2020 Volume 43 Issue 10 Pages 1591-1594
Japanese cedar (Cryptomeria japonica) pollen allergen Cry j1 increases the intracellular calcium concentration in human keratinocytes, and also impairs the epidermal barrier function. Here, we show that reduced glutathione (GSH) blocks both thrombin activation and the Cry j1-induced intracellular calcium elevation in cultured human keratinocytes, and also prevents the Cry j1-induced decrease of barrier function in ex vivo human skin.
The immunoglobulin E (IgE)-mediated type 1 hypersensitivity reaction to pollen allergens is well known to cause seasonal allergic rhinitis.1) In particular, over a third of Japanese people are allergic to pollen from the Japanese cedar (Cryptomeria japonica; sugi in Japanese), which is known as sugi-pollinosis.2) We have shown that the Japanese cedar pollen allergen Cry j1 induces an increased level of intracellular calcium in human keratinocytes, as well as impairing the epidermal permeability barrier function in ex vivo human skin.3) When the barrier function is impaired, Cry j1 might directly reach Langerhans cells in the epidermis, exacerbating the allergic reaction.
Mite or cockroach allergens exhibit serine-type protease activity and activate protease-activated receptor 2 (PAR-2), thereby increasing the intracellular calcium level in epidermal keratinocytes and damaging epidermal permeability barrier homeostasis.4) We previously showed that prothrombin is activated by Cry j1 and this activation results in a PAR-1-mediated increase of the intracellular calcium level in epidermal keratinocytes, leading to barrier dysfunction.5) Thrombin is a serine protease catalyzing the conversion of soluble fibrinogen to insoluble fibrin.6) Since glutathione prevents fibrinogen coagulation,7) we speculated that it might block Cry j1-induced thrombin activation and intracellular calcium elevation in keratinocytes. In the present study, we tested this hypothesis by evaluating the effects of reduced glutathione (GSH) on Cry j1-induced thrombin activation and on Cry j1-induced elevation of intracellular calcium in human keratinocytes. We also examined whether GSH can block the decrease of barrier function induced by Cry j1 in human skin ex vivo.
Normal human epithelial keratinocytes from Kurabo (Osaka, Japan) were seeded onto collagen-coated glass coverslips (Matsunami, Osaka, Japan), cultured in EPILIFE-KG2 (Kurabo), and used within 4 d. They were grown to 100% confluency in low-Ca2+ medium (0.06 mM) for 24–48 h, and then incubated in high-Ca2+ medium (1.8 mM) for 24–48 h.
Prothrombin Activation and Thrombin and Activated Coagulation Factor X (FXa) Activity AssaysFor thrombin activity assay, we used a Rox prothrombin kit (Rossix, Mölndal, Sweden). To examine the time course of prothrombin activation with or without GSH and oxidized glutathione (GSSG) (Nacalai Tesque, Inc., Kyoto, Japan) (final concentration 50–500 µg/mL), we mixed prothrombin (Invitrogen, Carlsbad, CA, U.S.A.) (40 ng) and the activating reagent with/without GSH, and then proceeded according to the instruction manual. To measure the direct effect of GSH on thrombin, GSH was added after prothrombin activation by the activating reagent. For FXa assay, we used a factor Xa activity assay kit (Abcam, Cambridge, U.K.). The time course of FXa activity with or without GSH (final concentration 50–500 µg/mL) was measured according to the instruction manual.
The absorption (405 nm) for thrombin activity assay and the fluorescence (Ex/Em = 350/450 nm) for FXa activity assay were measured with an ARVO X3 (PerkinElmer Japan Co., Ltd., Yokohama, Japan) every 40 s.
Determination of Intracellular Calcium by Ratiometric Fluorescence MeasurementIntracellular calcium concentration in individual cells was measured with Fura-2 AM (Molecular Probes Inc., OR, U.S.A.), according to the manufacturer’s instructions. Briefly, cells were loaded with 5 µM Fura-2 AM at 37 °C for 45 min, rinsed with balanced salt solution containing (in mM): NaCl 150, KCl 5, CaCl2 1.8, MgCl2 1.2, N-(2-hydroxyethyl)piperazine-N′-2-ethanesulfonic acid (HEPES) 25, and D-glucose 10, pH 7.4 [abbreviated as BSS(+)], and then incubated for 10 min more at room temperature for de-esterification of the loaded dye.
