Improvement of Damage in Human Dermal Fibroblasts by 3,5,7-Trimethoxyflavone from Black Ginger (Kaempferia parviflora)

Reactive oxygen species (ROS) are generated during intrinsic (chronological aging) and extrinsic (photoaging) skin aging. Therefore, antioxidants that inhibit ROS production may be involved in delaying skin aging. In this study, we investigated the potential effects of compounds isolated from black ginger, Kaempferia parviflora, a traditional medicinal plant, on normal human dermal fibroblasts in the context of inflammation and oxidative stress. The isolated compounds were structurally characterized as 5-hydroxy-7-methoxyflavone (1), 3,7-dimethoxy-5-hydroxyflavone (2), 5-hydroxy-3,7,3,4-tetramethoxyflavone (3), 7,4-dimethylapigenin (4), 3,7,4-trimethylkaempferol (5), and 3,5,7-trimethoxyflavone (6), using nuclear magnetic resonance spectroscopy (NMR) and liquid chromatography–mass spectrometry (LC/MS) analyses. These flavonoids were first evaluated for their ability to suppress extracellular matrix degradation in normal human dermal fibroblasts. Of these, 3,5,7-trimethoxyflavone (6) significantly inhibited the tumor necrosis factor (TNF)-α-induced high expression and secretion of matrix metalloproteinase (MMP)-1 by cells. We further found that 3,5,7-trimethoxyflavone suppressed the excessive increase in ROS, mitogen-activated protein kinases (MAPKs), Akt, and cyclooxygenase-2 (COX-2)and increased heme oxygenase (HO)-1 expression. The expression of pro-inflammatory cytokines, including interleukin (IL)-1β, IL-6, and IL-8, was also suppressed by 3,5,7-trimethoxyflavone (6). Taken together, our results indicate that 3,5,7-trimethoxyflavone (6) isolated from K. parviflora is a potential candidate for ameliorating skin damage.


Introduction
The skin is a major defensive organ formed from the human ectoderm; it performs sensory, control, and protective functions when in direct contact with latent harmful factors [1]. Intrinsic aging appears with a gradual decrease in cell activity of skin with age, and is caused by reactive oxygen species (ROS) produced from the cell metabolic process [2,3]. External aging and skin damage occur upon direct exposure to the external environment and are caused by environmental hazards, such as pollution, chemicals, smoking, and ultraviolet (UV) radiation [4,5]. UV radiation substantially contributes to various skin injuries and diseases, including skin aging and inflammatory skin diseases [6,7]. UV radiation induces intracellular ROS formation, causing widespread inflammatory damage from the epidermis to the dermis, resulting in accumulative skin damage, such as skin pigmentation and photoaging [8,9]. ROS generation causes intracellular oxidative damage and changes, induced aging damage to normal human dermal fibroblasts (NHDFs) and assessed the mechanism of action of these compounds.

Plant Material
K. parviflora rhizomes were collected in January 2020 from Chiang Mai City, northern Thailand. This material was authenticated by K. H. Kim. A voucher specimen (SKKU-BG 1908) was stored in the herbarium of the School of Pharmacy, Sungkyunkwan University, Suwon, Korea.

Cell Culture and Sample Preparations
We purchased NHDFs from PromoCell GmbH (Heidelberg, Germany). Cells were maintained in Dulbecco's modified Eagle medium (Corning, Manassas, VA, USA). The medium consisted of 10% fetal bovine serum (Atlas, Fort Collins, CO, USA) and 100 U/mL penicillin-streptomycin solution (Gibco, Grand Island, NY, USA). Cells were cultured in a humid atmosphere containing 5% CO 2 at 37 • C. The six isolated compounds for cell treatment were prepared by dissolving them in dimethyl sulfoxide (DMSO; Sigma-Aldrich, St. Louis, MO, USA), and the final concentration was kept below 0.1%. TNF-α (PeproTech, Rocky Hill, NJ, USA) was prepared by dissolving in 1% bovine serum albumin.

Cell Viability
We seeded NHDFs at 1 × 10 4 cells/well in 96-well plates and incubated them for 24 h. To create starvation conditions, the medium was then exchanged for serum-free. Serum starvation drives all cells to a phase of growth arrest, making all cells have the same cell cycle [39]. After 24 h, the NHDFs were exposed to each concentration of the compound. After 24 h, to measure cell viability, EZ-Cytox solution (Dogen, Seoul, Korea) was added to each cell-containing well, and absorbance was then determined after incubation for 2 h. The measurement was performed using a SPARK 10M device (Tecan Group Ltd., Männedorf, Switzerland), and the wavelength was set to 450/600 nm. Cell viability for each sample was presented as percent of the vehicle control.

