Mutation Analysis of the IL36RN Gene in 14 Japanese Patients with Generalized Pustular Psoriasis

ABSTRACT

Generalized pustular psoriasis (GPP) is a rare, potentially life threatening, and aggressive form of psoriasis, which is characterized by sudden onset with repeated episodic skin inflammation leading to pustule formation. Familial GPP is known to be caused by recessively inherited mutations in the IL36RN gene, which encodes interleukin 36 receptor antagonist (IL-36Ra). In this article, we performed mutation analysis of the IL36RN gene in 14 Japanese patients with GPP, and identified mutations in two of these patients analyzed. One patient was compound heterozygous for mutations c.115+6T>C and c.368C>G (p.Thr123Arg), whereas the other carried compound heterozygous mutations c.28C>T (p.Arg10*) and c.115+6T>C in the IL36RN gene. Expression studies using total RNA from the patients’ skin revealed that the mutation c.115+6T>C resulted in skipping of exon 3, leading to a frameshift and a premature termination codon (p.Arg10Argfs*1). The protein structure analysis suggested that the missense mutation p.Thr123Arg caused misfolding and instability of IL-36Ra protein. In vitro studies in cultured cells showed impaired expression of the p.Thr123Arg mutant IL-36Ra protein, which failed to antagonize the IL-36 signaling pathway. Our data further underscore the critical role of IL36RN in pathogenesis of GPP.

Introduction

Generalized pustular psoriasis (GPP) is a rare, but the most aggressive form of psoriasis which is characterized by sudden episodic diffuse erythematous skin eruption leading to sterile pustule formation, high fever, leukocytosis, and higher levels of C reactive protein in the serum [Griffiths and Barker, 2007]. Severe episodic attacks can be induced by pregnancy or infections. In addition, GPP can be frequently associated with psoriasis vulgaris and/or palmoplantar pustulosis [Griffiths and Barker, 2010].

Although psoriasis, in general, is considered to be a polygenic disorder, some cases of GPP are known to show a clear autosomal recessive inheritance trait [Familial GPP; MIM# 614204]. Importantly, it has recently been reported that mutations in IL36RN gene (MIM# 605507) underlie familial GPP in Tunisian and European populations [Marrakchi et al., 2011; Onoufriadis et al., 2011].

The IL36RN gene encodes interleukin 36 receptor antagonist (IL-36Ra), also known as IL-1F5. It belongs to interleukin 1 family and is structurally related to interleukin-36α (IL-36α), interleukin-36β (IL-36β), and interleukin-36γ (IL-36γ), all of which are inflammatory cytokines involved in modulating NF-κB signaling pathway via binding to their receptor named interleukin 1 receptor-like 2 (IL-1RL2) [Towne et al., 2004]. IL-36Ra protein has been shown to act as an antagonist for IL-36α, IL-36β, and IL-36γ through interaction with IL-1RL2 [Sims and Smith, 2010].

In addition to these findings, molecular basis of GPP in other populations remained largely elusive. The present study aims to identify the involvement of the IL36RN gene in the pathogenesis of GPP in Japanese population.

Materials and Methods

Source of DNA

After obtaining written informed consent, 14 unrelated Japanese patients with GPP (JPP1-JPP14) were enrolled in this study. The diagnosis of GPP was strictly determined by dermatologists in the Niigata University Hospital (A.M., N.K., A.A., H.F., and M.I.). The study was approved by the Ethics Committee of Niigata University School Graduate School of Medical and Dental Sciences and in adherence to the Declaration of Helsinki Principles. We collected peripheral blood sample from all the patients and 100 population-matched unrelated healthy control individuals in EDTA-containing tubes. Genomic DNA was isolated from peripheral blood lymphocytes according to standard extraction techniques.

