Frederic Valentin1#, Vinzenz Oji1#, Ingrid Hausser2, Eva Liebau3, Tatjana Tarinski1, Dieter Metze1, Dirk Breitkreutz4, Heiko Traupe1, Nathalie Jonca5 and Patrick Terheyden6
1Department of Dermatology, Von-Esmarch-Str. 58, DE-48149 Münster, 2Institute of Pathology, University Hospital Heidelberg, Heidelberg, 3Institute of Animal Physiology – Department of Molecular Physiology, University of Münster, Münster, 4Department of Genetics of Skin Carcinogenesis, German Cancer Research Center, Heidelberg, Germany, 5UMR5165/U1056 CNRS-INSERM-CNRS-Université Toulouse III, Différenciation Epidermique et Autoimmunité Rhumatoïde, Hôpital Purpan, Toulouse, France, and 6Department of Dermatology, University Hospital Schleswig-Holstein, Lübeck, Germany. E-mail: Frederic.Valentin@ukmuenster.de
#These authors contributed equally to this paper and should be considered as first authors.
Accepted May 11, 2015; Epub ahead of print May 27, 2015
Peeling skin syndromes (PSS) refer to a heterogeneous group of generalized and/or palmoplantar disorders. Inflammatory peeling skin disease (PSD) refers to PSS type B [MIM 270300] (1) and is an ichthyosiform erythroderma characterized by recurrent patchy peeling of the skin with accompanying pruritus. The disease persists throughout life, is accompanied by significant atopic manifestations, and can be reminiscent of Netherton syndrome (NTS) (1, 2). PSD is due to autosomal recessive loss-of-function mutations in the CDSN gene encoding corneodesmosin (CDSN) (3), as independently confirmed by several groups (4–7). Moreover, large deletions encompassing the CDSN gene have been identified as the molecular cause (8, 9).
CDSN is a basic phosphoprotein expressed in cornified epithelia and hair follicles (10). It is localized in the lamellar granules of the stratum granulosum and found in the corneodesmosomes of the stratum corneum. The protein has a high level of serine and glycine, with specific structure motifs rich in these amino acids at the N-terminal (aa 60–171) and C-terminal end (aa 375–450), designated as glycine loops. These structure motifs are suggested to mediate cell-cell cohesion. Progressive proteolytic degradation of CDSN by serine proteases (kallikrein 5 and 7) plays a key role in desquamation (11). Hence, PSD represents an impressive clinical example for the primary loss of function of CDSN.
This study investigated a new patient with a novel mutation of the CDSN gene and performed expression analysis to reveal information about the inflammatory component of the disorder.
METHODS AND RESULTS (for full details see Appendix S11)
The male patient had been treated for atopic dermatitis up to the age of 10 years until a clinical and histological diagnosis of PSD was first suggested. His parents were non-consanguine and healthy. Since the third week of life the patient suffered from peeling, scaling and eczema of the entire skin (Fig. 1a). The patient’s history showed sensitization and clinically relevant allergic reactions through contact with fish, peanut, house-dust mites or grass pollen. Moreover, he had palmoplantar hyperhidrosis. Routine laboratory parameters were normal, except for a highly elevated IgE level (3,204.0 kU/l, normal value < 26.3). Treatment was based on lipophilic ointment, antiseptics and antihistamines.
Direct sequencing of the CDSN gene revealed the heterozygous mutation c.175A>T (p.Lys59*) (3) and the novel heterozygous mutation c.1031delC, which predicts a frameshift with premature stop codon (p.Ile345Serfs*121) (method see (3)). The study was approved by the institutional review board of the University Hospital of Münster (2013-573-f-S). Skin biopsies of the affected patient analysed with the CDSN specific antibody G36-19 (10) revealed an absence of CDSN protein expression (Fig. 1b). Ultrastructural examination revealed an acanthotic epidermis, covered by a hyperkeratotic, loosely packed horny layer, exfoliating in variable levels, mostly after few layers of scales, but only rarely immediately after the last living cell layer. There was a continuous granular layer with normal amounts of keratohyalin granules. Corneodesmosomes were present in variable numbers, while the quality could not be evaluated (Fig. 1c). The subepidermal region showed a diffuse lymphocytic infiltrate with perivascular distribution including some eosinophilic granulocytes (not shown).
