Journal of Dermatological Science 77 (2015) 28–36

Contents lists available at ScienceDirect

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

Epidermal tight junction barrier function is altered by skin inflammation, but not by filaggrin-deficient stratum corneum Mariko Yokouchi a,b, Akiharu Kubo a,c,*, Hiroshi Kawasaki a,d, Kazue Yoshida a,e, Ken Ishii f, Mikio Furuse g, Masayuki Amagai a,c,h a

Department of Dermatology, Keio University School of Medicine, 35 Shinanomachi, Shinjuku, Tokyo 160-8582, Japan Nerima General Hospital, 1-24-1 Asahigaoka, Nerima-ku, Tokyo 176-8530, Japan Keio-Maruho Laboratory of Skin Barriology, Keio University School of Medicine, Tokyo, Japan d Kitasato University Kitasato Institute Hospital, 5-9-1 Shirokane, Minato-Ku, Tokyo 108-8642, Japan e Department of Dermatology, National Center for Child Health and Development, 2-10-1 Okura, Setagaya, Tokyo 157-8535, Japan f Department of Dermatology, School of Medicine Toho University, 5-21-16 Omori-nishi, Tokyo 143-8540, Japan g Division of Cerebral Structure, National Institute for Physiological Sciences, 5-1 Higashiyama, Myodaiji, Okazaki 444-8787, Japan h KOSE´ Endowed Program for Skin Care and Allergy Prevention, Keio University School of Medicine, Tokyo, Japan b c

A R T I C L E I N F O

A B S T R A C T

Article history: Received 17 October 2014 Received in revised form 12 November 2014 Accepted 13 November 2014

Background: The tight junction (TJ) barrier is located in the granular layer of the epidermis. Filaggrin deficiency predisposes patients to atopic dermatitis (AD) by impairing stratum corneum (SC) barrier function. Altered TJ barrier function has been observed in the skin of patients with AD; however, it remains unclear whether TJ function is influenced by filaggrin deficiency directly or secondarily via skin inflammation. Objective: To investigate the in vivo effects of filaggrin deficiency and skin inflammation on epidermal TJ function. Methods: Morphological changes in the TJ were investigated in filaggrin knockout mice and mice with hapten-induced dermatitis using en face visualization of epidermal sheets, and functional changes in the TJ were assessed with an in vivo permeation assay using tracers of various sizes. Results: In filaggrin knockout mice, there was no apparent change in the honeycomb morphology of the TJ, TJ component mRNA expression, or TJ barrier function in neonates and adults, indicating that filaggrin-deficiency had no direct effects on the TJ. By contrast, in mice with hapten-induced dermatitis, the mRNA expression of TJ components was decreased markedly and the TJ barrier function was size-dependently impaired: the TJ leaked small tracers (30 kDa). Conclusion: Filaggrin deficiency did not affect the epidermal TJ barrier directly, but once dermatitis occurred, the skin inflammation induced TJ dysfunction. Since TJ dysfunction induces the SC barrier impairment, skin inflammation will enhance skin permeability to external antigens and result in a vicious cycle of barrier dysfunction and skin inflammation. ß 2014 Japanese Society for Investigative Dermatology. Published by Elsevier Ireland Ltd. All rights reserved.

Keywords: Atopic dermatitis Tight junction Stratum corneum Barrier deficiency Filaggrin

1. Introduction In mammalian skin, the epidermis has two sets of physical barriers, the stratum corneum (SC) and tight junctions (TJs). These two barriers prevent the outside-in penetration of external antigens or the inside-out leakage of internal constituents. The SC, which is composed of terminally differentiated cornified cells

* Corresponding author at: Department of Dermatology, Keio University School of Medicine, Shinanomachi 35, Shinjuku, Tokyo 160-8582, Japan. Tel.: +81 3 53633426; fax: +81 3 53633426. E-mail addresses: [email protected], [email protected] (A. Kubo).

(corneocytes) and intercorneocyte lipids, serves as an air–liquid interface barrier to protect the viable cell layer underneath [1,2]. Under the SC, stratum granulosum (SG) cells form the outer layers of viable stratified keratinocytes. Naming the uppermost three SG layers SG1, SG2, and SG3 from the surface inward, TJs seal the intercellular spaces of the SG2 cells and limit the movement of water-soluble molecules through paracellular pathways [2–5]. Atopic dermatitis (AD) is a chronic relapsing eczematous skin disorder that is frequently associated with elevated serum IgE levels and a family history of AD, allergic rhinitis, or asthma [6]. Recent studies of Netherton syndrome [7,8], peeling skin syndrome type B [9], and filaggrin deficiency [10–12] have suggested that

http://dx.doi.org/10.1016/j.jdermsci.2014.11.007 0923-1811/ß 2014 Japanese Society for Investigative Dermatology. Published by Elsevier Ireland Ltd. All rights reserved.

