p-ERK1/2 or rabbit anti-p-p38 (dilution 1:1000, New England Biolabs, Beverly, MA). The src-kinase Lyn was detected as control for equal protein loading (anti-Lyn; Santa Cruz Biotechnology, Santa Cruz, CA; dilution 1:1000). After washing, antimouse horseradish peroxidase (0.2 mg per ml, Dianova) or antirabbit horseradish peroxidase (dilution 1: 2000, New England Biolabs) was added. All antibodies were diluted in blocking bu¡er. Blots were developed using chemiluminescence (ECL Plus, Amersham, Buckinghamshire, U.K.). For sequential detection of di¡erent antigens membranes were stripped (40 min at 501C in 62.5 mM Tris^HCl, 2% SDS, 100 mM b-mercaptoethanol, pH 6.7) and reprobed. Mouse ear swelling test Both sides of the left ears of 8^12-wk-old Balb/c mice of both sexes were painted with 15 ml each of 120 mM TNCB, NHM, sulfo-NHS, and 2-iminothiolane on days 1, 2, and 3. TNCB was dissolved in ethanol, the other compound in ethanol/aqua dest. (50%/50%). At day 6, 40 mM of the respective compounds were applied on both sides of the right ears. Ear thickness was recorded before and 24 h after challenge using an engineer’s micrometer. Control animals received vehicle alone, the sensitizing or challenging dose at day 1, and were measured at day 2 to monitor any irritant e¡ects. For this set of experiments the level of signi¢cance for a di¡erence between vehicle control and applied chemical was calculated using a paired Student t test. All animal experiments were performed according to the guidelines and with the permission of the government of Rheinland-Pfalz.
RESULTS Contact sensitizers and thiol-group-reactive compounds but not amino-group-reactive chemicals induce tyrosine phosphorylation in monocytes MCI/MI and TNCB as well as the thiol-group-speci¢c reagents NHM (Freed et al, 1986) and NEM (Majima et al, 1993) strongly induced tyrosine phosphorylation in CD14 þ monocytes (Fig 1). In contrast sulfoNHS and 2-iminothiolane, two common amino-group-speci¢c reagents (Donovan and Jennings, 1986; Meunier et al, 1999) failed to induce tyrosine phosphorylation (Fig 1). A clear dose-response relationship became apparent for the activating compounds whereas neither sulfo-NHS nor 2-iminothiolane showed any reactivity even in the highest concentration used (Fig 2). To exclude that the induction of tyrosine phosphorylation might be a speci¢c feature of maleimides, control experiments were performed using the structurally nonrelated thiol-reactive
VOL. 120, NO. 2 FEBRUARY 2003
Figure 1. Induction of tyrosine phosphorylation by strong contact sensitizers and thiol-reactive reagents but not by amino-reactive compounds. PBMC were incubated for 15 min with the strong contact sensitizers MCI/MI and TNCB, the thiol-reactive reagents NHM and NEM, and the amino-reactive compounds 2-iminothiolane and sulfoNHS and analyzed for the amount of tyrosine-phosphorylated proteins using £ow cytometric analysis. Histograms for the binding of phosphotyrosine-speci¢c antibodies to CD14+ monocytes are shown. Concentrations used for stimulation and the mean total £uorescence intensities are indicated.
compounds iodoacetamide (Crow et al, 2001) and sodium tetrathionate (Mammoto et al, 1997). Both were able to induce tyrosine phosphorylation in CD14 þ monocytes (iodoacetamide 370 mg per ml, relative mean increase 467%7131%, n ¼ 15, p ¼ 0.02; sodium tetrathionate 10 mg per ml, relative mean increase 254%794.5%, n ¼ 6, p ¼ 0.01). Toxic e¡ects were excluded in pilot experiments using propidium iodide as marker for cell damage (data not shown); all compounds were studied using the highest but still subtoxic range of concentration. Cysteine but not lysine blocks tyrosine phosphorylation induced by MCI/MI and TNCB The capacity of cysteine and lysine to modulate the degree of tyrosine phosphorylation in CD14 þ monocytes under stimulation with MCI/MI and TNCB was explored. Cysteine but not lysine completely blocked any tyrosine phosphorylation, suggesting a competitive in£uence of free thiol groups but not amino groups on the mechanism responsible for the activation of tyrosine kinases by both contact sensitizers (Fig 3). NHM and MCI/MI but not amino-group-reactive compounds induce IL-1b secretion by monocytes Human monocytes were enriched by adhesion and stimulated with MCI/MI, NHM, 2-iminothiolane, and sulfo-NHS before subculture for 24 h. IL-1b was measured in culture supernatants
ACTIVATION BY COUPLING TO THIOL GROUPS
235
Figure 2. Dose-response relationship for the induction of tyrosine phosphorylation by contact sensitizers NHM and NEM. CD14+ monocytes stimulated with the compounds and concentrations indicated were analyzed by £ow cytometry. Means7SEM of the total £uorescence intensities determined in ¢ve independent experiments are shown. A doserelated signi¢cant increase can be noticed for MCI/MI,TNCB, NHM, and NEM (p o 0.01 is indicated by an asterisk).
Figure 3. Tyrosine phosphorylation induced by MCI/MI and TNCB is blocked in the presence of cysteine but not lysine. PBMC were stimulated with MCI/MI and TNCB in the presence or absence of 10 mM cysteine or lysine. Selective analysis of tyrosine phosphorylation in CD14+ monocytes was performed by £ow cytometry. Means7SEM of the total £uorescence intensities determined in six independent experiments are shown. An asterisk indicates signi¢cance (p o 0.01) for the di¡erence between stimulated cells and the medium control.
by ELISA. In comparison to the basal production of unstimulated cells (mean 30.5 pg per ml, SEM 19.5 pg per ml) a signi¢cant liberation of IL-1b was found for MCI/MI and NHM (p o 0.05) but not for the two amino-reactive compounds (Fig 4). Therefore NHM resembles the e¡ect of a strong contact sensitizer by induction of this important cytokine for the sensitization phase of allergic contact dermatitis (Enk et al, 1993).
236
BECKER ET AL
Figure 4. Monocytes secret IL-1b following stimulation with MCI/ MI and NHM but not amino-reactive compounds. Enriched monocytes were stimulated as indicated and cultured for 24 h. IL-1b was determined in culture supernatants using an ELISA technique. Means7SEM of four independent experiments are shown (p o 0.05 is indicated by an asterisk).
Figure 5. Induction of contact hypersensitivity in Balb/c mice by TNCB and NHM but not sulfo-NHS and 2-iminothiolane. The capacity of NHM, sulfo-NHS, and 2-iminothiolane to induce contact hypersensitivity in a modi¢ed mouse ear swelling test was compared to TNCB as control for a potent contact sensitizer. Sensitization was performed with 120 mM and elicitation with 40 mM of each compound. The di¡erence in ear thickness between the solvent control and treated animals from three independent experiments with two animals in each group is indicated.
NHM but not amino-group-reactive compounds induce contact hypersensitivity in mice The capacity of NHM, sulfo-NHS, and 2-iminothiolane to sensitize Balb/c mice was compared to TNCB by testing equimolar concentrations in a modi¢ed mouse ear swelling test (Garrigue et al, 1994). All animals were sensitized by the strong experimental hapten TNCB using a moderate concentration (Enk et al, 1994; Steinbrink et al, 1996). Whereas at least three of six animals responded signi¢cantly to NHM, no evidence for a sensitizing potential of the other two compounds was found (Fig 5). Experiments performed to monitor any irritant potential showed that none of the compounds was irritant when used at 40 mM (elicitation). The sensitizing dose (120 mM) of TNCB was found to be moderately irritant (p o 0.01) whereas the three thiol- or amino-speci¢c reagents were well tolerated (Fig 6). NHM and NEM induce a contact-sensitizer-like pattern of tyrosine-phosphorylated proteins and activate MAPK Biochemical analysis on enriched monocytes stimulated with NHM revealed augmented phosphorylation of several protein bands in comparison to medium treated cells and resembled the
THE JOURNAL OF INVESTIGATIVE DERMATOLOGY
Figure 6. Only the sensitizing dose of TNCB is moderately irritant. Ear thickness measured 24 h after application of the sensitizing and elicitating doses of TNCB, NHM, sulfo-NHS, and 2-iminothiolane is shown in comparison to the solvent control. Means7SEM of three independent experiments with two animals in each group are shown (p o 0.01 is indicated by an asterisk).
