Protective role of Galectin‐7 for skin barrier impairment in atopic dermatitis

Abstract Background Atopic dermatitis (AD) patients have a barrier disorder in association with Th2 dominant skin inflammation. Galectin‐7 (Gal‐7), a soluble unglycosylated lectin, is highly expressed in the stratum corneum of AD patients. However, the biological significance of increased Gal‐7 expression in AD skin lesions remains unclear. Objective We aimed to investigate the production mechanism and functional role of Gal‐7 in AD patients and IL‐4/IL‐13–stimulated epidermal keratinocytes. Methods We assessed the Gal‐7 expression levels in skin lesions and sera from AD patients. Gal‐7 levels were also measured in monolayered normal human epidermal keratinocytes (NHEKs) and 3‐dimensional (3D)–reconstructed epidermis in the presence or absence of IL‐4/IL‐13 with or without Stat3, Stat6 or Gal‐7 gene silencing. Results Gal‐7 was highly expressed in the stratum corneum or intercellular space of AD lesional epidermis as assessed by the stratum corneum proteome analysis and immunohistochemistry. A positive correlation was noted between serum Gal‐7 level and transepidermal water loss in patients with AD. These clinical findings were corroborated by our in vitro data, which showed that IL‐4/IL‐13 facilitated the extracellular release of endogenous Gal‐7 in both monolayered NHEKs and 3D‐reconstructed epidermis. This machinery was caused by IL‐4/IL‐13–induced cell damage and inhibited by knockdown of Stat6 but not Stat3 in NHEKs. Moreover, we performed Gal‐7 knockdown experiment on 3D‐reconstructed epidermis and the result suggested that endogenous Gal‐7 serves as a protector from IL‐4/IL‐13–induced disruption of cell‐to‐cell adhesion and/or cell‐to‐extracellular matrix adhesion. Conclusion and Clinical Relevance Our study unveils the characteristic of Gal‐7 and its possible role as an alarmin that reflects the IL‐4/IL‐13–induced skin barrier impairment in AD.


| INTRODUC TI ON
Atopic dermatitis (AD) is caused by an impaired barrier function of the stratum corneum, which is partly represented by loss-of-function mutation of filaggrin (FLG). 1,2 Despite the presence or absence of FLG gene mutation, patients with AD show an abnormal skin condition based on Th2 skewing. 3,4 Th2 cytokines, interleukin (IL)-4 and IL-13, directly reduce FLG expression in epidermal keratinocytes, resulting in skin barrier disruption. 5 A skin barrier dysfunction leads to increased transepidermal water loss (TEWL) and xerosis, followed by a vicious circle of itch and scratch. 6 Furthermore, the disrupted skin barrier allows allergic or pathogenic antigens to penetrate the epidermis more easily and stimulates epidermal keratinocytes to produce inflammatory cytokines or antimicrobial peptides, including thymic stromal lymphopoietin (TSLP), IL-33 and β-defensin 2. [7][8][9] In particular, endogenous molecules such as IL-33, S100, high mobility group box 1 (HMGB-1) and β-defensin 2 are well known as "alarmins," which can be released from dead cells or via non-classical secretion pathways to induce immune responses under certain inflammatory milieu. 10 Galectin is a soluble unglycosylated lectin which its carbohydrate recognition domains bind with high affinity to β-galactosides. 11,12 Among various types of galectin, prototypic Galectin-7 (Gal-7) is highly expressed in the intercellular space, cytoplasm and nucleus of epidermal keratinocytes. 13 Importantly, Gal-7 is involved in the homeostasis of epidermal keratinocytes, such as proliferation, differentiation, apoptosis, and cell-to-cell or cell-to-extracellular matrix (ECM) adhesion. 14 In addition, Gal-7 serves as a target molecule of skin resident bacteria, Finegoldia magna, since β-galactosides on its cell wall have an affinity to  Intriguingly, Gal-7 is highly expressed in the stratum corneum of AD patients and positively associated with skin barrier defects. 16 We had previously identified and quantified 440 proteins from human stratum corneum by using proteome analysis of tape-stripped skin samples. 17 In a comparison between AD patients and healthy subjects, we have preliminarily shown the increased expression of Gal-7 in the stratum corneum of patients with AD. Meanwhile, a recent study demonstrated that Gal-7 participates in stabilization of morphological structure of epidermal keratinocytes by co-localizing with E-cadherin. 18 However, the regulation and functional significance of Gal-7 in AD skin lesions remain to be elucidated.
In this study, we first demonstrated, by using both proteome analysis and immunohistochemical (IHC) staining, that Gal-7 was deposited in the stratum corneum or intercellular space of AD lesional epidermis. We also revealed the positive correlation between serum Gal-7 levels and TEWL values in patients with AD. Our in vitro data showed that IL-4/IL-13 facilitated the extracellular release of Gal-7 from normal human epidermal keratinocytes (NHEKs) and 3-dimensional (3D)-reconstructed epidermis. It is further suggested that endogenous Gal-7 serves as a protector from IL-4/IL-13-induced disruption of cell-to-cell adhesion and/or cell-to-ECM adhesion. The outcome of this study shed light on a new role of Gal-7 as "alarmin" responding to the IL-4/IL-13-mediated skin barrier damage in AD.

