- Email: [email protected]
Stimulation of PPARα normalizes the skin lipid ratio and improves the skin barrier of normal and filaggrin deficient reconstructed skin
G Model DESC 2897 No. of Pages 9
Journal of Dermatological Science xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Journal of Dermatological Science journal homepage: www.jdsjournal.com
Stimulation of PPARa normalizes the skin lipid ratio and improves the skin barrier of normal and filaggrin deficient reconstructed skin Leonie Wallmeyera , Dominika Lehnena , Natascha Egera , Michaela Sochorováb , Lukáš Opálkab , Andrej Ková9cikb , Katerina Vávrováb , Sarah Hedtricha,* a b
Institute for Pharmaceutical Sciences, Pharmacology and Toxicology, Freie Universität Berlin, Germany Charles University Prague, Faculty of Pharmacy, Hradec Kralove, Czech Republic
A R T I C L E I N F O
A B S T R A C T
Article history: Received 9 April 2015 Received in revised form 19 August 2015 Accepted 30 September 2015
Background: Therapeutic options for atopic dermatitis mostly address the symptoms but causal therapies are still missing. Peroxisome proliferator activated receptor (PPAR) agonists exert beneficial effects in patients suffering this disease, whereas the stimulation of PPARa and g seemed most promising. Objectives: To elucidate the effects of the PPARa specific agonist WY14643, the PPARg agonist ciglitazone, and the dual PPARa + g agonist docosahexaenoic acid (DHA) on the homeostasis and barrier function of filaggrin deficient skin. Methods: The effects of the PPAR agonists on skin differentiation were evaluated via qPCR, Western blot, histological or immunofluorescence staining. Skin lipid organization was determined by ATR-FTIR and lipid composition was analyzed by HPTLC. Ultimately, the skin barrier function was assessed by skin absorption studies using the radioactively labeled compound testosterone. Results: Significant upregulation of filaggrin after DHA and WY14643 supplementation, but no effect of ciglitazone, on protein and mRNA level was detected. DHA and WY14643, but not ciglitazone, normalized the molar ratio of the main skin barrier lipids to 1:1:1 (free fatty acids:ceramides:cholesterol). Furthermore, DHA and WY14643 supplementation normalized the skin lipid profile in filaggrin deficient skin, but only WY14643 significantly improved the skin barrier function. Conclusion: Supplementation particularly with the PPARa agonist WY14643 improved the homeostasis and barrier function of filaggrin deficient skin models by normalization of the free fatty acid profile underlining the potential of PPAR agonists for the treatment of filaggrin-associated skin diseases. ã 2015 Japanese Society for Investigative Dermatology. Published by Elsevier Ireland Ltd. All rights reserved.
Keywords: Reconstructed skin models Filaggrin deficiency Skin barrier recovery PPAR agonists Skin lipids
1. Introduction Filaggrin is a late-stage differentiation marker of the human skin and plays a key role in epidermal barrier function. Its precursor pro-filaggrin is a large, highly phosphorylated polyprotein that is stored in keratohyalin granules in the granular cell layer of the viable epidermis. During terminal differentiation, profilaggrin is ultimately dephosphorylated and proteolytically cleaved into multiple functional filaggrin monomers. The
Abbreviations: AD, atopic dermatitis; DHA, docosahexaenoic acid; FFA, free fatty acids; WT, wild type; FLG siRNA, filaggrin knock down skin models; FTIR, Fourier transform infrared spectroscopy; HPTLC, high performance thin layer chromatography; IVL, involucrin; LOR, loricrin; PPAR, peroxisome proliferator-activated receptor; qPCR, real-time quantitative polymerase chain reaction; SC, stratum corneum; WB, Western blot. * Corresponding author. Fax: + 49 30 838 455065. E-mail address: [email protected] (S. Hedtrich).
monomers aggregate and bind to keratin forming the keratinmatrix-complex which is important for the integrity of the stratum corneum (SC) [1,2]. Loss-of-function mutations in the filaggrin gene (FLG) are a major predisposing factor for the manifestation of atopic dermatitis (AD) and the underlying cause of ichthyosis vulgaris [3]. However, not only patients with mutations show decreased filaggrin levels in the skin: inflammatory processes in general decrease filaggrin expression even in people not carrying the respective mutation [4]. Reduced levels or the lack of filaggrin disturbs the skin barrier function and triggers inflammatory responses [5–8]. Furthermore, we have recently shown that the lack of filaggrin leads to disturbed lipid organization and a significant increase in free fatty acids (FFA) due to an activation of secretory phospholipase A2 in filaggrin knock down skin models [9]. As for today, therapeutic options for filaggrin-associated skin diseases are limited and mostly address the symptoms by using e.g.
http://dx.doi.org/10.1016/j.jdermsci.2015.09.012 0923-1811/ ã 2015 Japanese Society for Investigative Dermatology. Published by Elsevier Ireland Ltd. All rights reserved.
Please cite this article in press as: L. Wallmeyer, et al., Stimulation of PPARa normalizes the skin lipid ratio and improves the skin barrier of normal and filaggrin deficient reconstructed skin, J Dermatol Sci (2015), http://dx.doi.org/10.1016/j.jdermsci.2015.09.012
G Model DESC 2897 No. of Pages 9
2
L. Wallmeyer et al. / Journal of Dermatological Science xxx (2015) xxx–xxx
anti-inflammatory or immunosuppressive drugs. No approved therapy exists to enhance or substitute missing parts of the filaggrin pathway or to specifically target the secondary abnormalities except for moisturizers [5]. Here, an upregulation of filaggrin expression in heterozygous patients could be a therapeutic option offering long-term benefits by reducing the barrier impairing effects due to the lack of filaggrin. Peroxisome proliferator-activated receptor (PPAR) agonists increase the expression of filaggrin and other structural proteins in the skin [10,11]. Furthermore, PPAR agonists are anti-inflammatory and exert beneficial effects on skin homeostasis and SC integrity [11–15] which makes them potential candidates for the treatment of inter alia filaggrin-associated skin diseases. Here, particularly the activation of PPARa and PPARg drew attention. Just recently, the upregulation of filaggrin in organotypic skin models following the incubation with PPAR modulators was described [16]. However, almost no data exist if the upregulation also works in filaggrin deficient skin and the specific effects of PPAR agonists on the skin barrier function are still poorly understood. So far, most data have been generated in mice but their predictivity for humans is ambiguous due to distinct interspecies-related differences [17] such as the abundance of hair follicles or the significantly thinner epidermal layer particularly in rodents. Hence, to enable studies in human-based tissue, reconstructed skin models emerged as a suitable alternative with steadily increasing relevance for fundamental dermatological research [9,18–20], skin absorption [21,22] or skin irritation testing [23]. In our current work, we report on the effects of the PPARa agonist WY14643, the PPARg agonist ciglitazone, and the dual PPARa + g agonist docosahexaenoic acid (DHA) in normal and filaggrin deficient reconstructed human skin for the first time. We were particularly interested in their effects on the expression of the structural proteins filaggrin, loricrin and involucrin, potential influences on the skin lipid organization and composition and their impact on the skin barrier function aiming for deeper insights into the action of PPAR agonists in normal and diseased human skin. 2. Materials and methods 2.1. Materials DHA, WY14643, ciglitazone, GW6471 and GW9662 were purchased from Sigma–Aldrich, St. Louis, USA. All solutions for H&E staining, formaldehyde solution 4%, Tween 20 and bovine serum albumin (BSA) were obtained from Carl Roth, Karlsruhe, Germany. Filaggrin antibody was purchased from Abcam, Cambridge, United Kingdom. Horseradish-peroxidase-conjugated secondary antibody and b-actin antibody were purchased from Cell Signaling, Frankfurt/Main, Germany. The secondary antibodies IgG DyLight 488 and IgG DyLight 594 as well as DAPI antifading mounting medium were bought from Dianova, Hamburg, Germany. The RNA isolation kit NucleoSpin1 RNA II was from Macherey-Nagel, Düren, Germany, the RevertAidTM First Strand cDNA Synthesis Kit from Fermentas, St. Leon-Rot, Germany. SYBR Green I Masterplus kit for qPCR was purchased from Roche, Penzberg, Germany. All primers for qPCR were provided by TibMolbiol, Berlin, Germany. Standards for HPTLC analysis (ceramide NS, AS, NP, AP, glucosylceramide and sphingomyelin) were obtained from Avanti Polar Lipids (Alabaster, AL, USA). Ceramides NH and EOS were prepared by a total synthesis; the synthetic procedures will be reported elsewhere. The structure and the purity of the prepared ceramides NH and EOS were determined by nuclear magnetic resonance spectra (Varian, polo Alto, CA, Mercury-Vx BB 300 Instrument, operating at 300 MHz for 1H and 75 MHz for 13C), infrared spectra (Nicolet 6700 FTIR spectrophotometer, Thermos Scientific, Waltham), and mass spectrometry
(Agilent 500 Ion Trap LC/MS, Santa Clara, CA). Cholesterol, fatty acid standards, L-a-phosphatidylcholine and all other agents were purchased from Sigma–Aldrich (St. Louis, USA) if not otherwise stated. 2.2. Preparation of skin models and supplementation with DHA, WY14643 or ciglitazone Normal, wild type (WT) and filaggrin deficient (FLG siRNA) skin models were generated according to previously published procedures [24] (see Supplementary material). The tissues were cultivated for 14 days with medium change every other day. Starting at day 2, 100 mM DHA, 50 mM WY14643 [16] or 5 mM ciglitazone were added to the cell culture medium [25]. To analyze the effect of cell differentiation, keratinocytes were pre-incubated with 1.3 mM calcium chloride for 5 days. After differentiation, the agonists were added and the cells were cultured for 48 h. 2.3. Histology, immunofluorescence and qPCR The models were embedded in tissue freezing medium, shockfrozen using liquid nitrogen, subsequently cut to vertical slices (5 mm) and stained with haematoxylin-eosin or immunofluorescence according to standard protocols. For RNA analysis, the skin models were punched to 10 mm discs, the epidermis was gently removed, frozen and then milled for 30 s at 25 Hz using a TissueLyzer (Qiagen, Hilden, Germany). Subsequently, RNA was isolated using NucleoSpin1 RNA II according to the manufacturer’s instructions. For cDNA synthesis, RevertAidTM First Strand cDNA Synthesis Kit was used. Subsequently, qPCR was performed using the SYBR Green I Masterplus kit. The primer sequences are listed in Table S1. YWHAZ served as house-keeping gene. 2.4. Western blot Western blot (WB) was performed according to standard protocols. Briefly, the skin models were lysed in radioimmunoprecipitation assay buffer and total protein concentrations were determined with the Pierce1 BCA Protein Assay Kit (Thermo Scientific, Waltham, USA). Subsequently, the samples (30 mg protein) were boiled in standard SDS-PAGE sample buffer and separated by 10% SDS polyacrylamide gel electrophoresis (Bio-Rad, Munich, Germany). The gels were blotted onto nitrocellulose membranes (Immobilon P, Carl Roth, Karlsruhe, Germany). After blocking with 5% skimmed-milk powder for 1 h at 37 C, the membranes were incubated with the primary filaggrin antibody at a concentration of 1:1.000 overnight at 4 C. Afterwards, the blots were washed and incubated with horseradish-peroxidase-conjugated secondary antibody for 1 h and subsequently, developed with SignalFireTM ECL reagent (Cell Signaling, Frankfurt/Main, Germany) and visualized by PXi/PXi Touch gel imaging system (Syngene, Cambridge, UK). 2.5. Receptor specificity of WY14643 and ciglitazone To verify PPAR specificity, cell culture studies with normal human keratinocytes were performed. Therefore, the cells were pre-incubated for 1 h with 5 mM of the selective PPARa-antagonist GW6471 or with 1 mM of the PPARg-specific antagonist GW9662. Subsequently, WY14643 or ciglitazone were added for 48 h. 2.6. SC isolation The models were placed on a filter paper soaked with 0.5% trypsin in phosphate-buffered saline (PBS, pH 7.4). The isolated SC
Please cite this article in press as: L. Wallmeyer, et al., Stimulation of PPARa normalizes the skin lipid ratio and improves the skin barrier of normal and filaggrin deficient reconstructed skin, J Dermatol Sci (2015), http://dx.doi.org/10.1016/j.jdermsci.2015.09.012
G Model DESC 2897 No. of Pages 9
L. Wallmeyer et al. / Journal of Dermatological Science xxx (2015) xxx–xxx
sheets were washed with PBS and any remaining keratinocytes were removed with a cotton swab. Subsequently, SC sheets were washed with acetone to remove surface contaminants, vacuumdried and stored at 20 C. Isolated human SC served as control. 2.7. FTIR spectroscopy IR spectra of the samples were collected on a Nicolet 6700 FT-IR spectrometer (Thermo Fisher Scientific, Waltham, USA) equipped with a single-reflection MIRacle attenuated total reflectance (ATR) germanium crystal at 23 C. The spectra were generated by coaddition of 256 scans collected at 4 cm 1 resolution and analyzed with the Bruker OPUS software. The exact peak positions were determined from second derivative spectra and by peak fitting if needed.