Mutation Analysis

Using the genomic DNA from the affected individuals (JPP1–JPP14) as templates, all five exons including exon–intron boundaries of the IL36RN gene (GenBank accession number, NG_031864.1) were amplified by polymerase chain reaction (PCR) using gene specific primers (Supp. Table S1). The amplified PCR products were directly sequenced in an ABI 3130xl genetic analyzer using the ABI Prism Big Dye Terminator Cycle Sequencing Ready Reaction Kit (PE Applied Biosystems, Foster City, CA). To find out whether two heterozygous mutations were located on different alleles in the patients JPP12 and JPP14, we amplified the genomic DNA encompassing both mutations in each patient using a forward primer (5′-GAGTCTACACCCTGTGGAGCT-3′) and a reverse primer (5′-TGCCCACTAAGTCAGACGTG-3′). The amplified products, 3,386 bp in size, were cloned into the pCRII-TOPO vector (Invitrogen, Carlsbad, CA), and the positive clones were analyzed by direct sequencing. To screen for the mutations c.28C>T, c.115+6T>C, and c.368C>G in the IL36RN gene, PCR-restriction fragment length polymorphism (PCR-RFLP) analysis was performed (Supp. Table S2). The digested products were run on polyacrylamide gels.

DNA mutation numbering system followed is based on reference cDNA sequence with +1 corresponding to the A of the ATG translation initiation codon, according to journal guidelines (www.hgvs.org/mutnomen). Mutation descriptions on the protein level reflect the initiation codon as codon 1. All the mutation/sequence variants have been checked by mutalyzer program (http://www.LOVD.nl/mutalyzer/) using batch mode. All the variants have been submitted to a locus-specific database for the IL36RN gene (http://www.lovd.nl/IL36RN).

Reverse Transcriptase PCR

Total RNA was isolated from lesional skin of the affected individuals JPP1-JPP5, JPP12, and JPP14, as well as from normal skin of a healthy Japanese individual, using the RNeasy Minikit (Qiagen Inc., Valencia, CA) in accordance with the manufacturer's guidelines. Reverse transcription was performed using oligo-dT primers and the SuperScript III (Invitrogen, Carlsbad, CA). Using the first-strand cDNA as templates, the IL36RN-cDNA (GenBank accession number, NM_012275) was PCR-amplified using a forward primer (5′-AGTCTACACCCTGTGGAGCT-3′) and a reverse primer (5′-GAGTCTGACAGGCTGATCGG-3′). The amplified products were run on 1.5% agarose gel. The products were extracted from the gel using the QIAquick Gel Extraction Kit (Quiagen Inc.), and were analyzed by direct sequencing.

Homology Modeling of Human IL-36Ra Protein

Sequence alignment was performed using CLUSTAL W [Thompson, et al., 1994]. The three-dimensional structure of human IL-36Ra protein was modeled (swiss model; http://swissmodel.expasy.org/) [Kiefer et al., 2009; Kopp and Schwede, 2004] using a reported mouse IL-1F5 structure (PDB ID: 1MD6 [Dunn et al., 2003]) as a template.

Generation of Expression Vectors

Using the first-strand cDNA from skin tissue of the affected individual JPP12 or a healthy Japanese individual as templates, coding sequences of the IL36RN, IL36G, and IL1RL2 were amplified by PCR (GenBank accession number, IL36G; NM_019618, IL1RL2; NM_003854) (Supp. Table S3). The amplified products were cloned into the mammalian expression vector pCXN2.1 [Niwa et al., 1991; Noguchi et al., 2003]. The sequences of all the generated constructs were confirmed by direct sequencing.

Transient Transfection and Western Blot

HEK293T (human embryonic kidney) and HeLa (human cervical cell carcinoma) cells were cultured in Dulbecco's modified Eagle's medium (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum, 100 IU/ml penicillin, and 100 µg/ml streptomycin. Expression vector for the N-terminal Flag-tagged wild type (Wt) or the p.Thr123Arg mutant (Mut) IL-36Ra (2 µg each) was transfected into the cells plated in six-well dishes using lipofectamine 2000 (Invitrogen, Carlsbad, CA). Cells were harvested 24 hr after the transfection. Total cell lysate was separated on 4–12% NuPAGE gels (Invitrogen, Carlsbad, CA), and Western Blots (WBs) were subsequently performed according to a previously described method [Shimomura et al., 2010]. The primary antibodies used were mouse monoclonal anti-Flag M2 (diluted 1:1,000; Sigma-Aldrich, St. Louis, MO) and rabbit polyclonal anti-β-actin (diluted 1:3,000; Sigma–Aldrich, St. Louis, MO).

