- Email: [email protected]
In vitro human T cell sensitization to haptens by monocyte-derived dendritic cells
Toxicology in Vitro 14 (2000) 517±522
www.elsevier.com/locate/toxinvit
In vitro human T cell sensitization to haptens by monocyte-derived dendritic cells G. GUIRONNET 1, C. DALBIEZ-GAUTHIER 1, F. ROUSSET 2, D. SCHMITT 1 and J. PEÂGUET-NAVARRO 1,* 1
INSERM U346, Pavillon R, HoÃpital E. Herriot, 69374, Lyon 03, and 2L'OreÂal, 1 avenue EugeÁne Schueller, 93190, Aulnay/Bois, France (Accepted 27 April 2000)
AbstractÐWe previously reported that in vitro primary sensitization of hapten-speci®c T cells by cultured human epidermal Langerhans cells (LC) provides an alternative approach to discriminate strong contact sensitizers from irritants (Krasteva et al., 1996; Moulon et al., 1993). However, this LC-based immunoassay was limited by the availability of human skin samples. In the present study, we used monocyte-derived dendritic cells (DC) to analyse the autologous proliferative T cell response to several allergens. Monocytes were puri®ed from the peripheral blood of healthy donors and cultured for 6±8 days in the presence of GM/CSF and IL-4 and then for 2 days in the presence of GM/CSF and TNF . The resulting cells exhibited the phenotype of mature DC, as assessed by the strong expression of HLA-DR, CD80, CD83 and CD86 antigens. We showed that trinitrophenyl (TNP)-treated mature DC induced a signi®cant T cell proliferative response in all experiments, while ¯uorescein isothiocyanate (FITC) gave positive results in about half of them. The prohaptens eugenol and isoeugenol induced signi®cant proliferation in one out of eight and in four out of 12 experiments, respectively. Interestingly, in 16 assays T cells never proliferated in the presence of sodium lauryl sulfate (SLS)-treated DC. Thus, this in vitro model allows discrimination between strong contact sensitizers and irritants. It might be very useful, therefore, for restriction of animal experimentation. # 2000 Elsevier Science Ltd. All rights reserved Keywords: contact hypersensitivity; haptens; monocyte-derived dendritic cells; T cell proliferation. Abbreviations: CHS=contact hypersensitivity; DC=dendritic cells; GM-CSF=granulocyte/macrophagestimulating factor; FITC=¯uorescein isothiocyanate; HBSS=Hanks' balanced salt solution; IL-4= interleukin 4; mAb=monoclonal antibody; LC=Langerhans cells; SLS=sodium lauryl sulfate; TNBS= 2,4,6-trinitrobenzene sulfonic acid; TNF=tumor necrosis factor; TNP=trinitrophenyl.
INTRODUCTION
Contact hypersensitivity (CHS) is a cell-mediated delayed-type reaction that occurs after repeated skin contact with small reactive molecules, the so-called contact allergens. There is now compelling evidence that Langerhans cells (LC), the dendritic cells (DC) from epidermis, play a key role in the elicitation phase of the reaction. Following primary application to the skin, LC trap allergens and then migrate to regional lymph nodes where they trigger speci®c T cell activation and proliferation (Kripke et al., 1990).
*Corresponding author. Fax: 04-72-11-02-90; e-mail: peguet @lyon151.inserm.fr
In vivo predictive tests are currently available to identify potential contact allergens, such as the guinea pig maximization test (Magnusson and Kligman, 1969), the mouse ear swelling test (Gad et al., 1986) and the local lymph node assay (Kimber and Weisenberger, 1989). Because of ethical objections to the use of laboratory animals, however, the interest in alternative methods, using full in vitro human assays, is considerable. Many biological assays have tried to reproduce in vitro the early events of CHS, that is, speci®c T cell activation by haptens. Hauser and Katz. (1990) ®rst used murine LC to induce in vitro primary sensitization of naive T cells to haptens. In previous papers, we reported that 2-day cultured human LC, but not freshly prepared LC, can induce in
0887-2333/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved. PII: S0887-2333(00)00043-6
518
G. Guironnet et al.
vitro primary proliferative T cell reactions to strong allergens such as trinitrophenyl (TNP) or ¯uorescein isothiocyanate (FITC), while an irritant such as sodium lauryl sulfate (SLS) failed to trigger any proliferative response (Krasteva et al., 1996; Moulon et al., 1993). DC are known to represent the most potent antigen-presenting cells for the elicitation of a primary T cell response (Steinman, 1991). A major advance over the past few years has been the development of methods for generating dendritic cells (DC) from hematopoietic progenitors or from peripheral monocytes. In humans, Caux et al. (1992) have demonstrated that GM-CSF and TNFa allowed the generation of DC from cord blood CD34+ progenitors. Some of these DC contain Birbeck granules, characteristics of LC. These in vitro generated DC can prime naive T cells from cord blood and induce a proliferative response to strong allergens (Rougier et al., 1998). Interestingly, however, this capacity was only acquired at the step of mature DC. A second way to generate DC is to culture peripheral blood monocytes with GM-CSF and IL-4. Cultured monocytes yield immature DC that can further mature in the presence of TNFa, LPS or CD40 ligand (Caux and Banchereau, 1996; Romani et al., 1994; Sallusto and Lanzavecchia, 1994). In the present study, we analyzed the capacity of mature DC derived from monocytes to sensitize human T cells to contact allergens. Signi®cant T cell proliferation could be obtained on stimulation by DC treated with strong allergens, while no response was obtained with an irritant. This in vitro system should be useful to discriminate potential allergens from irritants.
than 90% CD14+ cells, as assessed by ¯ow cytometry. Autologous T cells were recovered from the pellet of Percoll gradient and puri®ed by rosetting with 2amino-ethylisothiouronium bromide hydroxybromidetreated sheep red blood cells, as previously described (Kaplan and Clark, 1974). T cells were kept frozen until being used in the proliferative assay. Generation of DC from monocyte culture Monocytes (3 106 cells in 3 ml) were cultured for 6± 8 days in six-well tissue culture plates (Costar Corp., Cambridge, MA, USA), at 37 C in a humidi®ed atmosphere in the presence of 5% CO2. Complete RPMI medium was supplemented with recombinant human GM-CSF (500 U/ml, Sandoz, Basle, Switzerland) and IL-4 (50 U/ml, Schering-Plough, Dardilly, France). Every 2 or 3 days cells were fed by removing 2 ml medium and adding back 3 ml fresh medium with cytokines. Cells were then harvested, analysed, or transferred to new six-well plates and cultured for 2 further days in the presence of stimuli for maturation: 500 U/ml GM-CSF and 2.5 ng/ml TNFa (kindly provided by Dr G. Ghanem, Institiut Bordet, Brussels, Belgium). Immuno¯uorescence staining and ¯ow cytometry analysis
Culture medium was RPMI-1640 supplemented with 10% heat-inactivated foetal calf serum (FCS, Myoclone, GIBCO BRL Grand Island, NY, USA), 1 mm l-glutamine (GIBCO BRL) and 1% antibiotic antimycotic solution (Sigma, St Louis, MO, USA), called thereafter complete RPMI medium.
