Retinoid-related orphan receptor α is involved in induction of semaphorin 3A expression in normal human epidermal keratinocytes

Journal of Dermatological Science 79 (2015) 84–90

Contents lists available at ScienceDirect

Journal of Dermatological Science journal homepage: www.jdsjournal.com

Letter to the Editor Retinoid-related orphan receptor a is involved in induction of semaphorin 3A expression in normal human epidermal keratinocytes

Keywords: Semaphorin 3A; Keratinocytes; Antipruritic drugs

During neural development, nerve fibers (or axons) are regulated by both attraction and repulsion factors to reach their targets. Nerve growth factor (NGF) produced by keratinocytes is one of the major attraction factors that determine skin innervation [1]. NGF expression is greater in the epidermal keratinocytes of patients with atopic dermatitis (AD) than of healthy controls [1] and is involved in the sprouting of epidermal nerve fibers associated with pruritus in AD patients [1,2]. In addition, the expression of semaphorin 3A (Sema3A), which induces retraction of NGF-responsive sensory afferents, is decreased in the epidermis of AD patients [3]. These findings suggest that epidermal innervation is regulated by the balance between nerve elongation factors (e.g., NGF) and nerve repulsion factors (e.g., Sema3A) [3]. Recently, we showed that topical application of Sema3A ointment significantly inhibited itch-related behavior and improved dermatitis in NC/Nga mice with AD-like symptoms [4]. Although Sema3A may be a promising agent for the treatment of pruritus in the periphery, most protein drugs induce greater immune responses than small compounds, with prolonged application likely causing allergy reactions, such as contact dermatitis. Retinoid-related orphan receptor alpha (RORa) has been shown to function as a transcriptional regulator of Sema3F, and to be associated with the development and progression of breast cancer through inactivation of the RORa–Sema3F pathway [5]. This study therefore evaluated the ability of RORa to induce endogenous Sema3A in normal human epidermal keratinocytes (NHEKs). NHEKs derived from adult epidermis were cultured in keratinocyte basal medium (KBM-Gold; Lonza, Basel, Switzerland), supplemented with 0.15 mM calcium and KGM-Gold SingleQuot (Lonza), at 37 8C with 5% CO2 and used within three passages. NHEKs were incubated with various concentrations of the synthetic RORa/g agonist SR1078 (Calbiochem, Darmstadt, Germany), the synthetic RORa/g an inverse agonist SR1001 (Sigma Aldrich, St. Louis, MO, USA), or the RORg-specific inverse agonist ursolic acid (Tokyo Chemical Industry, Tokyo, Japan) for 48 h at 37 8C or with the endogenous RORa agonist cholesterol sulfate (Sigma Aldrich) for 96 h at 37 8C. The cells were subsequently harvested and Sema3A mRNA levels were quantified

by real-time PCR using SYBR Premix Ex Taq (Takara, Shiga, Japan) and the primers 50 -ACCCAACTATCAATGGGTGCCTTA-30 (forward) and 50 -AACACTGGATTGTACATGGCTGGA-30 (reverse). As controls, ribosome protein S18 (RPS18) mRNA was amplified using the primers 50 -ACTCAACACGGGAAACCTCA-30 (forward) and 50 -AACCAGACAAATCGCTCCAC-30 (reverse). The specificity of the amplified Sema3A fragments was confirmed by quantitative analysis of the melting curve with SDS software (Applied Biosystems, Foster City, CA, USA). The levels of Sema3A mRNA were normalized relative to those of RPS18 and expressed relative to ratios of untreated controls. Cultured NHEKs were also transfected with 40 nM siGENOME SMARTpool human RORa and control siRNA (Thermo Scientific, Waltham, MA, USA) using Lipofectamine RNAi Max (Life Technologies, Carlsbad, CA, USA). After culture in antibiotic-free medium for 48 h, total RNA was extracted and RORa mRNA levels analyzed by quantitative real-time PCR using the primers 50 -AAATCGCATCTGGAAACCTG-30 (forward) and

Fig. 1. Effect of cholesterol sulfate on Sema3A expression by cultured NHEKs. NHEKs were incubated with various concentrations of cholesterol sulfate for 96 h at 37 8C. The cells were harvested and Sema3A levels were examined by quantitative realtime PCR (a) and ELISA (b). Sema3A mRNA levels were normalized relative to RPS18 mRNA gene. Sema3A protein levels were normalized relative to total protein levels. **P < 0.01, ***P < 0.001 (Dunnett’s method).