Evaluation of Intracellular ROS
We seeded NHDFs at 1 × 10 4 cells/well in 96-well black plates, incubated them for 24 h, and then replaced the medium with serum-free medium to create starvation conditions. After 24 h, cells were treated with 3,5,7-trimethoxyflavone for 24 h. The cells were then exposed to 20 ng/mL TNF-α (PeproTech, Rocky Hill, NJ, USA) for 15 min. After 15 min of incubation, staining was carried out with dichlorofluorescein diacetate (DCFDA; Sigma-Aldrich) and washed with phosphate-buffered saline (Welgene, Gyeongsangbuk, Korea). Fluorescence imaging was performed using a fluorescence microscope IX51 (Olympus, Tokyo, Japan). Fluorescence was measured using a SPARK 10M, and the wavelength was set to 485/535 nm. Results of intracellular ROS were presented as percent of the vehicle control.

Real-Time Quantitative Reverse Transcription PCR
We seeded NHDFs at 3 × 10 5 cells/well in 6-well plates, incubated them for 24 h, and then replaced the medium with a serum-free medium to create starvation conditions. After 24 h, cells were treated with 3,5,7-trimethoxyflavone for 24 h. The cells were then exposed to 20 ng/mL TNF-α for 15 min. To measure expression of IL-1β, IL-6, IL-8, and β-actin, the cells were harvested after 4 h. To test for MMP-1 and procollagen I α1(COLIA1), the cells were harvested after 24 h. Next, the cells were harvested and isolated to cellular RNA by a RNeasy Mini Kit (QIAGEN, Hilden, Germany). Complementary DNA synthesis was performed using the RevertAid First Strand cDNA Synthesis Kit (Thermo Fisher Scientific, Waltham, MA, USA). Real-time PCR was carried out using PowerUp SYBR PCR Master Mix (Applied Biosystems, Waltham, MA, USA). The reaction was performed using the QuantStudio™ 3 Real-Time PCR System (Applied Biosystems), and the thermal conditions were set to 40 cycles (95 • C for 15 s, 60 • C for 30 s, 95 • C for 30 s). The primer sequences are shown in a previous paper [40], and β-actin was used as the reference gene.

Enzyme-Linked Immunosorbent Assay (ELISA)
We seeded NHDFs at 2 × 10 4 cells/well in 48-well plates, incubated them for 24 h, and then replaced the medium with a serum-free medium to create starvation conditions. After 24 h, the NHDF was treated with 3,5,7-trimethoxyflavone for 1 h, followed by exposure to 20 ng/mL of TNF-α. To measure protein expression for IL-1β, IL-6, and IL-8, the medium was collected from cells after 12 h. To measure protein expression for MMP-1 and procollagen I α1 (COLIA1), the medium was collected from cells after 24 h. Protein secretion was determined using a human Total MMP-1 DuoSet ELISA kit and a Human Pro-Collagen I alpha 1 DuoSet ELISA (R&D systems, Minneapolis, MN, USA). The absorbance was measured using a SPARK 10M spectrophotometer, and the wavelength was set to 550/600 nm. Results of protein secretion were presented as percent of the vehicle control.

Western Blotting
We seeded NHDFs at 3 × 10 5 cells/well in 6-well plates, incubated them for 24 h, and then replaced the medium with a serum-free medium to create starvation conditions. After 24 h, cells were treated with 3,5,7-trimethoxyflavone for 24 h. The cells were then exposed to 20 ng/mL TNF-α. To test for phospho-ERK, ERK, phospho-p38, p38, phospho-JNK, JNK, and GAPDH, the cells were collected after 15 min. To test for phospho-Akt, Akt, COX-2, heme oxygenase 1 (HO-1), and GAPDH, the cells were collected after 6 h. The cells were lysed with 1× radioimmunoprecipitation assay (RIPA) buffer (Tech & Innovation, Gangwon, Korea). The lysate was centrifuged, and the supernatant was prepared as eluted protein samples. The protein concentration was determined using a BCA Protein Assay Kit (Merck). Equal protein levels were analyzed using Western blotting. The primary antibodies, phospho-ERK, ERK, phospho-p38, p38, phospho-JNK, JNK, phospho-Akt, Akt, COX-2, HO-1, COX-2, and GAPDH (Cell Signaling Technology, Danvers, MA, USA) were reacted for 4 h at 20 ± 5 • C. The secondary antibodies, goat anti-rabbit IgG-HRP, and goat anti-mouse IgG-HRP (Santa Cruz Biotechnology, Dallas, TX, USA), were incubated for 1 h at room temperature. Protein bands were visualized using SuperSignal ® West Femto Maximum Sensitivity Chemiluminescent Substrate (Thermo Fisher Scientific) and Fusion Solo Chemiluminescence System (PEQLAB Biotechnologie GmbH, Erlangen, Germany). The band densities were calculated based on the ratio to the GAPDH band. Results were calculated from band density and were presented as percent of vehicle control.

