Research ArticleALLERGY

Clec10a regulates mite-induced dermatitis

See allHide authors and affiliations

Science Immunology  06 Dec 2019:
Vol. 4, Issue 42, eaax6908
DOI: 10.1126/sciimmunol.aax6908

Dermatitis details

House dust mite (HDM) is an allergen associated with a variety of diseases, including asthma and atopic dermatitis (AD). The NC/Nga mouse strain develops severe AD in response to HDM, and Kanemaru et al. have now identified a stop-gain mutation, the Clec10a C-type lectin receptor, that is associated with this response. HDM-induced dermatitis was driven by TLR4-mediated inflammatory responses, and asialoglycoprotein receptor 1 (Asgr1) functions as a Clec10a homolog in humans. They also identified a mucin-like molecule in HDM that can function as a ligand for Clec10a and Asgr1. These results provide insight into mechanisms associated with HDM-mediated dermatitis.

Abstract

House dust mite (HDM) is a major allergen that causes allergic diseases such as atopic dermatitis. However, the regulatory mechanisms of HDM-induced immune responses are incompletely understood. NC/Nga mice are an inbred strain that is more susceptible to HDM and develops more severe dermatitis than other strains. Using whole-exome sequencing, we found that NC/Nga mice carry a stop-gain mutation in Clec10a, which encodes a C-type lectin receptor, Clec10a (MGL1/CD301a). The repair of this gene mutation using the CRISPR-Cas9 system ameliorated HDM-induced dermatitis, indicating that the Clec10a mutation is responsible for hypersensitivity to HDM in NC/Nga mice. Similarly, Clec10a−/− mice on the C57BL/6J background showed exacerbated HDM-induced dermatitis. Clec10a expressed on skin macrophages inhibits HDM-induced Toll-like receptor 4 (TLR4)–mediated inflammatory cytokine production through the inhibitory immunoreceptor tyrosine activating motif in its cytoplasmic portion. We identified asialoglycoprotein receptor 1 (Asgr1) as a functional homolog of mouse Clec10a in humans. Moreover, we found that a mucin-like molecule in HDM is a ligand for mouse Clec10a and human Asgr1. Skin application of the ligand ameliorated a TLR4 ligand-induced dermatitis in mice. Our findings suggest that Clec10a in mice and Asgr1 in humans play an important role in skin homeostasis against inflammation associated with HDM-induced dermatitis.

INTRODUCTION

House dust mite (HDM) is one of the major allergens associated with allergic diseases, such as atopic dermatitis (AD), rhinitis, and asthma (1, 2). Most patients with AD and asthma are highly sensitive to HDM (1, 3). Allergenicity of HDM is associated with the mites themselves, including proteolysis and nonproteolytic proteins; exoskeletal components, such as chitin and fecal pellets of HDM consisting of food, debris, and proteolytic enzymes; immunogenic proteins; and mite-associated bacterial and fungal products (4). These components directly activate protease-activated receptors and pattern-recognition receptors (PRRs) for pathogen-associated molecule patterns or indirectly activate receptors for damage-associated molecule patterns (2). HDM contains several antigenic and nonantigenic components that induce or modify allergic diseases, but the exact mechanism of HDM-induced allergic responses is complicated and remains incompletely understood.

NC/Nga mice, an inbred strain established from Japanese fancy mice, are the most extensively studied animal model of human AD. They spontaneously develop AD-like skin symptoms such as itching, erythema, hemorrhage, scaling, dryness, and alopecia (57) when raised under conventional husbandry conditions but not when housed under specific pathogen–free (SPF) conditions (8), suggesting that epicutaneous exposure to certain allergens may be an important factor that induces symptoms in NC/Nga mice. HDM exposure of NC/Nga mice under SPF conditions induces more severe dermatitis than those in other mouse strains (e.g., ICR, C57BL/6, and BALB/c) (57). This evidence suggests that skin homeostasis is not controlled in NC/Nga mice during epicutaneous exposure to HDM. However, it is still unclear as to why NC/Nga mice are more sensitive to HDM, and whether findings from studies in NC/Nga mice can be applied to understanding dermatitis in other mouse strains and humans. In this study, we addressed these issues and showed that a C-type lectin receptor Clec10a (MGL1/CD301a) plays an important role in skin homeostasis against HDM exposure.

RESULTS

Clec10a in NC/Nga mice has a stop-gain mutation

To identify genetic factors associated with hypersensitivity to HDM in NC/Nga mice, we performed whole-exome sequencing of the NC/Nga genome and used the C57BL/6J genome as a reference. Among 70,772 variants detected in the NC/Nga genome, we found seven loss-of-function gene mutations that specifically exist in NC/Nga mice but not in other 19 mouse strains (table S1 and fig. S1A). Among them, we focused on a stop-gain mutation (C to T) at position c.706 (p.Q236X) of Clec10a (NM_010796) (Fig. 1A) in light of its relatively high expression on hematopoietic cells based on information in the BioGPS database (fig. S1A) (9). The encoded protein, Clec10a (MGL1/CD301a), is a member of the type II transmembrane C-type lectin receptor family and senses terminal galactose moieties of exogenous and endogenous antigens (1013). The c.706C>T mutation in the NC/Nga genome lies within the coding region of the C-type lectin-like domain (CTLD) (Fig. 1, A and B).

Fig. 1 Clec10a in NC/Nga mice has a stop-gain mutation.

(A and B) DNA sequencing of the stop-gain mutation site (c.706) in Clec10a (gene ID: 17312, NM_010796) (A) and Clec10a protein (B) of C57BL/6J and NC/Nga mice. White boxes indicate the coding region of Clec10a. TM, transmembrane domain. N, N terminus. (C) Cell surface Clec10a expression of MPs (CD64+MerTK+), cDCs (CD64MerTK), and monocyte-derived DCs (CD64MerTKlo) in the skin PICD45+MHCII+Lineage (CD3, CD19, NK1.1, and Ly-6G)EpCAM cells of C57BL/6J and NC/Nga mice. Shaded histograms show staining with isotype control Ab. (D) Fluorescence microscopy of tissue sections from the dorsal skin, stained with an anti-Clec10a mAb and the DNA binding dye DAPI. E, epidermis. D, dermis. Scale bars, 100 μm. (E and F) RAW264.7 transfectants expressing Flag-tagged Clec10ac.706C (derived from C57BL/6J mice)–IRES-GFP or Flag-tagged Clec10ac.706T (derived from NC/Nga mice)–IRES-GFP were analyzed for cell surface (E) or intracellular (F) expression of Flag by flow cytometry. Data are representative of two independent experiments.

Clec10ac.706C>T mutation in NC/Nga mice prevents Clec10a cell surface expression

Flow cytometry revealed the presence of Clec10a on the cell surfaces of macrophages (MPs) (CD64+MerTK+), conventional dendritic cells (cDCs) (CD64MerTK), and monocyte-derived DCs (CD64MerTKlo) in the skin CD45+MHCII+LineageEpCAM cells and on MPs in the peritoneal cavity of C57BL/6J, but not NC/Nga, mice (Fig. 1C and fig. S1, B and C). In contrast, nonhematopoietic cells (CD45), CD45+MHCII cells, CD45+MHCII+ (Lin, EpCAM)+ cells, and neutrophils (CD45+CD11b+Ly-6G+) in the skin of C57BL/6J mice lacked Clec10a expression (fig. S1, D and E). Immunohistochemistry also showed the expression of Clec10a in the skin cells of C57BL/6J but not NC/Nga mice (Fig. 1D). To further analyze characteristics of Clec10a expression in NC/Nga mice, we transfected RAW264.7 MP cells with C57BL/6J- and NC/Nga-derived Clec10a complementary DNAs (cDNAs) tagged with the Flag-encoding sequence and carrying the internal ribosomal entry site–green fluorescent protein (IRES-GFP) sequence at the 3′ end. Flag was expressed on the cell surfaces of transfectants expressing C57BL/6J Clec10a but not NC/Nga Clec10a (Fig. 1E). However, the transfectant expressing NC/Nga Clec10a showed intracellular expression of Flag (Fig. 1F). These results suggest that the Clec10a mutation impairs the transport of Clec10a to the cell surface.

