2021 年 44 巻 1 号 p. 140-143
To establish a system for assessing drug permeation and irritation of the skin, the permeation of benzoic acid and isosorbide dinitrate, which are listed in the Pharmacopoeia, and the chemical irritation were evaluated using skin generated from human induced pluripotent stem cells (iPSCs). Multilayer structures and cellular markers (keratin 14 and 10, which are in basal and suprabasal epidermal layers) were clearly detected in our iPSC-based skin. Transepidermal water loss (TEWL) decreased after iPSC-derived keratinocytes were cultured on collagen gels from human primary fibroblasts. These results indicate that the barrier function was partly increased by formation of the living epidermis. The cumulative amount of benzoic acid and isosorbide dinitrate across human iPSC-based skin gradually increased after an initial lag time. Moreover, the irritancy of various chemicals (non-irritants: ultrapure water, allyl phenoxy-acetate, isopropanol, and hexyl salicylate and irritants: 5% sodium dodecyl sulfate (SDS), heptanal, potassium hydroxide (5% aq.) and cyclamen aldehyde) to iPSC-based skin was almost met the irritation criteria of the Organisation for Economic Co-operation and Development (OECD) guideline. The results of our iPSC-based skin evaluation provide useful basic information for developing an assessment system to predict the permeation and safety of new transdermal drugs in human skin.
In developing new transdermal drug formulations, the skin permeation and safety of drugs must be evaluated. To predict the permeation and safety of transdermal drugs in human skin, new transdermal assessment systems are highly required.1,2) Of the various strategies available, human induced pluripotent stem cell (iPSC)-based skin seems to be a good approach for establishing new transdermal assessment systems.3) Presently, several research studies have reported three-dimensional skin produced using keratinocytes differentiated from human iPSCs for transplantation in the human body for wound healing and treatment of skin diseases such as epidermolysis bullosa and myelomeningocele caused by skin defect.4–7) However, iPSC-based skin has rarely been used for the evaluation of skin permeation and irritation by drugs. Recently, we attempted, for the first time, to use iPSC-based skin to evaluate drug permeation.3) We discovered that the relative permeation of model drugs with different molecular weights across the skin could be comparatively evaluated using iPSC-based skin in our previous study. However, the permeation of drugs listed in the Pharmacopoeia across iPSC-based skin and the chemical irritancy have not been examined.
Therefore, purpose of this study was to investigate the potential usefulness of iPSC-based skin for evaluating permeation of drugs listed in the Pharmacopoeia and the chemical irritancy based on Organisation for Economic Co-operation and Development (OECD) guidelines.8,9) To this end, we prepared human iPSC-based skin, and evaluated the transdermal permeation of benzoic acid (BA) and isosorbide dinitrate (ISDN), which are listed in the Japanese Pharmacopoeia, 17th Edition. Furthermore, we evaluated the skin irritation induced by various chemicals, which are listed in the OECD using the generated human iPSC-based skin.
The human iPSC line 201B7 was provided by Dr. S. Yamanaka, Kyoto University,10) and maintained on vitronectin with Essential 8 Flex medium (both from Thermo Fisher Scientific, Waltham, MA, U.S.A.). HFF-1 cells (fibroblast line derived from neonatal human foreskin) were purchased from American Type Culture Collection (ATCC, Manassas, VA, U.S.A.). HFF-1 cells were grown in Dulbecco’s modified Eagle’s medium (DMEM) with high-glucose (Nacalai Tesque, Inc., Kyoto, Japan) supplemented with 15% fetal bovine serum (FBS; Biosera, Marikina, Philippines) at 37 °C in humidified air containing 5% CO2.
iPSCs were differentiated into keratinocytes according to a previously described protocol with slight modifications.7,11)
Collagen gels were used as a dermis model and were generated using HFF-1 cells according to a previously described protocol12) with slight modifications.
After collagen gel culturing for 1 week, human iPSC-derived keratinocytes were added to collagen gels in cell culture inserts to generate the three-dimensional skin, as previously reported.7)
We performed histological and immunohistochemical examination of human iPSC-derived keratinocytes, fibroblasts, and three-dimensional skin according to methods previously reports7,11) with modifications.
The iPSC-based skin samples on 24-well inserts were placed in 24-well plates. The lower compartments of the inserts were filled with 250 µL phosphate buffered saline (PBS). Transepidermal water loss (TEWL) was measured using a VAPO SCAN (ASCH Japan Co., Ltd., Tokyo, Japan).
