Advertisement

Identification of an S100A8 Receptor Neuroplastin-β and its Heterodimer Formation with EMMPRIN

Open ArchivePublished:July 04, 2016DOI:https://doi.org/10.1016/j.jid.2016.06.617
      We previously reported a positive feedback loop between S100A8/A9 and proinflammatory cytokines mediated by extracellular matrix metalloproteinase inducer, an S100A9 receptor. Here, we identify neuroplastin-β as an unreported S100A8 receptor. Neuroplastin-β and extracellular matrix metalloproteinase inducer form homodimers and a heterodimer, and they are co-localized on the surface of cultured normal human keratinocytes. Knockdown of both receptors suppressed cell proliferation and proinflammatory cytokine induction. Upon stimulation with S100A8, neuroplastin-β recruited GRB2 and activated extracellular signal-regulated kinase, resulting in keratinocyte proliferation. Keratinocyte proliferation in response to inflammatory stimuli was accelerated in involucrin promoter-driven S100A8 transgenic mice. Further, S100A8 and S100A9 were strongly up-regulated and co-localized in lesional skin of atopic dermatitis patients. Our results indicate that neuroplastin-β and extracellular matrix metalloproteinase inducer form a functional heterodimeric receptor for S100A8/A9 heterodimer, followed by recruitment of specific adaptor molecules GRB2 and TRAF2, and this signaling pathway is involved in activation of both keratinocyte proliferation and skin inflammation in atopic skin. Suppression of this pathway might have potential for treatment of skin diseases associated with chronic inflammation such as atopic dermatitis.

      Abbreviations:

      AD (atopic dermatitis), EMMPRIN (extracellular matrix metalloproteinase inducer), ERK (extracellular signal-regulated kinase), NPTN (neuroplastin), RAGE (receptor for advance glycation end-products), TLR (Toll-like receptor), TNCB (trinitrochlorobenzene), TNF (tumor necrosis factor), TRAF (tumor necrosis factor receptor-associated factor)

      Introduction

      Emerging evidence indicates that inflammation in atopic dermatitis (AD) results primarily from a defect of skin barrier function (
      • Chan L.S.
      Atopic dermatitis in 2008.
      ,
      • Zheng T.
      • Yu J.
      • Oh M.H.
      • Zhu Z.
      The atopic march: progression from atopic dermatitis to allergic rhinitis and asthma.
      ), which results in excessive loss of moisture from the cornified layer, allowing the skin to become very dry and reducing its protective abilities. Large-scale DNA microarray and proteomics studies of AD skin lesions have shown overexpression of genes located at the epidermal differentiation complex and loss of expression of protective genes in the cornified envelope (
      • Broccardo C.J.
      • Mahaffey S.
      • Schwarz J.
      • Wruck L.
      • David G.
      • Schlievert P.M.
      • et al.
      Comparative proteomic profiling of patients with atopic dermatitis based on history of eczema herpeticum infection and Staphylococcus aureus colonization.
      ,
      • Khattri S.
      • Shemer A.
      • Rozenblit M.
      • Dhingra N.
      • Czarnowicki T.
      • Finney R.
      • et al.
      Cyclosporine in patients with atopic dermatitis modulates activated inflammatory pathways and reverses epidermal pathology.
      ). In particular, S100A8 and S100A9 are highly up-regulated.
      The S100 proteins modulate inflammation (
      • Donato R.
      S100: a multigenic family of calcium-modulated proteins of the EF-hand type with intracellular and extracellular functional roles.
      ,
      • Donato R.
      • Cannon B.R.
      • Sorci G.
      • Riuzzi F.
      • Hsu K.
      • Weber D.J.
      • et al.
      Functions of S100 proteins.
      ,
      • Goyette J.
      • Geczy C.L.
      Inflammation-associated S100 proteins: new mechanisms that regulate function.
      ,
      • Schiopu A.
      • Cotoi O.S.
      S100A8 and S100A9: DAMPs at the crossroads between innate immunity, traditional risk factors, and cardiovascular disease.
      ), and among them, S100A8 and S100A9 form a heterodimer, S100A8/A9 (known as calprotectin), in vivo (
      • Schiopu A.
      • Cotoi O.S.
      S100A8 and S100A9: DAMPs at the crossroads between innate immunity, traditional risk factors, and cardiovascular disease.
      ). We have shown that multiple proinflammatory cytokines are up-regulated in S100A8/A9-treated keratinocytes (
      • Nukui T.
      • Ehama R.
      • Sakaguchi M.
      • Sonegawa H.
      • Katagiri C.
      • Hibino T.
      • et al.
      S100A8/A9, a key mediator for positive feedback growth stimulation of normal human keratinocytes.
      ). Furthermore, S100A8/A9-induced proinflammatory molecules in turn stimulate keratinocytes to synthesize and secrete S100A8/A9, suggesting the existence of a positive feedback loop between S100A8/A9 and these proinflammatory factors.
      Receptor for advanced glycation end-products (RAGE) is known to be a general inflammation-related receptor (
      • Bierhaus A.
      • Stern D.M.
      • Nawroth P.P.
      RAGE in inflammation: a new therapeutic target?.
      ,
      • Clynes R.
      • Moser B.
      • Yan S.F.
      • Ramasamy R.
      • Herold K.
      • Schmidt A.M.
      Receptor for AGE (RAGE): weaving tangled webs within the inflammatory response.
      ,
      • Leclerc E.
      • Fritz G.
      • Vetter S.W.
      • Heizmann C.W.
      Binding of S100 proteins to RAGE: an update.
      ,
      • Sakaguchi M.
      • Murata H.
      • Yamamoto K.
      • Ono T.
      • Sakaguchi Y.
      • Motoyama A.
      • et al.
      TIRAP, an adaptor protein for TLR2/4, transduces a signal from RAGE phosphorylated upon ligand binding.
      ). However, its ligands also include advanced glycation end-products, high mobility group box-1 (HMGB1), and amyloid-β. We postulated that more specific receptor(s) might be involved in the positive feedback mechanism for S100A8/A9 (
      • Nukui T.
      • Ehama R.
      • Sakaguchi M.
      • Sonegawa H.
      • Katagiri C.
      • Hibino T.
      • et al.
      S100A8/A9, a key mediator for positive feedback growth stimulation of normal human keratinocytes.
      ), and we identified extracellular matrix metalloproteinase inducer (EMMPRIN) as a specific receptor for S100A9 in the inflammatory loop (
      • Hibino T.
      • Sakaguchi M.
      • Miyamoto S.
      • Yamamoto M.
      • Motoyama A.
      • Hosoi J.
      • et al.
      S100A9 is a novel ligand of EMMPRIN that promotes melanoma metastasis.
      ). In addition, the action of S100A8 and S100A9 as endogenous damage-associated molecular pattern molecules that link innate immunity and autoimmune responses (
      • Loser K.
      • Vogl T.
      • Voskort M.
      • Lueken A.
      • Kupas V.
      • Nacken W.
      • et al.
      The Toll-like receptor 4 ligands Mrp8 and Mrp14 are crucial in the development of autoreactive CD8+ T cells.
      ) is reported to be mediated by Toll-like receptor 4 (TLR4), which interacts with S100A8 (
      • Vogl T.
      • Tenbrock K.
      • Ludwig S.
      • Leukert N.
      • Ehrhardt C.
      • van Zoelen M.A.
      • et al.
      Mrp8 and Mrp14 are endogenous activators of Toll-like receptor 4, promoting lethal, endotoxin-induced shock.
      ). However, S100A8/A9 induce multiple effects, including cell proliferation, migration, apoptosis, and inflammation (
      • Halayko A.J.
      • Ghavami S.
      S100A8/A9: a mediator of severe asthma pathogenesis and morbidity?.
      ), and we hypothesized that other receptor(s) would also be involved.
      Therefore, we searched for a putative S100A8 receptor involved in the skin inflammation and abnormal proliferation of keratinocytes. We found that neuroplastin-β (NPTNβ) forms a heterodimer with EMMPRIN, and this heterodimer is suggested to work as a functional receptor for S100A8/A9, linking to S100A8/A9-induced skin inflammation and abnormal keratinocyte proliferation in AD.

