Delineating the Role of Histamine-1- and -4-Receptors in a Mouse Model of Th2-Dependent Antigen-Specific Skin Inflammation

Background Histamine drives pruritus in allergic skin diseases which clinically constitutes a most disruptive symptom. Skin pathology in allergic skin diseases is crucially influenced by different T-helper subsets. However, the contribution of different histamine-receptors to T-helper cell dependent skin pathology has not been definitively answered. Models which can specifically address the functional role of T-helper subsets and the mediators involved are therefore valuable to gain further insights into molecular pathways which contribute to allergic skin disease. They might also be helpful to probe amendable therapeutic interventions such as histamine-receptor antagonism. Objective Establishing an adoptive transfer model for antigen-specific Th cells, we aimed to delineate the role of histamine H1- and H4-receptors in Th2-dependent skin inflammation. Methods In-vitro differentiated and OVA primed Th2 cells were adoptively transferred into congenic recipient mice. In vivo treatment with specific histamine H1- and H4-receptor antagonists was performed to analyze the contribution of these histamine-receptors to Th2-dependent skin pathology in our model. Analysis four days after epicutaneous challenge comprised skin histology, flow cytometric detection of transferred T-helper cells and analysis of antigen-cytokine profiles in skin-draining lymph nodes. Results Use of specific H1- and H4-receptor antagonists revealed a crucial role for H1- and H4-receptors for Th2 migration and cytokine secretion in a Th2-driven model of skin inflammation. While H1- and H4-receptor antagonists both reduced Th2 recruitment to the site of challenge, local cytokine responses in skin-draining lymph nodes were only reduced by the combined application of H1- and H4-receptor antagonists and mast cell counts remained altogether unchanged by either H1R-, H4R- or combined antagonism. Conclusion Our model demonstrates a role for H1- and H4-receptors in Th2 cell infiltration and cytokine secretion in allergic skin diseases and suggests further studies to evaluate these findings for therapeutic approaches.


Introduction
Animal and human studies have demonstrated elevated histamine levels in atopic dermatitis (AD). Histamine is a central mediator in the complex signalling network that leads to the development and maintenance of pruritus [1]. Yet, pruritus in patients suffering with AD, contrary to the effects of antihistamines observed in patients with pruritus in allergic rhino-conjunctivitis, is often not relieved by antihistamines [2] which led to the assumption that histamine is binding to other histamine receptors, possibly expressed on the immune cells involved in AD. The H 4 R is expressed on different immune cells [3] and has thus been a focus of recent attention, as efficient targeting of this receptor is believed to be a promising approach for pruritus but also the inflammatory changes observed in AD. In this line, studies could show that patients with AD express increased levels of H 4 R on T-cells of the peripheral blood [4]. Moreover, Dunford et al. demonstrate that the H 4 R is involved in pruritic responses in mice to a greater extent than the H 1 R [5] and Ohsawa et al. could demonstrate a potent anti-inflammatory effect of combined administration of H 1 R and H 4 R antagonists in a mouse model of atopic dermatitis [6]. However, there have also been contradictory studies. For example, H 1 R or H 4 R antagonists had no impact on the development of acute skin lesions in an experimental canine atopic dermatitis model [7].
Skin contains around 20 billion T-cells in humans [8] which conduct immunosurveillance and are associated with the development of inflammatory disorders such as atopic dermatitis [9]. Amongst those T cells are antigen-specific T-helper (Th) subsets with different roles. The T-cell response in AD is biphasic with an initial phase predominated by Th2 cells and a chronic Th1dominated phase [10]. A number of animal models have been published which allow studies on the role of specific mediators in the skin's immune homeostasis and pathogenesis of AD [11]. The beneficial effects of a combined H 1 R and H 4 R application on pruritus have been demonstrated in such models [6,12]. However, the role of antigen-specific T-cell subsets cannot be specifically addressed in these models, as tracking of antigen-specific T-cells is not possible in polyclonal models. Studies which clarify the role of the H 4 R for antigen-specific Th2-mediated pathology in AD could emphasize their utility in the treatment of AD. In the study presented below, we describe the development of a murine model of Th2-dependent antigen-dependent skin inflammation which we utilized to demonstrate differential effects of the H 1 Rs and H 4 Rs on Th2 cell migration and cytokine secretion.