Statistical Analysis
The experimental data are presented as mean ± standard error of the mean (SEM). Differences of each group were assessed by using one-way analysis of variance (ANOVA) followed by Tukey's HSD (honestly significant difference) test. A p-value of 0.05 or less was considered statistically significant.

Effect of Flavonoids 1-6 Isolated from K. parviflora on Viability of NHDFs
Ahead of investigating the anti-aging effects of flavonoids 1-6 isolated from K. parviflora, their effects on NHDF cell viability were measured. As shown in Figure 2, flavonoids 1-6 did not exhibit considerable cytotoxicity against NHDFs at 100 µM. Therefore, all flavonoids were used in subsequent experiments at concentrations up to 100 µM.

Effect of Flavonoids 1-6 Isolated from K. parviflora on Viability of NHDFs
Ahead of investigating the anti-aging effects of flavonoids 1-6 isolated from K. parviflora, their effects on NHDF cell viability were measured. As shown in Figure 2, flavonoids 1-6 did not exhibit considerable cytotoxicity against NHDFs at 100 μM. Therefore, all flavonoids were used in subsequent experiments at concentrations up to 100 μM. oxidants 2022, 11, x FOR PEER REVIEW 8 of Figure 2. Survival of normal human dermal fibroblasts (NHDFs) under treatment with flavono 1-6 isolated from K. parviflora. We seeded NHDFs at 1 × 10 4 cells/well in 96-well plates, incubat them for 24 h, and replaced the medium with a serum-free medium to create starvation conditio After 24 h, NHDFs were exposed to each concentration of the compounds. After 24 h, the measu ment of cell viability was conducted with EZ-Cytox solution. Results of cell viability were present as percent of the vehicle control. The results were obtained through three replicate experiments, a the graphs are represented as mean ± SEM.

Effect of Flavonoids Isolated from K. parviflora on MMP-1 Secretion in TNF-α-Stimulated NHDFs
We then investigated the inhibitory effects of the six flavonoids (1-6) by measuri the MMP-1 secretion by TNF-α-stimulated NHDFs. As shown in Figure 3, TNF-α ( ng/mL) significantly increased MMP-1 secretion by 3.37 ± 0.08-fold (p < 0.01) compar with that of the vehicle control. Among the six flavonoids from K. parviflora, 3,5,7-t methoxyflavone (6) potently reduced the secretion of MMP-1 induced by TNF-α. Tre ment with 50 and 100 μM 6 suppressed the secretion of MMP-1 to 1.99 ± 0.05-fold (p < 0.0 and 1.76 ± 0.06-fold (p < 0. 01), compared with that of the control. Therefore, we focus on 3,5,7-trimethoxyflavone in further studies to elucidate the mechanism by which demonstrates protective effects against NHDF damage by TNF-α. Survival of normal human dermal fibroblasts (NHDFs) under treatment with flavonoids 1-6 isolated from K. parviflora. We seeded NHDFs at 1 × 10 4 cells/well in 96-well plates, incubated them for 24 h, and replaced the medium with a serum-free medium to create starvation conditions. After 24 h, NHDFs were exposed to each concentration of the compounds. After 24 h, the measurement of cell viability was conducted with EZ-Cytox solution. Results of cell viability were presented as percent of the vehicle control. The results were obtained through three replicate experiments, and the graphs are represented as mean ± SEM.
oxidants 2022, 11, x FOR PEER REVIEW 9 of Figure 3. Effects of flavonoids 1-6 isolated from K. parviflora on MMP-1 secretion in normal hum dermal fibroblasts (NHDFs). We seeded NHDFs at 2 × 10 4 cells/well in 48-well plates, incuba them for 24 h, and replaced the medium with a serum-free medium to create starvation conditio After 24 h, cells were treated with 50 and 100 μM of 6 for 1 h, and the cells were exposed to 20 ng/m TNF-α for 24 h. After 24 h, we measured MMP-1 secretion with an ELISA kit. Results of MM secretion were presented as percent of the vehicle control. The results were obtained through th replicate experiments, and the graphs are represented as mean ± SEM. ## p < 0.01 vs. vehicle contr * p < 0.05, ** p < 0.01 and *** p < 0.001 vs. TNF-α stimulated control.

