Nickel-Induced Contact Dermatitis Peripheral and Cutaneous T Cells

Preferential Usage of TCR-Vβ17 by Peripheral and Cutaneous T Cells in Nickel-Induced Contact Dermatitis

This information is current as Lioba Büdinger, Nicole Neuser, Uwe Totzke, Hans F. Merk of September 27, 2021. and Michael Hertl J Immunol 2001; 167:6038-6044; ; doi: 10.4049/jimmunol.167.10.6038 http://www.jimmunol.org/content/167/10/6038 Downloaded from

References This article cites 30 articles, 8 of which you can access for free at: http://www.jimmunol.org/content/167/10/6038.full#ref-list-1 http://www.jimmunol.org/ Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

by guest on September 27, 2021 *average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2001 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Preferential Usage of TCR-V␤17 by Peripheral and Cutaneous T Cells in Nickel-Induced Contact Dermatitis1

Lioba Bu¨dinger,*† Nicole Neuser,* Uwe Totzke,‡ Hans F. Merk,* and Michael Hertl2*§

Nickel (Ni) is one of the most common contact sensitizers in man, and Ni-induced contact dermatitis is considered as a model of hapten-induced delayed type hypersensitivity. Previous studies indicated that Ni-reactive T cells derived from Ni-allergic individuals preferentially express distinct TCR-V␤ chains. However, data on the TCR-V␤ repertoire of Ni-responsive T cells are not consistent. Therefore, the aim of this study was to identify the TCR-V␤ receptors of Ni-responsive peripheral and cutaneous T cells in a cohort of 17 donors with Ni-induced contact dermatitis in comparison with those of 6 healthy controls. Peripheral NiSO4- responsive T lymphocytes showed a significant overexpression of TCR-V␤17 and the frequency of TCR-V␤17؉ T cells correlated significantly with the in vitro reactivity of PBMC to NiSO4. In addition, the cutaneous infiltrate of Ni-induced patch test reactions ؉

␤ Downloaded from consisted primarily of V 17 T cells. The majority of patch test-derived NiSO4-responsive T cells of three allergic donors were ␤ ؉ ␤ TCR-V 17 , whereas patch test-derived NiSO4 unresponsive T cells of four additional donors did not express TCR-V 17. Skin-derived Ni-responsive T cell lines from three donors uniformly secreted the Th2 cytokine, IL-5, but no IFN-␥ or IL-10. These in vitro and in vivo ﬁndings strongly suggest that T cells with a restricted TCR-V␤ repertoire, i.e., V␤17, predominate in

NiSO4-induced contact dermatitis and may be crucial in the effector phase of Ni hypersensitivity. The Journal of Immunology, 2001, 167: 6038–6044. http://www.jimmunol.org/ ickel (Ni) is one of the most common contact sensitizers the aim of this study was a comprehensive analysis of TCR-V␤ causing delayed-type hypersensitivity in humans (1–4). expression of Ni-specific T cells by examining peripheral as well N Accordingly, the cellular immune response to Ni2ϩ ions as skin-derived T cells, by investigating sufficient numbers of in- has become a model system for studying hapten-induced hyper- dividuals for statistical evaluation, and by correlating both in vivo sensitivity (5–6). There are two key processes in hapten-induced and in vitro findings. immune responses: presentation of the hapten in association with endogenous or exogenous peptides by HLA molecules on the sur- face of the APC, and recognition of the Ag-HLA-complex by the Materials and Methods T cell via its specific Ag receptor (TCR). Ni-allergic and nonallergic individuals by guest on September 27, 2021 In Ni-sensitized individuals, the first step in this sequence ap- A total of 17 patients with a history of eczema and a positive epicutaneous pears to be rather unselective, because Ni is not preferentially pre- patch test reaction to NiSO4 and six healthy, nonallergic control individuals sented in association with distinct HLA-DR, -DQ, and -DP alleles participated in this study after giving written consent. Patch tests were performed with 5% NiSO4 in petrolatum and were evaluated after 48 and (4, 6). Therefore, much of the interest has recently been focusing 72 h. All patients, but none of the control persons, showed a positive result. on T cell recognition and the nature of the epitopes formed by Positive reactions ranged from erythema with few papules to multiple Ni2ϩ ions. Ni preferentially binds to cysteine and histidine resi- papulovesicles (Table I). Lymphocyte proliferation assays were performed dues of proteins/peptides and may be presented to T cells by pro- with PBMC immediately after patch testing. Skin biopsies were taken from cessing-dependent and -independent pathways, as shown for other six Ni-allergic patients and two nonallergic controls 24 h after performing an additional epicutaneous patch test with NiSO4. This particular time haptens (4, 6–7). With regard to the structural characteristics of the point was chosen because the number of Ni-specific T cells in the cutane- TCR used by Ni-specific T cells, previous studies strongly sug- ous infiltrate decreases over time due to the secondary influx of bystander gested that only a limited TCR-V␤ repertoire is expressed by Ni- T cells of unrelated Ag specificity. Biopsies were placed in the culture responsive peripheral T cells (8–11). However, the findings of the media without prior mechanical disaggregation, leading to the emigration of the cutaneous T cells into the culture media. After 7–10 days of culture, individual studies were inconsistent due to the different study de- 4–10 ϫ 106 skin-derived T cells were recovered from each 4-mm punch signs and the small numbers of individuals investigated. Therefore, biopsy.

