2.1 Tissues

Skin tissues without inflammation were obtained from Sapporo Medical University Hospital. Donors were a 3-year-old girl, a 32-year-old man, and a 50-year-old woman. All tissues were obtained with written informed consent according to the guidelines of the Declaration of Helsinki and with approval of the Institutional Review Board (IRB) of Sapporo Medical University Hospital (permit number 292-137; entitled “protein expression analysis in normal skin tissue”).

2.2 Cell culture and stimulation

Normal human epidermal keratinocytes (NHEKs) were purchased from Kurabo Industries Ltd. NHEKs were derived from the forehead skin of two Caucasian male donors under the age of 1 year. NHEKs derived from a 3-year-old female Japanese donor were isolated in our laboratory.²² The cells were cultured as a monolayer in keratinocyte-SFM supplemented with 5 ng/ml human recombinant epidermal growth factor 1–53 and 50 μg/ml bovine pituitary extract (Gibco) at 37°C in a humidified atmosphere with 5% CO₂. NHEKs were maintained for three or four passages in these conditions.

We defined the pre-differentiation stage as a semiconfluent monolayer of keratinocytes cultured in keratinocyte-SFM (low calcium). Keratinocyte differentiation was induced by calcium supplementation (1.0 mM).²³ The differentiating and differentiated stages were designated as NHEKs cultured in the differentiation conditions for 3 and 7 days, respectively. To develop culture of NHEKs, cells were seeded at a density of 5.0 × 10⁵ cells in a 12-mm-diameter polyester membrane with a pore size of 0.4 μm (Corning Costar) in the three-dimensional (3D) culture medium (keratinocyte-SFM:Dulbecco's modified Eagle's medium, 1:1) as previously reported.²⁴ After the cells had grown to complete confluence, medium in the apical compartment was removed to allow the cells to differentiate. NHEK stimulation was initiated when the apical medium was removed.

For cell stimulation, culture medium was supplemented with polyinosine–polycytidylic acid (poly I:C; Novus Biologicals), IL-4 (PeproTech), IL-13 (PeproTech), and/or IL-22 (PeproTech). The concentrations of each reagents are described in the figures or figure legends.

2.3 Small interfering RNA (siRNA) transfection

Human TP63 (ΔNp63)-specific siRNA was obtained from Invitrogen (Carlsbad; sense: 5ʹ-ACAAUGCCCAGACUCAAUU-3ʹ; antisense: 5ʹ-AAUUGAGUCUGGGCAUUGU-3ʹ). Scrambled sequence siRNA for the negative control was purchased from Invitrogen. Transfections were performed using Lipofectamine RNAiMAX (Invitrogen) in Opti-MEM (GIBCO) at 40 nmol/L according to the manufacturer's instructions. The culture medium was replaced at 6 h after siRNA transfection. Keratinocyte used in siRNA transfection experiment was at proliferation phase.

2.4 Complementary DNA (cDNA) microarray analysis

Messenger RNA (mRNA) was extracted from NHEKs transfected with TP63 (ΔNp63)-specific or scramble sequence siRNA at 72 h after transfection. Microarray slides were scanned using a 3D-GENE human 25k (TORAY) and microarray images were automatically analyzed with AROSTM, version 4.0 (Operon Biotechnologies). The result has been deposited in NCBIs Gene Expression Omnibus (GEO; http://www.ncbi.nlm.nih.gov/geo/) and are accessible through GEO Series accession number GSE162900.

2.5 Antibodies

The following antibodies were used: mouse anti-p63 monoclonal antibody (mAb) (clone: 4A4; Abcam), mouse anti-ΔNp63 mAb (clone: BC28; Biocare Medical), mouse anti-β-actin mAb (clone: AC-15; Sigma-Aldrich), mouse anti-claudin-4 mAb (clone: 3E2C1; Invitrogen), mouse anti-filaggrin mAb (clone: AKH1; Santa Cruz Biotechnology), mouse anti-caspase-14 mAb (clone: EPR12927; Abcam), mouse anti-Ki67 mAb (clone MIB-1; Dako), rabbit anti-claudin-1 mAb (clone: EPRR18871; Abcam), and rabbit anti-IL-33 mAb (clone: EPR20417; Abcam). Peroxidase-conjugated goat anti-mouse and anti-rabbit IgGs were purchased from Kirkegaard & Perry Lab Inc.

2.6 Polymerase chain reaction (PCR)

Total NHEK RNA was purified using RNeasy Mini Kit and RNase-free DNase (Qiagen) according to the manufacturer's protocol. Extracted total RNA was reverse-transcribed into cDNA using a RevertAid RT Kit containing random hexamer primers (Thermo Fisher Scientific). Quantitative PCR was performed with target gene-specific primers (Invitrogen) and SYBR green PCR master mix (Thermo Fisher Scientific) using a QuantStudio 3 Real-Time PCR System (Applied Biosystems) according to the manufacturer's instructions. Target gene expression was detected with primers (see Table S1). The amount of mRNA of elongation factor 1α was used to standardize the quantities of each transcript. The  method was used to calculate the relative gene expression level of triplicate specimens.