3
were dissolved in 100 ml CHCl3/MeOH 2:1. 10–30 ml of each lipid sample was sprayed on the plate using a Linomat IV (Camag, Muttenz, Switzerland). Standard lipids were dissolved in CHCl3/ MeOH 2:1 (v/v) (ceramides EOS, NS, AS, NP, NH, AP, cholesterol, and lignoceric acid), and CHCl3/MeOH 1:1 (v/v) (cholesterol sulfate, glucosylceramide, sphingomyelin and phospholipids), respectively, at 1 mg/ml. To generate calibration curves, the lipids were mixed in ratios that approximately corresponds to the composition of human SC [27]. For more detailed information, see Supplementary material. 2.10. HPLC analysis of FFA FFA were analyzed as their naphthacyl esters by a method based on Refs. [28] and [29]. For a detailed description, see Supplementary material.
2.8. Isolation of SC lipids 2.11. Skin absorption testing For the extraction of the SC intercellular lipids, a modified [26] method was used. The SC samples were extracted with 1 ml CHCl3/ MeOH 2:1 (v/v) per mg of SC for 1.5 h, filtered, separated and concentrated under a stream of nitrogen. The lipids were dried and stored at 20 C under argon. 2.9. HPTLC lipid analysis The lipid analysis was performed on silica gel 60 HPTLC plates (20 10 cm, Merck, Darmstadt, Germany). The extracted SC lipids
The skin permeability was evaluated according to validated test procedures [22,30]. Briefly, stock solutions of testosterone (40 mg/ml, 2% [v/v] Igepal1 CA-630, Sigma–Aldrich, St. Louis, USA) were spiked with an appropriate amount of the radiolabeled compound to achieve a total radioactivity of 2 mCi/ml. Permeation studies were performed at day 14 using a static setup (Franz type diffusion cells, diameter 15 mm, volume 12 ml, Permegear, Bethlehem, USA). The total amounts of the permeated test compound were quantified using radiochemical detection
Fig. 1. (a) Filaggrin expression in normal (WT) and filaggrin deficient (FLG siRNA) skin models following treatment with the dual PPARa + g agonist DHA, the PPARa agonist WY14643, and the PPARg agonist ciglitazone. Mean SEM, n = 4–5, * indicates statistical significance (p 0.05). (b) Western blot of filaggrin in normal (WT) and filaggrin deficient (FLG siRNA) skin models after supplementation with DHA, WY14643, or ciglitazone. (c) Representative immunostaining images against filaggrin in normal (WT) and filaggrin deficient (FLG siRNA) skin models and after supplementation with DHA, WY14643, or ciglitazone. SC: stratum corneum, Epi: viable epidermis, D: dermis equivalent, n = 3.
Please cite this article in press as: L. Wallmeyer, et al., Stimulation of PPARa normalizes the skin lipid ratio and improves the skin barrier of normal and filaggrin deficient reconstructed skin, J Dermatol Sci (2015), http://dx.doi.org/10.1016/j.jdermsci.2015.09.012
G Model DESC 2897 No. of Pages 9
4
L. Wallmeyer et al. / Journal of Dermatological Science xxx (2015) xxx–xxx
(Microbeta Plus, Wallac, Turku, Finland). The rate of testosterone permeation was depicted as the apparent permeability coefficient (Papp).
is well in line with previous findings [31,32]. Additionally, significantly increased epidermal thickness was observed with WY14643. Once more, the supplementation with ciglitazone showed no effect.
2.12. Statistical analysis 3.2. Influences of PPAR agonists on involucrin and loricrin expression The paired student’s t-test was used for direct comparisons between untreated and treated models or cells. For multiple comparisons, one-way analysis of variance, followed by Dunett's post hoc testing, was performed using GraphPad Prism 6.0 (GraphPad Software Inc., La Jolla, USA). p 0.05 (*) was considered significant. Data of at least three independent experiments are presented as means standard error of the mean (SEM). 3. Results 3.1. DHA and WY14643, but not ciglitazone, significantly enhanced filaggrin amounts Filaggrin expression in the knock down models was significantly reduced by 80%, respectively. The knock down was stable for at least 14 days [9,24]. Supplementation with DHA and WY14643 resulted in 16-fold and 2.3-fold increased filaggrin amounts in normal, wild type skin models (WT), respectively, (Fig. 1a). In filaggrin knock down models (FLG siRNA), filaggrin amounts also increased significantly about 5-fold after DHA and 3.5-fold after WY14643 treatment. In contrast, filaggrin amounts were only marginally influenced by ciglitazone. These results were also reproducible in monolayer cell cultures, in which ciglitazone treatment even reduced filaggrin in differentiated keratinocytes (Fig. S1). PPARa specificity of WY14643 was verified by inhibition studies using the PPARa selective antagonist GW6471. For ciglitazone, however, inhibition studies with the PPARg-selective antagonist GW9662 indicated off-target effects (Fig. S2). The PPAR antagonists themselves did not affect filaggrin amounts. Consistent with the qPCR results, immunofluorescence staining and WB also showed significantly increased amounts of filaggrin following DHA and WY14643 supplementation and no effects with ciglitazone (Fig. 1b and c). Furthermore, the SC thickness of FLG siRNA models significantly increased after DHA and WY14643 supplementation (Fig. 2) which
Involucrin (IVL) expression increased 2.4-fold in WT and 1.9fold in FLG siRNA models following WY14643 and ciglitazone supplementation. Loricrin (LOR) expression was induced 4.5-fold and 2.9-fold in FLG siRNA models with WY14643 and ciglitazone, respectively, but without statistical significance. DHA did not influence IVL and LOR expression (Fig. 3a and b). 3.3. DHA and WY14643 improved the SC lipid chain order in FLG SIRNA models To investigate the lipid chain order, SC samples from WT and FLG siRNA models and human skin were isolated and examined by Fourier transform infrared spectroscopy (FTIR). Recently, we showed that filaggrin deficiency alone leads to disturbed barrier lipid ratio and, consequently, lipid chain organization in reconstructed skin models [9]. The findings of the present study nicely underline these results: FLG siRNA models displayed a decreased lipid chain order with more gauche conformers as indicated by an increased methylene stretching wavenumber compared to WT (Fig. 4a and b). DHA and WY14643 treatment of FLG siRNA skin models increased the lipid chain order as reflected by a significant decrease of methylene stretching wavenumber to values found in WT skin models suggesting a normalization of the lipid organization. No significant changes in lipid chain order were found upon the treatment with ciglitazone. 3.4. DHA and WY14643 normalized the skin lipid ratio in FLG siRNA models To investigate possible reasons for the improved lipid chain order, the SC lipids of the skin models were extracted and analyzed by HPTLC. WT models contained less lipids per mg SC than human SC, but the major barrier lipids (ceramides/FFA/cholesterol) were present in an approximately 1:1:1 molar ratio. FLG siRNA models
Fig. 2. Thickness of stratum corneum (SC) and viable epidermis (VE) in untreated normal (WT) and filaggrin deficient (FLG siRNA) skin models and after treatment with DHA, WY14643, or ciglitazone. Magnification 10, scale bar = 100 mm, mean SEM, n = 5, * indicates statistical significance over normal, untreated skin models (WT) (p 0.05), + indicates statistical significance over FLG siRNA (p 0.05).
Please cite this article in press as: L. Wallmeyer, et al., Stimulation of PPARa normalizes the skin lipid ratio and improves the skin barrier of normal and filaggrin deficient reconstructed skin, J Dermatol Sci (2015), http://dx.doi.org/10.1016/j.jdermsci.2015.09.012
G Model DESC 2897 No. of Pages 9
L. Wallmeyer et al. / Journal of Dermatological Science xxx (2015) xxx–xxx
5
Fig. 3. Effects of DHA, WY14643 and ciglitazone on involucrin (IVL) (a) and loricrin (LOR) (b) expression in normal (WT) and filaggrin deficient (FLG siRNA) skin models. Mean SEM, n = 4–6.
showed elevated FFA concentrations compared to WT (21.5 2.4 mg/mg and 13.4 1.3 mg/mg) as described previously [9]. Upon DHA and WY14643 treatment, FFA levels were reduced to 15.3 1.0 mg/mg and 9.7 1.2 mg/mg, respectively, in FLG siRNA (Fig. 5a). Following ciglitazone treatment, we found slightly but insignificantly lower FFA values compared to untreated FLG siRNA skin models. HPLC analysis of the FFA profiles showed that more than 70% of FFA in the skin models were shorter (18 carbons or less) or unsaturated (10.4 1.4 mg/mg; Fig. 5b). FLG knock down led to an increase of the levels of those shorter and unsaturated FFA (to 17.6 3.9 mg/mg) while the levels of the very long FFA (with 20 carbons and more) remained relatively constant (approximately 3 mg/mg). Supplementation of FLG siRNA models with WY14643 normalized the FFA profile to similar levels seen in the WT models. Similar effect was also found for DHA treatment but without statistical significance. Other SC lipids such as cholesterol, ceramides and cholesterol sulfate were unaffected by DHA or WY14643 treatment (Fig. 5c–e). No significant changes were observed in the individual ceramide classes (Fig. S3) and on the lipid precursors (sphingomyelin, glucosylceramides and phospholipids) (Fig. 5c–h). Nevertheless, it is worth noting that the treatment with WY14643 showed trends towards increased levels of ceramides and cholesterol sulfate and decreased levels of the lipid precursors. Thus, treating the FLG
siRNA models with DHA and, in particular, WY14643 led to a normalization of FFA content, a normalization of the skin lipid ratio and, consequentially, improved skin lipid organization. 3.5. PPARa agonist WY14643 but not DHA and ciglitazone improved the skin barrier function Finally, the skin barrier function of normal and filaggrin deficient skin models after supplementation with DHA, WY14643 or ciglitazone was determined by skin permeation studies. Interestingly, the supplementation with WY14643 significantly improved the skin barrier function of normal and FLG siRNA models as seen from significantly lower Papp values compared to the untreated controls. No improvement of the skin barrier was observed after supplementation with ciglitazone or DHA (Fig. 6, Table 1). For the latter, a tendency towards reduced barrier function was seen. 4. Discussion Filaggrin mutations are a well-known risk factor for the development of AD and the underlying cause of ichthyosis vulgaris [3]. The exact mechanism, however, leading to the clinical manifestation of the skin diseases is poorly understood and
Fig. 4. Effects of DHA, WY14643 and ciglitazone on the stratum corneum lipid chain order as derived from the methylene symmetric stretching vibrations from infrared spectra of the normal (WT) and filaggrin deficient (FLG siRNA) skin models compared to human stratum corneum (hSC): (a) representative infrared spectra and (b) wavenumbers of methylene symmetric stretching of hydrated SC samples examined by infrared spectroscopy at 23 C. Mean SEM, n 3 batches, * indicates statistical significance (p 0.05) as indicated, h indicates statistically significant differences compared to hSC (p 0.05).
