Contact Dermatitis and Allergy - Phosphodiesterase-4-Inhibitoren zur Behandlung allergischer Ha

TARC and RANTES, but not CTACK, are induced in two models of allergic contact dermatitis. Effects of cilomilast and diflorasone diacetate on T-cell-attracting chemokines

W . B A¨ U M E R , U . S E E G E R S , M . B R A U N , T . T S C H E R N I G * A N D M . K I E T Z M A N N

Department of Pharmacology, Toxicology and Pharmacy, School of Veterinary Medicine, Buenteweg 17, 30559 Hannover, Germany

*Department of Functional and Applied Anatomy, Medical School of Hannover, Carl-Neuberg-Str. 1, 30625 Hannover, Germany

Accepted for publication 15 April 2004

Summary Background Skin-infiltrating T cells play a predominant role in allergic and inflammatory skin diseases such as atopic dermatitis and allergic contact dermatitis. These T cells are attracted by chemotactic factors, e.g. RANTES (regulation on activation, normal T cell expressed and secreted;

CCL5), TARC (thymus and activation regulated chemokine; CCL17) and CTACK (cutaneous T-cell attracting chemokine; CCL27).

Objectives To investigate which T-cell-attracting chemokines are involved in allergic contact dermatitis in mice.

Methods Allergic contact dermatitis was induced by application of dinitrochlorobenzene (DNCB) or toluene-2,4-diisocyanate (TDI), and chemokine concentrations were determined by enzyme-linked immunosorbent assay. The effects on chemokine concentrations of the highly selective phospho-diesterase 4 inhibitor cilomilast and the glucocorticoid diflorasone diacetate were studied in mouse ears.

Results RANTES and TARC were elevated in both models of allergic contact dermatitis 24 h after challenge, whereas CTACK remained unchanged. The increase in RANTES was diminished in mouse ears pretreated with cilomilast or diflorasone diacetate. TARC was reduced by diflorasone diacetate in the DNCB model but was highly induced in the TDI model; in contrast, TARC was not influenced by cilomilast.

Conclusions TARC and RANTES, but not CTACK, are involved in these two models of allergic contact dermatitis.

Key words: allergic contact dermatitis, cilomilast, CTACK, diflorasone diacetate, RANTES, TARC

Chemokines are central mediators of allergic skin diseases. Allergic contact dermatitis is a T-cell-mediated skin disorder. The recruitment of T cells and other leucocytes at the site of inflammation in skin is therefore a critical step in the elicitation phase of allergic contact dermatitis. The outcome of T-cell-dependent skin reactions also depends on the crosstalk

between infiltrating T cells and populations of resident skin cells such as keratinocytes, fibroblasts, mast cells and endothelial cells. These resident cells as well as dendritic cells contribute to attracting discrete T-cell subsets in the skin. Attracted T cells themselves secrete mediators that stimulate resident cells and lead to an inflammatory microenvironment.

RANTES (regulated on activation, normal T cell expressed and secreted; CCL5) is a CC chemokine and a British Journal of Dermatology2004; 151:823–830. DOI: 10.1111/j.1365-2133.2004.06220.x

monocytes.²RANTES is significantly elevated in serum of patients with atopic dermatitis.³ Studying the expression pattern of chemokines in the skin at different times after hapten application, Goebeleret al.⁴ detected RANTES 12 h after challenge concomitantly with the infiltration of mononuclear cells into the dermis and epidermis.

TARC (thymus and activation regulated chemokine;

CCL17) is a functional ligand for CCR4, a chemokine receptor expressed by most skin-homing [cutaneous lymphocyte-associated antigen (CLA)+] T-helper (Th) cells in the circulation and by skin-infiltrating lympho-cytes.⁵ TARC is expressed in keratinocytes and the secretion is induced by interferon (IFN)-c and tumour necrosis factor (TNF)-a. It has been suggested that TARC is a Th2 cytokine,^6,7 while others⁸ demonstra-ted a chemotactic response of interleukin (IL)-12-cultured T blasts to TARC after these cells returned to a resting state. Addition of IL-4 attenuates⁹ and addition of IL-10 augments¹⁰ the IFN-c- and TNF-a-induced secretion of TARC. TARC is also expressed in dendritic cells and may be involved in recruiting effector Th cells to dendritic cell networks at sites of inflammation.⁸The expression of TARC on endothelial cells of vessels in inflamed skin is controversial:

Campbell et al.⁵ found positively stained cells in psoriatic skin lesions, whereas Vestergaard et al.¹⁰ did not find positively stained cells in vessels in the dermis of patients with atopic dermatitis.

