DNFB was purchased from Sigma-Aldrich (St. Louis, MO, United States). Anti-mouse and anti-rabbit IgG horseradish peroxidase (HRP)-conjugated antibodies were purchased from Thermo Pierce (Rockford, IL, United States). All other antibodies used in this study were purchased from Santa Cruz Biotechnology (Dallas, TX, United States). The jetPRIME transfection reagent was purchased from Polyplus (NY, United States). Oligonucleotides and primers used in this study were commercially synthesized by Bioneer Company (Daejeon, South Korea).

Female BALB/C mice that were five week old were purchased from Nara Biotech (Seoul, South Korea). Animal experiments were approved by Institutional Animal Care and Use Committee (IACUC) of Kangwon National University (KW180823-1).

Isolations of mast cells and macrophages were performed according to the standard procedures with slight modifications (Eom et al., 2014a). Cultures were maintained in 5% CO₂ at 37°C. Human keratinocyte HaCaT cells were purchased (HDFa lot #1780051, Gibco, Grand Island, NY, United States) and expanded in Dulbecco's modified eagle medium (Gibco, Grand Island, NY, United States) containing 8% fetal bovine serum (Gibco, Mulgrave Victoria, Australia) at 37°C with 5% CO₂.

The microRNA (miRNA) expression analysis was performed by using miRNA Array III kit (Signosis, CA, United States) following the manufacturer's instructions.

TargetScan program identified targets of miR-183-5p.

Total miRNA was isolated with the miRNeasy Micro Kit (Qiagen, CA, United States). The extracted miRNA was reverse transcribed using a miScript II RT Kit (Qiagen, CA, United States). The expression level of miR-183-5p was quantified with SYBR Green Master Mix (Qiagen, CA, United States) Relative expression of miRNA was calculated using the 2^-ΔΔCT method (ΔCT = CT_miR − CT_reference).

For miR-183-5p knockdown, cells were transfected with 10 nM oligonucleotide (inhibitor) with Lipofectamine 2000 (Invitrogen) The sequences used were 5′-AGUGAAUUCUACCAGUGCCAUA-3′ (miR-183-5p inhibitor) and 5′-TAACACGTCTATACGCCCA-3′ (control inhibitor).

The pGL3 3′-UTR-BTG1 construct was made by cloning a 947-bp mouse BTG1 gene segment encompassing 3′-UTR into the XbaI site of pGL3 luciferase plasmid. The mutant pGL3 3′-UTR-BTG1 construct was made with the Quick Change site-directed mutagenesis kit (Stratagene). pFlag-BTG1construct was made by PCR amplification and cloning into Flag-tagged pcDNA3.1 vector.

MiR-183-5p promoter sequences, specific primers of miR-183-5p promoter-1 sequences [5′- GGCCCAGAATCTACTGATAGTG -3′ (sense) and 5′- TAAGTCTCTCTGGAGCTGGTG -3′ (antisense)], miR-183-5p promoter-2 sequences [5′- CACCAGCTCCAGAGAGACTT -3′ (sense) and 5′- AGAGGCCCAGAAGGTAAGAC -3′ (antisense)], miR-183-5p promoter-3 sequences [5′- GTCTTACCTTCTGGGCCTCT -3′ (sense) and 5′- GACTGATTTCTTGGGTTTGCAG -3′ (antisense)], and miR-183-5p promoter-4 sequences [5′- AGCCCCGTCTTTCTCCTT -3′ (sense) and 5′- CAGACCCTACAGAGAGGTCA -3′ (antisense)] were used for detection of binding of NF-κB to the promoter sequences of miR-183-5p.

Skin samples were fixed with 10% neutral buffered formalin, and embedded in paraffin. Sections (5 μm thickness) were stained with hematoxylin and eosin (H&E) or toluidine blue for leukocyte infiltration or mast cell infiltration and degranulation, respectively.

