- Email: [email protected]
T Cells and T Cell-Derived Cytokines as Pathogenic Factors in the Nonallergic Form of Atopic Dermatitis
T Cells and T Cell-Derived Cytokines as Pathogenic Factors in the Nonallergic Form of Atopic Dermatitis Cezmi A. Akdis,* Mu¨beccel Akdis,* Dagmar Simon,† Birgit Dibbert,* Martina Weber,* Stephanie Gratzl,* Oliver Kreyden,‡§ Rainer Disch,† Brunello Wu¨thrich,§ Kurt Blaser,* and Hans-Uwe Simon* *Swiss Institute of Allergy and Asthma Research (SIAF), Davos, Switzerland; †Clinic for Dermatology and Allergy (Alexanderhausklinik), Davos, Switzerland; ‡High-Altitude Clinic Davos-Clavadel, Davos, Switzerland; §Department of Dermatology, University of Zu¨rich, Zu¨rich, Switzerland
A subgroup of patients with atopic dermatitis are known to have normal serum total immunoglobulin E levels, undetectable specific immunoglobulin E, and negative skin prick tests towards allergens. This form of the disease has been termed nonallergic atopic dermatitis. In this study, we found that, among 1151 chronic atopic dermatitis patients, about 10% had normal serum immunoglobulin E levels with no evidence for immunoglobulin E sensitization. We investigated immunologic mechanisms of patients with ‘‘allergic’’ and ‘‘nonallergic’’ atopic dermatitis using peripheral blood and skin biopsy samples. Our data suggest that T cells are likely involved in the pathogenesis of both forms of atopic dermatitis. Skin T cells equally responded to superantigen, staphylococcal enterotoxin B, and produced interleukin-2, interleukin-5, interleukin-13, and interferon-γ in both forms of the disease. Interleukin-4, however, was not detectable in the skin biopsies of both atopic dermatitis
types and was secreted in very low amounts by T cells cultured from the skin biopsies. Moreover, skin T cells from nonallergic atopic dermatitis patients expressed lower interleukin-5 and interleukin-13 levels compared with allergic atopic dermatitis patients. Accordingly, T cells isolated from skin biopsies of atopic dermatitis, but not from the nonallergic atopic dermatitis, induced high immunoglobulin E production in cocultures with normal B cells that was mediated by interleukin-13. In addition, B cell activation with high CD23 expression was observed in the peripheral blood of atopic dermatitis, but not nonallergic atopic dermatitis patients. These data suggest, although high numbers of T cells are present in lesional skin of both types, a lack of interleukin-13-induced B cell activation and consequent immunoglobulin E production in nonallergic atopic dermatitis. Key words: atopic dermatitis/ cytokines/immunoglobulin E/superantigen/T cells. J Invest Dermatol 113:628–634, 1999
A
1993; Hamid et al, 1994; Akdis et al, 1997; Fujimura et al, 1997; Nakazawa et al, 1997). Although most of the patients with AD show high concentrations of total and allergen-specific IgE in blood and skin, some of the patients have normal total IgE levels and negative serum allergenspecific IgE (Wu¨thrich, 1978). Diagnostic criteria of AD by Hanifin and Rajka (1980) or Williams et al (1994) can be fulfilled in the absence of elevated total IgE and specific IgE to food or environmental allergens. This suggests that elevated IgE levels and IgE sensitization are not prerequisites in the pathogenesis of the disease. The subgroup of AD patients with normal IgE levels and without specific IgE sensitization has been termed nonallergic form of AD (NAD) or intrinsic-type AD (Ka¨gi et al, 1994). In this study, we reserved the term ‘‘AD’’ only for the patients with features of atopy. Patients were classified into the NAD group when they fulfilled the criteria of Hanifin and Rajka (1980), but did not exhibit any allergen-specific IgE sensitization and had normal serum total IgE levels. In this study, we investigated immunoregulatory mechanisms that are possibly involved in the pathogenesis of NAD in comparison with AD. The results suggest that T cells and T cell-derived cytokines do not only play an important part in allergic AD, but also in NAD. One important difference between the two forms of AD appears to be the different capacity of skin T cells to induce IgE production in B cells, possibly due to decreased interleukin (IL) -13 production in NAD.
topic dermatitis (AD) is a chronic relapsing inflammatory skin disease characterized by typically distributed eczematous skin lesions with lichenification, pruritic excoriations, severely dry skin, and a susceptibility to cutaneous infections (Hanifin and Rajka, 1980; Rudikoff and Lebwohl, 1998). Several lines of evidence suggest the contribution of immunologic mechanisms in the pathogenesis of AD. The skin lesions of AD patients are infiltrated by activated T cells, eosinophils, and antigen-presenting Langerhans cells that bind and present immunoglobulin (Ig) Ecomplexed allergens on their surface (Reinhold, 1992; Ka¨gi et al, 1994). In addition, activation of peripheral blood T cells that preferentially secrete T helper (Th) 2 cytokines and help B cells to produce large amounts of immunoglobulin (Ig) E has previously been reported to be associated with this skin disease (Walker et al,
Manuscript received March 18, 1999; revised June 10, 1999; accepted for publication June 14, 1999. Reprint requests to: Dr. Cezmi A. Akdis, Swiss Institute of Allergy and Asthma Research, Obere Strasse 22, CH-7270 Davos, Switzerland. Email: [email protected] Abbreviations: AD, atopic dermatitis; CLA, cutaneous lymphocyteassociated antigen; ECP, eosinophilic cationic protein; NAD, nonallergic form of AD; SEB, staphyloccoccal enterotoxin B.
·
0022-202X/99/$14.00 Copyright © 1999 by The Society for Investigative Dermatology, Inc.
628
VOL. 113, NO. 4 OCTOBER 1999
T CELLS AND CYTOKINES IN ALLERGIC AND NONALLERGIC ATOPIC DERMATITIS
MATERIALS AND METHODS Subjects In the time period from January 1997 until June 1998, we recruited 1151 patients among the three clinics in Davos and Allergy Unit of Zu¨rich University, Zu¨rich, Switzerland who fulfilled the diagnostic criteria of Hanifin and Rajka (1980). On the basis of the case history, total and specific IgE levels and skin prick tests, the subjects were further divided into AD and NAD patients. Characteristics of the AD patients were total serum IgE levels above 150 kU per liter and/or positive radioallergosorbent test or immunoassay (Pharmacia Diagnostics AG, Uppsala, Sweden), as well as a positive patient’s history for allergic diseases. Usually, these patients also had positive skin prick tests towards common environmental allergens. In many cases, however, skin prick tests could not be performed due to the severity of skin symptoms. Patients were termed NAD when they had a negative history for allergic diseases, low levels of total IgE (,150 kU per liter), no elevated specific IgE antibodies, and no immediate-type skin reactions towards a routine panel of relevant aeroallergens and food allergens. The possibility that the patients might have allergic contact dermatitis was ruled out by multiple evaluations of the patients at different times, disease history, clinical course, and features of the disease. Two of the patients were skin patch tested to exclude allergic contact dermatitis. For immunologic analysis, we obtained 10 ml of ethylenediamine tetraacetic acid anti-coagulated blood and skin biopsies from nine patients with AD (two female, seven male, mean age: 37 y) and eight patients with NAD (four female, four male, mean age: 49 y). One patient from the AD group and one patient from NAD group had asthma. Three patients with AD had allergic rhinoconjunctivitis. The clinical material was collected under standard hospital approved protocols. Skin biopsy specimens were obtained from acute AD lesions of less than 5 d onset. We also collected a group of six healthy individuals (three male, three female, mean age: 39 y) with no history or signs of atopy as a control group for immunophenotyping. All AD and NAD patients suffered from an acute exacerbation of their chronic disease. Duration of their disease was 18 6 4 y in AD and 16 6 6 y in NAD (mean 6 SEM). There was no significant difference in the age and age of onset between the AD versus NAD groups. Serum IgE levels were 12333 6 5663 kU per liter in AD and 42 6 13 kU per liter in NAD. The patients have not been treated with oral or topical steroids for at least 1 wk prior to vein puncture and skin biopsy. Informed consent was obtained from all patients and control individuals, and the study was approved by the Swiss Academy of Medical Science represented by the Medical Ethics Committee of Davos. Media, reagents, and antibodies Complete culture medium was RPMI 1640 supplemented with 2 mM L-glutamine, penicillin/streptomycin, and fetal bovine serum (all from Life Technologies, Basel, Switzerland). All conjugated monoclonal antibodies (MoAb) for flow-cytometric analysis were purchased from Coulter (Hialeah, FL), Immunotech Ltd. (Marseilles, France), or Pharmingen (San Diego, CA). Anti-CD2 (4B2 and 6G4) and anti-CD28 (15E8) MoAb were from the Red Cross Blood Transfusion Service in Amsterdam (the Netherlands). Anti-CD3 was produced by clone CRL 8001 obtained from ATCC (Washington, DC). Biotin-labeled MoAb and avidin was from Sigma (St. Louis, MO). Neutralizing anti-IL-4 MoAb (8F12) was provided by Novartis (Basel, Switzerland). Neutralizing anti-IL-13 MoAb (JES8–5A2 and JES10–2F9) and mutant IL-4 antagonist (Y124D; with inhibitory activity for both IL-4 and IL-13) were provided by DNAX Research Institute (Palo Alto, CA). The antibodies for immunohistochemistry, anti-IL-2, anti-interferon (IFN) -γ, and anti-IL-13 MoAb were from R&D Systems Europe Ltd. (Abingdon, U.K.). AntiIL-5 was a kind gift of Dr J. Tavernier (Gent, Belgium). Anti-cutaneous lymphocyte-associated antigen (CLA) MoAb was from Pharmingen, and anti-eosinophilic cationic protein (ECP; EG1) was from Pharmacia (Uppsala, Sweden). Anti-CD4 MoAb was obtained from Novocastra Laboratories (Newcastle, U.K.). Anti-CD8 and control MoAb were purchased from Dako (Zug, Switzerland). Analysis of leukocytes and lymphocyte subsets Blood leukocyte and differential counts were determined by automatic blood count analysis (Technicon H1, Technicon, Tarrytown, NY). To determine lymphocyte subsets, specific binding of MoAb was analyzed by direct immunofluorescence using standard methods (Oehling et al, 1997). The flow-cytometric analysis was performed by an EPICS XL (Coulter). Identification of the inflammatory cell infiltrate in the skin Inflammatory cells were identified using lineage-specific MoAb and immunohistochemistry. Briefly, fresh skin tissues were fixed in 4% paraformaldehyde solution. Paraffin sections were mounted on poly Llysin-coated slides, stored at room temperature, deparaffinized in xylene, and rehydrated through graded concentrations of ethanol before use.