Reporter Gene Assays

Using the genomic DNA of a healthy Japanese individual as a template, sequences of the IL8 promoter (−1554 to −1 from the transcription start site) was amplified by PCR (Supp. Table S3). The amplified product was cloned into the pGL3 basic luciferase reporter vector (Promega, Madison, WI). The generated vector was designated as pGL3-IL8-promoter. HeLa cells were seeded in 12-well dishes the day before transfection. A 100 ng of the pGL3-IL8-promoter or an empty pGL3-basic vector was transfected into each well along with the expression constructs for the IL-36γ, IL-1RL2, Wt-IL-36Ra, or p.Thr123Arg-Mut-IL-36Ra (200 ng each) using lipofectamine 2000 (Invitrogen, Carlsbad, CA). A construct for β-galactosidase reporter (100 ng; Promega, Madison, WI) was also transfected for normalization of transfection efficiency. Total amount of vectors were adjusted to 800 ng with an empty pCXN2.1 vector. The cells were lysed 24 hr after transfection and the signals were assayed as described previously [Shimomura et al., 2008]. The results represent triplicate determination of a single experiment that is representative a total of three similar experiments. Statistical analysis was performed using the Student's t-test. P < 0.01 was considered significant.

Real-Time PCR

To quantify the mRNA levels of the IL8, IL36A, IL36G in the patients’ skin, real-time PCR was performed using gene-specific primers (GenBank accession number, IL8; NM_000584, IL36A; NM_014440) (Supp. Table S4) and the first-strand cDNA from skin samples of the affected individuals JPP12 and JPP14, and a healthy Japanese individual, following the methods described previously [Shimomura et al., 2008]. The samples were run in triplicate and normalized to an internal control (B2M; GenBank accession number, NM_004048).

Immunohistochemistry

Paraffin-embedded skin sections from the affected individuals JPP1, JPP12, and JPP14, and a healthy Japanese individual were examined by immunohistochemistry with goat polyclonal anti-IL-36α/IL-1F6 antibody (diluted 1:200; R&D systems, Inc., Minneapolis, MN). Color development was performed with 3,3′-diamino-benzidine (Dako, Carpinteria, CA). Nuclei were stained with haematoxylin.

Results

Clinical Findings

A total of 14 unrelated Japanese GPP patients were included in this study. None of these patients had any familial history of GPP, and there were no consanguinities between the parents. All the affected individuals repeatedly exhibited fresh erythema and pustules on the whole body, accompanied by high fever, high C reactive protein, and neutrophilia (Table 1), and skin biopsy showed spongiform pustules of Kogoj in the subcorneal portion of the epidermis (Fig. 1; data not shown). Their medication records did not positively suggest drug-induced eruption. In addition, before the onset of GPP, none of the patients experienced withdrawal of oral or topical corticosteroid, which could precipitate the disease [Borges-Costa et al., 2011]. The detailed clinical information including the age at onset of GPP, predisposition to psoriasis vulgaris or palmoplantar pustulosis, body temperature, laboratory data, and response to the treatment given to each patient is summarized in Table 1.