Figure 7. Contact sensitizers and thiol-reactive but not amino-reactive reagents induce phosphorylation of tyrosine residues and MAPK. (a) Western blot analysis of enriched monocytes stimulated with TNCB and NHM reveals a similar pattern of tyrosine-phosphorylated proteins. Enriched monocytes were stimulated with 0.8 mg per ml TNCB or 25 mg per ml NHM. The pattern of tyrosine-phosphorylated proteins is shown in comparison to the medium control. The total £uorescence intensities for tyrosine-phosphorylated proteins of these cells as assessed by £ow cytometry are as follows: medium 4.4; TNCB 40.3; NHM 34.8. The blot is representative for three di¡erent experiments with comparable results. (b) Mature mo-DC were stimulated with 2 mg per ml sulfo-NHS, 400 mg per ml 2-iminothiolane, 25 mg per ml NHM, 25 mg per ml NEM, 0.8 mg per ml MCI/MI, and 0.8 mg per ml TNCB. Tyrosine-phosphorylated proteins, phosphorylated ERK1/2 MAPK, and p38 MAPK are shown in comparison to the medium control. The blot is representative for three di¡erent experiments with comparable results.
pattern seen for the strong contact sensitizer TNCB (Fig 7A). As reported before (Kˇhn et al, 1998) this correlated very well with the increase in total tyrosine phosphorylation as monitored by FACS analysis. To obtain data for professional antigen presenting cells £ow cytometric analysis on mature mo-DC was set up and demonstrated a similar response upon stimulation with contact sensitizers and thiol-reactive compounds as noticed for monocytes (data not shown). A detailed biochemical analysis on mature mo-DC stimulated with sulfo-NHS, 2-iminothiolane, NHM, NEM, MCI/MI, and TNCB revealed augmented
VOL. 120, NO. 2 FEBRUARY 2003
phosphorylation of several protein bands in comparison to medium treated cells for the contact sensitizers and the thiolreactive chemicals but not the amino-reactive compounds. The phosphorylation pattern induced by NHM and NEM resembled the pattern induced by haptens with regard to the molecular weight of several bands (Fig 7B). Only haptens and the thiolreactive compounds were able to induce phosphorylation of ERK1/2 MAPK whereas activation of p38 MAPK was induced by haptens but not NHM or NEM. Following stimulation with 2-iminothiolane some low but reproducible increase in phosphop38 became apparent. Expression of the src kinase Lyn was monitored as control for protein loading and was found to be comparable in all probes (data not shown). DISCUSSION
In contrast to protein allergens contact sensitizers are characterized as reactive compounds that must bind to skin proteins in order to be recognized by the immune system. Additionally, the coupling of strong contact sensitizers leads to the induction of signaling events in the cell. This could take place by activation of the corresponding receptors, kinases, or other intermediate components of de¢ned signaling cascades. The possible chemical coupling mechanisms of haptens to proteins have been described in detail (Roberts and Lepoittevin, 1998). Essentially, amino groups as well as thiol groups should be involved. Although some limited data have been presented for mercury compounds, MCI/MI, and strong experimental contact sensitizers, systematic studies on the preferential target structures of certain haptens or hapten groups and their importance for cell activation are missing. The induction of increased tyrosine phosphorylation by thiol-reactive compounds, but not sulfo-NHS and iminothiolane, suggests that the ability to couple to thiol groups is a prerequisite for a compound to resemble potent contact sensitizers with regard to their capacity for activation of cells. In a ¢rst set of experiments we used human monocytes as substitute for professional antigen presenting cells because of their excellent response to stimulation with haptens and their availability for £ow cytometric analysis without any need for enrichment, subculture, or stimulation with cytokines. Basic ¢ndings acquired for monocytes were con¢rmed using mature mo-DC to demonstrate their relevance for professional antigen presenting cells. Although epidermal Langerhans cells are characterized as an immature dendritic cell population we used mature mo-DC for our experiments because these cells are superior to immature DC with regard to cell number and purity of the populations harvested. In several control experiments we found no signi¢cant di¡erences between immature and mature DC in our experimental system. This missing correlation between the stage of maturation and induction of tyrosine phosphorylation or activation of MAPK by contact sensitizers is in good accordance with our former results on monocytes, murine Langerhans cells, and other MHC class II positive cell populations (Kˇhn et al, 1998; Neisius et al, 1999; Brand et al, 2002). The reagents studied in this paper are in use in organic chemistry for ligation or blocking of free amino or thiol groups (sulfoNHS, NHM, NEM, iodoacetamide, sodium tetrathionate) as well as introduction of free thiol groups instead of amino groups (2-iminothiolane) (Donovan and Jennings, 1986; Mammoto et al, 1997; Meunier et al, 1999; Crow et al, 2001). We are not aware of any clinical or experimental data on the sensitizing potency of these compounds in the literature. In our experiments the amino-group-speci¢c chemicals failed to induce any sensitization in comparison to the strong contact sensitizer TNCB, whereas at least half of the animals developed a convincing ear swelling following treatment with NHM. Irritation by haptens can play a decisive role for the induction of contact hypersensitivity as well as elicitation of allergic contact dermatitis (Grabbe et al, 1996). The moderately irritant concentration of TNCB used for sensitization in our experiments can be one reason for the distinct response of
ACTIVATION BY COUPLING TO THIOL GROUPS
237
all animals. The thiol- or amino-speci¢c reagents tested were nonirritant. Therefore, in comparison to sulfo-NHS and iminothiolane the moderately sensitizing e¡ect of NHM is not explained by a higher potential to irritate. On the other hand we cannot exclude that the amino-group-speci¢c compounds may sensitize in the presence of an adjuvant or if applied in di¡erent vehicles or higher concentrations. Preliminary data on NEM demonstrated the need for a higher concentration for sensitization in comparison to NHM (data not shown). Therefore, the reactivity to thiol or amino groups might be only one feature determining the sensitizing potential in vivo. Further systematic experiments are planned that should show whether and under what conditions sensitization to the amino- and thiol-reactive compounds used in this study can be induced. From experimental model systems on the immunogenic haptenpeptide complexes recognized by trinitrophenyl-speci¢c T cells it is well known that trinitrophenyl groups couple to lysin residues within peptides in the binding groove of MHC class I as well class II molecules (Martin et al, 1992; Ortmann et al, 1992; Kohler et al, 1995). Our data show that, for initiation of early activation events like increased tyrosine phosphorylation of cell proteins and production of IL-1b as an important functional result of this activation, binding to thiol groups seems to be crucial. This does not imply inevitably that these hapten-protein conjugates are also the source of the peptides recognized ¢nally by T cells. It is likely that both coupling mechanisms take place under physiologic conditions and indeed a good reactivity to both amino as well as thiol groups has been demonstrated for TNCB and other experimental contact sensitizers (Parker et al, 1983). Therefore cellular structures involved in each type of reaction seem to be important for di¡erent parts of the sensitization process. To obtain data on the reactivity of professional antigen presenting cells the initial biochemical experiments on monocytes designed as control for the £ow cytometric analysis were continued using mature mo-DC. The capacity of thiol-reactive compounds as well as the failure of amino-reactive compounds to induce tyrosine phosphorylation could be con¢rmed. Although not identical, the pattern of phosphorylated proteins in cells stimulated with strong sensitizers (MCI/MI and TNCB) resembled the pattern seen for NHM and NEM with regard to the molecular weight of some protein bands. The distinct pattern even for strong contact sensitizers might re£ect di¡erences in the a⁄nity for or access to target structures. The capacity of contact sensitizers to activate p38 MAPK has been described before (Arrighi et al, 2001; Brand et al, 2002) and is con¢rmed in this presentation. Additionally we demonstrate an activation of ERK1/2 MAPK by the haptens studied as well as NHM and NEM but not amino-reactive compounds. On the other hand the thiol-reactive reagents failed to induce a phosphorylation of p38 MAPK whereas a weak but reproducible phosphorylation of p38 was noticed following stimulation with 2-iminothiolane. At present, we have no explanation for the heterogeneous response of di¡erent MAPK families. The regulation of p38 activation by haptens seems to be more complex and less reliable for discrimination from nonsensitizing compounds than supposed to date. Further work has to identify the proteins modi¢ed at free thiol groups by haptens and the molecular events responsible for the initiation of intracellular signaling. At present one can only speculate on these mechanisms. For example, reaction with a thiol group might induce conformational changes in the structure of a receptor or receptor-associated kinase. Alternatively, regulatory sites of kinases or phosphatases might be blocked resulting in excess activity of the corresponding signaling cascades. Attempts to identify the proteins to which haptens have to bind for initiation of activation might be more successful by focusing on proteins with such functionally relevant thiol groups. In addition, it will be very interesting to know whether blocking of these early events also prevents sensitization at all. This might be a chance to develop new strategies for primary prevention of contact hypersensitivity, especially occupational diseases.