629
Antigen retrieval was achieved by the microwave oven heating procedure. CD4 and CD8 stainings were performed with the Dako catalyzed signal amplification (CSA) system according to the manufacturer’s instructions (Dako). CLA stainings were done using the Zymed Lab-SA System (Zymed Laboratories, South San Francisco, CA). ECP stainings were performed using the APAAP technique as described previously (Simon et al, 1997). Purification and stimulation of skin T cells The skin biopsy specimens from four AD patients and four NAD patients were minced with two scalpels; no proteolytic enzymes were used. The T cells of the obtained skin biopsy fractions were stimulated in complete culture medium with 25 U IL-2 per ml and the following combination of MoAb to T cell surface molecules: anti-CD2 (4B2 and 6G4, each 0.5 µg per ml), antiCD3 (5 µg per ml), and anti-CD28 (1 µg per ml) (Akdis et al, 1999). After 7–10 d, this treatment resulted in an expansion of skin T cells. These cells were washed and used for functional assays (IgE production), as well as for proliferative response to super antigens and cytokine analysis by flow cytometry and enzyme-linked immunosorbent assay (ELISA). T cells isolated from skin of AD and NAD patients were stimulated with different doses of staphylococcal enterotoxin B (SEB) in 96-well flat bottomed plates (5 3 104/200 µl per well) in triplicates by using rad irradiated autologous peripheral blood mononuclear cells (PBMC) (5 3 104) as antigen-presenting cells. [3H]thymidine incorporation was measured after 3 d following an 8 h incubation with 1 µCi per well of [3H]thymidine. The plates were harvested on glass-fiber plates and radioactivity was measured in a Beta Plate Reader (Pharmacia-Wallac, Turku, Finland). For the determination of cytokine production, the in vitro-expanded skin T cells (0.5 3 106 per ml; 48-well plates in triplicates) were again stimulated with anti-CD2, anti-CD3, and anti-CD28 MoAb for 72 h, and the supernatants were harvested and frozen at –80°C until ELISA measurements. For intracellular staining of cytokines, the cells were stimulated with the same combination of MoAb for 16 h before flow cytometric analysis. Measurements of cytokine levels ELISA The solid-phase sandwich ELISA for IFN-γ, IL-4, IL-5, and IL-10 were performed as described previously (Akdis et al, 1996, 1997, 1998). IL-13 and granulocyte-macrophage colony-stimulating factor were measured using commercial kits according to the manufacturer’s instructions (R&D Systems). Immunohistochemistry All cytokine stainings were performed using the APAAP technique as previously described (Yousefi et al, 1995; Simon et al, 1997). Sections were lightly counterstained with hematoxylin, mounted, and examined under a Zeiss Axioscope at a magnification of 3400 or 31000. The intensities of the stainings were evaluated by two independent investigators from low (0) to high (6). Flow cytometry During the last 14 h of cell culture, 1 µM monensin was added. The cells were washed with phosphate-buffered saline and surfacestained with phycoerythrin Texas red-labeled anti-CD4 or anti-CD8 MoAb, followed by fixing and permeabilizing of the cells with a formaldehyde/saponin solution (Ortho Permeafix, Ortho Diagnostic Systems, Raritan, NJ). After washing with phosphate-buffered saline containing 5% fetal bovine serum, 1.5% bovine serum albumin, and 0.0055% ethylenediamine tetraacetic acid, the cells were counter-stained with 0.5 µg per ml phycoerythrin- or fluorescein isothiocyanate-conjugated control, anti-IL-4, anti-IL-5, anti-IL-13, and anti-IFN-γ MoAb for 30 min at 4°C. The flowcytometric analysis was performed using an EPICS XL (Coulter). IgE production in vitro PBMC from normal control individuals were depleted from CD3-, CD4-, and CD8-bearing cells by negative selection using a magnetic cell separation system (MACS; Miltenyi Biotec, BergischGladbach, Germany) as described previously (Akdis et al, 1997). T celldepleted PBMC (0.9 3 105) were cocultured with anti-CD2-, anti-CD3-, and anti-CD28-stimulated skin T cells (0.1 3 105) derived from four AD and four NAD patients. IgE was determined in supernatants harvested after 12 d. Inhibition of IgE production was attempted by addition of neutralizing MoAb to IL-4 or IL-13 (both at 10 µg per ml) (Akdis et al, 1997). The IL-4 and IL-13 antagonist Y124D was used at 100 ng per ml (Zurawski and De Vries, 1994). Mouse IgG1 (10 µg per ml) was used as a control. IgE concentrations were measured by ELISA as described previously (Akdis et al, 1996, 1997). All experiments were performed in triplicate. Statistics All data are expressed as means 6 SD. Statistical analysis was performed using the Student’s t test (*p , 0.05).
630
AKDIS ET AL
THE JOURNAL OF INVESTIGATIVE DERMATOLOGY
Table I. Peripheral blood leukocyte and lymphocyte subpopulationsa Controls
AD
NAD
Leukocytes (per µl) 6850 6 1331 7544 6 2952 7525 6 3318 Lymphocytes (%) 34 6 8 37 6 19 36 6 7 74 6 4 74 6 5 73 6 8 CD31 (%) 47 6 6 50 6 6 45 6 15 CD31CD41 (%) 23 6 3 23 6 6 25 6 12 CD31CD81 (%) CD4/CD8 ratio 2.1 6 0.6 2.4 6 0.8 2.5 6 1.8 11 6 4 664 10 6 7 CD161 (%) 10 6 6 15 6 4 13 6 5 CD191 (%) 1 1 765 14 6 12 763 CD4 HLA-DR (%) 18 6 6 26 6 13 27 6 9 CD41CD251 (%) 664 16 6 24 19 6 16 CD81HLA-DR1 (%) 161 7 6 17 262 CD81CD251 (%) 1 1 24 6 16 67 6 12* 21 6 12 CD19 CD23 (%) Monocytes (%) 463 662 764 96 6 5 94 6 4 93 6 5 CD141HLA-DR1 (%) Neutrophils (%) 60 6 8 47 6 14 53 6 9 Eosinophils (%) 261 10 6 5* 4 6 3* aThese data are mean 6 SD from six normal control individuals, nine AD patients, and eight NAD patients. *Significantly different values from those of normal control individuals (p , 0.05).
RESULTS Peripheral blood leukocyte and lymphocyte subpopulations in AD and NAD We investigated 1151 patients who fulfilled the criteria of Hanifin and Rajka (1980) within a time period of 18 mo. From these patients, 116 NAD patients were identified. Accordingly, in this study about 10% of all atopic dermatitis patients belonged to the nonallergic group. The immunologic data described here were obtained from nine AD and eight NAD patients. The leukocyte differential counts revealed a characteristic blood eosinophilia in both types of atopic dermatitis. It was more pronounced in AD compared with NAD patients, although it did not show any statistically significant difference (Table I). Immunophenotyping using flow cytometry revealed increased numbers of CD231 B cells in AD, but not in NAD patients, in agreement with previously published work (Wu¨thrich, 1989; Ka¨gi et al, 1994). In contrast, CD41 and CD81 T cells expressed increased activation markers (HLA-DR and CD25) in both AD and NAD patients, although the differences had no statistical significance. T cells infiltrate into the skin of AD and NAD patients To identify the inflammatory cells present in lesional skin, we performed immunohistochemistry using skin biopsies from AD and NAD patients. T cells were identified with anti-CD4, anti-CD8, and anti-CLA MoAb, whereas eosinophils were stained with an antiECP MoAb. As shown in Fig 1 and Table II, AD and NAD patients exhibited similar cellular infiltrates in their lesional skins. The majority of the cells represented CD41 and CD81 cells, suggesting an important role for T cells in both groups. The CD4/ CD8 ratios between blood and skin appeared to be similar, suggesting that both CD41 and CD81 T subpopulations are equally recruited into the inflammatory sites of AD and NAD patients. In allergic AD, the pivotal role of CD45RO1 (memory/effector) T cells expressing the skin homing receptor, the CLA was demonstrated. Almost all T cells in benign and malignant cell infiltrations of the skin express CLA molecule on their surface (Picker et al, 1990, 1991). As previously observed in AD and other dermatologic diseases (Picker et al, 1990), most of the skin-infiltrating T cells of NAD patients were CLA positive. Eosinophil content of skin lesions were low in comparison with T cells. ECP positive cells were 15.1 6 18.3 per mm2 skin in AD and 8.3 6 11.4 per mm2 skin in NAD (mean 6 SD) (Table II). In vivo expression of cytokines in lesional skin of AD and NAD patients It has been repeatedly suggested that Th2
Figure 1. Inflammatory cells in lesional skin of AD and NAD patients. The majority of the cells represented CD41 and CD81 T cells. Many of the T cells were CLA1. The eosinophil infiltration was much less in comparison with the T cell infiltration. No differences were observed between AD and NAD patients. Data are shown in Table II.
cytokines play an important part in AD and other atopic diseases (Hamid et al, 1994; Akdis et al, 1997; Fujimura et al, 1997; Nakazawa et al, 1997). Therefore, we investigated the Th1/Th2 cytokine pattern of AD and NAD patients using immunohistochemistry. As shown in Fig 2, Table II, both Th1 (IL-2 and IFN-γ) and Th2 (IL-5 and IL-13) cytokines were expressed by inflammatory cells of AD and NAD patients. Higher IL-5 and IL-13 expressions were observed in AD than in NAD, although it did not reach statistical significance. As the cellular infiltrate mostly contained T cells (Table II), it is likely that these cells were the main cellular source for these cytokines. Interestingly, IL-4 was neither detectable in skin lesions of AD nor NAD patients. Superantigen response and cytokine expression by skin T cells isolated from lesional skin of AD and NAD patients To further investigate the expression of cytokines by skin T cells, we measured cytokine levels by flow cytometry and ELISA. Owing to the low cell numbers obtained from skin biopsies, however, T cells had to be expanded in vitro. Cells were stimulated with a combination of anti-CD2, anti-CD3, and anti-CD28 MoAb in the presence of IL-2 for optimal T cell proliferation. The in vitroexpanded T cells were then analyzed for intracytoplasmic cytokine expression. As assessed by flow cytometry, T cells from both AD
VOL. 113, NO. 4 OCTOBER 1999
T CELLS AND CYTOKINES IN ALLERGIC AND NONALLERGIC ATOPIC DERMATITIS
631
Table II. Inflammatory cells and cytokines in lesional skin of AD and NAD patientsa
AD NAD
AD NAD
CD4
CD8
CD4/CD8
ECP
CLA (%)
579 6 222 527 6 305
290 6 127 274 6 147
2.11 6 1.07 2.00 6 0.66
15.1 6 18.3 8.3 6 11.4
62.9 6 23.5 59.3 6 23.9
IL-2
IFN-γ
IL-4
IL-5
IL-13
2.75 6 1.83 2.13 6 0.99
1.00 6 1.07 0.50 6 0.76
0.00 0.00
2.50 6 1.77 1.13 6 1.25
3.00 6 1.20 2.13 6 1.36
aThe skin-infiltrating cells were determined by immunohistochemistry per mm2 skin section as described in Materials and Methods. These data are mean 6 SD from nine AD patients, and eight NAD patients The intensity of the cytokine stainings were independently evaluated by two investigators (M.W. and S.G.) from low (0) to high (6) grade.