*Department of Dermatology and †Interdisciplinary Center for Clinical Research, Ags and media ‡ Rheinisch-Westfa¨lische Technische Hochschule, Aachen, Germany; MDS Pharma ϫ ␮ § The following Ags were used in this study: NiSO4 6H2O (20 M), Services, Hamburg, Germany; and Department of Dermatology, University of ␮ ␮ ␮ Erlangen, Erlangen, Germany PdCl2 (10 M), CoCl2 (1 M), CuCl2 (0.1–5 M; all from Sigma-Aldrich, Deisenhofen, Germany), PHA (1%, Life Technologies, Rockville, MD), Received for publication May 18, 2001. Accepted for publication September tetanus toxoid (TT3; 3.2 Lf/ml, Behring Diagnostics, Marburg, Germany). 13, 2001. PBMC and patch test-derived T cells were cultured in RPMI 1640 (Life The costs of publication of this article were defrayed in part by the payment of page Technologies) supplemented with 2 mM L-glutamine, penicillin/streptomy- charges. This article must therefore be hereby marked advertisement in accordance cin, 1 mM Na-pyruvate, 1ϫ nonessential amino acids (all from Life Tech- with 18 U.S.C. Section 1734 solely to indicate this fact. nologies), and 10% heat-inactivated pooled human AB serum (Sigma-Al- 1 This study was supported by a grant from the Interdisciplinary Center for Clinical drich) or 10% FCS and 40 U/ml hIL-2 (Biotest Diagnostics, Dreieich, Research of the Rheinisch-Westfa¨lische Hochschule (to L.B.) and by the Deutsche Germany), respectively. Forschungsgemeinschaft (He 1602/5-2 to M.H.). 2 Address correspondence and reprint requests to Dr. Michael Hertl, Department of Dermatology, University of Erlangen, Hartmannstrasse 14, D-91054 Erlangen, Ger- many. E-mail address: [email protected] 3 Abbreviations used in this paper: TT, tetanus toxoid; SI, stimulation index.

Table I. Reactivity of 17 Ni-allergic donors to metal ions in vivo and in vitro

Patch Test Reactivitya Proliferative Response of PBMCb

c NiSO4 CoCl2 PdCl2 No Ag Patient NiSO4 CoCl2 PdCl2 (SI) (SI) (SI) (cpm) M8 ϩϩ ϩ Ϫ 16.5 0.5 NT 3153 M24 ϩϪϪ1.4 1.2 1.1 2884 M26 ϩϩ Ϫ Ϫ 2.4 2.4 NT 515 M27 ϩϩ Ϫ Ϫ 3.0 0.7 1.2 712 M32 ϩϩ Ϫ Ϫ 3.8 2.1 0.9 1858 M34 ϩϩϩ Ϫ Ϫ 40.3 3.7 1.4 746 M35 ϩϪϪ2.5 NT NT 1366 M36 ϩϩ Ϫ Ϫ 2.7 0.7 NT 891 M37 ϩϩ Ϫ Ϫ 2.2 0.7 0.5 1000 M38 ϩϩϩ Ϫ Ϫ 30.3 6.3 2.1 439 M41 ϩϩ Ϫ Ϫ 8.6 1.6 1.4 209 M42 ϩϩϩ ϩϩ ϩ 8.8 1.7 1.2 2551 M43 ϩϩ Ϫ Ϫ 1.3 0.9 0.9 1220 M46 ϩϩ Ϫ Ϫ 10.4 2.9 1.9 291 M51 ϩϩ Ϫ Ϫ 5.8 1.5 1.4 1623 M52 ϩϩ Ϫ Ϫ 6.0 1.0 1.8 1118 Downloaded from M53 ϩϩ Ϫ Ϫ 4.4 1.1 2.0 325

a Patch test reactivity 48 h after epicutaneous application of Ag; ϩ, erythema; ϩϩ, papules/eczema; ϩϩϩ, papulo-vesicles; Ϫ, no cutaneous reaction. b Ն ␮ SI (cpm culture with Ag/cpm in culture without Ag); a SI 3 was considered signiﬁcant (bold). Metal salts were used at ﬁnal concentrations as follows: NiSO4,20 M; ␮ ␮ CoCl2/CuCl2,1 M; PdCl2,10 M; NT, not tested. c cpm, [3H]thymidine incorporation. http://www.jimmunol.org/

Proliferative assays with peripheral and cutaneous T cells (TCRBV)23S1 (all from Immunotech, Hamburg, Germany). Following two washes with staining buffer, the cells were resuspended in staining PBMC were isolated from 20–40 ml of heparinized blood by density cen- buffer and incubated with a PE-labeled second Ab (PE rat anti-mouse IgG; trifugation on Lymphoflot (Biotest Diagnostics). Two 3ϫ105 PBMC were Immunotech) for 30 min. After blocking with pooled mouse IgG (1 mg/ml; cultured in the presence or absence of Ag in 96-well round-bottom culture Sigma-Aldrich), PBMC were stained with a FITC-labeled anti-CD3 mAb plates for 7 days (Falcon, Oxnard, CA). On day 6, triplicate microcultures were pulsed with 1 ␮Ci/well [3H]thymidine (DuPont Pharmaceuticals, (BD Biosciences, San Jose, CA). Following an additional washing step, ϫ 3 ϩ Mechelen, Belgium) and were harvested on GF/A filters after 18 h of cells were resuspended in staining buffer, and 5 10 CD3 T cells were incubation, and the incorporation of [3H]thymidine was measured in an analyzed on a FACScan flow cytometer (BD Biosciences) using a gate for