M. Yokouchi et al. / Journal of Dermatological Science 77 (2015) 28–36

impairment of the SC barrier predisposes patients to AD via increased percutaneous sensitization [2,13]. In this ‘‘impaired SC barrier model’’ of AD, it is postulated that the external antigens that penetrate the SC are taken up by skin dendritic cells for antigen presentation to the immune system [14]. External water-soluble antigens are believed to be held outside the TJ barrier and taken up by epidermal Langerhans cells (LCs) via the inside-out TJ penetration of LC dendrites for antigen presentation to T cells [2,15,16]. Once the TJ barrier is abrogated, these external antigens are ready to penetrate the dermis, where they are taken up by dermal dendritic cells [2,14,17,18]. Therefore, impaired epidermal TJ function is thought to affect the mode of the immune reaction with a distinct population of dendritic cells that take up external antigens. Some reports have investigated TJs in human AD skin or filaggrin-deficient ichthyosis vulgaris (IV) skin [19,20]. In the nonlesional skin of AD patients, reduced expression of claudin-1, a major transmembrane protein of epidermal TJs, has been reported [19]. An ex vivo analysis of blister roof epidermis containing both the SC and TJ barriers using Ussing chambers revealed a marked reduction in trans-blister roof resistance in AD patients, although it was not clear which barrier defects (i.e., the SC, TJ, or both) caused the decrease in trans-blister roof resistance [19]. In filaggrindeficient IV patients, immunofluorescence studies demonstrated decreased staining for occludin and zonula occludens-1 (ZO-1), and ex vivo electron microscopic studies demonstrated increased lanthanum permeation into the paracellular spaces of the SG and SC [20]. While these findings suggest impaired function of the epidermal TJ barrier in filaggrin-deficient IV patients and AD patients, it is unclear whether the TJ dysfunction is induced directly by filaggrin deficiency alone or is secondary to the skin inflammation. Here, we evaluated the in vivo effect of SC barrier impairment in filaggrin deficiency and skin inflammation on epidermal TJ function by examining the paracellular permeability of variously sized tracers in vivo in filaggrin knockout (Flg / ) mice [12] and dermatitis model mice induced with topical hapten application, which has no direct effect on TJ function per se.

2. Materials and methods 2.1. Animals The development of Flg / mice and Cldn1 / mice has been reported [3,12]. Eight-week-old female C57B6/J and Flg / mice were used to study adult mice, and neonatal mice within 24 h of birth were used to study neonatal mice. All animal protocols were approved by the Animal Ethics Review Board of Keio University and conformed to National Institutes of Health guidelines. 2.2. Hapten-induced dermatitis model mice Dermatitis was induced by the repeated topical application of 2,4-dinitrofluorobenzene (DNFB; Sigma-Aldrich, St. Louis, MO), as described previously, in NC/Nga mice [21]. Briefly, 25 mL of 0.15% DNFB in a 3:1 mixture of acetone/olive oil (150-00276; Wako Pure Chemical Industries, Ltd., Osaka, Japan) was applied to each side of both ears of female C57B6/J mice six times at 7-day intervals from 8 to 13 weeks of age under specific pathogen-free (SPF) conditions. Ear thickness was evaluated using a thickness gauge (PG-20, Teclock, Nagano, Japan), and the skin severity score of skin lesions was estimated by summing the individual scores, which were assigned a value of 0 (none), 1 (mild), 2 (moderate), or 3 (severe) for erythema, excoriation/erosion, hemorrhage, hardening, and swelling, as described previously [22], at 1 h before and 24 h after each application.

29

2.3. Antibodies The following primary antibodies were used: polyclonal antibodies against claudin-1 (ab15098; Invitrogen, Carlsbad, CA), desmoglein1 (sc20114; Santa Cruz Biotechnology, Santa Cruz, CA), and lipolysis-stimulated lipoprotein receptor (LSR [23]) at a 1:200 dilution, and monoclonal antibodies against ZO-1 (T8-754; provided by Dr. Itoh [24]), tricellulin (provided by Dr. Tsukita), occludin (MOC37 [25]), and HA (3F10, Roche Diagnostics, Mannheim, Germany). Species-specific secondary antibodies and streptavidin-labeled Alexa Fluor 488, 568, and 647 (Invitrogen) were used for detection at a 1:200 dilution. Cell nuclei were stained with Hoechst 33258 (Invitrogen). 2.4. Fluorescence microscopic observation Mouse skin samples were embedded in optimal cutting temperature compound (Sakura Finetek, Tokyo, Japan), frozen in liquid nitrogen, and sectioned using a cryostat, as described previously [4]. The frozen sections were processed immediately for immunostaining with incubation in 95% ethanol at 4 8C for 30 min, and then in 100% acetone at room temperature for 1 min before staining. Epidermal sheets were prepared from the ventral side of the mouse ear skin, as described previously with some modifications [26]. Mouse ear skin was split mechanically and the cartilage was removed. The ventral side of the ear was placed dermal side down on a solution of dispase (354235; BD Biosciences, Bedford, MA) at a concentration of 1500 U/mL in Hank’s Balanced Salt Solution (HBSS, 24020-117; Invitrogen) and incubated for 13– 16 min at room temperature. The epidermal sheet was peeled mechanically from the dermis and fixed in 95% ethanol at 4 8C for 30 min. All samples were immunostained as described previously [27]. Briefly, the samples were blocked with phosphate-buffered saline (PBS) containing 10% fetal bovine serum and 5% goat serum (Dako, Tokyo, Japan) for 30 min at room temperature, incubated with primary antibodies in blocking solution at 4 8C overnight, washed three times with PBS, and incubated with secondary antibodies in blocking solution at room temperature for 1 h. Samples were washed with PBS, mounted in Mowiol (Merck, Darmstadt, Germany), and observed using a Leica TCS sp5 laser scanning confocal microscope (Solms, Germany) equipped with a 63 objective using 0.4–0.5-mm optical slices. Three-dimensional (3D) reconstruction images were built using sp5 software (Leica). The images were processed using Adobe Photoshop CS4 (Adobe Systems Inc., San Jose, CA). Each immunofluorescence figure shows a representative image from at least five independent experiments. 2.5. Assessing TJ permeability in vivo using biotin tracers TJ permeability was assayed using the surface biotinylation technique developed by Chen et al. [28] with some modifications. First, 50 mL of 10 mg/mL Sulfo-NHS-LC-Biotin (biotin-SH [556 Da], 21335; Thermo Fisher Scientific, Rockford, IL), 30 mg/mL of BiotindPEGTM (24)-NHS (biotin-PEG1500 [1469 Da], 365441-71-0; Iris Biotech, Marktredwitz, Germany), and 50 mg/mL of Biotin-PEGNHS-5000 Da (biotin-PEG5000 [5000 Da], sc-281688; Santa Cruz Biotechnology) in PBS containing 1 mM CaCl2 were injected into the dermis on the back of the wild-type (WT) and Flg / newborn mice. In adult WT and Flg / mice, 30 mL of 10 mg/mL recombinant exfoliative toxin-A (ETA) [29] in PBS containing 1 mM CaCl2 were injected into the dermis on the ventral side of the ear, followed by the injection of biotin tracers 10 min later. After a 30-min incubation, skin samples were biopsied and embedded in optimal cutting temperature compound (Sakura Finetek), frozen in liquid nitrogen, sectioned with a cryostat, and immunostained as described above.