Immunohistochemical staining of tissues was performed using an established avidin-biotin detection method (Vectastain ABC kit, Vector Laboratories Inc., Burlingame, CA, United States). The following primary antibodies were used: anti-HDAC3 (1:100, Santa Cruz Biotechnology); anti-MCP1 (1:100, Invitrogen); anti-CD163 (1: 700, Abcam); anti-TSLP (1:500, Abcam); anti-BTG1 (1:200, Abcam), and anti-inducible nitric oxide synthase (iNOS) (1: 500, Santa Cruz Biotechnology). After washing, biotinylated secondary antibody was applied at 1:100 or 1:200 dilutions for 1 h. Color was developed with diaminobenzidine (Vector Laboratories, Inc.). Sections were counterstained with Mayer's hematoxylin. The sections incubated without primary antibody served as controls. To visualize tissue mast cells, the sections were stained with 0.1% toluidine blue (Sigma) in 0.1 N HCl for 15 min.

Cells were fixed with 4% paraformaldehyde (v/v) for 10 min and then permeabilized with 0.4% Triton X-100 for 10 min. Cells were incubated with primary antibody specific to CD163 (1:100; Abcam), iNOS (1:100; Santa Cruz Biotechnology), or NF-κB (1:200, Cell signaling Technology) for 2 h. Anti-rabbit Alexa Fluor 488 (for detection of iNOS) or anti-goat Alexa Fluor 546 (for detection of CD163 and NF-κB) secondary antibody (Molecular Probes) was added to cells and incubated for 1 h. Fluorescence images were acquired using a confocal laser scanning microscope and software (Fluoview version 2.0) with a X 60 objective (Olympus FV300, Tokyo, Japan).

Symptoms of atopic dermatitis (AD) were induced by using DNFB, as previously described (Kwon et al., 2015). Twenty-four hours after shaving, 150 μl of 1% DNFB in acetone: olive oil mixture (3:1 vol/vol) was topically applied on days 1 through 8. Later, the same dose of 0.2% DNFB was applied four times a week. Scores of 0 (none), 1 (mild), 2 (moderate), and 3 (severe) were given for each of the four symptoms: dryness, excoriation, erosion, and erythema and edema. Clinical severity represents the sum of the individual score.

The BALB/C mice were given an intradermal injection of 2, 4-dinitrophenol (DNP)-specific IgE (0.5 μg/kg). After 24 h, the mice were intravenously injected with DNP-human serum albumin (HSA) (250 μg/kg) and 2% (v/v) Evans blue solution. After 30 min of DNP-HSA challenge, the mice were euthanized, and the 2% (v/v) Evans blue dye was extracted from each dissected ear in 700 μl of acetone/water (7:3) overnight. The absorbance of Evans blue was measured at 620 nm. To determine the effect of miR-183-5p on the passive cutaneous anaphylaxis (PCA), BALB/C mice were given an intradermal injection of IgE and intravenous injection of the indicated inhibitor (each at 100 nM). The next day, BALB/C mice were given an intravenous injection of PBS or DNP-HSA along with Evans blue solution for determining the extent of vascular permeability accompanied by PCA.

To determine effect of miR-183-5p on passive systemic anaphylaxis, BALB/C mice were intravenously injected with control inhibitor (100 nM) or miR-183-5p inhibitor (100 nM). The next day, BALB/C mice were given an intravenous injection of IgE (0.5 μg/kg). The next day, BALB/C mice were intravenously injected with DNP-HSA (250 μg/kg). Changes in rectal temperatures were monitored by using a digital thermometer.

Data were analyzed and graphed using the GraphPad Prism statistics program (Version 7, GraphPad Prism software). Results are presented as means ± SE. One–way ANOVA was carried out when multiple comparisons were evaluated. Values were considered to be significant at p value less than 0.05.

To identify the chemical compounds that inhibit allergic inflammations, we screened chemical library, consisting of 1019 chemicals derived from natural products. We identified several compounds that prevented antigen (DNP-HSA) from increasing β-hexosaminidase activity in rat basophilic leukemia cells (RBL2H3). Among these compounds, homoharringtonine (HHT) showed the most potent effect on β-hexosaminidase activity (data not shown). HHT, in a concentration-dependent manner, prevented antigen from increasing levels Histone deacetylase 3 (HDAC3), Suppressor of cytokine signaling 1 (SOCS1) and Transglutaminase II (TGaseII) in RBL2H3 cells ( Figure 1A ). HHT showed the same effects on these molecules in lung mast cells ( Figure 1A ). Interactions of Fc epsilon receptor 1 (FcϵRI) with Src family kinase Lyn and HDAC3, induced by antigen, were inhibited by HHT ( Figure 1B ). The increased β-hexosaminidase activity, increased by antigen, was also inhibited by HHT ( Figure 1C ). The negative regulatory role of HHT in in vitro allergic reactions has not been reported.