Animals
Six to eight week-old female BALB/c mice were purchased from Charles River Laboratory (Charles River) and housed in the animal facility of the Hannover medical school. DO11.10 (BALB/ c-Tg(DO11.10)10Loh/J) mice on a BALB/c background with OVA-specific transgenic (Tg) TCR were bred in our facility. All experimental methods described in this manuscript were in accordance with the German Animal Welfare Legislation and performed as approved by the Lower Saxony State Office for Consumer Protection and Food Safety (LAVES; application no. 33.9-42502-04-09/1664). Animal treatments (patching, intranasal application) were performed under isoflurane anesthesia, and all efforts were made to minimize suffering.

BMDC preparation
For generation of bone marrow derived dendritic cells, bone marrow cells were isolated by perfusion of femur and tibia. Cells were cultured in RPMI 1640 (Invitrogen Corp.) in the presence of 1% culture supernatant from GM-CSF producing J558L cells [15]. After 6 days, cells were harvested and a fraction of DCs were subjected to FACS analysis for purity control with MHCII/ CD11c+ cells constituting between 60-70% of cells analyzed.
Animal treatment protocols for adoptive transfer and epicutaneous challenge 5610 6 in vitro differentiated, OVA primed Th2 cells of DO11.10 background were injected intraperitoneally (i.p.) into naïve BALB/ c mice. One day later, the back of the recipient mice was shaved with electric clippers and two days later, they were exposed epicutaneously (e.c.) to an occlusive patch (Hansaplast), fixed with an adhesive gel (Mastisol liquid adhesive, Eloquest) containing 100 mg OVA (Grade V; Sigma Aldrich) in 40 ml of PBS or PBS alone on shaved back skin as previously described [16]. Patches were left intact for 4 days (Fig. 1A). Intraperiteonal transfer was chosen as preliminary experiments had revealed efficient migration and local induction of cytokine secretion of intraperitoneally transferred Th cells compared to the migration efficiency of i.v. transferred cells ( Fig. 1B and C).

Animal treatment protocol for histamine antagonist administration
For the histamine antagonist experiments, H 1 R antagonist (Mepyramine, Sigma-Aldrich) and/or H 4 R antagonist (JNJ7777120) (gift from Dr. Armin Buschauer, University of Regensburg) were dissolved in 20% (v/v) DMSO and further diluted in PBS. 100 ml of antagonist solution was administered subcutaneously to the site of antigen challenge 30 minutes prior to OVA-challenge as well as 4 hrs, 24 hrs and 48 hrs after the challenge in the same dose (20 mg/kg, 10 mg/kg or 2 mg/kg body weight) [5]. Doses were chosen from previously published studies [7,17]. Control mice received vehicle treatment. Mice were killed and analyzed on day 7, as preliminary time-kinetic experiments had revealed optimal divergence of read-out parameters in sham-treated vs. antigen-treated animals at this time point (Fig. 1D). For histamine and mouse mastcellprotease-1 measurements animals were analyzed 6 h, 12 h and 24 h after administration of OVA patch.

Skin preparation
Skin sections (262 cm) taken from areas of antigen-/vehicleexposed skin were pooled from each group, manually minced into fragments by small surgical scissors, and incubated at 37uC for 2 hrs in digestion mix consisting of 450 U/mL collagenase IV (Worthington Biochemical Corporation), 20 mg/mL DNase I Type IV from bovine pancreas (Sigma-Aldrich), 25 mg/ml hyaluronidase (Sigma-Aldrich), 1% HEPES and 1% sodium pyruvate in RPMI medium. After digestion, the tissue was grated through a metal strainer and consecutively through a 100 mm nylon cell strainer (BD) to gain single-cell suspensions. Cells from each group and were cultured with IL-2 (25 U/ml) for 48 hrs prior to harvest of supernatants.

Lymph node preparation
Axillary lymph nodes (aLNs) were pooled from each group at the time of sacrifice. Single-cell suspensions were obtained by grating through a 100 mm nylon cell strainer (BD). 2610 5 cells were cultured with 200 mg/ml OVA presented by GM-CSF differentiated bone marrow-derived dendritic cells (2610 4 ) from wild-type mice for 48 hrs.

Cytokine ELISA
Cytokines levels of IFN-gamma, IL-4, IL-17 and IL-13 in culture supernatants were determined by means of ELISA DuoSet kits (R&D Systems) according to the manufacturer's instructions.

