Repetin Regulates Epidermal Barrier Homeostasis and Is Associated With Inflammatory Skin Diseases

ABSTRACT

Repetin (RPTN) is a member of the fusion S100 protein family encoded within the epidermal differentiation complex. Although genetic studies have revealed that RPTN is a susceptibility gene for atopic dermatitis (AD), its biological function remains poorly understood. In this study, we investigated the role of RPTN in epidermal homeostasis and inflammatory skin diseases. We examined RPTN expression in normal skin, inflammatory skin diseases, and differentiated normal human keratinocytes (NHKs). Functional analyses were performed using RPTN knockdown (KD) NHK and three‐dimensional (3D) skin‐equivalent models. Epidermal barrier function was assessed using a lucifer yellow permeability assay, and cytokine‐mediated regulation of RPTN expression was evaluated using 3D atopic dermatitis (AD) and psoriasis models. RPTN was primarily expressed in the granular layer of the normal epidermis and increased with keratinocyte differentiation. In the 3D skin‐equivalent model, RPTN deficiency impaired epidermal barrier function and induced the expression of differentiation‐related genes, including FLG, IVL, TGM1, CLDN1, KLK7, ALOX12, and TCHHL1. Notably, RPTN knockdown significantly increased IL‐25 expression. RPTN expression was elevated in chronic atopic dermatitis (AD) lesions and in hypergranular epithelia of psoriasis vulgaris, lichen planus, and epidermolytic ichthyosis. In a 3D AD model, IL‐4 and IL‐13 significantly induced RPTN expression. These findings indicate that RPTN contributes to maintaining epidermal barrier homeostasis and suggest that it may be associated with regulating inflammatory responses.

1. Introduction

Epidermal differentiation is a complex process requiring the regulated and sequential expression of various genes. A family of proteins described as ‘S100 fused‐type proteins’ containing the EF‐hand domain in the N‐terminus followed by multiple tandem peptide repeats is located on chromosome 1q21.3 [1, 2]. This protein family includes cornulin (CRNN), filaggrin (FLG), filaggrin‐2 (FLG2), hornerin (HRNR), repetin (RPTN), trichohyalin (TCHH), and trichohyalin‐like 1 (TCHHL1) [3, 4, 5, 6, 7, 8, 9]. Human RPTN is a 784‐amino‐acid protein that is synthesised in the granular layer of the epidermis, acrosyringium, inner root sheath, and tongue papillae; however, it is sparsely expressed in the homeostatic epidermis [5]. RPTN was hypothesised to be a cross‐binding protein involved in cornified cell envelope formation [5]. A genome‐wide association study for atopic dermatitis (AD) demonstrated that TCHH/RPTN was associated with AD in the European and Asian populations [10]. Trzeciak et al. demonstrated that the CC phenotype of RPTN rs3001978 was associated with early age at onset, pruritus, severity, and concomitant asthma in AD patients without FLG mutations and that the level of RPTN mRNA increased in skin lesions of AD [11]. Furthermore, the expression of RPTN was induced in the epidermis of Kruppel‐like factor 4‐deficient mice or loricrin‐deficient mice, both of which were characterised by an impaired epidermal barrier function [12, 13]. These findings suggest that RPTN may be involved in epidermal barrier formation and the pathogenesis of skin diseases; however, its exact role remains unknown. We investigated the role of RPTN in the epidermis and normal human keratinocytes (NHKs), as well as its potential relevance to skin disease pathogenesis. The present study suggests that RPTN plays an important role in homeostasis of the epidermal barrier and may be associated with the pathogenesis of skin diseases.

2. Methods and Materials

2.1. Clinical Materials

Tissue samples from patients with skin diseases (AD [n = 10], psoriasis vulgaris (PV) [n = 5], lichen planus (LP) [n = 5], ichthyosis vulgaris (IV) [n = 3], epidermolytic ichthyosis (EI) [n = 2], and actinic keratosis (AK) [n = 5]) were obtained from Toyama University Hospital. Five of the AD specimens were obtained from erythematous lesions (acute lesions); the others were obtained from lichenified lesions (chronic lesions). Skin diseases were diagnosed by experienced dermatologists and pathologists based on clinical and histological findings. Skin samples from unrelated Japanese individuals were used as controls. Patients provided their written informed consent in accordance with the principles of the Declaration of Helsinki. This study was approved by the Medical Ethics Committee of the University of Toyama, Japan (approval number: R2017139).