Clec10ac.706C>T mutation in NC/Nga mice causes HDM hypersensitivity

To examine whether the Clec10a mutation in NC/Nga mice is involved in HDM hypersensitivity, we used the CRISPR-Cas9 system (14) to generate mutant NC/Nga mice that carried the Clec10a of C57BL/6J mice (Clec10ac.706C) instead of their native sequence (Clec10ac.706T) (Fig. 2A). Clec10a mRNA expression in the skin of the mutant NC/Nga mice (Clec10ac.706T>C) was significantly increased compared with that of Clec10ac.706T mice (fig. S2A). NC/Nga mice with heterozygous Clec10a alleles (Clec10ac.706T/C) restored Clec10a expression on the cell surfaces of MPs in the skin and peritoneal cavity (Fig. 2, B and C, and fig. S2B). In addition, symptoms of HDM-induced skin inflammation, including erythema, dryness, and increased epidermal thickness, were ameliorated in the NC/Nga mice that carried Clec10a allele of C57BL/6J mice (Fig. 2, D to G). Furthermore, flow cytometry revealed that neutrophil infiltration into the skin on day 6 after HDM treatment was decreased in Clec10ac.706T/C compared with wild-type (WT) (Clec10ac.706T/T) NC/Nga mice (Fig. 2H). In contrast, serum levels of immunoglobulin E (IgE) were comparable between two genotypes of mice (fig. S2C). These results suggest that the Clec10a mutation (Clec10ac.706C>T) in NC/Nga mice causes hypersensitivity to HDM and severe dermatitis but not increased T helper 2 (TH2) responses.

Fig. 2 Clec10ac.706C>T mutation in NC/Nga mice causes HDM hypersensitivity.

(A) DNA sequencing of the stop-gain mutation site (c.706) in Clec10a of NC/Nga-Clec10ac.706T/C mice. (B) Cell surface Clec10a expression of MPs (CD64+MerTK+), cDCs (CD64MerTK), and monocyte-derived DCs (CD64MerTKlo) in the skin PICD45+MHCII+Lineage (CD3, CD19, NK1.1, and Ly-6G)EpCAM cells of NC/Nga- Clec10ac.706T/C mice. Shaded histograms show staining with isotype control Ab. (C) Fluorescence microscopy of tissue sections from the dorsal skin of NC/Nga-Clec10ac.706T/C mice, stained with an anti-Clec10a mAb and the DNA binding dye DAPI. (D to H) HDM ointment was applied twice each week to the dorsal skin of NC/Nga-Clec10ac.706T/T and Clec10ac.706T/C mice. Dermatitis score (D), the appearance on day 14 (E), histology (H&E) and epidermal thickness (F and G), populations of neutrophils (CD11b+Ly-6G+), eosinophils (CD11b+Siglec-F+), and Ly-6Chi monocytes (CD11b+Ly-6GSiglec-FLy-6Chi) among skin PICD45+ cells (H). Scale bars, 100 μm (C and F). *P < 0.05, **P < 0.01, and ***P < 0.001 (one-way ANOVA with post hoc test). Data show mean values ± SEM. Data are pooled from four (D, G, and H) experiments or representative of two independent experiments.

Clec10a inhibits HDM-induced dermatitis

To examine how the deficient cell surface expression of Clec10a is involved in HDM hypersensitivity, we used Clec10a-deficient (Clec10a−/−) mice on the C57BL/6J background and topically applied HDM to the dorsal skin. As seen in the skin of NC/Nga mice, Clec10a−/− mice showed exacerbated dermatitis, including enhanced erythema, dryness, and epidermal thickness, on days 5 to 6 and increased infiltration of neutrophils, but not other myeloid and lymphocyte lineage cells, in the skin on days 1 to 3 after HDM treatment (Fig. 3, A to E, and fig. S3, A to C) compared with WT C57BL/6J mice. In contrast, skin barrier function, as analyzed by the transepidermal water loss (TEWL) test, was comparable between WT and Clec10a−/− mice (fig. S3D). Depletion of neutrophils by injection of the anti–Ly-6G antibody ameliorated dermatitis in Clec10a−/− mice to a level comparable to that in WT mice (Fig. 3, F and G, and fig. S3E), indicating that exacerbated dermatitis in Clec10a−/− mice was dependent on neutrophils.

Fig. 3 Clec10a inhibits HDM-induced dermatitis.

(A to G) HDM ointment (A to G) or sodium dodecyl sulfate (SDS) alone (D) was applied twice each week to the dorsal skin of C57BL/6J-WT and Clec10a−/− mice with (F and G) or without (A to E) injecting control Ab or anti-Ly-6G (αLy-6G) Ab. The appearance (A) and dermatitis score (B) on day 5 (n = 18) and histology (H&E) and epidermal thickness at the indicated time points (C, D, F, and G) were compared between WT and Clec10a−/− mice. Scale bars, 100 μm. Flow cytometry identifying neutrophils (CD11b+Ly-6G+) among skin PICD45+ cells of WT and Clec10a−/− mice at the indicated time points (E). (H) Quantitative RT-PCR analysis of mRNA from WT and Clec10a−/− MHCII+ MPs and DCs sorted from the skin at 3 hours after HDM topical application. RQ, relative quantity. (I) Cytometric bead array (CBA) analysis of culture supernatants from HDM-stimulated WT and Clec10a−/− CD115+–enriched BMMPs (n = 3). *P < 0.05, **P < 0.01, ***P < 0.001. ns, not significant (one-way ANOVA with post hoc test). Data show mean values ± SEM. Data are pooled from five (B and E) or three (D, G, and H) experiments or representative of two (C) or three (I) independent experiments.

Although the serum levels of IgG1 were higher in Clec10a−/− mice than in WT mice at 2 weeks after the HDM treatment, IgE and IgG2c titers were comparable between the two genotypes of mice (fig. S3F). In addition, expression of Il4 and Gata3 by CD4+ T cells in the draining lymph nodes of HDM-treated skin was comparable between WT and Clec10a−/− mice (fig. S3G). Therefore, Clec10a likely suppresses dermatitis by reducing neutrophil infiltration rather than diminishing TH2 responses.

Clec10a is expressed on dermal MHCII+ MPs and DCs (Fig. 1, C and D). In contrast, neutrophils and nonhematopoietic cells in the skin did not express Clec10a even after HDM treatment (fig. S4, A and B). Therefore, we sorted MHCII+ MPs and DCs from skin cell samples of WT and Clec10a−/− C57BL/6J mice treated topically with HDM (fig. S4C). We found that gene expression levels of Il6, Cxcl1, and Cxcl2 were higher in Clec10a−/− MPs than in WT MPs, whereas the expression of these cytokines was comparable between both genotypes of DCs (Fig. 3H). These results suggest that Clec10a on MPs, rather than DCs, was involved in suppression of HDM-induced neutrophilic dermatitis. Because interleukin-17 (IL-17) causes neutrophilic dermatitis (15), we examined the involvement of IL-17 in exacerbated HDM-induced dermatitis in Clec10a−/− mice. However, the proportion of IL-17+ dermal γδ T cells, which are the major IL-17–secreting cells in the skin, were comparable between WT and Clec10a−/− skin (fig. S4, D and E), suggesting that IL-17 was not involved in the enhanced neutrophilic dermatitis in Clec10a−/− skin.

For in vitro analyses, we generated bone marrow–derived macrophages (BMMPs) from WT and Clec10a−/− C57BL/6J mice (fig. S4F). Because HDM contains group 2 allergens and endotoxin, both of which are Toll-like receptor 4 (TLR4) ligands (1618), we analyzed expression levels of TLR4 and MP markers. They were comparable between both genotypes of CD115+ BMMPs (fig. S4G). After HDM stimulation in vitro, Clec10a−/− BMMPs secreted larger amounts of neutrophil chemoattractants, such as IL-6, CXCL1, CCL3, and TNF-α (tumor necrosis factor-α), than did WT BMMPs (Fig. 3I). In addition, although colon MPs in Clec10a−/− mice were reported to produce less IL-10 compared with those in WT mice in a dextran sulfate sodium–induced colitis model (13), Il10 expressions in the skin and skin MHCII+ MPs were comparable between WT and Clec10a−/− mice after HDM application (fig. S4H). Similarly, BMMPs derived from Clec10ac.706T/T NC/Nga mice produced higher levels of cytokine production compared with Clec10ac.706T/C BMMPs (fig. S4I). Together, these results demonstrate that cell surface expression of Clec10a inhibits inflammatory cytokine production from dermal MPs and suppresses neutrophilic dermatitis after HDM application.

