Elevated Plasma Level of Interferon-λ1 in Chronic Spontaneous Urticaria: Upregulated Expression in CD8+ and Epithelial Cells and Induction of Inflammatory Cell Accumulation

Interferon- (IFN-) λ1 is regarded as a potent bio-active molecule in innate immunity. However, little is known about its role in chronic spontaneous urticaria (CSU). We therefore investigated expression of IFN-λ1 in CSU, its cellular location, and its influence on inflammatory cell accumulation by using flow cytometry analysis, skin tissue dispersion, immunohistochemical stain, and a mouse peritoneal inflammation model. The results showed that level of IFN-λ1 was 2.0-fold higher in plasma of the patients with CSU than the level in healthy control (HC) subjects. Among leukocytes examined, only CD8+ T cells expressed more IFN-λ1 in CSU blood. Double labeling immunohistochemical staining revealed that IFN-λ1+ inflammatory cells such as mast cells, eosinophils, B cells, neutrophils, and macrophages were mainly located in dermis, whereas epidermis tissue highly expressed IFN-λ1. IFN-λ1 induced a dose-dependent increase in number of eosinophils, lymphocytes, mast cells, macrophages, and neutrophils in the peritoneum of mice at 6 h following injection, which was inhibited by pretreatment of the animals with anti-intercellular adhesion molecule- (ICAM-) 1 and/or anti-L-selectin antibodies. In conclusion, IFN-λ1 is likely to play a role in the pathogenesis of CSU. Blocking IFN-λ1 production may help to reduce the accumulation of inflammatory cells in the involved CSU skin.


Introduction
CSU, which has no discernable external cause, comprises the majority of cases of chronic urticaria [1]. It is recognized to have an autoimmune cause including production of IgE autoantibodies against autoantigens and generation of IgG autoantibodies against Fc RI, IgE, or both, which might chronically activate mast cells and basophils [2]. These autoantibodies have been shown to activate blood basophils and cutaneous mast cells in vitro and induction of a cellular infiltration of CD4 + T lymphocytes, monocytes, neutrophils, eosinophils, and basophils [3]. However, the exact physiopathology of CSU remains unknown [4].
IFN-1 belongs to the Type-III IFN family [10] and is categorized to the superfamily of human Class II cytokines 2 Mediators of Inflammation  Median (range) data are shown for the number of subjects indicated. na: not applicable. [11]. The receptor for IFN-1 is composed of IFNLR1 and IL-10R2. IFN-exhibits several common features with type I IFNs: antiviral activity [12], antiproliferative activity, and in vivo antitumour activity [13]. It is also involved in immune regulatory activities, such as the regulation of T helper (h)1/Th2 responses [14], modulation of dendritic cell function [15,16], and induction of mast cell accumulation [17]. The previous reports that levels of IFN-1 were elevated in the plasma of the patients with asthma [17] and that sputum IL-29 mRNA levels were higher in the steroid-treated asthmatic patients than in healthy controls [18] implicate that the cytokine is likely to contribute to the pathogenesis of allergic airway disorders. Based on the "allergic march" concept, we anticipate that IFN-1 may be involved in the pathogenesis of CSU. The aim of the current study is to investigate the potential role of IFN-1 in CSU, its cellular location, and its influence on inflammatory cell accumulation. (3) to detect IFN-1 expression in monocytes, B cells, neutrophils, and eosinophils: PE/Cy7-antihuman CD14, APC-anti-human CD19, and BV421-antihuman CD16 before 200 L of whole blood being added at room temperature for 15 min in dark. Following ligation of red blood cells, white blood cells were fixed and permeabilized by using Cytofix/Cytoperm Fixation/Permeabilization Kit according to the manufacturer's instructions. The cell pellets were then resuspended and rabbit anti-human IFN-1 followed by PE-conjugated goat anti-rabbit IgG antibodies were added at 4 ∘ C for 30 min. Finally, cells were resuspended in fluorescence-activated cell sorting-(FACS-) flow solution and analyzed with FACS verse flow cytometer (BD Biosciences, San Jose, CA). A total of 10,000 events were analyzed per population for each sample. Data were analyzed with CellQuest software (BD Immunocytometry systems).

