Estradiol was purchased from Sigma-Aldrich (St. Louis, MO, USA), fetal bovine serum (FBS) was acquired from Thermo Fisher Scientific (Waltham, MA, USA), and rabbit and mouse serum solutions were procured from Nichirei Biosciences (Tokyo, Japan). Rabbit anti-ERα monoclonal antibody labeled with phycoerythrin (PE), mouse anti-ERβ monoclonal antibody labeled with PE, rabbit anti-loricrin polyclonal antibody, rabbit anti-involucrin polyclonal antibody, and mouse anti-cytokeratin 10 monoclonal antibodies were purchased from Abcam (Cambridge, UK). Thermo Fisher Scientific also provided Alexa Fluor 647-labeled donkey anti-mouse antibody and Alexa Fluor 488-labeled donkey anti-rabbit antibody. Stachyose and catalpol were standard for crude drug determination and were obtained from FUJIFILM Wako Pure Chemical (Osaka, Japan) and used as internal standards. All other chemicals were purchased from FUJIFILM Wako Pure Chemical.

Rehmannia root extract (RE, Lot.06282) was purchased from KOEI KOGYO (Tokyo, Japan). The roots of Rehmannia glutinosa Libosch. var. purpurea Makino were dried, cut into small pieces and extracted with a 1:1 solution of 1,3-butylene glycol and water at 50˚C for 5 h. The extract was stored at 4˚C for 3 d, filtered using a 5 μm filter.

Adult normal human dermal fibroblasts (NHDF; Kurabo, Osaka, Japan) were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% FBS, 100 U/mL penicillin, and 100 μg/mL streptomycin. In addition, NHDF displaying a high population doubling levels (PDL) were obtained by subculturing 15 times. Normal human epidermal keratinocytes immortalized by the pSV40 gene (immortalized NHEK; JCRB, Osaka, Japan) were cultured in an MCDB153 medium with growth supplements. Cells were seeded in culture plates and grown to subconfluence in a 5% CO₂ incubator at 37˚C. Reconstructed human epidermal models (LabCyte EPI-MODEL; J-TEC, Aichi, Japan) were pre-cultured in a 5% CO₂ incubator at 37˚C using the attached assay medium.

NHDF was seeded in LabTek Chamber slides (Thermo Fisher Scientific), cultured, and prior to UVA exposure, the medium was replaced with Hanks’ balanced salt solution and the cell was exposed to 0.26 J/cm² UVA using a sunlight irradiation system (SERIC, Tokyo, Japan). Sham-UVA irradiated NHDF was shielded from the lamp during UVA exposure.

Immortalized NHEK and NHDF were seeded in LabTek chamber slides and cultured, after which the medium was replaced with unaltered medium or that containing estrogen or RE, and incubation continued.

Following this, the cells were fixed in formalin and blocked with serum solution. They were then stained with PE-labeled anti-ERα antibody or PE-labeled anti-ERβ antibody. Cell nuclei were stained with 4’,6-diamidino-2-phenylindole dihydrochloride (DAPI). Cell nuclei and ERs fluoresced blue and yellow, respectively, when examined using an IX70 microscope coupled with a DP80 digital camera (Olympus, Tokyo, Japan). Images were acquired using WinROOF2015 image analysis software (Mitani, Fukui, Japan), and the average luminance value was calculated by dividing the luminance value by the cell number. We performed at least three independent studies, yielding similar results, which are shown in the figures.

Stachyose and catalpol were quantified by HPLC analysis using a HITACHI Chromaster (Hitachi High-Tech Science, Tokyo, Japan) coupled to a Chromaster 5430 diode array detector (Hitachi High-Tech Science) and a Chromaster 5450 refractive index detector (Hitachi High-Tech Science).

Analysis was carried out using Shodex Asahipak NH2P-50 4E columns (5 μm, 4.6 × 250 mm, Showa Denko, Tokyo, Japan). For quantification of stachyose and catalpol, a mobile phase consisting of 70% (v/v) acetonitrile in water was used at a flow rate of 1 mL/min and column temperature of 50˚C. UV detection was performed at a wavelength of 203 nm. The concentrations of stachyose and catalpol were calculated from standard curves of standard stachyose and catalpol solutions, respectively.

To investigate cell proliferation, immortalized NHEK and NHDF were seeded and cultured. A cell proliferation assay was performed using the Cell Counting Kit-8 (Dojindo, Kumamoto, Japan) according to the manufacturer’s instructions.

NHDF was seeded and cultured, and the supernatants were collected after sample treatment. Type I collagen was measured by direct ELISA using an anti-collagen type I rabbit antibody. The hyaluronic acid, bFGF, TGF-β1, and EGF contents were measured using the Quantikine ELISA kit of hyaluronic acid, human bFGF, human TGF-β1, or EGF (R&D Systems, Minneapolis, MN, USA), respectively, according to the manufacturer’s instructions.