Effect of 3,5,7-Trimethoxyflavone on Intracellular ROS Accumulation in TNF-α-Stimulat NHDFs
TNF-α is a stimulator that can cause oxidative stress in NHDFs. Therefore, we ev uated the effect of 3,5,7-trimethoxyflavone on ROS accumulation in TNF-α-stimulat NHDFs. As shown in the Figure 4, exposure of 20 ng/mL TNF-α did show dramatica increased ROS accumulation by 2.54 ± 0.06-fold (p < 0.05) compared with that in the vehi control. In contrast, 3,5,7-trimethoxyflavone (6) significantly reduced ROS accumulatio Treatment with 50 and 100 μM 6 suppressed ROS production to 1.88 ± 0.14-fold (p < 0.0 and 1.33 ± 0.09-fold (p < 0. 01), respectively, compared with that in the control. Effects of flavonoids 1-6 isolated from K. parviflora on MMP-1 secretion in normal human dermal fibroblasts (NHDFs). We seeded NHDFs at 2 × 10 4 cells/well in 48-well plates, incubated them for 24 h, and replaced the medium with a serum-free medium to create starvation conditions. After 24 h, cells were treated with 50 and 100 µM of 6 for 1 h, and the cells were exposed to 20 ng/mL TNF-α for 24 h. After 24 h, we measured MMP-1 secretion with an ELISA kit. Results of MMP-1 secretion were presented as percent of the vehicle control. The results were obtained through three replicate experiments, and the graphs are represented as mean ± SEM. ## p < 0.01 vs. vehicle control. ** p < 0.01 and *** p < 0.001 vs. TNF-α stimulated control.

Effect of 3,5,7-Trimethoxyflavone on TNF-α-Induced Phosphorylation of MAPKs in NHDFs
To improve the understanding of the mechanism of the cytoprotective action of 3,5,7-trimethoxyflavone (6), we evaluated its effect on MAPK phosphorylation in TNF-αstimulated NHDFs. MAPK protein expression was analyzed using Western blotting. As shown in Figure 5, exposure of 20 ng/mL TNF-α did show activation by of ERK, JNK, and p38 in NHDFs, which was reduced upon treatment with 6. ERK phosphorylation by TNF-α-stimulation was shown to increase by 2.72 ± 0.35-fold, whereas treatment with 6 suppressed ERK phosphorylation to 1.30 ± 0.42-fold (50 µM) and 0.77 ± 0.30-fold (100 µM, p < 0.05) compared with that in the control. JNK phosphorylation by TNF-α-stimulation was shown to increase by 8.15 ± 0.45-fold, whereas treatment with 6 suppressed it by 4.88 ± 0.24-fold (50 µM, p < 0.05), and 4.86 ± 0.13-fold (100 µM, p < 0.05) of the control. Further, TNF-α stimulation increased p38 phosphorylation by 6.43 ± 0.17-fold, whereas treatment with 6 suppressed it by 5.58 ± 0.42-fold (50 µM), and 5.20 ± 0.31-fold (100 µM) compared with the control; however, this difference was not significant. These results indicate that 3,5,7-trimethoxyflavone (6) can suppress the MAPK phosphorylation induced by TNF-α stimulation.  We seeded NHDFs at 1 × 10 4 cells/well in 96-well black plates incubated them for 24 h, and replaced the medium with a serum-free medium to create starvation conditions. After 24 h, cells were treated with 50 and 100 μM of 6 for 1h, and the cells were exposed to 20 ng/mL TNF-α for 15 min. The cell was stained with dichlorofluorescin diacetate (DCFDA) for 15 min, and photographs was observed with a microscope IX51. The measurement of fluorescence was conducted using a SPARK 10M. Results of intracellular ROS were presented as a percent of the vehicle control. The results were obtained through three replicate experiments, and the graphs are represented as mean ± SEM. # p < 0.05 vs. vehicle control. * p < 0.05 and * p < 0.05 and ** p < 0.01 vs. TNFα stimulated control.

