IL-31 is crucial for induction of pruritus, but not inflammation, in contact hypersensitivity

IL-31, which is a member of the IL-6 family of cytokines, is produced mainly by activated CD4+ T cells, in particular activated Th2 cells, suggesting a contribution to development of type-2 immune responses. IL-31 was reported to be increased in specimens from patients with atopic dermatitis, and IL-31-transgenic mice develop atopic dermatitis-like skin inflammation, which is involved in the pathogenesis of atopic dermatitis. However, the role of IL-31 in development of contact dermatitis/contact hypersensitivity (CHS), which is mediated by hapten-specific T cells, including Th2 cells, is not fully understood. Therefore, we investigated this using IL-31-deficient (Il31−/−) mice, which we newly generated. We demonstrated that the mice showed normal migration and maturation of skin dendritic cells and induction of hapten-specific T cells in the sensitization phase of FITC-induced CHS, and normal induction of local inflammation in the elicitation phase of FITC- and DNFB-induced CHS. On the other hand, those mice showed reduced scratching frequency and duration during FITC- and/or DNFB-induced CHS. Our findings suggest that IL-31 is responsible for pruritus, but not induction of local skin inflammation, during CHS induced by FITC and DNFB.

IL-31, which is a member of the IL-6 family of cytokines, is produced mainly by activated CD4 + T cells, in particular activated Th2 cells, mast cells, macrophages and dendritic cells [1][2][3][4] .Human and mouse IL-31 genes are located on chromosome 12q24.31 and chromosome 5, respectively, and their amino acid sequences show 31% homology 1 . Expression of IL-31 mRNA was reported in various human tissues, such as the testis, bone-marrow, skeletal muscle and kidney 1 . The receptor for IL-31 is a heterodimer, consisting of IL-31 receptor A (IL-31 RA) and oncostatin M receptor (OSMR) 1 . In humans and/or mice, IL-31 RA mRNA expression is found in various tissues, such as the testis, bone marrow, skin and dorsal root ganglia, and in various cells, such as activated monocytes, macrophages, dendritic cells (DCs), eosinophils, basophils and keratinocytes, while OSMR mRNA is broadly expressed in many tissues [5][6][7] . In cultures of various human cell lines such as keratinocytes 1 , intestinal and bronchial epithelial cell lines 8,9 , DCs 7 , monocytes and macrophages 10 , IL-31 has been shown to induce expression of such cytokines as IL-6, IL-8 and/or TNF and such chemokines as CCL2, CCL4, CCL5, CCL17 and/or CCL22.
IL-31 has been implicated to be involved in the pathogenesis of such allergic disorders as rhinitis, asthma and dermatitis 5 . Indeed, increased levels of IL-31 were observed in nasal secretions of patients with allergic rhinitis 11 . IL-31 levels in sera or PBMCs were significantly increased in patients with asthma 12 , and IL-31 can activate human bronchial epithelial cell lines, as described above. However, ovalbumin-induced airway inflammation SCIEnTIfIC REPORts | (2018) 8:6639 | DOI: 10.1038/s41598-018-25094-4 developed normally in mice treated with anti-IL-31RA neutralizing Ab 13 , suggesting that IL-31 is not crucial for induction of allergic airway inflammation in that model.