Clec10a inhibits TLR4-induced inflammatory cytokine secretions

Clec10a contains a single tyrosine (residue 3) in the cytoplasmic portion, which is a component of a putative hemi-immunoreceptor tyrosine-based activating motif (hemITAM; YxxL) (fig. S5A). Whereas BMMP transfectants expressing WT (C57BL/6J) Clec10a showed tyrosine phosphorylation of Clec10a after stimulation with HDM, tyrosine phosphorylation was not detected when Clec10a was mutated at residue 3 (Y3F), although Clec10a cell surface expression was unaffected (fig. S5, B to D). We found that tyrosine phosphorylation after HDM stimulation was substantially decreased when Clec10a was expressed on Tlr4−/− BMMPs (Fig. 4A and fig. S5E), indicating that HDM-induced tyrosine phosphorylation of Clec10a is dependent on TLR4. Moreover, TAK-242, a chemical compound that specifically inhibits TLR4 signaling by binding to the Toll/IL-1 receptor domain (fig. S5F) (19), decreased the HDM-induced production of inflammatory cytokines by both WT and Clec10a−/− BMMPs (Fig. 4B). Of note, TAK-242 decreased inflammatory cytokine production by Clec10a−/− BMMPs to a level comparable to that of WT BMMPs (Fig. 4B), suggesting that, whereas TLR4 mediates HDM-induced proinflammatory cytokine production, it is also involved in the Clec10a-associated suppression of cytokine production.

Fig. 4 Clec10a inhibits TLR4-induced inflammatory cytokine secretions.

(A) WT, Tlr4−/−, and Clec10a−/− BMMPs were stimulated with HDM for the indicated times, followed by immunoprecipitation (IP) of lysates with mAb to Clec10a and immunoblot (IB) analysis using Abs to phosphotyrosine (pTyr) or Clec10a. (B) CBA analysis of culture supernatants from WT and Clec10a−/− CD115+–enriched BMMPs pretreated with 0.5 μM TAK-242 and stimulated with HDM for 6 hours (n = 3). (C) WT and Clec10a−/− CD115+–enriched BMMPs were stimulated with HDM for the indicated times, followed by IB using mAb to phospho-Syk (pSyk; Y519/520) or Syk. (D and E) BMMP transfected with WT or Y3F Clec10a or with empty vector (EV) (D) or WT and Clec10a−/− BMMPs pretreated with or without Syk inhibitor IV (E) were stimulated with HDM, followed by IP of lysates with mAb to Clec10a and IB using Abs to Syk, SHP-1, or Clec10a. Arrowheads indicate the molecules of interest (black) or the heavy chain of IP-Ab (white). RQ, relative quantity. *P < 0.05 and **P < 0.01 (one-way ANOVA with post hoc test). Data show mean values ± SEM. Data are representative of three independent experiments.

Clec10a-associated suppression of HDM-induced cytokine production led us to ask whether the hemITAM sequence in Clec10a has an inhibitory, rather than activating, function. Stimulation with HDM led to activation of spleen tyrosine kinase (Syk) in WT (C57BL/6J) but not Clec10a−/− BMMPs (Fig. 4C and fig. S5G). Moreover, HDM stimulation induced the recruitment of Syk and Src homology region 2 domain-containing phosphatase-1 (SHP-1) to Clec10a, and this recruitment was dependent on the tyrosine residue of Clec10a (Fig. 4D and fig. S5H). Treatment of BMMPs with a Syk inhibitor suppressed HDM-induced SHP-1 recruitment to Clec10a (Fig. 4E and fig. S5I), suggesting that this event was dependent on Syk activation, consistent with regulatory signaling through the “inhibitory ITAM” (2022).

Clec10a recognizes a mucin-like molecule of HDM

Previous studies have shown that HDM components contained glycosylated protein antigens and polysaccharides (23, 24). To test whether Clec10a recognizes glycosylated components of HDM, we generated nuclear factor of activated T cells (NFAT)–GFP reporter cells (25) expressing the chimeric fusion protein comprising the extracellular and transmembrane portions of mouse Clec10a fused to the cytoplasmic portion of CD3ζ. The mouse Clec10a reporter cells expressed GFP in response to Lewis X, an oligosaccharide ligand of Clec10a (10), but not Lewis Y (fig. S6, A and B). We found that the mouse Clec10a reporter cells expressed GFP in response to plate-coated HDM in a dose-dependent manner (Fig. 5A). In addition, pretreatment of the reporter cells with anti-Clec10a monoclonal antibody (mAb) or galactose, but not glucose or mannose, inhibited GFP expression (Fig. 5, B and C). Moreover, treatment of HDM with galactosidase but not mannosidase is associated with a reduced ability to induce GFP expression in reporter cells (Fig. 5D). These results indicate that, through its galactose-binding site, Clec10a directly binds to a galactosylated component of HDM.

Fig. 5 Clec10a recognizes a mucin-like molecule of HDM.

(A to F and I) GFP expressions of mouse Clec10a-CD3ζ or parental reporter cells after stimulation. (A to D) Reporter cells were stimulated with plate-coated HDM of indicated concentrations (A); plate-coated HDM (20 μg/ml) in the presence of rat IgG2a or anti-Clec10a mAb (B); galactose (Gal), glucose (Glc), or mannose (Man) (C); or plate-coated HDM treated or not with either galactosidase (GALase) or mannosidase (MANase) (D). (E, F, and I) Reporter cells were stimulated with Clec10a-L in HDM pulled down (PD) with Clec10a-Fc or control human IgG (E), each size-based fraction of the Clec10a-L (F), or Clec10a-L treated with or without PNGase F or NaOH (I). Statistical analysis was performed by using the PBS-stimulated samples as control (F). (F to H) Clec10a-L was immunoblotted with Clec10a-Fc before (F to H) or after (H) treatment with PNGase F or NaOH or stained with silver with or without alcian blue (G). (J) Lectin microarray analysis of the Clec10a-L. Black bar indicate lectins, which bind to Galβ(1–3)GalNAc (T antigen). GalNAc, N-acetylgalactosamine. GlcNAc, N-acetylglucosamine. (K) Schematic of Clec10a-L in HDM. T, T antigen [Galβ(1–3)GalNAc]. Tn, Tn antigen (αGalNAc). LacNAc, N-acetyl-d-lactosamine [Galβ(1–4)GlcNAc]. *P < 0.05 and ***P < 0.001 (one-way ANOVA with post hoc test). Data show mean values ± SEM (n = 3). Data are representative of two (G to J) or three independent experiments.

To characterize the Clec10a ligand (Clec10a-L) in HDM, we generated the chimeric fusion protein comprising the extracellular portion of mouse Clec10a fused to the Fc portion of human IgG1 (Clec10a-Fc), which did bind to Lewis X but not to Lewis Y (fig. S6C). Using a pull-down assay from HDM with the Clec10a-Fc, we isolated the Clec10a-L with a molecular size of ~225 kDa, which was able to induce GFP expression in mouse Clec10a reporter cells (Fig. 5, E and F, and fig. S6D). The Clec10a-L was stained by Alcian blue and silver, but not by silver alone (Fig. 5G and fig. S6E), indicating that Clec10a-L is a negatively charged and heavily glycosylated protein (26). Treatment with NaOH, which dissociates O-linked glycans, abolished the Clec10a-Fc binding epitope (Fig. 5H and fig. S6F) and its activity to induce GFP expression in mouse Clec10a reporter cells (Fig. 5I). In contrast, peptide N-glycosidase F (PNGase F), which dissociates N-linked glycans, did not show any effect on the Clec10a-L (Fig. 5, H and I, and fig. S6F). These results suggest that the Clec10a recognizes an O-linked glycan of Clec10a-L. Furthermore, lectin microarray analysis demonstrated that Clec10a-L contains T antigen [Galβ(1–3)GalNAc] and Tn antigen (αGalNAc), which are the common mucin-type O-glycan core structures, with or without LacNAc epitope [Galβ(1–4)GlcNAc] that attaches to the mucin-type O-glycan (Fig. 5J and table S2). Notably, Clec10a-L most strongly bound to a lectin Maclura pomifera (MPA) that preferably recognizes the high densities of polyvalent T and Tn antigens (27). Conversely, glycan microarray analysis demonstrated that Clec10a-Fc bound to T and Tn antigens (table S3). Together, these results suggest that the Clec10a-L is a mucin-like molecule (Fig. 5K).