2.2. Cells and Cell Culture

NHKs (Kurabo Industries Ltd., Osaka, Japan) were cultured in Humedia‐KG2 (Kurabo Industries Ltd) at 37°C in a humidified atmosphere under 5% CO₂. All experiments were performed on cells at the third passage. To measure cell growth rates, 5.0 × 10⁴ cells were plated onto 35‐mm diameter plates. At each time point, cells were stained with trypan blue and counted. Differentiation was induced by the addition of 1.5 mM Ca²⁺ into the medium of semi‐confluent cultured cells, and the cells were cultured for 7 days. All experiments were independently performed in triplicate.

2.3. Reagents and Antibodies

Pre‐designed small interfering RNAs (siRNAs) targeting human RPTN (#1: s43060 and #2: s43061) and a negative control siRNA were purchased from Life Technologies (Carlsbad, CA, USA). Recombinant human IL‐4 was obtained from R&D Systems (Minneapolis, MN, USA), whereas recombinant human IL‐13, IL‐17A, IL‐22, and TNF‐α were purchased from PeproTech (Cranbury, NJ, USA).

Antibodies against FLG (ab218863), TCHH (ab58755), keratin 14 (119695), keratin 10 (ab9025), keratin 6 (ab93279), IVL (ab53112), TGM1 (183351), IL‐33 (ab207737), and TSLP (ab115700) were purchased from Abcam (Cambridge, UK). Antibodies against CRNN (11799‐1‐AP) and RPTN (HPA030483) were purchased from Proteintech (Rosemont, IL, USA) and Atlas Antibodies AB (Bromma, Sweden), respectively. An antibody against IL‐25 (NBP3‐13773) was purchased from Novus Biologicals (Centennial, CO, USA). Antibodies against TCHHL1, FLG2, and HRNR were generated as described previously [6, 9, 14]. An antibody against Ki67 was purchased from DAKO (Carpinteria, CA, USA).

2.4. RNA Interference

For transient knockdown, NHKs were transfected with siRNA duplexes using Lipofectoamine RNAiMAX (Thermo Fisher Scientific Inc., Waltham, MA) according to the manufacturer's instructions. Three days post‐transfection, cells were collected and used for the experiments.

2.5. Quantitative Reverse Transcription Polymerase Chain Reaction (qRT‐PCR)

Total RNA was isolated from cells using a RNeasy Mini Kit (QIAGEN, Venlo, Netherlands) and treated with DNase 1 (QIAGEN). Reverse transcription was performed using random hexamers and Superscript III (Thermo Fisher Scientific Inc., Waltham, MA). Complementary DNA samples were analysed using SYBR Premix Ex Taq II (Takara Bio Inc., Shiga, Japan) according to the manufacturer's instructions. All experiments were performed in triplicate and the β‐actin levels were normalised. The primers used for qRT‐PCR are listed in Table 1.

TABLE 1.

Primers for quantitative RT‐PCR.