238
BECKER ET AL
THE JOURNAL OF INVESTIGATIVE DERMATOLOGY
This work was supported by a grant from the Deutsche Forschungsgemeinschaft (SFB 548).The authors wish to thank S. Plochmann for skillful technical assistance.
REFERENCES Aiba S, Terunuma A, Manome H, Tagami H: Dendritic cells di¡erently respond to haptens and irritants by their production of cytokines and expression of costimulatory molecules. Eur J Immunol 27:3031^3038, 1997 Arrighi JF, Rebsamen M, Rousset F, Kindler V, Hauser C: A critical role for p38 mitogen-activated protein kinase in the maturation of human blood-derived dendritic cells induced by lipopolysaccharide, TNF-a, and contact sensitizers. J Immunol 166:3837^3845, 2001 Becker D, Neiss U, Neis S, Reske K, Knop J: Contact allergens modulate the expression of MHC class II molecules on murine epidermal Langerhans cells by endocytotic mechanisms. J Invest Dermatol 98:700^705, 1992 Becker D, Lempertz U, Enk A, Saloga J, Knop J: Contact sensitizers modulate mechanisms of receptor-mediated endocytosis but not £uid-phase endocytosis in murine epidermal Langerhans cells. Exp Dermatol 4:211^217, 1995 Brand P, Plochmann S, Valk E, Zahn S, Saloga J, Knop J, Becker D: Activation and translocation of p38 mitogen-activated protein kinase after stimulation of monocytes with contact sensitizers. J Invest Dermatol 119:99^106, 2002 Crow M, Taub DD, Cooper S, Broxmeyer HE, Sarris AH: Human recombinant interferon-inducible protein-10: intact disul¢de bridges are not required for inhibition of hematopoietic progenitors and chemotaxis of T lymphocytes and monocytes. J Hematother Stem Cell Res 10:147^156, 2001 Donovan JA, Jennings ML: N-hydroxysulfosuccinimido active esters and the L( þ )-lactate transport protein in rabbit erythrocytes. Biochemistry 25:1538^1545, 1986 Enk AH, Angeloni VL, Udey MC, Katz SI: An essential role for Langerhans cellderived IL-1b in the initiation of primary immune responses in skin. J Immunol 150:3698^3704, 1993 Enk AH, Saloga J, Becker D, Mohamadzadeh M, Knop J: Induction of hapten-speci¢c tolerance by interleukin 10 in vivo. J Exp Med 179:1397^1402, 1994 Freed BM, Mozayeni B, Lawrence DA, Wallach FR, Lempert N: Di¡erential inhibition of human T-lymphocyte activation by maleimide probes. Cell Immunol 101:181^194, 1986 Garrigue JL, Nicolas JF, Fraginals R, Benezra C, Bour H, Schmitt D: Optimization of the mouse ear swelling test for in vivo and in vitro studies of weak contact sensitizers. Contact Dermatitis 30:231^237, 1994 Grabbe S, Steinert M, Mahnke K, Schwartz A, Luger TA, Schwarz T: Dissection of antigenic and irritative e¡ects of epicutaneously applied haptens in mice. Evidence that not the antigenic component but nonspeci¢c proin£ammatory effects of haptens determine the concentration-dependent elicitation of allergic contact dermatitis. J Clin Invest 98:1158^1164, 1996 Gruvberger B, Bruze M: Can glutathione-containing emollients inactivate methylchloroisothiazolinone/methylisothiazolinone? Contact Dermatitis 38:261^265, 1998 Jonuleit H, Kuhn U, Muller G, et al: Pro-in£ammatory cytokines and prostaglandins induce maturation of potent immunostimulatory dendritic cells under fetal calf serum-free conditions. Eur J Immunol 27:3135^3142, 1997
Kohler J, Martin S, P£ugfelder U, Ruh H, Vollmer J, Weltzien HU: Cross-reactive trinitrophenylated peptides as antigens for class II major histocompatibility complex-restricted T cells and inducers of contact sensitivity in mice. Limited T cell receptor repertoire. Eur J Immunol 25:92^101, 1995 Kˇhn U, Brand P,Willemsen J, et al: Induction of tyrosine phosphorylation in human MHC class II-positive antigen-presenting cells by stimulation with contact sensitizers. J Immunol 160:667^673, 1998 Majima E, Koike H, Hong YM, Shinohara Y, Terada H: Characterization of cysteine residues of mitochondrial ADP/ATP carrier with the SH-reagents eosin 5-maleimide and N-ethylmaleimide. J Biol Chem 268:22181^22187, 1993 Mammoto A, Masumoto N, Tahara M, Yoneda M, Nishizaki T, Tasaka K, Miyake A: Involvement of a sperm protein sensitive to sulfhydryl-depleting reagents in mouse sperm-egg fusion. J Exp Zool 278:178^188, 1997 Martin S, Ortmann B, P£ugfelder U, Birsner U, Weltzien HU: Role of hapten-anchoring peptides in de¢ning hapten-epitopes for MHC-restricted cytotoxic. T cells. Cross-reactive TNP-determinants on di¡erent peptides. J Immunol 149:2569^2575, 1992 Meunier L, Bourgerie S, Mayer R, Roche AC, Monsigny M: Optimized conditions to couple two water-soluble biomolecules through alkylamine thiolation and thioetheri¢cation. Bioconjug Chem 10:206^212, 1999 Neisius U, Brand P, Plochmann S, Saloga J, Knop J, Becker D: Detection of increased tyrosine phosphorylation in murine Langerhans cells after stimulation with contact sensitizers. Arch Dermatol Res 291:22^27, 1999 Ortmann B, Martin S, von Bonin A, Schiltz E, Hoschutzky H, Weltzien HU: Synthetic peptides anchor T cell-speci¢c TNP epitopes to MHC antigens. J Immunol 148:1445^1450, 1992 Parker D, Long PV, Turk JL: A comparison of the conjugation of DNTB and other dinitrobenzenes with free protein radicals and their ability to sensitize or tolerize. J Invest Dermatol 81:198^201, 1983 Pichowski JS, Cumberbatch M, Dearman RJ, Basketter DA, Kimber I: Allergen-induced changes in interleukin 1 beta (IL-1b) mRNA expression by human blood-derived dendritic cells: inter-individual di¡erences and relevance for sensitization testing. J Appl Toxicol 21:115^121, 2001 Roberts DW, Lepoittevin J-P: Hapten^protein interactions. In: Lepoittevin J-P, Basketter DA, Goossens A, Karlberg A-T, eds. Allergic Contact Dermatitis ^ The Molecular Basis. Berlin, Heidelberg, New York: Springer, 1998: pp 81^111 Rosenspire AJ, Bodepudi S, Mathews M, McCabe MJ Jr: Low levels of ionic mercury modulate protein tyrosine phosphorylation in lymphocytes. Int J Immunopharmacol 20:697^707, 1998 Rougier N, Redziniak G, Mougin D, Schmitt D, Vincent C: In vitro evaluation of the sensitization potential of weak contact allergens using Langerhans-like dendritic cells and autologous T cells. Toxicol 145:73^82, 2000 Santucci B, Cannistraci C, Cristaudo A, Camera E, Picardo M: Thimerosal positivities: the role of SH groups and divalent ions. Contact Dermatitis 39:123^126, 1998 Santucci B, Cannistraci C, Cristaudo A, Camera E, Picardo M: Thimerosal positivities: patch testing to methylmercury chloride in subjects sensitive to ethylmercury chloride. Contact Dermatitis 40:8^13, 1999 Steinbrink K, Kolde G, Sorg C, Macher E: Induction of low zone tolerance to contact allergens in mice does not require functional Langerhans cells. J Invest Dermatol 107:243^247, 1996 Tuschl H, Kovac R, Weber E: The expression of surface markers on dendritic cells as indicators for the sensitizing potential of chemicals. Toxicol Vitro 14:541^549, 2000