| Hybrid quadrupole-orbitrap liquid chromatography/mass spectrometry (LC/MS/MS)
All the procedure for measurement of Gal-7 in tape-stripped stratum corneum samples from both AD and normal healthy subjects by hybrid quadrupole-orbitrap LC/MS/MS was performed as described earlier. 17

| Cells, cell culture and generation of 3D-reconstructed epidermis
NHEKs were obtained from Kurabo and maintained in EpiLife medium with supplemented with S7 and antibiotic-antimycotic (all from Gibco, Thermo Fisher Scientific). Sub-confluent monolayered NHEKs A 3D-reconstructed epidermis was generated as previously described. 20 In brief, normal human dermal fibroblasts (NHDFs) were obtained from Kurabo and maintained in Dulbecco's minimal essential medium (DMEM) supplemented with 10% foetal calf serum (FCS) and 1% penicillin/streptomycin (PC/SM) (Wako). After polymerization of the solution on the insert at 37°C, NHDFs containing collagen solution was applied onto the insert, followed again by polymerization at 37°C. DMEM supplemented with 10% FCS, 1% PC/SM and ascorbic acid (final concentration 50 ng/mL) was added to the fibroblast-rich gel. After 5-day culture, the NHDFs had contracted the gel. Five days after the dermal component was prepared, 1.0 × 10 5 NHEKs in 50 μL of EpiLife with S7 and antibiotic-antimycotic were seeded onto the concave surface of the contracted gel. The resulting living skin equivalent (LSE) was maintained submerged in culture medium for 2 days. When the keratinocytes reached confluence, the LSE was raised to the air-liquid interface, and cornification medium was added. The medium was changed every other day. On day 14 after airlift, 3D-reconstructed epidermis was stimulated with or without IL-4 + IL-13 (50 ng/mL) for 48 hours.

| ELISA
Patients' sera or culture supernatants were stored at −80°C before analysis of Gal-7 levels by RayBio ® Human Galectin-7 ELISA Kit (RayBiotech) according to the manufacturer's protocols.
Confocal microscopy analysis was carried out using a Leica TCS SP8 (Leica Microsystems). The images were analysed using Leica Application Suite (Leica) software.