      Results

      S100A8 specifically binds with NPTNβ

      We searched for a putative cell-surface receptor for S100A8 based on similarity with EMMPRIN using the protein BLAST program in the NCBI database (http://blast.ncbi.nlm.nih.gov/Blast.cgi) and found that NPTN showed considerable similarity with EMMPRIN. NPTN has two splicing variants, NPTNα and NPTNβ (
      • Langnaese K.
      • Beesley P.W.
      • Gundelfinger E.D.
      Synaptic membrane glycoproteins gp65 and gp55 are new members of the immunoglobulin superfamily.
      ,
      • Owczarek S.
      • Berezin V.
      Neuroplastin: cell adhesion molecule and signaling receptor.
      ) (see Supplementary Figure S1a and b online). We examined whether these isoforms interact with S100 proteins. S100A8 was co-immunoprecipitated with NPTNβ but not NPTNα (Figure 1a). S100A9 was found in the NPTNα fraction. NPTNα possesses two Ig domains in the extracellular region, whereas NPTNβ has one extra Ig domain (see Supplementary Figure S1a and b). We found that the additional Ig-3 domain in NPTNβ is responsible for the binding with S100A8 (Figure 1b). To assess more precisely the interaction about NPTNβ, we used purified recombinant proteins with an aim to rule out any cellular participant on the binding. ELISA with immobilized NPTNβ, soluble form (see Supplementary Figure S1c), confirmed dose-dependent bindings of S100A8 and S100A8/A9, which were much higher than that of S100A9 (see Supplementary Figure S1d). At the mRNA and protein levels, S100A8 and S100A9 are similarly expressed and strongly up-regulated in keratinocytes in response to differentiation stimuli (Figure 1c). We found that NPTNβ was the dominant isoform in cultured human keratinocytes, which was more obvious at protein levels, and showed rather constant expression levels under various conditions, whereas EMMPRIN was highly up-regulated by 1.5 mmol/L of calcium ions.
      Figure 1
      Figure 1S100A8 binds with NPTNβ, which is endogenously expressed in cultured keratinocytes. (a) Binding of NPTNα, NPTNβ, and EMMPRIN with S100 proteins was examined. After transfection of each S100 family construct into HEK293 cells, conditioned medium was immunoprecipitated with anti-His antibody, and secretion of the expressed S100 protein was verified by Western blot test. The collection of each conditioned medium was then added to another transfected HEK293 cells with NPTNα, NPTNβ, or EMMPRIN. Bindings of S100 proteins to the expressed receptors were analyzed by immunoprecipitation of the cell extracts with anti-HA antibody. (b) Identification of an indispensable region of NPTNβ for association with S100A8. Truncated forms of receptors, which are lacking the cytoplasmic tails, NPTNα (Δcyt), NPTNβ (Δcyt), and NPTNβ-Ig-3 (Δcyt) (left) were transfected to HEK293 cells. After addition of the conditioned medium containing S100A8 recombinant protein, which was prepared from another group of transfected HEK293 cells, to each truncated receptor-expressed group of cells, immunoprecipitation of the truncated receptors was done with anti-HA antibody. (c) Expression of S100A8, S100A9, NPTNβ, and EMMPRIN in cultured keratinocytes. mRNA and protein expressions were examined using real-time PCR (left) and Western blotting (right), respectively, in the growth phase (80%), at confluence (100%), 2 days after confluence (120%), after air exposure (Air), and in the presence of 1.5 mmol/L of calcium ions after confluence (Ca++). P > 0.05, ∗∗P > 0.01, ∗∗∗P > 0.001. EMMPRIN, extracellular matrix metalloproteinase inducer; IP, immunoprecipitation; NPTN, neuroplastin; WB, western blot.
      It has been reported that TLR4 transduces S100A8 signals in various cell types (
      • Gan N.
      • Yang L.
      • Omran A.
      • Peng J.
      • Wu L.
      • He F.
      • et al.
      Myoloid-related protein 8, an endogenous ligand of Toll-like receptor 4, is involved in epileptogenesis of mesial temporal lobe epilepsy via activation of the nuclear factor-kappaB pathway in astrocytes.
      ). We found that TLR4 was highly expressed in spleen and HeLa cells but that it was undetectable in cultured keratinocytes (see Supplementary Figures S2a and b and S3 online). Moreover, a TLR4-neutralizing antibody had no effect on S100A8-induced cytokine induction (see Supplementary Figure S2c). In addition, S100A8 did not show any affinity to RAGE (see Supplementary Figure S4 online). Binding of S100A8/A9 with RAGE was weak compared with that of HMGB1, S100A11, and S100A9 (see Supplementary Figure S4b). These results suggest that TLR4 and RAGE would not be key receptors for S100A8, at least in human keratinocytes.