Table 1. Clinical Information of the GPP Patients Analyzed in This Study

Patient	Age at onset Age/sex	of GPP	Age at onset of PV or PPP	Other manifestations	BT (°C)	Blood CRP (mg/dl)	Blood neutrophil (number/μl)	Treatment at the time of study	Triggering factors	Response to treatment
M, male; F, female; PV, psoriasis vulgaris; PPP, palmoplantar pustulosis; N/A, not applicable; BT, body temperature; CRP, C-reactive protein.
JPP1	80/M	74	N/A	Pulmonary edema	39.4	16.60	9,540	Etretinate, oral steroids, vitamin D3, topical steroids	Unknown	Excellent
JPP2	72/M	72	59 (PV)	None	38.3	30.57	8,100	Etretinate, oral steroids, vitamin D3, topical steroids	Infections	Excellent
JPP3	76/M	72	15 (PV)	None	38.2	7.81	15,380	Etretinate, vitamin D3, topical steroids	Infections	Excellent
JPP4	24/M	20	13 (PV)	None	38.0	4.31	9,430	Etretinate, tonsillectomy, psoralen ultraviolet A, vitamin D3, topical steroids	Unknown	Excellent
JPP5	50/M	45	N/A	Chronic kidney disease	39.4	20.75	18,830	Infliximab, tonsillectomy, vitamin D3, topical steroids	Unknown	Moderate
JPP6	48/F	43	N/A	Polyarthritis	38.1	12.08	26,690	Etretinate, tonsillectomy, vitamin D3, topical steroids	Infections	Excellent
JPP7	75/F	73	N/A	Dermatomyositis	38.6	16.69	16,420	Etretinate, vitamin D3, topical steroids	Unknown	Excellent
JPP8	44/M	42	N/A	Oligoarthritis, Down syndrome	38.8	35.77	46,670	Oral steroid, topical steroid	Infection	Moderate
JPP9	59/F	58	N/A	Diabetes mellitus	38.2	10.35	13,540	Etretinate, topical steroid, vitamin D3	Unknown	Excellent
JPP10	37/M	36	23 (PV)	Polyarthritis	38.7	38.08	21,160	Infliximab, MTX, topical steroid, vitamin D3	Infections	Moderate
JPP11	55/M	48	N/A	None	39.0	19.80	15,780	Cyclosporine A, etretinate, tonsillectomy, vitamin D3, topical steroids	Infections	Excellent
JPP12	76/F	51	N/A	None	38.5	14.60	12,820	Etretinate, topical steroid	Unknown	Moderate
JPP13	23/F	17	N/A	None	38.0	4.00	7,180	Topical steroid	Unknown	Excellent
JPP14	17/M	16	15 (PV)	None	39.3	9.04	18,680	Infliximab, oral steroid, vitamin D3	Infections	Moderate

Figure 1. Clinical features of a Japanese GPP patient, JPP12. A, B: Extensive erythematous skin lesions with pustules on belly and legs. C, D: Haematoxylin and eosin staining of a skin biopsy from a pustular eruption in JPP12 revealed a spongiform pustule of Kogoj (asterisks). Scale bars: 500 µm (C), 100 µm (D).

Identification of Compound Heterozygous Mutations in the IL36RN Gene

We searched for mutations in the IL36RN gene of 14 Japanese GPP patients (JPP1–JPP14). We did not find any sequence variants in the IL36RN gene in 12 of 14 patients analyzed (data not shown). However, we identified potential pathogenic mutations in the IL36RN gene in two patients (JPP12 and JPP14). Of these, JPP12 was heterozygous for c.115+6T>C in intron 3 and c.368C>G (p.Thr123Arg) in exon 5 of the IL36RN gene (Supp. Fig. S1A–D). JPP14 carried the same sequence variant c.115+6T>C in a heterozygous state (Supp. Fig. S1A and B), and additionally had a heterozygous substitution c.28C>T (p.Arg10*) in exon 2 of the IL36RN gene (Supp. Fig. S1E and F). Screening assays with restriction enzymes excluded all the three mutations from 100 healthy control individuals of Japanese origin (200 chromosomes) (Supp. Fig. S1B, D, and F; data not shown). We subsequently cloned larger genomic DNA fragments including both corresponding mutations in each patient into the pCRII–TOPO vectors, and separately sequenced each allele. The results confirmed that the two mutations were located on different alleles in each patient, thus revealed that both JPP12 and JPP14 carried compound heterozygous mutations in the IL36RN gene (Supp. Fig. S2).