30

M. Yokouchi et al. / Journal of Dermatological Science 77 (2015) 28–36

2.6. Assessing in vivo TJ permeability using anti-Dsg1 scFv and ETA

apical-basolateral cell membrane polarity of claudin-1apical-basoin WT mice, as described previously in humans [4]. In Flg / mice, while SC barrier impairment has been demonstrated, no spontaneous dermatitis was induced under normal humidity and SPF conditions [12]. No obvious change was detected in Flg / mice in the distribution of ZO-1, claudin-1, or Dsg1 or in the number of SG layers having ZO-1 positive junctions, both in neonatal (Supplementary Fig. S1) and adult (Fig. 1A) mice. Next, we elucidated the 3D structure of the epidermal TJs using whole-mount staining of epidermal cellular sheets from adult mouse ear skin [27]. ZO-1, claudin-1, and occludin, another TJ transmembrane protein, were concentrated at the bicellular cellcell contacts between two SG2 cells, forming a continuous network in WT mice (Fig. 1B). Tricellulin and the angulin family protein LSR, constituents of the tricellular TJs that seal tricellular spaces [23,31,32], were concentrated at the tricellular contacts among SG2 cells in WT mice (Fig. 1C). The bicellular localization of ZO-1, occludin, and claudin-1, the tricellular localization of tricellulin and LSR, and the honeycomb pattern of the TJ network showed no apparent changes in Flg / mice (Fig. 1A–C). We measured mRNA obtained from the epidermis of WT and Flg / mice to detect any changes in the expression of TJ-related proteins. In Flg / mice, while filaggrin mRNA was barely detected, the mRNAs of other differentiation markers such as Dsg1 and involucrin showed no significant change compared with WT mice (Fig. 1D). The mRNAs of TJ-related proteins, including claudin-1, claudin-4, ZO-1, occludin, tricellulin, and LSR, showed no significant change in Flg / mice when compared with WT mice (Fig. 1D). lateral+

For the single-chain variable fragment (scFv) permeability assay, recombinant scFv against desmoglein 1 (Dsg1; defined by scFv 3–30/3 h) [30] was obtained as follows. The scFv 3–30/3 h cDNA was cloned into a pET-16b expression vector (Novagen, Madison, WI). The BL21(DE3)pLysS strain of Escherichia coli (Promega, Madison, WI) was transformed with the expression vector and then cultured with 0.5 mM isopropylthio-b-galactoside to induce the expression of scFv and lysed using FastBreak (Promega). The insoluble fraction of the lysate was washed with 1% Triton X-100 in PBS twice and with distilled water. The scFv was solubilized from the pellet by incubating it in 4 M guanidine-HCl/ 10 mM 2-mercaptoethanol in 50 mM Tris–HCl (pH 8.0), and refolded via dialysis in PBS. Solubilized scFv (50 mL, concentration ca. 0.75 mg/mL) was injected intradermally into the backs of the newborn mice and the ventral side of the ears of adult mice. After 1 h, the skin was biopsied and prepared for immunostaining as described above. For the ETA permeability assay, 50 mL of 10 mg/mL recombinant ETA in PBS containing 1 mM CaCl2 were injected intradermally on the ventral side of the ears of adult mice [4,29]. After 1 h, the skin was biopsied and prepared for immunostaining as described above. 2.7. RNA isolation and real-time PCR analysis The ventral sides of the adult WT and Flg / mouse ears were split and the cartilage was removed, and placed dermal side down on 7.5 mL of HBSS. Five milliliters of 2.5% trypsin were added to obtain a final trypsin concentration of 1%. After a 60-min incubation at 37 8C, the ear halves were transferred to 10 mL of 80% FCS in HBSS on ice to stop the digestion. Epidermal sheets were peeled from the dermis mechanically and total RNA was purified from an 8-mm-punched epidermal sheet using an RNeasy micro kit (Invitrogen). For real-time PCR analysis, cDNA was synthesized with retrotranscriptase (SuperScript 3 First-Strand Synthesis SuperMix for qRT-PCR; Invitrogen), according to the manufacturer’s protocol. Real-time PCR analysis was conducted using Power SYBR Green PCR master mix (Applied Biosystems, Foster City, CA) and a StepOne Real-Time PCR system (Applied Biosystems). All primers (Supplementary Table 1) were designed using Primer Express software (Applied Biosystems) and the reactions were performed using the following cycling conditions: 10 min at 95 8C followed by 40 cycles of 15 s at 95 8C and 60 s at 60 8C. PCR for each template was done in duplicate in 96-well plates. The mRNA expression levels were normalized to b-actin expression based on the comparative CT method (Applied Biosystems).