Human Asgr1 is a structural and functional counterpart of mouse Clec10a

The Basic Local Alignment Search Tool (BLAST; https://blast.ncbi.nlm.nih.gov/Blast.cgi) revealed a highest similarity in the amino acid sequences of CTLD between human asialoglycoprotein receptor 1 (Asgr1) and mouse Clec10a, encoded by ASGR1 (gene ID: 432) and CLEC10A (gene ID: 10462), respectively. Although both human Asgr1 and Clec10a reporter cells expressed GFP in response to HDM stimulation (Fig. 6A and fig. S7A), addition of galactose inhibited HDM-induced GFP expression in only Asgr1 reporter cells (Fig. 6B and fig. S7B). Furthermore, the hemITAM sequence (YxxL) was observed only in human Asgr1 (Fig. 6C). These results suggest that Asgr1 in humans is a structural and functional counterpart of Clec10a in mice. Clec10a-L in HDM induced the GFP expression in human Asgr1 reporter cells (Fig. 6D). Similar to Clec10a in the mouse skin, immunohistochemical analyses showed that Asgr1 was expressed in both MP (CD68+) and non-MP (CD68) cells in human skin (Fig. 6E). Asgr1 mRNA was reported to have two splicing variants, which encode a membrane and a soluble type that lacks a transmembrane portion (28). We found that Asgr1 also seemed to be detected at the non–cell-associated region, which might be derived from the soluble type. Knockdown of ASGR1 in human monocyte-derived cultured MPs by small interfering RNA (siRNA) enhanced inflammatory cytokine secretion in response to HDM (Fig. 6F and fig. S7C), suggesting that Asgr1 regulates HDM-induced dermatitis in humans. Together, these results suggest that C-type lectin receptors such as Asgr1 in humans and Clec10a in mice recognize a mucin-like molecule in HDM and play an important role in skin homeostasis against inflammation induced by HDM invasion.

Fig. 6 Human Asgr1 is a structural and functional counterpart of mouse Clec10a.

(A and B) GFP expression of human Asgr1-CD3ζ reporter cells after stimulation with plate-coated HDM in the absence (A) or presence (B) of galactose, glucose, or mannose. (C) Amino acid sequences of the intracellular portion of mouse Clec10a, human Asgr1 (hAsgr1), and human Clec10a. The putative hemITAM sequence is underscored. (D) GFP expression of human Asgr1-CD3ζ reporter cells stimulated with plate-coated Clec10a-L pulled down from HDM with mouse Clec10a-Fc. (E) Fluorescence microscopy of tissue sections from the human skin, stained with an anti-Asgr1 pAb, anti-CD68 mAb, and the DNA binding dye DAPI. Scale bars, 100 μm. (F) CBA analysis of culture supernatants from human CD14+ monocyte-derived MPs treated with control siRNA or siRNA specific for ASGR1 and stimulated with HDM for 6 hours. *P < 0.05, **P < 0.01, and ***P < 0.001 [one-way ANOVA with post hoc test (B) or unpaired two-tailed Student’s t test (F)]. Data show mean values ± SEM (n = 3). Data are representative of two independent experiments.

HDM-derived Clec10a-L ameliorates LPS-induced dermatitis

To examine the functional role of the HDM component Clec10a-L in dermatitis induced by TLR4-mediated signaling, we topically applied Clec10a-L purified from HDM together with an Escherichia coli–derived lipopolysaccharide (LPS) as a TLR4 agonist on the dorsal skin of WT and Clec10a−/− mice on the C57BL/6J background. We found that treatment with the Clec10a-L decreased epidermal thickness and neutrophil infiltration in WT mice but not in Clec10a−/− mice (Fig. 7, A to C), indicating that the Clec10a-L ameliorated LPS-induced dermatitis through Clec10a. Of note, treatment with Clec10a-L, together with LPS, increased neutrophil recruitment, but not epidermal thickness, compared with LPS alone in Clec10a−/− mice (Fig. 7, B and C), suggesting that Clec10a-L might contain components other than a Clec10a-binding molecule that mediated Clec10a-independent signals for neutrophil recruitment. Nonetheless, because HDM contains TLR4 ligands as well, these results support a scenario in which an HDM component mucin-like molecule binds to Clec10a and suppresses pro-inflammatory cytokine production induced by HDM-derived TLR4 ligand (Fig. 7D).

Fig. 7 HDM-derived Clec10a-L ameliorates LPS-induced dermatitis.

(A and B) LPS with or without Clec10a-L was applied every day to the dorsal skin of C57BL/6J-WT and Clec10a−/− mice. Histology (H&E) and epidermal thickness on day 5 were compared between WT and Clec10a−/− mice. Scale bars, 100 μm. (C) Neutrophil (CD11b+Ly-6G+) numbers in the skin (1 cm2) of WT and Clec10a−/− mice 6 hours after application of LPS with or without Clec10a-L. (D) A hypothetical model of the role of Clec10a in skin homeostasis upon HDM exposure. *P < 0.05 and **P < 0.01 (one-way ANOVA with post hoc test). Data show mean values ± SEM. Data are representative of two independent experiments.

DISCUSSION

In this study, we found that the stop-gain mutation of Clec10a (c.706C>T, p.Q236X) in NC/Nga mice impairs the transport of Clec10a to the cell surface. These results suggest that the C-terminal region of CTLD (Q236 to S304) of Clec10a plays a critical role in the cell surface expression, consistent with a previous report that a shorter C-tail length of type II transmembrane proteins impairs their insertion into the endoplasmic reticulum membrane and cell surface localization (29). How the C-terminal region of Clec10a is involved in the cell surface expression is unclear. Our finding shows that the Clec10a mutation is responsible for HDM hypersensitivity and skin inflammation in NC/Nga mice. Of note, although NC/Nga mice also showed increased serum levels of IgE, Clec10a was not involved in IgE antibody response after HDM application in NC/Nga mice and C57BL/6J mice, consistent with previous reports that dermatitis and hyper-IgE production in NC/Nga mice are regulated differently (30).

We demonstrated that deficient cell surface expression of Clec10a on skin MPs enhanced inflammatory cytokine production, resulting in increased neutrophil recruitment in the skin. Depletion of neutrophils ameliorated the HDM-induced skin inflammation, indicating that neutrophils play a central role in the development of dermatitis in this model. These results are consistent with previous studies that neutrophils are involved in the pathophysiology of AD in certain situations of patients with severe AD (3133).

We showed that Clec10a suppressed inflammatory cytokine production induced by TLR4 signaling upon HDM stimulation. These results suggest that HDM contains both ligands for Clec10a and TLR4. It may be possible that Clec10a-L forms a complex with a TLR4 ligand in HDM and promotes the proximity between Clec10a and TLR4 for efficient signaling cross-talk in MPs. On the basis of the results in the current study, we propose a model for the regulatory mechanisms of HDM-stimulated MP activation by Clec10a and TLR4. Namely, although HDM stimulation activates both Clec10a and TLR4, TLR4 signaling is required for tyrosine phosphorylation of Clec10a and recruitment of Syk, which then induced recruitment and activation of SHP-1. SHP-1 in turn inhibited TLR4 signaling for inflammatory cytokine production. Therefore, the hemITAM of the cytoplasmic region of Clec10a functions as an inhibitory ITAM (2022). Previous reports demonstrated that the engagement of receptors by low-affinity or low-avidity ligands may induce transient recruitment of Syk to ITAM and transduce inhibitory signals through SHP-1 (21, 22), suggesting that Clec10a on skin MPs binds to the ligand in HDM with low affinity and/or avidity.