2.7 Immunohistochemical analysis

Immunohistochemical procedures were performed as previously described.²¹ Briefly, sections of formalin-fixed paraffin-embedded skin tissue or 3D-cultured NHEKs were immunostained using mAbs after epitope retrieval by Target Retrieval Solution pH 9 (Dako). ΔNp63, filaggrin, caspase-14, claudin-1, claudin-4, and Ki-67 proteins were detected using specific antibodies at the concentrations recommended by the manufacturers.

2.8 Western blot analysis

Cultured NHEKs were lysed with radioimmunoprecipitation assay buffer containing protease inhibitors (Roche) for 30 min on ice. Aliquots of the supernatants under reducing conditions were applied to SDS-5%–20% polyacrylamide gels (SuperSep™; FUJIFILM Wako Pure Chemical) and transferred onto polyvinylidene fluoride membranes (Millipore). The membranes were incubated with optimally diluted antibodies for 1 h at room temperature. After washing, the membranes were reacted with a peroxidase-labeled secondary antibody for 45 min. Signals were visualized using an enhanced chemiluminescence detection system (Amersham Life Science). The levels of intensity of signals detected in immunoblots were quantified using NIH ImageJ software and standardized to the corresponding levels of β-actin.

2.9 Enzyme-linked immunosorbent assay (ELISA)

Cells were cultured at a density of 3 × 10⁵ cells/well in a 12-well plate. Culture supernatants were collected at 48 h after the start of the experiment and analyzed in triplicate. Protein concentrations were determined with Human DUOset ELISA Kit (R&D Systems) for IL-1β according to the manufacturer's instructions. Absorbance was measured using an iMark microplate absorbance reader (Bio-Rad).

2.10 Statistical analysis

Data analysis was performed with Prism Version 6 software (GraphPad Software). Statistical significance was evaluated using the two-tailed unpaired Student's t test. Results are presented as mean ± SD; each set of results shown is representative of at least three separate experiments.

3.1 Distribution of ∆Np63 and AD-related barrier proteins in healthy human epidermis

To investigate the expression profile of ∆Np63 and AD-related barrier proteins, we examined the histopathology of skin tissues unaffected by inflammatory skin diseases, including AD (Figure 1A). ∆Np63 was expressed in the basal to spinous layers, whereas ∆Np63 expression was lost in the granular layer. In contrast, expression of claudin-4 and filaggrin was limited to the granular layer. Although claudin-1 and caspase-14 were detected in whole epidermis, their expression levels were slightly higher in the upper spinous layer compared with the basal layer.

Thus, we investigated the expression of these proteins in the cultured primary NHEK monolayer. As described above, NHEKs were cultured under two different conditions: “pre-differentiated” and “differentiated” conditions are models of basal and apical side keratinocytes, respectively. In the Western blot analysis, ∆Np63 expression level was higher in pre-differentiated keratinocytes compared with differentiated keratinocytes. In contrast, the barrier proteins claudin-1, claudin-4, pro-filaggrin, and pro-caspase-14 showed an inverse pattern compared with ∆Np63 (Figure 1B; mature filaggrin or caspase-14 were not detected in the two-dimensional culture system). These results were also confirmed by evaluating mRNA expression levels (Figure 1C).

3.2 ∆Np63 controls expression of barrier- and inflammation-related proteins

Based on the abovementioned inverse expression patterns between ∆Np63 and some barrier proteins, we hypothesized that ∆Np63 affects the expression of AD-related proteins. Indeed, we previously reported that expression levels of claudin-4 and TSLP receptor were modulated by ∆Np63 in HaCaT keratinocytes.^{19, 25} In this study, NHEKs expressing abundant ΔNp63 were transfected with TP63 (ΔNp63)-specific siRNA, which successfully downregulated ΔNp63 expression at the transcript and protein levels (Figure S1A,B). We could not find significant morphological alteration of the keratinocyte between these conditions. To investigate the transcriptional target of ΔNp63 in NHEKs, TP63 (ΔNp63) or control siRNA-transfected NHEKs were subjected to cDNA microarray expression analysis. As shown in Tables S2 and S3, ΔNp63 downregulated or upregulated approximately 1200 transcriptional gene expression levels by more than twofold. Interestingly, ΔNp63 negatively regulated the expression of the barrier-related molecules CASP14, CLDN1, and CLDN4, which encode caspase-14, claudin-1, and claudin-4, respectively. In addition to barrier-related genes, ΔNp63 positively regulated IL1B, which encodes the pleiotropic pro-inflammatory cytokine IL-1β. We then confirmed the expression of these genes by quantitative PCR (Figure 2A). Gene expression of FLG and IL33, which encode filaggrin and IL-33, respectively, were also modulated by ΔNp63; however, they were not detected in the microarray analysis (Figure 2A). At the protein level, expression of pro-filaggrin, pro-caspase-14, claudin-1, and claudin-4 were negatively modulated by ΔNp63, whereas ΔNp63 positively regulated IL-33 in the Western blot analysis (Figure 2B). According to a previous study, IL-1β release from NHEKs is induced by poly (I:C) stimulation.²⁶ As expected, ΔNp63 knockdown NHEKs released less IL-1β upon poly (I:C) stimulation (Figure 2C).