3.2. Development of an in vivo TJ barrier assay in adult mouse skin To date, in vivo permeation assays to evaluate TJ barrier function have been performed with neonatal, but not adult, mouse skin, mainly because most intradermally injected tracers that permeate the intercellular spaces to reach the TJs in neonatal skin fail to permeate the lower epidermal layers in adult mice (Fig. 2A). This methodological limitation has long hampered detailed study of epidermal TJ function in adult mouse skin. We overcame this difficulty by using ETA, a 31-kDa exotoxin produced by Staphylococcus aureus that specifically cleaves the extracellular domain of Dsg1 [33]. ETA specifically digests Dsg1 only under the TJ barrier when injected intradermally because ETA does not go beyond the TJ [4,16]. When we pre-injected recombinant ETA at a relatively low concentration that was not sufficient to cause microscopic blisters in the upper layers per se, biotin-SH was able to diffuse into the paracellular spaces in adult mouse epidermis (Fig. 2). Therefore, ETA pretreatment enabled us to evaluate the occlusive function of TJs in adult mouse skin. 3.3. No apparent loss of TJ function in neonatal and adult Flg

/

mice

3. Results 3.1. No apparent changes in TJ morphology or the expression of TJ components in Flg / mice We examined the morphological changes in epidermal TJs in neonatal and adult Flg / mice. First, we stained vertical sections of WT mouse skin. The desmosomal protein Dsg1 was distributed over the entire surface of SG1–SG3 cells, whereas claudin-1, a TJ transmembrane protein, was localized on the basolateral cell membrane of SG2 cells and was concentrated at the apical junctions, and the entire surface of SG3 cells (adult mouse, Fig. 1A; neonatal mouse, Supplementary Fig. S1). ZO-1, a cytoplasmic plaque protein of TJs, was specifically concentrated at the apical junctions between SG2 cells in adult (Fig. 1A) and neonatal (Supplementary Fig. S1) mice. Therefore, only SG2 cells showed

To elucidate the barrier functions of TJs in Flg / mice, we performed in vivo epidermal TJ function assays with the intradermal injection of a series of protein biotinylation reagents conjugated with polyethylene glycol (PEG) of various sizes: 556 Da (biotin-SH) [3,28], 1469 Da (biotin-PEG1500), and 5000 Da (biotin-PEG5000). When the tracers were injected intradermally into neonatal Flg / mice skin or into adult Flg / mice skin with ETA pretreatment, the tracers diffused into the intercellular spaces from the stratum basale to the SG2 layer, but stopped abruptly at the ZO-1-positive TJs of SG2 cells, just as in WT mice (biotin-SH in adult mice, Fig. 2B; biotin-SH in neonatal mice, Supplementary Fig. S2; others, data not shown). These findings indicate that the occlusive function of epidermal TJs showed no apparent change in terms of these water-soluble molecules in both neonate and adult Flg / mice.

M. Yokouchi et al. / Journal of Dermatological Science 77 (2015) 28–36

31

Fig. 1. Morphological assessment of epidermal TJs in adult Flg / mice. Distribution of TJ components in vertical sections (A) and epidermal sheets (B, C) of WT and Flg / mice. White arrows in (A) indicate ZO-1-positive junctions. In (A), the numbers 1, 2, and 3 indicate SG1, SG2, and SG3 cells, respectively. (D) Expression levels of TJ components and epidermal differentiation markers in epidermal sheets of WT and Flg / mice. Bars, 20 mm.

3.4. Attenuated expression of TJ components in mice with hapten-induced chronic dermatitis Since no significant change was observed in the morphology or function of the epidermal TJ barrier in Flg / mice, we examined whether TJ dysfunction is induced by skin inflammation. To address the question, we used hapten-induced chronic dermatitis model (HD model) mice using the C57B6/J strain with repeated topical application of DNFB [21], which does not show any direct effect on TJs with a one-time application (data not shown). The HD model mice developed chronic dermatitis-like skin lesions, including erythema followed by edema, excoriation, and scaling (Fig. 3A). Histological examinations of the affected skin showed acanthosis and hyperkeratosis with the infiltration of lymphocytes, eosinophils, and macrophages (data not shown). Immunofluorescence studies revealed a decrease in the intensity of claudin-1 cell surface staining (Fig. 3B). ZO-1 was restricted to cell junctions in the SG2 cell layer in control mice, while it was detected not only on the TJs of the SG2 cell layer, but

also on the apical junctions of SG2–SG3 and SG3–SG3 cells at low intensity in the HD model mice (Fig. 3B, lower panels). When the mRNA expression of TJ-related proteins was examined in the HD model mice, the expression of claudin-1, claudin-4, occludin, tricellulin, and LSR, but not that of ZO-1, decreased significantly in the epidermis, while the expression of filaggrin increased (Fig. 3C). 3.5. Impaired TJ barrier function in hapten-induced dermatitis model mice Next, we performed the in vivo TJ barrier function assay described above in the HD model mice. The diffusion of all three tracers of different molecular weights from the dermis (556, 1469, and 5000 Da) stopped at the TJ (ZO-1-positive junctions of the claudin-1apical-basolateral+ SG2 cells) in the untreated mice. In sharp contrast, all of these tracers permeated through several layers of ZO-1-positive junctions in the HD model mice, and reached the bottom of the SC (biotin-PEG1500, Fig. 4A; others, data not shown),