automatic beta-counter (Wallac, Freiburg, Germany). lymphocytes. by guest on September 27, 2021 Patch test-derived T cells from six allergic and two nonallergic individuals were obtained as previously described (12). In short, punch biopsies

from epicutaneous patch test reactions to NiSO4 were cultured in RPMI Immunohistochemical analysis of the T cell infiltrate in 1640/10% FCS supplemented with 40 U/ml IL-2 in 24-well culture plates epicutaneous patch test reactions to NiSO (Falcon) for 14 days. A total of 105 in vitro-expanded cutaneous T cells 4 ␮ ϫ 5 were stimulated with 20 M NiSO4 and 2 10 x-irradiated (5000 rad) Cryostat sections were prepared from skin biopsies 24 h after epicutaneous autologous PBMC as APC in round-bottom microculture plates (Falcon). application of 5% NiSO4. Skin sections were fixed in ice-cold acetone for After 4 days, NiSO4-specific proliferation was determined by the uptake of 10 min and were then incubated with mAb specific for various TCR-V␤ [3H]thymidine, which was added for 18 h. In some experiments, Ag-spe- (diluted at 1/20) and anti-CD3, anti-CD4, and anti-CD8 mAb (diluted at cific T cell proliferation was expressed as a stimulation index (SI), which 1/50; all from DAKO, Hamburg, Germany). mAbs were diluted in PBS/ 3 is the ratio of [ H]thymidine uptake (cpm) in cultures with Ag and without Tween/BSA and were added to each skin section in a volume of 100 ␮l for Ն Ag.ASIof 3 was considered to represent a significant proliferative 30 min at room temperature. Skin sections were washed three times in response. PBS/Tween for 5 min and were then incubated with a second, AP-conjugated Ab (rat anti-mouse IgG; DAKO) at 1/100 in PBS/Tween/BSA for 30 Cytokine secretion of NiSO4-reactive skin-derived T cells min. After three washes with PBS/Tween/BSA, an APAAP-complex solution (DAKO; 1/100 in PBS/Tween/BSA) was added to each skin section Cytokine secretion of cutaneous NiSO -reactive T cells was determined by 4 for 30 min. Unbound complex was removed by three washes with PBS/ coculture of 105 T cells with 2 ϫ 105 x-irradiated (5000 rad) autologous Tween/BSA and skin sections were stained with fuchsin solution (DAKO). PBMC as APC and Ag (NiSO , PdCl , CoCl , and CuCl ). After 48 h, 4 2 2 2 Skin sections were counterstained with hematoxylin for 2 min and were culture supernatants were collected and analyzed for IL-5, IL-10, and IFN-␥ immunoreactivity by ELISA (Beckmann, Hamburg, Germany) ac- mounted with glycerol. cording to the manufacturer’s instructions. Statistical analysis TCR-V␤ usage of peripheral and cutaneous T cells Peripheral T cells of the Ni-allergic donors expressing different TCR-V␤ Analysis of the TCR-V␤ repertoire of peripheral and cutaneous NiSO - 4 chains showed a differential proliferative response upon in vitro stimulation reactive T cells was performed 7–10 days (PBMC) or 4 days (cutaneous T ␤ with NiSO4 and TT, respectively. Accordingly, results of TCR-V usage cells) after in vitro stimulation of the T cells with NiSO4 or TT as a control Ag. For phenotypical analysis, 5 ϫ 104 T cells were resuspended in a are shown as box plots with medians and corresponding confidence inter- ␤ staining buffer consisting of PBS supplemented with 3% FCS and 0.1% vals (Fig. 1A). Differences in TCR-V frequencies of Ni- and TT-stimu- Na-azide (Sigma-Aldrich) and were incubated with mouse anti-human lated peripheral T cells were tested pairwise for statistical significance by TCR-V␤ mAb diluted at 1/10 for 30 min. The following mAb specific for Wilcoxon test. To control for serial testing, a sequential Bonferroni adjust- TCR-V␤ elements were used: (TCRBV)1S1, (TCRBV)2S1, (TCRBV)3S1, ment (13) with a table-wide significance level of 0.1 was applied. Over- (TCRBV)5S1, (TCRBV)5S2, (TCRBV)5S3, (TCRBV)7S1, (TCRBV)8S1/ expression of distinct TCR-V␤ chains of Ni-responsive peripheral T cells S2, (TCRBV)V9S1, (TCRBV)11S1, (TCRBV)12S2, (TCRBV)13S1, was further analyzed for a correlation with the extent of the proliferative (TCRBV)13S6, (TCRBV)14S1, (TCRBV)V16S1, (TCRBV)17S1, PBMC response to NiSO4 (SI) by a Spearman rank correlation analysis (TCRBV)18S1, (TCRBV)V20S1, (TCRBV)21S3, (TCRBV)22S1, and (Fig. 1B). 6040 TCR-V␤17ϩ T CELLS AND Ni ALLERGY

ducible peripheral T cell response to NiSO4 while six of 17 donors showed only suboptimal or marginal proliferative T cell responses

to NiSO4 (Table I). Two donors showed also positive patch test reactivity to CoCl2 and one to PdCl2; PBMC from two donors developed a proliferative response to CoCl2. None of the six healthy controls showed a positive patch test

reaction to NiSO4 or a signiﬁcant proliferative response of PBMC cultured with NiSO4.