      S100A8/A9 signaling pathway

      Motif analysis showed that the cytoplasmic tails of NPTN and EMMPRIN contain tumor necrosis factor (TNF) receptor associated factor (TRAF)-binding motifs. In addition, NPTN possesses an SH3 domain that binds to proline-rich sequences containing a conserved PxxP motif, where P is proline and x is any amino acid (
      • Kay B.K.
      • Williamson M.P.
      • Sudol M.
      The importance of being proline: the interaction of proline-rich motifs in signalling proteins with their cognate domains.
      ,
      • Mayer B.J.
      SH3 domains: complexity in moderation.
      ) (see Supplementary Figure S5a online). We expressed TRAF2, GRB2, CRK, NCK1, NCK2, and p38α proteins in HEK293 cells, together with EMMPRIN or NPTN, and examined complex formation (Figure 2a). Upon stimulation with S100A8/A9, EMMPRIN recruited only TRAF2, and NPTN recruited TRAF2 and GRB2, although NPTN-TRAF2 binding seemed to remain at relatively low levels compared with that of EMMPRIN. Mutation of the SH3 binding motif of NPTN did not affect TRAF2 binding but abolished GRB2 recruitment (Figure 2b, and see Supplementary Figure S5b). When the TRAF binding motif was mutated, only GRB2 was detected as an interactant with NPTN. S100A8/A9-induced activations of effector kinases p38 and extracellular signal-regulated kinase (ERK) were both suppressed by overexpression of the cytoplasmic domain of wild-type NPTN. That of mutated SH3 domain specifically inhibited the activation of p38, whereas the mutated TRAF2 domain functioned to mitigate ERK activation (Figure 2c). In these cases, mutated cytoplasmic domains of TRAFs and GRB2 trapped intact TRAF2 and GRB2, resulting in ERK inhibition in the former and p38 inhibition in the latter. Before receptor-mediated downstream analysis using siRNAs (see Supplementary Figure S6 online), we examined kinetics of activation levels of ERK and p38. Under stimulation with S100A8/A9, ERK and p38 were phosphorylated, and peak activity was reached at 30 minutes without noticeable attenuation thereafter (see Supplementary Figure S7a online). Knockdown of NPTN (see Supplementary Figure S6) caused profound down-regulation of S100A8/A9-mediated ERK phosphorylation, including the steady-state level (Figure 2d). In the case of S100A8 stimulation, knockdown of TRAF2 abolished p38 phosphorylation, and knockdown of GRB2 resulted in loss of ERK phosphorylation. Double knockdown gave essentially the same results as observed with a single knockdown. Similar results were also obtained for S100A9 stimulation and S100A8/A9 stimulation (see Supplementary Figures S6 and S7b). When keratinocytes were treated with small interfering RNA targeted to EMMPRIN (see Supplementary Figure S6), induction of mitogen-activated protein kinase phosphorylation by S100A8/A9 was considerably suppressed (Figure 2d). To rule out any contribution of S100A9-NPTNα axis to the downstream effects on S100A8/A9 in keratinocytes via the S100A8-NPTNβ interaction, we used truncated forms of receptors, NPTNα (Δcyt), NPTNβ (Δcyt) and NPTNβ-Ig-3 (Δcyt) for the assessment. By this approach, we found that both NPTNβ (Δcyt) and NPTNβ-Ig-3 (Δcyt) significantly inhibited the S100A8/A9-mediated ERK and p38 activation, whereas NPTNα (Δcyt) had almost no effect on ERK activation but impaired p38 phosphorylation, probably because of deprivation of S100A9 binding from EMMPRIN (Figure 2e). Together with the results of slight expression of NPTNα in keratinocytes (Figure 1c), we emphasize that the S100A8-NPTNβ axis plays a significant role in keratinocyte behavior upon S100A8/A9 stimulation. We also tested the involvement of endogenous proteins with co-immunoprecipitation studies. We used normal human keratinocytes and HeLa cells for comparison of signals in the presence or absence of S100A8/A9. NPTNβ was the only isoform pulled down with EMMPRIN in keratinocytes (Figure 2f). EMMPRIN was expressed strongly in keratinocytes, and it was considerably low in HeLa cells (see Supplementary Figure S3b). TLR4 was detected only in HeLa cells. RAGE and TLR4 recruited MyD88 and TRAF6. TRAF2 was found in keratinocytes and HeLa cells only in the pulled-down samples with EMMPRIN. GRB2 was co-immunoprecipitated with EMMPRIN solely in keratinocytes. Collectively, these results suggest that keratinocytes highly expressed NPTNβ and EMMPRIN and specifically recruited GRB2 and TRAF2 upon stimulation with S100A8/A9.
      Figure 2
      Figure 2Upon stimulation with S100A8/A9, NPTNβ and EMMPRIN recruit specific adaptor molecules and phosphorylate mitogen-activated protein kinases in cultured human keratinocytes. (a) Identification of adaptor molecules for signal transduction. Cytoplasmic domain of EMMPRIN (Emmprin-cyt) and various adaptor proteins as C-terminal 3×HA-6His–tagged forms were expressed in HEK293 cells and pulled down with anti-His tag antibody after S100A8/A9 stimulation for 30 minutes. (b) Effect of binding motif mutation on binding with adaptor molecules. Cytoplasmic domains of NPTN (NPTN-cyt) were mutated either at the SH3 domain or the TRAF binding motif and expressed in cultured keratinocytes. After stimulation with S100A8/A9, expressions of TRAF2 and GRB2 were verified with anti-TRAF antibody and anti-GRB2 antibody, respectively. (c) Inhibition of NPTN-mediated signal transduction in NHK cells by overexpression of NPTN cytoplasmic domains (NPTN-cyt) with or without mutations as decoys. S100A8/A9-induced activations of effector kinases, p38 and ERK, were both suppressed by overexpression of the NPTN-cyt (wt). The NPTN-cyt (mut SH3) specifically inhibited the activation of p38, and the NPTN-cyt (mut TRAF) functioned in the mitigation of ERK activation. (d) Involvement of NPTNβ and EMMPRIN in S100A8/A9 signal transduction. Under conditions of receptor knockdown, the effect of S100A8/A9 on ERK and p38 phosphorylation was examined in cultured keratinocytes. Tubulin was used as a loading control. Each level of phosphorylated ERK and phosphorylated p38 was expressed as p-ERK/ERK and p-p38/p38 ratio, respectively, after normalization with tubulin. (e) Assessment of downstream effects on S100A8/A9 in keratinocytes via S100A8-NPTNβ interaction. Truncated forms of receptors, NPTNα (Δcyt), NPTNβ (Δcyt), and NPTNβ-Ig-3 (Δcyt), were transfected to NHK cells, to prevent specific interactions of each S100 protein with the endogenous receptors. After addition of the recombinant S100A8, S100A9, or S100A8/A9 proteins, the effect of each S100 protein on ERK and p38 phosphorylation was examined. Tubulin was used as a loading control. Each level of phosphorylated ERK and phosphorylated p38 was expressed as p-ERK/ERK and p-p38/p38 ratio, respectively, after normalization with tubulin. (f) Binding profiles of endogenous receptors (RAGE, TLR4, EMMPRIN, and NPTNβ), ligand (S100A8/A9), and downstream adaptor proteins. NHK and HeLa cells (TLR4-positive cells, as shown in b online) were treated with 10 nmol/L of S100A8/A9 for 30 minutes. The treated cells were lysed and subjected to immunoprecipitation using the indicated biotinylated antibodies. In both NHK and HeLa cells, RAGE showed S100A8/A9 binding, which recruited MyD88 and TRAF6 adaptor proteins. RAGE exhibited no binding with TLR4, EMMPRIN, and NPTNα or -β. Similar bindings were also shown forTLR4 in HeLa cells. S100A8/A9 was bound with EMMPRIN in both NHK and HeLa cells. Interestingly, EMMPRIN/NPTNβ binding only appeared in NHK cells, at which the complex enabled recruitment of TRAF2 as well as GRB2. In HeLa cells, EMMPRIN did not immunoprecipitate NPTNβ and recruited TRAF2 only. ab, antibody; EMMPRIN, extracellular matrix metalloproteinase inducer; ERK, extracellular signal-regulated kinase; IP, immunoprecipitation; mut, mutated; NHK, normal human keratinocyte; NPTN, neuroplastin; p-, phosphorylated; RAGE, receptor for advanced glycation end-products; TLR, Toll-like receptor; TRAF, tumor necrosis factor receptor associated factor; WB, western blot.
      Thus, growth and inflammatory signals induced by these S100 proteins are mediated by NPTNβ and EMMPRIN in keratinocytes and are dependent on recruitment of GRB2 and TRAF2.