32

M. Yokouchi et al. / Journal of Dermatological Science 77 (2015) 28–36

Fig. 2. Functional assessment of epidermal TJs in adult Flg / mice. Infiltration of intradermally injected biotin-SH tracers into the paracellular spaces of the epidermis in control (ETA [ ]) and ETA-treated (ETA [+]) WT adult mice (A and B), and in ETA-treated (ETA [+]) Flg / mice (B). Bars, 20 mm (A) and 10 mm (B).

indicating that the epidermal TJ barrier became leaky in the HD model mice. As the barrier property of the TJ is dependent on the size, shape, or electrical charge of tracers, we evaluated whether the barrier function of the TJ was abrogated against larger molecules such as proteins in HD model skin. We used ETA (31 kDa) and anti-Dsg1 scFv 3-30/3h (32 kDa) as tracers. In normal human skin, both the ETA-dependent digestion of Dsg1 and scFv-dependent cell surface labeling have been observed to occur only beneath the TJ after their

intradermal injection [4,16]. When ETA was injected intradermally in WT control adult mice, the extracellular Dsg1 signals were markedly reduced beneath the TJ compared with non-injected controls (Fig. 4B compared with Fig. 1A), indicating that ETA digests Dsg1 only beneath the TJ barrier. When ETA was injected intradermally in the skin of HD model mice, the extracellular Dsg1 signals were reduced only beneath the ZO-1-positive junctions of SG2 cells, but remained on the apical side of SG2 cells and the entire surface of SG1 cells (Fig. 4B). Therefore, the ZO-1-positive

M. Yokouchi et al. / Journal of Dermatological Science 77 (2015) 28–36

33

Fig. 3. Development of the dermatitis model mice and morphological assessment of epidermal TJs. (A) Gross appearance, ear swelling, skin severity score of DNFB-treated dermatitis model mice and WT control mice. (B) Distribution of ZO-1, Cldn1, and Dsg1 in control and HD model mice. Bars, 20 mm (upper panels) and 10 mm (lower panels). (C) Expression levels of TJ components and epidermal differentiation markers in epidermal sheets of control and HD model mice.

junctions of SG2 cells function as an occlusive TJ barrier against the inside-out permeation of ETA, even in HD model skin. The anti-Dsg1 scFv injection assay also demonstrated that the inside-out permeation of the scFv, which binds to the extracellular portion of Dsg1 on the cell surface, stopped at the SG2–SG2 junctions in both HD model and WT control mice (Fig. 4C). Other ZO-1-positive junctions between SG2–SG3 cells or SG3–SG3 cells did not show any occlusive function. This size-dependent TJ leakage has not been described previously in Cldn1 / mice [3]. However, when we performed this assay in Cldn1 / mice, biotin-SH, biotin-PEG1500, and biotinPEG5000 permeated through the TJs, while ETA and scFv were stopped at the TJs (biotin-PEG1500, ETA, and scFv, Supplementary Fig. S3; others, data not shown), just as in our dermatitis model mice. The TJ leakiness and claudin-1 expression levels in each model mouse are summarized in Table 1. These findings indicate that the epidermal TJs in the HD model mice, where the expression of several TJ molecules was downregulated, leaked relatively small molecules, but still maintained occlusive functions against large molecules such as proteins, at least under our experimental conditions, in mice.

4. Discussion In this study, we found that the in vivo occlusive function of epidermal TJs was affected in dermatitis model mice with DNFB application, but not in Flg / mice. Of interest, the epidermal TJs of the dermatitis model mice still maintained an occlusive function against large (30-kDa protein) molecules. TJs control molecular movement through paracellular pathways. The tightness or leakiness of the TJ barrier differs in each type of epithelium. This is postulated to be determined by the combination of claudin molecules, the major TJ-specific transmembrane adhesion molecules forming TJ strands [34]. Twentyseven claudins have been identified in mice [35]. Epidermal TJs consist not only of claudin-1 but also claudin-4 and several other claudins [3,36,37]. The apical localization of occludin, ZO-1, and claudin-4 was maintained at SG2–SG2 junctions, even in Cldn1 / mice [3], and we found that the apical SG2–SG2 junctions of Cldn1 / mice still maintained their occlusive functions against the paracellular movement of 30-kDa proteins. These observations suggest that the epidermal TJs that formed without claudin-1 were leaky, but maintained their occlusive function against large protein

34

M. Yokouchi et al. / Journal of Dermatological Science 77 (2015) 28–36

Fig. 4. Size-dependent deficiency of the occlusive function of TJs in dermatitis model mice. The occlusive function of epidermal TJs in WT control and HD model mice was examined by the infiltration of intradermally injected biotin-PEG1500 tracers into the paracellular spaces of the epidermis (A), by the disappearance of Dsg1 digested with intradermally injected ETA (B), and by the distribution of Dsg1 marked with intradermally injected anti-Dsg1 scFv (C). The areas surrounded by dashed lines in the upper panels were enlarged and are shown in the lower panels. Bars, 10 mm.