The skin model was fixed in formalin and embedded in paraffin wax, and the tissues were cut, deparaffinized, and rehydrated. Hematoxylin and eosin (H&E) staining and immunohistochemical staining were performed according to standard procedures for tissue sections. For direct immunofluorescence, PE-labeled anti-ERα antibody and PE-labeled anti-ERβ antibody were used, whereas the primary antibodies used for indirect immunofluorescence were anti-loricrin antibody, anti-involucrin antibody, and anti-cytokeratin-10 antibody. The secondary antibody was Alexa Fluor 647-labeled rabbit anti-mouse antibody or Alexa Fluor 488-labeled goat anti-rabbit antibody. Cell nuclei was stained with DAPI. Cell nuclei, ERs, loricrin, involucrin, and cytokeratin-10 were observed in cells using an IX70 microscope coupled with a DP80 digital camera (Olympus). We performed at least three independent studies, which yielded similar results. A representative experiment is shown in the figures.

The experimental data were evaluated for statistical significance using Student’s t-test (two-tailed). A p value < 0.05 was considered statistically significant. Data are expressed as mean ± SD.

Estrogen can increase the expression of ER [16] [17], which may enhance its effect, and binding to the ER improves cell proliferation in the skin. Therefore, the effect of estrogen or RE on ER expression was evaluated using NHDF. First, the biotoxicity of E2 and RE were evaluated, and biotoxicity was observed at 500 μM E2 and 800 μg/mL RE. Therefore, the effects on ER expression and cell proliferation were evaluated at concentrations lower than 500 μM E2 and 800 μg/mL RE. And it was found that E2 upregulated ERα and ERβ expression in NHDF, reaching a maximum at 50 μM of E2. RE also upregulated ERα and ERβ expression to a maximum of 80 μg/mL of RE (Figure 1(A)). In addition, the 50 μM E2 treatment significantly upregulated the ER fluorescence intensity, whereas treatments with 10 μM and 100 μM E2 did not affect ER expression (Figure 1(B)). Similarly, the ER fluorescence intensity after RE treatment of 80 μg/mL was significantly upregulated, whereas other concentrations of RE did not affect the expression of ER (Figure 1(B)). We evaluated the effect of ER upregulation on cell proliferation and determined that E2 did not affect cell proliferation, although treatment with 80 μg/mL RE significantly promoted cell proliferation by upregulation of the ER. In contrast, other concentrations of RE did not affect cell proliferation since the ER was not regulated (Figure 1(C)).

To evaluate the effect of stachyose and catalpol on ER upregulation, the concentrations of stachyose and catalpol were measured using HPLC. The 80 μg/mL RE contained 15 μM stachyose and 2.8 μM catalpol. Therefore, we evaluated the effects of 15 μM stachyose and 2.8 μM catalpol on ER expression and cell proliferation and observed that 15 μM stachyose upregulated ERα and ERβ expression in NHDF (Figure 2(A)); however, 2.8 μM catalpol did not affect ER expression. We then evaluated the effect of ER upregulation on cell proliferation. The results indicated that 15 μM stachyose promoted cell proliferation in NHDF (Figure 2(B)), whereas the 2.8 μM catalpol had no effect. These results suggest that ER upregulation by RE is activated by stachyose. In fact, cell proliferation by stachyose was similar to the proliferation by RE.

The dermis ER is involved in the production of cell growth factors such as bFGF, TGF-β1, and EGF and is considered crucial for maintaining dermis function. We investigated whether RE enhances ER expression in NHDF and found that stimulation with 80 μg/mL RE upregulated ERα and ERβ after 24 h, significantly promoted cell proliferation by ER upregulation. Accordingly, we evaluated the time-dependent changes in cell proliferation, ER expression and cell growth factors production using NHDF. The cell proliferation increased in the vehicle and 80 μg/mL RE in a time-dependent manner, and cell proliferation of RE treatment significantly increased compared to that in the vehicle between 18 h and 24 h (Figure 3(A)). In addition, ER expression increased in the vehicle and 80 μg/mL RE in a time-dependent manner, and ER expression of RE treatment significantly increased compared to that in the vehicle between 6 h and 24 h (Figure 3(B)). The production of bFGF was most increased by vehicle and 80 μg/mL RE at 9 h, and the bFGF production of RE treatment significantly increased compared with that of the vehicle between 9 h and 24 h (Figure 3(C)). TGF-β1

production increased with the vehicle and 80 μg/mL RE in a time-dependent manner, and the TGF-β1 production of RE significantly increased compared with the vehicle between 9 h and 24 h (Figure 3(D)). Moreover, the 80 μg/mL RE treatment increased EGF production in a time-dependent manner, and the EGF production of RE treatment significantly increased compared with that of the vehicle between 18 h and 24 h (Figure 3(E)).