      S100A8/A9 receptor interaction is critical for keratinocyte proliferation and cytokine induction

      Even in the steady-state condition, double knockdown of these receptors had some effects on cell growth (see Supplementary Figure S8 online). On the other hand, suppression of NPTN or EMMPRIN alone showed little effect on cell proliferation induced by either S100 proteins or by S100A8/A9. When S100A8 was added to the culture medium and incubated for 24 hours, single receptor knockdown had little effect, but suppression of both receptors caused maximal growth inhibition. Similar results were obtained with S100A9. In the case of S100A8/A9 stimulation, double knockdown of these receptors caused more than 50% growth suppression. Analyses of receptor oligomerization showed that NPTNβ bound with itself, as well as with NPTNα (Figure 3a). Similar results were obtained with EMMPRIN. NPTN knockdown significantly suppressed the S100A8-mediated induction of CXCL-1, TNF-α, and IL-8, compared with the control (Figure 3b). Induction by S100A9 or S100A8/A9 was also suppressed. Knockdown of EMMPRIN had essentially the same effects on cytokine induction. Collectively, these results indicate that NPTNβ-EMMPRIN heterodimer serves as a receptor for S100A8/A9 and induces inflammatory cytokines and keratinocyte proliferation.
      Figure 3
      Figure 3S100A8/A9-dependent keratinocyte proliferation and cytokine induction are mediated via their receptors. (a) Receptor oligomerization. Myc-tagged and His-tagged EMMPRIN, NPTNα, or NPTNβ were expressed in cultured keratinocytes and pulled down with anti-Myc antibody. Dimer components were examined with anti-His tag antibody. (b) Suppression of cytokine induction by knockdown of specific receptors. After knockdown of either receptor or both, cultured keratinocytes were stimulated with S100A8, S100A9, or S100A8/A9. Quantitative PCR was carried out using specific primer pairs for CXCL1, TNF-α, and IL8. P > 0.05, ∗∗P > 0.01, ∗∗∗P > 0.001. EMMPRIN, extracellular matrix metalloproteinase inducer; IP, immunoprecipitation; NPTN, neuroplastin; RAGE, receptor for advanced glycation end-products; si, small interfering; TNF, tumor necrosis factor; WB, western blot.
      Because TRAF2 is also a signal transducer of the NF-κB pathway, we examined whether S100A8/A9 is capable of activating this pathway. After stimulation with S100A8/A9, NF-κB p50 subunit was translocated and accumulated in the nucleus as with TNF-α stimulation. Addition of parthenolide, which inhibits IκB phosphorylation, suppressed nuclear translocation of p50 by S100A8/A9 (see Supplementary Figure S9a and b online). Furthermore, NF-κB luciferase reporter assays showed that S100A9 is responsible for NF-κB signaling (see Supplementary Figure S9c). The effect of S100A8/A9 was less than that of S100A9, and S100A8 alone had little effect. Thus, S100A9-EMMPRIN interaction leads to activation of the NF-κB pathway.

      Localization of S100A8, S100A9, NPTN, and EMMPRIN

      We next examined the localization of S100 proteins and their receptors. All of these proteins were barely detectable in normal human skin and nonlesional skin with AD (see Supplementary Figure S10 online). In contrast, heavy staining of S100A8 and S100A9 was always observed in the upper epidermal layers of AD skin (Figure 4a). NPTN and EMMPRIN showed similar localization, although NPTN was also detected at the basal layer, especially beneath S100A8-positive areas of lesional skin. These results may imply that both molecules were induced at the onset of inflammation (see Supplementary Figure S10b). A monoclonal antibody, 27E10, which recognizes only the S100A8/A9 heterodimer, showed strong staining in the upper epidermal layers (Figure 4b). We also investigated localization of TLR4 and RAGE (see Supplementary Figure S11 online). Although cultured keratinocytes did not express TLR4, we recognized localization of TLR4 in the granular layer of normal epidermis (see Supplementary Figure S11a). However, TLR4 was markedly down-regulated both in the nonlesional and lesional skin with AD (see Supplementary Figure S11b). Expression of RAGE was considerably low in the normal and atopic skin (see Supplementary Figure S11). This was in contrast to the cultured cells, where human keratinocytes expressed a relatively high level of RAGE mRNA (see Supplementary Figure S3b). To investigate ligand-receptor relationships in vivo, we used proximity ligation assay (Figure 4c). Significant reactions between NPTN and S100A8 or S100A8/A9 were detected at epidermis covering the upper granular and the basal layers. Interaction between NPTN and S100A9 was strongly positive in the upper epidermis; EMMPRIN showed bright, almost linear reaction patterns, with S100 proteins at the granular layer (Figure 4d). S100A8/A9-EMMPRIN interaction showed a broader localization, including the basal layer. We also observed strong positive reactions for NPTN and EMMPRIN from the basal through the upper epidermis of atopic skin (Figure 4e). Positive reaction between NPTN and EMMPRIN in the lower epidermis was evident, especially in areas of acanthotic epidermis, where keratinocyte proliferation and parakeratotic changes took place. Combination of unrelated IgGs did not give any positive signals, showing specificity of these reactions (Figure 4f). The NPTNβ/EMMPRIN interaction was also studied by a co-immunoprecipitation experiment using tissue extract from a fresh atopic skin specimen (see Supplementary Figure S11c). We found that EMMPRIN co-immunoprecipitated with NPTNβ. Next, we investigated localization of endogenous NPTNβ and EMMPRIN in cultured keratinocytes with confocal microscopy (Figure 4g). NPTNβ and EMMPRIN showed membrane-associated localization, and these molecules were mostly co-localized, especially on the cell-cell contact area. A merged immunofluorescence image with DAPI staining is also shown in Figure 4h. Co-expression of these receptors resulted in extensive co-localization or association on the cell surface, consistent with heterodimer formation of these receptors.
      Figure 4
      Figure 4Immunohistochemical localization and PLA analysis of related molecules in lesional AD skin suggest heterodimer formation of both ligands and receptors. (a) Localization of S100A8, S100A9, NPTN, and EMMPRIN in lesional AD skin. Merged figures with nuclear staining are also shown. Scale bar = 100 μm. (b) Localization of S100A8/A9 (calprotectin). To clarify the presence of S100A8/A9, we used 27E10 antibody, which specifically recognizes this dimer. A merged figure with nuclear staining is shown. Scale bar = 100 μm. (c) PLA analysis for S100 proteins and NPTN. Close associations between S100A8 and NPTN, S100A9 and NPTN, and S100A8/A9 and NPTN were analyzed using the PLA method. (d) Similarly, combinations of S100A8 and EMMPRIN, S100A9 and EMMPRIN, and S100A8/A9 and EMMPRIN were investigated. (e) Close association of NPTN and EMMPRIN. A merged figure with nuclear staining is also shown. Scale bar = 50 μm. (f) PLA negative control. Combination of unrelated IgGs did not give any positive signals, showing specificity of these reactions. (g) Co-localization expression of NPTNβ and EMMPRIN in cultured keratinocytes. Human keratinocytes in culture were stained with anti-NPTN and EMMPRIN antibodies. (h) A merged figure with nuclear staining is also shown. Localization was examined with confocal microscopy. Scale bar = 10 μm. Dotted and dashed lines show top surface of the epidermis and interface between the epidermis and the dermis, respectively. AD, atopic dermatitis; EMMPRIN, extracellular matrix metalloproteinase inducer; NPTN, neuroplastin; PLA, proximity ligation assay.