Cell Dispersion from Skin Tissue and Flow Cytometry
Analysis of IFN-1 Expression. The procedure for dispersing skin tissue cells was mainly adopted from the procedure described previously by He et al. [21]. Briefly, skin tissues were digested with 2.0 mg/mL collagenase, 1.0 mg/mL hyaluronidase, and 1 g/mL DNase in DMEM for 70 min at 37 ∘ C. After centrifugation at 450 g for 6 min, the dispersed skin tissue cells were fixed by using a Cytofix/Cytoperm solution for 20 min at 4 ∘ C. Cells were then incubated with each labeled monoclonal antibody including BV421-antihuman CD117, PerCP-anti-human Fc RI , PE/Cy7-antihuman CD34, and rabbit anti-human IFN-1 (PE-conjugated goat anti-rabbit IgG) antibodies at 4 ∘ C for 30 min in the dark. Finally, cells were analyzed with FACS verse flow cytometer.

Immunohistochemical Stain and Cell Count.
Specimens fixed in Carnoy's fixative and embedded in paraffin wax. The staining procedure for double labeling immunohistochemistry was mainly adopted from the procedure described previously by He et al. [22]. For each section, the number of positively stained cells was counted in at least 30 fields (the area of each field equals 0.19 mm 2 ). Sections from six different donors were examined.
For detection of tryptase or chymase and IFN-1, sections of skin were incubated with biotin conjugated mouse antihuman IFN-1 antibody for 2 h followed by ExtrAvidin5peroxidase conjugate for 30 min and alkaline phosphatase conjugated G3 anti-tryptase antibody or alkaline phosphatase conjugated B7 anti-chymase antibody for 2 h. For detection of CD20 (marker of B cells) and IFN-1, the sections were incubated with biotin conjugated mouse anti-human IFN-1 antibody and L26 anti-CD20 (IgG2a isotype) antibody for 2 h and followed by adding ExtrAvidin-peroxidase conjugate for 30 min, and alkaline phosphatase conjugated goat anti-mouse IgG for 1 h. For detection of IFN-1 and CD8 (marker of cytotoxic T cells), lysozyme (marker of macrophages) or lactoferrin (marker of neutrophils), the sections were incubated with biotin conjugated mouse anti-human IFN-1 and anti-CD8, anti-lysozyme or anti-lactoferrin antibodies for 2 h, respectively, and followed by adding ExtrAvidin-peroxidase conjugate for 30 min and peroxidase conjugated sheep antirabbit IgG antibody for 1 h. Staining was developed over 4 min using DAB chromogen system and another 4 min with Fast Red TR/Naphthol AS-MX before being counterstained with Mayer's haematoxylin and mounted in aquamount.
For staining epidermis tissue, sequential sections were incubated with biotin conjugated mouse anti-human IFN-1 antibody for 2 h followed by ExtrAvidin-peroxidase conjugate for 30 min. Color was developed by using DAB chromogen system before being counterstained with Mayer's haematoxylin and mounted in aquamount.

Mouse Peritoneal Injection and Cell
Count. The procedure was adapted from that described previously [23]. Briefly, various concentrations of IFN-1 in the presence or absence of its specific antibody (anti-IFN-1, 3.0 g/mL), eotaxin (10 ng/mL), HSA (10 ng/mL), LPS (1.0 g/mL), and normal saline (NS) were injected in 0.5 mL volumes into the peritoneum of mice before their peritoneal lavage being collected for differential cell analysis. At 6 h following injection, animals were killed, and their peritoneal lavage fluids were collected and centrifuged. Cells were resuspended in 2.0 mL MEM, stained with 0.1% trypan blue, and enumerated using an Improved Neubauer haemocytometer (for total cell numbers). Cytocentrifuge preparations were made and stained with modified Wright's stain. Differential cell counts were performed for a minimum of 500 cells. The results were expressed as absolute numbers of lymphocytes, neutrophils and macrophages, eosinophils, and mast cells per mouse peritoneum. For the experiments investigating cell migration mechanism, groups of mice were pretreated intravenously with monoclonal antibodies against the adhesion molecules Lselectin (anti-CD62L, 1 mg⋅kg −1 ) and ICAM-1 (anti-CD54, 1 mg⋅kg −1 ) [24], respectively, for 30 min before intraperitoneal injection of IFN-1. At 6 h following injection, mice were sacrificed and their peritoneal lavages were processed as described above. Control animals received an equivalent dose of the corresponding normal rat or hamster IgG (isotype control).   are presented as scatter plot. Where Kruskal-Wallis analysis indicated significant differences between groups, for the preplanned comparisons of interest, the paired Mann-Whitney -test was employed. Where analysis of variance indicated significant differences between groups with ANOVA, Student's -test was applied. Correlations were determined by using Spearman rank correlation. For all analyses, < 0.05 was taken as significant.