In regard to skin diseases, elevated levels of IL-31 were observed in such specimens as plasma, sera and/ or cutaneous biopsies from patients with atopic dermatitis 14,15 or contact dermatitis 16 . In particular, IL-31 was implicated in itching by inducing activation of sensory nerve cells 5 . In support of that notion, transgenic mice (Tg mice), which express IL-31 systemically or specific for lymphocytes, spontaneously develop skin inflammation resembling human atopic dermatitis and show severe scratching behavior accompanied by exfoliation of epidermis 1 . Expression of IL-31 mRNA was increased in inflamed lesions of skin from mice that developed allergic dermatitis resembling human atopic dermatitis 17 . Administration of anti-IL-31 or anti-IL-31RA neutralizing Ab resulted in attenuation of scratching behavior, but not the severity of skin inflammation, in that mouse model 18,19 or in patients with atopic dermatitis 20 . On the other hand, the role of IL-31 in the development of contact dermatitis/contact hypersensitivity (CHS) is not well understood. CHS is a T-cell-mediated cutaneous allergic response 21 . Several studies using type-2-cytokine-deficient mice showed that type-2 cytokines are important for the development of CHS (reviewed in 22 ), suggesting that IL-31 may be involved in induction of CHS. Il31ra −/− mice showed enhanced type-2 immune responses due to hyperresponsiveness to OSM by increased formation of OSM receptors (the complex of gp130 and OSMR) rather than due to failure to form IL-31 receptors (the complex of IL-31RA and OSMR) 13 . Accordingly, we deemed that murine strain to be unsuitable for examining the role of IL-31 in CHS, and we instead used IL-31-deficient (Il31 −/− ) mice, which we newly generated.

Results
Generation of Il31 −/− mice. Il31ra −/− mice were reported to have not only loss of IL-31 function but also elevated responsiveness to OSM 13 , suggesting that it is necessary to elucidate the exact roles of IL-31 in vivo by using Il31 −/− mice rather than Il31ra −/− mice. Therefore, we newly generated Il31 −/− mice by replacing Il31 genes with a cassette consisting of IRES-EGFP and a neomycin resistance gene, flanked by loxP sequences (Fig. 1a). Il31 −/− mice were born at the expected Mendelian ratio, fertile, and did not show any gross phenotypic abnormalities under specific-pathogen-free housing conditions (data not shown). Expression of Il31 mRNA was below the limit of detection by qPCR in the lung and in PMA-ionomycin-stimulated spleen cells of Il31 −/− mice ( Fig. 1b and data not shown). No apparent abnormalities were found in the proportions of immune cells in the thymus, LNs, spleen and/or bone marrow between wild-type and Il31 −/− mice (Fig. 1c, and data not shown).
IL-31 is not essential for skin DC function or hapten-specific LN cell response in the sensitization phase of CHS. Treatment of mice that developed atopic dermatitis-like skin inflammation with anti-IL-31 or anti-IL-31RA neutralizing Ab resulted in attenuation of scratching behavior, but not the severity of skin inflammation 18,19 . On the other hand, the role of IL-31 in the development of CHS is not well understood. After epicutaneous exposure to haptens, skin DCs capture the haptens and migrate from the skin into draining LNs, followed by antigen presentation to naïve T cells to induce hapten-specific T cells 21 . To elucidate the role of IL-31 in skin DC migration and maturation, wild-type and Il31 −/− mice were treated epicutaneously with FITC. Twenty-four hours later, the proportion of FITC-positive cells among MHC class II hi CD11c + cells in draining LNs and their expression of co-stimulatory molecules were determined by flow cytometry. The proportions of FITC-positive cells were comparable between the wild-type and Il31 −/− mice (Fig. 2a). Such co-stimulatory molecules as CD86, CD40 and OX40L on the FITC-positive cells were also similarly expressed in the Il31 −/− and wild-type mice (Fig. 2b).
We next elucidated whether the lack of IL-31 influenced hapten-specific T-cell induction in the sensitization phase of CHS. After epicutaneous sensitization of Il31 −/− and wild-type mice with FITC, draining LN cells were harvested and cultured in the presence and absence of FITC. After 48 and 72 hours of culture, the levels of IFN-γ, IL-4 and IL-17 in the culture supernatants were comparable in the two mouse strains (Fig. 3). Therefore, these observations suggest that, for the sensitization phase of CHS, IL-31 is not essential for such skin DC functions as migration, maturation and hapten-specific T-cell induction.