Cells were imaged with an inverted epifluorescence microscope (ECLIPSE Ti, Nikon, Tokyo, Japan) equipped with a 75 W xenon lamp and band-pass filters (340, 380 nm), using a high-sensitivity CCD camera (ORCA-R2, Hamamatsu Photonics, Hamamatsu, Japan) controlled by a Ca2+ analyzing system (AQUACOSMOS/RATIO, Hamamatsu Photonics). Intracellular calcium concentration was measured repeatedly at intervals of one second.
Culture of Human Skin TissueExcised abdominal skin, which had been obtained from four healthy, Caucasian females aged 34, 35, 35, and 36 years old with informed consent following plastic surgery, was purchased from Biopredic International (Rennes, France) via KAC Co., Ltd. (Kyoto, Japan). It was dermatomed to a thickness of 340–440 µm (epidermis and dermis), and then discs (diameter 10 mm, thickness about 2 mm) were punched out and delivered to our laboratory. The Shiseido ethics committee approved this study, in accordance with National Institute of Health guidelines. The tissue discs were cultured in long-term skin culture medium (LTSC medium; Biopredic International).
Transepidermal Water LossWe measured gravimetric transepidermal water loss (TEWL) according to Hanley et al.8) Skin sections were placed dermis-side-down on glass-bottomed dishes. The lateral edges and dermal surface were sealed with petrolatum to ensure that water loss could occur only through the epidermal surface. The stratum corneum was stripped 20 times with adhesive tape, and then Cry j1 (Hayashibara Co., Ltd., Okayama, Japan) was applied with/without other reagents. The skin sections were kept at ambient temperature (37 °C) and humidity (30–35%), and weighed at intervals of 1 h. The TEWL levels were calculated as milligrams of water lost per square millimeter per hour. Skin sections from all four subjects were used.
Electron-Microscopic ObservationFor electron microscopy, skin samples were minced (pieces <0.5 mm3), fixed in modified Karnovsky’s fixative overnight, and then post-fixed with 2% aqueous osmium tetroxide or 0.2% ruthenium tetroxide as reported.9) These samples were dehydrated in graded ethanol solutions, and then embedded in Epon-epoxy. The area of secreted lipid domains between the stratum corneum (SC) and stratum granulosum (SG) was quantified using samples that had been post-fixed with osmium tetroxide. Measurements were made without knowledge of the experimental treatment, and parameters were extracted with NIH Image from photographs of randomly selected sections at constant magnification.
StatisticsWe determined the statistical significance of differences among three or more groups was by means of ANOVA with Scheffé’s method; p < 0.05 was considered significant.
GSH significantly and dose-dependently decreased the level of thrombin activity generated by activation of prothrombin when added to a mixture of prothrombin and the assay kit’s activating reagent, which contains activated coagulation factor V (FVa) and activated coagulation factor X (FXa) (Fig. 1). Oxidative glutathione (GSSG) (500 µg/mL) also decreased the level of thrombin activity with similar efficacy to GSH (approximately 99% inhibition). Application of GSH to activated thrombin or FXa did not alter either activity (data not shown).
GSH inhibited thrombin activity dose-dependently (n = 3). Bars and lines represent mean ± standard deviation (S.D.).
We previously demonstrated that many factors which induce elevation of intracellular calcium in keratinocytes delay stratum corneum barrier recovery after barrier disruption, and several factors which block the elevation of intracellular calcium accelerate the barrier recovery10) Therefore, calcium imaging in keratinocytes might be a good method to evaluate the effects of compounds on skin barrier homeostasis. Thus, we performed calcium imaging in keratinocytes to examine the effect of GSH on Cry j1-induced calcium elevation. The results of intracellular calcium imaging are shown in Fig. 2. Representative images, shown in Figs. 2A–D, indicate that application of GSH reduced the number of cells showing clearly elevated intracellular calcium in the presence of Cry j1. Representative profiles of intracellular calcium levels after application of Cry j1 with/without GSH (Fig. 2E), together with the results of quantification (Fig. 2F), confirm that co-application of GSH significantly reduced the Cry j1-induced elevation of intracellular calcium.