The formulation of cell growth factor by estrogen is related to ECM production and cell proliferation, and is considered crucial for maintaining dermal elasticity. To evaluate whether upregulation of ER improved the estrogen effect, we evaluated cell proliferation and ECM production. Treatment with 80 μg/mL RE increased cell proliferation in NHDF, while 10 μM E2 did not affect cell proliferation. Furthermore, cell proliferation was improved by the E2 and RE mixture and was significantly greater than that by E2 or RE separately (Figure 4(A)). In addition, 80 μg/mL RE increased hyaluronic acid production in NHDF, whereas 10 μM E2 had no effect on this production. Similar to the results for cell proliferation, the E2 and RE mixture improved hyaluronic acid production significantly more than E2 or RE alone (Figure 4(B)). Collagen production returned correlating results, in that 80 μg/mL RE increased type I collagen production in NHDF, whereas 10 μM E2 had no effect, and the E2 and RE mixture was superior to E2 or RE individually at improving collagen production (Figure 4(C)). These results may reflect the improved estrogen action due to ER upregulation of RE.

Aging and photoaging can cause wrinkles and sagging in the skin because they reduce ER expression and decrease the production of hyaluronic acid and collagen [3] [13] [14]. We evaluated the effects on ER expression, cell proliferation, and ECM production using NHDF with different PDLs. Although high PDL cells reduced the expression of ERα and ERβ compared to low PDL cells, treatment with 80 μg/mL RE significantly upregulated the ER fluorescence intensity in the high PDL cells and low PDL cells (Figure 5(A), Figure 5(B)). In addition, the high PDL cells decreased cell proliferation, hyaluronic acid and type I collagen production compared to low PDL cells. In contrast, RE treatment significantly increased cell proliferation and hyaluronic acid and type I collagen production in the high PDL cells and low PDL cells (Figures 5(C)-(E)).

UV radiation through sunlight is a strong accelerator of skin aging, and is termed photoaging. UVA is known to be responsible for the progression of photoaging because it directly affects dermal fibroblasts. Thus, we evaluated the effects on ER expression, cell proliferation, and production of ECM using NHDF irradiated with UVA. Although the expression of ERα and ERβ were downregulated by UVA irradiation, RE treatment significantly upregulated the ER fluorescence intensity in the UVA irradiated cells and sham-UVA irradiated cells (Figure 6(A), Figure 6(B)). In addition, cell proliferation was decreased by UVA irradiation, although when both cells were treated with RE, the proliferation significantly increased (Figure 6(C)). Moreover, the production of hyaluronic acid and type I collagen similarly lowered by UVA irradiation, whereas RE treatment significantly increased the production of hyaluronic acid and type I collagen in the UVA irradiated cells and sham-UVA irradiated cells (Figure 6(D), Figure 6(E)).

The epidermal ER is involved in epidermal differentiation and is considered crucial for maintaining epidermal homeostasis [12]. We investigated whether RE enhances ER expression in NHDF and found that stimulation with RE upregulates ERs. Thus, we evaluated the effect of RE on the epidermal ER using immortalized NHEK. The results showed that 50 μM E2 upregulated ERα and ERβ expression in NHDF; however, 10 μM E2 did not affect ER expression. Similarly, 80 μg/mL RE upregulated ERα and ERβ expression in immortalized NHEK, but 8 μg/mL had no effect (Figure 7(A)). In addition, the ER fluorescence intensity after E2 treatment was significantly upregulated by treatment with 50 μM E2. In contrast, treatment with 10 μM E2 did not affect the ER expression. The ER fluorescence intensity was significantly upregulated after RE treatment with 80 μg/mL RE, but treatment with 8 μg/mL again had no effect (Figure 7(B)). Next, we evaluated the effect of ER upregulation on cell proliferation. E2 did not affect cell proliferation. Treatment with 80 μg/mL RE significantly promoted cell proliferation by upregulating ER. In contrast, treatment with 8 μg/mL RE did not affect cell proliferation (Figure 7(C)). The results of NHEK coincided with the effect of ER upregulation on NHDF. These results suggest that RE activates the epidermis and dermis.

Epidermal ER is considered crucial for maintaining epidermal homeostasis. We investigated immortalized NHEK and RE and found that ERs were upregulated in NHEK following RE stimulation. Accordingly, we evaluated the time-dependent changes in ER expression and localization using a reconstructed human epidermal model, which revealed that RE increased ER expression in the skin model in a time-dependent manner. Interestingly, ER expression was observed in the stratum corneum from the basal layer of ERα and in the stratum corneum from the granular layer of ERβ (Figure 8(A)).

Furthermore, we evaluated the influence of ER upregulation on the expression of the epidermal differentiation markers loricrin, involucrin, and cytokeratin-10, to evaluate the influence of ER upregulation on epidermal differentiation. RE promoted the expression of loricrin, involucrin, and cytokeratin-10 proteins compared to their expression in the vehicle (Figure 8(B)).