IL-31 is involved in pruritus, but not inflammation, during CHS. We next investigated whether IL-31
is involved in the pathogenesis of FITC-induced CHS. The degree of skin inflammation was assessed by measuring the thickness of the ear skin after challenge with FITC. The ear thickness was comparable between wild-type and Il31 −/− mice at the indicated time points after the initial epicutaneous application of FITC (Fig. 4a). The degree of skin inflammation assessed by histological analysis was also similar between the wild-type and Il31 −/− mice at 24 hours after challenge with FITC or vehicle (Fig. 4b). We also used ELISA to examine the levels of FITC-specific IgG subsets and total IgE in sera from naïve and FITC-treated mice. After FITC challenge, the levels of FITC-specific IgG1, IgG2b and IgG2c and total IgE in sera from Il31 −/− mice increased similarly to those from wild-type mice (Fig. 4c). On the other hand, the scratching frequency and duration were significantly decreased in Il31 −/− mice compared with wild-type mice at 24 hours after FITC challenge (Fig. 5). Similar phenotypes were also observed for Il31 −/− mice during DNFB-induced CHS. Although the ear thickness was slightly, but not significantly, increased in Il31 −/− mice compared with wild-type mice at the indicated time points after the initial epicutaneous application of DNFB (Fig. 6a), the scratching frequency and duration were significantly decreased in Il31 −/− mice compared with wild-type mice at 24 hours after DNFB challenge (Fig. 6b). GFP + cells were hardly detected in LNs from Il31 gfp/+ mice during DNFB-induced CHS (data not shown). We found that a part of F4/80 + and CD11c + cells, but not CD3 + cells, Gr1 + cells and tryptase + cells (data not shown), were producers of IL-31 in skin from wild-type mice, but not Il31 −/− mice, at 24 h after DNFB challenge (Fig. 7). The signals in keratinocytes from wild-type mice were due to non-specific binding of the anti-IL-31 Ab, since they were detected even in Il31 −/− mice (Fig. 7). These observations suggest that IL-31 is important for induction of pruritus, but not inflammation, in CHS induced by FITC and DNFB. Discussion IL-31 is preferentially produced by Th2 cells in Th subsets, and excessive IL-31 production in IL-31 Tg mice resulted in development of dermatitis due to enhancement of type-2 immune responses 1 . By contrast, Il31ra −/− mice showed enhanced type-2 immune responses during nematode infection 23 , suggesting that IL-31R signaling has a suppressive role in type-2 immune responses. Regarding the apparent discrepancies in phenotype between IL-31 Tg mice 1 and Il31ra −/− mice 23 , the latter were reported to be hyperresponsive to OSM due to increased formation of OSM receptors (gp130 and OSMR) caused by the lack of IL-31R (IL-31RA and OSMR), resulting in OSM-mediated suppression of type-2 immune responses 13 . Therefore, to investigate the precise role of IL-31 in vivo, we newly generated Il31 −/− mice for the present study. Using those mice, we demonstrated for the first time that IL-31 is important for induction of pruritus, but not inflammation, in CHS induced by FITC and DNFB.
IL-31 is considered to be involved in the pathogenesis of type-2 cytokine-associated allergic disorders such as rhinitis, asthma and dermatitis 5 . In support of that notion, the levels of IL-31 were increased in patients with allergic rhinitis 11 , asthma 12 and atopic dermatitis 14,15 . DNFB-and/or FITC-induced CHS was attenuated in Il4 −/− mice 24,25 and/or stat6 −/− mice 26 , suggesting that type-2 cytokines are important for induction of DNFBand/or FITC-induced CHS. In addition, the levels of IL-31 were also elevated in patients with contact dermatitis 16 . These observations suggest that IL-31 may be involved in induction of type-2-cytokine-associated CHS. However, we demonstrated that Il31 −/− mice showed normal migration and maturation of skin DCs and induction of hapten-specific T cells in the sensitization phase of FITC-induced CHS, and normal induction of local inflammation in the elicitation phase of FITC-and DNFB-induced CHS, indicating that IL-31 is not essential for induction of skin inflammation during FITC-or DNFB-induced CHS.