responsiveness to superantigenic stimuli released by Staphylococcus aureus colonizing the skin. This was analyzed by stimulating the T cells from skin biopsies of AD and NAD patients with SEB, by using irradiated autologous PBMC as antigen-presenting cells. As shown in Fig 4(A), T cells of both AD and NAD patients responded to SEB in a concentration-dependent manner. SEB-specific stimulation only involves T cells that express certain Vβ elements of the TCR. To better quantitate the cytokines expressed by skin T cells, in vitro-expanded skin T cells were stimulated with the combination of anti-CD2/anti-CD3/anti-CD28 MoAb for 72 h, and released cytokines were measured by ELISA. As shown in Fig 4(B), increased IL-5 and IL-13 concentrations were measured in the supernatants of AD compared with NAD patients. T cells from NAD patients, however, also produced detectable amounts of these two cytokines. In contrast, the released amounts of IFN-γ and IL-4 were less, and no significant differences were found between AD and NAD patients. Skin T cells help B cells to produce high amounts of IgE in AD but not NAD As we had evidence for increased IL-13 expression in AD compared with NAD patients, we hypothesized that the threshold to help B cells to produce high amounts of IgE may be reached by skin T cells from AD, but not NAD patients. As shown in Fig 5, no significant IgE production is observed in the absence of T cells. Skin T cells from AD patients induced normal blood B cells to produce significantly high amounts of IgE in comparison with NAD. Whereas the B cell help of skin T cells from AD patients could be neutralized by anti-IL-13 MoAb and recombinant IL-4/IL-13 antagonist, however, the marginal induced IgE production by NAD T cells could not be blocked by these reagents. Neutralizing IL-4 by anti-IL-4 MoAb had no effect in this system. Together, these data suggest that IL-13 is an important cytokine involved in increased IgE production in AD patients. Although T cells from NAD patients also produced significant amounts of IL-13, these levels may not have been elevated enough to stimulate increased IgE production. DISCUSSION Figure 2. IL-13, but not IL-4, is expressed in lesional skin of AD and NAD patients. The positive control for IL-4 was a bladder cancer tissue expressing IL-4. In average, the expression of IL-13 was less in NAD in comparison with AD patients. All data regarding cytokine expression are shown in Table II.
and NAD patients did not express detectable amounts of IL-4 protein (Fig 3), confirming the data obtained by immunohistochemistry (Fig 2, Table II). Moreover, significant amounts of IL-5 were usually detected in T cells from AD (CD41 . CD81), but not NAD patients (Fig 3). In contrast, IFN-γ was frequently detected in T cells from NAD (CD81 . CD41), but not AD patients (Fig 3). Furthermore, under these experimental conditions, we observed increased expression of IL-13 in AD compared with NAD patients. Because NAD patients are not sensitized to aeroallergens and food antigens, one explanation for the existence and in vivo activation of T cells in skin lesions of NAD could be their
A number of immunologic parameters were demonstrated to be abnormal and related to the clinical severity of allergic and nonallergic forms of AD (Kapp et al, 1988; Rousset et al, 1991; Jujo et al, 1992; Walker et al, 1993; Hamid et al, 1994; Ka¨gi et al, 1994; Morren et al, 1994; Akdis et al, 1997; Fujimura et al, 1997; Nakazawa et al, 1997). Based on skin biopsies, the present study demonstrates that T cells are important pathogenetic factors in both forms of the disease. The majority of cells in the cellular infiltrate in AD were T cells. The numbers of infiltrating eosinophils compared with T cells were marginal in both AD and NAD patients. This is in contrast to other eosinophilic tissues from patients with bronchial asthma (Bousquet et al, 1990) or nasal polyps (Simon et al, 1997). Allergen-specific T cells derived from skin lesions in patients with the allergic variant of AD were found to produce predominantly type 2 Th cytokines (Sager et al, 1992; Van Reijsen et al, 1992). These studies were mainly performed by T cells cloned after epicutaneous application of inhalant allergens. In contrast,
632
AKDIS ET AL
THE JOURNAL OF INVESTIGATIVE DERMATOLOGY
Figure 3. Intracytoplasmic cytokine content of CD4F and CD8F skin T cells following stimulation with antiCD2, anti-CD3, and anti-CD28 MoAb for 16 h. The results are representative of four independent experiments in each group of patients. The skin T cells of AD patients expressed much more IL-5 and IL-13 compared with NAD patients. IL-4 expressing cells were ø 3% in both AD and NAD patients. The expression of IFN-γ was lower in AD in comparison with NAD patients.
allergen-specific T cell clones obtained from peripheral blood of the same patients (Sager et al, 1992) and allergen-specific T cell clones from chronic lesions and older patch tests displayed IFN-γ (Werfel et al, 1996). It appears that, the age of the patient population, the age of the eczema lesion (Hamid et al, 1994), the isolation site of T cells and T cell cloning procedure might influence the cytokine pattern of T cells. The present study demonstrates by immunohistochemistry, flow cytometry, and ELISA measurements, that, skin T cells from AD and NAD patients produce significant amounts of IL-5 and IL-13. In all three test systems, AD patients expressed higher levels of these two cytokines than NAD patients. In both groups of patients, IL-4 was not significantly expressed by skin T cells. In contrast to IL-4, IL-13, that can replace IL-4 in the induction of IgE mRNA (Punnonen et al, 1993), was detected in skin lesions of both forms of AD; however, The overall IL-13 expression was lower in NAD compared with AD patients. IL-5 was detected in many, but not in all AD and NAD patients. The Th1 cytokines IL-2 and IFN-γ were both detected in lesional skin, although the IFN-γ levels were relatively low in both AD and NAD patients and was slightly increased in NAD compared with AD patients. Similarly, high amounts of intracytoplasmic IL-5 and IL-13 were detected in activated skin homing CLA1 T cells of AD patients purified from peripheral blood (Akdis et al, 1997, 1999). The leukocyte differential blood counts revealed a characteristic blood eosinophilia in both AD and NAD patients, that was more pronounced in AD compared with NAD patients. High IL-5 production in both forms of AD and even higher levels in allergic form of AD may suggest a part for blood eosinophilia in both diseases as demonstrated by delayed apoptosis of blood eosinophils (Wedi et al, 1997). The functional IgE regulatory capacity of T cells from skin lesions of AD patients was investigated by coculturing isolated skin T cells with purified B cells. The IgE production by B cells mainly depended on secreted IL-13. A number of studies have shown that IL-4 plays an important part in the induction of IgE synthesis in AD in PBMC cultures (Rousset et al, 1991; Jujo et al, 1992; Morren et al, 1994). Previous reports and our present findings, however, indicate that IL-4 inhibition is not sufficient to suppress IgE in atopic diseases. An IgE-inducing activity which could be attributed to IL-13, was found in PBMC cultures of atopic but not normal individuals (Zhang et al, 1992; Van der Pouw-Kraan et al, 1994). In contrast to IL-4, there is no feedback or priming mechanism exerted by IL-13 on T cells, because human T cells do not display specific IL-13 receptors (Zurawski et al, 1994).
Figure 4. Superantigen response and cytokine profile of skin T cells in AD and NAD. There was no difference in SEB induced proliferation between skin T cells of AD and NAD patients. IL-5 and IL-13 production following stimulation with anti-CD2, anti-CD3, and anti-CD28 MoAb for 72 h of the skin T cells from AD patients was significantly high compared with NAD patients. The levels of released IFN-γ and IL-4 were low, and no differences were observed between AD and NAD patients. The data are presented as mean 6 SD of four independent experiments in each patient group (*p , 0.05).
VOL. 113, NO. 4 OCTOBER 1999
T CELLS AND CYTOKINES IN ALLERGIC AND NONALLERGIC ATOPIC DERMATITIS
Figure 5. Skin T cells from both AD and NAD patients induce IgE production in T cell-depleted PBMC populations (T Depl. PBMC). Only the skin T cells from AD patients helped B cells to produce very high amounts of IgE. The induced IgE production by T cells from AD patients could be blocked with neutralizing anti-IL-13 MoAb and recombinant IL-4/IL-13 antagonist (Y124D), suggesting that IL-13 may play an important part in the induction of IgE synthesis. Results represent means of triplicate cultures 6 SD from four independent experiments in each patient group (*p , 0.05).
Similarly, a central role for IL-13 in the expression of murine asthma is demonstrated (Gru¨nig et al, 1998; Wills-Karp et al, 1998), suggesting that IL-4 may be decisive at the initial phase of allergic responses and in the priming and development of Th2 cells (Le Gros et al, 1990; Jung et al, 1996), whereas IL-13 becomes more prominent in IgE induction in chronic AD. In addition, IgE Fc receptor II (CD23) on B cells is significantly expressed in AD compared with NAD (Ka¨gi et al, 1994). It is known that both IL-4 and IL-13 may upregulate this receptor (Zurawski et al, 1994). As IL-4 is not expressed in skin T cells, as demonstrated by immunohistochemistry and intracytoplasmic cytokine staining, IL-13 appears to be the major cytokine in the activation of B cells and IgE production in AD. Although IL-13 was produced in higher amounts in both AD and NAD, B cells of healthy individuals did not respond to the IL-13 produced from skin T cells of NAD. This can be ascribed to IL-13 produced over a threshold to activate B cells in AD. Furthermore, the genetical variations on the quality of IL-13 produced might play a part in increased activity of the cytokine. Moreover, the IL-4–IL-13 receptor common α chain has been demonstrated to have mutations that might induce high B cell responsiveness to IL-13 in AD (Tony et al, 1994; Aman et al, 1996). A number of pathogenetic mechanisms leading to T cell activation in AD including aeroallergens, food allergens, and superantigens have been emphasized. The role of aeroallergens and food allergens in the exacerbation of AD by T cell activation has been extensively studied (Sager et al, 1992; Van der Heijden et al, 1991; Varney et al, 1993; Abernathy-Carver et al, 1995). It is apparent that allergenspecific T cell responses in food and aeroallergen allergy are confined to AD and does not explain the activation and recruitment of T cells in NAD skin lesions. The present study shows that T cells isolated from the skin of AD and NAD patients efficiently proliferate by superantigenic stimulation. From a number of studies it could be concluded that bacterial superantigens contribute to the pathogenesis and exacerbation of AD. Staphylococcal superantigens were isolated from AD skin (Leyden et al, 1974; Kotzin et al, 1993). Superantigen patch test elicits skin inflammation in AD patients (Strange et al, 1996) and in human severe combined immunodeficiency mouse model (Herz et al, 1998). In addition, superantigens upregulate the skin homing ligand (Leung et al, 1995). Together, these data suggest that bacterial superantigens can lead to activation and proliferation of T cells in NAD skin as well as AD skin. The clinical manifestations of NAD were the same as AD suggesting that mechanisms causing eczema is independent of IgE overproduction. Provocation factors other than antigens and
633
superantigens such as emotional factors (Koblenzer, 1983), irritating substances (Agner, 1991), hormones, and climatic changes (Morren et al, 1994) were shown to influence the course of eczema. Taken together, T cells appear to be important players regulating the inflammatory processes in both AD and NAD patients. This hypothesis is further supported by the observation that drugs, such as FK-506, that block the activation of T cells, are effective in AD treatment (Ruzicka et al, 1997). In conclusion, we established a role for T cells in the nonallergic form of AD. Although the cellular infiltrate in lesional skin of both forms of AD patients contained high numbers of T cells, it appeared that the expression of cytokines IL-5 and IL-13 as well as the capacity to produce these cytokines by skin T cells is reduced in NAD patients. T cells that express low IL-13 are unable to help B cells to produce large amounts of IgE in NAD patients. The reasons for the different cytokine expression patterns observed between the two forms of AD patients could involve underlying genetic differences or additional factors that limit IL-5 and IL-13 production at the inflammatory site in NAD patients.