Coupling of Contact Sensitizers to Thiol Groups is a Key Event for the Activation of Monocytes and Monocyte-Derived Dendritic Cells Detlef Becker, Elke Valk, Sabine Zahn, Pia Brand, and Jˇrgen Knop Department of Dermatology, University of Mainz, Germany
Strong contact sensitizers are able to induce distinct signal transduction mechanisms in antigen-presenting cells by coupling to cell proteins. The predominant target structures of haptens are thought to be thiol and amino groups in cysteine and lysine residues.We studied whether coupling of small reactive chemicals to thiol or amino groups might be responsible for the activation of monocytes and mature monocyte-derived dendritic cells. Human peripheral blood mononuclear cells were stimulated in vitro with subtoxic concentrations of the strong haptens 5-chloro-2-methylisothiazolinone plus 2-methylisothiazolinone and 2, 4, 6-trinitrochlorobenzene, the thiol-reactive reagents N-hydroxymaleimide and N-ethylmaleimide, as well as the amino-reactive compounds sulfosuccinimidyl acetate and 2-iminothiolane. Flow cytometric quanti¢cation of tyrosine phosphorylation in CD14 þ monocytes showed that 5-chloro-2-methylisothiazolinone plus 2-methylisothiazolinone, 2, 4, 6-trinitrochlorobenzene, N-hydroxymaleimide, and N-ethylmaleimide but not sulfosuccinimidyl acetate and 2-iminothiolane strongly induced this process. Tyrosine phosphorylation induced by 5-
chloro-2-methylisothiazolinone plus 2-methylisothiazolinone and 2, 4, 6-trinitrochlorobenzene was completely prevented in the presence of cysteine but not lysine, suggesting a competitive mechanism between cysteine and sulfhydryl groups of cell proteins. Using the mouse ear swelling test N-hydroxymaleimide could be classi¢ed as a signi¢cant contact allergen in comparison to 2, 4, 6-trinitrochlorobenzene, whereas no sensitizing potential became apparent for sulfosuccinimidyl acetate and 2-iminothiolane. Western blot analysis on monocytes and mature monocyte-derived dendritic cells con¢rmed the £ow cytometric data for tyrosine phosphorylation and demonstrated a selective capacity of haptens and thiol-reactive compounds to activate ERK1/2 mitogen-activated protein kinase. Our data show that strong a⁄nity of a small reactive chemical toward thiol groups is important for the activation of monocytes and monocyte-derived dendritic cells and can support the process of sensitization. Key words: thiol groups/amino groups/signal transduction/hapten. J Invest Dermatol 120:233 ^238, 2003
T
(Pichowski et al, 2001). We were able to demonstrate a vigorous phosphorylation of tyrosine residues in di¡erent populations of human MHC class II positive cells following stimulation with structurally nonrelated contact sensitizers (Kˇhn et al, 1998) and con¢rmed the relevance of these ¢ndings for murine Langerhans cells (Neisius et al, 1999). Recently, Arrighi et al (2001) reported on the activation of the p38 mitogen-activated protein kinase (MAPK) in human mo-DC by dinitro£uorobenzene and nickel sulfate. Using human monocytes as model cells we found similar results and especially a remarkable translocation of p38 MAPK from the cytoplasm to membrane-associated compartments (Brand et al, 2002). At present, the structures to which contact sensitizers have to bind for initiation of these signaling cascades are unknown. From a theoretical point of view haptens are thought to bind predominantly to thiol and amino groups in cysteine and lysine residues of cell proteins (Roberts and Lepoittevin, 1998). A good reactivity to amino as well as thiol groups has been demonstrated for 2, 4, 6 -trinitrochlorobenzene (TNCB) and other experimental contact sensitizers in vitro (Parker et al, 1983). For mercury compounds (Rosenspire et al, 1998; Santucci et al, 1998; 1999) and 5-chloro-2methylisothiazolinone plus 2-methylisothiazolinone (MCI/MI) (Gruvberger and Bruze, 1998) a direct interaction between hapten and thiol groups has been shown or at least supposed.
here is plenty of evidence for the capacity of contact sensitizers to activate epidermal Langerhans cells. Although some events might be induced indirectly by cytokines derived from other epidermal cell populations several studies demonstrated a direct in£uence of haptens on cellular and molecular changes in dendritic cells (DC) and other major histocompatibility complex (MHC) class II positive cell populations. This includes increased receptor-mediated endocytosis of membrane molecules by Langerhans cells (Becker et al, 1992; 1995), cytokine production and upregulation of membrane markers in monocyte-derived dendritic cells (mo-DC) (Aiba et al, 1997; Rougier et al, 2000; Tuschl et al, 2000), and induction of mRNA for interleukin-1b (IL-1b) Manuscript received October 31, 2001; revised October 10, 2002; accepted for publication October 14, 2002 Reprint requests to: Detlef Becker, M.D., UniversitHts-Hautklinik, Langenbeckstr. 1, D-55131 Mainz, Germany; Email: [email protected]. uni-mainz.de Abbreviations: MAPK, mitogen-activated protein kinase; MCI/MI, 5chloro-2-methylisothiazolinone plus 2-methylisothiazolinone; mo-DC, monocyte-derived dendritic cells; NEM, N-ethylmaleimide; NHM, Nhydroxymaleimide; PBMC, peripheral blood mononuclear cells; sulfoNHS, sulfosuccinimidyl acetate; TNCB, 2, 4, 6 -trinitrochlorobenzene.