Please cite this article in press as: L. Wallmeyer, et al., Stimulation of PPARa normalizes the skin lipid ratio and improves the skin barrier of normal and filaggrin deficient reconstructed skin, J Dermatol Sci (2015), http://dx.doi.org/10.1016/j.jdermsci.2015.09.012
G Model DESC 2897 No. of Pages 9
6
L. Wallmeyer et al. / Journal of Dermatological Science xxx (2015) xxx–xxx
Fig. 5. Effects of DHA, WY14643 and ciglitazone on the stratum corneum lipid profiles of normal (WT) and filaggrin deficient (FLG siRNA) skin models compared to human stratum corneum (hSC). SC lipids were extracted from the skin models and analyzed by HPTLC for the content of barrier lipids (FFA: free fatty acids, Chol: cholesterol, Cer: ceramides, CholS: cholesterol sulfate) and their precursors (SM: sphingomyelin, PL: phospholipids, GCer: glucosylceramides). FFA profiles were determined by reversed phase HPLC of their naphthacyl esters. Mean SEM, n 3 batches, * indicates statistical significance (p 0.05) as indicated, h indicates statistically significant differences compared to hSC at p 0.05.
therapeutic options targeting the underlying cause are still missing [5]. During the last decade, PPAR agonists gained interest as candidates for the treatment of filaggrin-associated and inflammatory skin diseases due to anti-inflammatory effects in models of irritant and allergic contact dermatitis [12], accelerated skin acidification [32] and stimulated skin lipid synthesis [11,31,33]. Moreover, simultaneous application of glucocorticoids and PPAR agonists counterbalance side effects of glucocorticoids such as decreased filaggrin expression and epidermal thinning [13,34]. In human skin, the isotypes PPARa, PPARb/d and PPARg are expressed. PPARb/d is present throughout all epidermal layers, whereas PPARa and PPARg are mainly located in suprabasal compartments [10,11]. PPARs are crucial for skin homeostasis and mediate e.g. keratinocyte differentiation and skin repair [25,35]. Hence, PPARs emerged as promising targets for the treatment of hyperproliferative or inflammatory skin disorders where particularly the activation of PPARa and PPARg play a role. PPARg modulates cellular differentiation and acts as an inhibitory regulator of immune cells [14,15,35]. PPARa governs cell proliferation and differentiation and exerts potent anti-inflammatory effects in AD as demonstrated in mice [13,36]. Moreover, a small
clinical trial with AD patients that were topically treated with the PPARa agonist clofibrate also showed a significant decrease of the disease severity score and chemokine levels after two weeks of topical application [37]. The specific effects on the human skin barrier function, however, are not fully understood. Hence, in this study we investigated the effects of the PPARa specific agonist WY14643, the PPARg agonist ciglitazone and the dual PPARa + g modulator DHA in normal and filaggrin deficient skin models. Filaggrin deficient skin models are characterized by disturbed epidermal maturation and differentiation and increased susceptibility to skin irritation [24,38]. Just recently, we studied the impact of filaggrin deficiency on skin lipids and found significantly increased expression of NHE-1, secretory phospholipase A2 and, consequently, increased FFA levels in the knock down model and disturbed skin lipid organization contributing to the disturbed skin barrier function of the models [9]. The presence of PPARs in reconstructed skin was previously demonstrated and increased filaggrin expression following supplementation with PPAR agonists in normal models was described [16,31]. Here, we first demonstrated that this upregulation also works in filaggrin deficient skin models after supplementation with DHA and
Please cite this article in press as: L. Wallmeyer, et al., Stimulation of PPARa normalizes the skin lipid ratio and improves the skin barrier of normal and filaggrin deficient reconstructed skin, J Dermatol Sci (2015), http://dx.doi.org/10.1016/j.jdermsci.2015.09.012
G Model DESC 2897 No. of Pages 9
L. Wallmeyer et al. / Journal of Dermatological Science xxx (2015) xxx–xxx
7
Fig. 6. Skin permeability of untreated, normal (WT) ( ) and filaggrin deficient (FLG siRNA) (&) skin models following the supplementation with WY14643 (* WT/ + WY14643, & FLG siRNA/ + WY14643) (a), DHA (* WT/ + DHA, & FLG siRNA/ + DHA) (b), and ciglitazone (* WT/ + ciglitazone, & FLG siRNA/ + ciglitazone) (c). Mean SEM, n = 3–5.
WY14643 (Fig. 1). Selective PPARa and g stimulation also increased loricrin and involucrin expression—other important structural proteins of the cornified envelope (Fig. 3 and S3). DHA was without effect. Well in line with previous findings, we also observed a thickening of the SC after DHA and WY14643 supplementation [31,32] and increased epidermal thickness with WY14643. In terms of the skin barrier lipids, we confirmed our previous finding that filaggrin deficient skin models display less ordered lipid chains and a disturbed ratio of the main barrier lipids (FFA: ceramides:cholesterol) compared to WT models [9]. Upon treatment of the FLG siRNA models with DHA and WY14643, the levels of FFA in SC decreased leading to lipid ratio and lipid chain order values similar to those in the WT models. PPARg agonist ciglitazone, which was not able to upregulate filaggrin, did not change the SC lipid profile of FLG siRNA. To obtain deeper insights into the changes of the FFA profiles by the action of WY14643 and DHA, or more precisely how could an increase in FFA levels decrease lipid chain order, we determined the proportions of very long FFA (with 20 carbons and more) and shorter (with 18 carbons or less) or unsaturated FFA. The FFA pool in all the skin models was mainly (over 70%) shorter or unsaturated, which could explain the lower lipid order compared to human SC. This altered FFA profile in the skin models compared to human SC is in agreement with [39]. In FLG siRNA models, the proportion of the very long FFA did not change compared to WT,
Table 1 Papp values of the model drug testosterone in untreated, normal (WT) and FLGdeficient (FLG siRNA) skin models after treatment with DHA, WY14643 or ciglitazone. * indicates statistical significance (p 0.05). Papp (x10
WT FLG siRNA
6
cm/s) for testosterone
Control
DHA
WY14643
Ciglitazone
3.4 0.9 3.2 0.7
7.6 0.9 5.6 1.1
1.7 0.1* 1.8 0.5
3.0 0.9 2.9 1.1
but the levels of the shorter or unsaturated FFA increased. Again, this is in good agreement with the lower lipid chain order of the FLG siRNA compared to WT and with our previous finding [9]. Increased levels of shorter and unsaturated FFA were previously found in AD patients (although without association with FLG mutations) and their negative effects on the lipid order was proved, too [40]. Treatment of FLG siRNA models with WY14643 (and also with DHA, but without statistical significance) decreased the levels of shorter/unsaturated FFA to the values seen in WT models. Consequentially, the skin lipid ratio and, thus, the skin lipid organization improved significantly (Fig. 4). It is well established that the stimulation of PPAR receptors affects the skin lipid metabolism. Rivier and coworkers found a dose-dependent increase of cholesterol sulfate and ceramide levels in reconstructed skin models after supplementation with the PPARa agonist WY14643 [31]. At the same concentration of WY14643 that we used here, they observed slightly but insignificantly increased levels of ceramides (in particular of ceramide NS) and cholesterol sulfate, which is very similar to our findings. We did not observe any detrimental effects of DHA on other skin lipid classes (Fig. 5). This is noteworthy since the (topical) application of fatty acids can adversely affect the lamellar body formation and skin barrier function [41]. Ultimately, the effects of the PPAR agonists on skin barrier function were investigated. For WY14643, significantly decreased permeability for the model drug testosterone was observed indicating an improvement of skin barrier (Fig. 6). Interestingly, despite filaggrin upregulation, SC thickening, and amended skin lipid ratio, the skin barrier was not improved by DHA. Potential explanations are apoptotic effects of DHA on proliferative cells hampering epidermal differentiation [42] or the lack of effect on loricrin and involucrin expression (Fig. 3a and b). The ineffectiveness of ciglitazone observed in our study was concordant with previously published data in healthy [43] and AD mouse models [36] and might be explained with its rather low
Please cite this article in press as: L. Wallmeyer, et al., Stimulation of PPARa normalizes the skin lipid ratio and improves the skin barrier of normal and filaggrin deficient reconstructed skin, J Dermatol Sci (2015), http://dx.doi.org/10.1016/j.jdermsci.2015.09.012
G Model DESC 2897 No. of Pages 9
8
L. Wallmeyer et al. / Journal of Dermatological Science xxx (2015) xxx–xxx
PPARg binding affinity [35]. Nevertheless, PPARg activation efficiently suppressed dendritic cell and nuclear factor kB mediated immune responses [14,15] implying strong anti-inflammatory effects which, however, cannot be studied in reconstructed skin models so far. In conclusion, the effects of PPAR agonists on the skin barrier were investigated in filaggrin deficient skin for the first time. Well in line with previous studies, PPARa agonist WY14643 upregulated filaggrin and promoted skin differentiation. We could further show that WY14643 normalized the disturbed FFA profile present in filaggrin deficient skin models by reducing the levels of short and unsaturated FFA to levels found in control skin models which ultimately resulted in significantly improved skin barrier function. PPARg agonist ciglitazone and the dual agonist DHA failed in terms of skin barrier function improvement. Funding sources Financial support by the Foundation for the Promotion of Alternate and Complementary Methods to Reduce Animal Testing (Foundation SET) is gratefully acknowledged (S.H., L.W.). This work was further supported by the Czech Science Foundation (project No. 13-23891S) and Charles University (SVV 260 183). Conflict of interest The authors state no conflict of interest. Acknowledgement We greatly acknowledge the scientific support of Priv. Doz. Dr. Joachim Fluhr, Charité Berlin. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j. jdermsci.2015.09.012. References [1] G.M. O’Regan, A. Sandilands, W.H. McLean, A.D. Irvine, Filaggrin in atopic dermatitis, J. Allergy Clin. Immunol. 124 (2009) R2–R6. [2] S.J. Brown, W.H. McLean, One remarkable molecule: filaggrin, J. Invest. Dermatol. 132 (2012) 751–762. [3] C.N. Palmer, A.D. Irvine, A. Terron-Kwiatkowski, Y. Zhao, H. Liao, S.P. Lee, et al., Common loss-of-function variants of the epidermal barrier protein filaggrin are a major predisposing factor for atopic dermatitis, Nat. Genet. 38 (2006) 441–446. [4] M.D. Howell, B.E. Kim, P. Gao, A.V. Grant, M. Boguniewicz, A. Debenedetto, et al., Cytokine modulation of atopic dermatitis filaggrin skin expression, J. Allergy Clin. Immunol. 120 (2007) 150–155. [5] P.M. Elias, Therapeutic implications of a barrier-based pathogenesis of atopic dermatitis, Ann. Dermatol. 22 (2010) 245–254. [6] S. Kezic, Function of filaggrin and its metabolites, in: J.P. Thyssen, H.I. Maibach (Eds.), Filaggrin, Springer, 2014, 2015, pp. 3–8. [7] M.A. McAleer, A.D. Irvine, The multifunctional role of filaggrin in allergic skin disease, J. Allergy Clin. Immunol. 131 (2013) 280–291. [8] H. Kawasaki, K. Nagao, A. Kubo, T. Hata, A. Shimizu, H. Mizuno, et al., Altered stratum corneum barrier and enhanced percutaneous immune responses in filaggrin-null mice, J. Allergy Clin. Immunol. 129 (2012) 1538–1546 e1536. [9] K. Vávrová, D. Henkes, K. Strüver, M. Sochorová, B. Školová, M.Y. Witting, et al., Filaggrin deficiency leads to impaired lipid profile and altered acidification pathways in a 3D skin construct, J. Invest. Dermatol. 134 (2014) 746–753. [10] L. Michalik, W. Wahli, Peroxisome proliferator-activated receptors (PPARs) in skin health, repair and disease, Biochim. Biophys. Acta 1771 (2007) 991–998. [11] M. Schmuth, Y.J. Jiang, S. Dubrac, P.M. Elias, K.R. Feingold, Thematic review series: skin lipids. Peroxisome proliferator-activated receptors and liver X receptors in epidermal biology, J. Lipid Res. 49 (2008) 499–509. [12] M.Y. Sheu, A.J. Fowler, J. Kao, M. Schmuth, K. Schoonjans, J. Auwerx, et al., Topical peroxisome proliferator activated receptor-alpha activators reduce inflammation in irritant and allergic contact dermatitis models, J. Invest. Dermatol. 118 (2002) 94–101. [13] Y. Hatano, P.M. Elias, D. Crumrine, K.R. Feingold, K. Katagiri, S. Fujiwara, Efficacy of combined peroxisome proliferator-activated receptor-alpha ligand and glucocorticoid therapy in a murine model of atopic dermatitis, J. Invest. Dermatol. 131 (2011) 1845–1852.