We are grateful to Dr. Jan de Vries and Dr. Gerard Zurawsky (DNAX Research Institute, Palo Alto, CA) for anti-IL-13 MoAb and Y124D mutant IL-4. We also thank Dr. Christoph Heusser and Dr. Sefik S. Alkan (Novartis, Basel, Switzerland) for anti-IL-4 and anti-IFN-γ MoAb, as well as Dr. Jan Tavernier (University of Gent, Gent, Belgium) for anti-IL-5 MoAb. This work was supported by grants 32-49210.96 and 31-50590.97/1, both from the Swiss National Science Foundation.
REFERENCES Abernathy-Carver KJ, Sampson HA, Picker LJ, Leung DYM: Milk-induced eczema is associated with the expansion of T cells expressing cutaneous lymphocyte antigen. J Clin Invest 95:913–918, 1995 Agner T: Susceptibility of atopic dermatitis patients to irritant dermatitis caused by sodium lauryl sulphate. Acta Derm Venereol (Stockh) 71:296–300, 1991 Akdis CA, Akdis M, Blesken T, Wymann D, Alkan SS, Mu¨ller U, Blaser K: Epitope specific T cell tolerance to phospholipase A2 in bee venom immunotherapy and recovery by IL-2 and IL-15 in vitro. J Clin Invest 98:1676–1683, 1996 Akdis M, Akdis CA, Weigl L, Disch R, Blaser K: Skin-homing, CLA1 memory T cells are activated in atopic dermatitis and regulate IgE by an IL-13dominated cytokine pattern. IgG4 counter-regulation by CLA– memory T cells. J Immunol 159:4611–4619, 1997 Akdis CA, Blesken T, Akdis M, Wu¨thrich B, Blaser K: The role of IL-10 in specific immunotherapy. J Clin Invest 102:98–106, 1998 Akdis M, Simon H-U, Weigl L, Kreyden O, Blaser K, Akdis CA: Skin homing (Cutaneous Lymphocyte-Associated Antigen-positive) CD81 T cells respond to superantigen and contribute to eosinophilia and IgE production in atopic dermatitis. J Immunol 163:466–475, 1999 Aman MJ, Tayebi N, Obiri NI, Puri RK, Modi WS, Leonard WJ: cDNA cloning and characterization of the human interleukin 13 receptor alpha chain. J Biol Chem 271:29265–29270, 1996 Bousquet J, Chanez P, Lacoste JY, et al: Eosinophilic inflammation in asthma. N Engl J Med 323:1033–1039, 1990 Fujimura T, Yamanashi R, Masuzava M, et al: Conversion of the CD41 T cell profile from Th2-dominant type to Th1-dominant type after varicella-zoster virus infection in atopic dermatitis. J Allergy Clin Immunol 100:274–282, 1997 Gru¨nig G, Warnock M, Wakil AE, et al: Requirement of IL-13 independently of IL-4 in experimental asthma. Science 282:2261–2263, 1998 Hamid Q, Boguniewicz M, Leung DYM: Differential in situ cytokine gene expression in acute versus chronic atopic dermatitis. J Clin Invest 94:870–876, 1994 Hanifin JM, Rajka G: Diagnostic features of atopic dermatitis. Acta Derm Venereol (Stockh) 92:44–47, 1980 Herz U, Schnoy N, Borelli S, et al: A hu-SCID mouse model for allergic immune responses: Bacterial superantigen enhances skin inflammation and suppresses IgE production. J Invest Dermatol 110:224–231, 1998 Jujo K, Renz H, Abe J, Gelfand EW, Leung DYM: Decreased interferon gamma and increased interleukin-4 production in atopic dermatitis promotes IgE synthesis. J Allergy Clin Immunol 90:323–331, 1992 Jung T, Wijdenes J, Neumann C, De Vries JE, Yssel H: Interleukin-13 is produced by activated human CD45RA1 and CD45RO1 T cells: modulation by interleukin-4 and interleukin-12. Eur J Immunol 26:571–577, 1996 Ka¨gi MK, Wu¨thrich B, Montano E, Barandun J, Blaser K, Walker C: Differential cytokine profiles in peripheral blood supernatants and skin biopsies from patients with different forms of atopic dermatitis, psoriasis and normal individuals. Int Arch Allergy Immunol 103:332–340, 1994 Kapp A, Piskorski A, Scho¨pf E: Elevated levels of interleukin 2 receptor in sera of patients with atopic dermatitis. Br J Dermatol 119:707–710, 1988
634
AKDIS ET AL
Koblenzer C: Stress and the skin: psychosomatic concepts in dermatology. Arch Dermatol 119:501–512, 1983 Kotzin BL, Leung DYM, Kappler J, Marrack P: Superantigens and human disease. Adv Immunol 54:99–166, 1993 Le Gros G, Ben Sason SZ, Seder R, Finkelman FD, Paul WE: Generation of IL-4 producing cells in vivo and in vitro: IL-2 and IL-4 are required for the in vitro generation of IL-4 producing cells. J Exp Med 172:921–929, 1990 Leung DYM, Gately M, Trumble A, Ferguson-Darnell B, Schlievert PM, Picker LJ: Bacterial superantigens induce T cell expression of the skin-selective homing receptor, the cutaneous lymphocyte-associated antigen, via stimulation of interleukin 12 production. J Exp Med 181:747–753, 1995 Leyden JE, Marpies RR, Kligman AM: Staphylococcus aureus in the lesions of atopic dermatitis. Br J Dermatol 90:525–530, 1974 Morren MA, Przybilla B, Bamelis M, Heykants B, Reynaers A, Degreef H: Atopic dermatitis: triggering factors. J Am Acad Dermatol 31:467–473, 1994 Nakazawa M, Sugi N, Kawaguchi H, Ishii N, Nakajima H, Minami M: Predominance of type 2 cytokine-producing CD41 and CD81 cells in patients with atopic dermatitis. J Allergy Clin Immunol 99:673–682, 1997 Oehling AG, Akdis CA, Schapowal A, Blaser K, Schmitz M, Simon H-U: Suppression of the immune system by oral glucocorticoid therapy in bronchial asthma. Allergy 52:144–152, 1997 Picker LJ, Michie SA, Rott LS, Butcher EC: A Unique phenotype of skin associated lymphocytes in humans: preferential expression of the HECA-452 epitope by benign and malignant T-cells at cutaneous sites. Am J Pathol 136:1053– 1061, 1990 Picker LJ, Kishimoto TK, Smith CW, Warnock RA, Butcher EC: ELAM-1 is an adhesion molecule for skin homing T cells. Nature 349:796–799, 1991 Punnonen J, Aversa G, Cocks BG, et al: Interleukin 13 induces interleukin 4 independent IgG4 and IgE synthesis and CD23 expression by human B cells. Proc Natl Acad Sci USA 90:3730–3734, 1993 Reinhold U. T cell mediated immunoregulation in atopic dermatitis. In: Wu¨thrich B (ed.). Highlights in Allergy and Clinical Immunology. Seattle: Hogrefe and Huber, 1992, p 3 Rousset F, Robert J, Andary M, et al: Shifts in interleukin-4 and interferon-γ production by T cells of patients with elevated serum IgE levels and the modulatory effects of these lymphokines on spontaneous IgE synthesis. J Allergy Clin Immunol 87:58–69, 1991 Rudikoff D, Lebwohl M: Atopic dermatitis. Lancet 351:1715–1721, 1998 Ruzicka T, Bieber T, Scho¨pf E, et al: A short-term trial of tacrolimus ointment for atopic dermatitis. N Engl J Med 337:816–821, 1997 Sager N, Feldmann A, Schilling G, Kreitsch P, Neumann C: House dust-mite specific T cells in the skin of subjects with atopic dermatitis: frequency and lymphokine profile in the allergen patch test. J Allergy Clin Immunol 89:801– 810, 1992 Simon H-U, Yousefi S, Schranz C, Schapowal A, Bachert C, Blaser K: Direct demonstration of delayed eosinophil apoptosis as a mechanism causing tissue eosinophilia. J Immunol 158:3902–3908, 1997
THE JOURNAL OF INVESTIGATIVE DERMATOLOGY
Strange P, Skov L, Lisby S, Nielsen PL, Baadsgaard O: Staphylococcal enterotoxin B applied on intact normal and intact atopic skin induces dermatitis. Arch Dermatol 132:27–33, 1996 Tony HP, Shen BJ, Reusch P, Sebald W: Design of human interleukin-4 antagonists inhibiting interleukin-4 dependent and interleukin-13 dependent responses in T-cells and B-cells with high efficiency. Eur J Biochem 225:659–665, 1994 Van der Heijden FL, Wierenga EA, Bos JD, Kapsenberg ML: High frequency of IL-4-producing CD41 allergen-specific T lymphocytes in atopic dermatitis lesional skin. J Invest Dermatol 97:389–394, 1991 Van der Pouw-Kraan TC, Aalberse RC, Aarden LA: IgE production in atopic patients is not related to IL-4 production. Clin Exp Immunol 97:254–259, 1994 Van Reijsen FC, Bruijnzeel-Komen CAFM, Kalthoff FS, Maggi E, Romagnani S, Westland JKT, Mudde GC: Skin-derived aero-allergen specific T cell clones of Th2 phenotype in patients with atopic dermatitis. J Allergy Clin Immunol 90:184–193, 1992 Varney VA, Hamid QA, Gaga M, et al: Influence of grass pollen immunotherapy on cellular infiltration and cytokine mRNA expression during allergen-induced late-phase cutaneous responses. J Clin Invest 92:644–651, 1993 Walker C, Ka¨gi MK, Ingold P, Braun P, Blaser K, Bruijnzeel-Koomen CAFM, Wu¨thrich B: Atopic dermatitis: correlation of peripheral blood T cell activation, eosinophilia and serum factors with clinical severity. Clin Exp Allergy 23:145– 153, 1993 Wedi B, Raap U, Lewrick H, Kapp A: Delayed eosinophil programmed cell death in vitro: a common feature of inhalant allergy and extrinsic or intrinsic atopic dermatitis. J Allergy Clin Immunol 100:536–543, 1997 Werfel T, Morita A, Grewe M, Renz H, Wahn U, Krutmann J, Kapp A: Allergenspecificity of skin-infiltrating T-cells is not restricted to a type 2 cytokine pattern in chronic skin lesions of atopic dermatitis. J Invest Dermatol 107:871– 876, 1996 Williams HC, Burney PG, Hay RJ, et al: The U.K. Working Party’s Diagnostic Criteria for Atopic Dermatitis. I. Derivation of a minimum set of discriminators for atopic dermatitis. Br J Dermatol 131:383–396, 1994 Wills-Karp M, Luyimbazi J, Xu X, Schofield B, Neben TY, Karp KL, Donaldson DD: Interleukin 13: Central mediator of allergic asthma. Science 282:2258– 2261, 1998 Wu¨thrich B: Serum IgE in atopic dermatitis. Clin Allergy 8:241–248, 1978 Wu¨thrich B: Atopic dermatitis flare provoked by inhalant allergens. Dermatologica 178:51–53, 1989 Yousefi S, Hemmann S, Weber M, Ho¨lzer C, Hartung K, Blaser K, Simon H-U: IL-8 is expressed by human peripheral blood eosinophils. Evidence for increased secretion in asthma. J Immunol 154:5481–5490, 1995 Zhang X, Polla B, Hauser C, Zubler RH: T cells from atopic individuals produce an IgE inducing activity incompletely blocked by anti-interleukin-4 antibody. Eur J Immunol 22:829–833, 1992 Zurawski G, De Vries JE: Interleukin 13, an interleukin 4-like cytokine that acts on monocytes and B cells, but not on T cells. Immunol Today 15:19–26, 1994