0022-202X/03/$15.00 . Copyright r 2003 by The Society for Investigative Dermatology, Inc. 233
234
BECKER ET AL
From experiments on the capacity of thiol antioxidants to block the activation of cells by contact sensitizers (Bruchhausen et al, submitted) we concluded that coupling to thiol groups might be very important for this process. In this paper we studied whether de¢ned thiol- or amino-reactive reagents would be able to resemble the activation process induced by contact sensitizers. MATERIALS AND METHODS Culture of mo-DC Human mo-DC were prepared from peripheral blood mononuclear cells (PBMC) by short-term adherence in the modi¢cation described by Jonuleit et al (1997). Brie£y, PBMC were isolated from leukocyte-enriched bu¡y coats (Transfusion Center, Mainz, Germany) by Ficoll-Paque 1.077 (Biochrom, Berlin, Germany) and incubated (45 min, 371C, 5% CO2) in six-well plates (Greiner, Frickenhausen, Karlstahe, Germany) using RPMI 1640 medium (Life Technologies) supplemented with 3% autologous plasma. Non-adherent cells were removed by washing gently with phosphate-bu¡ered saline (PBS). The remaining cells were cultured in RPMI with 5% autologous serum, 1000 U per ml IL- 4 (Strathmann Biotech, Hannover, Germany), and 800 U per ml granulocyte-macrophage colony stimulating factor (GM-CSF) (Leucomax, Sandoz, Basel, Switzerland). Cells were fed every other day with RPMI, 5% autologous serum, IL- 4 (1000 U per ml), and GM-CSF (800 U per ml). On day 7 the medium was supplemented with 10 ng per ml IL-1b (Strathmann Biotech), 10 ng per ml tumor necrosis factor a (Strathmann Biotech), and 1 mg per ml prostaglandin E2 (Minprostin, Pharmacia & Upjohn, Heppenheim, Germany). Mature DC were harvested on day 9. The phenotype of DC was determined by £ow cytometry as 70% CD83 þ , 4 80% CD86 þ , and 4 90% MHCII þ ; viability routinely exceeded 90% as determined by propidium iodide staining. Stimulation of PBMC with chemicals Human PBMC (106 per ml) were stimulated in vitro with subtoxic concentrations of the strong haptens MCI/MI (Hermal, Reinbeck, Germany) and TNCB (ICN, Costa Mesa, CA), the thiol-reactive reagents N-hydroxymaleimide (NHM), Nethylmaleimide (NEM), iodoactetamide, and sodium tetrathionate (all obtained from Sigma, St. Louis, MO), as well as the amino-reactive compounds sulfosuccinimidyl acetate (sulfo-NHS, Pierce, Rockford, IL) and 2-iminothiolane (Pierce) for 15 min at 371C in RPMI medium. If indicated 10 mM cysteine (Sigma) or 10 mM lysine (Roth, Karlsruhe, Germany) was present during incubation. Stock solutions of TNCB were prepared in dimethylsulfoxide; the other chemicals were dissolved in culture medium immediately before use. The ¢nal concentration of dimethylsulfoxide never exceeded 0.1%. Flow cytometric analysis To perform intracellular £uorescenceactivated cell sorter (FACS) analysis, PBMC were permeabilized immediately after stimulation with PBS/1% fetal bovine serum (FBS)/ 0.25% saponine (Roth) and stained by phycoerythrin-conjugated antiCD14 (10 mg per ml; Immunotech, Marseille, France) in combination with £uorescein isothiocyanate conjugated antiphosphotyrosine (PY20FITC; 4 mg per ml; Leinco Technologies, St. Louis, MO) dissolved in PBS/1% FBS/0.25% saponine for 20 min at 41C. Finally, samples were washed twice in cold washing bu¡er (PBS/1% FBS/0.25% saponine). Fluorescence intensities in channel 1 (phosphotyrosine) were recorded for cells gated in channel 2 (CD14 þ ) using a FACScalibur £ow cytometer (Becton Dickinson, Heidelberg, Germany) equipped with the CellQuest software. Standard calibration procedures were used following the recommendation of Becton Dickinson. The level of signi¢cance for a di¡erence between medium treated and stimulated cells was calculated using a paired Student t test. Measurement of IL-1b production by enzyme-linked immunosorbent assay (ELISA) PBMC were resuspended in Iscove’s medium/ 1% autologous plasma and seeded onto a 12-well plate (8 106 cells per well). After 45 min the nonadherent cells were removed by washing with warm PBS. After stimulation with MCI/MI (0.4 mg per ml), NHM (25 mg per ml), 2-iminothiolane (400 mg per ml), and sulfo-NHS (2 mg per ml) for 15 min, the medium was aspirated and replaced by 1.5 ml fresh medium containing 1% autologous plasma. Supernatants were collected after 24 h and assessed for IL-1b production by ELISA according to the recommendations of the distributor of the pair of antibodies employed (R&D Systems, Wiesbaden, Germany). Brie£y, Maxisorp Nunc Immunoplates (Nunc, Wiesbaden, Germany) were coated overnight at 41C with 2 mg per ml capture antibody diluted in 0.1 M NaHCO3, pH 8.2. After washing with PBS/0.05% Tween-20, the plates
THE JOURNAL OF INVESTIGATIVE DERMATOLOGY
were blocked for 2 h at 371C with PBS/1% bovine serum albumin. Serial dilutions of the provided standards (R&D Systems) and collected supernatants were incubated overnight at 41C. The next day the plates were washed and incubated at 371C for 1 h with 10 ng per ml biotinylated detection antibody, diluted in PBS/0.1% bovine serum albumin. After washing, 2.5 mg per ml peroxidase-conjugated avidin (Sigma) was added for 30 min. The plates were then developed with ABTS and read in a microplate reader model 450 (Bio-Rad) at 405 nm. IL-1b concentration was calculated according to the standard curves with a detection limit of 16 pg per ml. IL-1b values after stimulation were calculated as percentage of the corresponding medium control. The level of signi¢cance for a positive di¡erence between medium treated and stimulated cells was calculated using a paired Student t test. Preparation of cell extracts, sodium dodecyl sulfate (SDS) polyacrylamide gel electrophoresis, and Western blot 5 106 enriched monocytes or 1 106 mature mo-DC were stimulated as indicated in culture medium or PBS. Cells were resuspended in 50 ml icecold lysis bu¡er (50 mM HEPES, pH 7.4, 150 mM NaCl, 1.5 mM MgCl2, 1 mM ethylenediamine tetraacetic acid, 10% glycerol, 1% Triton X-100, 1 mM phenylmethylsulfonyl £uoride, 10 mg per ml aprotinin, and 1 mM sodium orthovanadate) and incubated on ice for 30 min. Insoluble material was spun down (10 min, 14,000 rpm, 41C) and clear supernatants were stored at ^701C. Protein content of the lysates was determined by BCA protein assay (Pierce). Cell lysates were mixed with loading bu¡er (Roti Load, Roth), heated for 5 min at 961C, and resolved on 8% SDS polyacrylamide gels (100 mg per lane) using standard procedures. Proteins were transferred to PVDF membranes (Millipore, Bedford, MA) by semidry blotting (Bio-Rad, Hercules, CA) in blotting bu¡er (Roti-Blot A, Roti-Blot K, Roth). Membranes were blocked for 1 h in blocking bu¡er (1 PBS, 0,1% Tween-20, 5% nonfat dry milk powder) and incubated with the following primary antibodies: mouse antiphosphotyrosine (clone 4G10; 0.5 mg per ml; Upstate Biotechnology, Lake Placid, NY), rabbit anti-p-ERK1/2 or rabbit anti-p-p38 (dilution 1:1000, New England Biolabs, Beverly, MA). The src-kinase Lyn was detected as control for equal protein loading (anti-Lyn; Santa Cruz Biotechnology, Santa Cruz, CA; dilution 1:1000). After washing, antimouse horseradish peroxidase (0.2 mg per ml, Dianova) or antirabbit horseradish peroxidase (dilution 1: 2000, New England Biolabs) was added. All antibodies were diluted in blocking bu¡er. Blots were developed using chemiluminescence (ECL Plus, Amersham, Buckinghamshire, U.K.). For sequential detection of di¡erent antigens membranes were stripped (40 min at 501C in 62.5 mM Tris^HCl, 2% SDS, 100 mM b-mercaptoethanol, pH 6.7) and reprobed. Mouse ear swelling test Both sides of the left ears of 8^12-wk-old Balb/c mice of both sexes were painted with 15 ml each of 120 mM TNCB, NHM, sulfo-NHS, and 2-iminothiolane on days 1, 2, and 3. TNCB was dissolved in ethanol, the other compound in ethanol/aqua dest. (50%/50%). At day 6, 40 mM of the respective compounds were applied on both sides of the right ears. Ear thickness was recorded before and 24 h after challenge using an engineer’s micrometer. Control animals received vehicle alone, the sensitizing or challenging dose at day 1, and were measured at day 2 to monitor any irritant e¡ects. For this set of experiments the level of signi¢cance for a di¡erence between vehicle control and applied chemical was calculated using a paired Student t test. All animal experiments were performed according to the guidelines and with the permission of the government of Rheinland-Pfalz.