[14] A. Mastrofrancesco, D. Kovacs, M. Sarra, E. Bastonini, G. Cardinali, N. Aspite, et al., Preclinical studies of a specific PPARgamma modulator in the control of skin inflammation, J. Invest. Dermatol. 134 (2014) 1001–1011. [15] K. Jung, A. Tanaka, H. Fujita, A. Matsuda, K. Oida, K. Karasawa, et al., Peroxisome proliferator-activated receptor gamma-mediated suppression of dendritic cell function prevents the onset of atopic dermatitis in NC/Tnd mice, J. Allergy Clin. Immunol. 127 (2011) 420–429 e421–426. [16] C. Zhang, I. Gurevich, B.J. Aneskievich, Organotypic modeling of human keratinocyte response to peroxisome proliferators, Cells Tissues Organs 196 (2012) 431–441. [17] M. Leist, T. Hartung, Inflammatory findings on species extrapolations: humans are definitely no 70-kg mice, Altex 30 (2013) 227–230. [18] M.O. Danso, V. van Drongelen, A. Mulder, J. van Esch, H. Scott, J. van Smeden, et al., TNF-alpha and Th2cytokines induce atopic dermatitis-like features on epidermal differentiation proteins and stratum corneum lipids in human skin equivalents, J. Invest. Dermatol. 134 (2014) 1941–1950. [19] D. van, V. rongelen, M. Alloul-Ramdhani, M.O. Danso, A. Mieremet, A. Mulder, S. van, J. meden, et al., Knock-down of filaggrin does not affect lipid organization and composition in stratum corneum of reconstructed human skin equivalents, Exp. Dermatol. 22 (2013) 807–812. [20] E.H. van den Bogaard, G.S. Tjabringa, I. Joosten, M. Vonk-Bergers, E. van Rijssen, H.J. Tijssen, et al., Crosstalk between keratinocytes and T cells in a 3D microenvironment: a model to study inflammatory skin diseases, J. Invest. Dermatol. 134 (2014) 719–727. [21] S. Küchler, K. Strüver, W. Friess, Reconstructed skin models as emerging tools for drug absorption studies, Expert Opin. Drug Metab. Toxicol. 9 (2013) 1255–1263. [22] M. Schäfer-Korting, U. Bock, W. Diembeck, H.J. Dusing, A. Gamer, E. HaltnerUkomadu, et al., The use of reconstructed human epidermis for skin absorption testing: results of the validation study, Altern. Lab. Anim.: ATLA 36 (2008) 161–187. [23] S.N. Kolle, K.M. Sullivan, A. Mehling, B. van Ravenzwaay, R. Landsiedel, Applicability of in vitro tests for skin irritation and corrosion to regulatory classification schemes: substantiating test strategies with data from routine studies, Regul. Toxicol. Pharmacol.: RTP 64 (2012) 402–414. [24] S. Küchler, D. Henkes, K.M. Eckl, K. Ackermann, J. Plendl, H.C. Korting, et al., Hallmarks of atopic skin mimicked in vitro by means of a skin disease model based on FLG knock-down, Altern. Lab. Anim.: ATLA 39 (2011) 471–480. [25] M. Mao-Qiang, A.J. Fowler, M. Schmuth, P. Lau, S. Chang, B.E. Brown, et al., Peroxisome-proliferator-activated receptor (PPAR)-gamma activation stimulates keratinocyte differentiation, J. Invest. Dermatol. 123 (2004) 305–312. [26] E.G. Bligh, W.J. Dyer, A rapid method of total lipid extraction and purification, Can. J. Biochem. Physiol. 37 (1959) 911–917. ková, M. Pospíšilová, B. Školová, J. Zbytovská, K. [27] P. Pullmannová, K. Stan Vávrová, Effects of sphingomyelin/ceramide ratio on the permeability and microstructure of model stratum corneum lipid membranes, Biochim. Biophys. Acta (BBA)—Biomembr. 1838 (2014) 2115–2126. [28] V. Rioux, D. Catheline, M. Bouriel, P. Legrand, High performance liquid chromatography of fatty acids as naphthacyl derivatives, Analusis 27 (1999) 186–193. [29] R.F. Borch, Separation of long chain fatty acids as phenacyl esters by high pressure liquid chromatography, Anal. Chem. 47 (1975) 2437–2439. [30] OECD, Test No. 428: Skin Absorption: In Vitro Method, OECD Publishing, 2015. [31] M. Rivier, I. Castiel, I. Safonova, G. Ailhaud, S. Michel, Peroxisome proliferatoractivated receptor-alpha enhances lipid metabolism in a skin equivalent model, J. Invest. Dermatol. 114 (2000) 681–687. [32] J.W. Fluhr, M.Q. Man, J.P. Hachem, D. Crumrine, T.M. Mauro, P.M. Elias, et al., Topical peroxisome proliferator activated receptor activators accelerate postnatal stratum corneum acidification, J. Invest. Dermatol. 129 (2009) 365–374. [33] M. Schmuth, V. Moosbrugger-Martinz, S. Blunder, S. Dubrac, Role of PPAR, LXR, and PXR in epidermal homeostasis and inflammation, Biochim. Biophys. Acta 1841 (2014) 463–473. [34] M. Demerjian, E.H. Choi, M.Q. Man, S. Chang, P.M. Elias, K.R. Feingold, Activators of PPARs and LXR decrease the adverse effects of exogenous glucocorticoids on the epidermis, Exp. Dermatol. 18 (2009) 643–649. [35] Y. Yan, M. Furumura, S. Numata, K. Teye, T. Karashima, B. Ohyama, et al., Various peroxisome proliferator-activated receptor (PPAR)-gamma agonists differently induce differentiation of cultured human keratinocytes, Exp. Dermatol. 24 (2014) 62–65. [36] T. Chiba, S. Takeuchi, H. Esaki, K. Yamamura, Y. Kurihara, Y. Moroi, et al., Topical application of PPARalpha (but not beta/delta or gamma) suppresses atopic dermatitis in NC/Nga mice, Allergy 67 (2012) 936–942. [37] M. Fukaya, H. Kimata, Topical clofibrate improves symptoms in patients with atopic dermatitis and reduces serum TARC levels: a randomized, double-blind, placebo-controlled pilot study, J. Drugs Dermatol.: JDD 13 (2014) 259–263. [38] M. Mildner, J.A. Jin, L. Eckhart, S. Kezic, F. Gruber, C. Barresi, et al., Knockdown of filaggrin impairs diffusion barrier function and increases UV sensitivity in a human skin model, J. Invest. Dermatol. 130 (2010) 2286–2294. [39] J. van Smeden, W.A. Boiten, T. Hankemeier, R. Rissmann, J.A. Bouwstra, R.J. Vreeken, Combined LC/MS-platform for analysis of all major stratum corneum lipids: and the profiling of skin substitutes, Biochim. Biophys. Acta 1841 (2014) 70–79.
Please cite this article in press as: L. Wallmeyer, et al., Stimulation of PPARa normalizes the skin lipid ratio and improves the skin barrier of normal and filaggrin deficient reconstructed skin, J Dermatol Sci (2015), http://dx.doi.org/10.1016/j.jdermsci.2015.09.012
G Model DESC 2897 No. of Pages 9
L. Wallmeyer et al. / Journal of Dermatological Science xxx (2015) xxx–xxx [40] J. van Smeden, M. Janssens, E.C. Kaye, P.J. Caspers, A.P. Lavrijsen, R.J. Vreeken, et al., The importance of free fatty acid chain length for the skin barrier function in atopic eczema patients, Exp. Dermatol. 23 (2014) 45–52. [41] M. Mao-Qiang, K.F. Feingold, P.M. Elias, Exogenous lipids influence permeability barrier recovery in acetone-treated murine skin, Arch. Dermatol. 129 (1993) 728–738.
9
[42] S. Serini, E. Piccioni, N. Merendino, G. Calviello, Dietary polyunsaturated fatty acids as inducers of apoptosis: implications for cancer, Apoptosis: Int. J. Programmed Cell Death 14 (2009) 135–152. [43] M.Q. Man, E.H. Choi, M. Schmuth, D. Crumrine, Y. Uchida, P.M. Elias, et al., Basis for improved permeability barrier homeostasis induced by PPAR and LXR activators: liposensors stimulate lipid synthesis, lamellar body secretion, and post-secretory lipid processing, J. Invest. Dermatol. 126 (2006) 386–392.