(A) Representative image of intracellular calcium before the application of 100 ng/mL Cry j1 (expressed as the ratio of fluorescence intensities at 340 and 380 nm). (B) Two minutes after application of 100 ng/mL Cry j1. Intracellular calcium is elevated in many cells. (C) Representative image of intracellular calcium before the application of both 100 ng/mL Cry j1 and 5 µg/mL GSH. (D) Two minutes after the application, only a few cells show elevation of intracellular calcium. Bar = 20 µm. (E) Representative profiles of intracellular calcium after Cry j1 treatment with or without GSH. (F) Percentage of activated cells after treatment with Cry j1, with or without GSH (n = 20 cells). Similar results were obtained in three independent experiments. The Cry j1-induced elevation of intracellular calcium was significantly reduced by pre-application of GSH. Bars and lines represent mean ± S.D.
Figure 3A shows changes of TEWL following application of Cry j1 alone or with GSH. Application of Cry j1 dramatically impaired the barrier function, compared with the control (tape stripping + water). However, application of GSH with Cry j1 almost completely blocked the decrease of barrier function induced by Cry j1. Electron-microscopic observations were consistent with this finding. In Cry j1-treated, tape-stripped skin, the area of secreted lipid domains between SG and SC were significantly decreased. In contrast, large area of secreted lipid domains were observed between SG and SC when GSH was applied together with Cry j1 (Figs. 3B–E).
(A) The vertical axis shows the amount of water loss during 0–1, 1–2 or 2–3 h after tape stripping (n = 4). Application of 1 µg/mL Cry j1 markedly increased water loss, but the Cry j1-induced increase was significantly reduced by co-application of GSH (50 µg/mL). (B) Tape-stripping (TS) 20 times followed by application of water. (C) TS followed by application of 1 µg/mL Cry j1. (D) TS followed by application of 1 µg/mL Cry j1 and 50 µg/mL GSH. Asterisks show secreted lipids located between the stratum corneum (SC) and stratum granulosum (SG). Bars = 1 µm. (E) Quantified results of evaluation of the area of secreted lipid domains (n = 4). Application of Cry j1 significantly decreased the area of secreted lipid domains between SC and SG. Co-application of GSH significantly blocked the effect of Cy j1.
GSH inhibited prothrombin activation, but did not inhibit thrombin after activation. This is consistent with a previous finding that GSH has no effect on thrombin activity.7) GSH also had no effect on FXa activity. To get insight into the mechanism of GSH’s inhibition of prothrombin activation, we performed docking simulation of GSH to prothrombin, thrombin, FV and FX. The results indicated that GSH does not bind to prothrombin or thrombin, but might bind to the C2 domain in FV and the Gla domain in FX (Supplementary Fig. 1). These domains are essential for membrane binding of coagulation proteins, which is an essential step in prothrombinase complex formation.11–13) Since the complex is essential for prothrombin activation, GSH may disturb prothrombinase complex formation by binding to the C2 domain in FV and Gla domain in FX.
GSH is an important cellular antioxidant that protects cells from reactive oxygen species, such as oxygen free radicals and hydrogen peroxide.14) Furthermore, oral application of GSH or GSSG reduces the melanin index and wrinkling, and increases skin elasticity and the water content of the stratum corneum.15) Topical application of GSSG also reduces the melanin index and wrinkle formation, and increases the water content of the stratum corneum.16) Although these observations are interesting from the standpoint of clinical dermatology, the mechanisms of these effects have not been clarified. Interestingly, we found that GSSG inhibited prothrombin activation in vitro as potently as GSH. Although we did not test the effect of GSSG in keratinocytes or ex vivo human skin in the present study, this result, together with the reported findings, suggests that GSSG might be as effective as GSH to block the Cry j1-induced decrease of barrier function in human skin ex vivo.
Our findings here show that GSH inhibits the activation of prothrombin and reduces the Cry j1-induced intracellular calcium elevation in cultured human keratinocytes, as well as blocking the Cry j1-induced decrease of barrier function in human skin ex vivo. These effects are similar to those that we previously observed with tranexamic acid.5) In addition, topical application of tranexamic acid accelerated barrier recovery,17) and tranexamic acid also inhibited melanogenesis in cultured melanoma cells.18) It will be interesting to see if GHS also inhibits melanogenesis or skin aging. Although further work is needed to establish the mechanisms of GSH’s actions, our findings suggest that GSH might be a promising candidate to treat the epidermal barrier pathology induced by Japanese cedar pollen.
This work was supported by JST CREST (Grant No. JPMJCR15D2).
Shinobu Nakanishi, Kanna Kurihara and Mitsuhiro Denda are employees of Shiseido. The authors declare no conflict of interest.
The online version of this article contains supplementary materials.