ACTB	FW	TGGCACCCAGCACAATGAA
	RV	CTAAGTCATAGTCCGCCTAGAAGCA
ALOX12	FW	GGAGCTCACCTATGACAGCCTCTAC
	RV	CAAGACTGCAATTCCGGATCAC
CASP14	FW	GCGAGCTAAGCCCAAGGTGTA
	RV	AACGTGCAAGGCATCTGTGTA
CLDN1	FW	GGGCAGATCCAGTGCAAAG
	RV	GGATGCCAACCACCATCAAG
CRNN	FW	GCCTTCAGGCGCTATGCAA
	RV	TCATCCAGCAGACGCAGGAC
FLG	FW	GGAATTTCGGCAAATCCTG
	RV	GCTTGAGCCAACTTGAATACCA
FLG2	FW	GCACACTGAGCAAGGGTGAACTAA
	RV	AGGACCTTGTTGCAGGCCATA
HRNR	FW	ACGTTGAACAAGGCAGAGCTGA
	RV	CTCGATCCAGACTTTGCAAGATGA
IL‐25	FW	AGTCCTGTAGGGCCAGTGAAGAT
	RV	GCCGGTTCAAGTCTCTGTCCA
IL‐33	FW	GTTGCATGCCAACAACAACAAGGA
	RV	GCATTCAAATGAAACACAGTTGG
IVL	FW	GCTGGAGCAGCCTGTGTTTG
	RV	CTGGACACTGCGGGTGGTTA
K10	FW	TTTAGCCGTGGGAGCTGTG
	RV	CCACCAAAGCTGCTACTTCCATA
K14	FW	AACGCCGACCTGGAAGTGA
	RV	TTGTCCACTGTGGCTGTGAGAA
K16	FW	GACCATCGAGGACCTGAGGA
	RV	AGCCTGGCATTGTCAATCTG
KLK7	FW	GCGTGGATGTCAAGCTCATCTC
	RV	CAGGGTACCTCTGCACACCAAC
RPTN	FW	GTTGA AAC AAC TGC TCT TGGCTGA
	RV	AGGCT TGG ACC AAC TGG AACAC
TCHH	FW	TGTGATGGAGCAGCATTAACTAAGA
	RV	GCCCATTACTGTCAAGATCCAGAA
TCHHL1	FW	AATTA AAGGTCCAAGGCCCAAG
	RV	GGAGGCTGAATTGTCCTCATCTA
TGM1	FW	TGGCCAGGAGTGTGAAGTACAGA
	RV	CACTGTTTCATTGCCTCCAATGTC
TSLP	FW	CGCCTATGAGCAGCCACATT
	RV	TCTTCTTGATTGCCTGAGTAGCA

2.6. Preparation of RPTN Knockdown Skin‐Equivalent Models

NHKs, which were transfected with RPTN siRNA or control siRNA, were plated in chambers (CELLnTEC, Bern, Switzerland) 1 day after transfection. Two days later, the cell surface was exposed to air and the cells were cultured for 12 days according to the manufacturer's instructions. All experiments were independently performed in triplicate.

2.7. Analysis of Lucifer Yellow Permeability

At day 12 after initiation of the organotypic skin culture, 100 μL of Lucifer Yellow (Thermo Fisher Scientific Inc., Waltham, MA) was added onto the horny layer of the organotypic skin samples and incubated for 60 min at room temperature. The skin samples were fixed in 4% paraformaldehyde for 30 min. They were directly dipped into OCT Compound (Ted Pella, Redding, CA) and rapidly frozen. Sections (10 μm) were inspected under a confocal laser microscope (LSM780, Carl Zeiss, Oberkochen, Germany).

2.8. Preparation of 3D Skin Disease Models

For the 3D AD skin model, EpiDerm was purchased from the MatTek Corporation (Ashland, MA, USA). Recombinant human IL‐4 (50 ng/mL) and IL‐13 (50 ng/mL) were added to the medium (EPI‐100: MatTek Corporation) and culturing was continued for 5 days according to a previous study.¹⁵ Similarly, to generate a 3D psoriatic skin model, EpiDerm was cultured with recombinant human TNF‐α (50 ng/mL), IL‐17A (50 ng/mL), and IL‐22 (50 ng/mL) for 5 days [15]. All experiments were independently performed in triplicate.

2.9. Western Blotting

Protein extracts were prepared as described previously [6]. Ten micrograms of protein was separated on 10% SDS‐polyacrylamide gels and transferred onto PolyScreen Transfer Membranes (NEN life science Products, Boston, MA). The membranes were treated with primary antibodies at room temperature for 1 h, and positive signals were visualised using ECL prime Western Blotting Detection Reagents (GE Healthcare UK Ltd., Buckinghamshire, England).

2.10. Immunostaining and TUNEL Assays

Paraffin‐embedded tissue sections were incubated with Dako Target Retrieval Solution (DAKO) at pH 9 and 121°C for 4 min for antigen retrieval, and subsequently blocked with Protein Block Serum‐Free (DAKO) for 15 min. Sections were then incubated with primary antibodies for 60 min. Alexa Fluor 488 goat anti‐rabbit IgG (H + L) or Alexa Fluor 555 goat anti‐mouse IgG (H + L) (Molecular Probes, Eugene, OR) were used as secondary antibodies. Thereafter, 6‐diamidine‐2′‐phenylindole dihydrochloride (DAPI) (Molecular Probes Inc.) was used to visualise the nuclei. The TUNEL reaction was performed using a fluorescein in situ cell death detection kit (Roche Diagnostics Ltd., Basel, Switzerland) according to the manufacturer's instructions. Tissue sections were observed under a fluorescence microscope (Olympus IX71; Olympus Co., Tokyo, Japan). Cultured NHKs were fixed with 4% paraformaldehyde; then, the cells were stained in the same manner. The cells were observed under a confocal laser microscope (LSM780, Carl Zeiss, Oberkochen, Germany). The ratio of Ki67‐positive or TUNEL‐positive cells to DAPI‐stained nuclei was calculated as follows: in a typical experiment, > 1500 cells in 7 fields were counted, and the results are presented as the mean ± standard deviation (SD). The experiments were independently repeated in triplicate.