␤ TCR-V 17 is preferentially used by NiSO4-reactive peripheral T cells PBMC from 10 Ni-allergic donors (Table I) showed overexpression of TCR-V␤1, 3, 7, 8, 9, 12, 17, 21, and 23 upon stimulation

with NiSO4 as compared with stimulation with TT (Wilcoxon test, p Ͻ 0.05). Bonferroni adjustment for serial testing identiﬁed a more restricted preferential usage of the TCR-V␤ elements 12, 17,

and 21 by the NiSO4 responsive T cells (Fig. 1A). A Spearman rank correlation analysis with data of all 17 Ni-allergic donors Downloaded from revealed a signiﬁcant relationship of TCR-V␤17 but neither of TCR-V␤12 (rs ϭϪ0.277, p Ͼ 0.1) nor of TCR-V␤21 (rs ϭ 0.063, Ͼ p 0.1) frequencies with the extent of NiSO4-induced T cell proliferation (Fig. 1B). This correlation was still signiﬁcant when the values of donor M34 were excluded from the analysis. Periph- eral T cells of the two donors M34 (Fig. 2) and M52 with severe Ni-induced contact dermatitis showed a Ͼ10-fold increase of http://www.jimmunol.org/ ␤ TCR-V 17 expression after stimulation with NiSO4 compared with stimulation with TT (M34: Ni, 14.11% and TT, 1.1%; M52: Ni, 6.9% and TT, 0.6%; percentages related to total CD3ϩ cells). In contrast, there was no comparable increase of TCR-V␤12 (M34, 1.3%; M52, 0.7%) or V␤21 (M34, 2.0%; M52, 0.5%) usage by peripheral T cells from these donors upon in vitro stimulation with

NiSO4. by guest on September 27, 2021 TCR-V␤17ϩ T cells are a major component of the T cell

infiltrate in NiSO4-induced contact dermatitis Histological examination of skin sections of five Ni-allergics (M26, M27, M38, M42, and M52) with Ni-induced contact dermatitis showed a cellular infiltrate, which was primarily composed ϩ FIGURE 1. A, Preferential usage of restricted TCR-V␤ chains by pe- of CD3/CD4 T cells (Fig. 3). In all these patients, the cutaneous ϩ ␤ ϩ ripheral NiSO4-responsive T cells. NiSO4-responsive PBMC from 10 Ni- CD3 T cell infiltrate contained a high proportion of TCR-V 17 allergic donors (Table I) were either cocultured with NiSO4 ( )orTTas T cells (Fig. 3). In contrast, TCR-V␤12 and V␤21 (which were a control Ag (Ⅺ) for 7 days and were stained with mAb against various also expressed by peripheral T cells) were only occasionally de- TCR-V␤ followed by a second, PE-conjugated Ab. Afterward, PBMC tected on the cutaneous T cell infiltrate in Ni-induced contact der- were stained with a FITC-conjugated anti-CD3 mAb. TCR-V␤ expression ␤ ϩ matitis (not shown). In addition, TCR-V 2, which is overex- of peripheral CD3 T cells was finally determined by FACS analysis (dou- pressed by T cells in normal human skin, was rarely detected in the ble green/red fluorescence). The number of PBMC samples that were sub- cutaneous T cell infiltrate in the Ni-allergic donors (Fig. 3). Biop- jected to TCR-V␤ analysis is given on the right of each plot (n). Median, vertical line; 25th and 75th percentiles, box; 5th and 95th percentiles, hor- sies of patch tests with NiSO4 from the two nonallergic control izontal lines. Values of p derived from Wilcoxon tests are listed in the right donors Co1 (Fig. 3) and Co2 (not shown) contained only a sparse column; significant differences after Bonferroni adjustment are printed in T cell infiltrate that did not express TCR-V␤17. bold letters. B, TCR-V␤17 expression of peripheral T cells of 17 Ni-allergic donors correlates with their proliferative response to NiSO4. The pro- NiSO4-responsive cutaneous T cells preferentially express TCR- liferative in vitro response to NiSO4 of PBMC of 17 Ni-allergic donors, V␤17 which is expressed as an SI (data shown in Table I), is plotted against the percentage of TCR-V␤17ϩ PBMC upon in vitro stimulation with NiSO . T cells derived from patch tests with NiSO4 of four of six Ni- 4 allergic donors developed a vigorous proliferative in vitro response

to NiSO4 (Table II). In contrast, patch test-derived T cells from the two Ni-allergic donors, M26 and M27, and from the nonallergic Results controls, Co1 and Co2, were not responsive to NiSO (Table II). Epicutaneous patch test reactivity correlates with the 4 To address the question of whether T cell responsiveness to NiSO4 proliferative in vitro T cell response to NiSO4 of Ni-allergic correlated with the expression of TCR-V␤17, the cutaneous T cells patients were double stained with mAb against CD3 and various TCR-V␤ All 17 Ni-allergic donors studied showed positive patch test reac- chains. In fact, the Ni-responsive cutaneous T cell lines from do- ␤ tions of variable intensity to epicutaneously applied NiSO4. Eleven nors M38, 42, and 52 preferentially used TCR-V 17, whereas the of these individuals additionally showed a signiﬁcant and repro- Ni-unresponsive cutaneous T cells from donors M26, M27, Co1, The Journal of Immunology 6041 Downloaded from http://www.jimmunol.org/