molecules. Together with the finding that Cldn1 knockdown keratinocyte cellular sheets still maintained the water flux barrier, but became slightly leaky against 40-kDa FITC-dextran [5], these results indicate the importance of evaluating the tightness of TJs against molecules of various size and various properties for evaluating the dysfunction of the epidermal TJ barrier in vivo. Here, we demonstrated that the SC barrier impairment caused by filaggrin deficiency alone induced no detectable change in the morphology or occlusive functions of epidermal TJs in neonatal and adult mice under SPF and normal humidity conditions. This is consistent with the fact that the SC of Flg / mice does not show compact hyperkeratosis [12], which has been observed in TJ impairment in vivo (e.g., Cldn1 / mice [3,38] and Cldn6

overexpression mice [39]). On the other hand, the downregulation of occludin expression has been observed in flaky tail mice [40] and IV patients [20]. Since flaky tail mice have homozygous mutations not only in Flg (ft) but also in Tmem79 (ma) [41,42] and the spontaneous dermatitis observed in flaky tail mice is dependent on the Tmem79 mutation, but not on the Flg mutation [42], the downregulation of occludin in flaky tail mice might be due to secondary effects of the spontaneous dermatitis induced by the Tmem79 mutation. We speculate that the downregulation of occludin expression and leakage of the paracellular barrier reported in IV patients [20] is induced by subtle skin inflammation. Contrasting Flg / mice, the epidermal TJs of the HD model mice showed TJ leakiness. UVB irradiation has been reported to induce

Table 1 Summary of the tightness and leakiness of epidermal TJs in the mouse models.

Real-time PCR analysis Expression level of claudin-1 Tracer permeability assay Sulfo-NHS-LC-Biotin (Biotin-SH) Biotin-dPEGTM (24)-NHS (Biotin-PEG1500) Biotin-PEG-NHS-5000Da (Biotin-PEG5000) ETA scFv 3-30/3h [30]

556 Da 1469 Da 5 kDa 31 kDa 32 kDa

Dermatitis model (adult)

Cldn1-/- (neonate)

No detectable change

Markedly reduced

Null

No No No No No

Leakage Leakage Leakage No leaks No leaks

Leakage Leakage Leakage No leaks No leaks

Wild type (adult/neonate)

Flg

No detectable change No No No No No

leaks leaks leaks leaks leaks

/

(adult/neonate)

leaks leaks leaks leaks leaks

Abbreviations: PEG, polyethylene glycol; scFv, single-chain variable fragment; ETA, exfoliative toxin-A.

M. Yokouchi et al. / Journal of Dermatological Science 77 (2015) 28–36

TJ dysfunction in human skin xenografted on severe combined immunodeficiency mice [43], suggesting that not only chronic allergic dermatitis but also other non-allergic inflammatory reactions induce TJ leakiness. Although ZO-1-positive junctions in the HD model mice were multilayered, only SG2–SG2 ZO-1positive junctions, and not other ZO-1-positive junctions between SG2–SG3 or SG3–SG3 cells, exhibited an occlusive function. Such SG2–SG2-specific occlusive functions among ZO-1-positive multilayered junctions have been demonstrated in human epidermis [4], suggesting the existence of a mechanism to maintain occlusive junctions as single-layered in stratified epithelia. TJ leakiness observed in the HD model mice was size-dependent; the TJs were permeated by the 556-, 1469-, and 5000-Da tracers, but not 30kDa protein molecules. We found that Cldn1 / mice had similar molecular size- and property-dependent TJ leakiness in the epidermis. The mRNA expression of TJ constituents, including claudin-1, was decreased markedly in the HD model mice, suggesting that the downregulation of TJ proteins resulted in the leaky TJ barrier in the HD model mice. Interestingly, several reports have shown that TJ barrier defects induce impairment of the SC barrier (e.g., defects in the SC barrier in Cldn1 / mice [38] and defects in the SC barrier in 3D-cultured keratinocytes treated with Clostridium perfringens enterotoxin to abrade the TJ barrier [44]). Therefore, chronic dermatitis-induced TJ barrier defects are thought to induce subsequent SC barrier defects. As TJs do not show directional discrimination [45], it is speculated that the outside-in penetration of external antigens or allergens through the TJ barrier is facilitated in HD model mice. Since the TJ barrier maintains its occlusive function against large proteins, external antigens of significant size are probably kept outside the TJ barrier, even in the presence of dermatitis. Such external antigens located outside of the TJ barrier could be taken up by the trans-TJ antigen uptake by epidermal LCs [2,15]. Further study is needed to investigate the precise relationship between the molecular size- and property-dependent outside-in penetration of external antigens and the efficiency of percutaneous immunization/sensitization in chronic dermatitis conditions, especially AD. In this study, we found that the epidermal TJ barrier was not affected by filaggrin deficiency directly, but once dermatitis occurred the skin inflammation induced TJ leakiness. Together

Fig. 5. Vicious cycle between the barrier deficiency and skin inflammation in chronic dermatitis. Filaggrin deficiency induces SC barrier impairment. Damage to the SC barrier facilitates the outside-in penetration of external antigens and allergic sensitization. Once inflammation occurs, it induces TJ barrier leakage and subsequent SC barrier impairment, which triggers a vicious cycle in AD pathogenesis.