Flow cytometry analysis
All staining procedures were performed on ice. Cells were incubated with anti-FcR (clone 24G2, lab grown) antibody for 20 minutes on ice, then stained with antibodies. The presence of OVA-transgenic T-cells was determined by staining with fluorescently labeled antibodies: CD4 (clone RM4-5; BD) in combination with the anti-DO11.10 TCR antibody clone KJ1-26 (lab grown). The cells were analyzed on a LSRII ow cytometer (BD) or FACS Canto II (BD) in association with FlowJo software (Treestar).

Skin Histology
Paraffin-embedded skin samples were sectioned (5 mm thick) and stained with hematoxylin and eosin (H&E) or giemsa. Pictures were examined at 200X and 600X magnifications using an Olympus BX51 microscope (Olympus) or at 100x, 200x and 1000x magnifications using a Keyence BZ-9000 microscope (Keyence). Mast cells were counted on giemsa stained sections in a blinded fashion by analyzing 10 high power fields (1000x magnification) per section.

Mass spectrometry
Skin sections (262 cm) taken from areas of antigen-exposed skin were processed individually, weighed and snap-frozen. Prior to histamine analysis in serum and skin tissue, sample preparation was performed as follows: 50 mL of serum were treated with 200 mL of 50/50 acetonitrile/methanol [v/v]. After centrifugation (20,8006g; 10 min; 4uC) supernatant fluid was dried under a constant nitrogen stream. The residual pellet was resolved in 50 mL of 80/20 acetonitrile/water [v/v] containing 0.25 mM d4histamine as internal standard and analyzed by liquid chromatography coupled with tandem mass spectrometry. Tissue sample preparation was performed using a FastPrepH-24 instrument (MP Biomedicals LLC, Illkirch, France). Skin samples were transferred into a FastPrepH tube (Lysing Matrix A). 800 mL of ice cold extraction solvent (70/30 ethanol/water [v/v]) were added and tissue was homogenized (5.0 m/sec; 2x for 30 sec). Samples were centrifuged (20,8006g; 10 min; 4uC) and 600 mL supernatant fluid were treated as described above. Skin samples were diluted (1:50) in 80/20 acetonitrile/water [v/v] containing 0.25 mM d4histamine for analysis.

Statistical analysis
Unless indicated otherwise, 4-5 mice were used for each condition studied in an individual experiment. The number of repeat experiments is disclosed in the individual figure legends. For statistical analysis t-tests or one way ANOVA (analysis of variance) with Bonferroni's multiple comparison test was performed with the GraphPad PrismH software to determine statistical differences between means or proportions between two groups of data. A p,0.05 was considered to be significant.

Adoptively transferred polarized T-cells migrate to the site of antigen-challenge in the skin
To investigate the role of H 1 and H 4 -receptors in Th2dependent skin inflammation, we established a model of Thdependent antigen-specific pathology. To this end, we adapted an adoptive transfer model used to examine allergic lung inflammation [14]. Splenic naïve T-cells derived from DO11.10 mice were subjected to culture conditions for Th2 differentiation. Th polarization status was confirmed by in vitro stimulation with PMA and ionomycin (data not shown). Polarized Th2 cells were adoptively transferred into the corresponding recipient mouse strains which was e.c. challenged and analyzed (Fig. 1A).
Upon flow cytometric analysis of single-cell suspensions from the skin, we could detect increased percentages of the CD4 + KJ1-26 + cells in BALB/c mice which received Th2 cells and were OVA-treated compared to their respective sham-treated controls ( Fig. 2A). Sham-treatment of Th2-recipient mice compared to animals which had not received Th2 cells but were antigen-treated also revealed Th2 recruitment ( Fig. 2A) suggesting an enhanced skin-migratory propensity of Th2 cells regardless of antigencontact. Migration of Th2 cells to the site of antigen-contact was neither polarization-nor strain-specific though, as we could also detect increased migration of Th1 and Th17 cells upon OVA treatment (Fig. 2B) and OTII derived Th cells also migrated efficiently to the skin in C57BL/6 mice (data not shown).