Then, cells were washed with cold PBS and stained with Annexin V Phycoerythrin (PE) Apoptosis Detection Kit I (BD Biosciences)
according to the manufacturer's instructions. Dead cells, which are positive for both 7-amino-actinomycin D (7-AAD) and Annexin V-PE, were quantified by flow cytometry using a FACS Canto II (BD Biosciences).

| RNA isolation and real-time quantitative reverse transcriptase-PCR (RT-qPCR)
Total RNA was prepared from cells in culture by using PureLink RNA Mini Kit (Thermo Fisher Scientific) according to manufacturer's protocol. Next, 1 μg of total RNA was reverse-transcribed into cDNA by using TaqMan Reverse Transcription Reagents (Thermo Fisher Scientific). Real-time PCR was performed with a Thermal Cycler Dice Real-Time System II (Takara Bio Inc). The relative change in the level of target gene was calculated using the 2 -deltadeltaCT method.

| siRNA transfection
Synthetic Stat3, Stat6 siRNA oligonucleotides (SignalSilence ® Stat3 siRNA II, SignalSilence ® Stat6 siRNA II) or control siRNA (SignalSilence ® Control siRNA) was purchased from Cell Signaling Technology. 1 × 10 5 cells/mL of NHEKs in 6-cm dishes were seeded in culture medium with S7 and antibiotics, and incubated for 24 hours to allow cells to settle. After changing the medium, NHEKs were transfected with 10 nM siStat3, siStat6 or siControl using HiPerFect Transfection Reagent (Qiagen) and incubated for 24 hours. After adding another preparation of the same siRNA mix in the same way to each well, the cells were incubated for 24 hour and conducted to each experiment.

| Statistical analysis
All statistical analyses were performed by using GraphPad Prism software version 7 (GraphPad Software). Differences between two groups were evaluated by using unpaired t test, paired t test or Mann-Whitney U test (for non-parametric data). When experimental groups were more than three, differences were evaluated using one-way ANOVA with Tukey's multiple comparisons test or with Dunnett's multiple comparison test (comparisons with control group) for parametric data. For repeated data, two-way ANOVA with Tukey's or Sidak's multiple comparison test for parametric data was used. Pearson's correlation analysis or Spearman's rank correlation (for non-parametric data) was calculated to assess the correlation between the data. A P-values < .05 was considered significant.

| AD patients have increased levels of Gal-7 in both skin lesions and sera
We fist compared the levels of Gal-7 expression between the AD and HC skin. By LC/MS/MS proteome analysis, we found that the amount of Gal-7 in patients with AD was significantly higher than that in HC ( Figure 1A). The IHC staining showed diffuse expression of Gal-7 at a moderate level in the cytoplasm and nucleus of epidermal keratinocytes in normal skin ( Figure 1B, lower panel). In contrast, Gal-7 was abundantly found in the intercellular space of the epidermis and also accumulated in the stratum corneum of AD patient ( Figure 1B, upper panel). Quantitative image analysis revealed no significant difference in the percentage of intercellular Gal-7 + keratinocytes between AD and HC groups ( Figure 1C). We next measured serum Gal-7 level in 20 patients with AD and 7 HC, and then examined the correlations with barrier impairment and disease severity as previously reported. 21 In consistent with our LC/MS/MS data on the skin lesions, serum Gal-7 level in patients with AD was significantly higher than that in HC ( Figure 1D). In addition, serum Gal-7 level positively correlated with TEWL value, a specific marker for skin barrier impairment, and tended to correlate with a clinical severity marker, SCORAD index ( Figure 1E,F).
Taken together, these findings suggested that Gal-7 was highly secreted from the epidermis and diffused into the peripheral blood, resulting in the elevation of serum Gal-7 in patients with AD.

| Th2 cytokines, IL-4/IL-13, promote Gal-7 release from NHEKs
Th2 cytokines, IL-4 and IL-13, are thought to play a crucial role in skin barrier dysfunction of patients with AD. 5 However, Th1-, Th17-or Th22-derived cytokines may also be involved in the pathogenesis of AD. 22,23 Therefore, we conducted an in vitro study to address the responsible cytokines that regulate production and secretion of Gal-7 in epidermal keratinocytes. We cultured NHEKs in the presence or ab-  Figure 2D).
We further generated 3D-reconstructed epidermis as previously described 20 and stimulated it with IL-4/IL-13. In consistent with the results from AD skin lesions, the IL-4/IL-13-treated epidermis for 48 hours clearly showed the intercellular deposition of Gal-7 at a concentration of 50 ng/mL compared with the control epidermis ( Figure 2E). These findings indicate that IL-4/IL-13 primarily promotes the release of endogenous Gal-7 from keratinocytes via necrotic cell death.