2.11. Statistical Analysis

Statistical significance was evaluated using the Mann–Whitney U test. Statistical significance was set at p < 0.05. Values are expressed as the mean ± standard deviation (SD). Three independent experiments were conducted under all assay conditions.

3. Results

3.1. RPTN Expression in the Normal Skin and Cultured Keratinocytes

First, we examined RPTN expression in normal human skin obtained from various regions. Immunohistochemistry clearly revealed the expression of RPTN in the granular layer of the epidermis of the palms and soles (Figure 1A, top and middle panels). RPTN was detected in some FLG‐expressing cells (Figure 1A, right panels). In the scalp, face, and back, RPTN signals were sparsely observed although FLG was regularly detected (Figures 1A, bottom panels and S1). On immunoblotting, a band corresponding to RPTN (~90 kDa) was detected in the differentiated NHKs (Figure 1B). RPTN was distributed on presumptive keratohyalin granules in differentiated NHKs. The signals for RPTN were partially co‐localised with those of FLG; however, most RPTN was distributed in a different population from FLG (Figure 1C).

FIGURE 1.

Expression of RPTN in normal human skin and keratinocytes. (A) Normal skin tissues obtained from the palm, sole, and back were doubly immunostained for RPTN and FLG. The tissue sections were also stained with DAPI to visualise nuclei. Scale bar: 100 μm (in all panels). (B) Primary cultures of NHKs were induced to differentiate with 1.5 mM Ca²⁺. The protein extracts were prepared before, 3 and 7 days after the addition of Ca²⁺. The membranes were blotted with anti‐RPTN or anti‐β‐actin antibodies. Day 0: before the addition of Ca²⁺; Day 3: 3 days after the addition of Ca²⁺; Day 7: 7 days after the addition of Ca²⁺. (C) RPTN immunostaining of NHKs induced to undergo differentiation with 1.5 mM Ca²⁺ for 7 days. RPTN (green) was present in association with keratohyalin granules in the differentiating cells, and was partially co‐localised with FLG (red). Scale bar: 20 μm (in all panels).

3.2. Role of RPTN in the Epidermis of 3D Skin‐Equivalent Model

To investigate the function of RPTN in keratinocytes, we transfected NHK cells with RPTN siRNA or negative control siRNA. One day after transfection, 1.5 mM Ca²⁺ was added to induce keratinocyte differentiation, and cells were harvested after 7 days. Both RPTN siRNA#1 and siRNA#2 significantly reduced RPTN mRNA and protein expression compared to control siRNA (Figure S2). However, RPTN knockdown (KD) did not significantly affect the number of viable cells, Ki67‐positive cells, or TUNEL‐positive cells (Figures S3–, S5). This suggests that RPTN does not substantially affect keratinocyte proliferation or apoptosis under these conditions.

To investigate the role of RPTN in epidermal tissue formation, a three‐dimensional (3D) skin‐equivalent model was generated using RPTN‐KD NHKs. Because both siRNAs demonstrated equivalent knockdown efficiency, we used siRNA#1 in subsequent experiments. RPTN expression remained significantly suppressed throughout the culture period (Figure S6). Histologically, the 3D RPTN‐KD skin model showed a mild tendency toward epidermal thickening compared with the 3D control skin model, and RPTN expression was significantly reduced (Figure 2A,B). Analysis of barrier function revealed that in the 3D RPTN‐KD skin model, there was increased penetration of Lucifer Yellow through the horny layer into the upper spinous layer, whereas in the control model, little or no penetration was observed (Figure 2C). This indicates impaired epidermal barrier function.

FIGURE 2.