Phenotypic analysis was performed using the following monoclonal antibodies (mAbs): anti-HLA-DRFITC, B8.12.2, (IgG1); anti-CD54, 84H10, (IgG1); anti-CD83, HB15A, (IgG1); anti HLA-ABC, W6.32, (IgG2a); all from Immunotech, Marseille, France; anti-CD1a-FITC, BB5 (IgG1, Diaclone, BesancËon, France); anti-CD80-FITC, BB-1, anti-CD86, IT2.2, (IgG1, Pharmingen, San Diego, CA, USA); anti-CD40, G28.5, (IgG1, Schering-Plough); anti-CD14-FITC, TUÈK 4, (IgG1, Dako, Denmark); anti-CD32 AT10 (IgG1, Serotec). Cells were incubated for 20 min at 4 C with puri®ed mouse mAbs at the appropriate concentration or with control isotype-matched irrelevant mAb at the same concentration. For indirect staining, cells were washed and incubated for 20 min at 4 C with FITC-conjugated goat F(ab)0 2 anti-mouse Ig. Fluorescence analysis was performed using a FACScan (Becton Dickinson, Le Pont de Claix, France).
Cell preparation
Test chemicals
Mononuclear cells were obtained from peripheral blood of healthy donors by centrifugation on FicollHypaque (Pharmacia, St Quentin en Yvelines, France). They were then separated on multistep Percoll gradient (Pharmacia Fine Chemicals). The light density fraction from the 40±50% interface was recovered and monocytes were further puri®ed by magnetic depletion using anti-CD3, CD7, CD19, CD45RA, CD56, and anti-IgE antibodies (MACS; Miltenyi Biotec, Bergisch Gladbach, Germany) and a magnetic cell separator (MidiMACS) according to the manufacturer's instructions. The resulting suspension routinely contained more
2,4,6-Trinitrobenzene sulfonic acid (TNBS, 5 mm, Sigma) was used to generate trinitrophenlyl (TNP)conjugated proteins. The other allergens chosen for the study were ¯uorescein isothiocyanate (FITC), eugenol and isoeugenol all from Sigma. An irritant, sodium lauryl sulfate (SLS, Sigma) was used for control of hapten speci®city. All chemicals were used in aqueous solutions, except isoeugenol and eugenol, which were dissolved in dipropylene glycol and dimethyl sulfoxide, respectively. For all chemicals, the working concentrations included the highest nontoxic concentration.
MATERIALS AND METHODS
Culture medium
In vitro model for contact sensitization
To circumvent problems related to the toxicity of TNP and FITC, DC were haptenated before they were cultured with T cells. TNP modi®cation was performed according to the method of Shearer (1974). Cell pellets were resuspended in Hanks' balanced salt solution (HBSS) (pH 7.2) containing 5 mm TNP. Cells were incubated for 10 min at 37 C and then washed out extensively. FITC modi®cation was performed in the same way with HBSS containing 100 or 200 mg/ml of the hapten. DC viability was checked using the typan blue exclusion test. For SLS, eugenol and isoeugenol, preliminary experiments were carried out to determine the highest non-toxic concentrations. With this aim, DC were incubated overnight with increasing doses of the chemicals and cell viability was assessed using the typan blue exclusion test. Accordingly, SLS, eugenol and isoeugenol were added directly to DC-T co-cultures at the concentrations of 2.5, 5 and 10 mg/ml. In some experiments, DC were treated for 30 mn with SLS, eugenol or isoeugenol at the same concentration as FITC, that is, 200 mg/ml. Such treatment did not aect DC viability, as assessed by the typan blue exclusion test. Cells were then washed and added to the allogeneic T cells. Lymphoproliferative assays T cells and monocyte-derived DC were co-cultured in round-bottomed microtitre plates (105 autologous T cells and graded numbers of dendritic cells in a ®nal volume of 200 ml complete RPMI medium per well) at 37 C in a humidi®ed atmosphere of 5% CO2. Triplicate cultures for each allergen concentration were maintained for 4±6 days. A kinetic study of T cell proliferation was performed by pulsing the cells with 1 mCi 3H-thymidine for the ®nal 18 hr of culture. Results from triplicate wells were expressed as mean cpm SD. In most experiments, SD did not exceed 10±20% of the mean values. Antigen-speci®c stimulation was calculated as proliferation indexes (PI=uptake of isotope in stimulated culture/uptake of isotope in non-stimulated culture). A PI 5 2, which refers to cpm values with non-overlapping errors bars, was considered as the criterion for the induction of allergen-speci®c T cell sensitization.