As a unique function of IL-31, it is known to be involved in itching via signal activation of IL-31 receptors on sensory nerve cells 5,27 . Indeed, IL-31 Tg mice show severe scratching behavior accompanied by exfoliation of epidermis 1 . Administration of anti-IL-31 or anti-IL-31RA neutralizing Ab resulted in attenuation of scratching behavior, but not the severity of skin inflammation, in mice that developed atopic dermatitis-like skin inflammation 18,19 . Consistent with this, we showed that Il31 −/− mice had reduced scratching frequency and duration in FITC-and/or DNFB-induced CHS, indicating that IL-31 is responsible for pruritus in the setting. However, the underlying mechanism of this remains unclear and should be further investigated in future studies.
In summary, we demonstrated here that IL-31 is involved in pruritus reactions without affecting induction of local skin inflammation in CHS.

Methods
Mice. C57BL/6 N wild-type mice were obtained from Japan SLC, Inc. (Hamamatsu, Japan). Il31 −/− mice on the C57BL/6 N background were generated as follows. The Il31 gene was disrupted by replacement of the region from a part of exon 2 to a part of the intron behind exon 3 with a cassette consisting of IRES-EGFP and a neomycin resistance gene, flanked by loxP sequences (Fig. 1a). Homologous regions were amplified by PCR using the following primers: 5′-CAACCTATTCCTAGTTCCCTCACC-3′ and 5′-AAAGCAGCACCAGGGTAGGCTTCG-3′ to generate a 6-kb fragment, and 5′-TGAATGAGGGGAACAGAAAATTACC-3′ and 5′-TGATGAATAATGATATTCCTTACC-3′ to generate a 2-kb fragment. The targeting vector was electroporated into C57BL/6 N ES cells (EGR-101, kindly provided by Dr. Masaru Okabe, Osaka University). Male chimeric mice were obtained from three distinct targeted clones and mated with C57BL/6 N female mice. Genotyping of Il31 −/− mice was performed by PCR using the following primers: common 1 (5′-GACAACCCTTTATATATTGCCTGG-3′), WT 1 (5′-GCTTCCATAGCTGGCTTTATATCG-3′) and IL-31 KO (5′-CGACACCGGCCTTATTCCAAGCGG-3′). The common 1 and WT 1 primers were used for detection of wild-type alleles (~550 bp), and the common and IL-31 KO primers were used for detection of mutant alleles (~390 bp). For generation of Il31 egfp/egfp mice, a plasmid carrying Cre cDNA (pCAG-Cre, kindly provided by Dr. Jun-ichi Miyazaki, Osaka University) was injected into fertilized eggs from Il31 −/− mice to deplete the neomycin resistance gene, and it was flanked by loxP sequences. The eggs were then transferred into pseudopregnant female mice. Genotyping of Il31 egfp/egfp mice was performed by PCR using the following primers: common 2 (5′-TGGGAAGATACATGAAACCAATCC-3′), WT 2 (5′-CACTGGCTGCTCTTCCAGAGGACC-3′) and Neo r (5′-GACGTGCTACTTCCATTTGTCACG-3′). The common and WT primers were used for detection of wild-type alleles (~493 bp), and the common and Neo r primers were used for detection of mutant alleles (no band). All mice were housed in a specific-pathogen-free environment at The Institute of Medical Science, The University of Tokyo. The animal protocol for experiments was approved by the Institutional Review Board of the Institute (A11-28 and A14-10), and all experiments were conducted according to the ethical and safety guidelines of the Institute.