      S100A8 transgenic mice showed abnormal proliferation of keratinocytes

      Because S100A8 is able to bind with NPTNβ and recruit GRB2, we examined whether S100A8 is involved in excessive cell proliferation at the onset of inflammation by using involucrin promoter-driven S100A8 transgenic mice (
      • Hibino T.
      • Sakaguchi M.
      • Miyamoto S.
      • Yamamoto M.
      • Motoyama A.
      • Hosoi J.
      • et al.
      S100A9 is a novel ligand of EMMPRIN that promotes melanoma metastasis.
      ) (see Supplementary Figure S12a online). We applied SDS as a simple inflammatory stimulant and trinitrochlorobenzene (TNCB) as an inflammatory and sensitizing stimulant. In the transgenic mice, the expression of human S100A8 for SDS and TNCB tended to be induced less than 2-fold and 2.5-fold, respectively (see Supplementary Figure S12b). Only the human transgenic S100A8 showed an increased tendency after stimulation. Intrinsic mouse S100a8 showed rather suppressive tendency in the case of TNCB stimulation. We confirmed that human S100A8 and S100A9 bound with mouse Nptnβ and Emmprin, respectively (see Supplementary Figure S12c). Measurement of ear swelling indicated that SDS treatment tended to have only a very mild effect in wild-type HR-1 mice, whereas a stronger response was observed in S100A8 transgenic mice, peaking at 96 hours after application. There was a strong response to TNCB treatment. Ear swelling gradually increased up to 72 hours and then showed a declined tendency in the wild-type mice. In contrast, ear swelling remained high in S100A8 transgenic mice, even at the 144-hour time point (Figure 5a). Histological sections of the treated dorsal skin exhibited mild spongiosis and dermal edema in TNCB-treated skin of both wild and transgenic mice at the 24-hour time point (Figure 5b), although nontreated S100A8 transgenic mice did not show any noticeable changes compared with wild-type mice (see Supplementary Figure S13a online). Acanthotic epidermis was evident in the skin of S100A8-positive transgenic mice (Figure 5b, and see Supplementary Figure S13b). Numbers of proliferating cells identified with Ki-67 staining were markedly increased at 96 hours, and many basal cells had positive results in the dorsal skin of S100A8 transgenic mice (see Supplementary Figure S14a online). Some spinous cells also had positive results in these mice (see Supplementary Figure S14b). Quantitative analysis clearly showed that Ki67-positive cells were highly up-regulated after stimulation with either SDS or TNCB in the skin of transgenic mice (Figure 5c).
      Figure 5
      Figure 5S100A8 transgenic mice driven by the involucrin promoter are hyperreactive to inflammatory stimuli. (a) Ear thickness of wild-type and transgenic mice after SDS or TNCB treatment. Ear thickness was measured at appropriate times (n = 5, mean ± standard deviation of three measurements). P > 0.05, ∗∗P > 0.01. (b) Hematoxylin and eosin staining of dorsal skin of wild-type and transgenic mice after irritant stimulation. Skin samples were taken from the treated area at 24, 96, and 144 hours. Scale bar = 50 μm. (c) Quantitative analysis of Ki67-positive cells in dorsal skins in a online. Numbers of Ki67-positive cells were counted in 500 μm-long fields × 8. Mean values of Ki67-positive cells in the 500-μm length were plotted. Error bars indicate standard deviation. ∗∗P < 0.01, ∗∗∗P < 0.001. h, hours; TG, transgenic; TNCB, trinitrochlorobenzene; wild, wild type.