A subgroup of patients with atopic dermatitis are known to have normal serum total immunoglobulin E levels, undetectable specific immunoglobulin E, and negative skin prick tests towards allergens. This form of the disease has been termed nonallergic atopic dermatitis. In this study, we found that, among 1151 chronic atopic dermatitis patients, about 10% had normal serum immunoglobulin E levels with no evidence for immunoglobulin E sensitization. We investigated immunologic mechanisms of patients with ‘‘allergic’’ and ‘‘nonallergic’’ atopic dermatitis using peripheral blood and skin biopsy samples. Our data suggest that T cells are likely involved in the pathogenesis of both forms of atopic dermatitis. Skin T cells equally responded to superantigen, staphylococcal enterotoxin B, and produced interleukin-2, interleukin-5, interleukin-13, and interferon-γ in both forms of the disease. Interleukin-4, however, was not detectable in the skin biopsies of both atopic dermatitis
types and was secreted in very low amounts by T cells cultured from the skin biopsies. Moreover, skin T cells from nonallergic atopic dermatitis patients expressed lower interleukin-5 and interleukin-13 levels compared with allergic atopic dermatitis patients. Accordingly, T cells isolated from skin biopsies of atopic dermatitis, but not from the nonallergic atopic dermatitis, induced high immunoglobulin E production in cocultures with normal B cells that was mediated by interleukin-13. In addition, B cell activation with high CD23 expression was observed in the peripheral blood of atopic dermatitis, but not nonallergic atopic dermatitis patients. These data suggest, although high numbers of T cells are present in lesional skin of both types, a lack of interleukin-13-induced B cell activation and consequent immunoglobulin E production in nonallergic atopic dermatitis. Key words: atopic dermatitis/ cytokines/immunoglobulin E/superantigen/T cells. J Invest Dermatol 113:628–634, 1999
A
1993; Hamid et al, 1994; Akdis et al, 1997; Fujimura et al, 1997; Nakazawa et al, 1997). Although most of the patients with AD show high concentrations of total and allergen-specific IgE in blood and skin, some of the patients have normal total IgE levels and negative serum allergenspecific IgE (Wu¨thrich, 1978). Diagnostic criteria of AD by Hanifin and Rajka (1980) or Williams et al (1994) can be fulfilled in the absence of elevated total IgE and specific IgE to food or environmental allergens. This suggests that elevated IgE levels and IgE sensitization are not prerequisites in the pathogenesis of the disease. The subgroup of AD patients with normal IgE levels and without specific IgE sensitization has been termed nonallergic form of AD (NAD) or intrinsic-type AD (Ka¨gi et al, 1994). In this study, we reserved the term ‘‘AD’’ only for the patients with features of atopy. Patients were classified into the NAD group when they fulfilled the criteria of Hanifin and Rajka (1980), but did not exhibit any allergen-specific IgE sensitization and had normal serum total IgE levels. In this study, we investigated immunoregulatory mechanisms that are possibly involved in the pathogenesis of NAD in comparison with AD. The results suggest that T cells and T cell-derived cytokines do not only play an important part in allergic AD, but also in NAD. One important difference between the two forms of AD appears to be the different capacity of skin T cells to induce IgE production in B cells, possibly due to decreased interleukin (IL) -13 production in NAD.
topic dermatitis (AD) is a chronic relapsing inflammatory skin disease characterized by typically distributed eczematous skin lesions with lichenification, pruritic excoriations, severely dry skin, and a susceptibility to cutaneous infections (Hanifin and Rajka, 1980; Rudikoff and Lebwohl, 1998). Several lines of evidence suggest the contribution of immunologic mechanisms in the pathogenesis of AD. The skin lesions of AD patients are infiltrated by activated T cells, eosinophils, and antigen-presenting Langerhans cells that bind and present immunoglobulin (Ig) Ecomplexed allergens on their surface (Reinhold, 1992; Ka¨gi et al, 1994). In addition, activation of peripheral blood T cells that preferentially secrete T helper (Th) 2 cytokines and help B cells to produce large amounts of immunoglobulin (Ig) E has previously been reported to be associated with this skin disease (Walker et al,
Manuscript received March 18, 1999; revised June 10, 1999; accepted for publication June 14, 1999. Reprint requests to: Dr. Cezmi A. Akdis, Swiss Institute of Allergy and Asthma Research, Obere Strasse 22, CH-7270 Davos, Switzerland. Email: [email protected] Abbreviations: AD, atopic dermatitis; CLA, cutaneous lymphocyteassociated antigen; ECP, eosinophilic cationic protein; NAD, nonallergic form of AD; SEB, staphyloccoccal enterotoxin B.
·
0022-202X/99/$14.00 Copyright © 1999 by The Society for Investigative Dermatology, Inc.
628
VOL. 113, NO. 4 OCTOBER 1999
T CELLS AND CYTOKINES IN ALLERGIC AND NONALLERGIC ATOPIC DERMATITIS
MATERIALS AND METHODS Subjects In the time period from January 1997 until June 1998, we recruited 1151 patients among the three clinics in Davos and Allergy Unit of Zu¨rich University, Zu¨rich, Switzerland who fulfilled the diagnostic criteria of Hanifin and Rajka (1980). On the basis of the case history, total and specific IgE levels and skin prick tests, the subjects were further divided into AD and NAD patients. Characteristics of the AD patients were total serum IgE levels above 150 kU per liter and/or positive radioallergosorbent test or immunoassay (Pharmacia Diagnostics AG, Uppsala, Sweden), as well as a positive patient’s history for allergic diseases. Usually, these patients also had positive skin prick tests towards common environmental allergens. In many cases, however, skin prick tests could not be performed due to the severity of skin symptoms. Patients were termed NAD when they had a negative history for allergic diseases, low levels of total IgE (,150 kU per liter), no elevated specific IgE antibodies, and no immediate-type skin reactions towards a routine panel of relevant aeroallergens and food allergens. The possibility that the patients might have allergic contact dermatitis was ruled out by multiple evaluations of the patients at different times, disease history, clinical course, and features of the disease. Two of the patients were skin patch tested to exclude allergic contact dermatitis. For immunologic analysis, we obtained 10 ml of ethylenediamine tetraacetic acid anti-coagulated blood and skin biopsies from nine patients with AD (two female, seven male, mean age: 37 y) and eight patients with NAD (four female, four male, mean age: 49 y). One patient from the AD group and one patient from NAD group had asthma. Three patients with AD had allergic rhinoconjunctivitis. The clinical material was collected under standard hospital approved protocols. Skin biopsy specimens were obtained from acute AD lesions of less than 5 d onset. We also collected a group of six healthy individuals (three male, three female, mean age: 39 y) with no history or signs of atopy as a control group for immunophenotyping. All AD and NAD patients suffered from an acute exacerbation of their chronic disease. Duration of their disease was 18 6 4 y in AD and 16 6 6 y in NAD (mean 6 SEM). There was no significant difference in the age and age of onset between the AD versus NAD groups. Serum IgE levels were 12333 6 5663 kU per liter in AD and 42 6 13 kU per liter in NAD. The patients have not been treated with oral or topical steroids for at least 1 wk prior to vein puncture and skin biopsy. Informed consent was obtained from all patients and control individuals, and the study was approved by the Swiss Academy of Medical Science represented by the Medical Ethics Committee of Davos. Media, reagents, and antibodies Complete culture medium was RPMI 1640 supplemented with 2 mM L-glutamine, penicillin/streptomycin, and fetal bovine serum (all from Life Technologies, Basel, Switzerland). All conjugated monoclonal antibodies (MoAb) for flow-cytometric analysis were purchased from Coulter (Hialeah, FL), Immunotech Ltd. (Marseilles, France), or Pharmingen (San Diego, CA). Anti-CD2 (4B2 and 6G4) and anti-CD28 (15E8) MoAb were from the Red Cross Blood Transfusion Service in Amsterdam (the Netherlands). Anti-CD3 was produced by clone CRL 8001 obtained from ATCC (Washington, DC). Biotin-labeled MoAb and avidin was from Sigma (St. Louis, MO). Neutralizing anti-IL-4 MoAb (8F12) was provided by Novartis (Basel, Switzerland). Neutralizing anti-IL-13 MoAb (JES8–5A2 and JES10–2F9) and mutant IL-4 antagonist (Y124D; with inhibitory activity for both IL-4 and IL-13) were provided by DNAX Research Institute (Palo Alto, CA). The antibodies for immunohistochemistry, anti-IL-2, anti-interferon (IFN) -γ, and anti-IL-13 MoAb were from R&D Systems Europe Ltd. (Abingdon, U.K.). AntiIL-5 was a kind gift of Dr J. Tavernier (Gent, Belgium). Anti-cutaneous lymphocyte-associated antigen (CLA) MoAb was from Pharmingen, and anti-eosinophilic cationic protein (ECP; EG1) was from Pharmacia (Uppsala, Sweden). Anti-CD4 MoAb was obtained from Novocastra Laboratories (Newcastle, U.K.). Anti-CD8 and control MoAb were purchased from Dako (Zug, Switzerland). Analysis of leukocytes and lymphocyte subsets Blood leukocyte and differential counts were determined by automatic blood count analysis (Technicon H1, Technicon, Tarrytown, NY). To determine lymphocyte subsets, specific binding of MoAb was analyzed by direct immunofluorescence using standard methods (Oehling et al, 1997). The flow-cytometric analysis was performed by an EPICS XL (Coulter). Identification of the inflammatory cell infiltrate in the skin Inflammatory cells were identified using lineage-specific MoAb and immunohistochemistry. Briefly, fresh skin tissues were fixed in 4% paraformaldehyde solution. Paraffin sections were mounted on poly Llysin-coated slides, stored at room temperature, deparaffinized in xylene, and rehydrated through graded concentrations of ethanol before use.