FIGURE 2. Preferential usage of TCR-V␤17 by peripheral T cells of donor M34 with severe Ni-induced contact dermatitis. PBMC from donor M34 ␤ with strong epicutaneous patch test reaction to NiSO4 (Table I) were cocultured with NiSO4 or TT and were stained with mAb against various TCR V followed by a second, PE-conjugated Ab. Afterward, PBMC were stained with a FITC-conjugated anti-CD3 mAb, and TCR-V␤ expression of CD3ϩ T cells was determined by FACS analysis (double green/red ﬂuorescence).

␤ and Co2 (not shown) did not express TCR-V 17 (Fig. 4). TCR- trast, NiSO4-unresponsive cutaneous T cells from two allergic do- ␤ ␤ by guest on September 27, 2021 V 12 and V 21 were only expressed by a minor fraction of the nors with less severe skin reactivity to NiSO4 and from two ␤ cutaneous T cells from NiSO4-induced patch test reactions (Fig. nonallergic donors did not express TCR-V 17. The preferential ␤ ϩ ␤ 4), which is consistent with the low frequency of V 12 and usage of TCR-V 17 by peripheral and cutaneous NiSO4-respon- ␤ ϩ V 21 T cells found in NiSO4-induced contact dermatitis in vivo. sive T cells strongly suggests that T cells carrying this particular Thus, these findings strongly suggest that Ni responsiveness of TCR may be critically involved in the effector phase of Ni cutaneous T cells correlates with the expression of TCR-V␤17. hypersensitivity. Several independent investigations have tried to define the struc- NiSO4-responsive cutaneous T cells secrete a Th2-type cytokine tural diversity of the TCR of Ni-specific peripheral and cutaneous pattern T cells. Zollner et al. (10) investigated a cohort of 10 Ni-allergic Cutaneous T cells from four Ni-allergic donors showed a vigorous individuals. Similarly to the present study, PBMC from the donors proliferative in vitro response to NiSO4 (Table II). Skin-derived T were cocultured with NiSO4 and IL-2 for 10 days before pheno- cells from the three Ni-allergic donors, M38, M42, and M52, that typical analysis with mAb against 11 distinct TCR-V␤ chains. Ni- showed an increased expression of TCR-V␤17 (Fig. 4) were responsive PBMC from two of 10 Ni-allergic donors showed an ␤ strongly stimulated by NiSO4 and to a lesser extent by PdCl2 (M38 increased usage of V 2, and Ni-responsive PBMC from three ad- and M42) and by CoCl2 (M42; Fig. 5). NiSO4-specific cutaneous ditional Ni-allergic individuals showed a preferential usage of ei- ␤ ␤ ␤ T cells of donors M38, M42, and M46 produced substantial ther V 17, V 18, or V 21.3 upon in vitro culture with NiSO4/ amounts of IL-5 but no IFN-␥ or IL-10 upon stimulation with IL-2 compared with culture with IL-2 alone. Immunohisto-

NiSO4, suggesting a Th2 cytokine proﬁle. In contrast, upon Ag- chemical analysis of NiSO4-induced patch test reactions revealed independent stimulation with PHA, they produced both IL-5 and a signiﬁcant under-representation of TCR-V␤2 and V␤3 and an IFN-␥ (Fig. 5). over-representation of V␤8 in skin-derived compared with blood- derived T cells from these Ni-allergic donors (10). Because skin

Discussion biopsies in their study were taken 72 h after epicutaneous NiSO4 Ni-responsive peripheral T cells from 10 Ni-allergic donors challenge (in contrast to 24 h in this study), the cutaneous infiltrate showed a preferential usage of TCR-V␤ 12, 17, and V␤21 upon may have contained substantial numbers of bystander cells of stimulation with NiSO4. However, a positive correlation was only unrelated specificity. Werfel et al. (8) investigated the TCR rep- ϩ found between the extent of T cell reactivity to NiSO4 and the ertoire of a panel of blood- and skin-derived CD4 Ni-reactive T ␤ usage of TCR-V 17. Moreover, the cellular infiltrate of NiSO4- cell clones from two Ni-allergic donors. Their findings suggested ␤ induced contact dermatitis contained substantial numbers of TCR- a less restricted usage of TCR-V chains by NiSO4-responsive T ␤ ϩ ␤ ϩ ␤ ϩ V 17 , but not of V 12 or V 21 T cells. NiSO4-responsive cells. Of 79 blood- and skin-derived Ni-reactive T cell clones, skin-derived T cells from three donors with strong contact derma- ϳ40% expressed the TCR elements V␤2, V␤6.7/V␤14, V␤13.1/ ␤ ␤ titis to NiSO4 also preferentially expressed TCR-V 17. In con- 13.2, and V 20. In a more recent study, peripheral T cells from 6042 TCR-V␤17ϩ T CELLS AND Ni ALLERGY Downloaded from