Mutation c.115+6T>C Leads to Aberrant Splicing in JPP12 and JPP14

To investigate the effect of the c.115+6T>C mutation on splicing of the IL36RN in JPP12 and JPP14, we performed reverse transcriptase PCR (RT-PCR) analysis using total RNA from the patients’ skin. PCR for the IL36RN-cDNA amplified two distinct fragments, 424 bp and 338 bp in size, in both the patients, whereas a clear single fragment, 424 bp in size, was detected in a control individual (Fig. 2A). Each fragment was extracted from the gel and analyzed by direct sequencing. The results showed that, in JPP12, the 424 bp fragment was the product of normal splicing event and carried the mutation c.368C>G (p.Thr123Arg), whereas the 338 bp fragment lacked the entire sequences of exon 3, leading to an immediate premature termination codon (PTC) (Fig. 2B). Similarly, sequences of the 338 bp fragment in JPP14 also lacked exon 3, whereas the 424 bp fragment turned out to be the product from the c.28C>T (p.Arg10^*) allele (Fig. 2C). Collectively, we conclude that the splice site mutation c.115+6T>C resulted in skipping of exon 3 at the mRNA level.

Figure 2. The mutation c.115+6T>C caused skipping of exon 3. A: RT-PCR for the IL36RN in JPP12 and JPP14 amplified 424 bp and 338 bp fragments, whereas only the 424 bp fragment was detected in a control individual. B2M was also amplified to check the quality of cDNA. B: In JPP12, sequencing of the 424 bp fragment revealed normal splicing with the c.368C>G mutation, whereas the 338 bp fragment has exon 3 skipping without having the c.368C>G mutation. C: In JPP14, sequencing of the 424 bp fragment revealed normal splicing with the c.28C>T mutation, whereas 338 bp fragment has exon 3 skipping without having the c.28C>T mutation. PTC, premature termination codon (B) (C).

Effect of the Mutation p.Thr123Arg on Structure of IL36Ra

IL-36Ra (IL-1F5) protein is highly conserved between mice and humans (Fig. 3A), thus we modeled the three-dimensional structure of human IL36-Ra protein using the previously reported structure of mouse IL-1F5 (PDB ID: 1MD6 [Dunn et al., 2003]) as a template, and tried to predict how the mutation p.Thr123Arg would affect the structure of human IL-36Ra protein. Thr123 is situated in a core environment, where a large number of amino acid residues are densely embedded. Thr123 Cγ is predicted to be involved in formation of the core hydrophobic patch, consisting of residues Met11 in β1, Leu19 in β2, Leu26 and Ala28 in β3, Val131 in loop10, Phe149 in β12, to stabilize the conformation of human IL-36Ra protein (Fig. 3B and C). In addition, Thr123 Oγ is also considered to contribute to the stabilization by participating in hydrogen bond network (shown as dotted lines in Fig. 3B and C) together with Asp13 in β1 and Lys17 in loop1 (Physical distance: Thr123 Oγ – Asp13 Oε1, 2.7 Å; Asp13 Oε1 – Lys17 NH2, 2.8 Å). The mutation of Thr123 to Arg with bulky side chain would most likely cause steric clashing with adjacent amino acid residues. Therefore, the mutation is predicted to disrupt the hydrophobic interaction and hydrogen bond network contributing to the stabilization of IL-36Ra protein, resulting in core conformational change and misfolding. In contrast, it is unlikely that the mutation p.Thr123Arg directly affects the interaction with the receptor because Thr123 is obviously away from critical residues that mediate the receptor interaction, such as His32, Lys38, Tyr89, Glu94, and Lys96 (Fig. 3B).