Several C-type lectin receptors such as Dectin-1, Dectin-2, mannose receptor (CD206), and DC-SIGN expressed on DCs in the lung bind to HDM components and are involved in allergic asthma (23, 3436). These receptors activate DCs to exacerbate allergic airway inflammation. In contrast, Dectin-1 expressed on epithelial cells recognizes tropomyosin derived from arthropods, including HDM, and suppresses IL-33 production (37). However, the carbohydrate moiety of HDM components recognized by these C-type lectin receptors remains undetermined. In the current study, we showed that Clec10a recognizes a mucin-like molecule in HDM and inhibits, rather than activates, MPs in the skin, resulting in suppression of the development of HDM-induced dermatitis.

We showed that the Clec10a-L in HDM is a heavily glycosylated mucin-like molecule and has a suppressive function against LPS-induced dermatitis. Recent evidences demonstrated that the mucin-type O-glycans show various biological properties. They regulate immune responses through the structure with various terminal sugars that mediate interactions with cell surface receptors and other molecules (38). For example, the mucin-like O-glycans CA125 and TAG-72 expressed on tumors bind to C-type lectin receptors on tumor-associated MPs (TAMs) and modulate cytokine productions by TAM toward an immune-suppressive profile (39) or inhibit natural killer (NK) cell–mediated cytotoxicity against tumor cells (40). These observations suggest that the mucin-like O-glycans expressed in tumors contribute to the immune escape. The surface of many protozoan parasites, including Trypanosoma cruzi, is covered with mucins (41), which interact with Siglec-E through the sialic acid of the T. cruzi mucins and contribute to immune evasion of T. cruzi by suppression of immune activation (41).

We demonstrated that the human Clec10a binds to HDM through a carbohydrate binding specificity different from mouse Clec10a, consistent with a previous report (10, 42). Instead, we showed that human Asgr1 and mouse Clec10a bound to HDM through its galactose-binding site. Given the expression profile and molecular and functional characteristics of human Asgr1, we concluded that Asgr1, rather than Clec10a, is a human counterpart of mouse Clec10a. Although human Asgr1 is known to be expressed on hepatocytes and myeloid cells (4345) and associated with the development of coronary heart disease and liver cirrhosis and cancers (45, 46), the function of Asgr1 remains entirely obscure. We showed that not only mouse Clec10a but also human Asgr1 bound to the same mucin-like O-glycan in HDM and suppressed inflammatory cytokine secretions of MPs after HDM stimulation. Together, these results suggest that both mouse Clec10a and human Asgr1 may play an important role in the skin homeostasis during epicutaneous exposure to HDM. Whether Asgr1 is involved in the human AD pathology is an important question. Decreased expression of Asgr1 might lead to the development of HDM-induced skin inflammation in AD. We demonstrated that Clec10a is involved in TLR4 ligand-induced neutrophilic skin inflammation, rather than TH2 responses, in mice. Therefore, down-regulation of Asgr1 expression might also be involved in the development of dermatitis other than AD. In these patients as well as in patients with AD, itching and scratching impair the skin barrier function and increase the permeability of TLR4 ligands of HDM or Gram-negative bacteria, further exacerbating the skin inflammation. However, genome-wide association study showed no single-nucleotide polymorphism of ASGR1 that is associated with AD. The regulatory mechanism of Asgr1 expression remains undetermined and is an interesting issue to be clarified; this would facilitate the study of skin inflammation in many types of dermatitis.

In the current study, we showed that a mucin-like molecule contained in HDM is a ligand for Clec10a in mice and Asgr1 in humans and suppresses TLR4 ligand-induced inflammatory cytokine production from MPs. However, there are still several questions to be clarified: (i) whether the ligand is indeed derived from HDM or other HDM-associated organisms, (ii) whether it is a glycoprotein or a glycosylated molecule without a core protein, (iii) what are its molecular and structural characteristics, and (iv) whether mucins derived from other organisms or cells have a similar property to the ligand in HDM. Future studies are required to address these important issues.

MATERIALS AND METHODS

Study design

The objective of this study was to understand the regulatory mechanism of immune responses induced by HDM. For this objective, we analyzed the genome of NC/Nga mouse that shows hypersensitivity to HDM and found a causal mutation in Clec10a. Therefore, we examined the function of mouse Clec10a and its human homolog, Asgr1, in response to the stimulation with their ligand contained in HDM in vivo and in vitro in the context of HDM-induced dermatitis by using genetically engineered mice and human samples, as described in this section.

Mice

NC/Nga mice were purchased from Charles River Laboratories Japan (Yokohama, Japan). NC/Nga-Clec10a706T>C mice were generated as described in Supplementary Materials and Methods, and NC/Nga-Clec10a706T/T (WT) littermate controls were used for experiments. Clec10a−/− mice in C57BL/6J background were generated, as previously described (13, 47). Clec10a−/− mice and control WT mice were bred under SPF conditions in the same room of the animal facility of the University of Tsukuba. Tlr4−/− mice on a C57BL/6J background were purchased from Oriental Bioservice (Kyoto, Japan). All animal experiments in this study were performed humanely after receipt of approval from the Animal Ethics Committee of the Laboratory Animal Resource Center, University of Tsukuba (no. 18-266) and in accordance with the “Fundamental Guideline for Proper Conduct of Animal Experiment and Related Activities in Academic Research Institutions” under the Jurisdiction of the Ministry of Education, Culture, Sports, Science, and Technology.

Human samples

Peripheral blood mononuclear cells were isolated from the blood of healthy volunteers. Skin tissues were obtained from the healthy area of skin biopsy specimen from patients with skin tumor at the University of Tsukuba Hospital, Japan. Written informed consent was obtained from the patients and healthy volunteers. This study was approved by the ethical review boards of the University of Tsukuba (no. 234-1, H30-127).

Antibodies

The antibodies used in this study are available in the Supplementary Materials.

Exome sequencing and identification of Clec10a mutation

DNA was extracted from the blood of NC/Nga mice by using the QIAamp DNA Blood Mini Kit (QIAGEN, Venlo, The Netherlands) with appropriate optimization for mouse DNA extraction. Exome sequencing was performed according to the protocol described in the SureSelect Library Prep Kit (post-pool version 4, Agilent Technologies, Santa Clara, CA) and SureSelect Mouse All Exon Kit (Agilent Technologies). The DNA library underwent emulsion polymerase chain reaction (PCR) (SOLiD EZ Bead Emulsifier kit, Thermo Fisher Scientific, Waltham, MA) to generate clonal DNA fragments on beads, followed by bead enrichment (SOLiD EZ Bead Enrichment Kit, Thermo Fisher Scientific). Enriched template beads were sequenced on a SOLiD 5500xl sequencer as single-end, 60–base pair (bp) reads (Thermo Fisher Scientific). The SOLiD 5500xl output reads were aligned against the mouse genome reference sequence (NCBI37/mm9) using LifeScope version 2.5.1 (Life Technologies) to generate BAM files. Variant calling was performed according to the protocol described in the Genome Analysis Toolkit (GATK), Picard (http://broadinstitute.github.io/picard), and SAMtools, and only reads that mapped to a unique position in the reference genome were used. Variants were annotated by using ANNOVAR software (48). Genetic variations in inbred mouse strains other than NC/Nga were obtained from Release REL-1211, which is numbered according to NCBIm37 assembly (www.sanger.ac.uk/science/data/mouse-genomes-project), and the Mouse Genome Informatics website (www.informatics.jax.org/). Sanger sequencing was performed using the BigDye Terminator v1.1 Cycle Sequencing Kit (Thermo Fisher Scientific) on an ABI 3130xl Genetic Analyzer (Thermo Fisher Scientific).

Comparison against the C57BL/6J mouse genome revealed 70,772 variants in the exons of the NC/Nga genome. The variants with low-quality values generated by GATK output were filtered out, resulting in a new total of 64,518 variants. The loss-of-function variants including stop-gain and frameshift mutations were selected using ANNOVAR software. The 34,816 exonic or splicing variants contained a total of 35 stop-gain mutations, and 48 frameshift mutations that introduced stop codons were identified. Each of these 83 genes was manually inspected, and 46 variants in 43 genes were selected for further Sanger sequence validation. The lists of the primers for Sanger sequence validations are shown in table S1.