CTACK (cutaneous T-cell attracting chemokine;

CCL27) is described as a skin-associated chemokine that preferentially attracts skin-homing memory T cells.¹¹ Additionally, it was demonstrated that most skin-infiltrating lymphocytes in patients with psoriasis, atopic or allergic contact dermatitis express CCR10, the receptor for CTACK.¹² However, it was recently dem-onstrated that CCR10 is expressed by only 10% of CLA+ Th cells¹³ in two models of inflammatory skin diseases. It was concluded that CCR10 is unlikely to be necessary for cutaneous homing of Th cells tested in these models of skin inflammation.

The following study was performed to evaluate the role of the T-cell-attracting chemokines RANTES, TARC and CTACK in two models of allergic contact dermatitis established by our group,^14–16 and the influence of the dermatocorticoid diflorasone diacetate and the phosphodiesterase 4 inhibitor cilomilast on the chemokine concentration in inflamed skin, as these two substances display strong inhibitory action, at least in allergic contact dermatitis induced by toluene-2,4-diisocyanate (TDI).¹⁶

Materials and methods

Sensitization procedure

Female BALB⁄c mice (Charles River, Sulzfeld, Germany) were sensitized to TDI as described previ-ously.¹⁶ In short, the abdominal skin of the mice was shaved, depilated and tape stripped. For active sensiti-zation, 100 lL of 5% TDI in acetone was administered to the stripped epidermis on three consecutive days.

The allergic reaction was boosted 21 days later by administration of 20 lL of 0Æ5% TDI in acetone to the left ears. Before and 24 h after challenge, the ear thickness was measured with a cutimeter (model 7309;

Mitutoyo, Neuss, Germany). The mice were divided into two equal treatment groups (n¼6) according to swelling intensity, so that each group contained animals which had responded to varying extents. They were rested for 7 days until the ear thickness had reached an almost normal level. To exclude residues of the allergen on the ears, the untreated right ears were used for the main experiment.

The sensitization procedure for dinitrochlorobenzene (DNCB)¹⁴ was as follows: 100lL of 0Æ5% DNCB [in acetone⁄dimethylsulphoxide (DMSO) 1 : 1] was administered to the stripped epidermis on three con-secutive days. The animals were treated with 50 lL of Freund’s complete adjuvant intradermally to induce a Th1 response. On day 14 the reaction was boosted by administration of 20lL of 0Æ5% DNCB in acet-one⁄DMSO to the left ears. After resting for 7 days the right ears were used for the main experiment.

The animal experiment had been registered by the Bezirksregierung Hannover, Germany (Ref. no.

509i-42502-98A839). All procedures were carried out in agreement with the German Law on the Protection of Animals.

Topical treatment

One group of mice (n¼6) was neither sensitized nor challenged. A sensitized second group (n¼6) was treated with acetone⁄DMSO (9 : 1; 20 lL). Two hours later the mice were challenged topically by adminis-tration of 20 lL of 0Æ5% TDI in acetone or 0Æ5% of DNCB in acetone⁄DMSO to the right ears. Two hours before TDI challenge, a third and fourth group (each n¼6) were treated additionally with 20lL of cilom-ilast (3% ¼600lg in acetone⁄DMSO 9 : 1; Elbion AG, Radebeul, Germany) or 20lL of diflorasone diacetate (0Æ05%¼10lg in acetone; Basotherm, 8 2 4 W . B A¨ U M E R et al.

Biberach an der Riss, Germany) on the right ears. The same treatment was performed 2 h before the DNCB challenge.

Mouse ear swelling test and chemokine determination

The ear thickness was determined 6 and 24 h after challenge. Next, the mice were killed by cervical dislocation 24 h after TDI and DNCB challenge and the ears were snap-frozen in liquid nitrogen for immunohistology (embedded in Tissue-Tek) and chem-okine determination. The ears were homogenized under liquid nitrogen and protein was extracted in Tris–HCl and protease inhibitor for 24 h at 4C.¹⁷This protein extraction method resulted in a generally higher yield of protein in comparison with the former extraction method.^15,16 The protein content was determined with a protein assay (Biorad, Munich, Germany). Additionally, the epidermis of seven vehicle-treated and seven TDI-vehicle-treated mouse ears was separ-ated from the dermis with 0Æ5 mol L⁾¹ ammonium thiocyanate (10 min, 37C; Riedel de Haen, Hann-over, Germany) for the determination of CTACK in the epidermis. CTACK, RANTES and TARC were deter-mined by enzyme-linked immunosorbent assay (ELISA) (R&D Systems, Wiesbaden, Germany) according to the manufacturer’s protocols.