Please cite this article in press as: L. Wallmeyer, et al., Stimulation of PPARa normalizes the skin lipid ratio and improves the skin barrier of normal and filaggrin deficient reconstructed skin, J Dermatol Sci (2015), http://dx.doi.org/10.1016/j.jdermsci.2015.09.012
Journal of Dermatological Science xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Journal of Dermatological Science journal homepage: www.jdsjournal.com
Stimulation of PPARa normalizes the skin lipid ratio and improves the skin barrier of normal and filaggrin deficient reconstructed skin Leonie Wallmeyera , Dominika Lehnena , Natascha Egera , Michaela Sochorováb , Lukáš Opálkab , Andrej Ková9cikb , Katerina Vávrováb , Sarah Hedtricha,* a b
Institute for Pharmaceutical Sciences, Pharmacology and Toxicology, Freie Universität Berlin, Germany Charles University Prague, Faculty of Pharmacy, Hradec Kralove, Czech Republic
A R T I C L E I N F O
A B S T R A C T
Article history: Received 9 April 2015 Received in revised form 19 August 2015 Accepted 30 September 2015
Background: Therapeutic options for atopic dermatitis mostly address the symptoms but causal therapies are still missing. Peroxisome proliferator activated receptor (PPAR) agonists exert beneficial effects in patients suffering this disease, whereas the stimulation of PPARa and g seemed most promising. Objectives: To elucidate the effects of the PPARa specific agonist WY14643, the PPARg agonist ciglitazone, and the dual PPARa + g agonist docosahexaenoic acid (DHA) on the homeostasis and barrier function of filaggrin deficient skin. Methods: The effects of the PPAR agonists on skin differentiation were evaluated via qPCR, Western blot, histological or immunofluorescence staining. Skin lipid organization was determined by ATR-FTIR and lipid composition was analyzed by HPTLC. Ultimately, the skin barrier function was assessed by skin absorption studies using the radioactively labeled compound testosterone. Results: Significant upregulation of filaggrin after DHA and WY14643 supplementation, but no effect of ciglitazone, on protein and mRNA level was detected. DHA and WY14643, but not ciglitazone, normalized the molar ratio of the main skin barrier lipids to 1:1:1 (free fatty acids:ceramides:cholesterol). Furthermore, DHA and WY14643 supplementation normalized the skin lipid profile in filaggrin deficient skin, but only WY14643 significantly improved the skin barrier function. Conclusion: Supplementation particularly with the PPARa agonist WY14643 improved the homeostasis and barrier function of filaggrin deficient skin models by normalization of the free fatty acid profile underlining the potential of PPAR agonists for the treatment of filaggrin-associated skin diseases. ã 2015 Japanese Society for Investigative Dermatology. Published by Elsevier Ireland Ltd. All rights reserved.
Keywords: Reconstructed skin models Filaggrin deficiency Skin barrier recovery PPAR agonists Skin lipids
1. Introduction Filaggrin is a late-stage differentiation marker of the human skin and plays a key role in epidermal barrier function. Its precursor pro-filaggrin is a large, highly phosphorylated polyprotein that is stored in keratohyalin granules in the granular cell layer of the viable epidermis. During terminal differentiation, profilaggrin is ultimately dephosphorylated and proteolytically cleaved into multiple functional filaggrin monomers. The
Abbreviations: AD, atopic dermatitis; DHA, docosahexaenoic acid; FFA, free fatty acids; WT, wild type; FLG siRNA, filaggrin knock down skin models; FTIR, Fourier transform infrared spectroscopy; HPTLC, high performance thin layer chromatography; IVL, involucrin; LOR, loricrin; PPAR, peroxisome proliferator-activated receptor; qPCR, real-time quantitative polymerase chain reaction; SC, stratum corneum; WB, Western blot. * Corresponding author. Fax: + 49 30 838 455065. E-mail address: [email protected] (S. Hedtrich).
monomers aggregate and bind to keratin forming the keratinmatrix-complex which is important for the integrity of the stratum corneum (SC) [1,2]. Loss-of-function mutations in the filaggrin gene (FLG) are a major predisposing factor for the manifestation of atopic dermatitis (AD) and the underlying cause of ichthyosis vulgaris [3]. However, not only patients with mutations show decreased filaggrin levels in the skin: inflammatory processes in general decrease filaggrin expression even in people not carrying the respective mutation [4]. Reduced levels or the lack of filaggrin disturbs the skin barrier function and triggers inflammatory responses [5–8]. Furthermore, we have recently shown that the lack of filaggrin leads to disturbed lipid organization and a significant increase in free fatty acids (FFA) due to an activation of secretory phospholipase A2 in filaggrin knock down skin models [9]. As for today, therapeutic options for filaggrin-associated skin diseases are limited and mostly address the symptoms by using e.g.
http://dx.doi.org/10.1016/j.jdermsci.2015.09.012 0923-1811/ ã 2015 Japanese Society for Investigative Dermatology. Published by Elsevier Ireland Ltd. All rights reserved.
Please cite this article in press as: L. Wallmeyer, et al., Stimulation of PPARa normalizes the skin lipid ratio and improves the skin barrier of normal and filaggrin deficient reconstructed skin, J Dermatol Sci (2015), http://dx.doi.org/10.1016/j.jdermsci.2015.09.012
G Model DESC 2897 No. of Pages 9
2
L. Wallmeyer et al. / Journal of Dermatological Science xxx (2015) xxx–xxx
anti-inflammatory or immunosuppressive drugs. No approved therapy exists to enhance or substitute missing parts of the filaggrin pathway or to specifically target the secondary abnormalities except for moisturizers [5]. Here, an upregulation of filaggrin expression in heterozygous patients could be a therapeutic option offering long-term benefits by reducing the barrier impairing effects due to the lack of filaggrin. Peroxisome proliferator-activated receptor (PPAR) agonists increase the expression of filaggrin and other structural proteins in the skin [10,11]. Furthermore, PPAR agonists are anti-inflammatory and exert beneficial effects on skin homeostasis and SC integrity [11–15] which makes them potential candidates for the treatment of inter alia filaggrin-associated skin diseases. Here, particularly the activation of PPARa and PPARg drew attention. Just recently, the upregulation of filaggrin in organotypic skin models following the incubation with PPAR modulators was described [16]. However, almost no data exist if the upregulation also works in filaggrin deficient skin and the specific effects of PPAR agonists on the skin barrier function are still poorly understood. So far, most data have been generated in mice but their predictivity for humans is ambiguous due to distinct interspecies-related differences [17] such as the abundance of hair follicles or the significantly thinner epidermal layer particularly in rodents. Hence, to enable studies in human-based tissue, reconstructed skin models emerged as a suitable alternative with steadily increasing relevance for fundamental dermatological research [9,18–20], skin absorption [21,22] or skin irritation testing [23]. In our current work, we report on the effects of the PPARa agonist WY14643, the PPARg agonist ciglitazone, and the dual PPARa + g agonist docosahexaenoic acid (DHA) in normal and filaggrin deficient reconstructed human skin for the first time. We were particularly interested in their effects on the expression of the structural proteins filaggrin, loricrin and involucrin, potential influences on the skin lipid organization and composition and their impact on the skin barrier function aiming for deeper insights into the action of PPAR agonists in normal and diseased human skin. 2. Materials and methods 2.1. Materials DHA, WY14643, ciglitazone, GW6471 and GW9662 were purchased from Sigma–Aldrich, St. Louis, USA. All solutions for H&E staining, formaldehyde solution 4%, Tween 20 and bovine serum albumin (BSA) were obtained from Carl Roth, Karlsruhe, Germany. Filaggrin antibody was purchased from Abcam, Cambridge, United Kingdom. Horseradish-peroxidase-conjugated secondary antibody and b-actin antibody were purchased from Cell Signaling, Frankfurt/Main, Germany. The secondary antibodies IgG DyLight 488 and IgG DyLight 594 as well as DAPI antifading mounting medium were bought from Dianova, Hamburg, Germany. The RNA isolation kit NucleoSpin1 RNA II was from Macherey-Nagel, Düren, Germany, the RevertAidTM First Strand cDNA Synthesis Kit from Fermentas, St. Leon-Rot, Germany. SYBR Green I Masterplus kit for qPCR was purchased from Roche, Penzberg, Germany. All primers for qPCR were provided by TibMolbiol, Berlin, Germany. Standards for HPTLC analysis (ceramide NS, AS, NP, AP, glucosylceramide and sphingomyelin) were obtained from Avanti Polar Lipids (Alabaster, AL, USA). Ceramides NH and EOS were prepared by a total synthesis; the synthetic procedures will be reported elsewhere. The structure and the purity of the prepared ceramides NH and EOS were determined by nuclear magnetic resonance spectra (Varian, polo Alto, CA, Mercury-Vx BB 300 Instrument, operating at 300 MHz for 1H and 75 MHz for 13C), infrared spectra (Nicolet 6700 FTIR spectrophotometer, Thermos Scientific, Waltham), and mass spectrometry
(Agilent 500 Ion Trap LC/MS, Santa Clara, CA). Cholesterol, fatty acid standards, L-a-phosphatidylcholine and all other agents were purchased from Sigma–Aldrich (St. Louis, USA) if not otherwise stated. 2.2. Preparation of skin models and supplementation with DHA, WY14643 or ciglitazone Normal, wild type (WT) and filaggrin deficient (FLG siRNA) skin models were generated according to previously published procedures [24] (see Supplementary material). The tissues were cultivated for 14 days with medium change every other day. Starting at day 2, 100 mM DHA, 50 mM WY14643 [16] or 5 mM ciglitazone were added to the cell culture medium [25]. To analyze the effect of cell differentiation, keratinocytes were pre-incubated with 1.3 mM calcium chloride for 5 days. After differentiation, the agonists were added and the cells were cultured for 48 h. 2.3. Histology, immunofluorescence and qPCR The models were embedded in tissue freezing medium, shockfrozen using liquid nitrogen, subsequently cut to vertical slices (5 mm) and stained with haematoxylin-eosin or immunofluorescence according to standard protocols. For RNA analysis, the skin models were punched to 10 mm discs, the epidermis was gently removed, frozen and then milled for 30 s at 25 Hz using a TissueLyzer (Qiagen, Hilden, Germany). Subsequently, RNA was isolated using NucleoSpin1 RNA II according to the manufacturer’s instructions. For cDNA synthesis, RevertAidTM First Strand cDNA Synthesis Kit was used. Subsequently, qPCR was performed using the SYBR Green I Masterplus kit. The primer sequences are listed in Table S1. YWHAZ served as house-keeping gene. 2.4. Western blot Western blot (WB) was performed according to standard protocols. Briefly, the skin models were lysed in radioimmunoprecipitation assay buffer and total protein concentrations were determined with the Pierce1 BCA Protein Assay Kit (Thermo Scientific, Waltham, USA). Subsequently, the samples (30 mg protein) were boiled in standard SDS-PAGE sample buffer and separated by 10% SDS polyacrylamide gel electrophoresis (Bio-Rad, Munich, Germany). The gels were blotted onto nitrocellulose membranes (Immobilon P, Carl Roth, Karlsruhe, Germany). After blocking with 5% skimmed-milk powder for 1 h at 37 C, the membranes were incubated with the primary filaggrin antibody at a concentration of 1:1.000 overnight at 4 C. Afterwards, the blots were washed and incubated with horseradish-peroxidase-conjugated secondary antibody for 1 h and subsequently, developed with SignalFireTM ECL reagent (Cell Signaling, Frankfurt/Main, Germany) and visualized by PXi/PXi Touch gel imaging system (Syngene, Cambridge, UK). 2.5. Receptor specificity of WY14643 and ciglitazone To verify PPAR specificity, cell culture studies with normal human keratinocytes were performed. Therefore, the cells were pre-incubated for 1 h with 5 mM of the selective PPARa-antagonist GW6471 or with 1 mM of the PPARg-specific antagonist GW9662. Subsequently, WY14643 or ciglitazone were added for 48 h. 2.6. SC isolation The models were placed on a filter paper soaked with 0.5% trypsin in phosphate-buffered saline (PBS, pH 7.4). The isolated SC
Please cite this article in press as: L. Wallmeyer, et al., Stimulation of PPARa normalizes the skin lipid ratio and improves the skin barrier of normal and filaggrin deficient reconstructed skin, J Dermatol Sci (2015), http://dx.doi.org/10.1016/j.jdermsci.2015.09.012
G Model DESC 2897 No. of Pages 9
L. Wallmeyer et al. / Journal of Dermatological Science xxx (2015) xxx–xxx
sheets were washed with PBS and any remaining keratinocytes were removed with a cotton swab. Subsequently, SC sheets were washed with acetone to remove surface contaminants, vacuumdried and stored at 20 C. Isolated human SC served as control. 2.7. FTIR spectroscopy IR spectra of the samples were collected on a Nicolet 6700 FT-IR spectrometer (Thermo Fisher Scientific, Waltham, USA) equipped with a single-reflection MIRacle attenuated total reflectance (ATR) germanium crystal at 23 C. The spectra were generated by coaddition of 256 scans collected at 4 cm 1 resolution and analyzed with the Bruker OPUS software. The exact peak positions were determined from second derivative spectra and by peak fitting if needed.