␤ ϩ FIGURE 4. Disparate in vitro responsiveness to NiSO4 of TCR V 17 and TCR-V␤17Ϫ cutaneous T cells. Skin-derived T cells were subjected to analysis of TCR-V␤ expression by FACS after in vitro stimulation with http://www.jimmunol.org/

NiSO4. Cutaneous T cells from donors M38, M42, and M52 that showed a vigorous proliferative response to NiSO4 (Table II) showed a preferential expression of TCR-V␤17. In contrast, patch test-derived T cells from donors M26 and M27 and the nonallergic control Co1 that were unresponsive ␤ to NiSO4 (Table II) showed marginal or no expression of TCR-V 17.

three donors with severe Ni allergy showed an impressive over-representation of the TCR-V␤17 element (12–16%) upon by guest on September 27, 2021 in vitro stimulation with NiSO4 (9). Moreover, the authors showed that TCR-V␤17 expression correlated with the clinical severity of Ni-induced contact dermatitis. This finding is in line with and is extended by the present study showing that TCR- V␤17 is overexpressed by both peripheral and cutaneous Ni- responsive T cells and correlates with the clinical severity of Ni-induced contact dermatitis and the proliferative T cell re- FIGURE 3. The majority of cutaneous T cells in NiSO4-induced contact dermatitis express TCR-V␤17. Skin biopsies were taken from five Ni-allergic sponse to NiSO4. donors (M26, M27, M38, M42, and M52) and a nonallergic donor (Co1) 24 h In this study, TCR other than V␤17 were not consistently over- after epicutaneous application of NiSO4. Cryostat sections of the biopsies were expressed by skin-derived Ni-responsive T cells. In particular, ␤ stained with mAb against human CD4 and against various TCR-V . Skin TCR-V␤2, which was previously identified to be preferentially sections were then incubated with a second, AP-conjugated Ab followed by a final incubation with APAAP-complex solution. Shown is the expression of used by peripheral (10) and cutaneous (8) T cells of selected Ni- CD4, TCR-V␤17, and TCR-V␤2 of the cutaneous T cell infiltrate. allergic donors, was neither overexpressed by peripheral nor by

Table II. In vivo and in vitro reactivity to metal ions of six Ni-allergic donors and two nonallergic controls

In Vitrob

In Vivoa PBMC Skin-derived T cells

c c Pat. NiSO4 NiSO4 (SI) No Ag (cpm) NiSO4 (SI) No Ag (cpm) M26 ϩϩ 2.4 515 0.8 391 M27 ϩϩ 8.3 712 0.8 222 M38 ϩϩϩ 30.3 439 40.6 405 M42 ϩϩϩ 8.8 2551 15.6 597 M46 ϩϩ 10.4 291 4.9 731 M52 ϩϩ 6.0 1118 12.1 950 Co1 – 1.8 446 0.9 135 Co2 – 0.8 1085 0.9 2171 a ϩ ϩϩ ϩϩϩ Ϫ Epicutaneous patch test with NiSO4; , erythema; , papules/eczema; , papulo/vesicles; , no cutaneous reaction. b SI (cpm culture with Ag/cpm in culture without Ag); a SI Ն 3 was considered signiﬁcant (bold). Metal salts were used at ␮ ␮ ␮ ﬁnal concentrations as follows: NiSO4,20 M; CoCl2/CuCl2,1 M; PdCl2,10 M; NT, not tested. c cpm, [3H]thymidine incorporation. The Journal of Immunology 6043

FIGURE 5. NiSO4-responsive cutaneous T cells secrete a Th2-type cytokine pattern. Cutaneous Ni-reactive T cells of the Ni-allergic donors M38, M42, and M46 were cocultured with autologous, x-irradiated PBMC as

APC and NiSO4, PdCl2, CuCl2, and CoCl2 for 4 days, and the proliferative response was determined by the uptake of [3H]thymidine. Cytokine (IFN-␥ and IL-5) analysis was performed in supernatants from cultures with signiﬁcant T cell responses to Ag. Downloaded from http://www.jimmunol.org/

cutaneous Ni-responsive T cells. There is a bias toward expression cytes, whereas Ni-reactive Th2 cells are not (19). Moreover, IL- of TCR-V␤2 by peripheral T cells in healthy individuals (14), as 10-secreting CD4ϩ T regulatory cells seem to be involved in the by guest on September 27, 2021 well as a preferential usage of V␤2 and V␤6 by cutaneous T cells regulation of Ni-induced contact dermatitis by inhibiting the mat- in normal skin (15). uration of dendritic cells (24). Thus, Ni allergy represents a unique In addition to TCR-V␤17, TCR-V␤12 and TCR-V␤21 that were paradigm of human allergic contact dermatitis with many solved found to be preferentially used by peripheral Ni-reactive T cells in questions regarding the pathogenic role that T cells play in hapten- this study were not overexpressed in vivo and in vitro by cutaneous induced contact sensitivity (25).