Altered expression of differentiation‐associated proteins in the 3D RPTN‐KD skin model. (A) Histological findings of 3D RPTN‐KD skin models (upper panels). Scale bar: 100 μm (original magnification 200×). 3D RPTN‐KD skin models were stained with anti‐RPTN antibodies (lower panels). The dotted lines indicate the basement membrane. Scale bar: 100 μm (original magnification 200×). (B) Relative mRNA expression levels of RPTN in 3D RPTN‐KD skin models were analysed by quantitative RT‐PCR, and normalised to the β‐actin value. All data represent the mean ± SD of three independent experiments. *p < 0.05; **p < 0.01; ***p < 0.001 versus NHKs or 3D skin models, which were transfected with control siRNA. NS, not significant. (C) Lucifer Yellow penetrated through the horny layer down to the upper spinous layer of the 3D RPTN‐KD skin models. In the 3D control skin model, Lucifer yellow was retained in the horny layer. The dotted lines indicate the basement membrane. Scale bar: 20 μm (original magnification 640×). (D) 3D RPTN‐KD skin models were stained with antibodies against differentiation‐associated proteins. The dotted lines indicate the basement membrane. Scale bar: 100 μm (original magnification 200×). (E) Relative mRNA expression levels of differentiation‐associated proteins in 3D RPTN‐KD skin models were analysed by quantitative RT‐PCR, and normalised to the β‐actin value. All data represent the mean ± SD of three independent experiments. *p < 0.05; **p < 0.01; ***p < 0.001 versus 3D skin models with control siRNA. (F) 3D RPTN‐KD skin models were stained with antibodies against keratinocyte‐driven cytokines, IL‐25, IL‐33, and TSLP. The dotted lines indicate the basement membrane. Scale bar: 100 μm (original magnification 200×). (G) Relative mRNA levels of IL‐25, IL‐33, and TSLP in 3D RPTN‐KD skin models were analysed by quantitative RT‐PCR, and normalised to the β‐actin value. All data represent the mean ± SD of three independent experiments. *p < 0.05; **p < 0.01; ***p < 0.001 versus 3D skin models with control siRNA.

To determine the effect of RPTN deficiency on epidermal differentiation, we examined the expression of differentiation‐associated proteins in the 3D RPTN‐KD skin model. Immunohistochemistry and quantitative RT‐PCR demonstrated increased expression of TGM1, IVL, and K10, whereas K14 expression was unchanged (Figure 2D,E). In addition, CLDN1, KLK7, and ALOX12 expression levels were significantly increased, while CASP14 expression was significantly decreased (Figure 2E).

We also examined the effects of RPTN deficiency on S100 fused‐type proteins. Immunohistochemistry revealed increased expression of FLG, FLG2, and TCHHL1 in the 3D RPTN‐KD model (Figure 2D). Quantitative RT‐PCR confirmed significant upregulation of FLG and TCHHL1, whereas the expression levels of other fused S100 proteins were unchanged (Figure 2D,E).

Finally, the expression of epidermis‐derived cytokines was evaluated. Among the cytokines examined, IL‐25 expression was markedly increased in the 3D RPTN‐KD skin model, as demonstrated by both immunofluorescence and quantitative RT‐PCR (Figure 2F,G). In contrast, only slight and non‐significant increases in TSLP and IL‐33 expression were observed. These findings indicate that RPTN deficiency is associated with selective induction of IL‐25 in epidermal tissue.

3.3. Expression of RPTN in Skin Diseases

To clarify the association of RPTN with the pathogenesis of skin diseases, we examined the expression of RPTN, FLG, and keratin 6 (K6) in tissue samples with skin diseases (Figures 3 and S7). In chronic AD lesions, RPTN expression was markedly increased in the granular layers. In PV, the increased expression of RPTN was observed in the hypergranular region of the epidermis but not in the parakeratotic region. RPTN was strongly expressed in the granular layers of the epidermis in LP and EI. In the epidermis of these diseases, a marked expression of K6 was observed. RPTN expression in acute lesions of AD was slightly higher than in the normal back skin, despite the absence of K6 expression. In contrast, RPTN signals were not observed in IV or AK. RPTN signals were mostly colocalised with FLG signals in atopic dermatitis, psoriasis vulgaris, and lichen planus. In epidermolytic ichthyosis, RPTN was irregularly detected in a different cell population from that expressing FLG.

FIGURE 3.