RESULTS Contact sensitizers and thiol-group-reactive compounds but not amino-group-reactive chemicals induce tyrosine phosphorylation in monocytes MCI/MI and TNCB as well as the thiol-group-speci¢c reagents NHM (Freed et al, 1986) and NEM (Majima et al, 1993) strongly induced tyrosine phosphorylation in CD14 þ monocytes (Fig 1). In contrast sulfoNHS and 2-iminothiolane, two common amino-group-speci¢c reagents (Donovan and Jennings, 1986; Meunier et al, 1999) failed to induce tyrosine phosphorylation (Fig 1). A clear dose-response relationship became apparent for the activating compounds whereas neither sulfo-NHS nor 2-iminothiolane showed any reactivity even in the highest concentration used (Fig 2). To exclude that the induction of tyrosine phosphorylation might be a speci¢c feature of maleimides, control experiments were performed using the structurally nonrelated thiol-reactive
VOL. 120, NO. 2 FEBRUARY 2003
Figure 1. Induction of tyrosine phosphorylation by strong contact sensitizers and thiol-reactive reagents but not by amino-reactive compounds. PBMC were incubated for 15 min with the strong contact sensitizers MCI/MI and TNCB, the thiol-reactive reagents NHM and NEM, and the amino-reactive compounds 2-iminothiolane and sulfoNHS and analyzed for the amount of tyrosine-phosphorylated proteins using £ow cytometric analysis. Histograms for the binding of phosphotyrosine-speci¢c antibodies to CD14+ monocytes are shown. Concentrations used for stimulation and the mean total £uorescence intensities are indicated.
compounds iodoacetamide (Crow et al, 2001) and sodium tetrathionate (Mammoto et al, 1997). Both were able to induce tyrosine phosphorylation in CD14 þ monocytes (iodoacetamide 370 mg per ml, relative mean increase 467%7131%, n ¼ 15, p ¼ 0.02; sodium tetrathionate 10 mg per ml, relative mean increase 254%794.5%, n ¼ 6, p ¼ 0.01). Toxic e¡ects were excluded in pilot experiments using propidium iodide as marker for cell damage (data not shown); all compounds were studied using the highest but still subtoxic range of concentration. Cysteine but not lysine blocks tyrosine phosphorylation induced by MCI/MI and TNCB The capacity of cysteine and lysine to modulate the degree of tyrosine phosphorylation in CD14 þ monocytes under stimulation with MCI/MI and TNCB was explored. Cysteine but not lysine completely blocked any tyrosine phosphorylation, suggesting a competitive in£uence of free thiol groups but not amino groups on the mechanism responsible for the activation of tyrosine kinases by both contact sensitizers (Fig 3). NHM and MCI/MI but not amino-group-reactive compounds induce IL-1b secretion by monocytes Human monocytes were enriched by adhesion and stimulated with MCI/MI, NHM, 2-iminothiolane, and sulfo-NHS before subculture for 24 h. IL-1b was measured in culture supernatants
ACTIVATION BY COUPLING TO THIOL GROUPS
235
Figure 2. Dose-response relationship for the induction of tyrosine phosphorylation by contact sensitizers NHM and NEM. CD14+ monocytes stimulated with the compounds and concentrations indicated were analyzed by £ow cytometry. Means7SEM of the total £uorescence intensities determined in ¢ve independent experiments are shown. A doserelated signi¢cant increase can be noticed for MCI/MI,TNCB, NHM, and NEM (p o 0.01 is indicated by an asterisk).
Figure 3. Tyrosine phosphorylation induced by MCI/MI and TNCB is blocked in the presence of cysteine but not lysine. PBMC were stimulated with MCI/MI and TNCB in the presence or absence of 10 mM cysteine or lysine. Selective analysis of tyrosine phosphorylation in CD14+ monocytes was performed by £ow cytometry. Means7SEM of the total £uorescence intensities determined in six independent experiments are shown. An asterisk indicates signi¢cance (p o 0.01) for the di¡erence between stimulated cells and the medium control.
by ELISA. In comparison to the basal production of unstimulated cells (mean 30.5 pg per ml, SEM 19.5 pg per ml) a signi¢cant liberation of IL-1b was found for MCI/MI and NHM (p o 0.05) but not for the two amino-reactive compounds (Fig 4). Therefore NHM resembles the e¡ect of a strong contact sensitizer by induction of this important cytokine for the sensitization phase of allergic contact dermatitis (Enk et al, 1993).
236
BECKER ET AL
Figure 4. Monocytes secret IL-1b following stimulation with MCI/ MI and NHM but not amino-reactive compounds. Enriched monocytes were stimulated as indicated and cultured for 24 h. IL-1b was determined in culture supernatants using an ELISA technique. Means7SEM of four independent experiments are shown (p o 0.05 is indicated by an asterisk).
Figure 5. Induction of contact hypersensitivity in Balb/c mice by TNCB and NHM but not sulfo-NHS and 2-iminothiolane. The capacity of NHM, sulfo-NHS, and 2-iminothiolane to induce contact hypersensitivity in a modi¢ed mouse ear swelling test was compared to TNCB as control for a potent contact sensitizer. Sensitization was performed with 120 mM and elicitation with 40 mM of each compound. The di¡erence in ear thickness between the solvent control and treated animals from three independent experiments with two animals in each group is indicated.
NHM but not amino-group-reactive compounds induce contact hypersensitivity in mice The capacity of NHM, sulfo-NHS, and 2-iminothiolane to sensitize Balb/c mice was compared to TNCB by testing equimolar concentrations in a modi¢ed mouse ear swelling test (Garrigue et al, 1994). All animals were sensitized by the strong experimental hapten TNCB using a moderate concentration (Enk et al, 1994; Steinbrink et al, 1996). Whereas at least three of six animals responded signi¢cantly to NHM, no evidence for a sensitizing potential of the other two compounds was found (Fig 5). Experiments performed to monitor any irritant potential showed that none of the compounds was irritant when used at 40 mM (elicitation). The sensitizing dose (120 mM) of TNCB was found to be moderately irritant (p o 0.01) whereas the three thiol- or amino-speci¢c reagents were well tolerated (Fig 6). NHM and NEM induce a contact-sensitizer-like pattern of tyrosine-phosphorylated proteins and activate MAPK Biochemical analysis on enriched monocytes stimulated with NHM revealed augmented phosphorylation of several protein bands in comparison to medium treated cells and resembled the
THE JOURNAL OF INVESTIGATIVE DERMATOLOGY
Figure 6. Only the sensitizing dose of TNCB is moderately irritant. Ear thickness measured 24 h after application of the sensitizing and elicitating doses of TNCB, NHM, sulfo-NHS, and 2-iminothiolane is shown in comparison to the solvent control. Means7SEM of three independent experiments with two animals in each group are shown (p o 0.01 is indicated by an asterisk).