35

with the reports demonstrating that the SC barrier is impaired by TJ barrier dysfunction [38,44], dermatitis is speculated to impair the SC barrier via TJ barrier leakage and disturbed differentiation of keratinocytes, which facilitates percutaneous sensitization and triggers a vicious cycle between barrier deficiencies and skin inflammation (Fig. 5). Further clarification of the cascades in skin barrier dysfunction will lead to the development of a novel therapeutic strategy to stop the vicious cycle of AD aggravation. Acknowledgments We thank Minae Suzuki, Hiromi Itoh, and Showbu Sato for technical support and animal care. This work was supported by Grants-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science, and Technology of Japan; Health Labour Sciences Research Grants for Research on rare and intractable diseases and for Research on Allergic Diseases and Immunology from the Ministry of Health, Labour, and Welfare of Japan; the Nakatomi Foundation; and the Mochida Memorial Foundation for Medical and Pharmaceutical Research.

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.2014. 11.007. References [1] Elias PM. Stratum corneum defensive functions: an integrated view. J Invest Dermatol 2005;125:183–200. [2] Kubo A, Nagao K, Amagai M. Epidermal barrier dysfunction and cutaneous sensitization in atopic diseases. J Clin Invest 2012;122:440–7. [3] Furuse M, Hata M, Furuse K, Yoshida Y, Haratake A, Sugitani Y, et al. Claudinbased tight junctions are crucial for the mammalian epidermal barrier: a lesson from claudin-1-deficient mice. J Cell Biol 2002;156:1099–111. [4] Yoshida K, Yokouchi M, Nagao K, Ishii K, Amagai M, Kubo A. Functional tight junction barrier localizes in the second layer of the stratum granulosum of human epidermis. J Dermatol Sci 2013;71:89–99. [5] Kirschner N, Rosenthal R, Furuse M, Moll I, Fromm M, Brandner JM. Contribution of tight junction proteins to ion, macromolecule, and water barrier in keratinocytes. J Invest Dermatol 2013;133:1161–9. [6] Bieber T. Atopic dermatitis. N Engl J Med 2008;358:1483–94. [7] Chavanas S, Bodemer C, Rochat A, Hamel-Teillac D, Ali M, Irvine AD, et al. Mutations in SPINK5, encoding a serine protease inhibitor, cause Netherton syndrome. Nat Genet 2000;25:141–2. [8] Walley AJ, Chavanas S, Moffatt MF, Esnouf RM, Ubhi B, Lawrence R, et al. Gene polymorphism in Netherton and common atopic disease. Nat Genet 2001; 29:175–8. [9] Oji V, Eckl K-M, Aufenvenne K, Na¨tebus M, Tarinski T, Ackermann K, et al. Loss of corneodesmosin leads to severe skin barrier defect, pruritus, and atopy: unraveling the peeling skin disease. Am J Hum Genet 2010;87:274–81. [10] Brown SJ, McLean WHI. One remarkable molecule: filaggrin. J Invest Dermatol 2012;132:751–62. [11] Palmer CNA, Irvine AD, Terron-Kwiatkowski A, Zhao Y, Liao H, Lee SP, 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: 441–446. [12] Kawasaki H, Nagao K, Kubo A, Hata T, Shimizu A, Mizuno H, et al. Altered stratum corneum barrier and enhanced percutaneous immune responses in filaggrin-null mice. J Allergy Clin Immunol 2012;129:1538–46. e1536. [13] De Benedetto A, Kubo A, Beck LA. Skin barrier disruption: a requirement for allergen sensitization. J Invest Dermatol 2012;132:949–63. [14] McGrath JA, Uitto J. The filaggrin story: novel insights into skin-barrier function and disease. Trends Mol Med 2008;14:20–7. [15] Kubo A, Nagao K, Yokouchi M, Sasaki H, Amagai M. External antigen uptake by Langerhans cells with reorganization of epidermal tight junction barriers. J Exp Med 2009;206:2937–46. [16] Ouchi T, Kubo A, Yokouchi M, Adachi T, Kobayashi T, Kitashima D, et al. Langerhans cell antigen capture through tight junctions confers preemptive immunity in experimental staphylococcal scalded skin syndrome. J Exp Med 2011;208:2607–13. [17] Kaplan DH. In vivo function of Langerhans cells and dermal dendritic cells. Trends Immunol 2010;31:446–51.