Expression of RPTN in skin diseases with hyperkeratosis. Immunostaining of RPTN in pathological skin tissue samples from patients with hyperkeratosis‐associated diseases. Tissue samples from patients with AD, psoriasis vulgaris, lichen planus, ichthyosis vulgaris, epidermolytic ichthyosis, and actinic keratosis were double immunostained for RPTN and FLG. Similar results were obtained for all six diseases. Tissue sections were stained with 4′,6‐diamidine‐2′‐phenylindole dihydrochloride (DAPI) to visualise the nuclei. Scale bar, 200 μm (in all panels).

3.4. Altered Expression of RPTN in 3D AD and Psoriatic Skin Models

To examine the effect of cytokines on RPTN expression, we generated a 3D AD skin model, in which the 3D skin‐equivalent model was stimulated with IL‐4 and IL‐13, and a 3D psoriatic skin model, in which the 3D skin‐equivalent model was stimulated with TNF‐α, IL‐17A, and IL‐22. Additionally, we generated 3D skin‐equivalent models stimulated with IL‐4, IL‐13, TNF‐α, IL‐17A, or IL‐22.

Immunohistochemistry revealed the increased expression of RPTN in the 3D AD skin model, although the FLG expression was decreased (Figure 4A). Similarly, the RPTN mRNA expression was significantly increased in the 3D AD skin model (Figure 4B). RPTN expression in the 3D IL‐4 and IL‐13 skin models showed no significant difference from that in the 3D control skin model, although the FLG expression was decreased in the 3D IL‐4 and IL‐13 skin models (Figure 4A,B). K16 expression in the 3D AD, IL‐4, and IL‐13 skin models was not significantly different from that in the 3D control skin model (Figure 4B).

FIGURE 4.

Expression of RPTN in 3D AD and psoriatic skin models. (A) 3D skin models were double‐stained with anti‐RPTN and anti‐FLG antibodies. The dotted line indicates the boundary between the granular layer and the horny layer. Scale bar: 100 μm (original magnification 400×). (B) Relative mRNA expression levels of RPTN, FLG, and K16 in 3D skin models were analysed by quantitative RT‐PCR, and normalised to the β‐actin value. All data represent the mean ± SD of three independent experiments. (C) 3D skin models were double‐stained with anti‐RPTN and anti‐FLG antibodies. The dotted line indicates the boundary between the granular layer and the horny layer. Scale bar: 100 μm (original magnification 400×). (D) Relative mRNA expression levels of RPTN, FLG, and K16 in 3D skin models were analysed by quantitative RT‐PCR, and normalised to the β‐actin value. All data represent the mean ± SD of three independent experiments. **p < 0.01; ***p < 0.001; ****p < 0.0001 versus 3D control skin models. AD, 3D AD skin model; Control, 3D control skin model; IL‐4, 3D IL‐4 skin model; IL‐13, 3D IL‐13 skin model; IL‐17A, 3D IL‐17A skin model; IL‐22, 3D IL‐22 skin model for all experiments; pso, 3D psoriatic skin model; TNF‐α, 3D TNF‐α skin model.

In the 3D psoriatic skin model, immunohistochemistry revealed the markedly decreased expression of RPTN and FLG relative to the 3D control skin model (Figure 4C). The expression of RPTN and FLG mRNAs was rarely observed in the 3D psoriatic skin model (Figure 4D). Furthermore, immunohistochemistry revealed the increased expression of RPTN in the 3D IL‐22 skin models, while it was decreased in the 3D TNF‐α and IL‐17A skin models (Figure 4C). The RPTN mRNA expression levels in the 3D IL‐22 skin models were significantly increased relative to the 3D control skin model (Figure 4D). The K16 expression in the 3D IL‐22 skin models showed a significant increase, although that in the 3D psoriatic, TNF‐α, and IL‐17A skin models showed a significant decrease relative to the 3D control skin model (Figure 4D).

4. Discussion

The present study demonstrated that RPTN plays an important role in epidermal barrier homeostasis and may be associated with the regulation of epidermis‐derived inflammatory responses. RPTN is primarily expressed in the granular layer of normal epidermis and was upregulated in hyperproliferative and inflammatory skin diseases. Furthermore, RPTN knockdown in a 3D skin model impaired epidermal barrier function and induced IL‐25 expression. This suggests that RPTN plays a previously unrecognised role in the interaction between epidermal barrier integrity and type 2 inflammatory responses.