For the skin permeation assays, 0.5 mL saturated isosorbide dinitrate (ISDN, molecular weight (Mw); 236.14, log Pow; 1.225, The United States Pharmacopeia, North Bethesda, MD, U.S.A.) or benzoic acid (BA, Mw; 122.12, log Pow; 1.410, Nacalai Tesque, Inc.) in PBS and 5 mL 40% polyethylene glycol (PEG) 400 (solubilizer) in PBS was added into the donor and receptor compartments, respectively.13) The cells were maintained at 32 °C and the plates with cell culture inserts were continuously agitated at 100 rpm. Receptor solution samples were periodically collected and replaced with an equal volume of fresh 40% PEG400 in PBS and the permeability of these drugs in dehaired Wistar rat skin was evaluated as controls as reported previously.14)
The skin irritation induced by various chemicals was evaluated using human iPSC-based skin as reported previously.9,15) Briefly, the eight reference test chemicals (non-irritants: ultrapure water, allyl phenoxy-acetate, isopropanol and hexyl salicylate and irritants: 5% sodium dodecyl sulfate (SDS), heptanal, potassium hydroxide (5% aq.) and cyclamen aldehyde) were selected from the list in the OECD test guideline 439 (TG439). The chemicals were added to the epidermal site of the human iPSC-based skin on the culture insert set in the 24-well plate.
After a 5-min incubation at 15–20 °C, the washed tissues were incubated in the three-dimensional culture medium for 42 h at 37 °C and then the cell viability was evaluated using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. A chemical was determined to be an irritant when the mean cell viability of the tested tissues was reduced to <50% of that of the untreated sample and a nonirritant when the cell viability remained >50%.
All animal experiments were conducted in accordance with the principles and procedures outlined in the National Institutes of Health (NIH) Guidelines for the Care and Use of Laboratory Animals. The animal experimental protocols were approved by the Animal Experimentation Committee at the Kyoto Pharmaceutical University.
All the experiments in the study were performed at least in triplicate, and the data are expressed as the mean ± standard error (S.E.) or standard deviation (S.D.). The statistical comparisons were evaluated using an ANOVA with subsequent Dunnett’s multiple comparison tests.
Detailed Materials and Methods are summarized in Supplementary Materials.
Figure 1 shows the characteristics of the generated human iPSC-based skin. The tissues in the inserts were yellowish with a dry surface (Fig. 1A). The histological images of the iPSC-based skin showed that a typical multilayered structure was clearly formed, although iPSC-based skin did not show sufficient formation of stratum corneum. Keratin 14 and 10 were also detected in the basal and suprabasal epidermal layers of the human iPSC-based skin, respectively (Figs. 1B, C). To evaluate the epidermal barrier function of the human iPSC-based skin, we measured the TEWL values of the collagen gel generated from primary fibroblasts and our iPSC-based skin, which were 35.5 ± 1.4 and 20.3 ± 0.5 g/(m2·h), respectively.
Representative photomicrograph and histological images of (A) iPSC-based skin, (B) iPSC-based skin stained with hematoxylin and eosin, (C) iPSC-based skin after immunohistochemical analysis of keratin 14 (green) and keratin 10 (red). Scale bars, 50 µm. (Color figure can be accessed in the online version.)
Figure 2 shows the drug permeability of the generated iPSC-based and rat skin samples. The permeation of BA was higher than that of ISDN in rat skin, while there was no significant difference in the permeation after an initial lag time in generated iPSC-based skin between BA and ISDN.
Permeation rates of these drugs across human iPSC-based skin were similar to that across rat skin. Results are means ± standard errors (S.Es.) of three to five experiments. (◆) BA, (■) ISDN.
To evaluate utility of our iPSC-based skin for the irritation study, the irritancy of various chemicals listed in OECD TG439 was evaluated in the generated human iPSC-based skin samples (Fig. 3). To determine the optimal incubation time for chemicals in human iPSC-based skin, it was incubated with allyl phenoxy-acetate, a representative non-irritant chemical. Although the cell viability was significantly decreased after a 10-min incubation with allyl phenoxy-acetate (false positive), it was maintained after 5-min incubation (negative). Consequently, 5 min was selected as the incubation time for the human iPSC-based skin irritant test and Fig. 3 shows the results of the four irritants and four non-irritants listed in the OECD TG439 after 5-min incubation in human iPSC-based skin. Cell viabilities of ultrapure water, allyl phenoxy-acetate, isopropanol, hexyl salicylate, and heptanal were >50% (negative), while cell viabilities of 5% SDS, potassium hydroxide (5% aq.), and cyclamen aldehyde were <50% (heptanal; false negative, other chemicals: positive) after 5-min incubation. Five percent SDS, potassium hydroxide (5% aq.), and cyclamen aldehyde significantly decreased the cell viabilities after 5-min incubation. The irritancy of these chemicals except for heptanal was in good agreement with the OECD guideline criteria.