Among the 46 variants identified, 24 were confirmed by Sanger sequencing (table S1). To distinguish potentially pathogenic variants from other variants, we filtered out variants that existed in 17 inbred strains of laboratory mice and two strains of Japanese fancy mice and thus focused on the remaining seven variants in seven genes (table S1, shown in bold). Last, we chose Clec10a from among the seven genes in light of its expression in hematopoietic cells, on the basis of information in the BioGPS database (fig. S1A) (9).

Production of NC/Nga-Clec10a706c.T>C mice by using CRISPR-Cas9 technology

The pX330 plasmid (catalog no. 42230, Cambridge, MA) was used as the CRISPR expression vector. To cleave 1341 bp upstream and 682 bp downstream of c. T706, we individually inserted a 20-nucleotide sequence (5′-GGATACTGGTGAAGACACGG-3′ and 5′-AGAGGATAAATGTTAGATTG-3′) into pX330 at each of these sites, thus deriving the plasmids pX330-up-1314 and pX330-down-682, respectively. In addition, we constructed the donor plasmid DNA (p706-donor) to induce the single-base substitution (c.706 T>C in NC/Nga-Clec10a). The p706-donor plasmid carried 5′ (1255 bp), central (2024 bp), and 3′ (1364 bp) homology arms such that the c.706C sequence was in the central homology arm and the restriction enzyme recognition sequences (Xho I and Nhe I) were at each border of the homology arms.

We injected pregnant mare serum gonadotropin (5 U) into each of the 20 female NC/Nga mice (10 weeks of age), injected them 48 hours later with human chorionic gonadotropin (5 U), and collected 756 unfertilized oocytes from their oviducts. In vitro fertilization of these oocytes was performed by using NC/Nga sperm, and 551 pronuclear stage embryos were obtained. Consequently, the DNA mixture of pX330-up-1314 (5 ng/μl), pX330-down-682 (5 ng/μl), and p706-donor (10 ng/μl) was injected into their pronucleus; the 276 two-cell embryos that survived were then transferred into pseudopregnant ICR mice, and 28 pups were obtained. A single pup carried the target point mutation without pX330 random integration.

Skin and peritoneal cavity cell preparations

To isolate skin cells from naïve and HDM-applied mice, shaved dorsal skin samples were minced by using scissors and incubated for 60 min in collagenase II (200 U/ml; Worthington Biochemical, Lakewood, NJ) in RPMI 1640 medium containing deoxyribonuclease I (50 U/ml; Worthington Biochemical) and 10% fetal bovine serum (FBS). Additional dissociation and homogenization were performed by using a gentleMACS Dissociator (Miltenyi Biotec, Bergisch Gladbach, Germany). Cell preparations were filtered through a 55-μm nylon mesh to obtain single-cell suspensions. Peritoneal cavity cells were harvested from naïve mice and then filtered through a 55-μm nylon mesh to obtain single-cell suspensions.

Flow cytometry

Flow cytometric analyses and cell sorting were performed by using FACS LSRFortessa and FACS Aria flow cytometers (BD Biosciences), respectively. FlowJo software (Tree Star, Ashland, OR) was used for data analyses. Dead cells were stained and excluded using propidium iodide solution (catalog no. P4864, Sigma-Aldrich).

Histology and immunohistochemistry

For histologic analysis, dorsal skin samples from mice were fixed in formalin, embedded in paraffin, and sectioned at 4 μm. Fixed sections were stained with hematoxylin and eosin (H&E) and analyzed by light microscopy. Epidermal thickness was measured at five locations per field and in three or five fields per mouse.

For immunohistochemistry, skin tissues from mice were embedded in Tissue-Tek Optimal Cutting Temperature Compound (Sakura Finetek Japan, Tokyo, Japan) and stored at −80°C. For immunohistochemistry, 4-μm sections were used. The sections were washed and rehydrated in phosphate-buffered saline (PBS) containing 0.05% Tween 20 (PBS-T; pH 7.4) and incubated in Blocking One Histo (Nacalai, Kyoto, Japan) at room temperature for 10 min; sections were then washed with PBS-T, incubated overnight at 4°C with anti-Clec10a mAb, washed with PBS-T, and then incubated for 1 hour with Alexa Fluor 546–conjugated anti-rat IgG polyclonal antibody (pAb). After thorough washing with PBS-T, sections were counterstained with the nuclear counterstain 4,6-diamidino-2-phenylindole (DAPI) (VectaShield, Vector Laboratories).

Human skin tissues were resected at the marginal healthy area of biopsy specimen from patients with skin tumor, fixed in formalin, embedded in paraffin, and sectioned at 4 μm. The sections were deparaffinized with xylene, rehydrated with ethanol, and blocked for endogenous peroxidase with methanol; sections were then stained with anti-CD68 and anti-Asgr1 antibodies (Abs) according to the manufacturer’s instructions of the Opal 4-Color Automation IHC Kit (catalog no. NEL820001KT, PerkinElmer, Waltham, MA). Briefly, the sections were incubated for 15 min at 95°C, washed with tris-buffered saline (TBS) containing 0.05% Tween 20 (TBS-T) (catalog no. T9142, Takara Bio, Shiga, Japan), and treated with the blocking solution for 10 min at room temperature. The sections were then incubated with anti-CD68 Ab or mouse IgG1 overnight at 4°C, washed with TBS-T, and incubated with the Opalpolymer horseradish peroxidase (HRP) for 30 min at room temperature in a humidified chamber. After washing with TBS-T, the sections were incubated with Opal Fluorophore Working Solution for 10 min at room temperature in a humidified chamber and washed with TBS-T. Abs for first staining were removed from the sections by heating at 95°C, and the sections were subsequently stained with anti-Asgr1 Ab or rabbit IgG in the same manner as Abs for first staining. Last, the sections were counterstained with spectral DAPI solution.

Dermatitis

At the first induction (day 0), the back skin of anesthetized mice was clipped by using electric clippers, and residual hair was depilated by using a hair removal cream. For HDM-induced dermatitis, antigen challenge was performed by using topical application of 100 mg of HDM (Dermatophagoides farinae) ointment (Biostir, Kobe, Japan) on the dorsal shaved skin. From the second induction downward, barrier disruption was achieved by applying 150 μl of 4% sodium dodecyl sulfate to the dorsal skin at 2 hours before HDM ointment application. These procedures were repeated twice each week. Each of several factors—erythema/hemorrhage and scarring/dryness—was scored on days 3, 6, 14, and 21 as 0 (none), 1 (mild), 2 (moderate), or 3 (severe) according to the manufacturer’s instructions (Biostir). The sum of these individual scores was taken as the overall dermatitis score. For depletion of neutrophils, 400 μg of control rat IgG (catalog no. BE0094, Bio X cell, West Lebanon, NH) or anti–Ly-6G Ab (1A8) (catalog no. BP0075, Bio X cell) was injected intraperitoneally 1 day before HDM application.

For LPS-induced dermatitis, antigen challenge was performed after tape stripping on the dorsal shaved skin by topical application of 50 μg of LPS (from E. coli O111:B4, catalog no. L2630, Sigma-Aldrich), which mimics a TLR4 agonist contained in HDM (17, 18), together with or without Clec10a-L pulled down from 170 μg of HDM by using the chimeric fusion protein consisting the extracellular portion of mouse Clec10a fused with the Fc portion of human IgG (Clec10a-Fc). Tape stripping and antigen challenge were repeated every day. The neutrophil recruitment was analyzed 6 hours after the first antigen challenge, and histological analysis was performed using the skin on day 5.

TEWL test

TEWL was measured on the dorsal skin of WT and Clec10a−/− mice by Tewameter TM 300 (Integral, Tokyo, Japan) after hair removal by using electric clippers and hair removal cream. Measurements were performed in triplicate for each mouse.