Adoptive transfer of Th2 cells elicits a Th-specific cytokine secretion profile and histological changes
Having shown that the transferred cells migrate to the site of antigen-challenge, we analyzed the cytokine profile at the site of challenge and in the draining lymph nodes (dLNs) to confirm the local development of a Th2-polarized cytokine milieu. To this end, we stimulated single-cell suspensions obtained from the skin of mice which received Th2 cells and were either OVA-or PBSpatched with IL-2 for 48 hrs. Skin cells from mice which had received Th2 cells and were OVA-treated secreted mainly IL-13 and also small but significant amounts of IL-17. IFN-gamma levels were low regardless of patch-treatment and did not differ significantly between groups. IL-4 levels were consistently below the detection limit in these cultures (Fig. 3A).
In the sham-treated group IFN-gamma levels were low and IL-17 levels were below the detection limit. We observed IL-13 secretion in this group at intermediate levels, suggesting that shaving and patching alone induced a baseline secretion of IL-13 in the sham control group leading to a ''skin background'' secretion level of IL-13 [16,18], as also suggested by background levels of IL-13 in animals which had received Th1 or Th17 cells and were sham-treated ( Fig. 3B and C).
Upon restimulation of cells from dLNs, we also observed a Th2polarized cytokine pattern. Cytokine analysis after antigen-specific in vitro restimulation revealed that, in line with their initial polarization status before transfer, cells from mice which had received Th2 cells secreted significant levels of IL-4 and IL-13 while IFN-gamma and IL-17 levels were negligible and did not differ significantly from sham-treated controls (Fig. 3D). Shamtreated mice showed low to undetectable cytokine levels for all cytokines assayed (Fig. 3D).
Experiments with Th1 and Th17 polarized cells confirmed cytokine secretion patterns specific to the transferred cells in dLN ( Fig. 3E and F) and skin ( Fig. 3B and C). However, again we observed background secretion of IL-13 in the skin for both Th1and Th17-recipients and background secretion of IFN-gamma and IL-13 for Th17-recipients in the dLNs. In summary, these results suggest this model to be well-suited to study the effects of pharmacological targets, such as the histamine receptors on Th cell-specific driven skin inflammation.
Additionally, we performed H&E stainings of histological sections to examine phenotypical changes of the antigenchallenged skin of Th2-recipient mice. The sections showed an increased inflammatory cell infiltration in the epidermis compared to the sham-treated controls (Fig. 4A). Moreover, we detected increased numbers of degranulated mast cells both on the epidermal and dermal side of skin sections of Th2-recipient OVA-treated mice in comparison to the sham-treated groups ( Fig. 4A and B). Finally, ELISA measurement of mouse mast cell protease-1 levels as a product of mast cell degranulation in sera of the same mice, revealed an increase in this mast cell product, peaking at 12 h post patching suggesting that mast cells not only increased in numbers but also degranulated (Fig. 4 C). We thus conclude that in our model migration of transferred Th2 cells is associated with an increased inflammatory cell influx into the skin, particularly mast cells which show evidence for degranulation.

H 1 -and H 4 -receptor antagonists decrease Th2 migration and cytokine secretion but not mast cell numbers in the skin
Having confirmed that our model is suitable to study Th2dependent skin pathology, we strove to dissect the role of H 1 R and H 4 R on Th2 cell migration in our model. To this end, Th2 recipient mice were treated with the H 1 R antagonist mepyramine and/or the H 4 R antagonists JNJ7777120 prior and after e.c. challenge with OVA (Fig. 1D). Flow cytometric analysis of singlecell suspensions from the challenged skin revealed decreased percentages of CD4 + KJ1-26 + transferred cells after either H 1 R or H 4 R antagonist treatment in comparison to the sham-treated group. The percentage of transferred cells in skin further declined upon administration of both the antagonists (Fig. 5A).
We also assessed the antigen-specific cytokine secretion pattern in the dLNs. IL-4 and IL-13 secretion were unaffected in cells from the dLNs of mice which had been treated with either one of the antagonists. However, we observed significantly decreased IL-4 and IL-13 secretion from cells from the H 1 R and H 4 R combined treatment group (Fig. 5B). We did not observe statistical changes in IFN-gamma secretion upon combined administration of both antagonists compared to vehicle treatment (Fig. 5B).
Furthermore, we performed giemsa stainings, a method widely accepted for identification of mast cells [19], of the antigenchallenged skin to examine the effect of histamine receptor antagonism on mast cell numbers. Mast cell counts in the skin of mice subjected to Th2 transfer and OVA patch were increased compared to a naïve controls, but we did not observe changes in mast cell counts due to histamine receptor antagonist treatment (Fig. 5C).
In order to strengthen our conclusion that the observed effects on Th2 migration are dependent on histamine receptor antagonism, we performed two additional sets of experiments: Firstly, to assure that the observed effects of histamine receptor antagonism were due to pertinent histamine levels acting on its receptors, we quantified histamine in serum and skin samples of mice, which received Th2 cells, 6 h, 12 h and 24 h after application of OVA patch by means of mass spectrometry. Indeed we were able to detect significant levels of histamine in our skin samples suggesting that the histamine receptor antagonists could act upon significant levels of this mediator in our model. Albeit in comparison to histamine levels in naïve mice, we found no significant difference in serum or skin from treated mice at the indicated times ( Fig. 6 A,  B). Secondly, we repeated our experiments with different doses of H4R or H1R antagonists and compared the influx of Th2 cell via FACS analysis with skin from vehicle treated and naïve animals. The 20 mg/kg dose of H4R antagonist, which was used in the previous experiments led to a strong decrease in the migration of transferred T cells to the skin (0.69%) compared to vehicle treatment (2.28%), with the percentage reaching background level as measured in naïve skin (0.65%; Fig. 6C). Reduction of the antagonist dose led to a step-wise recovery of T cell migration to the skin (1.19% and 1.41%; Fig. 6C). In contrast we did not observe any dose-dependent effects by H1R antagonism (data not shown).
Thus, we can conclude that our model allowed us to demonstrate a reduction of Th2 migration to the skin and Th2dependent cytokine secretion by histamine receptor antagonism. Antagonism of either the H 1 R or the H 4 R alone affected Th2 migration with an additive effect observed upon application of both antagonists. Cytokine secretion and mast cells counts, on the other hand side, were unaffected by H 1 R or H 4 R antagonism alone. Only upon combined receptor antagonism, could we demonstrate effects on Th2 cytokine secretion while mast cell counts remained unaffected by this measure.