3.3 IL-13 inhibited downregulation of ∆Np63 expression during differentiation of keratinocytes

A growing body of studies have demonstrated that abundant IL-13, a representative cytokine constituting type 2 inflammatory environment, derived from ILC2 or Th2 plays a pivotal role in the pathogenesis of AD.²⁷ On the contrary, accumulating evidence has demonstrated disturbed keratinocyte differentiation in AD.^{28, 29} Therefore, we investigated how IL-13 affects ΔNp63-mediated keratinocyte differentiation. We treated NHEKs with IL-13. Morphologically, IL-13-stimulated NHEKs showed a smaller and tightly packed polygonal shape compared with control NHEKs (Figure 3A). Next, we investigated TP63 (ΔNp63) gene expression in response to IL-13 at various stages of differentiation of NHEKs. Of interest, the gene expression of TP63 (ΔNp63) was slightly, but significantly, upregulated at all stages of differentiation (Figure 3B). IL-13 exposure during the 7 days of differentiation inhibited attenuation of ΔNp63 expression, which was confirmed by Western blot analysis (Figure 3C). IL-4, another type 2 cytokine sharing IL-4Rα, also inhibited the downregulation of ΔNp63 (Figure 3D). However, transient IL-13 stimulation at the “pre-differentiated” or “differentiated” stage did not increase protein expression of ΔNp63 (Figure S2A,B). IL-22, which is also considered to contribute to AD pathophysiology, did not affect gene or protein expression levels of ΔNp63 (Figure S3A,B).

3.4 IL-13 exposure increased the thickness of 3D-cultured human NHEKs

To investigate the morphological influence of prolonged stimulation with IL-13, we stimulated the 3D NHEK culture with IL-13 (Figure 4A). Of interest, IL-13-exposed NHEK culture was statistically significantly thicker compared with the control (Figure 4B,C). Both keratinized and non-keratinized NHEK layers were thickened (Figure S4A,B). To investigate proliferation potency of NHEK, we performed Ki-67 expression. While Ki-67 expression was limited in the basal layer of the control group, in IL-13-stimulated condition, parabasal NHEK also expressed Ki-67 (Figure S4C). The positive rate of Ki-67 in NHEK was significantly increased in IL-13-exposed group (Figure S4D). Similarly, IL-22-exposed NHEKs were also hypertrophic (Figure S3C–E). However, although IL-13-stimulated NHEKs preserved the granular layer, this layer was lost in IL-22-stimulated NHEKs. In addition, IL-22-exposed NHEKs showed parakeratosis.

Next, we investigated ΔNp63 expression in the 3D-cultured human NHEKs with or without IL-13. Signal intensity of ΔNp63 in the basal layer did not differ between groups. However, we found increased number of decreased but still preserved ΔNp63 expression (ΔNp63^low cell in Figure 4D) in the nucleus of IL-13-stimulated NHEKs in the upper layer of stratification, where ΔNp63 expression should have ceased and the nuclei of keratinocytes should have degraded under normal conditions. If there are more cells in the field, then accordingly there are more ΔNp63^low cells. Therefore, we investigated the ratio of ΔNp63^low nuclei to total number of ΔNp63⁺ nuclei, which was significantly higher in IL-13 stimulation group (Figure 4E).

3.5 IL-13 modulated the expression of ∆Np63-controlled barrier and inflammation-related proteins

We investigated how IL-13 exposure affects the ∆Np63-controlled barrier and inflammation-related molecules during keratinocyte differentiation. As expected, in Western blot analysis, expression levels of filaggrin, caspase-14 were lower, and IL-33 expression level was higher in the IL-13-stimulated group compared with the control group (Figure 5A). Furthermore, release of IL-1β was increased in response to IL-13 after stimulation with poly (I:C) (Figure 5B). However, claudin-1 expression was increased in response to IL-13 exposure (Figure 5A). In addition, IL-13 stimulation did not alter claudin-4 expression (Figure 5A). Next, we confirmed that expression of filaggrin and caspase-14 was decreased by IL-13 exposure in the 3D-cultured NHEKs. The expression patterns of filaggrin and caspase-14 in control groups were similar to those seen in human epidermis. Levels of these proteins were decreased in the IL-13-exposed keratinocytes (Figure 5C).