HHT exerted a negative effect on passive cutaneous anaphylaxis (PCA) employing BALB/C mouse ( Figure 2A ). HHT prevented antigen (DNA-HSA) from increasing β-hexosaminidase activity ( Figure 2B ). HHT exerted negative effects of DNP-HSA on hallmarks of allergic inflammation and interactions of FcϵRI with Lyn, HDAC3, and SOCS1 in a mouse model of PCA ( Figure 2C ). HHT prevented DNP-HSA from decreasing rectal temperature in BALB/C mouse model of passive systemic anaphylaxis (PSA) ( Figure S1A ). HHT prevented antigen from increasing β-hexosaminidase activity ( Figure S1B ) and levels of hallmarks of allergic inflammations ( Figure S1C ), and from inducing interactions of FcϵRI with Lyn and SOCS1 ( Figure S1C ). Immunohistochemical staining showed that HHT prevented antigen from increasing expression of HDAC3 ( Figure S1D ).

AD shares common molecular features with anaphylaxis (Kim et al., 2018). We examined the effect of HHT on AD. HHT suppressed clinical symptoms associated with AD induced by 2, 4,-dinitrofluorobenzene (DNFB) ( Figure 3A ). HHT exerted negative effects on the increased β-hexosaminidase activity ( Figure 3B ) and the amount of histamine released by DNFB ( Figure 3C ). HHT suppressed the effects of AD on hallmarks of allergic inflammations, such as thymic stromal lymphopoietin protein (TSLP), Cyclooxygenase 2 (COX2), Monocyte Chemoattractant Protein-1 (MCP1), and CXC chemokine ligand 10 (CXCL10) ( Figure 3D ). TSLP is a marker of AD. AD is accompanied by the increased expression of Th1 cytokines such as CXCL10 (Brunner et al., 2017). HHT suppressed the effects of AD on the expression levels of CD163 and iNOS, markers of M1 (classical) and M2 macrophages (alternative), respectively ( Figure 3D ), and the interactions of FcϵRI with Lyn, HDAC3, and SOCS1 ( Figure 3D ). HHT exerted negative effects on molecular features associated with AD ( Figure 3E ). AD increased levels of T-bet and GATA, but decreased level of Forkhead Box p3 (FoxP3) ( Figure 3F ). HHT prevented AD from increasing expression levels of HDAC3 and MCP1 ( Figure S2A ), regulating expression levels of inducible Nitric oxide synthase (iNOS) and CD163 ( Figure S2A ), inducing epidermal hyperplasia ( Figure S2B ) and increasing number of activated mast cells ( Figure S2B ). We examined whether DNFB, just like DNP-HSA, would induce features of allergic inflammation. DNFB induced features of in vitro allergic reactions ( Figure S3 ). HHT suppressed the effects of DNFB on features of in vitro allergic reactions ( Figure S3 ). Thus, HHT suppresses clinical symptoms of AD by regulating molecular features of allergic inflammation.

Increased expression levels of T_H1 (IFN-γ), T_H2 (IL-31), and T_H17/T_H22 (IL-23p19/IL-8/S100A12) mRNA expression are molecular features of human AD (Guttman-Yassky et al., 2019). AD increased expression levels of Th1 cytokines such as Interleuikin-1β (IL-1β), Interferon-γ (IFN-γ) and Tumor necrosis factor- α (TNF-α) ( Figure S4 ). Expression levels of Th2 cytokines, such as IL-4, IL-5, IL-6, and IL-13, were also increased by AD ( Figure S4 ). AD decreased expression levels of FoxP3 and IL-10 ( Figure S4 ). HHT suppressed the effects of AD on the expression levels of these cytokines. Thus, HHT suppresses AD by regulating expression levels of inflammatory cytokines.