629
Antigen retrieval was achieved by the microwave oven heating procedure. CD4 and CD8 stainings were performed with the Dako catalyzed signal amplification (CSA) system according to the manufacturer’s instructions (Dako). CLA stainings were done using the Zymed Lab-SA System (Zymed Laboratories, South San Francisco, CA). ECP stainings were performed using the APAAP technique as described previously (Simon et al, 1997). Purification and stimulation of skin T cells The skin biopsy specimens from four AD patients and four NAD patients were minced with two scalpels; no proteolytic enzymes were used. The T cells of the obtained skin biopsy fractions were stimulated in complete culture medium with 25 U IL-2 per ml and the following combination of MoAb to T cell surface molecules: anti-CD2 (4B2 and 6G4, each 0.5 µg per ml), antiCD3 (5 µg per ml), and anti-CD28 (1 µg per ml) (Akdis et al, 1999). After 7–10 d, this treatment resulted in an expansion of skin T cells. These cells were washed and used for functional assays (IgE production), as well as for proliferative response to super antigens and cytokine analysis by flow cytometry and enzyme-linked immunosorbent assay (ELISA). T cells isolated from skin of AD and NAD patients were stimulated with different doses of staphylococcal enterotoxin B (SEB) in 96-well flat bottomed plates (5 3 104/200 µl per well) in triplicates by using rad irradiated autologous peripheral blood mononuclear cells (PBMC) (5 3 104) as antigen-presenting cells. [3H]thymidine incorporation was measured after 3 d following an 8 h incubation with 1 µCi per well of [3H]thymidine. The plates were harvested on glass-fiber plates and radioactivity was measured in a Beta Plate Reader (Pharmacia-Wallac, Turku, Finland). For the determination of cytokine production, the in vitro-expanded skin T cells (0.5 3 106 per ml; 48-well plates in triplicates) were again stimulated with anti-CD2, anti-CD3, and anti-CD28 MoAb for 72 h, and the supernatants were harvested and frozen at –80°C until ELISA measurements. For intracellular staining of cytokines, the cells were stimulated with the same combination of MoAb for 16 h before flow cytometric analysis. Measurements of cytokine levels ELISA The solid-phase sandwich ELISA for IFN-γ, IL-4, IL-5, and IL-10 were performed as described previously (Akdis et al, 1996, 1997, 1998). IL-13 and granulocyte-macrophage colony-stimulating factor were measured using commercial kits according to the manufacturer’s instructions (R&D Systems). Immunohistochemistry All cytokine stainings were performed using the APAAP technique as previously described (Yousefi et al, 1995; Simon et al, 1997). Sections were lightly counterstained with hematoxylin, mounted, and examined under a Zeiss Axioscope at a magnification of 3400 or 31000. The intensities of the stainings were evaluated by two independent investigators from low (0) to high (6). Flow cytometry During the last 14 h of cell culture, 1 µM monensin was added. The cells were washed with phosphate-buffered saline and surfacestained with phycoerythrin Texas red-labeled anti-CD4 or anti-CD8 MoAb, followed by fixing and permeabilizing of the cells with a formaldehyde/saponin solution (Ortho Permeafix, Ortho Diagnostic Systems, Raritan, NJ). After washing with phosphate-buffered saline containing 5% fetal bovine serum, 1.5% bovine serum albumin, and 0.0055% ethylenediamine tetraacetic acid, the cells were counter-stained with 0.5 µg per ml phycoerythrin- or fluorescein isothiocyanate-conjugated control, anti-IL-4, anti-IL-5, anti-IL-13, and anti-IFN-γ MoAb for 30 min at 4°C. The flowcytometric analysis was performed using an EPICS XL (Coulter). IgE production in vitro PBMC from normal control individuals were depleted from CD3-, CD4-, and CD8-bearing cells by negative selection using a magnetic cell separation system (MACS; Miltenyi Biotec, BergischGladbach, Germany) as described previously (Akdis et al, 1997). T celldepleted PBMC (0.9 3 105) were cocultured with anti-CD2-, anti-CD3-, and anti-CD28-stimulated skin T cells (0.1 3 105) derived from four AD and four NAD patients. IgE was determined in supernatants harvested after 12 d. Inhibition of IgE production was attempted by addition of neutralizing MoAb to IL-4 or IL-13 (both at 10 µg per ml) (Akdis et al, 1997). The IL-4 and IL-13 antagonist Y124D was used at 100 ng per ml (Zurawski and De Vries, 1994). Mouse IgG1 (10 µg per ml) was used as a control. IgE concentrations were measured by ELISA as described previously (Akdis et al, 1996, 1997). All experiments were performed in triplicate. Statistics All data are expressed as means 6 SD. Statistical analysis was performed using the Student’s t test (*p , 0.05).
630
AKDIS ET AL
THE JOURNAL OF INVESTIGATIVE DERMATOLOGY
Table I. Peripheral blood leukocyte and lymphocyte subpopulationsa Controls
AD
NAD
Leukocytes (per µl) 6850 6 1331 7544 6 2952 7525 6 3318 Lymphocytes (%) 34 6 8 37 6 19 36 6 7 74 6 4 74 6 5 73 6 8 CD31 (%) 47 6 6 50 6 6 45 6 15 CD31CD41 (%) 23 6 3 23 6 6 25 6 12 CD31CD81 (%) CD4/CD8 ratio 2.1 6 0.6 2.4 6 0.8 2.5 6 1.8 11 6 4 664 10 6 7 CD161 (%) 10 6 6 15 6 4 13 6 5 CD191 (%) 1 1 765 14 6 12 763 CD4 HLA-DR (%) 18 6 6 26 6 13 27 6 9 CD41CD251 (%) 664 16 6 24 19 6 16 CD81HLA-DR1 (%) 161 7 6 17 262 CD81CD251 (%) 1 1 24 6 16 67 6 12* 21 6 12 CD19 CD23 (%) Monocytes (%) 463 662 764 96 6 5 94 6 4 93 6 5 CD141HLA-DR1 (%) Neutrophils (%) 60 6 8 47 6 14 53 6 9 Eosinophils (%) 261 10 6 5* 4 6 3* aThese data are mean 6 SD from six normal control individuals, nine AD patients, and eight NAD patients. *Significantly different values from those of normal control individuals (p , 0.05).
RESULTS Peripheral blood leukocyte and lymphocyte subpopulations in AD and NAD We investigated 1151 patients who fulfilled the criteria of Hanifin and Rajka (1980) within a time period of 18 mo. From these patients, 116 NAD patients were identified. Accordingly, in this study about 10% of all atopic dermatitis patients belonged to the nonallergic group. The immunologic data described here were obtained from nine AD and eight NAD patients. The leukocyte differential counts revealed a characteristic blood eosinophilia in both types of atopic dermatitis. It was more pronounced in AD compared with NAD patients, although it did not show any statistically significant difference (Table I). Immunophenotyping using flow cytometry revealed increased numbers of CD231 B cells in AD, but not in NAD patients, in agreement with previously published work (Wu¨thrich, 1989; Ka¨gi et al, 1994). In contrast, CD41 and CD81 T cells expressed increased activation markers (HLA-DR and CD25) in both AD and NAD patients, although the differences had no statistical significance. T cells infiltrate into the skin of AD and NAD patients To identify the inflammatory cells present in lesional skin, we performed immunohistochemistry using skin biopsies from AD and NAD patients. T cells were identified with anti-CD4, anti-CD8, and anti-CLA MoAb, whereas eosinophils were stained with an antiECP MoAb. As shown in Fig 1 and Table II, AD and NAD patients exhibited similar cellular infiltrates in their lesional skins. The majority of the cells represented CD41 and CD81 cells, suggesting an important role for T cells in both groups. The CD4/ CD8 ratios between blood and skin appeared to be similar, suggesting that both CD41 and CD81 T subpopulations are equally recruited into the inflammatory sites of AD and NAD patients. In allergic AD, the pivotal role of CD45RO1 (memory/effector) T cells expressing the skin homing receptor, the CLA was demonstrated. Almost all T cells in benign and malignant cell infiltrations of the skin express CLA molecule on their surface (Picker et al, 1990, 1991). As previously observed in AD and other dermatologic diseases (Picker et al, 1990), most of the skin-infiltrating T cells of NAD patients were CLA positive. Eosinophil content of skin lesions were low in comparison with T cells. ECP positive cells were 15.1 6 18.3 per mm2 skin in AD and 8.3 6 11.4 per mm2 skin in NAD (mean 6 SD) (Table II). In vivo expression of cytokines in lesional skin of AD and NAD patients It has been repeatedly suggested that Th2
Figure 1. Inflammatory cells in lesional skin of AD and NAD patients. The majority of the cells represented CD41 and CD81 T cells. Many of the T cells were CLA1. The eosinophil infiltration was much less in comparison with the T cell infiltration. No differences were observed between AD and NAD patients. Data are shown in Table II.
cytokines play an important part in AD and other atopic diseases (Hamid et al, 1994; Akdis et al, 1997; Fujimura et al, 1997; Nakazawa et al, 1997). Therefore, we investigated the Th1/Th2 cytokine pattern of AD and NAD patients using immunohistochemistry. As shown in Fig 2, Table II, both Th1 (IL-2 and IFN-γ) and Th2 (IL-5 and IL-13) cytokines were expressed by inflammatory cells of AD and NAD patients. Higher IL-5 and IL-13 expressions were observed in AD than in NAD, although it did not reach statistical significance. As the cellular infiltrate mostly contained T cells (Table II), it is likely that these cells were the main cellular source for these cytokines. Interestingly, IL-4 was neither detectable in skin lesions of AD nor NAD patients. Superantigen response and cytokine expression by skin T cells isolated from lesional skin of AD and NAD patients To further investigate the expression of cytokines by skin T cells, we measured cytokine levels by flow cytometry and ELISA. Owing to the low cell numbers obtained from skin biopsies, however, T cells had to be expanded in vitro. Cells were stimulated with a combination of anti-CD2, anti-CD3, and anti-CD28 MoAb in the presence of IL-2 for optimal T cell proliferation. The in vitroexpanded T cells were then analyzed for intracytoplasmic cytokine expression. As assessed by flow cytometry, T cells from both AD
VOL. 113, NO. 4 OCTOBER 1999
T CELLS AND CYTOKINES IN ALLERGIC AND NONALLERGIC ATOPIC DERMATITIS
631
Table II. Inflammatory cells and cytokines in lesional skin of AD and NAD patientsa
AD NAD
AD NAD
CD4
CD8
CD4/CD8
ECP
CLA (%)
579 6 222 527 6 305
290 6 127 274 6 147
2.11 6 1.07 2.00 6 0.66
15.1 6 18.3 8.3 6 11.4
62.9 6 23.5 59.3 6 23.9
IL-2
IFN-γ
IL-4
IL-5
IL-13
2.75 6 1.83 2.13 6 0.99
1.00 6 1.07 0.50 6 0.76
0.00 0.00
2.50 6 1.77 1.13 6 1.25
3.00 6 1.20 2.13 6 1.36
aThe skin-infiltrating cells were determined by immunohistochemistry per mm2 skin section as described in Materials and Methods. These data are mean 6 SD from nine AD patients, and eight NAD patients The intensity of the cytokine stainings were independently evaluated by two investigators (M.W. and S.G.) from low (0) to high (6) grade.