To generate three-dimensional (3D) skin equivalents, primary keratinocytes and fibroblasts were taken. After 24 h, skin models were raised to the air-liquid interface and grown for 9 days (methods adapted from (3)). Similar to the patient’s biopsy 3D skin equivalents revealed mild acanthosis and signs of increased desquamation. Moreover, a higher number of dyskeratoses can be observed in skin equivalents of PSD (Fig. 1d). In contrast to healthy skin equivalents immunofluorescence analysis of skin equivalents of the patient showed a loss of CDSN expression (Fig. 1e). This lack of CDSN was further confirmed by Western blot analysis (Fig. 1f).
Quantitative real-time PCR (qRT-PCR) with keratinocytes obtained from the patient was performed with an iCycler iQ™ Real-Time PCR Detection System (BioRad) using gene-specific primers. Keratinocytes of 4 different healthy persons served as controls. Analysis revealed a decreased expression level of CDSN mRNA (fold change (FC) 2-log –2.49). In addition, the expression levels of caspase 1 (CASP1), an inflammatory caspase, and interleukin 18 (IL18), a proinflammatory cytokine, were analysed. For both, increased expression levels (CASP1 (FC 2-log 1.31) and IL18 (FC 2-log 1.65)) were detected. Their expression levels were compared with that of keratin 1, which did not show a significant difference in expression (FC 2-log –0.81) (Fig. S11).
DISCUSSION
Loss of corneodesmosin as a molecular cause of PSD was identified in 2010 by Oji et al. (3). Since then 6 different CDSN mutations, including large deletions of the CDSN locus on chromosome 6p21.3, have been published for PSD (4–9). The novel loss of function mutation c.1031delC investigated in this study is located directly before the C-terminal glycine loop of CDSN. The mutation predicts a protein with a long nonsense sequence (p.Ile345Sers*121) (Fig. S21). Quantitative RT-PCR experiments with primary keratinocytes of the patient showed decreased expression of CDSN mRNA (Fig. S11), suggesting nonsense-mediated mRNA decay as a possible reason. This is in contrast to the previously described stop codon mutation p.Gly142* residing within the first glycine loop, which showed an elevated (2-fold more) expression level of CDSN mRNA and a residual presence of the truncated protein (7).
Our immunohistological findings in skin biopsies and 3D skin equivalents revealed absent protein expression of CDSN, as confirmed by Western blot analysis (Fig. 1). The ultrastructural findings of this study confirm the pathological impact of CDSN deficiency on corneodesmosomal adherens, as shown previously (3, 9).
Fig. 1. Clinical presentation, immunohistochemistry, ultrastructure, three-dimensional (3D) skin equivalents and corneodesmosin (CDSN) protein expression. (a) Clinical presentation. (b) Immunohistochemistry revealed absence of CDSN in the skin of the patient. (c) Ultrastructure indicates an exfoliating horny layer with corneodesmosomes as punctual contact between scales (arrows). (d) Similar to the patient’s biopsy 3D skin equivalents revealed mild acanthosis and signs of increased desquamation (haematoxylin and eosin (H&E) staining). (c) Immunofluorescence of corneodesmosin (green) (DAPI nuclear staining blue) and (d) Western blot analysis of 3D skin equivalents confirmed the immunohistological results of an absent CDSN protein expression. (Scale bars: 50 µm (b, d, e), 1000 nm (c)).
PSD and NTS, as well as the recently described SAM (severe dermatitis, multiple allergies and metabolic wasting) syndrome (12) share some characteristic clinical features, i. e. patients with these diseases often have pruritus and atopic manifestations (1, 2). Caspase-1 (CASP1) is activated by the inflammasome, a multiprotein complex, and promotes the secretion and processing of the proinflammatory cytokines like IL1β and IL18 (13). Dysregulation of epidermal CASP1 activity and elevated serum levels of IL18 have been observed in NTS (14); elevated IL18 levels have been shown in atopic dermatitis (15). To investigate whether the balance of CASP1 and IL18 is altered in PSD, we studied their relative expression levels. Interestingly, both CASP1 and IL18 expression levels were found to be up-regulated in the keratinocytes of the patient. These molecular findings correspond well with the clinical symptoms of PSD. Hence, this study supports the idea that peeling skin disease represents a novel monogenetic disease model for inflammatory and atopic skin diseases in general.
ACKNOWLEDGEMENTS
The authors would like to thank the patient for his cooperation. The expert clinical assistance of Christa Post, Andrea Wissel and Kira Süßmuth is gratefully acknowledged.
Funding resources. This work was supported by the Selbsthilfe Ichthyose e. V. (www.ichthyose.de) and the Medical Faculty (OJ111409). Use of EM900 was courtesy of the Electron Microscopy Core Facility (EMCF), University Heidelberg.
1http://www.medicaljournals.se/acta/content/?doi=10.2340/00015555-2142
REFERENCES