0923-1811/ß 2015 Japanese Society for Investigative Dermatology. Published by Elsevier Ireland Ltd. All rights reserved.

Letters to the Editor / Journal of Dermatological Science 79 (2015) 84–90

85

Fig. 2. Effects of agonists and inverse agonists of RORa/g, and of RORa siRNA, on Sema3A expression in cultured NHEKs. NHEKs were incubated with various concentrations of SR1078 (a), SR1001 (b) and ursolic acid (c) for 48 h at 37 8C. The cells were harvested and Sema3A mRNA levels were assessed by quantitative real-time PCR. (d) NHEKs were transfected with control or RORa siRNA (40 nM). After 48 h, the cells were harvested and RORa (left panel) and Sema3A (right panel) mRNA levels were examined by quantitative real-time PCR. The data were normalized relative to RPS18. ***P < 0.001 (Dunnett’s method in panels (a)–(c); Student’s t test in panel (d)).

50 -TTGGCAAACTCCACCACATA-30 (reverse). Concentrations of Sema3A protein were measured using an enzyme-linked immunosorbent assay (ELISA) kit for Sema3A (Uscn Life Science, Uhan, China). Statistical analyses were performed using Student’s t tests and Dunnett’s multiple comparison tests with GraphPad Prism 5 software (GraphPad Software, La Jolla, CA, USA), with P < 0.05 defined as statistically significant. Quantitative real-time PCR analysis showed that Sema3A expression in cultured keratinocytes was dose-dependently induced by cholesterol sulfate (7.6-fold, P < 0.001) (Fig. 1a) and SR1078 (3.4-fold, P < 0.001) (Fig. 2a), but was reduced by SR1001 (44%, P < 0.001) (Fig. 2b). The expression of Sema3A mRNA and Sema3A protein (Fig. 1b) both plateaued at a cholesterol sulfate concentration of 40 mM. In contrast, Sema3A mRNA expression was unaltered when these cells were incubated with ursolic acid, an RORg inverse agonist (Fig. 2c). Treatment of keratinocytes with RORa specific siRNA significantly reduced the Sema3A mRNA level (24.0%, P < 0.001) compared with control siRNA-treated cells (Fig. 2d). The ROR family of nuclear receptors consists of three members, RORa, RORb, and RORg [6,7]. This study found that the endogenous RORa agonist cholesterol sulfate dose-dependently upregulated Sema3A expression at both the mRNA and protein levels. RORa was previously reported to function as an inducer of keratinocyte differentiation, and to be involved in the transcriptional regulation of target genes such as filaggrin [6,8]. To our knowledge, however, cholesterol sulfate was not previously shown to activate RORb and g [8,9]. This study also found that the synthetic RORa/g agonist, SR1078 dose-dependently upregulated the expression of Sema3A mRNA, whereas the synthetic RORa/g inverse agonist SR1001 dose-dependently inhibited Sema3A expression in cultured NHEKs. In contrast, the RORg inverse agonist ursolic acid had no effect on Sema3A expression in the NHEKs. Ursolic acid has been reported to suppress IL-17 production by inhibiting RORg activity [9]. Although our results suggest that RORa may be involved in the transcriptional

regulation of Sema3A in NHEKs, as also shown by siRNA knockdown of RORa, we could not completely exclude the involvement of RORg in Sema3A expression. Cholesterol sulfate is a membrane lipid synthesized by cholesterol sulfotransferase from cholesterol during the process of keratinization [9]. The distribution of cholesterol sulfate in skin of patients with AD, however, remains unclear. RORa recognizes and binds as a monomer to specific sequences of DNA, typically of an AGGTCA half-site with a 50 AT-rich extension in the target gene [6]. Treatment with an RORa agonist such as cholesterol sulfate results in the further recruitment of activator proteins, increasing transcriptional activity [10]. Inverse agonists induce conformational changes within the receptor, repressing transcription [7]. Taken together, these findings suggest that an RORa agonist may be useful as a topical antipruritic treatment of skin diseases with peripheral itch sensitization involving epidermal hyperinnervation, such as AD, by inducing endogenous Sema3A expression. However, transcriptional activator may result in genotoxicity and affect genes other than the target gene in vivo.