Figure 7. Contact sensitizers and thiol-reactive but not amino-reactive reagents induce phosphorylation of tyrosine residues and MAPK. (a) Western blot analysis of enriched monocytes stimulated with TNCB and NHM reveals a similar pattern of tyrosine-phosphorylated proteins. Enriched monocytes were stimulated with 0.8 mg per ml TNCB or 25 mg per ml NHM. The pattern of tyrosine-phosphorylated proteins is shown in comparison to the medium control. The total £uorescence intensities for tyrosine-phosphorylated proteins of these cells as assessed by £ow cytometry are as follows: medium 4.4; TNCB 40.3; NHM 34.8. The blot is representative for three di¡erent experiments with comparable results. (b) Mature mo-DC were stimulated with 2 mg per ml sulfo-NHS, 400 mg per ml 2-iminothiolane, 25 mg per ml NHM, 25 mg per ml NEM, 0.8 mg per ml MCI/MI, and 0.8 mg per ml TNCB. Tyrosine-phosphorylated proteins, phosphorylated ERK1/2 MAPK, and p38 MAPK are shown in comparison to the medium control. The blot is representative for three di¡erent experiments with comparable results.
pattern seen for the strong contact sensitizer TNCB (Fig 7A). As reported before (Kˇhn et al, 1998) this correlated very well with the increase in total tyrosine phosphorylation as monitored by FACS analysis. To obtain data for professional antigen presenting cells £ow cytometric analysis on mature mo-DC was set up and demonstrated a similar response upon stimulation with contact sensitizers and thiol-reactive compounds as noticed for monocytes (data not shown). A detailed biochemical analysis on mature mo-DC stimulated with sulfo-NHS, 2-iminothiolane, NHM, NEM, MCI/MI, and TNCB revealed augmented
VOL. 120, NO. 2 FEBRUARY 2003
phosphorylation of several protein bands in comparison to medium treated cells for the contact sensitizers and the thiolreactive chemicals but not the amino-reactive compounds. The phosphorylation pattern induced by NHM and NEM resembled the pattern induced by haptens with regard to the molecular weight of several bands (Fig 7B). Only haptens and the thiolreactive compounds were able to induce phosphorylation of ERK1/2 MAPK whereas activation of p38 MAPK was induced by haptens but not NHM or NEM. Following stimulation with 2-iminothiolane some low but reproducible increase in phosphop38 became apparent. Expression of the src kinase Lyn was monitored as control for protein loading and was found to be comparable in all probes (data not shown). DISCUSSION
In contrast to protein allergens contact sensitizers are characterized as reactive compounds that must bind to skin proteins in order to be recognized by the immune system. Additionally, the coupling of strong contact sensitizers leads to the induction of signaling events in the cell. This could take place by activation of the corresponding receptors, kinases, or other intermediate components of de¢ned signaling cascades. The possible chemical coupling mechanisms of haptens to proteins have been described in detail (Roberts and Lepoittevin, 1998). Essentially, amino groups as well as thiol groups should be involved. Although some limited data have been presented for mercury compounds, MCI/MI, and strong experimental contact sensitizers, systematic studies on the preferential target structures of certain haptens or hapten groups and their importance for cell activation are missing. The induction of increased tyrosine phosphorylation by thiol-reactive compounds, but not sulfo-NHS and iminothiolane, suggests that the ability to couple to thiol groups is a prerequisite for a compound to resemble potent contact sensitizers with regard to their capacity for activation of cells. In a ¢rst set of experiments we used human monocytes as substitute for professional antigen presenting cells because of their excellent response to stimulation with haptens and their availability for £ow cytometric analysis without any need for enrichment, subculture, or stimulation with cytokines. Basic ¢ndings acquired for monocytes were con¢rmed using mature mo-DC to demonstrate their relevance for professional antigen presenting cells. Although epidermal Langerhans cells are characterized as an immature dendritic cell population we used mature mo-DC for our experiments because these cells are superior to immature DC with regard to cell number and purity of the populations harvested. In several control experiments we found no signi¢cant di¡erences between immature and mature DC in our experimental system. This missing correlation between the stage of maturation and induction of tyrosine phosphorylation or activation of MAPK by contact sensitizers is in good accordance with our former results on monocytes, murine Langerhans cells, and other MHC class II positive cell populations (Kˇhn et al, 1998; Neisius et al, 1999; Brand et al, 2002). The reagents studied in this paper are in use in organic chemistry for ligation or blocking of free amino or thiol groups (sulfoNHS, NHM, NEM, iodoacetamide, sodium tetrathionate) as well as introduction of free thiol groups instead of amino groups (2-iminothiolane) (Donovan and Jennings, 1986; Mammoto et al, 1997; Meunier et al, 1999; Crow et al, 2001). We are not aware of any clinical or experimental data on the sensitizing potency of these compounds in the literature. In our experiments the amino-group-speci¢c chemicals failed to induce any sensitization in comparison to the strong contact sensitizer TNCB, whereas at least half of the animals developed a convincing ear swelling following treatment with NHM. Irritation by haptens can play a decisive role for the induction of contact hypersensitivity as well as elicitation of allergic contact dermatitis (Grabbe et al, 1996). The moderately irritant concentration of TNCB used for sensitization in our experiments can be one reason for the distinct response of
ACTIVATION BY COUPLING TO THIOL GROUPS
237
all animals. The thiol- or amino-speci¢c reagents tested were nonirritant. Therefore, in comparison to sulfo-NHS and iminothiolane the moderately sensitizing e¡ect of NHM is not explained by a higher potential to irritate. On the other hand we cannot exclude that the amino-group-speci¢c compounds may sensitize in the presence of an adjuvant or if applied in di¡erent vehicles or higher concentrations. Preliminary data on NEM demonstrated the need for a higher concentration for sensitization in comparison to NHM (data not shown). Therefore, the reactivity to thiol or amino groups might be only one feature determining the sensitizing potential in vivo. Further systematic experiments are planned that should show whether and under what conditions sensitization to the amino- and thiol-reactive compounds used in this study can be induced. From experimental model systems on the immunogenic haptenpeptide complexes recognized by trinitrophenyl-speci¢c T cells it is well known that trinitrophenyl groups couple to lysin residues within peptides in the binding groove of MHC class I as well class II molecules (Martin et al, 1992; Ortmann et al, 1992; Kohler et al, 1995). Our data show that, for initiation of early activation events like increased tyrosine phosphorylation of cell proteins and production of IL-1b as an important functional result of this activation, binding to thiol groups seems to be crucial. This does not imply inevitably that these hapten-protein conjugates are also the source of the peptides recognized ¢nally by T cells. It is likely that both coupling mechanisms take place under physiologic conditions and indeed a good reactivity to both amino as well as thiol groups has been demonstrated for TNCB and other experimental contact sensitizers (Parker et al, 1983). Therefore cellular structures involved in each type of reaction seem to be important for di¡erent parts of the sensitization process. To obtain data on the reactivity of professional antigen presenting cells the initial biochemical experiments on monocytes designed as control for the £ow cytometric analysis were continued using mature mo-DC. The capacity of thiol-reactive compounds as well as the failure of amino-reactive compounds to induce tyrosine phosphorylation could be con¢rmed. Although not identical, the pattern of phosphorylated proteins in cells stimulated with strong sensitizers (MCI/MI and TNCB) resembled the pattern seen for NHM and NEM with regard to the molecular weight of some protein bands. The distinct pattern even for strong contact sensitizers might re£ect di¡erences in the a⁄nity for or access to target structures. The capacity of contact sensitizers to activate p38 MAPK has been described before (Arrighi et al, 2001; Brand et al, 2002) and is con¢rmed in this presentation. Additionally we demonstrate an activation of ERK1/2 MAPK by the haptens studied as well as NHM and NEM but not amino-reactive compounds. On the other hand the thiol-reactive reagents failed to induce a phosphorylation of p38 MAPK whereas a weak but reproducible phosphorylation of p38 was noticed following stimulation with 2-iminothiolane. At present, we have no explanation for the heterogeneous response of di¡erent MAPK families. The regulation of p38 activation by haptens seems to be more complex and less reliable for discrimination from nonsensitizing compounds than supposed to date. Further work has to identify the proteins modi¢ed at free thiol groups by haptens and the molecular events responsible for the initiation of intracellular signaling. At present one can only speculate on these mechanisms. For example, reaction with a thiol group might induce conformational changes in the structure of a receptor or receptor-associated kinase. Alternatively, regulatory sites of kinases or phosphatases might be blocked resulting in excess activity of the corresponding signaling cascades. Attempts to identify the proteins to which haptens have to bind for initiation of activation might be more successful by focusing on proteins with such functionally relevant thiol groups. In addition, it will be very interesting to know whether blocking of these early events also prevents sensitization at all. This might be a chance to develop new strategies for primary prevention of contact hypersensitivity, especially occupational diseases.