The roles of miRNAs in allergic inflammations have been reported (Nimpong et al., 2017; Kim et al., 2019). MiRNA array analysis was employed to identify miRNAs that were regulated by HHT in RBL2H3 cells ( Figure 4A ). Among these, miR-183-5p was significantly upregulated by antigen and decreased by HHT in RBL2H3 cells ( Figure 4A ). MiR-183-5p is closely associated with occupational asthma (OA) (Lin et al., 2019). HHT exerted negative effect on the increased expression of miR-183-5p by antigen ( Figure 4B ). MiR-183-5p inhibitor suppressed the effects of antigen on the β-hexosaminidase activity ( Figure 4C ) and hallmarks of allergic inflammation ( Figure 4D ). MiR-183-5p inhibitor suppressed antigen-induced interactions involving FcϵRI ( Figure 4D ). Thus, miR-183-5p is necessary for in vitro allergic reactions.

MiR-183-5p inhibitor suppressed antigen-induced PCA ( Figure 5A ). PCA increased the expression of miR-183-5p ( Figure 5B ). MiR-183-5p inhibitor suppressed the effects of PCA on β-hexosaminidase activity and the amount of histamine released ( Figure 5B ). MiR-183-5p suppressed the effects of PCA on the expression levels of CD163 and iNOS, and interactions involving FcϵRI ( Figure 5C ). MiR-183-5p inhibitor suppressed the effects of antigen on rectal temperatures ( Figure S5A ) and β-hexosaminidase activity ( Figure S5B ) in BALB/C mouse model of PSA ( Figure S5A ). PSA increased the expression of miR-183-5p ( Figure S5B ). MiR-183-5p inhibitor suppressed the effects of PSA on the hallmarks of allergic inflammation, CD163, and iNOS ( Figure S5C ). MiR-183-5p inhibitor negatively regulated effect of PSA on interactions involving FcϵRI ( Figure S5C ). Thus, miR-183-5p mediates anaphylaxis.

MiR-183-5p inhibitor attenuated clinical symptoms associated with of AD ( Figure 6A ). MiR-183-5p inhibitor suppressed the effects of DNFB on β-hexosaminidase activity ( Figure 6B ), the hallmarks of allergic inflammation and CD163, and iNOS ( Figure 6C ). MiR-183-5p inhibitor suppressed the effect of DNFB on inducing interactions of FcϵRI with HDAC3, Lyn, and SOCS1 ( Figure 6C ). AD increased expression levels of IL-1β, IL-2, IL-5, IL-6, IL-8, IL-17, and IFN-γ in BALB/C mouse in a miR-183-5p-dependent manner ( Figure 6D ). MiR-183-5p inhibitor prevented DNFB from increasing expression levels of HDAC3, MCP1, and CD163, and decreasing expression of iNOS ( Figure S6A ). Toluidine blue staining showed that DNFB increased the number of activated mast cells in a miR-183-5p-dependent manner ( Figure S6B ). H&E staining showed that miR-183-5p inhibitor suppressed the effect of DNFB on skin hyperplasia ( Figure S6B ). Thus, miR-183-5p mediates AD by regulating molecular and cellular features of AD.

HHT binds to NF-κB repressing factor (Chen et al., 2019). AD activates NF-κB signaling in Nc/Nga mouse model (Sung and Kim, 2018). PCA activates NF-κB signaling (Kang et al., 2019). Mitogen activated protein kinase (MAPK), PI3K/AKT, and NF-κB function as downstream signaling pathways of FcϵRI (Ye et al., 2017). We hypothesized that NF-κB would be involved in the expression regulation of miR-183-5p. HHT and miR-183-5p inhibitor suppressed the effect of DNFB on the expression of NF-κB in BALB/C mouse model of AD ( Figure S7A ). HHT also prevented DNFB from increasing expression of NF-κB in RBL2H3 cells ( Figure S7A ). Toll-like receptor 2 (TLR2) activates mast cells and mediates AD (Tsurusaki et al., 2016). NF-κB signaling mediates TLR2 ligand–mediated skin inflammation (Jiao et al., 2016). DNFB increased the expression of TLR2 in BALB/C mouse and RBL2H3 cells ( Figure S7A ). HHT and miR-183-5p inhibitor suppressed the effect of DNFB on the expression of TLR2 in BALB/C mouse and RBL2H3 cells ( Figure S7A ). BAY11-7082, an inhibitor of NF-κB, prevented DNFB from inducing nuclear translocation of NF-κB in RBL2H3 cells ( Figure S7B ). NF-kB was shown to bind to the promoter sequences of miR-183-5p ( Figure S7C ). Thus, NF-κB and miR-183-5p form a positive feedback loop and mediate AD.