Elevation of IFN-1 Levels in the Plasma of the Patients with CSU.
IFN-1 is a newly identified immune active cytokine that plays a role in antivirus [12] and some other immune regulatory activities [14]. With ELISA, we found that the level of IFN-1 was 2.0-fold higher in the plasma of the patients with CSU than the level in HC subjects (Figure 1(a)). In contrast, plasma IFN-1 level of the patients with eczema was similar to that of HC subjects (Figure 1(a)). The levels of IL-4 (Figure 1(b)) and TSLP (Figure 1(d)) in the plasma of the patients with CSU were lower than that of HC subjects. Levels of IFN-1 correlated negatively with IL-4, IL-10, and TSLP in the plasma of CSU and HC subjects ( Table 2). Levels of IFNin the plasma of CSU, eczema, and HC subjects were low and inconsistent (data not shown).

Enhanced Expression of IFN-1 in Peripheral Blood
Leukocytes of the Patients with Urticaria. In order to identify the potential source of IFN-1, we investigated the expression of IFN-1 in peripheral blood leukocytes. The results showed that the mean fluorescence intensity (MFI) of IFN-1 expression was not increased in CD4 + , CD14 + , CD16 + , CD19 + , CD16 − , and CD123 + HLA-DR − cell populations of CSU in comparison with that in HC subjects. However, IFN-1 expression in CD8 + cells of CSU was enhanced by 2.6-fold ( Figure 2).

Flow Cytometric Analysis of IFN-1 Containing Cells in Skin.
Mildly upregulated expression of IFN-1 in CD8 + , but not in CD4 + , CD14 + , CD16 + , CD19 + , CD16 − , and CD123 + HLA-DR − blood cells of CSU, indicated that elevated plasma level of IFN-1 was unlikely generated from peripheral blood leukocytes. In order to further identify IFN-1 containing cells we investigated the various cell types in human skin tissues. The results showed that CD14 + , CD16 + , CD19 + , and CD16 − cells and mast cells clearly expressed IFN-1 (Figure 3).

Immunohistochemical Analysis of IFN-1 Containing Cells in Skin.
In order to confirm the above flow cytometric analysis result, double labeling immunohistochemical analysis    technique was employed. The results revealed that IFN-1 containing cells were mainly located in the connective tissue of dermis (Figure 4). It was observed that approximately 3.8 tryptase + mast cells, 58.7 chymase + mast cells, 3.5 eosinophils, 1.6 B cells, 0.8 neutrophils, and 0.6 macrophages per mm 2 skin tissue expressed IFN-1 (Table 3). However, T cells in the tissues examined did not express IFN-1 (data not shown). Using adult foreskin, it was found that human epidermis tissue highly expressed IFN-1 ( Figure 5).

Induction of Inflammatory Cell Accumulation in Mouse
Peritoneum by IFN-1. In order to examine the potential proinflammatory action of IFN-1, IFN-1 was injected into mouse peritoneum. The results showed that IFN-1 induced a dose-dependent increase in the number of lymphocytes (Figure 6(a)), macrophages (Figure 6(b)), neutrophils (Figure 6(c)), mast cells (Figure 6(d)), and eosinophils (Figure 6(e)) in the peritoneum of mice at 6 h following injection.