Discussion
There are many models to study different aspects of allergic skin inflammation as a proxy for AD, yet so far neither of these models allowed us to specifically address the role of Th polarized antigenspecific T-cell subsets [11]. Our aim was to establish such a murine model to dissect the contribution of H 1 R-and H 4 Rs to Th2 cellmediated skin inflammation. In that line, we demonstrate that the combined action of H 1 R and H 4 R antagonists has additive effects on Th2 migration to the skin. Furthermore, we show that the combined application of H 1 R and H 4 R antagonists modulates local Th2 cytokine secretion.
The H 1 R is expressed on diverse cell types and the H 4 R predominantly on keratinocytes and haematopoetic cells [3]. A role for the H 1 R in Th2 cell migration into the lung has been demonstrated in a mouse model of pulmonary inflammation [20]. Different studies have investigated the combined role of H 1 R and H 4 R antagonist in skin inflammation. Combined administration of the antagonists attenuated pruritic and inflammatory responses in different models of skin inflammation and allergic contact dermatitis [6,12]. However, neither of these studies specifically addressed the contribution of these receptors to Th2 migration to the skin, a question made feasible to address by establishment of an adoptive transfer model only. Our results show for the first time that blockade of either, the H 1 R or the H 4 R, leads to a reduction of H1R-and H4R-dependent migration of antigen-specific Th2cells to the skin, accompanied by a decrease in the Th2-specific cytokines IL-4 and IL-13 upon combined action of both receptor antagonists.
Several studies support our observations and provide concepts how H 1 R and H 4 R antagonism might act on a cellular level for example by blocking the signals on DCs necessary for Th2 polarization (H1R antagonism) [21] or inhibiting DC chemotaxis (H4R antagonism) [22]. The use of H 4 R antagonists can significantly decrease scratching in mice [23]. Thus in our model, administration of an H 4 R antagonist could lead to a decrease in scratching and in turn, decreased abrasion of the skin, reducing secretion of Th2 cytokines induced by this mechanism [24], thus breaking a vicious cycle. Additionally, decreases in Th2 migration by H 4 R antagonism, as demonstrated by our results, would further dampen Th2 cytokine secretion. Moreover, studies have shown an increased Th2 polarization and increased Langerhans cell activation by H 4 R stimulation [4,25]. Hence, combined blockade of the H 1 R and H 4 R affects Th2 polarized skin inflammation at several check-points, leading to a more effective reduction in Th2 cytokine secretion than either antagonist alone which is supported by our results.
The lack of effectiveness of classical H 1 R and H 2 R antihistamines in alleviating chronic allergic skin conditions has led the focus of research for antipruritic drugs to shift from histamine to other putative mediators. Although more work is necessary, our results suggest a beneficial effect of a combined H 1 R and H 4 R blockade for the treatment of such conditions. Thus, the development of a new adoptive transfer system for allergic skin inflammation has allowed us to discover previously unappreciated effects of long-standing pharmacological inhibitors, underscoring the usefulness of our model, which will allow us to address other important questions regarding the role of Th cells in allergic skin disease in the near future.
Whilst our model demonstrates increased mast cell counts and mast cell protease-1 levels after Th2 transfer and antigen challenge suggesting increased mast cell degranulation, levels of histamine, the hallmark of mast cell degranulation were not further increased by Th2 transfer and antigen patching compared to naïve controls. This could be due to several reasons. Although measurement of increased histamine levels in serum is an accepted parameter for anaphylactic reactions for example in murine models of food allergy [26], it is less often used in models of skin inflammation. Imai et al. describe increased blood histamine levels in transgenic mice spontaneously developing long-lasting dermatitis (by skinspecific IL-33 expression) [27], a phenotype which resembles the chronic phase of atopic dermatitis. Other groups could show elevated histamine levels in skin or serum only after repeated epicutaneous challenge [28,29], suggesting that re-exposure might be necessary to induce measurable increases in histamine levels. It is thus conceivable that repeated patching could also lead to histamine increases in our model, a read-out parameter we did not address, though. Thus, the histamine receptor antagonists we employed seem to exert their functions on the background histamine levels we detected in the skin irrespective of patching.