Immunohistochemical staining

Cryostat sections of mouse ears (7lm) were cut at )20C (n¼6 each group). After being fixed in acetone at)20C, the sections were stained using the alkaline phosphatase–anti-alkaline phosphatase (APAAP) technique. The skin was stained for CD4+

T cells (L3T4), CD8+ T cells (Lyt-2) and for neutrophils (GR-1) (all three from Dianova, Hamburg, Germany) as described previously.¹⁸ After a 30-min incubation at room temperature with the primary antibodies and washing with Tris-buffered saline (containing 0Æ5%

Tween; Serva, Heidelberg, Germany), the slices were incubated with bridging antibody (rat antimouse IgG, Z494; Dako, Hamburg, Germany) for another 30 min and then again, after washing, with the APAAP complex (D488; Dako) for 25 min. In order to increase the staining intensity, the incubations with the brid-ging antibody and the APAAP complex were repeated for 15 min each. Fast blue (Sigma, Deisenhofen, Germany) served as substrate for the alkaline phos-phatase. Sections were counterstained with

haematox-CTACK and TARC were detected in skin with a primary rat antimouse IgG2a (CTACK, 1 : 200; clone 68706, R&D Systems) or primary goat antimouse IgG (TARC, 1 : 50; AF529, R&D Systems) in acetone-fixed cryostat sections. After blocking with 5% rabbit serum (Dako) and 10% goat serum (Dako) (for TARC) for 60 min the slices were incubated with the primary antibody overnight at 4 C. Next, sections were rinsed in Tris-buffered saline and placed in biotin-labelled secondary antiserum (rabbit antirat IgG for CTACK and donkey antigoat for TARC, both 1 : 500; Dako) for 60 min. They were then washed again and incubated in horseradish peroxidase-labelled streptavidin (1 : 375; Dako) for 60 min, followed by coupling the antibodies to a fluorochrome (carbocyanin 3 for TARC and carbocyanin 2 for CTACK, both 1 : 4000; Jackson Immunoresearch Laboratories, West Grove, PA, U.S.A.). Analyses were performed by using the Kontron KS 400 image analysis system. For quantification of intensity the nickel-intensified diaminobenzidine reac-tion (0Æ05% 3,3¢-diaminobenzidine and 0Æ6% ammo-nium nickel sulphate; both from Sigma, Taufkirchen, Germany) was performed in the presence of 0Æ01%

H₂O₂ for 15 min. Sections were washed, air-dried, treated with toluene for 1 min, and coverslipped with Entellan (Merck, Darmstadt, Germany). As negative control, the primary antibodies were replaced by the same amount of serum to exclude any artefactual labelling by the secondary antibody.

Differences in positively stained cells or staining intensity were evaluated by a blinded examination (range: +, weak to +++++, strong staining intensity).

Considering the ELISA results, it was decided to forgo the histological examination of RANTES in mouse skin.

Statistical evaluation

Results are presented as mean ± SD and as mean

± SEM for the ear swelling. A one-wayANOVAon ranks was performed followed by a multiple comparison method (Dunn’s test). The Mann–Whitney U-test was used for comparison of two different groups.

Results

Challenge with both TDI and DNCB resulted in a distinct increase in ear swelling 6 and 24 h after treatment. It was demonstrated immunohistologically that in both models of contact sensitivity there was an increase in neutrophil granulocytes, CD4+ and CD8+

T A R C , R A N T E S , C T A C K I N A L L E R G I C C O N T A C T D E R M A T I T I S 8 2 5

DNCB-treated mice. Pretreatment with diflorasone diacetate markedly inhibited the influx, whereas cil-omilast induced only a slight inhibition of the inflam-matory cell influx (Table 1 and Fig. 1). When mouse ears were treated with cilomilast or diflorasone diace-tate 2 h before challenge, the ear swelling was also diminished. However, cilomilast failed to produce a significant inhibition of swelling in DNCB-induced dermatitis (Fig. 2).

IL-12, but not IL-4, was induced in DNCB-treated mouse ears, and IL-4, but not IL-12, was induced in TDI-treated ears (Table 2).

There was no significant difference in CTACK con-centration in mouse ear skin between the different treatment groups 24 h after challenge with either TDI

or DNCB (Table 3). Earlier studies with a limited number of mice had shown that there was no distinct increase in CTACK concentration in TDI-treated mouse ears 1, 4, 8 or 16 h after challenge (data not shown).

