IL-33 impacts on the skin barrier by downregulating the expression of filaggrin

IL-33 is a member of the IL-1 family of cytokines that is constitutively expressed in healthy skin and was found to be increased in the skin of patients with atopic dermatitis (AD). Because it can be released after tissue damage or physical stress including scratching of the skin,1Dickel H. Gambichler T. Kamphowe J. Altmeyer P. Skrygan M. Standardized tape stripping prior to patch testing induces upregulation of Hsp90, Hsp70, IL-33, TNF-alpha and IL-8/CXCL8 mRNA: new insights into the involvement of ‘alarmins’.Contact Dermatitis. 2010; 63: 215-222Crossref PubMed Scopus (84) Google Scholar it has been classified as an alarmin concerned with alerting the immune system.2Haraldsen G. Balogh J. Pollheimer J. Sponheim J. Kuchler A.M. Interleukin-33— cytokine of dual function or novel alarmin?.Trends Immunol. 2009; 30: 227-233Abstract Full Text Full Text PDF PubMed Scopus (257) Google Scholar It enhances TH2 responses by inducing IL-5 and IL-13 as well as TH1 responses via upregulation of IFN-γ. Keratinocytes are known producer cells of IL-33 and also express the receptor complex consisting of ST2 and IL-1RAcP on their surface. The aim of this study was to investigate the effect of IL-33 on keratinocytes, skin biopsies, and living skin equivalents with regard to the regulation of the skin barrier molecule filaggrin (FLG). FLG is a structural protein promoting the outer epidermal barrier by aggregation of intermediate filaments3Mack J.W. Steven A.C. Steinert P.M. The mechanism of interaction of filaggrin with intermediate filaments: the ionic zipper hypothesis.J Mol Biol. 1993; 232: 50-66Crossref PubMed Scopus (94) Google Scholar and influencing cell differentiation, which is considered to be a key step in establishing the structure and function of the stratum corneum.E1Houben E. De Paepe K. Rogiers V. A keratinocyte's course of life.Skin Pharmacol Physiol. 2007; 20: 122-132Crossref PubMed Scopus (68) Google Scholar Furthermore, processed FLG contributes to natural moisturing factors important for skin hydration.E2Denecker G. Ovaere P. Vandenabeele P. Declercq W. Caspase-14 reveals its secrets.J Cell Biol. 2008; 180: 451-458Crossref PubMed Scopus (159) Google Scholar There is evidence that a lack of FLG breakdown products favors transepidermal water loss,E3Cork M.J. Danby S.G. Vasilopoulos Y. Hadgraft J. Lane M.E. Moustafa M. et al.Epidermal barrier dysfunction in atopic dermatitis.J Investig Dermatol. 2009; 129: 1892-1908Crossref PubMed Scopus (519) Google Scholar allergen penetration,4Scharschmidt T.C. Man M.Q. Hatano Y. Crumrine D. Gunathilake R. Sundberg J.P. et al.Filaggrin deficiency confers a paracellular barrier abnormality that reduces inflammatory thresholds to irritants and haptens.J Allergy Clin Immunol. 2009; 124 (e1-6): 496-506Abstract Full Text Full Text PDF PubMed Scopus (212) Google Scholar and skin colonization with Staphylococcus aureus.E4Miajlovic H. Fallon P.G. Irvine A.D. Foster T.J. Effect of filaggrin breakdown products on growth of and protein expression by Staphylococcus aureus.J Allergy Clin Immunol. 2010; 126: 1184-1190.e3Abstract Full Text Full Text PDF PubMed Scopus (152) Google Scholar FLG loss of function may also play a role in the development of AD5Palmer C.N. Irvine A.D. Terron-Kwiatkowski A. Zhao Y. Liao H. Lee S.P. et al.Common loss-of-function variants of the epidermal barrier protein filaggrin are a major predisposing factor for atopic dermatitis.Nat Genet. 2006; 38: 441-446Crossref PubMed Scopus (2214) Google Scholar, 6Weidinger S. Illig T. Baurecht H. Irvine A.D. Rodriguez E. Diaz-Lacava A. et al.Loss-of-function variations within the filaggrin gene predispose for atopic dermatitis with allergic sensitizations.J Allergy Clin Immunol. 2006; 118: 214-219Abstract Full Text Full Text PDF PubMed Scopus (491) Google Scholar because it has been shown that up to 50% of the patients with AD carry such a mutation. Several cytokines such as IL-4,7Howell M.D. Kim B.E. Gao P. Grant A.V. Boguniewicz M. Debenedetto A. et al.Cytokine modulation of atopic dermatitis filaggrin skin expression.J Allergy Clin Immunol. 