Cell viability of iPSC-based skin after 5-min incubation with various chemicals (non-irritants: ultrapure water, allyl phenoxy-acetate, isopropanol and hexyl salicylate; irritants: 5% SDS, heptanal, potassium hydroxide (5% aq.) and cyclamen aldehyde). Results are means ± standard deviations (S.Ds.) of at least three experiments. **, a statistically significant difference compared with untreated group (p < 0.01).
In the present study, we evaluated characteristics of human iPSC-based skin in terms of the permeability of drugs listed in the Pharmacopoeia and the chemical irritancy based on OECD guidelines. In a previous study, iPSC-based skin was generated from iPSC-derived keratinocytes and fibroblasts.3) However, collagen gel generated from human iPSC-derived fibroblasts sometimes has difficulty generating uniform three-dimensional skin with iPSC-derived keratinocytes because of differences in their shrinkage. Therefore, in the present study, three-dimensional skin was established more consistently using HFF-1 cells (human primary fibroblasts) to generate the collagen gel instead of iPSC-derived fibroblasts.7) The epidermis generated from iPSC-derived keratinocytes, but not collagen gel, plays a key role in the evaluation of drug permeability and irritation.16,17) Therefore, we think that replacing iPSC-derived fibroblasts with HFF-1 cells did not unduly affect the functions of the generated three-dimensional skin. The iPSC-based skin generated from iPSC-derived keratinocytes and primary fibroblasts was uniformly covered with the membrane of the cell culture inserts (Fig. 1A). Furthermore, the multilayered structure of the human iPSC-based skin in the present study was in good agreement with that obtained in a previous study.3)
TEWL is widely used to evaluate the skin barrier function and Yokota et al.18) reported its usefulness for supporting the quality of reconstructed epidermal models and three-dimensional skin. The decrease in TEWL after iPSC-derived keratinocytes were cultured on collagen gels from human primary fibroblasts indicated that the barrier function was partly increased by the formation of a living epidermis. Our preliminary data showed that the TEWL of EpiDerm, a commercial human three-dimensional skin product, was approximately 18 g/(m2·h), which was similar to that observed in our iPSC-based skin. These results indicate that the iPSC-based skin had partial barrier function for the evaluation of drug permeability, although the barrier function of iPSC-based skin was not completed due to the insufficient formation of stratum corneum. These results together with those of the structural analysis of iPSC-based skin, indicate that the combination of iPSC-derived keratinocytes and collagen gel generated from HFF-1 cells is a good approach for the generation of iPSC-based skin.
The physicochemical properties of BA and ISDN are well known and they are widely used as markers for in vitro skin permeation studies using animal or three-dimensional skin instead of human skin.13,19,20) Kano et al. reported that the apparent permeability coefficient (Papp) of BA across rat and human skin was higher than that of ISDN.13) Furthermore, previous studies reported that the skin permeation of drugs depends on their lipophilicities in the log Pow range between 1 and 3.21) Therefore, further studies on the permeation of various drugs are needed to elucidate the relationships between permeability and lipophilicity of drug in iPSC-based skin. However, the drug permeations with an initial lag time in generated iPSC-based skin indicate that our iPSC-based skin has the potential to be a model for evaluating the permeation of drugs.
The OECD has suggested the use of in vitro skin irritation test methods using reconstructed human epidermis (OECD TG439). Li et al. reported that the irritant chemicals listed in the OECD guidelines could be appropriately evaluated using human three-dimensional skin.22) The irritancy of chemicals except for heptanal was in good agreement with the OECD guideline criteria.9) Although further studies are required to validate the skin irritation using iPSC-based skin, these results, together with the results of structural analysis of iPSC-based skin (insufficient formation of stratum corneum), indicate that our iPSC-based skin has the potential for be use in quick evaluating skin irritation induced by chemicals.
In conclusion, we showed characteristics of our generated iPSC-based skin in terms of evaluating drug permeability and irritancy. These results of our iPSC-based skin analysis provide useful basic information for developing an assessment system to predict the permeation and safety of new transdermal drugs in human skin using iPSC-based skin.
This work was supported in part by the Promotion and Mutual Aid Corporation for Private Schools of Japan. The authors are grateful to Drs. Yukiko Takahashi, Kazuhiro Kajiwara, and Prof. Akihiro Umezawa of the National Research Institute for Child Health and Development, for their helpful comments on three-dimensional skin modeling. The authors are also grateful to Drs. Akiko Kanda, Mariko Akita, Naoko Kawaguchi, Koichi Nakaoji, and Kazuhiko Hamada of Pias Corporation for supporting the TEWL measurements.
The authors declare no conflict of interest.
The online version of this article contains supplementary materials.