BTG1 was predicted to be a target of miR-183-5p based on TargetScan analysis. Luciferase activity assays showed the direct expression regulation of BTG1 by miR-183-5p ( Figure S8A ). HHT and miR-183-5p inhibitor suppressed the effect of DNFB on the expression of BTG1 in BALB/C mouse model of AD ( Figure S8B ). HHT prevented antigen from decreasing the expression of BTG1 in RBL2H3 cells ( Figure S8C ). BTG1 suppressed the effects of antigen on hallmarks of allergic inflammation ( Figure 7A ), interactions involving FcϵRI ( Figure 7A ), and β-hexosaminidase activity and the expression of miR-183-5p ( Figure 7B ). DNFB decreased the expression of BTG1 in HaCaT cells ( Figure 7C ). BTG1 prevented suppressed the effects of DNFB on the hallmarks of allergic inflammation and miR-183-5p in HaCaT cells ( Figure 7C ). Thus, BTG1acts as a negative regulator of in vitro allergic reactions.

TLR4-NF-κB signaling is necessary for maintaining epithelial barrier in allergic inflammation (Tao et al., 2017). BAY11-7082, an inhibitor of NF-κB, suppressed the effects of antigen on the expression levels of NF-κB, phospho I kappa B^{Ser 32} (pIkB^Ser32), TLR2, and TLR4 ( Figure 8A ), and BTG1 in RBL2H3 cells ( Figure 8A ). BAY11-07082 also inhibited antigen-induced interactions involving FcϵRI ( Figure 8A ). BAY11-7082 suppressed the effect of antigen on the level of miR-183-5p and β-hexosaminidase activity ( Figure 8B ). BAY11-7082 rather increased the level of BTG1 mRNA in unstimulated RBL2H3 cells and prevented antigen from decreasing the level of BTG1 mRNA ( Figure 8B ). BAY11-7082 prevented antigen from inducing nuclear translocation of NF-κB ( Figure 8C ). BAY11-7082 suppressed the effect of DNFB on the expression levels of pIkB^Ser32, hallmarks of allergic inflammation, and miR-183-5p ( Figure 8D ). BAY11-07082 prevented DNFB from decreasing the level of BTG1 in HaCaT cells ( Figure 8D ). BAY11-7082 suppressed antigen-induced PCA ( Figure S9A ) and prevented PCA from increasing β-hexosaminidase activity and the levels of BTG1 mRNA and miR-183-5p ( Figure S9B ), increasing the expression levels of BTG1, NF-κB, and pIkBα^Ser32 ( Figure S9C ) and decreasing the expression level of BTG1 ( Figure S9C ). BAY11-7082 inhibited DNFB-induced nuclear translocation of NF-κB in HaCaT cells ( Figure S9D ).

We examined whether cellular interactions would be necessary for AD. Culture medium of skin mast cells isolated from AD-induced BALB/C mouse increased expression levels of CD163 and hallmarks of allergic inflammation, but decreased the expression of iNOS in macrophages ( Figures 9A, B ). HHT inhibited activation of macrophages by culture medium of mast cells from AD-induced BALB/C mouse ( Figures 9A, B ). Culture medium of mast cells from AD-induced BALB/C mouse decreased expression levels of BTG1 and iNOS in macrophages ( Figure S10A, C ). Culture medium of mast cells from AD-induced BALB/C mouse injected with miR-183-5p inhibitor had no significant effect on expression level of BTG1, CD163 or iNOS in macrophages ( Figures S10A, C ). Thus, HHT and miR-183-5p inhibitor exert anti-atopic effect by inhibiting cellular interactions involving mast cells and macrophages.