36

M. Yokouchi et al. / Journal of Dermatological Science 77 (2015) 28–36

[18] Merad M, Ginhoux F, Collin M. Origin, homeostasis and function of Langerhans cells and other langerin-expressing dendritic cells. Nat Rev Immunol 2008;8:935–47. [19] De Benedetto A, Rafaels NM, McGirt LY, Ivanov AI, Georas SN, Cheadle C, et al. Tight junction defects in patients with atopic dermatitis. J Allergy Clin Immunol 2011;127:773–86. [20] Gruber R, Elias PM, Crumrine D, Lin T-K, Brandner JM, Hachem J-P, et al. Filaggrin genotype in ichthyosis vulgaris predicts abnormalities in epidermal structure and function. Am J Pathol 2011;178:2252–63. [21] Tomimori Y, Tanaka Y, Goto M, Fukuda Y. Repeated topical challenge with chemical antigen elicits sustained dermatitis in NC/Nga mice in specificpathogen-free condition. J Invest Dermatol 2005;124:119–24. [22] Matsuda H, Watanabe N, Geba GP, Sperl J, Tsudzuki M, Hiroi J, et al. Development of atopic dermatitis-like skin lesion with IgE hyperproduction in NC/Nga mice. Int Immunol 1997;9:461–6. [23] Masuda S, Oda Y, Sasaki H, Ikenouchi J, Higashi T, Akashi M, et al. LSR defines cell corners for tricellular tight junction formation in epithelial cells. J Cell Sci 2011;124:548–55. [24] Itoh M, Yonemura S, Nagafuchi A, Tsukita S. A 220-kD undercoat-constitutive protein: its specific localization at cadherin-based cell-cell adhesion sites. J Cell Biol 1991;115:1449–62. [25] Saitou M, Ando-Akatsuka Y, Itoh M, Furuse M, Inazawa J, Fujimoto K, et al. Mammalian occludin in epithelial cells: its expression and subcellular distribution. Eur J Cell Biol 1997;73:222–31. [26] Kitano Y, Okada N. Separation of the epidermal sheet by dispase. Br J Dermatol 1983;108:555–60. [27] Kubo A, Nagao K, Amagai M. 3D visualization of epidermal Langerhans cells. Methods Mol Biol 2013;961:119–27. [28] Chen Y, Merzdorf C, Paul DL, Goodenough DA. COOH terminus of occludin is required for tight junction barrier function in early Xenopus embryos. J Cell Biol 1997;138:891–9. [29] Hanakawa Y, Schechter NM, Lin C, Garza L, Li H, Yamaguchi T, et al. Molecular mechanisms of blister formation in bullous impetigo and staphylococcal scalded skin syndrome. J Clin Invest 2002;110:53–60. [30] Ishii K, Lin C, Siegel DL, Stanley JR. Isolation of pathogenic monoclonal antidesmoglein 1 human antibodies by phage display of pemphigus foliaceus autoantibodies. J Invest Dermatol 2008;128:939–48. [31] Higashi T, Tokuda S, Kitajiri S-i, Masuda S, Nakamura H, Oda Y, et al. Analysis of the ‘angulin’ proteins LSR, ILDR1 and ILDR2—tricellulin recruitment, epithelial barrier function and implication in deafness pathogenesis. J Cell Sci 2013;126: 966–977.

[32] Ikenouchi J, Furuse M, Furuse K, Sasaki H, Tsukita S, Tsukita S. Tricellulin constitutes a novel barrier at tricellular contacts of epithelial cells. J Cell Biol 2005;171:939–45. [33] Amagai M, Matsuyoshi N, Wang ZH, Andl C, Stanley JR. Toxin in bullous impetigo and staphylococcal scalded-skin syndrome targets desmoglein 1. Nat Med 2000;6:1275–7. [34] Tsukita S, Yamazaki Y, Katsuno T, Tamura A. Tight junction-based epithelial microenvironment and cell proliferation. Oncogene 2008;27:6930–8. [35] Mineta K, Yamamoto Y, Yamazaki Y, Tanaka H, Tada Y, Saito K, et al. Predicted expansion of the claudin multigene family. FEBS Lett 2011;585:606–12. [36] Brandner JM, Kief S, Grund C, Rendl M, Houdek P, Kuhn C, et al. Organization and formation of the tight junction system in human epidermis and cultured keratinocytes. Eur J Cell Biol 2002;81:253–63. [37] Kirschner N, Brandner JM. Barriers and more: functions of tight junction proteins in the skin. Ann NY Acad Sci 2012;1257:158–66. [38] Sugawara T, Iwamoto N, Akashi M, Kojima T, Hisatsune J, Sugai M, et al. Tight junction dysfunction in the stratum granulosum leads to aberrant stratum corneum barrier function in claudin-1-deficient mice. J Dermatol Sci 2013;70:12–8. [39] Turksen K, Troy T-C. Permeability barrier dysfunction in transgenic mice overexpressing claudin 6. Development 2002;129:1775–84. [40] Nakai K, Yoneda K, Hosokawa Y, Moriue T, Presland RB, Fallon PG, et al. Reduced expression of epidermal growth factor receptor, E-cadherin, and occludin in the skin of flaky tail mice is due to filaggrin and loricrin deficiencies. Am J Pathol 2012;181:969–77. [41] Fallon PG, Sasaki T, Sandilands A, Campbell LE, Saunders SP, Mangan NE, et al. A homozygous frameshift mutation in the mouse Flg gene facilitates enhanced percutaneous allergen priming. Nat Genet 2009;41:602–8. [42] Sasaki T, Shiohama A, Kubo A, Kawasaki H, Ishida-Yamamoto A, Yamada T, et al. A homozygous nonsense mutation in the gene for Tmem79, a component for the lamellar granule secretory system, produces spontaneous eczema in an experimental model of atopic dermatitis. J Allergy Clin Immunol 2013;132: 1111–1120. e1114. [43] Yuki T, Hachiya A, Kusaka A, Sriwiriyanont P, Visscher MO, Morita K, et al. Characterization of tight junctions and their disruption by UVB in human epidermis and cultured keratinocytes. J Invest Dermatol 2011;131:744–52. [44] Yuki T, Komiya A, Kusaka A, Kuze T, Sugiyama Y, Inoue S. Impaired tight junctions obstruct stratum corneum formation by altering polar lipid and profilaggrin processing. J Dermatol Sci 2013;69:148–58. [45] Anderson JM. Molecular structure of tight junctions and their role in epithelial transport. News Physiol Sci 2001;16:126–30.

Epidermal tight junction barrier function is altered by skin inflammation, but not by filaggrin-deficient stratum corneum.

The tight junction (TJ) barrier is located in the granular layer of the epidermis. Filaggrin deficiency predisposes patients to atopic dermatitis (AD)...
4MB Sizes 0 Downloads 9 Views