responsiveness to superantigenic stimuli released by Staphylococcus aureus colonizing the skin. This was analyzed by stimulating the T cells from skin biopsies of AD and NAD patients with SEB, by using irradiated autologous PBMC as antigen-presenting cells. As shown in Fig 4(A), T cells of both AD and NAD patients responded to SEB in a concentration-dependent manner. SEB-specific stimulation only involves T cells that express certain Vβ elements of the TCR. To better quantitate the cytokines expressed by skin T cells, in vitro-expanded skin T cells were stimulated with the combination of anti-CD2/anti-CD3/anti-CD28 MoAb for 72 h, and released cytokines were measured by ELISA. As shown in Fig 4(B), increased IL-5 and IL-13 concentrations were measured in the supernatants of AD compared with NAD patients. T cells from NAD patients, however, also produced detectable amounts of these two cytokines. In contrast, the released amounts of IFN-γ and IL-4 were less, and no significant differences were found between AD and NAD patients. Skin T cells help B cells to produce high amounts of IgE in AD but not NAD As we had evidence for increased IL-13 expression in AD compared with NAD patients, we hypothesized that the threshold to help B cells to produce high amounts of IgE may be reached by skin T cells from AD, but not NAD patients. As shown in Fig 5, no significant IgE production is observed in the absence of T cells. Skin T cells from AD patients induced normal blood B cells to produce significantly high amounts of IgE in comparison with NAD. Whereas the B cell help of skin T cells from AD patients could be neutralized by anti-IL-13 MoAb and recombinant IL-4/IL-13 antagonist, however, the marginal induced IgE production by NAD T cells could not be blocked by these reagents. Neutralizing IL-4 by anti-IL-4 MoAb had no effect in this system. Together, these data suggest that IL-13 is an important cytokine involved in increased IgE production in AD patients. Although T cells from NAD patients also produced significant amounts of IL-13, these levels may not have been elevated enough to stimulate increased IgE production. DISCUSSION Figure 2. IL-13, but not IL-4, is expressed in lesional skin of AD and NAD patients. The positive control for IL-4 was a bladder cancer tissue expressing IL-4. In average, the expression of IL-13 was less in NAD in comparison with AD patients. All data regarding cytokine expression are shown in Table II.
and NAD patients did not express detectable amounts of IL-4 protein (Fig 3), confirming the data obtained by immunohistochemistry (Fig 2, Table II). Moreover, significant amounts of IL-5 were usually detected in T cells from AD (CD41 . CD81), but not NAD patients (Fig 3). In contrast, IFN-γ was frequently detected in T cells from NAD (CD81 . CD41), but not AD patients (Fig 3). Furthermore, under these experimental conditions, we observed increased expression of IL-13 in AD compared with NAD patients. Because NAD patients are not sensitized to aeroallergens and food antigens, one explanation for the existence and in vivo activation of T cells in skin lesions of NAD could be their
A number of immunologic parameters were demonstrated to be abnormal and related to the clinical severity of allergic and nonallergic forms of AD (Kapp et al, 1988; Rousset et al, 1991; Jujo et al, 1992; Walker et al, 1993; Hamid et al, 1994; Ka¨gi et al, 1994; Morren et al, 1994; Akdis et al, 1997; Fujimura et al, 1997; Nakazawa et al, 1997). Based on skin biopsies, the present study demonstrates that T cells are important pathogenetic factors in both forms of the disease. The majority of cells in the cellular infiltrate in AD were T cells. The numbers of infiltrating eosinophils compared with T cells were marginal in both AD and NAD patients. This is in contrast to other eosinophilic tissues from patients with bronchial asthma (Bousquet et al, 1990) or nasal polyps (Simon et al, 1997). Allergen-specific T cells derived from skin lesions in patients with the allergic variant of AD were found to produce predominantly type 2 Th cytokines (Sager et al, 1992; Van Reijsen et al, 1992). These studies were mainly performed by T cells cloned after epicutaneous application of inhalant allergens. In contrast,
632
AKDIS ET AL
THE JOURNAL OF INVESTIGATIVE DERMATOLOGY
Figure 3. Intracytoplasmic cytokine content of CD4F and CD8F skin T cells following stimulation with antiCD2, anti-CD3, and anti-CD28 MoAb for 16 h. The results are representative of four independent experiments in each group of patients. The skin T cells of AD patients expressed much more IL-5 and IL-13 compared with NAD patients. IL-4 expressing cells were ø 3% in both AD and NAD patients. The expression of IFN-γ was lower in AD in comparison with NAD patients.
allergen-specific T cell clones obtained from peripheral blood of the same patients (Sager et al, 1992) and allergen-specific T cell clones from chronic lesions and older patch tests displayed IFN-γ (Werfel et al, 1996). It appears that, the age of the patient population, the age of the eczema lesion (Hamid et al, 1994), the isolation site of T cells and T cell cloning procedure might influence the cytokine pattern of T cells. The present study demonstrates by immunohistochemistry, flow cytometry, and ELISA measurements, that, skin T cells from AD and NAD patients produce significant amounts of IL-5 and IL-13. In all three test systems, AD patients expressed higher levels of these two cytokines than NAD patients. In both groups of patients, IL-4 was not significantly expressed by skin T cells. In contrast to IL-4, IL-13, that can replace IL-4 in the induction of IgE mRNA (Punnonen et al, 1993), was detected in skin lesions of both forms of AD; however, The overall IL-13 expression was lower in NAD compared with AD patients. IL-5 was detected in many, but not in all AD and NAD patients. The Th1 cytokines IL-2 and IFN-γ were both detected in lesional skin, although the IFN-γ levels were relatively low in both AD and NAD patients and was slightly increased in NAD compared with AD patients. Similarly, high amounts of intracytoplasmic IL-5 and IL-13 were detected in activated skin homing CLA1 T cells of AD patients purified from peripheral blood (Akdis et al, 1997, 1999). The leukocyte differential blood counts revealed a characteristic blood eosinophilia in both AD and NAD patients, that was more pronounced in AD compared with NAD patients. High IL-5 production in both forms of AD and even higher levels in allergic form of AD may suggest a part for blood eosinophilia in both diseases as demonstrated by delayed apoptosis of blood eosinophils (Wedi et al, 1997). The functional IgE regulatory capacity of T cells from skin lesions of AD patients was investigated by coculturing isolated skin T cells with purified B cells. The IgE production by B cells mainly depended on secreted IL-13. A number of studies have shown that IL-4 plays an important part in the induction of IgE synthesis in AD in PBMC cultures (Rousset et al, 1991; Jujo et al, 1992; Morren et al, 1994). Previous reports and our present findings, however, indicate that IL-4 inhibition is not sufficient to suppress IgE in atopic diseases. An IgE-inducing activity which could be attributed to IL-13, was found in PBMC cultures of atopic but not normal individuals (Zhang et al, 1992; Van der Pouw-Kraan et al, 1994). In contrast to IL-4, there is no feedback or priming mechanism exerted by IL-13 on T cells, because human T cells do not display specific IL-13 receptors (Zurawski et al, 1994).
Figure 4. Superantigen response and cytokine profile of skin T cells in AD and NAD. There was no difference in SEB induced proliferation between skin T cells of AD and NAD patients. IL-5 and IL-13 production following stimulation with anti-CD2, anti-CD3, and anti-CD28 MoAb for 72 h of the skin T cells from AD patients was significantly high compared with NAD patients. The levels of released IFN-γ and IL-4 were low, and no differences were observed between AD and NAD patients. The data are presented as mean 6 SD of four independent experiments in each patient group (*p , 0.05).
VOL. 113, NO. 4 OCTOBER 1999
T CELLS AND CYTOKINES IN ALLERGIC AND NONALLERGIC ATOPIC DERMATITIS
Figure 5. Skin T cells from both AD and NAD patients induce IgE production in T cell-depleted PBMC populations (T Depl. PBMC). Only the skin T cells from AD patients helped B cells to produce very high amounts of IgE. The induced IgE production by T cells from AD patients could be blocked with neutralizing anti-IL-13 MoAb and recombinant IL-4/IL-13 antagonist (Y124D), suggesting that IL-13 may play an important part in the induction of IgE synthesis. Results represent means of triplicate cultures 6 SD from four independent experiments in each patient group (*p , 0.05).
Similarly, a central role for IL-13 in the expression of murine asthma is demonstrated (Gru¨nig et al, 1998; Wills-Karp et al, 1998), suggesting that IL-4 may be decisive at the initial phase of allergic responses and in the priming and development of Th2 cells (Le Gros et al, 1990; Jung et al, 1996), whereas IL-13 becomes more prominent in IgE induction in chronic AD. In addition, IgE Fc receptor II (CD23) on B cells is significantly expressed in AD compared with NAD (Ka¨gi et al, 1994). It is known that both IL-4 and IL-13 may upregulate this receptor (Zurawski et al, 1994). As IL-4 is not expressed in skin T cells, as demonstrated by immunohistochemistry and intracytoplasmic cytokine staining, IL-13 appears to be the major cytokine in the activation of B cells and IgE production in AD. Although IL-13 was produced in higher amounts in both AD and NAD, B cells of healthy individuals did not respond to the IL-13 produced from skin T cells of NAD. This can be ascribed to IL-13 produced over a threshold to activate B cells in AD. Furthermore, the genetical variations on the quality of IL-13 produced might play a part in increased activity of the cytokine. Moreover, the IL-4–IL-13 receptor common α chain has been demonstrated to have mutations that might induce high B cell responsiveness to IL-13 in AD (Tony et al, 1994; Aman et al, 1996). A number of pathogenetic mechanisms leading to T cell activation in AD including aeroallergens, food allergens, and superantigens have been emphasized. The role of aeroallergens and food allergens in the exacerbation of AD by T cell activation has been extensively studied (Sager et al, 1992; Van der Heijden et al, 1991; Varney et al, 1993; Abernathy-Carver et al, 1995). It is apparent that allergenspecific T cell responses in food and aeroallergen allergy are confined to AD and does not explain the activation and recruitment of T cells in NAD skin lesions. The present study shows that T cells isolated from the skin of AD and NAD patients efficiently proliferate by superantigenic stimulation. From a number of studies it could be concluded that bacterial superantigens contribute to the pathogenesis and exacerbation of AD. Staphylococcal superantigens were isolated from AD skin (Leyden et al, 1974; Kotzin et al, 1993). Superantigen patch test elicits skin inflammation in AD patients (Strange et al, 1996) and in human severe combined immunodeficiency mouse model (Herz et al, 1998). In addition, superantigens upregulate the skin homing ligand (Leung et al, 1995). Together, these data suggest that bacterial superantigens can lead to activation and proliferation of T cells in NAD skin as well as AD skin. The clinical manifestations of NAD were the same as AD suggesting that mechanisms causing eczema is independent of IgE overproduction. Provocation factors other than antigens and
633
superantigens such as emotional factors (Koblenzer, 1983), irritating substances (Agner, 1991), hormones, and climatic changes (Morren et al, 1994) were shown to influence the course of eczema. Taken together, T cells appear to be important players regulating the inflammatory processes in both AD and NAD patients. This hypothesis is further supported by the observation that drugs, such as FK-506, that block the activation of T cells, are effective in AD treatment (Ruzicka et al, 1997). In conclusion, we established a role for T cells in the nonallergic form of AD. Although the cellular infiltrate in lesional skin of both forms of AD patients contained high numbers of T cells, it appeared that the expression of cytokines IL-5 and IL-13 as well as the capacity to produce these cytokines by skin T cells is reduced in NAD patients. T cells that express low IL-13 are unable to help B cells to produce large amounts of IgE in NAD patients. The reasons for the different cytokine expression patterns observed between the two forms of AD patients could involve underlying genetic differences or additional factors that limit IL-5 and IL-13 production at the inflammatory site in NAD patients.
We are grateful to Dr. Jan de Vries and Dr. Gerard Zurawsky (DNAX Research Institute, Palo Alto, CA) for anti-IL-13 MoAb and Y124D mutant IL-4. We also thank Dr. Christoph Heusser and Dr. Sefik S. Alkan (Novartis, Basel, Switzerland) for anti-IL-4 and anti-IFN-γ MoAb, as well as Dr. Jan Tavernier (University of Gent, Gent, Belgium) for anti-IL-5 MoAb. This work was supported by grants 32-49210.96 and 31-50590.97/1, both from the Swiss National Science Foundation.