Quantitative PCR. Total RNA was prepared from lungs using Sepasol (Nacalai Tesque, Inc.) and treated with DNase (TURBO DNA-free-kit; Thermo Fisher Scientific Inc., MA). cDNA was synthesized from the isolated RNA by RT-PCR (PrimeScript RT Reagent kit; TAKARA BIO Inc., Shiga, Japan). Quantitative real-time PCR was performed with SYBER Premix Ex Taq (TAKARA BIO Inc.) or SYBER Premix DimerEraser (TAKARA BIO Inc.) using a CFX384 TM Touch Real-time PCR Detection System (BioRad Laboratories, Inc., Hercules, CA). Relative gene expression was determined against HPRT gene expression. The following PCR primers were designed: forward primer 5′-ATACAGCTGCCGTGTTTCAG-3′ and reverse primer 5′-AGCCATCTTATCACCCAAGAA-3′ for Il31 mRNA; and forward primer 5′-GGCCAGACTTTGTTGGATTTG-3′ and reverse primer 5′-CGCTCATCTTA GGCTTTGTATTTG-3′ for Hprt mRNA.   Induction of contact hypersensitivity (CHS). CHS was induced with FITC and DNFB as described elsewhere 28,29 . Briefly, 2 days after shaving the back hair with electrical clippers, the back skin was treated with 150 µl of 1.0% FITC isomer I in a 1:1 mixture of acetone and dibutyl phthalate or 25 µl of 0.5% DNFB (SIGMA) in a 4:1 mixture of acetone and olive oil. Five days later, the respective mice were challenged with 40 µl of 0.5% FITC isomer I or 0.2% DNFB (each surface of the left ear) and 40 µl of the vehicle alone (each surface of the right ear). The ear thickness was measured before and after FITC challenge using dial thickness gauges (Ozaki MFG. Co., Ltd., Tokyo, Japan).
Measurement of levels of serum immunoglobulins. Sera were collected from mice 4 days after FITC challenge during FITC-induced CHS. The levels of FITC-specific Igs in the sera were determined by ELISA as described elsewhere 29 . In brief, 96-well flat-bottom plates (Thermo Fisher Scientific Inc.) were coated with 20 µg/ml FITC-conjugated OVA at 4 °C overnight. After the wells were blocked with PBS containing 1% FBS, optimally diluted serum samples (IgG1 = 1/100; IgG2b = 1/100; IgG2c = 1/10) were placed in the wells, and the plates were incubated at room temperature for 2 hours. After washing, HRP-conjugated anti-mouse IgG1, IgG2b or IgG2c mAb (Bethyl Laboratories, Inc., TX) was added, followed by incubation at room temperature for 1 hour. For enzymatic reaction, TMB substrate (Nacalai Tesque Inc.) was used as a substrate. The reaction was stopped by addition of 0.2 M H 2 SO 4 , and then the absorbance (450 nm) was measured using a VersaMax (Molecular Devices, LLC, Sunnyvale, CA). Data show the absorbance value at 450 nm. The levels of total IgE in sera were determined using an ELISA kit (Bethyl Laboratories, Inc.) in accordance with the manufacturer's instructions.
FITC-specific LN cell responses. FITC-specific LN cell responses were determined as described elsewhere 28,29 . Briefly, mice were epicutaneously treated with 1.0% FITC isomer I in a 1:1 mixture of acetone and dibutyl phthalate on both the left and right ears (20 μl on one surface of each ear). Six days later, submaxillary LNs were collected. The LN cells were cultured in the presence and absence of 40 μg/ml FITC at 37 °C for 72 hours. The levels of cytokines in each culture supernatant were determined with mouse IFN-γ, IL-4 and IL-17A ELISA kits obtained from Biolegend (San Diego, CA) or Peprotech (Rocky Hill, NJ). Assessment of scratching behaviour. SCLABA-Real (NOVELTEC) was used to assess scratching behavior (frequency and duration) at 24 and 48 hours after challenge with FITC or DNFB during FITC-or DNFB-induced CHS.
Histology. Twenty-four hours after FITC or vehicle challenge, the ear skins were harvested, fixed in Carnoy's fluid and embedded in paraffin. Sections were prepared and stained with hematoxylin-eosin.