RESULTS
DC phenotype We ®rst compared the phenotype of monocytederived DC before and after addition of the maturation stimulus, i.e. TNFa. As previously described (Romani et al., 1994; Sallusto and Lanzavecchia, 1994), after a 6-day culture in the presence of GM-CSF and IL-4, the cells appeared as non-adherent cells with a typical dendritic morphology (data not shown). They expressed high levels of CD1a, HLA-DR, HLA class I, CD32, CD40 and CD54 antigens, while
519
only low levels of CD14, CD80, CD86 and CD83 could be detected at the cell surface (Fig. 1). Following addition of TNFa, the cells acquired the phenotype of mature DC, as evidenced by the strong up-regulation of HLA-DR, CD40, CD54, CD80 and CD86 expression. Furthermore, most of the cells expressed CD83, the speci®c marker of mature DC. Primary T-cell response to TNP-modi®ed mature DC TNP was chosen to determine the optimal conditions for in vitro sensitization of T cells by haptentreated mature DC. To this end, the kinetics of the proliferative T-cell response induced by a graded number of TNP-modi®ed DC was analyzed. As shown in Fig. 2, which was representative of three experiments, the addition of TNP-modi®ed DC to autologous T cells induced a signi®cant T cell response at all days of culture. The highest proliferation index was recorded at day 5, by adding 103 TNP-treated DC to 105 T cells. Indeed, in most cases, increasing DC numbers to 2103 enhanced the autologous T cell response and led, therefore, to lower proliferation indexes. Similar results were obtained using 4103 or 8103 DC per well (not shown).Very interestingly, SLS did not induce any T cell response at any time of culture. In some experiments, DC have been generated from monocytes and some of them were kept frozen. The proliferation index induced by the thawed, then TNP-treated DC was quite similar to that obtained with the fresh DC (data not shown). In vitro primary T-cell responses to autologous monocyte-derived DC modi®ed by various chemicals We studied the T cell response to other chemicals by applying the conditions in which we had obtained the highest response to TNP, i.e. after 5 days of culture using 103 or 2 103 hapten-treated DC. The calculation of proliferation indexes makes possible the comparison of the proliferative T-cell response induced by the dierent haptens at dierent concentrations in several experiments carried out with dierent donors. We have shown in Table 1 the highest proliferation index recorded within a single experiment. A proliferation index of two or more, was considered as the criterion for the induction of T cell sensitization. The highest proliferation indexes were always recorded with TNP and, using TNPtreated DC, signi®cant proliferation was obtained in 16 out of 16 experiments. The other highly sensitizing chemical FITC gave positive responses in about half of experiments. It should be noted that the levels of proliferation to the strong allergens varied largely according to the donors. When added directly in the DC-T cell co-culture, the prohaptens eugenol and isoeugenol induced a signi®cant proliferative response in one out of eight and four out of 12 experiments, respectively. By contrast, addition of SLS did not induce signi®cant T cell proliferation in the 16 experiments.
520 G. Guironnet et al.
Fig. 1. Comparative phenotype of immature vs mature monocyte-derived DC. Puri®ed peripheral blood monocytes were cultured for 6 days in the presence of GM/CSF and IL-4 (immature DC, white histogram) and for further 2 days in the presence of GM/CSF and TNFa (mature DC, black histogram). Phenotypic analysis was carried out by staining with a panel of monoclonal antibodies and ¯uorescence was analysed on a FACScan.
In vitro model for contact sensitization
521
In ®ve additional experiments, DC were pretreated with SLS, isoeugenol or eugenol at 200 mg/ml and washed before being added to the T cells. Such treatment did not aect DC viability, as assessed by the trypan blue excusion test. In these conditions, however, no signi®cant T cell responses were obtained with any of the chemicals (data not shown).
DISCUSSION
Fig. 2. Kinetic analysis of the T cell proliferative response induced by TNP-treated monocyte-derived DC. T lymphocytes (105 cells) were cocultured with TNP or SLS-treated mature DC (1000 or 2000 DC/well). T cell proliferation was assessed at days 4, 5 or 6 by 3H-thymidine incorporation for the ®nal 18 hr of culture. Results are expressed as the mean cpm SD of triplicate cultures. The proliferation indexes (PI) for TNP are indicated on the graph (PI=uptake of isotope in hapten-stimulated culture/uptake of isotope in non-stimulated control culture). Table 1. In vitro T cell proliferative response induced by monocytederived mature DC treated with dierent chemicals Experiments 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 a
TNP a
22.1 4.4 2.5 2.0 149 3.5 5.0 2.9 2.3 2.9 2.3 2.2 17.0 3.8 12.2 44.0
FITC
SLS
Eugenol
Isoeugenol
2.4 3.7 nd nd 105 1.6 nd nd 0.9 nd 0.9 1.1 1.0 1.25 2.2 2.2
0.8 1.2 1.0 0.6 1.0 1.1 1.0 1.0 0.9 1.0 0.9 1.0 1.2 1.1 0.85 1.2
3.5 1.8 1.4 1.3 nd nd nd nd nd nd nd nd 0.9 1.1 0.9 1.5
nd nd nd nd 0.9 2.3 1.0 1.1 1.0 1.1 1.0 1.0 2.5 3.0 2.0 1.6
Allergen-speci®c T cell proliferation was calculated as proliferation indexes (PI=uptake of isotope in stimulated culture/uptake of isotope in non-stimulated culture). Experiments have been carried out using cells from 16 dierent donors. In each of them, 1 103 and 2 103 DC were added to the T cells after treatment with dierent concentrations of allergens and the highest PI was recorded.
Penetration of an allergen into the skin results in the clonal expansion of allergen-speci®c T cells, a process which is measured in the in vivo local lymph node assay (Kimber and Weisenberger, 1989). Using mature LC as antigen-presenting cells, we previously reported that induction of in vitro T cell proliferation provides an alternative approach to discriminate strong contact sensitizers from irritants (Krasteva et al., 1996; Moulon et al., 1993). This human LCbased immunoassay was limited, however, by the availability of this cell type. Therefore, in the present study, we used monocytederived DC, as LC equivalents, to analyse the proliferative T cell responses to several allergens. We showed that in all experiments, a signi®cant T cell proliferative response was obtained using TNP-treated mature DC, but not SLS-treated DC. FITC gave positive results in about half of the experiments. The large variability in the proliferative response to strong chemicals could not be related to the culture technique. Indeed, similar proliferation indexes could be repeatedly obtained using frozen DC and T cells. Thus, the response level to the strong allergens was related to the donors, revealing the existence of highly responsive, as opposed to little responsive donors. Peripheral blood has been recognized as the most useful source to generate DC. A large number of DC can be generated from peripheral blood monocytes following culture with GM-CSF and IL-4. In agreement with many authors (Romani et al., 1994; Sallusto and Lanzavecchia, 1994), we found that these cells display a characteristic dendritic cell phenotype, as evidenced by the expression of HLA-DR, CD1a, CD40 and CD80 expression. In addition, further cultivation in the presence of GM/CSF and TNFa resulted in DC maturation which is characterized by upregulation of co-stimulatory molecules and CD83 expression. Previous experiments have suggested that maturation of the antigen-presenting cell is a necessary condition for T cell sensitization. Indeed, in previous experiments we showed that 2-day cultured LC, but not freshly isolated LC, can induce T cell proliferation to allergens (Moulon et al., 1993). Furthermore, DC derived from cord blood CD34+ cells were ecient in T cell priming only when they have acquired the mature DC phenotype (Rougier et al., 1988). Upregulation of co-stimulatory molecules on mature DC is the most probable explanation for