2007; 120: 150-155Abstract Full Text Full Text PDF PubMed Scopus (564) Google Scholar IL-17,E5Gutowska-Owsiak D. Schaupp A.L. Salimi M. Selvakumar T.A. McPherson T. Taylor S. et al.IL-17 downregulates filaggrin and affects keratinocyte expression of genes associated with cellular adhesion.Exp Dermatol. 2012; 21: 104-110Crossref PubMed Scopus (147) Google Scholar IL-22,E5Gutowska-Owsiak D. Schaupp A.L. Salimi M. Selvakumar T.A. McPherson T. Taylor S. et al.IL-17 downregulates filaggrin and affects keratinocyte expression of genes associated with cellular adhesion.Exp Dermatol. 2012; 21: 104-110Crossref PubMed Scopus (147) Google Scholar IL-25,E6Deleuran M. Hvid M. Kemp K. Christensen G.B. Deleuran B. Vestergaard C. IL-25 induces both inflammation and skin barrier dysfunction in atopic dermatitis.Chem Immunol Allergy. 2012; 96: 45-49Crossref PubMed Scopus (43) Google Scholar, E7Hvid M. Vestergaard C. Kemp K. Christensen G.B. Deleuran B. Deleuran M. IL-25 in atopic dermatitis: a possible link between inflammation and skin barrier dysfunction?.J Investig Dermatol. 2011; 131: 150-157Crossref PubMed Scopus (157) Google Scholar and IL-31E8Cornelissen C. Marquardt Y. Czaja K. Wenzel J. Frank J. Luscher-Firzlaff J. et al.IL-31 regulates differentiation and filaggrin expression in human organotypic skin models.J Allergy Clin Immunol. 2012; 129 (433.e1-8): 426-433Abstract Full Text Full Text PDF PubMed Scopus (195) Google Scholar have been described to negatively affect the expression of FLG in keratinocytes. In this study, we analyzed the effect of IL-33 on FLG expression in subconfluent and differentiated monolayer keratinocytes, skin biopsies, and skin equivalents. In the first approach, monolayer keratinocytes and skin samples were stimulated with IL-33 or IL-4 to analyze FLG mRNA expression (see additional information in the Methods section in this article's Online Repository at www.jacionline.org). We confirmed the effect of IL-4 on FLG in Ca2+-differentiated human primary keratinocytes (Fig 1, A). On investigating the impact of IL-33 on FLG transcription, we observed a tendency toward downregulation using the same experimental setup, whereas no effect of IL-33 on FLG expression was observable in undifferentiated keratinocytes cultured in the absence of calcium (Fig 1, B). When we analyzed keratinocytes from patients with AD that were kept in cell culture in a semiconfluent manner with or without Ca2+, we observed significant decreases in FLG mRNA on stimulation with IL-33 (Fig 1, B). The median effect on mRNA reduction was more pronounced compared with keratinocytes from healthy donors (median 0.54 vs 0.73, respectively). Subsequently, we investigated the expression of FLG mRNA in skin biopsies after incubation for 24 hours with IL-33 (Fig 1, C). Here, we could also confirm the downregulatory effect of IL-4. Although the effect of IL-33 was not significant, which may be explained by the time point of measurement after 24-hour incubation, a clear tendency could be observed (P = .08). An even shorter stimulation (4 hours) of skin samples with IL-33 or IL-4 did not show a clear effect in the regulation of FLG mRNA expression (data not shown). Histologic staining of FLG expression in the skin confirmed our results on the protein level. When compared with the nonstimulated skin, addition of IL-4 or IL-33 led to a decrease in FLG protein expression in the epidermis (Fig 2). We analyzed the staining regarding intensity and area. Because existing FLG within the stratum corneum is not expected to be affected by 24-hour stimulation, evaluation was restricted to keratinocytes with detectable nuclear counterstaining. By immunohistochemical staining and software-assisted analysis, we measured the stained area within the epidermis and observed a significant difference between nonstimulated and IL-33–stimulated samples (Fig 2, A). Stimulation with IL-4 also led to FLG loss as expected (nonstimulated, 17.72% ± 11.92; IL-4, 8.97% ± 7.06%; % epidermis stained, n = 6). In the next step, we studied the expression of FLG with fluorescence antibodies, in which the signal strength is not dependent on an enzymatic reaction. Here, the intensity of staining with anti-FLG was