There was also no difference in CTACK staining intensity between untreated control and DNCB- or TDI-treated skin in immunohistological sections (Fig. 3). Histological examination of TDI- and DNCB-treated mouse ears reveal a distinct oedema and cell influx mainly in the dermis (Fig. 1). As there was no difference in CTACK concentration when the protein content of the whole ear was considered and as CTACK is found only in the epidermis,¹²only the epidermis was examined for CTACK. There was no significant increase of CTACK in the epidermis of TDI-treated mouse ears 24 h after challenge (vehicle, 592 ± 195 pg mg⁾¹vs.

TDI, 802 ± 480 pg mg⁾¹). The absolute values are lower in comparison with the whole ear (Table 3). This is due to a shorter protein extraction time (2 h for epidermis vs. 24 h for whole ear).

The concentration of TARC increased significantly in TDI- and DNCB-treated mouse ears 24 h after chal-lenge. The elevation was more distinct in DNCB-treated ears than in TDI-treated ears (Fig. 4). Cilomilast had no significant influence on either model. Diflorasone diacetate inhibited the DNCB-induced elevation of TARC in the DNCB model. In contrast, diflorasone diacetate administration in the TDI model resulted in a significant increase in TARC concentration. As this

Table 1. Cell influx in mouse ears and effects of cilomilast and di-florasone diacetate 24 h after challenge with toluene-2,4-diisocya-nate (TDI) or dinitrochlorobenzene (DNCB)

Cell type Control Elicitor

GR-1 stains for neutrophils. Reactions are scored: +, weak to +++++, strong.

Figure 1. Staining of CD4+, CD8+ and GR-1+ cells in mouse skin after challenge with toluene-2,4-diisocyanate (TDI) and dinitro-chlorobenzene (DNCB). In comparison with the control (a) there is an influx of CD4+ T cells in both models of allergic contact dermatitis (b, TDI; c, DNCB). The increase in CD8+ cells is more distinct in the DNCB model (f) compared with TDI (e) (d, control). The influx of GR-1+ cells is strong in both models (g, control; h, TDI; i, DNCB). For quantifica-tion see Table 1. Bar¼50lm.

8 2 6 W . B A¨ U M E R et al.

result was not anticipated the whole experiment was repeated for the TDI model with similar results (data

The increase in TARC concentration in TDI- and DNCB-treated mouse ears correlated with a stronger staining intensity in skin sections staining positive for TARC (Fig. 3).

RANTES was significantly elevated 24 h after TDI and DNCB challenge. Pretreatment with diflorasone diacetate resulted in a significant inhibition in both models. Cilomilast significantly inhibited the increase in RANTES in the TDI but not in the DNCB model (Fig. 4).

Discussion

For the invasion of lymphocytes in allergic skin diseases such as contact sensitivity, chemokines and chemokine receptors play a pivotal role in both attracting the lymphocytes and differentiating between them.^19,20 Therefore the T-cell-attracting chemokines RANTES, TARC and CTACK were studied in the models of TDI-and DNCB-induced allergic contact dermatitis which differ immunologically in their cytokine pattern as far as Th1 and Th2 dominance is concerned.¹⁹ It was previously demonstrated that repetitive dermal TDI administration leads to increased IgE serum levels,^21,22 whereas DNCB administration failed to provoke an IgE response in Balb⁄c mice²³or the IgE response was only moderate compared with TDI.^22,24Most comparisons of TDI- and DNCB-induced allergic contact dermatitis refer to the cytokine secretion in the draining lymph nodes¹⁹ or the mRNA expression of cytokines in the lymph nodes.²⁵However, these results correspond well to our findings in inflamed ear skin (Table 2), confirm-ing the Th1⁄Th2 differentiation between classical contact allergens such as DNCB (Th1) and the chem-ical respiratory allergen TDI (Th2) irrespective of the route of exposure, as both haptens were administered epicutaneously. A recent report has revealed that DNCB and trimellitic anhydride (a chemical respiratory allergen like TDI) differ with respect to the time course of induced Langerhans cell migration and dendritic cell accumulation in draining lymph nodes. The mobiliza-tion of Langerhans cells is more rapid following topical exposure of mice to DNCB.²⁶The authors suggest that the kinetics of Langerhans cell migration could corre-late with a preferential Th1 or Th2 response. However, the authors stress that these preliminary results do not prove a general causative relationship between the dynamics of Langerhans cell migration and the quality of immune response elicited by a chemical allergen.²⁶

Our results also demonstrate that TDI- and DNCB-induced immune responses differ in their inflammatory

* *

ear swelling (%)ear swelling (%)

Figure 2.Mean ± SEM ear swelling 6 h (white bars) and 24 h (grey bars) after challenge with (a) toluene-2,4-diisocyanate (TDI) and (b) dinitrochlorobenzene (DNCB). There is a significant increase in ear swelling in TDI- and DNCB-treated mice compared with untreated controls. Both cilomilast and diflorasone diacetate inhibited the swelling significantly in the TDI model. In the DNCB model only diflorasone diacetate caused a significant inhibition. *P <0Æ05 in comparison with treated animals (n¼6 for each group; n¼5 for diflorasone diacetate + DNCB).