522
G. Guironnet et al.
these results. In the present study, we showed that mature DC could induce in vitro T cell proliferation to strong allergens. However, since immature DC possess a higher capacity for antigen capturing and processing, it would be interesting to add allergens to DC during their maturation step. Further experiments are in progress to ascertain whether this could improve the results. Eugenol and isoeugenol are perfume chemicals that appear structurally very similar. The compounds were prohaptens and the metabolic pathways which leads to immunogenic metabolites are not completely understood (Bertrand et al., 1997). In the present study, isoeugenol and eugenol induced signi®cant T cell responses in four out of 12 and in only one out of eight experiments, respectively. The low frequency of positive responses, especially for eugenol, might re¯ect a limitation of our model, inasmuch as it could not reproduce the metabolization of prohaptens which normally takes place within epidermis. Alternatively, and although the number of experiments may be too limited, the results might be relevant to in vivo ®ndings showing dierent sensitizing properties of the two chemicals. Indeed, based on the guinea pig sensitization test (Barratt and Basketter, 1992), as well as human clinical studies (Marzulli and Maibach, 1980), isoeugenol appears as a stronger sensitizer than eugenol. SLS, which is exclusively an irritant in human pathology, has been widely used as the negative control for the validation of a number of animal sensitizing experiments. While it never induces CHS reactions in the guinea pig model (Magnusson and Kligman, 1969), it was shown to induce false positive results in the local lymph node assay (Basketter et al., 1994; Montelius et al., 1994). An interesting ®nding in this study was that SLS did not induce signi®cant T cell response in any experiments. So, the present in vitro assay did not overestimate the sensitizing potential of irritant substances. In conclusion, the present in vitro model of T cell priming using monocyte-derived autologous DC could dierentiate strong allergens from irritants. It might be therefore very useful for restriction of animal experimentation. REFERENCES
Barratt M. D. and Basketter D. A. (1992) Possible origin of the sensitization potential of isoeugenol and related compounds. 1: Preliminary studies of potential reaction mechanisms. Contact Dermatitis 27, 98±104. Basketter D. A., Scholes E. W. and Kimber I. (1994) The performance of the local lymph node assay with chemicals identi®ed as contact allergens in the human maximization test. Food and Chemical Toxicology 32, 543± 547. Bertrand F., Basketter D. A., Roberts D. W. and Lepoitevin J. P. (1997) Skin sensitization to eugenol and isoeugenol in mice: possible metabolic pathways involving orthoquinone and quinone methide intermediates. Chemical Research in Toxicology 10, 335±343.
Caux C. and Banchereau J. (1996) In vitro regulation of dendritic cell development and function. In Blood Cell Chemistry: Hematotopoieic Cell Growth Factors and their Receptors, ed. A. D. Whetton and J. Gordon, pp. 263± 301. Plenum Press, New-York. Caux C., Dezutter-Dambuyant C., Schmitt D. and Banchereau J. (1992) GM-CSF and TNFa cooperate in the generation of dendritic Langerhans cells. Nature 360, 258±261. Gad S. C., Dunn B. J., Dobbs D. W., Reilly C. and Walsh R. D. (1986) Development and validation of an alternative dermal sensitization test: the mouse ear swelling test. Toxicology and Applied Pharmacology 84, 93±114. Hauser C. and Katz S. I. (1990) Generation and characterization of T-helper cells by primary in vitro sensitization using Langerhans cells. Immunological Review 117, 67±71. Kaplan M. E. and Clark C. (1974) An improved rosetting assay for detection of human T lymphocytes. Journal of Immunological Methods 5, 131±136. Kimber I. and Weisenberger C. (1989) A murine local lymph node assay for the identi®cation of contact allergens. Assay development and results of an initial validation study. Archives of Toxicology 63, 274±282. Krasteva M., PeÂguet-Navarro J., Moulon C., Courtellemont P., Redziniak G. and Schmitt D. (1996) In vitro primary sensitization of hapten-speci®c T cells by cultured human epidermal Langerhans cells Ð a screening predictive assay for contact sensitizers. Clinical and Experimental Allergy 26, 563±570. Kripke M. L., Munn C. G., Jeevan A., Tang J. M. and Bucana C. (1990) Evidence that cutaneous antigen-presenting cells migrate to regional lymph nodes during contact sensitization. Journal of Immunology 145, 2833±2838. Magnusson B. and Kligman A. M. (1969) The identi®cation of contact allergens by animal assay. The guinea pig maximization test. Journal of Investigative Dermatology 52, 268±276. Marzulli F. and Maibach H. I. (1980) Contact allergy. Predictive testing of ¯agrance ingredients in humans by Draize and maximization methods. Journal of Environmental Pathology and Toxicology 3, 235±245. Montelius J., Wahlquist H., Boman A., FernstroÈm P., Grabergs L. and Wahlberg J. E. (1994) Experience with the murine local lymph node assay: inability to discriminate between allergens and irritants. Acta Dermatologica Venereologica 74, 22±27. Moulon C., PeÂguet-Navarro J., Courtellemont P., Redziniak G. and Schmitt D. (1993) In vitro primary sensitization and restimulation of hapten-speci®c T cells by fresh and cultured human epidermal Langerhans cells. Immunology 80, 373±379. Romani N., Gruner S., Brang D., KaÈmpgen E., Lenz A., Trockenbacher B., Konwalinka G., Fritsch P. O., Steinman R. M. and Schuler G. (1994) Proliferating dendritic cell progenitors in human blood. Journal of Experimental Medicine 180, 83±93. Rougier N., Redziniak G., Schmitt D. and Vincent C. (1998) Evaluation of the capacity of dendritic cells derived from cord blood CD34+ precursors to present haptens to unsensitized autologous T cells in vitro. The Journal of Investigative Dermatology 110, 348±352. Sallusto F. and Lanzavecchia A. (1994) Ecient presentation of soluble antigen by cultured human dendritic cells is maintained by granulocyte/macrophage-colony stimulating factor plus interleukin-4 and down-regulated by tumar necrosis factor a. Journal of Experimental Medicine 179, 1109±1118. Shearer G. M. (1974) Cell-mediated cytotoxicity to trinitrophenyl-modi®ed syngeneic lymphocytes. European Journal of Immunology 4, 527±533. Steinman R. M. (1991) The dendritic cell system and its role in immunogenicity. Annual Review of Immunology 9, 271± 296.