reduced in skin samples on IL-33 incubation (Fig 2, B). Under these conditions, IL-4 stimulation did not lead to a distinct decrease in FLG expression (nonstimulated, 14.07 ± 4.3; IL-4, 13.08 ± 5.29; fluorescence intensity, n = 9). Finally, in vitro skin equivalents were continuously stimulated with IL-33 to investigate a putative effect on keratinocyte differentiation and we observed a disturbed barrier formation by trend. After 8 days of culture in the air-liquid interphase, FLG was reduced by IL-33 stimulation (see Fig E1 in this article's Online Repository at www.jacionline.org). This effect was still observable after 14 days. Software-assisted measurement of FLG staining depicts a tendency toward reduced protein expression under the influence of IL-33. To investigate the effect on the penetrance of allergens, we applied biotinylated natural peanut allergen Ara h 2 onto the skin equivalent. After 2 hours of incubation, the allergen was detected in fixed paraffin sections by streptavidin-phycoerythrin. In control skin equivalents, we were not able to detect allergen penetrance, while IL-33 treatment led to a modest but visible effect on facilitated allergen penetrance into the epidermal part of the skin equivalent (see Fig E2 in this article's Online Repository at www.jacionline.org). On the basis of these results, we hypothesize that IL-33 contributes to eczematous inflammation not only by acting as an alarmin but also by virtue of its barrier-weakening function. We here observed a direct action of IL-33 on downregulation of FLG expression independent of cytokines from mast cells or TH2 cells. The observed stronger reduction in cells derived from patients with AD than in cells derived from healthy donors fits our findings of a higher expression of ST2 on the surface of AD keratinocytes in a previous study.8Seltmann J. Werfel T. Wittmann M. Evidence for a regulatory loop between IFN-gamma and IL-33 in skin inflammation.Exp Dermatol. 2013; 22: 102-107Crossref PubMed Scopus (50) Google Scholar Filaggrin expression was shown to be controlled by the activator protein 1 (AP1) family of transcriptional regulators,E11Jang S.I. Steinert P.M. Markova N.G. Activator protein 1 activity is involved in the regulation of the cell type-specific expression from the proximal promoter of the human profilaggrin gene.J Biol Chem. 1996; 271: 24105-24114Crossref PubMed Scopus (88) Google Scholar POU-domain transcription factors,E12Jang S.I. Karaman-Jurukovska N. Morasso M.I. Steinert P.M. Markova N.G. Complex interactions between epidermal POU domain and activator protein 1 transcription factors regulate the expression of the profilaggrin gene in normal human epidermal keratinocytes.J Biol Chem. 2000; 275: 15295-15304Crossref PubMed Scopus (31) Google Scholar and p63.E13Candi E. Rufini A. Terrinoni A. Dinsdale D. Ranalli M. Paradisi A. et al.Differential roles of p63 isoforms in epidermal development: selective genetic complementation in p63 null mice.Cell Death Differ. 2006; 13: 1037-1047Crossref PubMed Scopus (218) Google Scholar Signaling of IL-33 via ST-2 may affect these pathways because it has been shown to result in the recruitment of MyD88, IRAK, IRAK4, and TRAF6, followed by phosphorylation of the downstream targets extracellular-signal–regulated kinase1/2, p38, IκBα, and JUN amino-terminal kinase, finally resulting in nuclear factor-κB activation.E14Schmitz J. Owyang A. Oldham E. Song Y. Murphy E. McClanahan T.K. et al.IL-33, an interleukin-1-like cytokine that signals via the IL-1 receptor-related protein ST2 and induces T helper type 2-associated cytokines.Immunity. 2005; 23: 479-490Abstract Full Text Full Text PDF PubMed Scopus (2732) Google Scholar The mitogen-activated protein kinases Erk1/2, p38, and JNK stimulate AP1 activity and transcription,E15Eferl R. Wagner E.F. AP-1: a double-edged sword in tumorigenesis.Nat Rev Cancer. 