Table 2.Concentration (mean ± SD) of interleukin (IL)-4 and IL-12 (pg mg⁾¹protein) in mouse ears 24 h after challenge with toluene-2,4-diisocyanate (TDI) or dinitrochlorobenzene (DNCB)

Cytokine Control TDI Control DNCB

IL-4 18Æ5 ± 6Æ7 164Æ9 ± 77Æ2* 56Æ3 ± 21Æ9 83Æ7 ± 34Æ3 IL-12 p70 5Æ7 ± 3Æ1 7Æ1 ± 2Æ2 7Æ4 ± 1Æ4 19Æ5 ± 2Æ5*

*P <0Æ01 compared with control, Mann–WhitneyU-test.

Table 3.Concentration (mean ± SD) of CTACK (pg mg⁾¹protein) in mouse ears 24 h after challenge with toluene-2,4-diisocyanate (TDI) or dinitrochlorobenzene (DNCB)

striking that DNCB induces a significant increase in CD8+ cells, whereas TDI challenge results in only a slight increase in the influx of CD8+ cells. It has been postulated that in many instances the most important effectors are CD8+ lymphocytes, while CD4+ cells perform regulatory functions in allergic contact der-matitis.²⁰This theory can also be assumed to be valid for the DNCB-induced contact dermatitis. TDI seems to differ fromclassicalcontact allergens such as DNCB in that it induces a delayed-type hypersensitivity clearly mediated by CD4+ T lymphocytes (Tables 1 and 2).

This corresponds to findings with the contact allergen fluorescein isothiocyanate, and so it can be concluded that the distribution of CD4+ and CD8+ cells differs in various models of allergic contact dermatitis.²⁰

As a functional parameter, the ear swelling reaction was measured in analogy to former studies.^14–16 As demonstrated previously, the magnitude of the re-sponse corresponded to the inflammatory reaction observed in histological sections. Although the con-centration of DNCB used (0.5%) is described as not being irritant,²⁷ the distinct swelling in the DNCB model as early as 6 h after challenge can be interpreted as an additional irritatant⁄cytotoxic effect of DNCB. It was observed in vitro that as little as 30lmol L⁾¹ DNCB was cytotoxic for a murine keratinocyte cell line (MSC-P5), whereas up to 1200lmol L⁾¹TDI did not induce any decrease in cell viability (measured by Owen’s reagent) or in cell proliferation (measured by bromodeoxyuridine incorporation) determined 24 h after treatment (own unpublished observations).

It was demonstrated that RANTES and TARC are induced in mouse skin after challenge with either TDI or DNCB (Fig. 4). Although a strong CTACK expression in the epidermis was observed in a model of dinitro-fluorobenzene-induced allergic dermatitis 24 h after

challenge,¹²it was not possible to confirm a pivotal role of CTACK in either model of allergic contact dermatitis tested in this study. However, even if there is no induction of CTACK, it is striking that the basal levels of CTACK are very high (Table 3). All cytokines and chemokines measured before were detected at concen-trations measured in pg mg⁾¹protein.^15,16In contrast, the concentration of CTACK ranged between 3 and 4 ng mg⁾¹ protein. Therefore, CTACK might play a central role in steady-state trafficking of T cells into the skin. Our results support findings of Soler et al.¹³who investigated the expression of the chemokine receptors CCR4 (ligand: TARC) and CCR10 (ligand: CTACK). In contrast to CCR4, only a small subpopulation of the infiltrated lymphocytes expressed CCR10 in models of delayed-type hypersensitivity and bacterial chancroid skin lesions. It was concluded that CLA and CCR4 are intimately associated with the process by which most memory Th cells specifically enter the skin. These findings are confirmed by a distinct increase in the ligand for CCR4 (TARC) in the skin in both models of allergic contact dermatitis and an unchanged concen-tration of the ligand for CCR10 (CTACK). TARC is

Im Dokument Phosphodiesterase-4-Inhibitoren zur Behandlung allergischer Hauterkrankungen bei Mensch und Tier (Seite 88-108)