Acknowledgements This work was partly supported by a Research Fellowship from the Japan Society for the Promotion of Science (13J09449), and by KAKENHI (26860387), and by a Health Labour Sciences Research Grant for Research on Allergy and Immunology from the Japanese Ministry of Health Labour and Welfare, and by a grant of Strategic Research Foundation Grant-aided Project for Private Universities from MEXT (S1311011). References [1] Ikoma A, Steinhoff M, Sta¨nder S, Yosipovitch G, Schmelz M. The neurobiology of itch. Nat Rev Neurosci 2006;7:535–47. [2] Lewin GR, Mendell LM. Nerve growth factor and nociception. Trends Neurosci 1993;16:353–9.

86

Letters to the Editor / Journal of Dermatological Science 79 (2015) 84–90

[3] Tominaga M, Takamori K. Itch and nerve fibers with special reference to atopic dermatitis: therapeutic implications. J Dermatol 2014;41:205–12. [4] Negi O, Tominaga M, Tengara S, Kamo A, Taneda K, Suga Y, et al. Topically applied semaphorin 3A ointment inhibits scratching behavior and improves skin inflammation in NC/Nga mice with atopic dermatitis. J Dermatol Sci 2012;66:37–43. [5] Xiong G, Wang C, Evers BM, Zhou BP, Xu R. RORa suppresses breast tumor invasion by inducing SEMA3F expression. Cancer Res 2012;72:1728–39. [6] Hanyu O, Nakae H, Miida T, Higashi Y, Fuda H, Endo M, et al. Cholesterol sulfate induces expression of the skin barrier protein filaggrin in normal human epidermal keratinocytes through induction of RORa. Biochem Biophys Res Commun 2012;428:99–104. [7] Jetten AM. Retinoid-related orphan receptors (RORs): critical roles in development, immunity, circadian rhythm, and cellular metabolism. Nucl Recept Signal 2009;7:e003. [8] Strott CA, Higashi Y. Cholesterol sulfate in human physiology: what’s it all about. J Lipid Res 2003;44:1268–78. [9] Solt LA, Burris TP. Action of RORs and their ligands in (patho)physiology. Trends Endocrinol Metab 2012;23:619–27. [10] Gigue`re V. Orphan nuclear receptors: from gene to function. Endocr Rev 1999;20:689–725.

Yayoi Kamataa, Mitsutoshi Tominagaa, Azumi Sakaguchia, Yoshie Umeharaa, Osamu Negib, Hideoki Ogawaa, Kenji Takamoria,b,* a Institute for Environmental and Gender-Specific Medicine, Juntendo University Graduate School of Medicine, 2-1-1 Tomioka, Urayasu, Chiba 279-0021, Japan; bJuntendo University Urayasu Hospital, 2-1-1 Tomioka, Urayasu, Chiba 279-0021, Japan *Corresponding author at: Juntendo University Urayasu Hospital, 2-1-1 Tomioka, Urayasu, Chiba 279-0021, Japan. Tel.: +81 47 353 3171; fax: +81 47 353 3178 E-mail address: [email protected] (K. Takamori).