www.elsevier.com/locate/toxinvit
In vitro human T cell sensitization to haptens by monocyte-derived dendritic cells G. GUIRONNET 1, C. DALBIEZ-GAUTHIER 1, F. ROUSSET 2, D. SCHMITT 1 and J. PEÂGUET-NAVARRO 1,* 1
INSERM U346, Pavillon R, HoÃpital E. Herriot, 69374, Lyon 03, and 2L'OreÂal, 1 avenue EugeÁne Schueller, 93190, Aulnay/Bois, France (Accepted 27 April 2000)
AbstractÐWe previously reported that in vitro primary sensitization of hapten-speci®c T cells by cultured human epidermal Langerhans cells (LC) provides an alternative approach to discriminate strong contact sensitizers from irritants (Krasteva et al., 1996; Moulon et al., 1993). However, this LC-based immunoassay was limited by the availability of human skin samples. In the present study, we used monocyte-derived dendritic cells (DC) to analyse the autologous proliferative T cell response to several allergens. Monocytes were puri®ed from the peripheral blood of healthy donors and cultured for 6±8 days in the presence of GM/CSF and IL-4 and then for 2 days in the presence of GM/CSF and TNF . The resulting cells exhibited the phenotype of mature DC, as assessed by the strong expression of HLA-DR, CD80, CD83 and CD86 antigens. We showed that trinitrophenyl (TNP)-treated mature DC induced a signi®cant T cell proliferative response in all experiments, while ¯uorescein isothiocyanate (FITC) gave positive results in about half of them. The prohaptens eugenol and isoeugenol induced signi®cant proliferation in one out of eight and in four out of 12 experiments, respectively. Interestingly, in 16 assays T cells never proliferated in the presence of sodium lauryl sulfate (SLS)-treated DC. Thus, this in vitro model allows discrimination between strong contact sensitizers and irritants. It might be very useful, therefore, for restriction of animal experimentation. # 2000 Elsevier Science Ltd. All rights reserved Keywords: contact hypersensitivity; haptens; monocyte-derived dendritic cells; T cell proliferation. Abbreviations: CHS=contact hypersensitivity; DC=dendritic cells; GM-CSF=granulocyte/macrophagestimulating factor; FITC=¯uorescein isothiocyanate; HBSS=Hanks' balanced salt solution; IL-4= interleukin 4; mAb=monoclonal antibody; LC=Langerhans cells; SLS=sodium lauryl sulfate; TNBS= 2,4,6-trinitrobenzene sulfonic acid; TNF=tumor necrosis factor; TNP=trinitrophenyl.
INTRODUCTION
Contact hypersensitivity (CHS) is a cell-mediated delayed-type reaction that occurs after repeated skin contact with small reactive molecules, the so-called contact allergens. There is now compelling evidence that Langerhans cells (LC), the dendritic cells (DC) from epidermis, play a key role in the elicitation phase of the reaction. Following primary application to the skin, LC trap allergens and then migrate to regional lymph nodes where they trigger speci®c T cell activation and proliferation (Kripke et al., 1990).
*Corresponding author. Fax: 04-72-11-02-90; e-mail: peguet @lyon151.inserm.fr
In vivo predictive tests are currently available to identify potential contact allergens, such as the guinea pig maximization test (Magnusson and Kligman, 1969), the mouse ear swelling test (Gad et al., 1986) and the local lymph node assay (Kimber and Weisenberger, 1989). Because of ethical objections to the use of laboratory animals, however, the interest in alternative methods, using full in vitro human assays, is considerable. Many biological assays have tried to reproduce in vitro the early events of CHS, that is, speci®c T cell activation by haptens. Hauser and Katz. (1990) ®rst used murine LC to induce in vitro primary sensitization of naive T cells to haptens. In previous papers, we reported that 2-day cultured human LC, but not freshly prepared LC, can induce in
0887-2333/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved. PII: S0887-2333(00)00043-6
518
G. Guironnet et al.
vitro primary proliferative T cell reactions to strong allergens such as trinitrophenyl (TNP) or ¯uorescein isothiocyanate (FITC), while an irritant such as sodium lauryl sulfate (SLS) failed to trigger any proliferative response (Krasteva et al., 1996; Moulon et al., 1993). DC are known to represent the most potent antigen-presenting cells for the elicitation of a primary T cell response (Steinman, 1991). A major advance over the past few years has been the development of methods for generating dendritic cells (DC) from hematopoietic progenitors or from peripheral monocytes. In humans, Caux et al. (1992) have demonstrated that GM-CSF and TNFa allowed the generation of DC from cord blood CD34+ progenitors. Some of these DC contain Birbeck granules, characteristics of LC. These in vitro generated DC can prime naive T cells from cord blood and induce a proliferative response to strong allergens (Rougier et al., 1998). Interestingly, however, this capacity was only acquired at the step of mature DC. A second way to generate DC is to culture peripheral blood monocytes with GM-CSF and IL-4. Cultured monocytes yield immature DC that can further mature in the presence of TNFa, LPS or CD40 ligand (Caux and Banchereau, 1996; Romani et al., 1994; Sallusto and Lanzavecchia, 1994). In the present study, we analyzed the capacity of mature DC derived from monocytes to sensitize human T cells to contact allergens. Signi®cant T cell proliferation could be obtained on stimulation by DC treated with strong allergens, while no response was obtained with an irritant. This in vitro system should be useful to discriminate potential allergens from irritants.
than 90% CD14+ cells, as assessed by ¯ow cytometry. Autologous T cells were recovered from the pellet of Percoll gradient and puri®ed by rosetting with 2amino-ethylisothiouronium bromide hydroxybromidetreated sheep red blood cells, as previously described (Kaplan and Clark, 1974). T cells were kept frozen until being used in the proliferative assay. Generation of DC from monocyte culture Monocytes (3 106 cells in 3 ml) were cultured for 6± 8 days in six-well tissue culture plates (Costar Corp., Cambridge, MA, USA), at 37 C in a humidi®ed atmosphere in the presence of 5% CO2. Complete RPMI medium was supplemented with recombinant human GM-CSF (500 U/ml, Sandoz, Basle, Switzerland) and IL-4 (50 U/ml, Schering-Plough, Dardilly, France). Every 2 or 3 days cells were fed by removing 2 ml medium and adding back 3 ml fresh medium with cytokines. Cells were then harvested, analysed, or transferred to new six-well plates and cultured for 2 further days in the presence of stimuli for maturation: 500 U/ml GM-CSF and 2.5 ng/ml TNFa (kindly provided by Dr G. Ghanem, Institiut Bordet, Brussels, Belgium). Immuno¯uorescence staining and ¯ow cytometry analysis
Culture medium was RPMI-1640 supplemented with 10% heat-inactivated foetal calf serum (FCS, Myoclone, GIBCO BRL Grand Island, NY, USA), 1 mm l-glutamine (GIBCO BRL) and 1% antibiotic antimycotic solution (Sigma, St Louis, MO, USA), called thereafter complete RPMI medium.