Establishment of RAW264.7 transfectants

Clec10a cDNAs obtained from C57BL/6J or NC/Nga mice and tagged with the Flag-encoding sequence were subcloned into the pMXs-IRES-GFP retroviral vector (49, 50). RAW264.7 transfectants that stably expressed either C57BL/6J-type or NC/Nga-type Clec10a were established as previously described (51).

Enzyme-linked immunosorbent assay

Serum IgE was measured using a capture antibody to mouse IgE (R35-72) and biotinylated anti-mouse IgE (R35-118), followed by HRP-conjugated streptavidin (catalog no. RPN1231V, GE Healthcare, Chicago, IL). Purified mouse IgE (C38-2, BD Biosciences) was used for standard. Serum IgG1 was measured using a capture antibody to mouse IgG1 (A85-3) and HRP-conjugated antibody to mouse IgG1. Purified mouse IgG1 (107.3, BD Biosciences) was used for standard. Serum IgG2c was detected by Mouse ELISA Quantitation Set (catalog no. E90-136, Bethyl, Montgomery, TX).

Intracellular IL-17A staining

Skin cells of naïve or HDM-applied mice were isolated from the dorsal skin and incubated in complete RPMI 1640 containing 10% FBS and brefeldin A (10 μg/ml; Sigma-Aldrich) for 3 hours.

Cells were then harvested, resuspended, stained for intracellular cytokines (FIX & PERM Cell Permeabilization Kit, catalog no. GAS004, Thermo Fisher Scientific) according to the manufacturer’s instructions, and analyzed by flow cytometry.

Generation of BMMPs

Bone marrow cells were cultured in a culture dish [product no. 430166 (6 cm diameter) or 430167 (10 cm diameter), Corning, Corning, NY] in complete RPMI 1640 containing 10% FBS in the presence of granulocyte-macrophage colony-stimulating factor (GM-CSF) (10 ng/ml; R&D Systems) and IL-4 (7 ng/ml; Wako, Osaka, Japan). On day 2, 70% of the nonadherent cells were removed, and fresh medium with GM-CSF and IL-4 was added. On day 5, 100% of the nonadherent cells were removed by washing with PBS, and fresh medium with GM-CSF was added. On day 7, all nonadherent cells were removed by washing with PBS, and adherent cells were used in the following experiments. For analyzing cytokine secretion and Syk phosphorylation, CD115+ BMMPs were enriched by using anti-CD115 Ab (BioLegend) and anti-rat IgG microbeads (Miltenyi Biotec).

Cytokine secretion assay

CD115+-enriched BMMPs were stimulated with HDM (100 μg/ml) (D. farinae) extract (Cosmo Bio, Tokyo, Japan), LPS (1 ng/ml), or Pam2CSK4 (200 pg/ml) in the presence or absence of 0.5 μM TAK-242 (TLR4 inhibitor, Merck) for 15 min. After 6 hours of stimulation, culture supernatants were collected, and the concentration of each cytokine was determined by using cytometric bead array analysis (BD Biosciences).

cDNA synthesis and real-time PCR

Total RNA was extracted by using Isogen reagent (Nippon Gene, Tokyo, Japan) from skin tissues or sorted cells by flow cytometry: skin MPs (CD45+MHCII+CD3CD19NK1.1Ly-6GEpCAMCD64+), DCs (CD45+MHCII+CD3CD19NK1.1Ly-6GEpCAMCD64), or CD3+CD4+ cells in the skin-draining lymph nodes. cDNA was synthesized by using the High Capacity RNA-to-cDNA Kit (Applied Biosystems, Carlsbad, CA). Expression of Clec10a and inflammatory cytokine genes was measured through quantitative real-time PCR (RT-PCR) analysis, performed by using SYBR Green Master Mix (Applied Biosystems) and specific primers. The Gapdh expression level was used as an internal control to normalize data. Primer sequences for the target genes are available in the Supplementary Materials.

Retroviral gene transduction

WT Clec10a cDNA was subcloned into the pMXs-puro retroviral vector (Cell Biolabs, San Diego, CA). To generate a site-specific Clec10a mutant, a sense PCR primer was designed that contained the codon for phenylalanine (TTC) instead of tyrosine (TAC; Y3F construct). WT and mutant cDNAs were verified by sequencing. Retroviruses were produced by transfecting 293GP packaging cells with the generated WT or mutant Y3F cDNA or the empty G glycoprotein of the vesicular stomatitis virus (VSV-G) expression vector, pCMV-VSV-G (52). BMMPs were infected with retroviral supernatants supplemented with polybrene (8 μg/ml; Sigma-Aldrich) on days 2 and 4, respectively. The plate was centrifuged at 1100g at 32°C for 2 hours, after which the virus-containing supernatant was removed and replaced with fresh BMMP culture medium. On day 5, medium was replaced with fresh BMMP culture medium. On day 7, nonadherent cells were removed by washing with PBS, and adherent cells were used for experiments.

Preparation of Clec10a-Fc chimera

The mouse Clec10a-Fc chimeric (Clec10a-Fc) construct was generated by cloning the extracellular portion of mouse Clec10a into the pME18S expression vector containing the Fc portion of human IgG1. The Clec10a-Fc constructs were transfected with Lipofectamine 2000 (catalog no. 11668019, Thermo Fisher Scientific) into HEK293T cells cultured in Opti-MEM (catalog no. 31985070, Thermo Fisher Scientific). Transfection medium was replaced with GIT medium (catalog no. 16041000, Kohjin-Bio, Saitama, Japan), and Clec10a-Fc protein was then purified from culture supernatant by using protein A agarose (catalog no. 1536153, Bio-Rad Laboratories, Hercules, CA).

Isolation of the Clec10a-L

HDM extract dissolved in the buffer containing 150 mM NaCl, 50 mM tris base, 1 mM CaCl2, and 0.01% Tween 20 was subjected to a pull-down assay by using a fusion protein composed of the extracellular portion of mouse Clec10a fused with the Fc portion of human IgG1 at the N terminus (Clec10a-Fc), which was coupled to Dynabeads Protein G (catalog no. 10009D, Thermo Fisher Scientific), followed by elution with 30 mM EDTA or 200 mM galactose. Effluent was dialyzed with PBS using a centrifugal filter unit (catalog no. UFC503024, Merck) and defined as the Clec10a-L.

Alcian blue and silver staining

The Clec10a-L was resolved by SDS–polyacrylamide gel electrophoresis (SDS-PAGE). Immediately after electrophoresis, the gels were washed with deionized water and 10% acetic acid buffer (10% acetic acid and 30% ethanol in deionized water) and stained or not with Alcian Blue solution (pH 2.5) (catalog no. 013-13801, Wako) at room temperature for 2 hours, followed by destaining with 3% acetic acid buffer (3% acetic acid in deionized water) and 10% acetic acid buffer. The gels were then stained with silver according to the manufacturer’s instructions (Pierce Silver Stain Kit, catalog no. 24612, Thermo Fisher Scientific).

Fractionation of Clec10a-L

The Clec10a-L was resolved by SDS-PAGE, and the gel was cut off and separated into fractions according to their size. Separated gels were mechanically crushed and incubated in PBS using a stirrer (catalog nos. 300-95441 and 382-01271, Nippi, Tokyo, Japan) overnight. Supernatants were collected after centrifugal separation at 17,400g for 10 min and dialyzed with PBS using a centrifugal filter unit (catalog no. UFC503024, Merck).

Immunoblotting

To analyze Syk phosphorylation, BMMPs were stimulated with HDM extract (100 μg/ml) for 0, 10, or 30 min at 37°C; lysed with 1% NP-40 lysis buffer; resolved by SDS-PAGE; transferred onto polyvinylidene difluoride (PVDF) membranes by electroblotting; and immunoblotted with anti–phospho-Syk and anti-Syk Ab followed by HRP-conjugated anti-rabbit IgG Ab.

To analyze the tyrosine phosphorylation of Clec10a, BMMPs were stimulated with HDM (100 μg/ml) for 0, 3, or 10 min at 37°C; lysed with 1% NP-40 lysis buffer; and immunoprecipitated with anti-Clec10a mAb. Immunoprecipitates were transferred onto PVDF membranes as described earlier and immunoblotted with HRP-conjugated anti-pTyr Ab and anti-Clec10a pAb, followed by HRP-conjugated anti-goat IgG Ab.