3
were dissolved in 100 ml CHCl3/MeOH 2:1. 10–30 ml of each lipid sample was sprayed on the plate using a Linomat IV (Camag, Muttenz, Switzerland). Standard lipids were dissolved in CHCl3/ MeOH 2:1 (v/v) (ceramides EOS, NS, AS, NP, NH, AP, cholesterol, and lignoceric acid), and CHCl3/MeOH 1:1 (v/v) (cholesterol sulfate, glucosylceramide, sphingomyelin and phospholipids), respectively, at 1 mg/ml. To generate calibration curves, the lipids were mixed in ratios that approximately corresponds to the composition of human SC [27]. For more detailed information, see Supplementary material. 2.10. HPLC analysis of FFA FFA were analyzed as their naphthacyl esters by a method based on Refs. [28] and [29]. For a detailed description, see Supplementary material.
2.8. Isolation of SC lipids 2.11. Skin absorption testing For the extraction of the SC intercellular lipids, a modified [26] method was used. The SC samples were extracted with 1 ml CHCl3/ MeOH 2:1 (v/v) per mg of SC for 1.5 h, filtered, separated and concentrated under a stream of nitrogen. The lipids were dried and stored at 20 C under argon. 2.9. HPTLC lipid analysis The lipid analysis was performed on silica gel 60 HPTLC plates (20 10 cm, Merck, Darmstadt, Germany). The extracted SC lipids
The skin permeability was evaluated according to validated test procedures [22,30]. Briefly, stock solutions of testosterone (40 mg/ml, 2% [v/v] Igepal1 CA-630, Sigma–Aldrich, St. Louis, USA) were spiked with an appropriate amount of the radiolabeled compound to achieve a total radioactivity of 2 mCi/ml. Permeation studies were performed at day 14 using a static setup (Franz type diffusion cells, diameter 15 mm, volume 12 ml, Permegear, Bethlehem, USA). The total amounts of the permeated test compound were quantified using radiochemical detection
Fig. 1. (a) Filaggrin expression in normal (WT) and filaggrin deficient (FLG siRNA) skin models following treatment with the dual PPARa + g agonist DHA, the PPARa agonist WY14643, and the PPARg agonist ciglitazone. Mean SEM, n = 4–5, * indicates statistical significance (p 0.05). (b) Western blot of filaggrin in normal (WT) and filaggrin deficient (FLG siRNA) skin models after supplementation with DHA, WY14643, or ciglitazone. (c) Representative immunostaining images against filaggrin in normal (WT) and filaggrin deficient (FLG siRNA) skin models and after supplementation with DHA, WY14643, or ciglitazone. SC: stratum corneum, Epi: viable epidermis, D: dermis equivalent, n = 3.
Please cite this article in press as: L. Wallmeyer, et al., Stimulation of PPARa normalizes the skin lipid ratio and improves the skin barrier of normal and filaggrin deficient reconstructed skin, J Dermatol Sci (2015), http://dx.doi.org/10.1016/j.jdermsci.2015.09.012
G Model DESC 2897 No. of Pages 9
4
L. Wallmeyer et al. / Journal of Dermatological Science xxx (2015) xxx–xxx
(Microbeta Plus, Wallac, Turku, Finland). The rate of testosterone permeation was depicted as the apparent permeability coefficient (Papp).
is well in line with previous findings [31,32]. Additionally, significantly increased epidermal thickness was observed with WY14643. Once more, the supplementation with ciglitazone showed no effect.
2.12. Statistical analysis 3.2. Influences of PPAR agonists on involucrin and loricrin expression The paired student’s t-test was used for direct comparisons between untreated and treated models or cells. For multiple comparisons, one-way analysis of variance, followed by Dunett's post hoc testing, was performed using GraphPad Prism 6.0 (GraphPad Software Inc., La Jolla, USA). p 0.05 (*) was considered significant. Data of at least three independent experiments are presented as means standard error of the mean (SEM). 3. Results 3.1. DHA and WY14643, but not ciglitazone, significantly enhanced filaggrin amounts Filaggrin expression in the knock down models was significantly reduced by 80%, respectively. The knock down was stable for at least 14 days [9,24]. Supplementation with DHA and WY14643 resulted in 16-fold and 2.3-fold increased filaggrin amounts in normal, wild type skin models (WT), respectively, (Fig. 1a). In filaggrin knock down models (FLG siRNA), filaggrin amounts also increased significantly about 5-fold after DHA and 3.5-fold after WY14643 treatment. In contrast, filaggrin amounts were only marginally influenced by ciglitazone. These results were also reproducible in monolayer cell cultures, in which ciglitazone treatment even reduced filaggrin in differentiated keratinocytes (Fig. S1). PPARa specificity of WY14643 was verified by inhibition studies using the PPARa selective antagonist GW6471. For ciglitazone, however, inhibition studies with the PPARg-selective antagonist GW9662 indicated off-target effects (Fig. S2). The PPAR antagonists themselves did not affect filaggrin amounts. Consistent with the qPCR results, immunofluorescence staining and WB also showed significantly increased amounts of filaggrin following DHA and WY14643 supplementation and no effects with ciglitazone (Fig. 1b and c). Furthermore, the SC thickness of FLG siRNA models significantly increased after DHA and WY14643 supplementation (Fig. 2) which
Involucrin (IVL) expression increased 2.4-fold in WT and 1.9fold in FLG siRNA models following WY14643 and ciglitazone supplementation. Loricrin (LOR) expression was induced 4.5-fold and 2.9-fold in FLG siRNA models with WY14643 and ciglitazone, respectively, but without statistical significance. DHA did not influence IVL and LOR expression (Fig. 3a and b). 3.3. DHA and WY14643 improved the SC lipid chain order in FLG SIRNA models To investigate the lipid chain order, SC samples from WT and FLG siRNA models and human skin were isolated and examined by Fourier transform infrared spectroscopy (FTIR). Recently, we showed that filaggrin deficiency alone leads to disturbed barrier lipid ratio and, consequently, lipid chain organization in reconstructed skin models [9]. The findings of the present study nicely underline these results: FLG siRNA models displayed a decreased lipid chain order with more gauche conformers as indicated by an increased methylene stretching wavenumber compared to WT (Fig. 4a and b). DHA and WY14643 treatment of FLG siRNA skin models increased the lipid chain order as reflected by a significant decrease of methylene stretching wavenumber to values found in WT skin models suggesting a normalization of the lipid organization. No significant changes in lipid chain order were found upon the treatment with ciglitazone. 3.4. DHA and WY14643 normalized the skin lipid ratio in FLG siRNA models To investigate possible reasons for the improved lipid chain order, the SC lipids of the skin models were extracted and analyzed by HPTLC. WT models contained less lipids per mg SC than human SC, but the major barrier lipids (ceramides/FFA/cholesterol) were present in an approximately 1:1:1 molar ratio. FLG siRNA models
Fig. 2. Thickness of stratum corneum (SC) and viable epidermis (VE) in untreated normal (WT) and filaggrin deficient (FLG siRNA) skin models and after treatment with DHA, WY14643, or ciglitazone. Magnification 10, scale bar = 100 mm, mean SEM, n = 5, * indicates statistical significance over normal, untreated skin models (WT) (p 0.05), + indicates statistical significance over FLG siRNA (p 0.05).
Please cite this article in press as: L. Wallmeyer, et al., Stimulation of PPARa normalizes the skin lipid ratio and improves the skin barrier of normal and filaggrin deficient reconstructed skin, J Dermatol Sci (2015), http://dx.doi.org/10.1016/j.jdermsci.2015.09.012
G Model DESC 2897 No. of Pages 9
L. Wallmeyer et al. / Journal of Dermatological Science xxx (2015) xxx–xxx
5
Fig. 3. Effects of DHA, WY14643 and ciglitazone on involucrin (IVL) (a) and loricrin (LOR) (b) expression in normal (WT) and filaggrin deficient (FLG siRNA) skin models. Mean SEM, n = 4–6.
showed elevated FFA concentrations compared to WT (21.5 2.4 mg/mg and 13.4 1.3 mg/mg) as described previously [9]. Upon DHA and WY14643 treatment, FFA levels were reduced to 15.3 1.0 mg/mg and 9.7 1.2 mg/mg, respectively, in FLG siRNA (Fig. 5a). Following ciglitazone treatment, we found slightly but insignificantly lower FFA values compared to untreated FLG siRNA skin models. HPLC analysis of the FFA profiles showed that more than 70% of FFA in the skin models were shorter (18 carbons or less) or unsaturated (10.4 1.4 mg/mg; Fig. 5b). FLG knock down led to an increase of the levels of those shorter and unsaturated FFA (to 17.6 3.9 mg/mg) while the levels of the very long FFA (with 20 carbons and more) remained relatively constant (approximately 3 mg/mg). Supplementation of FLG siRNA models with WY14643 normalized the FFA profile to similar levels seen in the WT models. Similar effect was also found for DHA treatment but without statistical significance. Other SC lipids such as cholesterol, ceramides and cholesterol sulfate were unaffected by DHA or WY14643 treatment (Fig. 5c–e). No significant changes were observed in the individual ceramide classes (Fig. S3) and on the lipid precursors (sphingomyelin, glucosylceramides and phospholipids) (Fig. 5c–h). Nevertheless, it is worth noting that the treatment with WY14643 showed trends towards increased levels of ceramides and cholesterol sulfate and decreased levels of the lipid precursors. Thus, treating the FLG
siRNA models with DHA and, in particular, WY14643 led to a normalization of FFA content, a normalization of the skin lipid ratio and, consequentially, improved skin lipid organization. 3.5. PPARa agonist WY14643 but not DHA and ciglitazone improved the skin barrier function Finally, the skin barrier function of normal and filaggrin deficient skin models after supplementation with DHA, WY14643 or ciglitazone was determined by skin permeation studies. Interestingly, the supplementation with WY14643 significantly improved the skin barrier function of normal and FLG siRNA models as seen from significantly lower Papp values compared to the untreated controls. No improvement of the skin barrier was observed after supplementation with ciglitazone or DHA (Fig. 6, Table 1). For the latter, a tendency towards reduced barrier function was seen. 4. Discussion Filaggrin mutations are a well-known risk factor for the development of AD and the underlying cause of ichthyosis vulgaris [3]. The exact mechanism, however, leading to the clinical manifestation of the skin diseases is poorly understood and
Fig. 4. Effects of DHA, WY14643 and ciglitazone on the stratum corneum lipid chain order as derived from the methylene symmetric stretching vibrations from infrared spectra of the normal (WT) and filaggrin deficient (FLG siRNA) skin models compared to human stratum corneum (hSC): (a) representative infrared spectra and (b) wavenumbers of methylene symmetric stretching of hydrated SC samples examined by infrared spectroscopy at 23 C. Mean SEM, n 3 batches, * indicates statistical significance (p 0.05) as indicated, h indicates statistically significant differences compared to hSC (p 0.05).