| Endogenous Gal-7 serves as a protector from IL-4/IL-13-induced disruption of cell-to-cell adhesion and/or cell-to-ECM adhesion
To elucidate the functional role of IL-4/IL-13-induced Gal-7 release from keratinocytes, we prepared Gal-7 gene silenced 3D-reconstructed epidermis. Our RT-qPCR showed that the Gal-7 We next hypothesized that extracellular Gal-7 contributes to stabilization of IL-4/IL-13-induced, E-cadherin-mediated cell-tocell and/or cell-to-ECM adhesion. However, we could not detect significant reduction in E-cadherin expression at the basal and suprabasal layer of the shGal-7-transduced epidermis compared to shCtr epidermis with or without IL-4/IL-13 treatment ( Figure S1A).
Furthermore, recombinant Gal-7 supplement had no effect on the expression levels of either Gal-7 or E-cadherin in the shGal-7-transduced epidermis ( Figure S1B). Thus, our results do not support for the notion that exogenous Gal-7 stabilizes the IL-4/Il-13-induced, E-cadherin-dependent disrupted keratinocyte adhesion.

| D ISCUSS I ON
In this study, we showed that Gal-7 was released from epidermal keratinocytes and highly accumulated at the intercellular space and/ or the stratum corneum in AD skin lesions. Consistently, a previous study showed that the amount of Gal-7 in tape-stripped stratum corneum from patients with AD was significantly higher than that from normal subjects and positively correlated with the disease severity or TEWL. 16 We further clearly demonstrated that serum Gal-7 level was elevated in AD patients and reflected their skin barrier impairment. As for the other clinical severity markers, we also found that the serum Gal-7 level positively correlated with the serum levels of activation-regulated chemokine (TARC) and lactate dehydrogenase (LDH), circulating eosinophil percentage and visual analogue scale (VAS) of pruritus ( Figure S2). Thus, we indicate that the serum Gal-7 level is a potential biomarker for AD.
NHEKs constantly express Gal-7 in the cytoplasm and nucleus for maintaining the homeostasis and release it extracellularly to evoke biological responses to glycoproteins expressing other cell surface. 13 Our in vitro study provides supportive evidence that Th2 cytokine IL-4/ IL-13 may promote this machinery via Stat6 activation. Since we found no evidence that IL-4/IL-13 serves as a direct inducer of Gal-7 transcription, we proposed that IL-4/IL-13-induced Stat6-mediated cell damage might have a crucial role in the endogenous Gal-7 release from NHEKs.
As mentioned in introduction, alarmins are categorized as endogenous molecules, which are released from dead cells or via non-classical secretion pathways, and activate the immune responses. 10 From this point of view, it has been reported that Gal-7 exerts an effect on T cells to proliferate, polarize towards Th1 cells and even down-modulate their cytokine production. 26,27 Taken together, Gal-7 could be considered as an alarmin that responds to the IL-4/IL-13-mediated tissue damage, as seen in Gal-3 and Gal-9. 28 E-cadherin, DSG1, DSC1 or CLDN1. [30][31][32][33][34] In particular, IL-4/IL-13 decreases the expression of membrane E-cadherin, and simultaneously increases the intercellular accumulation of hyaluronan (HA), which is a key pathogenic event of spongiosis. 34 A recent study clearly demonstrated that Gal-7 stabilizes E-cadherin at the plasma membrane by restraining its endocytosis, and Gal-7 silence decreases E-cadherin-mediated intercellular adhesion. 18 In addition, Gal-7 is also thought to interact with ECM via β1-integrin or matrix metalloproteinase-9 (MMP-9) in other cell types. 14