To analyze the associations between Clec10a and Syk or SHP-1, BMMPs were preincubated or not with Syk inhibitor IV (catalog no. 574714, Merck) (5 mM) for 30 min at 37°C; stimulated with HDM (100 μg/ml) for 0, 3, 5, or 10 min at 37°C; lysed with 0.2% digitonin buffer; and immunoprecipitated with anti-Clec10a mAb. Immunoprecipitates were transferred onto PVDF membranes as described earlier and immunoblotted with anti-Syk, anti–SHP-1, or anti-Clec10a Ab, followed by HRP-conjugated anti-rabbit IgG or anti-goat IgG Ab. All proteins were detected by using enhanced chemiluminescence (Thermo Fisher Scientific).

The Clec10a-L was pretreated or not with PNGase F (PRIM, catalog no. NZS1, N-Zyme Szientifics, Doylestown, PA) at 37°C for 16 hours or with 0.05 M NaOH at 40°C for 16 hours after heat denaturation at 95°C for 5 min, then resolved by SDS-PAGE, transferred onto PVDF membranes by electroblotting, and immunoblotted with biotinylated Clec10a-Fc followed by HRP-conjugated streptavidin.

Establishment and stimulation of reporter cells

The intracellular portion of human CD3ζ was obtained from a vector provided by L.L. Lanier (University of California, San Francisco), and the transmembrane through extracellular portion of mouse or human Clec10a, or human Asgr1 cDNA was subcloned into the pMXs-puro retroviral vector (Cell Biolabs, San Diego, CA). 2B4-NFAT-GFP reporter cells were provided by H. Arase (University of Osaka, Japan) (25). 2B4-NFAT-GFP reporter transfectants stably expressing mouse or human Clec10a or human Asgr1 were established as previously described (51). Reporter cells were incubated for 18 hours, with or without anti-Clec10a mAb, Lewis X (GlycoTech, Gaithersburg, MD), Lewis Y (GlycoTech), galactose (Sigma-Aldrich), glucose (Sigma-Aldrich), or mannose (Sigma-Aldrich), on plate-coated with HDM extract treated or not treated with galactosidase (catalog nos. 5704-GH and 5549-GH, R&D Systems) or mannosidase (catalog no. P0768S, New England BioLabs) or plate-coated Clec10a-L or size-fractionated Clec10a-L. The activation of NFAT-GFP was monitored by using flow cytometry.

Lectin microarray analysis

The lectin microarray was produced using a noncontact microarray printing robot (MicroSys4000, Genomic Solutions) as described previously with minor modification (53). Samples were fluorescently labeled with Cy3 Mono-Reactive Dye (GE), and excess Cy3 was removed with Sephadex G-25 desalting columns (GE). After dilution at 10 times with probing solution [25 mM tris-HCl (pH 7.5) and 140 mM NaCl (TBS) containing 2.7 mM KCl, 1 mM CaCl2, 1 mM MnCl2, and 1% Triton X-100], the Cy3-labeled samples were applied to the lectin microarray and incubated at 20°C overnight. After washing with probing solution, fluorescence images were acquired using an evanescent field–activated fluorescence scanner (Bio-Rex scan 200, Rexxam Co. Ltd.). The lectin signals of triplicate spots were averaged for each protein sample and normalized relative to the mean value of 96 lectins. The lists of the lectins are shown in table S5.

Glycoconjugate microarray analysis

Glycoconjugate microarray containing 98 glycoconjugates (table S6) was produced using MicroSys4000 (Genomic Solutions) as previously described with minor modification (54). Clec10a-Fc chimera (10 μg/ml) precomplexed with Cy3-conjugated goat anti-human IgG (1 μg/ml) and Fc (catalog no. 109-165-098, Jackson ImmunoResearch) in TBST (20 mM tris-HCl buffer, 0.15 M NaCl, and 1% Triton X-100) containing either 10 mM CaCl2 or 10 mM EDTA was incubated with the glycoconjugate microarray (80 μl per well) at 20°C overnight. After washing, fluorescent images were immediately acquired using Bio-Rex scan 200.

Knockdown of human ASGR1

CD14+ monocytes were enriched from the peripheral blood mononuclear cells by using anti-CD14 microbeads (catalog no. 130-050-201, Miltenyi Biotec), cultured with GM-CSF (10 ng/ml) for 2 days, transferred and cultured with Accell siRNA-delivery media (catalog no. B-005000, Dharmacon) containing GM-CSF (10 ng/ml) and 1 μM siRNA specific for ASGR1 (catalog no. E-013268, SMARTpool Accell siRNA, Dharmacon, Lafayette, CO) or control siRNA for 3 days, and then stimulated with HDM extract (100 μg/ml) for 6 hours. The concentration of each cytokine in the culture supernatants was determined by using cytometric bead array analysis (BD Biosciences).

Statistical analysis

Statistical analyses were performed by using one-way analysis of variance (ANOVA) with post hoc test or two-tailed Student’s t test (GraphPad Prism 5, GraphPad Software, La Jolla, CA) as specified in the figure legends.

SUPPLEMENTARY MATERIALS

immunology.sciencemag.org/cgi/content/full/4/42/eaax6908/DC1

Materials and Methods

Fig. S1. Characterization of Clec10a in C57BL/6J and NC/Nga mice.

Fig. S2. Phenotypes of NC/Nga-Clec10ac.706T>C mice.

Fig. S3. Phenotypes of HDM-induced dermatitis in Clec10a-deficient mice.

Fig. S4. Characterization of WT and Clec10a−/− MPs.

Fig. S5. Generation of BMMPs transfected with WT or mutated (Y3F) Clec10a.

Fig. S6. Generation of Clec10a-CD3ζ reporter cells and Clec10a-Fc chimera.

Fig. S7. HDM stimulation of human Clec10a-CD3ζ reporter cells and knockdown efficiency of human ASGR1.

Table S1. The stop-gain and frameshift mutations in NC/Nga mouse confirmed by Sanger sequence.

Table S2. Lectin microarray analysis of Clec10a-L in HDM.

Table S3. Glycan microarray analysis of Clec10a-Fc.

Table S4. The primers for Sanger sequencing validation of stop-gain and frameshift mutations in the NC/Nga genome.

Table S5. Lectins used for lectin microarray.

Table S6. Glycans used for glycoconjugate microarray.

Table S7. Raw data file (Excel file).

REFERENCES AND NOTES

Acknowledgments: We thank S. Tochihara and W. Saito for secretarial assistance, R. Sato for exome sequencing of the NC/Nga genome and analyzing the data, R. Hirochika and K. Hiemori for technical assistance to the generation of NC/Nga-Clec10ac.706T>C mice and the lecin microarray analysis, respectively, and M.S. Almeida for the illustration of the graphic. Funding: This research was supported in part by grants provided by the Japan Agency for Medical Research and Development (AMED-CREST) (to A.S.) and the Ministry of Education, Culture, Sports, Science, and Technology of Japan [to A.S. (16H06387 and 18H05022), K.K. (17K17625 and 19H03695), and H.M. (16H06383)]. The sponsors had no control over the interpretation, writing, or publication of this work. Author contributions: K.K. designed and conducted the experiments, analyzed the data, performed statistical analysis, and wrote the paper. E.N. carried out the exome sequencing of the NC/Nga genome and analyzed the data. S.M., F.S., and S.T. generated NC/Nga-Clec10a706T>C mice. Y.F. and Y.N. contributed the analyses of human skin samples. H.T. performed the lectin and glycoconjugate array analyses. K.D.-N, T.I., and H.M. provided materials. S.T.-H. and K.S. analyzed the data. A.S. supervised the overall project and wrote the paper. Competing interests: The authors declare that they have no competing interests associated with this study. Data and materials availability: Raw BAM files for the NC/Nga exome sequencing were deposited in DNA Data Bank of Japan under the accession number DRA008740. Materials and mouse strains are available through a material transfer agreement. All other data needed to evaluate the conclusions in the paper are present in the paper or the Supplementary Materials.
View Abstract

Stay Connected to Science Immunology

Navigate This Article