238
BECKER ET AL
THE JOURNAL OF INVESTIGATIVE DERMATOLOGY
This work was supported by a grant from the Deutsche Forschungsgemeinschaft (SFB 548).The authors wish to thank S. Plochmann for skillful technical assistance.
REFERENCES Aiba S, Terunuma A, Manome H, Tagami H: Dendritic cells di¡erently respond to haptens and irritants by their production of cytokines and expression of costimulatory molecules. Eur J Immunol 27:3031^3038, 1997 Arrighi JF, Rebsamen M, Rousset F, Kindler V, Hauser C: A critical role for p38 mitogen-activated protein kinase in the maturation of human blood-derived dendritic cells induced by lipopolysaccharide, TNF-a, and contact sensitizers. J Immunol 166:3837^3845, 2001 Becker D, Neiss U, Neis S, Reske K, Knop J: Contact allergens modulate the expression of MHC class II molecules on murine epidermal Langerhans cells by endocytotic mechanisms. J Invest Dermatol 98:700^705, 1992 Becker D, Lempertz U, Enk A, Saloga J, Knop J: Contact sensitizers modulate mechanisms of receptor-mediated endocytosis but not £uid-phase endocytosis in murine epidermal Langerhans cells. Exp Dermatol 4:211^217, 1995 Brand P, Plochmann S, Valk E, Zahn S, Saloga J, Knop J, Becker D: Activation and translocation of p38 mitogen-activated protein kinase after stimulation of monocytes with contact sensitizers. J Invest Dermatol 119:99^106, 2002 Crow M, Taub DD, Cooper S, Broxmeyer HE, Sarris AH: Human recombinant interferon-inducible protein-10: intact disul¢de bridges are not required for inhibition of hematopoietic progenitors and chemotaxis of T lymphocytes and monocytes. J Hematother Stem Cell Res 10:147^156, 2001 Donovan JA, Jennings ML: N-hydroxysulfosuccinimido active esters and the L( þ )-lactate transport protein in rabbit erythrocytes. Biochemistry 25:1538^1545, 1986 Enk AH, Angeloni VL, Udey MC, Katz SI: An essential role for Langerhans cellderived IL-1b in the initiation of primary immune responses in skin. J Immunol 150:3698^3704, 1993 Enk AH, Saloga J, Becker D, Mohamadzadeh M, Knop J: Induction of hapten-speci¢c tolerance by interleukin 10 in vivo. J Exp Med 179:1397^1402, 1994 Freed BM, Mozayeni B, Lawrence DA, Wallach FR, Lempert N: Di¡erential inhibition of human T-lymphocyte activation by maleimide probes. Cell Immunol 101:181^194, 1986 Garrigue JL, Nicolas JF, Fraginals R, Benezra C, Bour H, Schmitt D: Optimization of the mouse ear swelling test for in vivo and in vitro studies of weak contact sensitizers. Contact Dermatitis 30:231^237, 1994 Grabbe S, Steinert M, Mahnke K, Schwartz A, Luger TA, Schwarz T: Dissection of antigenic and irritative e¡ects of epicutaneously applied haptens in mice. Evidence that not the antigenic component but nonspeci¢c proin£ammatory effects of haptens determine the concentration-dependent elicitation of allergic contact dermatitis. J Clin Invest 98:1158^1164, 1996 Gruvberger B, Bruze M: Can glutathione-containing emollients inactivate methylchloroisothiazolinone/methylisothiazolinone? Contact Dermatitis 38:261^265, 1998 Jonuleit H, Kuhn U, Muller G, et al: Pro-in£ammatory cytokines and prostaglandins induce maturation of potent immunostimulatory dendritic cells under fetal calf serum-free conditions. Eur J Immunol 27:3135^3142, 1997
Kohler J, Martin S, P£ugfelder U, Ruh H, Vollmer J, Weltzien HU: Cross-reactive trinitrophenylated peptides as antigens for class II major histocompatibility complex-restricted T cells and inducers of contact sensitivity in mice. Limited T cell receptor repertoire. Eur J Immunol 25:92^101, 1995 Kˇhn U, Brand P,Willemsen J, et al: Induction of tyrosine phosphorylation in human MHC class II-positive antigen-presenting cells by stimulation with contact sensitizers. J Immunol 160:667^673, 1998 Majima E, Koike H, Hong YM, Shinohara Y, Terada H: Characterization of cysteine residues of mitochondrial ADP/ATP carrier with the SH-reagents eosin 5-maleimide and N-ethylmaleimide. J Biol Chem 268:22181^22187, 1993 Mammoto A, Masumoto N, Tahara M, Yoneda M, Nishizaki T, Tasaka K, Miyake A: Involvement of a sperm protein sensitive to sulfhydryl-depleting reagents in mouse sperm-egg fusion. J Exp Zool 278:178^188, 1997 Martin S, Ortmann B, P£ugfelder U, Birsner U, Weltzien HU: Role of hapten-anchoring peptides in de¢ning hapten-epitopes for MHC-restricted cytotoxic. T cells. Cross-reactive TNP-determinants on di¡erent peptides. J Immunol 149:2569^2575, 1992 Meunier L, Bourgerie S, Mayer R, Roche AC, Monsigny M: Optimized conditions to couple two water-soluble biomolecules through alkylamine thiolation and thioetheri¢cation. Bioconjug Chem 10:206^212, 1999 Neisius U, Brand P, Plochmann S, Saloga J, Knop J, Becker D: Detection of increased tyrosine phosphorylation in murine Langerhans cells after stimulation with contact sensitizers. Arch Dermatol Res 291:22^27, 1999 Ortmann B, Martin S, von Bonin A, Schiltz E, Hoschutzky H, Weltzien HU: Synthetic peptides anchor T cell-speci¢c TNP epitopes to MHC antigens. J Immunol 148:1445^1450, 1992 Parker D, Long PV, Turk JL: A comparison of the conjugation of DNTB and other dinitrobenzenes with free protein radicals and their ability to sensitize or tolerize. J Invest Dermatol 81:198^201, 1983 Pichowski JS, Cumberbatch M, Dearman RJ, Basketter DA, Kimber I: Allergen-induced changes in interleukin 1 beta (IL-1b) mRNA expression by human blood-derived dendritic cells: inter-individual di¡erences and relevance for sensitization testing. J Appl Toxicol 21:115^121, 2001 Roberts DW, Lepoittevin J-P: Hapten^protein interactions. In: Lepoittevin J-P, Basketter DA, Goossens A, Karlberg A-T, eds. Allergic Contact Dermatitis ^ The Molecular Basis. Berlin, Heidelberg, New York: Springer, 1998: pp 81^111 Rosenspire AJ, Bodepudi S, Mathews M, McCabe MJ Jr: Low levels of ionic mercury modulate protein tyrosine phosphorylation in lymphocytes. Int J Immunopharmacol 20:697^707, 1998 Rougier N, Redziniak G, Mougin D, Schmitt D, Vincent C: In vitro evaluation of the sensitization potential of weak contact allergens using Langerhans-like dendritic cells and autologous T cells. Toxicol 145:73^82, 2000 Santucci B, Cannistraci C, Cristaudo A, Camera E, Picardo M: Thimerosal positivities: the role of SH groups and divalent ions. Contact Dermatitis 39:123^126, 1998 Santucci B, Cannistraci C, Cristaudo A, Camera E, Picardo M: Thimerosal positivities: patch testing to methylmercury chloride in subjects sensitive to ethylmercury chloride. Contact Dermatitis 40:8^13, 1999 Steinbrink K, Kolde G, Sorg C, Macher E: Induction of low zone tolerance to contact allergens in mice does not require functional Langerhans cells. J Invest Dermatol 107:243^247, 1996 Tuschl H, Kovac R, Weber E: The expression of surface markers on dendritic cells as indicators for the sensitizing potential of chemicals. Toxicol Vitro 14:541^549, 2000