Please cite this article in press as: L. Wallmeyer, et al., Stimulation of PPARa normalizes the skin lipid ratio and improves the skin barrier of normal and filaggrin deficient reconstructed skin, J Dermatol Sci (2015), http://dx.doi.org/10.1016/j.jdermsci.2015.09.012
G Model DESC 2897 No. of Pages 9
6
L. Wallmeyer et al. / Journal of Dermatological Science xxx (2015) xxx–xxx
Fig. 5. Effects of DHA, WY14643 and ciglitazone on the stratum corneum lipid profiles of normal (WT) and filaggrin deficient (FLG siRNA) skin models compared to human stratum corneum (hSC). SC lipids were extracted from the skin models and analyzed by HPTLC for the content of barrier lipids (FFA: free fatty acids, Chol: cholesterol, Cer: ceramides, CholS: cholesterol sulfate) and their precursors (SM: sphingomyelin, PL: phospholipids, GCer: glucosylceramides). FFA profiles were determined by reversed phase HPLC of their naphthacyl esters. Mean SEM, n 3 batches, * indicates statistical significance (p 0.05) as indicated, h indicates statistically significant differences compared to hSC at p 0.05.
therapeutic options targeting the underlying cause are still missing [5]. During the last decade, PPAR agonists gained interest as candidates for the treatment of filaggrin-associated and inflammatory skin diseases due to anti-inflammatory effects in models of irritant and allergic contact dermatitis [12], accelerated skin acidification [32] and stimulated skin lipid synthesis [11,31,33]. Moreover, simultaneous application of glucocorticoids and PPAR agonists counterbalance side effects of glucocorticoids such as decreased filaggrin expression and epidermal thinning [13,34]. In human skin, the isotypes PPARa, PPARb/d and PPARg are expressed. PPARb/d is present throughout all epidermal layers, whereas PPARa and PPARg are mainly located in suprabasal compartments [10,11]. PPARs are crucial for skin homeostasis and mediate e.g. keratinocyte differentiation and skin repair [25,35]. Hence, PPARs emerged as promising targets for the treatment of hyperproliferative or inflammatory skin disorders where particularly the activation of PPARa and PPARg play a role. PPARg modulates cellular differentiation and acts as an inhibitory regulator of immune cells [14,15,35]. PPARa governs cell proliferation and differentiation and exerts potent anti-inflammatory effects in AD as demonstrated in mice [13,36]. Moreover, a small
clinical trial with AD patients that were topically treated with the PPARa agonist clofibrate also showed a significant decrease of the disease severity score and chemokine levels after two weeks of topical application [37]. The specific effects on the human skin barrier function, however, are not fully understood. Hence, in this study we investigated the effects of the PPARa specific agonist WY14643, the PPARg agonist ciglitazone and the dual PPARa + g modulator DHA in normal and filaggrin deficient skin models. Filaggrin deficient skin models are characterized by disturbed epidermal maturation and differentiation and increased susceptibility to skin irritation [24,38]. Just recently, we studied the impact of filaggrin deficiency on skin lipids and found significantly increased expression of NHE-1, secretory phospholipase A2 and, consequently, increased FFA levels in the knock down model and disturbed skin lipid organization contributing to the disturbed skin barrier function of the models [9]. The presence of PPARs in reconstructed skin was previously demonstrated and increased filaggrin expression following supplementation with PPAR agonists in normal models was described [16,31]. Here, we first demonstrated that this upregulation also works in filaggrin deficient skin models after supplementation with DHA and
Please cite this article in press as: L. Wallmeyer, et al., Stimulation of PPARa normalizes the skin lipid ratio and improves the skin barrier of normal and filaggrin deficient reconstructed skin, J Dermatol Sci (2015), http://dx.doi.org/10.1016/j.jdermsci.2015.09.012
G Model DESC 2897 No. of Pages 9
L. Wallmeyer et al. / Journal of Dermatological Science xxx (2015) xxx–xxx
7
Fig. 6. Skin permeability of untreated, normal (WT) ( ) and filaggrin deficient (FLG siRNA) (&) skin models following the supplementation with WY14643 (* WT/ + WY14643, & FLG siRNA/ + WY14643) (a), DHA (* WT/ + DHA, & FLG siRNA/ + DHA) (b), and ciglitazone (* WT/ + ciglitazone, & FLG siRNA/ + ciglitazone) (c). Mean SEM, n = 3–5.
WY14643 (Fig. 1). Selective PPARa and g stimulation also increased loricrin and involucrin expression—other important structural proteins of the cornified envelope (Fig. 3 and S3). DHA was without effect. Well in line with previous findings, we also observed a thickening of the SC after DHA and WY14643 supplementation [31,32] and increased epidermal thickness with WY14643. In terms of the skin barrier lipids, we confirmed our previous finding that filaggrin deficient skin models display less ordered lipid chains and a disturbed ratio of the main barrier lipids (FFA: ceramides:cholesterol) compared to WT models [9]. Upon treatment of the FLG siRNA models with DHA and WY14643, the levels of FFA in SC decreased leading to lipid ratio and lipid chain order values similar to those in the WT models. PPARg agonist ciglitazone, which was not able to upregulate filaggrin, did not change the SC lipid profile of FLG siRNA. To obtain deeper insights into the changes of the FFA profiles by the action of WY14643 and DHA, or more precisely how could an increase in FFA levels decrease lipid chain order, we determined the proportions of very long FFA (with 20 carbons and more) and shorter (with 18 carbons or less) or unsaturated FFA. The FFA pool in all the skin models was mainly (over 70%) shorter or unsaturated, which could explain the lower lipid order compared to human SC. This altered FFA profile in the skin models compared to human SC is in agreement with [39]. In FLG siRNA models, the proportion of the very long FFA did not change compared to WT,
Table 1 Papp values of the model drug testosterone in untreated, normal (WT) and FLGdeficient (FLG siRNA) skin models after treatment with DHA, WY14643 or ciglitazone. * indicates statistical significance (p 0.05). Papp (x10
WT FLG siRNA
6
cm/s) for testosterone
Control
DHA
WY14643
Ciglitazone
3.4 0.9 3.2 0.7
7.6 0.9 5.6 1.1
1.7 0.1* 1.8 0.5
3.0 0.9 2.9 1.1
but the levels of the shorter or unsaturated FFA increased. Again, this is in good agreement with the lower lipid chain order of the FLG siRNA compared to WT and with our previous finding [9]. Increased levels of shorter and unsaturated FFA were previously found in AD patients (although without association with FLG mutations) and their negative effects on the lipid order was proved, too [40]. Treatment of FLG siRNA models with WY14643 (and also with DHA, but without statistical significance) decreased the levels of shorter/unsaturated FFA to the values seen in WT models. Consequentially, the skin lipid ratio and, thus, the skin lipid organization improved significantly (Fig. 4). It is well established that the stimulation of PPAR receptors affects the skin lipid metabolism. Rivier and coworkers found a dose-dependent increase of cholesterol sulfate and ceramide levels in reconstructed skin models after supplementation with the PPARa agonist WY14643 [31]. At the same concentration of WY14643 that we used here, they observed slightly but insignificantly increased levels of ceramides (in particular of ceramide NS) and cholesterol sulfate, which is very similar to our findings. We did not observe any detrimental effects of DHA on other skin lipid classes (Fig. 5). This is noteworthy since the (topical) application of fatty acids can adversely affect the lamellar body formation and skin barrier function [41]. Ultimately, the effects of the PPAR agonists on skin barrier function were investigated. For WY14643, significantly decreased permeability for the model drug testosterone was observed indicating an improvement of skin barrier (Fig. 6). Interestingly, despite filaggrin upregulation, SC thickening, and amended skin lipid ratio, the skin barrier was not improved by DHA. Potential explanations are apoptotic effects of DHA on proliferative cells hampering epidermal differentiation [42] or the lack of effect on loricrin and involucrin expression (Fig. 3a and b). The ineffectiveness of ciglitazone observed in our study was concordant with previously published data in healthy [43] and AD mouse models [36] and might be explained with its rather low
Please cite this article in press as: L. Wallmeyer, et al., Stimulation of PPARa normalizes the skin lipid ratio and improves the skin barrier of normal and filaggrin deficient reconstructed skin, J Dermatol Sci (2015), http://dx.doi.org/10.1016/j.jdermsci.2015.09.012
G Model DESC 2897 No. of Pages 9
8
L. Wallmeyer et al. / Journal of Dermatological Science xxx (2015) xxx–xxx
PPARg binding affinity [35]. Nevertheless, PPARg activation efficiently suppressed dendritic cell and nuclear factor kB mediated immune responses [14,15] implying strong anti-inflammatory effects which, however, cannot be studied in reconstructed skin models so far. In conclusion, the effects of PPAR agonists on the skin barrier were investigated in filaggrin deficient skin for the first time. Well in line with previous studies, PPARa agonist WY14643 upregulated filaggrin and promoted skin differentiation. We could further show that WY14643 normalized the disturbed FFA profile present in filaggrin deficient skin models by reducing the levels of short and unsaturated FFA to levels found in control skin models which ultimately resulted in significantly improved skin barrier function. PPARg agonist ciglitazone and the dual agonist DHA failed in terms of skin barrier function improvement. Funding sources Financial support by the Foundation for the Promotion of Alternate and Complementary Methods to Reduce Animal Testing (Foundation SET) is gratefully acknowledged (S.H., L.W.). This work was further supported by the Czech Science Foundation (project No. 13-23891S) and Charles University (SVV 260 183). Conflict of interest The authors state no conflict of interest. Acknowledgement We greatly acknowledge the scientific support of Priv. Doz. Dr. Joachim Fluhr, Charité Berlin. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j. jdermsci.2015.09.012. References [1] G.M. O’Regan, A. Sandilands, W.H. McLean, A.D. Irvine, Filaggrin in atopic dermatitis, J. Allergy Clin. Immunol. 124 (2009) R2–R6. [2] S.J. Brown, W.H. McLean, One remarkable molecule: filaggrin, J. Invest. Dermatol. 132 (2012) 751–762. [3] C.N. Palmer, A.D. Irvine, A. Terron-Kwiatkowski, Y. Zhao, H. Liao, S.P. Lee, et al., Common loss-of-function variants of the epidermal barrier protein filaggrin are a major predisposing factor for atopic dermatitis, Nat. Genet. 38 (2006) 441–446. [4] M.D. Howell, B.E. Kim, P. Gao, A.V. Grant, M. Boguniewicz, A. Debenedetto, et al., Cytokine modulation of atopic dermatitis filaggrin skin expression, J. Allergy Clin. Immunol. 120 (2007) 150–155. [5] P.M. Elias, Therapeutic implications of a barrier-based pathogenesis of atopic dermatitis, Ann. Dermatol. 22 (2010) 245–254. [6] S. Kezic, Function of filaggrin and its metabolites, in: J.P. Thyssen, H.I. Maibach (Eds.), Filaggrin, Springer, 2014, 2015, pp. 3–8. [7] M.A. McAleer, A.D. Irvine, The multifunctional role of filaggrin in allergic skin disease, J. Allergy Clin. Immunol. 131 (2013) 280–291. [8] H. Kawasaki, K. Nagao, A. Kubo, T. Hata, A. Shimizu, H. Mizuno, et al., Altered stratum corneum barrier and enhanced percutaneous immune responses in filaggrin-null mice, J. Allergy Clin. Immunol. 129 (2012) 1538–1546 e1536. [9] K. Vávrová, D. Henkes, K. Strüver, M. Sochorová, B. Školová, M.Y. Witting, et al., Filaggrin deficiency leads to impaired lipid profile and altered acidification pathways in a 3D skin construct, J. Invest. Dermatol. 134 (2014) 746–753. [10] L. Michalik, W. Wahli, Peroxisome proliferator-activated receptors (PPARs) in skin health, repair and disease, Biochim. Biophys. Acta 1771 (2007) 991–998. [11] M. Schmuth, Y.J. Jiang, S. Dubrac, P.M. Elias, K.R. Feingold, Thematic review series: skin lipids. Peroxisome proliferator-activated receptors and liver X receptors in epidermal biology, J. Lipid Res. 49 (2008) 499–509. [12] M.Y. Sheu, A.J. Fowler, J. Kao, M. Schmuth, K. Schoonjans, J. Auwerx, et al., Topical peroxisome proliferator activated receptor-alpha activators reduce inflammation in irritant and allergic contact dermatitis models, J. Invest. Dermatol. 118 (2002) 94–101. [13] Y. Hatano, P.M. Elias, D. Crumrine, K.R. Feingold, K. Katagiri, S. Fujiwara, Efficacy of combined peroxisome proliferator-activated receptor-alpha ligand and glucocorticoid therapy in a murine model of atopic dermatitis, J. Invest. Dermatol. 131 (2011) 1845–1852.