      Discussion

      We identified NPTNβ, but not NPTNα, as an unreported receptor for S100A8. Physiological roles of NPTNs have been explored in neuronal cells (
      • Owczarek S.
      • Berezin V.
      Neuroplastin: cell adhesion molecule and signaling receptor.
      ) and include cell adhesion in the plexiform layers during histogenesis (
      • Kreutz M.R.
      • Langnaese K.
      • Dieterich D.C.
      • Seidenbecher C.I.
      • Zuschratter W.
      • Beesley P.W.
      • et al.
      Distribution of transcript and protein isoforms of the synaptic glycoprotein neuroplastin in rat retina.
      ), promoting long-term changes in synaptic activity (
      • Smalla K.H.
      • Matthies H.
      • Langnase K.
      • Shabir S.
      • Bockers T.M.
      • Wyneken U.
      • et al.
      The synaptic glycoprotein neuroplastin is involved in long-term potentiation at hippocampal CA1 synapses.
      ), and promoting synaptic plasticity (
      • Owczarek S.
      • Kiryushko D.
      • Larsen M.H.
      • Kastrup J.S.
      • Gajhede M.
      • Sandi C.
      • et al.
      Neuroplastin-55 binds to and signals through the fibroblast growth factor receptor.
      ). To our knowledge, this is the first report that NPTNβ is dominantly expressed in human keratinocytes and transduces signals for S100A8.
      Because S100A8 forms a stable heterodimer with S100A9 (
      • Manitz M.P.
      • Horst B.
      • Seeliger S.
      • Strey A.
      • Skryabin B.V.
      • Gunzer M.
      • et al.
      Loss of S100A9 (MRP14) results in reduced interleukin-8-induced CD11b surface expression, a polarized microfilament system, and diminished responsiveness to chemoattractants in vitro.
      ), we considered the possible involvement of the S100A9 receptor EMMPRIN in the present signaling pathway. Immunoprecipitation studies indicated that NPTNβ and EMMPRIN could form heterodimers in cultured keratinocytes and AD skin. Proximity ligation assay also supported the possibility of fashioning the receptors into heterodimers in the granular and basal layers of epidermis in AD skin. Thus, a characteristic feature of this ligand-receptor relationship may appear to be heterodimer formation of both the ligands and the receptors.
      We also examined the signaling pathways upon S100A8/A9 stimulation and found that NPTNβ recruits GRB2, whereas EMMPRIN recruits TRAF2. Knockdown of either receptor or adaptor molecules markedly suppressed both ERK and p38 phosphorylation. Double knockdown of the two receptors had a profound effect on mitogen-activated protein kinase activation and keratinocyte proliferation. These results further support the idea that NPTNβ/EMMPRIN heterodimer functions as a receptor for S100A8/A9.
      TLR4 and RAGE have been established as general receptors for many S100 proteins. Although we found that TLR4 is expressed in the granular layer of normal skin, it is markedly down-regulated in the atopic skin. These results show sharp contrast to the expression profiles of NPTN, in which it was hardly detectable in the normal skin and highly up-regulated in the lesional skin with AD. As for RAGE signaling (
      • Foell D.
      • Wittkowski H.
      • Vogl T.
      • Roth J.
      S100 proteins expressed in phagocytes: a group of damage-associated molecular pattern molecules.
      ,
      • Heizmann C.W.
      • Ackermann G.E.
      • Galichet A.
      Pathologies involving the S100 proteins and RAGE.
      ,
      • Ibrahim Z.A.
      • Armour C.L.
      • Phipps S.
      • Sukkar M.B.
      RAGE and TLRs: relatives, friends or neighbours?.
      ), RAGE did not bind with S100A8 and did not form a heterodimer with NPTNα and -β or EMMPRIN. These findings are consistent with a recent report showing that loss of TLR4 signaling or RAGE deficiency did not appreciably affect S100A9-mediated lung pathology or inflammatory cell infiltration in the alveolar space (
      • Chen B.
      • Miller A.L.
      • Rebelatto M.
      • Brewah Y.
      • Rowe D.C.
      • Clarke L.
      • et al.
      S100A9 induced inflammatory responses are mediated by distinct damage associated molecular patterns (DAMP) receptors in vitro and in vivo.
      ). Taken together, we consider that the NPTNβ/EMMPRIN heterodimer, rather than TLR4 and RAGE, has a critical role in S100A8/A9-dependent physiological reactions in human keratinocytes toward the atopic state.
      TRAF2 is required for S100A9-induced signaling via EMMPRIN and is also a well-known adaptor for the TNF-α signaling pathway to induce an activation of NF-κB (
      • Bradley J.R.
      • Pober J.S.
      Tumor necrosis factor receptor-associated factors (TRAFs).
      ,
      • Wajant H.
      • Grell M.
      • Scheurich P.
      TNF receptor associated factors in cytokine signaling.
      ). Our results show that S100A8/A9 activates NF-κB. This effect is solely S100A9 dependent, because NF-κB luciferase assay clearly showed that S100A9 but not S100A8 has the ability to induce NF-κB pathway activation. Because of the strong proinflammatory activities of the TNF-α–NF-κB system, the S100A9-EMMPRIN system appears to have similar functions to the system. However, we found that S100A8-NPTNβ binding recruits the adaptor protein GRB2. GRB2 is an essential factor for induction of cell proliferation by various growth factors including epidermal growth factor (
      • Fridell Y.W.
      • Jin Y.
      • Quilliam L.A.
      • Burchert A.
      • McCloskey P.
      • Spizz G.
      • et al.
      Differential activation of the Ras/extracellular-signal-regulated protein kinase pathway is responsible for the biological consequences induced by the Axl receptor tyrosine kinase.
      ), linking to the Ras–mitogen-activated protein kinase/ERK–ERK pathway. Thus, S100A8 and S100A9 possess distinct and cross-reactive functions. Formation of the receptor heterodimer would be highly effective to integrate complex signals and to promote strong cell reactions leading to proliferation and inflammation.
      To examine further the action of S100A8, we used involucrin promoter-driven S100A8 transgenic mice, because intrinsic S100a8 is expressed mostly in the upper epidermis. The human transgene showed considerable up-regulation after SDS or TNCB treatment, but to our surprise, TNCB treatment rather suppressed the level of endogenous S100A8. These changes may be due to compensatory reactions. The skin of S100A8 transgenic mice consistently exhibited hyperreactive characteristics. S100a8-deficient mice show embryonic lethality, possibly because loss of S100a8 in developmental stages affects essential cell proliferation.
      Our results indicate that NPTNβ/EMMPRIN heterodimer could function as a receptor for S100A8/A9, leading to activation of both keratinocyte proliferation and skin inflammatory pathways via recruitment of specific adaptor molecules, GRB2 and TRAF2. This idea is consistent with increasing evidence that chemokine receptors in general form homo- or hetero-oligomeric complexes, resulting in complex networks and crosstalk with other orthogonal signaling complexes (
      • Kraemer S.
      • Alampour-Rajabi S.
      • El Bounkari O.
      • Bernhagen J.
      Hetero-oligomerization of chemokine receptors: diversity and relevance for function.
      ). Thus, our findings offer the latest insight into the roles of S100A8 and S100A9 in chronic inflammation of atopic skin. Targeting this ligand-receptor axis would be a promising strategy to ameliorate skin inflammatory diseases.

      Materials and Methods

      Cell culture

      Normal human keratinocytes were purchased from Kurabo (Osaka, Japan). Human embryonic kidney cell line (HEK293) and cervical cancer cell line (HeLa) were purchased from ATCC (Manassas, VA). Details of cell culture are provided in the Supplementary Materials online.

      Immunohistochemistry

      Immunohistochemical studies using human tissue specimens were approved by the Ethical Committee of Tokyo Medical University, methods were carried out in accordance with the approved guidelines, and only samples in the university were used. Preparation of tissue specimens and immunostaining were performed as detailed in the Supplementary Materials.