Figure 3. Homology modeling of human IL-36Ra protein and expression studies in cultured cells. A: Structure-based sequence alignment of human and mouse IL-36Ra proteins. Both sequences were aligned using CLUSTAL W [Thompson et al., 1994] and visualized using ESPript 2.2 (http://espript.ibcp.fr/ESPript/ESPript/). The β-strands and helical structures of IL-36Ra protein are indicated with arrows and coils, respectively, and numbered according to their succession. B: Cartoon stereo diagrams of human IL-36Ra protein. Each of three β-hairpins (residues 1–46, 47–104, 105–155) is colored in cyan, purple, and green, respectively. Highlighted stick representation are amino acid residues contributing to stabilize the conformation of human IL-36Ra protein together with Thr123. Suggested hydrogen bond is shown by dotted lines. C: Three-dimensional structure of the core hydrophobic patch in human IL-36Ra protein suggested the involvement of Thr123 Cγ and Thr123 Oγ in stabilization of the protein. D: Expression of the p.Thr123Arg mutant IL-36Ra protein is severely impaired as compared to that of the wild-type (Wt) IL-36Ra in HEK293T cells.

Expression of IL-36Ra Protein is Severely Impaired by the p.Thr123Arg Mutation

To determine the effect of the missense mutation p.Thr123Arg on the expression of IL-36Ra, we overexpressed either the wild type (Wt) or the p.Thr123Arg mutant (Mut) IL-36Ra with an N-terminal Flag-tag (Flag-IL-36Ra) in HEK293T cells, collected the cell lysate, and performed WBs with an anti-Flag antibody. Expression of the p.Thr123Arg-Mut-IL-36Ra was much lower than that of the Wt-IL-36Ra, suggesting instability of the Mut-IL-36Ra, possibly due to misfolding (Fig. 3D). Similar results were also obtained in HeLa cells (Supp. Fig. S3).

The p.Thr123Arg Mutant IL-36Ra Protein Loses Its Ability to Antagonize the IL-36/IL-1RL2 Signaling

To investigate how the mutation p.Thr123Arg would affect its function, we performed reporter gene assays using the IL8 promoter, which is a bona fide target gene of the IL-36/IL-1RL2 signaling [Marrakchi et al., 2011]. The IL8 promoter was basically active in HeLa cells and the activity was further upregulated by cotransfection with a ligand (IL-36γ) and its receptor (IL-1RL2) (Fig. 4A). The Wt-IL-36Ra protein significantly repressed the promoter activity induced by IL-36γ and IL-1RL2 (Fig. 4A). In contrast, the Mut-IL-36Ra protein did not reduce the activity at all (Fig. 4A).

Figure 4. The p.Thr123Arg mutant IL-36Ra protein loses its ability to antagonize the IL-36/IL-1RL2 signaling, and the expression of inflammatory cytokines are upregulated in the patients’ skin. A: Results of IL8 promoter–reporter gene assays in HeLa cells. The IL8 promoter activity induced by IL-36γ and IL-1RL2 was significantly downregulated by the wild-type (Wt) IL-36Ra, but not by the p.Thr123Arg mutant IL-36Ra. N.S., not significant. B: Results of real-time PCR to measure the mRNA expression of the IL8, IL36A, and IL36G genes in the patients’ skin. All the three genes are markedly upregulated in all the patients’ skin analyzed. C–E: Results of immunohistochemistry with an anti-IL-36α antibody in skin sections of JPP1 without IL36RN mutations (C), JPP12 with IL36RN mutations (D) and a control individual (E). Note that the strong positive signals were detected in keratinocytes adjacent to the pustule (asterisk) in both the patients’ skin (C, D). Similar results were also obtained in the lesional skin of JPP14 (data not shown). Scale bar: 100 µm (C).

Expression of Inflammatory Cytokines is Significantly Upregulated in the Patients’ Skin

Finally, we assessed if expression of inflammatory cytokines related to the IL-36/IL-1RL2 signaling was increased in the patients’ skin. We examined not only JPP12 and JPP14, but also two mutation-negative patients—JPP1 and JPP2. Real-time PCR showed that expression levels of IL8, IL36A, and IL36G-transcripts were markedly upregulated in both the patients’ skin with IL36RN mutations (JPP12 and JPP14) (Fig. 4B). Interestingly, similar overexpression of these cytokines was also detected in the patients’ skin without IL36RN mutations (JPP1 and JPP2) (Fig. 4B). In line with these results, immunohistochemistry with an anti-IL-36α antibody demonstrated an elevated expression of IL-36α in the patients’ skin, especially in keratinocytes adjacent to the pustule (Fig. 4C–E).