Discussion
We have observed for the first time that the level of IFN-1 was markedly higher in the plasma of the patients with CSU, but not with eczema. Since CSU is an immunological disease and IFN-1 is an immune active cytokine, which is associated with various diseases such as cancer, viral infection, and Mediators of Inflammation  Table 3.  Figure 6: Interferon-(IFN-) 1 induces inflammatory cell accumulation in mouse peritoneum. Mice were treated with IFN-1 (ng/mL) in the presence or absence of its specific antibody (3.0 g/mL), eotaxin (10 ng/mL), human serum albumin (HSA, 10 ng/mL), lipopolysaccharide (LPS, 1.0 g/mL), and normal saline (NS) for 6 h before their peritoneal lavage being collected for differential cell analysis. For certain mice, anti-CD54 antibody (anti-CD54, 1 mg⋅kg −1 ) and anti-CD62L antibody (anti-CD62L, 1 mg⋅kg −1 ) were injected via tail vein 30 min before IFN-allergy, it is most likely that IFN-1 is involved in the pathogenesis of CSU. Lack of significant alteration of IFN-1 in plasma of patients with eczema and low concentration of IFN-in the plasma of CSU indicate that increased level of IFN-1 in CSU is specific.
To our surprise, the levels of IL-4 and TSLP were significantly lower in the plasma of CSU, which was very different from that observed in allergic disorders. It was previously shown that level of IL-4 was increased in the serum of allergic rhinitis [25] and allergic children [26], and TSLP were significantly increased in the serum of atopic dermatitis children [27]. Our findings provided further evidence that CSU is unlikely an allergic disease [28] though its IgE levels in blood appear to be elevated. The negative correlations between plasma IFN-1 and IL-4 and IL-10 and TSLP in the patients with CSU suggested that IFN-1 may not be released from the same cell source as the other three cytokines.
It is not too difficult to think that the increased level of IFN-1 was released from peripheral blood leukocytes as eosinophils [29] mononuclear cells [13], and monocytederived dendritic cells [10] were reported to secrete IFN-1. However, among the leukocytes examined, only CD8 + T cells expressed more IFN-1 in blood from CSU than blood from HC subjects, which can hardly explain the elevated plasma level of IFN-1. In order to find the potential source of IFN-1, we investigated expression of IFN-1 in human foreskin tissue. Because CD14 + and CD19 + cells and CD16 + and CD16 − granular cells and mast cells clearly expressed IFN-1 and there are a large number of mast cells, CD4 + T cells, monocytes, neutrophils, basophils [3], eosinophils [30], and B cells [31] in CSU skin tissue, we believe that the enhanced level of IFN-1 was most likely from these cells. It is worthy emphasizing that human epidermis tissue expresses huge amount of IFN-1, which could well be the source of plasma IFN-1 in CSU. Obviously, further study is required to clearly address the issue.
We also observed that IFN-1 was able to induce lymphocyte and neutrophil infiltration via an ICAM-1-and L-selectin-dependent mechanism. Since little information on cell migration induced by IFN-is available the reports that ICAM-1 is involved in each step of neutrophil extravasation [32] and CXC chemokine-induced neutrophil accumulation is dependent on neutrophil L-selectin [33] and that ICAM-1 mediates migration of Th1 and Th17 cells across human vascular endothelium [34] and L-selectin recruits Th1 cells in contact hypersensitivity of skin [35] may support our current observation. Similarly, the previous finding that monocyte migration to inflamed skin and lymph nodes is controlled by L-selectin [36] may help to understand our current observation that IFN-1 induced macrophage accumulation appeared to rely on activation of L-selectin. It was observed in the present study that IFN-1 induced recruitment of eosinophils and mast cells was through an ICAM-1-dependent manner. The findings that hapten-induced colonic eosinophilic inflammation is critically dependent on ICAM-1 [37] and that IL-4 provoked aggregation of human mast cells by promoting LFA-1/ICAM-1 adhesion molecules [38] may help to explain our current observation.
In conclusion, markedly elevated IFN-1 level in the plasma of patients with CSU suggests that IFN-1 plays a role in the pathogenesis of CSU through its ability in recruiting inflammatory cells. Tissue cells such as mast cells and macrophages rather than peripheral blood leukocytes were likely the source of plasma IFN-1. Blocking IFN-1 production may help to reduce the accumulation of inflammatory cells in the involved skin in CSU.

Bulleted Statements
(i) Role of interferon-(IFN-) 1 in innate immunity is recognized recently and regarded as a potent bioactive molecule. However, little is known about its role in chronic spontaneous urticaria (CSU).
(ii) Markedly elevated IFN-1 level in CSU plasma suggests that IFN-1 is involved in CSU.
(iii) The ability in recruiting inflammatory cells suggests that IFN-1 plays a role in the pathogenesis of CSU.
(iv) IFN-1 could be generated from tissue inflammatory cells.