Phenotypic analysis was performed using the following monoclonal antibodies (mAbs): anti-HLA-DRFITC, B8.12.2, (IgG1); anti-CD54, 84H10, (IgG1); anti-CD83, HB15A, (IgG1); anti HLA-ABC, W6.32, (IgG2a); all from Immunotech, Marseille, France; anti-CD1a-FITC, BB5 (IgG1, Diaclone, BesancËon, France); anti-CD80-FITC, BB-1, anti-CD86, IT2.2, (IgG1, Pharmingen, San Diego, CA, USA); anti-CD40, G28.5, (IgG1, Schering-Plough); anti-CD14-FITC, TUÈK 4, (IgG1, Dako, Denmark); anti-CD32 AT10 (IgG1, Serotec). Cells were incubated for 20 min at 4 C with puri®ed mouse mAbs at the appropriate concentration or with control isotype-matched irrelevant mAb at the same concentration. For indirect staining, cells were washed and incubated for 20 min at 4 C with FITC-conjugated goat F(ab)0 2 anti-mouse Ig. Fluorescence analysis was performed using a FACScan (Becton Dickinson, Le Pont de Claix, France).
Cell preparation
Test chemicals
Mononuclear cells were obtained from peripheral blood of healthy donors by centrifugation on FicollHypaque (Pharmacia, St Quentin en Yvelines, France). They were then separated on multistep Percoll gradient (Pharmacia Fine Chemicals). The light density fraction from the 40±50% interface was recovered and monocytes were further puri®ed by magnetic depletion using anti-CD3, CD7, CD19, CD45RA, CD56, and anti-IgE antibodies (MACS; Miltenyi Biotec, Bergisch Gladbach, Germany) and a magnetic cell separator (MidiMACS) according to the manufacturer's instructions. The resulting suspension routinely contained more
2,4,6-Trinitrobenzene sulfonic acid (TNBS, 5 mm, Sigma) was used to generate trinitrophenlyl (TNP)conjugated proteins. The other allergens chosen for the study were ¯uorescein isothiocyanate (FITC), eugenol and isoeugenol all from Sigma. An irritant, sodium lauryl sulfate (SLS, Sigma) was used for control of hapten speci®city. All chemicals were used in aqueous solutions, except isoeugenol and eugenol, which were dissolved in dipropylene glycol and dimethyl sulfoxide, respectively. For all chemicals, the working concentrations included the highest nontoxic concentration.
MATERIALS AND METHODS
Culture medium
In vitro model for contact sensitization
To circumvent problems related to the toxicity of TNP and FITC, DC were haptenated before they were cultured with T cells. TNP modi®cation was performed according to the method of Shearer (1974). Cell pellets were resuspended in Hanks' balanced salt solution (HBSS) (pH 7.2) containing 5 mm TNP. Cells were incubated for 10 min at 37 C and then washed out extensively. FITC modi®cation was performed in the same way with HBSS containing 100 or 200 mg/ml of the hapten. DC viability was checked using the typan blue exclusion test. For SLS, eugenol and isoeugenol, preliminary experiments were carried out to determine the highest non-toxic concentrations. With this aim, DC were incubated overnight with increasing doses of the chemicals and cell viability was assessed using the typan blue exclusion test. Accordingly, SLS, eugenol and isoeugenol were added directly to DC-T co-cultures at the concentrations of 2.5, 5 and 10 mg/ml. In some experiments, DC were treated for 30 mn with SLS, eugenol or isoeugenol at the same concentration as FITC, that is, 200 mg/ml. Such treatment did not aect DC viability, as assessed by the typan blue exclusion test. Cells were then washed and added to the allogeneic T cells. Lymphoproliferative assays T cells and monocyte-derived DC were co-cultured in round-bottomed microtitre plates (105 autologous T cells and graded numbers of dendritic cells in a ®nal volume of 200 ml complete RPMI medium per well) at 37 C in a humidi®ed atmosphere of 5% CO2. Triplicate cultures for each allergen concentration were maintained for 4±6 days. A kinetic study of T cell proliferation was performed by pulsing the cells with 1 mCi 3H-thymidine for the ®nal 18 hr of culture. Results from triplicate wells were expressed as mean cpm SD. In most experiments, SD did not exceed 10±20% of the mean values. Antigen-speci®c stimulation was calculated as proliferation indexes (PI=uptake of isotope in stimulated culture/uptake of isotope in non-stimulated culture). A PI 5 2, which refers to cpm values with non-overlapping errors bars, was considered as the criterion for the induction of allergen-speci®c T cell sensitization.