2003; 3: 859-868Crossref PubMed Scopus (1576) Google Scholar whereas nuclear factor-κB has been shown to counteract AP1 by downregulation of cFos via Elk-1.E16Fujioka S. Niu J. Schmidt C. Sclabas G.M. Peng B. Uwagawa T. et al.NF-kappaB and AP-1 connection: mechanism of NF-kappaB-dependent regulation of AP-1 activity.Mol Cell Biol. 2004; 24: 7806-7819Crossref PubMed Scopus (323) Google Scholar Under most experimental conditions, the effect of IL-33 on FLG reduction was weaker than that observed for IL-4. We recently described another IL-4–mimetic action of IL-33 regarding the regulation of human beta defensin-2,9Alase A. Seltmann J. Werfel T. Wittmann M. Interleukin-33 modulates the expression of human beta-defensin 2 in human primary keratinocytes and may influence the susceptibility to bacterial superinfection in acute atopic dermatitis.Br J Dermatol. 2012; 167: 1386-1389Crossref PubMed Scopus (31) Google Scholar supporting the idea of IL-33 being a dichotomous cytokine having IL-1 family-like and TH2-associated properties. In conclusion, we propose that IL-33 has skin barrier–modulating effects apart from its well-described function as an innate alarmin mediating the activation of the adaptive immune system by promoting TH2 as well as TH1 responses. IL-33 bears the potential to directly or indirectly promote epidermal barrier impairment and thereby plays a crucial role in sustaining and enhancing the local inflammatory responses in diseases such as AD. We thank Ute Staar, Gabriele Begemann, and Petra Kienlin for excellent technical assistance. Foreskin for keratinocyte culture was obtained from anonymized donors. Keratinocytes from patients with AD were isolated from plucked hairs as described before.E9Wang D. Drenker M. Eiz-Vesper B. Werfel T. Wittmann M. Evidence for a pathogenetic role of interleukin-18 in cutaneous lupus erythematosus.Arthritis Rheum. 2008; 58: 3205-3215Crossref PubMed Scopus (68) Google Scholar Cells were cultured in serum-free medium (KGM-2, Promocell, Heidelberg, Germany). Human dermal fibroblasts were isolated from foreskin samples by incubation of 1 g of tissue for 2 hours in 5 mg/mL of Collagenase (Worthington, Freehold, NJ) and 2.5 mg/mL of hyaluronidase (Sigma-Aldrich, St Louis, Mo). Released fibroblasts were cultured in RPMI (Lonza, Basel, Switzerland) supplemented with 10% FCS. Full-thickness skin from otherwise discarded tissue in the context of routine surgery was obtained from anonymized healthy donors. Punch biopsies (4 mm) were taken from full-thickness skin, placed in serum-free medium (KGM-2, Promocell), and cultured at 37°C 5% CO2. The study was approved by the medical ethics committee of Hannover Medical School and conducted according to the principles of Declaration of Helsinki. Informed written consent was given by the patients or their parents. For differentiation experiments, keratinocytes were expanded until they reached confluency. The CaCl2 concentration in the medium was increased to 1.5 mM for a total of 5 days before the cells were used for experiments. Hydrocortisone and epidermal growth factor were omitted from the culture medium before stimulation. In vitro skin equivalents were prepared as described before.E10Mildner M. Ballaun C. Stichenwirth M. Bauer R. Gmeiner R. Buchberger M. et al.Gene silencing in a human organotypic skin model.Biochem Biophys Res Commun. 2006; 348: 76-82Crossref PubMed Scopus (66) Google Scholar Briefly, 1 × 106 dermal fibroblasts in 1 volume FCS per gel were mixed with 8 volumes of collagen (Pure Col, Advanced BioMatrix, San Diego, Calif) and 1 volume 10× HBSS (Gibco/Invitrogen, Darmstadt, Germany) neutralized with 1 mol/L NaOH and poured into 6-well inserts (BD Falcon, Franklin Lakes, NJ) in a deep 6-well tray. After gelation in a humified atmosphere at 37°C (without CO2) for 2 hours, the gels were equilibrated with KGM (Lonza) for another 2 hours at 37°C/5% CO2. Afterwards, 1 × 106 keratinocytes in 2 mL per gel were added to the inserts. The skin equivalents were incubated overnight. The next day the medium was removed from the inserts and the external well. The medium of the external well was replaced by a modified serum-free medium without bovine pituitary extract, but supplemented with 1.3 mM of calcium (Promocell) and 10 μg/mL of transferrin and 50 μg/mL of ascorbic acid (both Sigma-Aldrich), for the air-liquid interphase culture. The medium was changed every second day up to 14 days. To investigate allergen penetrance into the skin equivalent, we applied 