T cells in NiSO4-induced contact dermatitis. This finding is in line In summary, Ni hypersensitivity may be considered a T cell- with the aforementioned previous studies that did not consistently mediated immune response characterized by the preferential usage detect an overexpression of TCR-V␤12 or V␤21 by peripheral or of a restricted TCR-V␤ repertoire involved in Ag recognition. cutaneous Ni-responsive T cells (8–10). Vollmer et al. (26–27) demonstrated that Ni activates TCR- The Ni-responsive cutaneous T cells produced the Th2 cytokine, V␤17ϩ T cells in a nonsuperantigen-like manner because specific IL-5, but no IFN-␥ and IL-10 upon stimulation with Ni, which is combinations of TCR ␣- and ␤-chains were required for proper T consistent with the previous identification of Ni-responsive cuta- cell activation. In addition, mutations of the complementarity-de- neous Th2 cells in Ni-induced contact dermatitis (8, 16). A recent termining region 3 of TCR-V␤17 (the site of the TCR that pre- study suggested that Ni-reactive peripheral T cells also belonged to sumably interacts with peptide bound to MHC) abrogated T cell 2ϩ the Th2 type because they produced more IL-4 and IL-5 than pe- recognition of NiSO4 (26). Pd , a bivalent metal ion closely re- ripheral T cells from nonallergic donors (17). The preferential de- lated to Ni2ϩ, seems to exhibit similar conformation because the tection of cutaneous Th2 cells in Ni hypersensitivity (8, 16) sug- Ni-responsive cutaneous T cells were also stimulated by Pd. This gests a role for these cells in the effector phase of Ni contact observation is also consistent with previous observations showing dermatitis, although their biological function still remains to be that Ni-reactive T cell clones frequently cross-react with Pd and elucidated. However, there is evidence that Ni hypersensitivity is Cu (28–29). Thus, Ni-induced contact dermatitis holds great also associated with the presence of Ni-reactive Th1 cells (18). promise as a model system to establish therapeutic concepts to Cavani et al. (19) identified both Ni-responsive CD8ϩ and CD4ϩ specifically modulate hypersensitivity reactions to haptens such as T cells in the blood of Ni-allergic donors, whereas nonallergic various ubiquitous and occupational allergens, including metals, individuals had only Ni-responsive peripheral CD4ϩ T cells. Both contact sensitizers, drugs, etc. Evidence for the general feasibility CD4ϩ and CD8ϩ may be critical in Ni-induced contact dermatitis of this approach has been provided by Preckel et al. (30) who as demonstrated in murine models of contact sensitivity (20–21). identified altered hapten ligands that antagonized hapten-specific Human Ni-reactive CD8ϩ T cells displayed cytotoxicity against cytotoxic T cells. Ni-pulsed epidermal keratinocytes and may thus contribute to the pathology of Ni-induced contact dermatitis (22–23). Ni-responsive Acknowledgment ϩ CD4 Th1 cells are also potentially cytotoxic against keratino- We thank Sybille Thoma-Uszynski for critical review of the manuscript. 6044 TCR-V␤17ϩ T CELLS AND Ni ALLERGY