RPTN expression was particularly prominent in the epidermis of the palms and soles, as well as in hyperproliferative skin diseases. Similar expression patterns have been reported for keratins associated with epidermal activation, such as K6, K16, and K17 [12, 16, 17]. These findings suggest that RPTN expression is associated with the epidermal stress response and the enhancement of barrier function.

RPTN deficiency impaired epidermal barrier function in the 3D skin model and induced the expression of several differentiation‐associated proteins, including FLG, IVL, TGM1, CLDN1, KLK7, and ALOX12. FLG promotes aggregation of keratin filaments [3], whereas IVL and TGM1 are involved in the formation of the cornified cell envelope [18]. CLDN1 is an essential component of epidermal tight junctions and contributes to barrier function [19]. KLK7 regulates corneocyte desquamation through cleavage of adhesion molecules [20], and ALOX12 participates in the formation of the corneocyte lipid envelope [21]. The coordinated upregulation of these genes following RPTN knockdown suggests the activation of a compensatory differentiation program aimed at restoring epidermal barrier function.

Interestingly, TCHHL1 expression was significantly increased in the 3D RPTN‐KD skin model. TCHHL1 promotes keratinocyte proliferation and inhibits apoptosis [22]. Accordingly, the induction of TCHHL1 expression may represent an adaptive response to maintain epidermal homeostasis following RPTN deficiency. These findings suggest that RPTN contributes not only to epidermal barrier formation but also to the regulation of epidermal homeostasis.

An important finding of the present study is the marked induction of IL‐25 following RPTN knockdown. IL‐25 functions as an epithelial‐derived alarmin and plays an important role in the initiation and amplification of type 2 immune responses [23]. Increased IL‐25 expression is thought to be involved in the pathogenesis of atopic dermatitis and promotes the production of type 2 cytokines, including IL‐4 and IL‐13 [23]. Therefore, the induction of IL‐25 in RPTN‐deficient epidermis suggests that RPTN deficiency may influence not only epidermal barrier integrity but also immune homeostasis. Because barrier dysfunction is a hallmark of atopic dermatitis, RPTN may be involved in the association between impaired barrier function and type 2 inflammation. Although the underlying mechanism remains to be elucidated, these results point to a previously unrecognised role for RPTN in regulating epithelium‐derived inflammatory signalling.

The clinical significance of these findings was supported by the expression patterns of RPTN in skin diseases. RPTN expression was markedly increased in chronic lesions of AD and in hypergranular epidermis of PV, LP, and EI. Increased expression of K6, K16, and K17 has also been reported in these disorders [24, 25, 26], supporting the notion that RPTN expression is associated with epidermal activation and hyperproliferation. Interestingly, a slight increase in RPTN expression was also observed in acute AD lesions although no significant epidermal hyperplasia was observed. Acute AD lesions are characterised by barrier dysfunction primarily induced by IL‐4 and IL‐13 [27]. Similar findings have been reported in KLF4‐deficient epidermis, where barrier dysfunction caused changes in the expression of RPTN [13]. The expression of RPTN in acute AD lesions may reflect stress on the epidermal barrier rather than hyperproliferation.

In contrast, RPTN expression was not detected in the epidermis of patients with IV carrying FLG loss‐of‐function mutations [28]. Furthermore, our previous study demonstrated that RPTN expression was not increased in a 3D FLG‐KD skin model [29]. These findings suggest that RPTN may be involved in a distinct regulatory pathway for maintaining the epidermal barrier, rather than simply compensating for the absence of FLG.

To further elucidate the regulatory mechanisms of RPTN, we examined cytokine‐induced changes in its expression using disease‐specific 3D skin models. In the AD model, IL‐4 and IL‐13 significantly increased RPTN expression. Previous studies have shown that IL‐4 and IL‐13 suppress OVOL1, which in turn reduces the expression of FLG, IVL, and loricrin, thereby impairing epidermal barrier function [30]. The induction of RPTN may represent a compensatory response of the epidermis to cytokine‐induced barrier dysfunction. This mechanism is thought to explain the increased expression of RPTN observed in acute AD lesions.