[14] A. Mastrofrancesco, D. Kovacs, M. Sarra, E. Bastonini, G. Cardinali, N. Aspite, et al., Preclinical studies of a specific PPARgamma modulator in the control of skin inflammation, J. Invest. Dermatol. 134 (2014) 1001–1011. [15] K. Jung, A. Tanaka, H. Fujita, A. Matsuda, K. Oida, K. Karasawa, et al., Peroxisome proliferator-activated receptor gamma-mediated suppression of dendritic cell function prevents the onset of atopic dermatitis in NC/Tnd mice, J. Allergy Clin. Immunol. 127 (2011) 420–429 e421–426. [16] C. Zhang, I. Gurevich, B.J. Aneskievich, Organotypic modeling of human keratinocyte response to peroxisome proliferators, Cells Tissues Organs 196 (2012) 431–441. [17] M. Leist, T. Hartung, Inflammatory findings on species extrapolations: humans are definitely no 70-kg mice, Altex 30 (2013) 227–230. [18] M.O. Danso, V. van Drongelen, A. Mulder, J. van Esch, H. Scott, J. van Smeden, et al., TNF-alpha and Th2cytokines induce atopic dermatitis-like features on epidermal differentiation proteins and stratum corneum lipids in human skin equivalents, J. Invest. Dermatol. 134 (2014) 1941–1950. [19] D. van, V. rongelen, M. Alloul-Ramdhani, M.O. Danso, A. Mieremet, A. Mulder, S. van, J. meden, et al., Knock-down of filaggrin does not affect lipid organization and composition in stratum corneum of reconstructed human skin equivalents, Exp. Dermatol. 22 (2013) 807–812. [20] E.H. van den Bogaard, G.S. Tjabringa, I. Joosten, M. Vonk-Bergers, E. van Rijssen, H.J. Tijssen, et al., Crosstalk between keratinocytes and T cells in a 3D microenvironment: a model to study inflammatory skin diseases, J. Invest. Dermatol. 134 (2014) 719–727. [21] S. Küchler, K. Strüver, W. Friess, Reconstructed skin models as emerging tools for drug absorption studies, Expert Opin. Drug Metab. Toxicol. 9 (2013) 1255–1263. [22] M. Schäfer-Korting, U. Bock, W. Diembeck, H.J. Dusing, A. Gamer, E. HaltnerUkomadu, et al., The use of reconstructed human epidermis for skin absorption testing: results of the validation study, Altern. Lab. Anim.: ATLA 36 (2008) 161–187. [23] S.N. Kolle, K.M. Sullivan, A. Mehling, B. van Ravenzwaay, R. Landsiedel, Applicability of in vitro tests for skin irritation and corrosion to regulatory classification schemes: substantiating test strategies with data from routine studies, Regul. Toxicol. Pharmacol.: RTP 64 (2012) 402–414. [24] S. Küchler, D. Henkes, K.M. Eckl, K. Ackermann, J. Plendl, H.C. Korting, et al., Hallmarks of atopic skin mimicked in vitro by means of a skin disease model based on FLG knock-down, Altern. Lab. Anim.: ATLA 39 (2011) 471–480. [25] M. Mao-Qiang, A.J. Fowler, M. Schmuth, P. Lau, S. Chang, B.E. Brown, et al., Peroxisome-proliferator-activated receptor (PPAR)-gamma activation stimulates keratinocyte differentiation, J. Invest. Dermatol. 123 (2004) 305–312. [26] E.G. Bligh, W.J. Dyer, A rapid method of total lipid extraction and purification, Can. J. Biochem. Physiol. 37 (1959) 911–917. ková, M. Pospíšilová, B. Školová, J. Zbytovská, K. [27] P. Pullmannová, K. Stan Vávrová, Effects of sphingomyelin/ceramide ratio on the permeability and microstructure of model stratum corneum lipid membranes, Biochim. Biophys. Acta (BBA)—Biomembr. 1838 (2014) 2115–2126. [28] V. Rioux, D. Catheline, M. Bouriel, P. Legrand, High performance liquid chromatography of fatty acids as naphthacyl derivatives, Analusis 27 (1999) 186–193. [29] R.F. Borch, Separation of long chain fatty acids as phenacyl esters by high pressure liquid chromatography, Anal. Chem. 47 (1975) 2437–2439. [30] OECD, Test No. 428: Skin Absorption: In Vitro Method, OECD Publishing, 2015. [31] M. Rivier, I. Castiel, I. Safonova, G. Ailhaud, S. Michel, Peroxisome proliferatoractivated receptor-alpha enhances lipid metabolism in a skin equivalent model, J. Invest. Dermatol. 114 (2000) 681–687. [32] J.W. Fluhr, M.Q. Man, J.P. Hachem, D. Crumrine, T.M. Mauro, P.M. Elias, et al., Topical peroxisome proliferator activated receptor activators accelerate postnatal stratum corneum acidification, J. Invest. Dermatol. 129 (2009) 365–374. [33] M. Schmuth, V. Moosbrugger-Martinz, S. Blunder, S. Dubrac, Role of PPAR, LXR, and PXR in epidermal homeostasis and inflammation, Biochim. Biophys. Acta 1841 (2014) 463–473. [34] M. Demerjian, E.H. Choi, M.Q. Man, S. Chang, P.M. Elias, K.R. Feingold, Activators of PPARs and LXR decrease the adverse effects of exogenous glucocorticoids on the epidermis, Exp. Dermatol. 18 (2009) 643–649. [35] Y. Yan, M. Furumura, S. Numata, K. Teye, T. Karashima, B. Ohyama, et al., Various peroxisome proliferator-activated receptor (PPAR)-gamma agonists differently induce differentiation of cultured human keratinocytes, Exp. Dermatol. 24 (2014) 62–65. [36] T. Chiba, S. Takeuchi, H. Esaki, K. Yamamura, Y. Kurihara, Y. Moroi, et al., Topical application of PPARalpha (but not beta/delta or gamma) suppresses atopic dermatitis in NC/Nga mice, Allergy 67 (2012) 936–942. [37] M. Fukaya, H. Kimata, Topical clofibrate improves symptoms in patients with atopic dermatitis and reduces serum TARC levels: a randomized, double-blind, placebo-controlled pilot study, J. Drugs Dermatol.: JDD 13 (2014) 259–263. [38] M. Mildner, J.A. Jin, L. Eckhart, S. Kezic, F. Gruber, C. Barresi, et al., Knockdown of filaggrin impairs diffusion barrier function and increases UV sensitivity in a human skin model, J. Invest. Dermatol. 130 (2010) 2286–2294. [39] J. van Smeden, W.A. Boiten, T. Hankemeier, R. Rissmann, J.A. Bouwstra, R.J. Vreeken, Combined LC/MS-platform for analysis of all major stratum corneum lipids: and the profiling of skin substitutes, Biochim. Biophys. Acta 1841 (2014) 70–79.
Please cite this article in press as: L. Wallmeyer, et al., Stimulation of PPARa normalizes the skin lipid ratio and improves the skin barrier of normal and filaggrin deficient reconstructed skin, J Dermatol Sci (2015), http://dx.doi.org/10.1016/j.jdermsci.2015.09.012
G Model DESC 2897 No. of Pages 9
L. Wallmeyer et al. / Journal of Dermatological Science xxx (2015) xxx–xxx [40] J. van Smeden, M. Janssens, E.C. Kaye, P.J. Caspers, A.P. Lavrijsen, R.J. Vreeken, et al., The importance of free fatty acid chain length for the skin barrier function in atopic eczema patients, Exp. Dermatol. 23 (2014) 45–52. [41] M. Mao-Qiang, K.F. Feingold, P.M. Elias, Exogenous lipids influence permeability barrier recovery in acetone-treated murine skin, Arch. Dermatol. 129 (1993) 728–738.
9
[42] S. Serini, E. Piccioni, N. Merendino, G. Calviello, Dietary polyunsaturated fatty acids as inducers of apoptosis: implications for cancer, Apoptosis: Int. J. Programmed Cell Death 14 (2009) 135–152. [43] M.Q. Man, E.H. Choi, M. Schmuth, D. Crumrine, Y. Uchida, P.M. Elias, et al., Basis for improved permeability barrier homeostasis induced by PPAR and LXR activators: liposensors stimulate lipid synthesis, lamellar body secretion, and post-secretory lipid processing, J. Invest. Dermatol. 126 (2006) 386–392.
Please cite this article in press as: L. Wallmeyer, et al., Stimulation of PPARa normalizes the skin lipid ratio and improves the skin barrier of normal and filaggrin deficient reconstructed skin, J Dermatol Sci (2015), http://dx.doi.org/10.1016/j.jdermsci.2015.09.012