References 17. Borg, L., J. M. Christensen, J. Kristiansen, N. H. Nielsen, T. Menne, and L. K. Poulsen. 2000. Nickel-induced cytokine production from mononuclear cells 1. Bo¨hm, I., M. Brody, and R. Bauer. 1997. Comparison of personal history with in nickel-sensitive individuals and controls. Arch. Dermatol. Res. 292:285. patch test results in metal allergy. J. Dermatol. 24:510. 18. Kapsenberg, M. L., E. A. Wierenga, F. E. Stiekema, A. M. Tiggelman, and 2. Mathelier-Fusade, P., and F. Leynadier. 1997. Crossreactions in contact derma- J. D. Bos. 1992. Th1 lymphokine production profiles of nickel-specific CD4ϩ titis. Clin. Rev. Allergy Immunol. 15:477. T-lymphocyte clones from nickel contact allergic and non-allergic individuals. 3. Krasteva, M., J. Kehren, M. T. Ducluzeau, M. Sayag, M. Cacciapuoti, H. Akiba, J. Invest. Dermatol. 98:59. J. Descotes, and J. F. Nicolas. 1999. Contact dermatitis: pathophysiology of con- 19. Cavani, A., D. Mei, E. Guerra, S. Corinti, M. Giani, L. Pirrotta, P. Puddu, and tact sensitivity. Eur. J. Dermatol. 9:65. G. Girolomoni. 1998. Patients with allergic contact dermatitis to nickel and non- 4. Bu¨dinger, L., and M. Hertl. 2000. Immunologic mechanisms in hypersensitivity allergic individuals display different nickel-specific T-cell responses: evidence ϩ ϩ reactions to metal ions: an overview. Allergy 55:108. for the presence of effector CD8 and regulatory CD4 T cells. J. Invest. Der- matol. 111:621. 5. Sinigaglia, F., D. Scheidegger, G. Garotta, R. Scherper, M. Pletscher, and ϩ ϩ A. Lanzavecchia. 1985. Isolation and characterization of Ni-specific clones from 20. Gocinski, B. L., and R. E. Tigelaar. 1990. Roles of CD4 and CD8 cells in patients with Ni contact dermatitis. J. Immunol. 135:3929. murine contact sensitivity revealed by in vivo monoclonal antibody depletion. 6. Sinigaglia, F. 1994. The molecular basis of metal recognition by T cells. J. Invest. J. Immunol. 144:4121. Dermatol. 102:398. 21. Kehren, J., C. Desvignes, M. Krasteva, M. T. Ducluzeau, O. Assossou, F. Horand, M. Hahne, D. Ka¨gi, D. Kaiserlian, and J. F. Nicolas. 1999. Cytotox- 7. Kalish, R. S., J. A. Wood, and A. LaPorte. 1994. Processing of urushiol (poison icity is mandatory for CD8ϩ T cell-mediated contact hypersensitivity. J. Exp. ivy) hapten by both endogenous and exogenous pathways for presentation human Med. 189:779. T-cells. J. Clin. Invest. 93:2039. 22. Moulon, C., D. Wild, A. Dormoy, and H. U. Weltzien. 1998. MHC-dependent 8. Werfel, T., M. Hentschel, A. Kapp, and H. Renz. 1997. Dichotomy of blood- and and -independent activation of human nickel-specific CD8ϩ cytotoxic T cells ␤ skin-derived IL-4-producing allergen-specific T cells and restricted V repertoire from allergic donors. J. Invest. Dermatol. 111:360. in nickel-mediated contact dermatitis. J. Immunol. 158:2500. 23. Traidl, C., S. Sebastiani, C. Albanesi, H. F. Merk, P. Puddu, G. Girolomoni, and 9. Vollmer, J., M. Fritz, A. Dormoy, H. U. Weltzien, and C. Moulon. 1997. Dom- A. Cavani. 2000. Disparate cytotoxic activity of nickel-specific CD8ϩ and CD4ϩ inance of the BV17 element in nickel-specific human T cell receptors relates to T cell subsets against keratinocytes. J. Immunol. 165:3058. severity of contact sensitivity. Eur. J. Immunol. 27:1865. 24. Cavani, A., F. Nasorri, C. Prezzi, S. Sebastiani, C. Albanesi, and G. Girolomoni. Downloaded from 10. Zollner, T. M., C. Neubert, A. Wettstein, W. H. Boencke, W. H. Manfras, B. O. 2000. Human CD4ϩ T lymphocytes with remarkable regulatory functions on den- Bo¨hm, and W. Sterry. 1998. The T cell receptor V␤ repertoire of nickel-specific dritic cells and nickel-specific Th1 immune responses. J. Invest. Dermatol. 114:295. T cells. Arch. Dermatol. Res. 290:397. 25. Cavani, A., C. Albanesi, C. Traidl, S. Sebastiani, and G. Girolomoni. 2001. Effector 11. Silvennoien-Kassinen, S., J. Karvonen, and I. Ika¨heimo. 1998. Restricted and and regulatory T cells in allergic contact dermatitis. Trends Immunol. 22:118. individual usage of T-cell receptor ␤-gene variables in nickel-induced CD4ϩ and 26. Vollmer, J., H. U. Weltzien, and C. Moulon. 1999. TCR reactivity in human CD8ϩ cells. Scand. J. Immunol. 48:99. nickel allergy indicates contact with complementarity-determining region 3 but 12. Hertl, M., H. Bohlen, F. Jugert, C. Boecker, R. Knaup, and H. F. Merk. 1993. excludes superantigen-like recognition. J. Immunol. 163:2723. Predominance of epidermal CD8ϩ T lymphocytes in bullous cutaneous reactions 27. Vollmer, J., H. U. Weltzien, K. Gamerdinger, S. Lang, Y. Choleva, and http://www.jimmunol.org/ ␣ caused by ␤-lactam antibiotics. J. Invest. Dermatol. 101:794. C. Moulon. 2000. Antigen contacts by Ni-reactive TCR: typical -chain coop- eration versus ␣-chain-dominated specificity. Int. Immunol. 12:1723. 13. Rice, W. R. 1989. Analyzing tables of statistical tests. Evolution 43:223. 28. Moulon, C., J. Vollmer, and H. U. Weltzien. 1995. Characterization of processing 14. Dunn, D. A., A. S. Gadenne, S. Simha, E. A. Lerner, M. Bigby, and requirements and metal cross-reactivities in T cell clones from patients with ␤ P. A. Bleicher. 1993. T cell receptor V expression in normal human skin. Proc. allergic contact dermatitis to nickel. Eur. J. Immunol. 25:3308. Natl. Acad. Sci. USA 90:1267. 29. Pistoor, F. H. M., M. L. Kapsenberg, J. D. Bos, M. M. H. M. Meinardi, 15. Menssen, A., P. Trommler, S. Vollmer, D. Schendel, E. Albert, L. Gurtler, B. M. E. von Blomberg, and R. S. Scheper. 1995. Cross-reactivity of human G. Riethmuller, and J. C. Prinz. 1995. Evidence for an antigen-specific cellular nickel-reactive T-lymphocyte clones with copper and palladium. J. Invest. Der- immune response in skin lesions of patients with psoriasis vulgaris. J. Immunol. matol. 105:92. 15:155. 30. Preckel, T., R. Grimm, S. Martin, and H. U. Weltzien. 1997. Altered hapten 16. Probst, P., D. Kuntzlin, and B. Fleischer. 1995. Th2-type infiltrating T cells in ligands antagonize trinitrophenyl-specific cytotoxic T cells and block internal-

nickel-induced contact dermatitis. Cell. Immunol. 165:134. ization of hapten-speciﬁc receptors. J. Exp. Med. 185:1803. by guest on September 27, 2021