RESULTS
DC phenotype We ®rst compared the phenotype of monocytederived DC before and after addition of the maturation stimulus, i.e. TNFa. As previously described (Romani et al., 1994; Sallusto and Lanzavecchia, 1994), after a 6-day culture in the presence of GM-CSF and IL-4, the cells appeared as non-adherent cells with a typical dendritic morphology (data not shown). They expressed high levels of CD1a, HLA-DR, HLA class I, CD32, CD40 and CD54 antigens, while
519
only low levels of CD14, CD80, CD86 and CD83 could be detected at the cell surface (Fig. 1). Following addition of TNFa, the cells acquired the phenotype of mature DC, as evidenced by the strong up-regulation of HLA-DR, CD40, CD54, CD80 and CD86 expression. Furthermore, most of the cells expressed CD83, the speci®c marker of mature DC. Primary T-cell response to TNP-modi®ed mature DC TNP was chosen to determine the optimal conditions for in vitro sensitization of T cells by haptentreated mature DC. To this end, the kinetics of the proliferative T-cell response induced by a graded number of TNP-modi®ed DC was analyzed. As shown in Fig. 2, which was representative of three experiments, the addition of TNP-modi®ed DC to autologous T cells induced a signi®cant T cell response at all days of culture. The highest proliferation index was recorded at day 5, by adding 103 TNP-treated DC to 105 T cells. Indeed, in most cases, increasing DC numbers to 2103 enhanced the autologous T cell response and led, therefore, to lower proliferation indexes. Similar results were obtained using 4103 or 8103 DC per well (not shown).Very interestingly, SLS did not induce any T cell response at any time of culture. In some experiments, DC have been generated from monocytes and some of them were kept frozen. The proliferation index induced by the thawed, then TNP-treated DC was quite similar to that obtained with the fresh DC (data not shown). In vitro primary T-cell responses to autologous monocyte-derived DC modi®ed by various chemicals We studied the T cell response to other chemicals by applying the conditions in which we had obtained the highest response to TNP, i.e. after 5 days of culture using 103 or 2 103 hapten-treated DC. The calculation of proliferation indexes makes possible the comparison of the proliferative T-cell response induced by the dierent haptens at dierent concentrations in several experiments carried out with dierent donors. We have shown in Table 1 the highest proliferation index recorded within a single experiment. A proliferation index of two or more, was considered as the criterion for the induction of T cell sensitization. The highest proliferation indexes were always recorded with TNP and, using TNPtreated DC, signi®cant proliferation was obtained in 16 out of 16 experiments. The other highly sensitizing chemical FITC gave positive responses in about half of experiments. It should be noted that the levels of proliferation to the strong allergens varied largely according to the donors. When added directly in the DC-T cell co-culture, the prohaptens eugenol and isoeugenol induced a signi®cant proliferative response in one out of eight and four out of 12 experiments, respectively. By contrast, addition of SLS did not induce signi®cant T cell proliferation in the 16 experiments.
520 G. Guironnet et al.
Fig. 1. Comparative phenotype of immature vs mature monocyte-derived DC. Puri®ed peripheral blood monocytes were cultured for 6 days in the presence of GM/CSF and IL-4 (immature DC, white histogram) and for further 2 days in the presence of GM/CSF and TNFa (mature DC, black histogram). Phenotypic analysis was carried out by staining with a panel of monoclonal antibodies and ¯uorescence was analysed on a FACScan.
In vitro model for contact sensitization
521
In ®ve additional experiments, DC were pretreated with SLS, isoeugenol or eugenol at 200 mg/ml and washed before being added to the T cells. Such treatment did not aect DC viability, as assessed by the trypan blue excusion test. In these conditions, however, no signi®cant T cell responses were obtained with any of the chemicals (data not shown).
DISCUSSION
Fig. 2. Kinetic analysis of the T cell proliferative response induced by TNP-treated monocyte-derived DC. T lymphocytes (105 cells) were cocultured with TNP or SLS-treated mature DC (1000 or 2000 DC/well). T cell proliferation was assessed at days 4, 5 or 6 by 3H-thymidine incorporation for the ®nal 18 hr of culture. Results are expressed as the mean cpm SD of triplicate cultures. The proliferation indexes (PI) for TNP are indicated on the graph (PI=uptake of isotope in hapten-stimulated culture/uptake of isotope in non-stimulated control culture). Table 1. In vitro T cell proliferative response induced by monocytederived mature DC treated with dierent chemicals Experiments 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 a
TNP a
22.1 4.4 2.5 2.0 149 3.5 5.0 2.9 2.3 2.9 2.3 2.2 17.0 3.8 12.2 44.0
FITC
SLS
Eugenol
Isoeugenol
2.4 3.7 nd nd 105 1.6 nd nd 0.9 nd 0.9 1.1 1.0 1.25 2.2 2.2
0.8 1.2 1.0 0.6 1.0 1.1 1.0 1.0 0.9 1.0 0.9 1.0 1.2 1.1 0.85 1.2
3.5 1.8 1.4 1.3 nd nd nd nd nd nd nd nd 0.9 1.1 0.9 1.5
nd nd nd nd 0.9 2.3 1.0 1.1 1.0 1.1 1.0 1.0 2.5 3.0 2.0 1.6
Allergen-speci®c T cell proliferation was calculated as proliferation indexes (PI=uptake of isotope in stimulated culture/uptake of isotope in non-stimulated culture). Experiments have been carried out using cells from 16 dierent donors. In each of them, 1 103 and 2 103 DC were added to the T cells after treatment with dierent concentrations of allergens and the highest PI was recorded.
Penetration of an allergen into the skin results in the clonal expansion of allergen-speci®c T cells, a process which is measured in the in vivo local lymph node assay (Kimber and Weisenberger, 1989). Using mature LC as antigen-presenting cells, we previously reported that induction of in vitro T cell proliferation provides an alternative approach to discriminate strong contact sensitizers from irritants (Krasteva et al., 1996; Moulon et al., 1993). This human LCbased immunoassay was limited, however, by the availability of this cell type. Therefore, in the present study, we used monocytederived DC, as LC equivalents, to analyse the proliferative T cell responses to several allergens. We showed that in all experiments, a signi®cant T cell proliferative response was obtained using TNP-treated mature DC, but not SLS-treated DC. FITC gave positive results in about half of the experiments. The large variability in the proliferative response to strong chemicals could not be related to the culture technique. Indeed, similar proliferation indexes could be repeatedly obtained using frozen DC and T cells. Thus, the response level to the strong allergens was related to the donors, revealing the existence of highly responsive, as opposed to little responsive donors. Peripheral blood has been recognized as the most useful source to generate DC. A large number of DC can be generated from peripheral blood monocytes following culture with GM-CSF and IL-4. In agreement with many authors (Romani et al., 1994; Sallusto and Lanzavecchia, 1994), we found that these cells display a characteristic dendritic cell phenotype, as evidenced by the expression of HLA-DR, CD1a, CD40 and CD80 expression. In addition, further cultivation in the presence of GM/CSF and TNFa resulted in DC maturation which is characterized by upregulation of co-stimulatory molecules and CD83 expression. Previous experiments have suggested that maturation of the antigen-presenting cell is a necessary condition for T cell sensitization. Indeed, in previous experiments we showed that 2-day cultured LC, but not freshly isolated LC, can induce T cell proliferation to allergens (Moulon et al., 1993). Furthermore, DC derived from cord blood CD34+ cells were ecient in T cell priming only when they have acquired the mature DC phenotype (Rougier et al., 1988). Upregulation of co-stimulatory molecules on mature DC is the most probable explanation for