In a psoriatic skin model, stimulation with TNF‐α, IL‐17A, and IL‐22 decreased RPTN expression; however, IL‐22 alone increased RPTN expression in parallel with K6 expression. IL‐22 activates the STAT3 and ERK1/2 signalling pathways and induces hyperproliferation‐associated keratins [31]. Because IL‐22 is abundantly expressed in lesions of chronic AD and psoriasis [32], the expression of RPTN in chronic inflammatory skin diseases may be regulated, at least in part, through IL‐22‐mediated activation of the epidermis.

There are several limitations in this study. This study is primarily based on an siRNA knockdown model, and the mechanism linking RPTN deficiency to IL‐25 induction remains unclear. Furthermore, the functional significance of elevated IL‐25 expression in the immune response has not been directly evaluated. Further research is needed to elucidate the molecular mechanisms underlying RPTN‐mediated regulation of epidermal barrier function and inflammatory signalling.

In conclusion, the present study demonstrated that RPTN contributes to maintaining epidermal barrier homeostasis and may be involved in the regulation of the epidermal immune response. Therefore, RPTN may play a potential role in the interaction between barrier disruption and type 2 inflammatory signalling. These findings provide new insight into the biological function of RPTN and suggest that RPTN may be associated with the pathogenesis of AD and other inflammatory skin diseases.

Author Contributions

T.M. and T.S. conceptualised this study. T.M., M.M., and S.Y. were involved in the formal analysis. T.M. and M.M. were involved in funding acquisition. T.M., K.T., and S.Y. were involved in the methodology. T.M. and S.Y. contributed essential reagents and tools. T.M., K.T., S.Y., and M.M. were involved in the data analysis. T.M., S.Y., and T.S. supervised this study. T.M. and S.Y. were involved in the visualisation. T.M. and M.M. were involved in the writing – original draft. All the authors were involved in the writing – review and editing.

Funding

This work was supported by JSPS KAKENHI grant numbers JP18K08265 (to T.M.) and JP23K16051 (to M.M.).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Figure S1: Normal skin tissues obtained from face and scalp were doubly immunostained for RPTN and FLG. The tissue sections were also stained with DAPI to visualise nuclei. Scale bar: 100 μm (in all panels).

Figure S2: (A) RPTN siRNA#1 and RPTN siRNA#2 significantly suppressed the mRNA expression of RPTN 7 days post‐transfection in comparison to the control level. (B) Immunoblotting showed a marked reduction of RPTN proteins (siRNA#1: 8.3%, siRNA#2: 10.2%) compared to control siRNA 7 days post‐transfection of siRNA. β‐actin was used as a loading control.

Figure S3: (A) Representative micrographs of growing NHKs at 1, 3, and 5 days post‐transfection of RPTN siRNA or control siRNA. (B) Quantification of the number of growing NHKs at 1‐, 3‐, and 5‐days post‐transfection with RPTN siRNA or control siRNA.

Figure S4: (A) RPTN siRNA or control siRNA were transfected into growing NHKs that were stained with an anti‐Ki67 antibody (left panels). (B) The quantification of Ki67‐positive cells after transfection of RPTN siRNA or control siRNA. The ratio of Ki67‐positive cells to DAPI‐stained nuclei was calculated. In a typical experiment, over 1500 cells in 7 fields were counted and the results show mean values ± standard deviation (SD). The experiments were independently repeated three times.

Figure S5: (A) Growing NHKs transfected with RPTN siRNA or control siRNA were subjected to TUNEL staining. (B) Quantification of TUNEL‐positive cells after transfection of RPTN siRNA or control siRNA. In each case, the ratio of TUNEL‐positive cells to DAPI‐stained nuclei was calculated. In a typical experiment, over 1500 cells in 7 fields were counted and the results shows mean values ± standard deviation (SD). The experiments were independently repeated three times.

Figure S6: Growing NHKs in 50% confluency were transfected with RPTN siRNA#1 or negative control siRNA. At one day after transfection, 1.5 mM Ca²⁺ was added to induce the expression of RPTN, and the cells were harvested at 4, 8, and 12 days post‐transfection. RPTN siRNA treatment significantly suppressed the mRNA expression of RPTN in comparison to the control level until 12 days post‐transfection.

Figure S7: Tissue samples from patients with skin diseases were immunostained for K6. Tissue sections were stained with 4′,6‐diamidine‐2′‐phenylindole dihydrochloride (DAPI) to visualise the nuclei. Scale bar: 200 μm (in all panels).

Data Availability Statement

The datasets generated and/or analysed during the current study are available from the corresponding author upon reasonable request.