REFERENCES Abernathy-Carver KJ, Sampson HA, Picker LJ, Leung DYM: Milk-induced eczema is associated with the expansion of T cells expressing cutaneous lymphocyte antigen. J Clin Invest 95:913–918, 1995 Agner T: Susceptibility of atopic dermatitis patients to irritant dermatitis caused by sodium lauryl sulphate. Acta Derm Venereol (Stockh) 71:296–300, 1991 Akdis CA, Akdis M, Blesken T, Wymann D, Alkan SS, Mu¨ller U, Blaser K: Epitope specific T cell tolerance to phospholipase A2 in bee venom immunotherapy and recovery by IL-2 and IL-15 in vitro. J Clin Invest 98:1676–1683, 1996 Akdis M, Akdis CA, Weigl L, Disch R, Blaser K: Skin-homing, CLA1 memory T cells are activated in atopic dermatitis and regulate IgE by an IL-13dominated cytokine pattern. IgG4 counter-regulation by CLA– memory T cells. J Immunol 159:4611–4619, 1997 Akdis CA, Blesken T, Akdis M, Wu¨thrich B, Blaser K: The role of IL-10 in specific immunotherapy. J Clin Invest 102:98–106, 1998 Akdis M, Simon H-U, Weigl L, Kreyden O, Blaser K, Akdis CA: Skin homing (Cutaneous Lymphocyte-Associated Antigen-positive) CD81 T cells respond to superantigen and contribute to eosinophilia and IgE production in atopic dermatitis. J Immunol 163:466–475, 1999 Aman MJ, Tayebi N, Obiri NI, Puri RK, Modi WS, Leonard WJ: cDNA cloning and characterization of the human interleukin 13 receptor alpha chain. J Biol Chem 271:29265–29270, 1996 Bousquet J, Chanez P, Lacoste JY, et al: Eosinophilic inflammation in asthma. N Engl J Med 323:1033–1039, 1990 Fujimura T, Yamanashi R, Masuzava M, et al: Conversion of the CD41 T cell profile from Th2-dominant type to Th1-dominant type after varicella-zoster virus infection in atopic dermatitis. J Allergy Clin Immunol 100:274–282, 1997 Gru¨nig G, Warnock M, Wakil AE, et al: Requirement of IL-13 independently of IL-4 in experimental asthma. Science 282:2261–2263, 1998 Hamid Q, Boguniewicz M, Leung DYM: Differential in situ cytokine gene expression in acute versus chronic atopic dermatitis. J Clin Invest 94:870–876, 1994 Hanifin JM, Rajka G: Diagnostic features of atopic dermatitis. Acta Derm Venereol (Stockh) 92:44–47, 1980 Herz U, Schnoy N, Borelli S, et al: A hu-SCID mouse model for allergic immune responses: Bacterial superantigen enhances skin inflammation and suppresses IgE production. J Invest Dermatol 110:224–231, 1998 Jujo K, Renz H, Abe J, Gelfand EW, Leung DYM: Decreased interferon gamma and increased interleukin-4 production in atopic dermatitis promotes IgE synthesis. J Allergy Clin Immunol 90:323–331, 1992 Jung T, Wijdenes J, Neumann C, De Vries JE, Yssel H: Interleukin-13 is produced by activated human CD45RA1 and CD45RO1 T cells: modulation by interleukin-4 and interleukin-12. Eur J Immunol 26:571–577, 1996 Ka¨gi MK, Wu¨thrich B, Montano E, Barandun J, Blaser K, Walker C: Differential cytokine profiles in peripheral blood supernatants and skin biopsies from patients with different forms of atopic dermatitis, psoriasis and normal individuals. Int Arch Allergy Immunol 103:332–340, 1994 Kapp A, Piskorski A, Scho¨pf E: Elevated levels of interleukin 2 receptor in sera of patients with atopic dermatitis. Br J Dermatol 119:707–710, 1988
634
AKDIS ET AL
Koblenzer C: Stress and the skin: psychosomatic concepts in dermatology. Arch Dermatol 119:501–512, 1983 Kotzin BL, Leung DYM, Kappler J, Marrack P: Superantigens and human disease. Adv Immunol 54:99–166, 1993 Le Gros G, Ben Sason SZ, Seder R, Finkelman FD, Paul WE: Generation of IL-4 producing cells in vivo and in vitro: IL-2 and IL-4 are required for the in vitro generation of IL-4 producing cells. J Exp Med 172:921–929, 1990 Leung DYM, Gately M, Trumble A, Ferguson-Darnell B, Schlievert PM, Picker LJ: Bacterial superantigens induce T cell expression of the skin-selective homing receptor, the cutaneous lymphocyte-associated antigen, via stimulation of interleukin 12 production. J Exp Med 181:747–753, 1995 Leyden JE, Marpies RR, Kligman AM: Staphylococcus aureus in the lesions of atopic dermatitis. Br J Dermatol 90:525–530, 1974 Morren MA, Przybilla B, Bamelis M, Heykants B, Reynaers A, Degreef H: Atopic dermatitis: triggering factors. J Am Acad Dermatol 31:467–473, 1994 Nakazawa M, Sugi N, Kawaguchi H, Ishii N, Nakajima H, Minami M: Predominance of type 2 cytokine-producing CD41 and CD81 cells in patients with atopic dermatitis. J Allergy Clin Immunol 99:673–682, 1997 Oehling AG, Akdis CA, Schapowal A, Blaser K, Schmitz M, Simon H-U: Suppression of the immune system by oral glucocorticoid therapy in bronchial asthma. Allergy 52:144–152, 1997 Picker LJ, Michie SA, Rott LS, Butcher EC: A Unique phenotype of skin associated lymphocytes in humans: preferential expression of the HECA-452 epitope by benign and malignant T-cells at cutaneous sites. Am J Pathol 136:1053– 1061, 1990 Picker LJ, Kishimoto TK, Smith CW, Warnock RA, Butcher EC: ELAM-1 is an adhesion molecule for skin homing T cells. Nature 349:796–799, 1991 Punnonen J, Aversa G, Cocks BG, et al: Interleukin 13 induces interleukin 4 independent IgG4 and IgE synthesis and CD23 expression by human B cells. Proc Natl Acad Sci USA 90:3730–3734, 1993 Reinhold U. T cell mediated immunoregulation in atopic dermatitis. In: Wu¨thrich B (ed.). Highlights in Allergy and Clinical Immunology. Seattle: Hogrefe and Huber, 1992, p 3 Rousset F, Robert J, Andary M, et al: Shifts in interleukin-4 and interferon-γ production by T cells of patients with elevated serum IgE levels and the modulatory effects of these lymphokines on spontaneous IgE synthesis. J Allergy Clin Immunol 87:58–69, 1991 Rudikoff D, Lebwohl M: Atopic dermatitis. Lancet 351:1715–1721, 1998 Ruzicka T, Bieber T, Scho¨pf E, et al: A short-term trial of tacrolimus ointment for atopic dermatitis. N Engl J Med 337:816–821, 1997 Sager N, Feldmann A, Schilling G, Kreitsch P, Neumann C: House dust-mite specific T cells in the skin of subjects with atopic dermatitis: frequency and lymphokine profile in the allergen patch test. J Allergy Clin Immunol 89:801– 810, 1992 Simon H-U, Yousefi S, Schranz C, Schapowal A, Bachert C, Blaser K: Direct demonstration of delayed eosinophil apoptosis as a mechanism causing tissue eosinophilia. J Immunol 158:3902–3908, 1997
THE JOURNAL OF INVESTIGATIVE DERMATOLOGY
Strange P, Skov L, Lisby S, Nielsen PL, Baadsgaard O: Staphylococcal enterotoxin B applied on intact normal and intact atopic skin induces dermatitis. Arch Dermatol 132:27–33, 1996 Tony HP, Shen BJ, Reusch P, Sebald W: Design of human interleukin-4 antagonists inhibiting interleukin-4 dependent and interleukin-13 dependent responses in T-cells and B-cells with high efficiency. Eur J Biochem 225:659–665, 1994 Van der Heijden FL, Wierenga EA, Bos JD, Kapsenberg ML: High frequency of IL-4-producing CD41 allergen-specific T lymphocytes in atopic dermatitis lesional skin. J Invest Dermatol 97:389–394, 1991 Van der Pouw-Kraan TC, Aalberse RC, Aarden LA: IgE production in atopic patients is not related to IL-4 production. Clin Exp Immunol 97:254–259, 1994 Van Reijsen FC, Bruijnzeel-Komen CAFM, Kalthoff FS, Maggi E, Romagnani S, Westland JKT, Mudde GC: Skin-derived aero-allergen specific T cell clones of Th2 phenotype in patients with atopic dermatitis. J Allergy Clin Immunol 90:184–193, 1992 Varney VA, Hamid QA, Gaga M, et al: Influence of grass pollen immunotherapy on cellular infiltration and cytokine mRNA expression during allergen-induced late-phase cutaneous responses. J Clin Invest 92:644–651, 1993 Walker C, Ka¨gi MK, Ingold P, Braun P, Blaser K, Bruijnzeel-Koomen CAFM, Wu¨thrich B: Atopic dermatitis: correlation of peripheral blood T cell activation, eosinophilia and serum factors with clinical severity. Clin Exp Allergy 23:145– 153, 1993 Wedi B, Raap U, Lewrick H, Kapp A: Delayed eosinophil programmed cell death in vitro: a common feature of inhalant allergy and extrinsic or intrinsic atopic dermatitis. J Allergy Clin Immunol 100:536–543, 1997 Werfel T, Morita A, Grewe M, Renz H, Wahn U, Krutmann J, Kapp A: Allergenspecificity of skin-infiltrating T-cells is not restricted to a type 2 cytokine pattern in chronic skin lesions of atopic dermatitis. J Invest Dermatol 107:871– 876, 1996 Williams HC, Burney PG, Hay RJ, et al: The U.K. Working Party’s Diagnostic Criteria for Atopic Dermatitis. I. Derivation of a minimum set of discriminators for atopic dermatitis. Br J Dermatol 131:383–396, 1994 Wills-Karp M, Luyimbazi J, Xu X, Schofield B, Neben TY, Karp KL, Donaldson DD: Interleukin 13: Central mediator of allergic asthma. Science 282:2258– 2261, 1998 Wu¨thrich B: Serum IgE in atopic dermatitis. Clin Allergy 8:241–248, 1978 Wu¨thrich B: Atopic dermatitis flare provoked by inhalant allergens. Dermatologica 178:51–53, 1989 Yousefi S, Hemmann S, Weber M, Ho¨lzer C, Hartung K, Blaser K, Simon H-U: IL-8 is expressed by human peripheral blood eosinophils. Evidence for increased secretion in asthma. J Immunol 154:5481–5490, 1995 Zhang X, Polla B, Hauser C, Zubler RH: T cells from atopic individuals produce an IgE inducing activity incompletely blocked by anti-interleukin-4 antibody. Eur J Immunol 22:829–833, 1992 Zurawski G, De Vries JE: Interleukin 13, an interleukin 4-like cytokine that acts on monocytes and B cells, but not on T cells. Immunol Today 15:19–26, 1994