      S100A8/A9 proteins

      High-purity human S100A8 and S100A9 recombinant proteins were prepared as reported previously (
      • Hibino T.
      • Sakaguchi M.
      • Miyamoto S.
      • Yamamoto M.
      • Motoyama A.
      • Hosoi J.
      • et al.
      S100A9 is a novel ligand of EMMPRIN that promotes melanoma metastasis.
      ,
      • Nukui T.
      • Ehama R.
      • Sakaguchi M.
      • Sonegawa H.
      • Katagiri C.
      • Hibino T.
      • et al.
      S100A8/A9, a key mediator for positive feedback growth stimulation of normal human keratinocytes.
      ,
      • Sakaguchi M.
      • Murata H.
      • Aoyama Y.
      • Hibino T.
      • Putranto E.W.
      • Ruma I.M.
      • et al.
      DNAX-activating protein 10 (DAP10) membrane adaptor associates with receptor for advanced glycation end products (RAGE) and modulates the RAGE-triggered signaling pathway in human keratinocytes.
      ). The amount of contaminating endotoxins (lipopolysaccharides and β-glucans) was confirmed to be less than 0.01 endotoxin units/μg, as determined with a Limulus amebocyte lysate assay (Seikagaku Corporation, Tokyo, Japan).

      S100A8-NPTN binding assay

      Quantitative binding analysis between S100A8 and NPTNβ was performed as detailed in the Supplementary Materials.

      Keratinocyte proliferation

      Keratinocyte proliferation was assessed as described in the Supplementary Materials.

      Quantitative PCR

      Primers used are listed on Supplementary Table S1 online. Real-time PCR details are provided in the Supplementary Materials.

      Small interfering RNA-mediated knockdown

      Knockdown experiments using small interfering RNAs were performed as described in the Supplementary Materials.

      Vector constructs

      cDNAs were inserted into the pIDT-SMART (C-TSC) vector, also named pCMViR-TSC (
      • Sakaguchi M.
      • Watanabe M.
      • Kinoshita R.
      • Kaku H.
      • Ueki H.
      • Futami J.
      • et al.
      Dramatic increase in expression of a transgene by insertion of promoters downstream of the cargo gene.
      ). The inserts are listed in the Supplementary Materials.

      Co-immunoprecipitation

      Co-immunoprecipitation and Western blot experiments were performed as described in the Supplementary Materials.

      Proximity ligation assay

      The proximity ligation assay method was used to detect in situ interaction of target proteins. Details are provided in the Supplementary Materials.

      Inflammatory response in S100A8 transgenic mice

      Responses to simple inflammatory stimulation and inflammation-sensitization stimulation were analyzed using involucrin promoter-driven S100A8 transgenic mice with a hairless phenotype (HR-1, n = 5/group) (
      • Hibino T.
      • Sakaguchi M.
      • Miyamoto S.
      • Yamamoto M.
      • Motoyama A.
      • Hosoi J.
      • et al.
      S100A9 is a novel ligand of EMMPRIN that promotes melanoma metastasis.
      ). Details are provided in the Supplementary Materials.

      Statistical analysis

      Data are expressed as mean ± standard deviation. We used simple pair-wise comparison with Student t test (two-tailed distribution with equal variance in the two samples). P < 0.05 was considered significant.

      Conflict of Interest

      The authors state no conflict of interest.

      Acknowledgments

      We thank Masuyoshi Saito, Mami Saito, and Yukari Okubo (Tokyo Medical University) for providing biopsy skin samples from patients with atopic dermatitis. This work was supported in part by the Japan Society for the Promotion of Science Grant-in-Aid for Scientific Research (JSPS KAKENHI), grant number 26860899 to MY and grant numbers 26290039 15K14382 to MS.