5 μL of biotinylated natural peanut allergen Ara h 2 (1.1 mg/mL, Indoor Biotech, Charlottesville, Va) onto IL-33–stimulated or unstimulated skin equivalents. After 2 hours of incubation followed by 4% formaldehyde/PBS fixation, the allergen was detected in paraffin sections by immunofluorescence as described below. Keratinocytes and skin biopsies were stimulated with 50 ng/mL of IL-33, IL-4 (both R&D Systems, Wiesbaden, Germany), or their combination for 4 hours or 24 hours for either mRNA isolation or immunohistochemistry. Skin equivalents were either not stimulated or treated with 50 ng/mL of IL-33 every 2 days once the air-liquid interphase culture was started for a total of 8 or 14 days. A punch biopsy was taken and immediately embedded in TissueTek O.C.T Compound (Sakura Finetek, Torrance, Calif) to prepare frozen sections for immunohistochemistry. The remaining epidermis was detached, and RNA was isolated from the keratinocytes for PCR analysis. RNA from keratinocytes was isolated with the Mini RNA Isolation II Kit (Zymo Research, Freiburg, Germany). RNA from skin biopsies was isolated with the RNeasy Fibrous Tissue Mini Kit (Qiagen, Hilden, Germany). Reverse transcription was performed using the Quantitect Reverse Transcription Kit (Qiagen). Quantitative real-time PCR was run on a Light Cycler using SYBR Green with Quantitect primer assays for filaggrin FLG and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (both Qiagen). Embedded tissues were cut into 6-μm cryosections and fixed for 10 minutes at 4°C in acetone. Staining was performed according to the DAKO EnVision+ System-HRP protocol for mouse or rabbit antibody (DAKO, Glostrup, Denmark). Briefly, sections were blocked for 5 minutes with Peroxidase Block before primary anti-human FLG antibody (clone SPM181, Abcam, Cambridge, United Kingdom) (0.5 μg/mL) or the isotype control IgG was applied for 1 hour. Afterwards, the slides were incubated for 30 minutes each, first with peroxidase-labeled polymer and substrate chromogen. Slides were counterstained with hematoxylin and mounted with DAKO Ultramount medium. The percentage of area stained within the epidermis was measured using cell Sens Dimension software (version 1.8.1, Olympus, Hamburg, Germany). The area stained with the specific color spectrum was measured within the epidermis except the stratum corneum (skin biopsies, see Fig 2) or within the epidermis and the stratum corneum (skin equivalents, see Fig E1). Cryosections (6 μm) of skin biopsies were fixed for 10 minutes at 4°C in acetone. Initial blocking was performed by incubation with 2% BSA and 10% normal goat serum for 30 minutes. Anti-human FLG antibody (see above) or isotype control was applied (4 μg/mL) in antibody diluent (DAKO) overnight at 4°C. AlexaFluor546-labeled goat-anti-mouse IgG (Life Technologies, Carlsbad, Calif) was applied subsequently for 1 hour. Slides were mounted with VectaShield (VectorLabs, Burlingame, Calif). Intensity of fluorescence staining was measured using ImageJ version 1.44o (NIH, Bethesda, Md). Staining of the epidermis except the stratum corneum was analyzed. Background staining was substracted. Sections (5 μm) of paraffin-embedded skin equivalents were subjected to deparaffinization and a prior blocking step by applying antibody diluent (Dako, Glostrup, Denmark). Staining of biotinylated Ara h 2 was achieved by incubation with streptavidin-phycoerythrin (Jackson Laboratories, Bar Harbor, Me). Statistical analysis was performed with PRISM 5.0 (GraphPad Prism, La Jolla, Calif). Data that passed the normality test were analyzed with unpaired t test or paired t test when indicated. P values of less than .05 were considered significant (*P < .05, **P < .005, ***P < .001).Fig E2Penetration of Ara h 2 into the skin equivalent. Skin equivalents were either stimulated with IL-33 at the beginning of the air-liquid interphase for a total of 14 days or left unstimulated. Biotin-Ara h 2 that entered the skin equivalent was detected by streptavidin-PE. Area of positive staining is indicated by white arrows. auto, Autofluorescence; n.s., not stimulated; PE, phycoerythrin.View Large Image Figure ViewerDownload Hi-res image Download (PPT)