522
G. Guironnet et al.
these results. In the present study, we showed that mature DC could induce in vitro T cell proliferation to strong allergens. However, since immature DC possess a higher capacity for antigen capturing and processing, it would be interesting to add allergens to DC during their maturation step. Further experiments are in progress to ascertain whether this could improve the results. Eugenol and isoeugenol are perfume chemicals that appear structurally very similar. The compounds were prohaptens and the metabolic pathways which leads to immunogenic metabolites are not completely understood (Bertrand et al., 1997). In the present study, isoeugenol and eugenol induced signi®cant T cell responses in four out of 12 and in only one out of eight experiments, respectively. The low frequency of positive responses, especially for eugenol, might re¯ect a limitation of our model, inasmuch as it could not reproduce the metabolization of prohaptens which normally takes place within epidermis. Alternatively, and although the number of experiments may be too limited, the results might be relevant to in vivo ®ndings showing dierent sensitizing properties of the two chemicals. Indeed, based on the guinea pig sensitization test (Barratt and Basketter, 1992), as well as human clinical studies (Marzulli and Maibach, 1980), isoeugenol appears as a stronger sensitizer than eugenol. SLS, which is exclusively an irritant in human pathology, has been widely used as the negative control for the validation of a number of animal sensitizing experiments. While it never induces CHS reactions in the guinea pig model (Magnusson and Kligman, 1969), it was shown to induce false positive results in the local lymph node assay (Basketter et al., 1994; Montelius et al., 1994). An interesting ®nding in this study was that SLS did not induce signi®cant T cell response in any experiments. So, the present in vitro assay did not overestimate the sensitizing potential of irritant substances. In conclusion, the present in vitro model of T cell priming using monocyte-derived autologous DC could dierentiate strong allergens from irritants. It might be therefore very useful for restriction of animal experimentation. REFERENCES
Barratt M. D. and Basketter D. A. (1992) Possible origin of the sensitization potential of isoeugenol and related compounds. 1: Preliminary studies of potential reaction mechanisms. Contact Dermatitis 27, 98±104. Basketter D. A., Scholes E. W. and Kimber I. (1994) The performance of the local lymph node assay with chemicals identi®ed as contact allergens in the human maximization test. Food and Chemical Toxicology 32, 543± 547. Bertrand F., Basketter D. A., Roberts D. W. and Lepoitevin J. P. (1997) Skin sensitization to eugenol and isoeugenol in mice: possible metabolic pathways involving orthoquinone and quinone methide intermediates. Chemical Research in Toxicology 10, 335±343.
Caux C. and Banchereau J. (1996) In vitro regulation of dendritic cell development and function. In Blood Cell Chemistry: Hematotopoieic Cell Growth Factors and their Receptors, ed. A. D. Whetton and J. Gordon, pp. 263± 301. Plenum Press, New-York. Caux C., Dezutter-Dambuyant C., Schmitt D. and Banchereau J. (1992) GM-CSF and TNFa cooperate in the generation of dendritic Langerhans cells. Nature 360, 258±261. Gad S. C., Dunn B. J., Dobbs D. W., Reilly C. and Walsh R. D. (1986) Development and validation of an alternative dermal sensitization test: the mouse ear swelling test. Toxicology and Applied Pharmacology 84, 93±114. Hauser C. and Katz S. I. (1990) Generation and characterization of T-helper cells by primary in vitro sensitization using Langerhans cells. Immunological Review 117, 67±71. Kaplan M. E. and Clark C. (1974) An improved rosetting assay for detection of human T lymphocytes. Journal of Immunological Methods 5, 131±136. Kimber I. and Weisenberger C. (1989) A murine local lymph node assay for the identi®cation of contact allergens. Assay development and results of an initial validation study. Archives of Toxicology 63, 274±282. Krasteva M., PeÂguet-Navarro J., Moulon C., Courtellemont P., Redziniak G. and Schmitt D. (1996) In vitro primary sensitization of hapten-speci®c T cells by cultured human epidermal Langerhans cells Ð a screening predictive assay for contact sensitizers. Clinical and Experimental Allergy 26, 563±570. Kripke M. L., Munn C. G., Jeevan A., Tang J. M. and Bucana C. (1990) Evidence that cutaneous antigen-presenting cells migrate to regional lymph nodes during contact sensitization. Journal of Immunology 145, 2833±2838. Magnusson B. and Kligman A. M. (1969) The identi®cation of contact allergens by animal assay. The guinea pig maximization test. Journal of Investigative Dermatology 52, 268±276. Marzulli F. and Maibach H. I. (1980) Contact allergy. Predictive testing of ¯agrance ingredients in humans by Draize and maximization methods. Journal of Environmental Pathology and Toxicology 3, 235±245. Montelius J., Wahlquist H., Boman A., FernstroÈm P., Grabergs L. and Wahlberg J. E. (1994) Experience with the murine local lymph node assay: inability to discriminate between allergens and irritants. Acta Dermatologica Venereologica 74, 22±27. Moulon C., PeÂguet-Navarro J., Courtellemont P., Redziniak G. and Schmitt D. (1993) In vitro primary sensitization and restimulation of hapten-speci®c T cells by fresh and cultured human epidermal Langerhans cells. Immunology 80, 373±379. Romani N., Gruner S., Brang D., KaÈmpgen E., Lenz A., Trockenbacher B., Konwalinka G., Fritsch P. O., Steinman R. M. and Schuler G. (1994) Proliferating dendritic cell progenitors in human blood. Journal of Experimental Medicine 180, 83±93. Rougier N., Redziniak G., Schmitt D. and Vincent C. (1998) Evaluation of the capacity of dendritic cells derived from cord blood CD34+ precursors to present haptens to unsensitized autologous T cells in vitro. The Journal of Investigative Dermatology 110, 348±352. Sallusto F. and Lanzavecchia A. (1994) Ecient presentation of soluble antigen by cultured human dendritic cells is maintained by granulocyte/macrophage-colony stimulating factor plus interleukin-4 and down-regulated by tumar necrosis factor a. Journal of Experimental Medicine 179, 1109±1118. Shearer G. M. (1974) Cell-mediated cytotoxicity to trinitrophenyl-modi®ed syngeneic lymphocytes. European Journal of Immunology 4, 527±533. Steinman R. M. (1991) The dendritic cell system and its role in immunogenicity. Annual Review of Immunology 9, 271± 296.