Discussion

In the present study, we performed genetic analysis for the IL36RN gene in 14 Japanese patients with GPP and identified compound heterozygous mutations in 2 of 14 patients analyzed (JPP12 and JPP14). Both patients were heterozygous for a point mutation in intron 3, designated as c.115+6T>C (Supp. Fig. S1A and B), which was demonstrated to cause skipping of exon 3, leading to a frameshift and an immediate PTC (p.Arg10Argfs*1) (Fig. 2A–C). In addition to the mutation c.115+6T>C, JPP14 carried another heterozygous mutation c.28C>T (p.Arg10*) (Supp. Fig. S1E, F and Fig. 2C). Most recently, the same mutation has been identified in a Japanese family with GPP [Sugiura et al., 2012], which together with our results, suggest the possibility that both the c.115+6T>C and the c.28C>T may be a common founder mutation in the Japanese population. Our RT-PCR experiments led us to postulate that inactive truncated IL-36Ra proteins lacking most of the C-terminus could be translated from the mutant transcripts.

The missense mutation p.Thr123Arg identified in JPP12 occurred at an evolutionary conserved amino acid residue (Fig. 3A). Our in vitro studies clearly showed that the p.Thr123Arg mutant protein was poorly expressed and failed to antagonize the IL-36/IL1-RL2 signaling in cultured cells, most likely due to instability of the mutant protein rather than disruption of interaction with the receptor (Figs. 3D and 4A and Supp. Fig. S3). These data strongly suggest that the p.Thr123Arg is a loss of function mutation, similar to three other missense mutations (p.Leu27Pro, p.Arg48Trp, and p.Ser113Leu) reported recently [Marrakchi et al., 2011; Onoufriadis et al., 2011]. Loss of expression and/or function of IL-36Ra is predicted to induce hyperactivation of the IL-36/IL-1RL2 signaling and subsequent overexpression of inflammatory cytokines, as shown by our expression studies in the patients’ skin (Fig. 4B and D).

Although IL36RN mutations are strongly associated with early onset GPP in the Tunisian and Caucasian populations [Marrakchi et al., 2011; Onoufriadis et al., 2011], this correlation appears to be unclear in the Japanese population. In our study, the age at onset of GPP in JPP12 was 51 years old, and that in JPP14 was 16 years old, respectively (Table 1). In addition, the recently reported Japanese case with the homozygous mutation p.Arg10* in the IL36RN exhibited relatively late onset at the age of 34 [Sugiura et al., 2012]. It can be postulated that in addition to IL36RN mutations, many different factors including genetic modifiers, environmental factors, and/or genetic background are also involved in causing GPP. As the number of GPP cases with IL36RN mutations is currently so limited, it is yet premature to evaluate how IL36RN mutations are correlated with the age at onset and severity of GPP, and response to treatment. Association between IL36RN mutations and other forms of psoriasis also needs to be clarified in the future.

The present study did not identify mutations in the IL36RN gene in 12 of 14 GPP patients analyzed. We extracted total RNA from skin samples of five patients without IL36RN mutations (JPP1–JPP5) and performed RT-PCR, which showed expression of the IL36RN -transcripts with normal sequences in all these patients (data not shown). However, it is noteworthy that these mutation-negative patients also showed overexpression of inflammatory cytokines, similar to the patients with IL36RN mutations (Fig. 4B and C). The results may suggest the existence of other causative gene(s) responsible for GPP in the Japanese population, of which mutations would result in abnormal upregulation of the IL-36/IL-1RL2 axis. Extensive studies on larger scale including many GPP patients will further enhance the involvement of the IL36RN gene in pathogenesis of GPP and also disclose the genetic heterogeneity of the disease.

ACKNOWLEDGMENTS

We acknowledge the patients involved in this study. We thank Dr. Satoshi Ishii (Akita University, Japan) and Dr. Junichi Miyazaki (Osaka University, Japan) for supplying pCXN2.1 vector.

Disclosure Statement: The authors declare no conflict of interest.