      Supplementary Material

      References

        • Bierhaus A.
        • Stern D.M.
        • Nawroth P.P.
        RAGE in inflammation: a new therapeutic target?.
        Curr Opin Investig Drugs. 2006; 7: 985-991
        • Bradley J.R.
        • Pober J.S.
        Tumor necrosis factor receptor-associated factors (TRAFs).
        Oncogene. 2001; 20: 6482-6491
        • Broccardo C.J.
        • Mahaffey S.
        • Schwarz J.
        • Wruck L.
        • David G.
        • Schlievert P.M.
        • et al.
        Comparative proteomic profiling of patients with atopic dermatitis based on history of eczema herpeticum infection and Staphylococcus aureus colonization.
        J Allergy Clin Immunol. 2011; 127 (193.e1-11): 186-193
        • Chan L.S.
        Atopic dermatitis in 2008.
        Curr Dir Autoimmun. 2008; 10: 76-118
        • Chen B.
        • Miller A.L.
        • Rebelatto M.
        • Brewah Y.
        • Rowe D.C.
        • Clarke L.
        • et al.
        S100A9 induced inflammatory responses are mediated by distinct damage associated molecular patterns (DAMP) receptors in vitro and in vivo.
        PloS One. 2015; 10: e0115828
        • Clynes R.
        • Moser B.
        • Yan S.F.
        • Ramasamy R.
        • Herold K.
        • Schmidt A.M.
        Receptor for AGE (RAGE): weaving tangled webs within the inflammatory response.
        Curr Mol Med. 2007; 7: 743-751
        • Donato R.
        S100: a multigenic family of calcium-modulated proteins of the EF-hand type with intracellular and extracellular functional roles.
        Int J Biochem Cell Biol. 2001; 33: 637-668
        • Donato R.
        • Cannon B.R.
        • Sorci G.
        • Riuzzi F.
        • Hsu K.
        • Weber D.J.
        • et al.
        Functions of S100 proteins.
        Curr Mol Med. 2013; 13: 24-57
        • Foell D.
        • Wittkowski H.
        • Vogl T.
        • Roth J.
        S100 proteins expressed in phagocytes: a group of damage-associated molecular pattern molecules.
        J Leukoc Biol. 2007; 81: 28-37
        • Fridell Y.W.
        • Jin Y.
        • Quilliam L.A.
        • Burchert A.
        • McCloskey P.
        • Spizz G.
        • et al.
        Differential activation of the Ras/extracellular-signal-regulated protein kinase pathway is responsible for the biological consequences induced by the Axl receptor tyrosine kinase.
        Mol Cell Biol. 1996; 16: 135-145
        • Gan N.
        • Yang L.
        • Omran A.
        • Peng J.
        • Wu L.
        • He F.
        • et al.
        Myoloid-related protein 8, an endogenous ligand of Toll-like receptor 4, is involved in epileptogenesis of mesial temporal lobe epilepsy via activation of the nuclear factor-kappaB pathway in astrocytes.
        Mol Neurobiol. 2014; 49: 337-351
        • Goyette J.
        • Geczy C.L.
        Inflammation-associated S100 proteins: new mechanisms that regulate function.
        Amino Acids. 2011; 41: 821-842
        • Halayko A.J.
        • Ghavami S.
        S100A8/A9: a mediator of severe asthma pathogenesis and morbidity?.
        Can J Physiol Pharmacol. 2009; 87: 743-755
        • Heizmann C.W.
        • Ackermann G.E.
        • Galichet A.
        Pathologies involving the S100 proteins and RAGE.
        Subcell Biochem. 2007; 45: 93-138
        • Hibino T.
        • Sakaguchi M.
        • Miyamoto S.
        • Yamamoto M.
        • Motoyama A.
        • Hosoi J.
        • et al.
        S100A9 is a novel ligand of EMMPRIN that promotes melanoma metastasis.
        Cancer Res. 2013; 73: 172-183
        • Ibrahim Z.A.
        • Armour C.L.
        • Phipps S.
        • Sukkar M.B.
        RAGE and TLRs: relatives, friends or neighbours?.
        Mol Immunol. 2013; 56: 739-744
        • Kay B.K.
        • Williamson M.P.
        • Sudol M.
        The importance of being proline: the interaction of proline-rich motifs in signalling proteins with their cognate domains.
        FASEB J. 2000; 14: 231-241
        • Khattri S.
        • Shemer A.
        • Rozenblit M.
        • Dhingra N.
        • Czarnowicki T.
        • Finney R.
        • et al.
        Cyclosporine in patients with atopic dermatitis modulates activated inflammatory pathways and reverses epidermal pathology.
        J Allergy Clin Immunol. 2014; 133: 1626-1634
        • Kraemer S.
        • Alampour-Rajabi S.
        • El Bounkari O.
        • Bernhagen J.
        Hetero-oligomerization of chemokine receptors: diversity and relevance for function.
        Curr Med Chem. 2013; 20: 2524-2536
        • Kreutz M.R.
        • Langnaese K.
        • Dieterich D.C.
        • Seidenbecher C.I.
        • Zuschratter W.
        • Beesley P.W.
        • et al.
        Distribution of transcript and protein isoforms of the synaptic glycoprotein neuroplastin in rat retina.
        Invest Ophthalmol Vis Sci. 2001; 42: 1907-1914
        • Langnaese K.
        • Beesley P.W.
        • Gundelfinger E.D.
        Synaptic membrane glycoproteins gp65 and gp55 are new members of the immunoglobulin superfamily.
        J Biol Chem. 1997; 272: 821-827
        • Leclerc E.
        • Fritz G.
        • Vetter S.W.
        • Heizmann C.W.
        Binding of S100 proteins to RAGE: an update.
        Biochim Biophys Acta. 2009; 1793: 993-1007
        • Loser K.
        • Vogl T.
        • Voskort M.
        • Lueken A.
        • Kupas V.
        • Nacken W.
        • et al.
        The Toll-like receptor 4 ligands Mrp8 and Mrp14 are crucial in the development of autoreactive CD8+ T cells.
        Nature Med. 2010; 16: 713-717
        • Manitz M.P.
        • Horst B.
        • Seeliger S.
        • Strey A.
        • Skryabin B.V.
        • Gunzer M.
        • et al.
        Loss of S100A9 (MRP14) results in reduced interleukin-8-induced CD11b surface expression, a polarized microfilament system, and diminished responsiveness to chemoattractants in vitro.
        Mol Cell Biol. 2003; 23: 1034-1043
        • Mayer B.J.
        SH3 domains: complexity in moderation.
        Journal Cell Sci. 2001; 114: 1253-1263
        • Nukui T.
        • Ehama R.
        • Sakaguchi M.
        • Sonegawa H.
        • Katagiri C.
        • Hibino T.
        • et al.
        S100A8/A9, a key mediator for positive feedback growth stimulation of normal human keratinocytes.
        J Cell Biochem. 2008; 104: 453-464
        • Owczarek S.
        • Berezin V.
        Neuroplastin: cell adhesion molecule and signaling receptor.
        Int J Biochem Cell Biol. 2012; 44: 1-5
        • Owczarek S.
        • Kiryushko D.
        • Larsen M.H.
        • Kastrup J.S.
        • Gajhede M.
        • Sandi C.
        • et al.
        Neuroplastin-55 binds to and signals through the fibroblast growth factor receptor.
        FASEB J. 2010; 24: 1139-1150
        • Sakaguchi M.
        • Murata H.
        • Aoyama Y.
        • Hibino T.
        • Putranto E.W.
        • Ruma I.M.
        • et al.
        DNAX-activating protein 10 (DAP10) membrane adaptor associates with receptor for advanced glycation end products (RAGE) and modulates the RAGE-triggered signaling pathway in human keratinocytes.
        J Biol Chem. 2014; 289: 23389-23402
        • Sakaguchi M.
        • Murata H.
        • Yamamoto K.
        • Ono T.
        • Sakaguchi Y.
        • Motoyama A.
        • et al.
        TIRAP, an adaptor protein for TLR2/4, transduces a signal from RAGE phosphorylated upon ligand binding.
        PloS One. 2011; 6: e23132
        • Sakaguchi M.
        • Watanabe M.
        • Kinoshita R.
        • Kaku H.
        • Ueki H.
        • Futami J.
        • et al.
        Dramatic increase in expression of a transgene by insertion of promoters downstream of the cargo gene.
        Mol Biotech. 2014; 56: 621-630
        • Schiopu A.
        • Cotoi O.S.
        S100A8 and S100A9: DAMPs at the crossroads between innate immunity, traditional risk factors, and cardiovascular disease.
        Mediators Inflamm. 2013; 2013: 828354
        • Smalla K.H.
        • Matthies H.
        • Langnase K.
        • Shabir S.
        • Bockers T.M.
        • Wyneken U.
        • et al.
        The synaptic glycoprotein neuroplastin is involved in long-term potentiation at hippocampal CA1 synapses.
        Proc Natl Acad Sci USA. 2000; 97: 4327-4332
        • Vogl T.
        • Tenbrock K.
        • Ludwig S.
        • Leukert N.
        • Ehrhardt C.
        • van Zoelen M.A.
        • et al.
        Mrp8 and Mrp14 are endogenous activators of Toll-like receptor 4, promoting lethal, endotoxin-induced shock.
        Nature Med. 2007; 13: 1042-1049
        • Wajant H.
        • Grell M.
        • Scheurich P.
        TNF receptor associated factors in cytokine signaling.
        Cytokine Growth Factor Rev. 1999; 10: 15-26
        • Zheng T.
        • Yu J.
        • Oh M.H.
        • Zhu Z.
        The atopic march: progression from atopic dermatitis to allergic rhinitis and asthma.
        Allergy Asthma Immunol Res. 2011; 3: 67-73