2 December 2014 http://dx.doi.org/10.1016/j.jdermsci.2015.03.015

Letter to the Editor Superficial epidermolytic ichthyosis caused by a novel KRT2 mutation Dear Editor Superficial epidermolytic ichthyosis (SEI, OMIM 146800), previously known as ichthyosis bullosa of Siemens (IBS), is a rare autosomal dominant skin disorder caused by mutations in the keratin 2 (KRT2) gene. SEI shows similar, but milder, clinical and histopathological findings to epidermolytic ichthyosis (EI, OMIM 113800), which is caused by mutations in the keratin 1 or 10 genes. SEI is clinically characterized by mild epidermal hyperkeratosis over flexural areas, blister formation, and the development of superficially denuded areas of hyperkeratotic skin [1]. Herein, we report a sporadic case of SEI harboring a novel mutation in the 1A rod region of the KRT2 gene. A 10-year-old Japanese boy was referred to our department with a history of skin rash on his extremities since early childhood. There was no definite history of blistering. Physical examination revealed dark to light gray hyperkeratosis covering the extremities (Fig. 1A–D), as well as mild hyperkeratosis with lichenification on the knees and ankles (Fig. 1B–D). Superficial denuded areas (Fig. 1B, arrowhead) and hypopigmented spots (Fig. 1A and B, arrow) were observed in the hyperkeratotic regions. His palms, soles, and trunk were not involved, and his nails, teeth, and hair were normal. No other family members showed any skin symptoms. A skin biopsy specimen obtained from his right ankle revealed marked hyperkeratosis and vacuolated keratinocytes with large keratohyalin granules in the upper spinous layer and granular layer (Fig. 1E–F). A tentative diagnosis of SEI was made and, following informed consent, mutation analysis was performed. Genomic DNA was isolated by standard methods from the peripheral blood of the patient and his unaffected parents. The coding regions of the KRT2 gene were amplified by polymerase chain reaction (PCR). All PCR products were sequenced in an ABI 310 automated sequencer. Direct sequencing of the KRT2 gene from the proband, revealed a novel heterozygous missense mutation in exon 1. This A to G transition at nucleotide position 557 (c.557A>G) resulted in the change of a codon from asparagine to serine at amino acid position 186 (p.N186S) (Fig. 2A). In order to confirm that the mutation detected in this study was not a common polymorphism, we screened 102 normal controls by restriction fragment length polymorphism (RFLP)

analysis, such that DdeI would specifically digest the resulting 478-bp PCR product from the mutated allele but not from the normal allele. RFLP analysis by DdeI digestion analysis revealed that the heterozygous c.557G>A mutation was present in the proband but not in his unaffected parents or in any of the 102 controls (Fig. 2B), thereby suggesting that the missense mutation is not a common polymorphism but rather a pathogenic mutation. In addition, Polyphen-2, the software for prediction of functional effects of human non-synonymous single nucleotide polymorphisms, predicted that this mutation was probably damaging for function of keratin 2 (http://genetics. bwh.harvard.edu/pph2/index.shtml). Therefore, we diagnosed a sporadic case of SEI. Hypopigmented macules observed in present case are not common clinical features of SEI. The pathomechanism of hypopigmented lesions remains unclear because skin biopsy specimen was not obtained from these areas. However it is presumed that the hypopigmented patch occurred because surrounding skin came to have mild pigmentation and dark gray hyperkeratosis, a small island of normal skin presented as if it were a hypopigmented lesions. The a-helical rod domain is composed of heptad repeats (seven amino acid repeats) denoted as (abcdefg)n. Positions a and d are typically occupied by hydrophobic amino acids, whereas e and g are frequently charged residues. Our patient had mutations in position e. According to a previous study of SEI mutations, 14 KRT2 mutations have been reported to date and almost all KRT2 mutations were located at the position of a, d, e, and g. Sung et al. reported a mild EI phenotype mimicking SEI which was related to a mutation in the KRT1 gene, pE478D, in position g, and thus speculated that the diversity of phenotypic severity among EI cases may be associated with specific heptad positions, namely that mutations affecting positions a and d were associated with severe phenotype [2]. Four distinct KRT2 mutations at positions a or d, including I182N, L484P, E487K, and E487D, were reported to date. However, there are no previous reports of SEI that demonstrate variety in clinical severity related to heptad position, because patients with SEI generally present with milder phenotypes compared to patients with EI. The asparagine residue, N186, is highly conserved in type I and II keratin intermediate filaments [3], suggesting that minor changes at this residue may affect keratin conformation. Indeed, mutations in this asparagine residue of other keratins have been reported in the pathogenesis of several other keratin disorders,