We also did not observe effects of the histamine receptor antagonists on mast cell counts. This aspect, however is not altogether surprising, as we were blocking effects downstream of mast cell infiltration.
Apart from our findings on the contribution of different histamine-receptors in Th2-dependent allergic skin inflammation, our model provided us with additional interesting insights into Thdependent allergic skin pathology. Our model, which we adapted from widely-used adoptive transfer models for allergic lung inflammation [14], allows tracking of the transferred cells in the patched skin regardless of the strain background, similar to what has enabled researchers in the past decades to dissect the role of different T-cell subsets in analogous lung models. We were also able to demonstrate that the transferred cells induce a distinct cytokine secretion profile in the recipient, largely corresponding to the transferred T-cell subtype. Interestingly, we observed detectable levels of IL-13 regardless of the transfer, polarization status or skin challenge of the mice. These results are in line with other studies which suggest that shaving and patching alone elicit a background level of IL-13 secretion from endogenous skin cells [30], underscoring the critical role of IL-13 in inflammatory skin disease [16]. The additional increase observed in the Th2 recipient OVA treated group most likely results from secretion of the transferred cells which is efficiently abrogated by combined application of the H 1 R and H 4 R antagonists. Upon Th17 cell transfer, we not only observe the expected IL-17 but also significant amounts of IFN-gamma. Prior to transfer, upon in vitro restimulation of the in vitro generated Th17 cells, we can detect secretion of IFN-gamma as well as IL-17 (data not shown). Hence, after transfer IFN-gamma secretion is possibly due to these transferred cells. IFN-gamma appears to be the most ''exclusive'' cytokine in the skin as significant increases were only detectable in the mice which had received Th1 cells and were antigen-treated. Taken together, we speculate that the distinct ''Th specific'' cytokine profiles we observe reflect cytokine secretion by the transferred cells and the ''background secretion'' of IL-13 and IL-17 is provided by endogenous cells; a hypothesis where future studies in our model will be able to make contributions towards an increased understanding of the contribution of different Th subsets in antigen-specific skin inflammation. Hence we conclude that our model warrants further investigation with regards to the cytokine profiles and histological patterns associated with Th1-and Th17transfer.
In summary, in addition to the development of an interesting new tool for future experiments to dissect the role of different Th subsets and their mediators in allergic skin disease, our results show that the combined treatment with H1R antagonist and H4R antagonist is superior to treatment with single antagonists in reducing the migration of antigen-specific Th2 cells to skin, as well as reducing Th2-dependent cytokine secretion. These findings corroborate recent results from other groups describing a similar effect on pruritus and skin pathology in models of chronic dermatitis and allergic contact dermatitis [6,31]. Furthermore, a similar conclusion has been drawn in the study of Beermann et al. [32] who investigated the role of H 1 R and H 4 R antagonists in a model of allergen-induced asthma. Together, we and others provide evidence that the combined treatment approach is a promising new venue for the treatment of different types of allergic diseases.