Effect of 3,5,7-Trimethoxyflavone on TNF-α-Induced Phosphorylation of MAPKs in NHDFs
To improve the understanding of the mechanism of the cytoprotective action of 3,5,7trimethoxyflavone (6), we evaluated its effect on MAPK phosphorylation in TNF-α-stimulated NHDFs. MAPK protein expression was analyzed using Western blotting. As We seeded NHDFs at 1 × 10 4 cells/well in 96-well black plates incubated them for 24 h, and replaced the medium with a serum-free medium to create starvation conditions. After 24 h, cells were treated with 50 and 100 µM of 6 for 1h, and the cells were exposed to 20 ng/mL TNF-α for 15 min. The cell was stained with dichlorofluorescin diacetate (DCFDA) for 15 min, and photographs was observed with a microscope IX51. The measurement of fluorescence was conducted using a SPARK 10M. Results of intracellular ROS were presented as a percent of the vehicle control. The results were obtained through three replicate experiments, and the graphs are represented as mean ± SEM. # p < 0.05 vs. vehicle control. * p < 0.05 and ** p < 0.01 vs. TNF-α stimulated control.  NHDFs were seeded at a density of 3 × 10 5 cells/well in 6-well plates and incubated for 24 h, and then the medium was replaced with a serum-free medium to create starvation conditions. After 24 h, cells were treated with 50 and 100 μM of 6 for 1 h, and the cells were exposed to 20 ng/mL TNF-α for 15 min. Relative comparison of expression levels of p-ERK, ERK, p-JNK, JNK, p-p38, p38, and GAPDH proteins were determined with Western blotting. Results of each protein expression was presented as percent of the vehicle control. The results were obtained through three replicate experiments, and the graphs are represented as mean ± SEM. # p < 0.05 vs. vehicle control. * p < 0.05 vs. TNF-α stimulated control.

Effect of 3,5,7-Trimethoxyflavone on Akt Phosphorylation and Expression of COX-2 and HO-1 in TNF-α-Stimulated NHDFs
Immoderate production of ROS activates Akt, which causes inflammation by upregulating COX-2 [46]. To determine the protective effect of 3,5,7-trimethoxyflavone (6) against the inflammatory response, we evaluated its effect on Akt phosphorylation and COX-2 expression in TNF-α-stimulated NHDFs. Western blotting was used to measure protein expression levels of p-Akt, Akt, and COX-2. As shown in Figure 6, 20 ng/mL TNFα exposure induced Akt phosphorylation in NHDFs compared to that with the vehicle control, which was reduced upon treatment with 6. TNF-α stimulation increased Akt phosphorylation by 2.61 ± 0.38-fold, whereas treatment with 6 suppressed it by 1.13 ± 0.28- NHDFs were seeded at a density of 3 × 10 5 cells/well in 6-well plates and incubated for 24 h, and then the medium was replaced with a serum-free medium to create starvation conditions. After 24 h, cells were treated with 50 and 100 µM of 6 for 1 h, and the cells were exposed to 20 ng/mL TNF-α for 15 min. Relative comparison of expression levels of p-ERK, ERK, p-JNK, JNK, p-p38, p38, and GAPDH proteins were determined with Western blotting. Results of each protein expression was presented as percent of the vehicle control. The results were obtained through three replicate experiments, and the graphs are represented as mean ± SEM. # p < 0.05 vs. vehicle control. * p < 0.05 vs. TNF-α stimulated control.
An increase in HO-1 through the transcriptional activation of Nrf2 suppresses free radical production, thereby preventing inflammatory damage and apoptosis in human skin cells [47,48]. In Figure 6, exposure of 20 ng/mL TNF-α did not show alteration of HO-1 expression in NHDFs compared to that in the vehicle control; however, this was increased by treatment with 6. HO-1 expression increased by 1.68 ± 0.29-fold (50 μM) and 3.03 ± 0.36-fold (100 μM, p < 0.05) of the control upon treatment with 6. These results suggest that 3,5,7-trimethoxyflavone (6) inhibits the ROS accumulation induced by TNF-α stimulation through free radical trapping by HO-1. Figure 6. Effects of 3,5,7-trimethoxyflavone (6) on MAPK phosphorylation in TNF-α stimulated normal human dermal fibroblasts (NHDFs). (A) Bands of protein expressions. (B) Bar graphs of the relative expression. We seeded NHDFs at 3 × 10 5 cells/well in 6-well plates, incubated for 24 h, and then replaced the medium with a serum-free medium to create starvation conditions. After 24 h, cells were treated with 50 and 100 μM of 6 for 1 h, and the cells were exposed to 20 ng/mL TNF-α for 6 h. Protein expressions of p-Akt, Akt, COX-2, HO-1, and GAPDH were determined with Western blotting. Results of each protein expression was presented as percent of the vehicle control. The results were obtained through three replicate experiments, and the graphs are represented as mean ± SEM. # p < 0.05 and ## p < 0.01 vs. vehicle control. * p < 0.05 vs. TNF-α stimulated control. We seeded NHDFs at 3 × 10 5 cells/well in 6-well plates, incubated for 24 h, and then replaced the medium with a serum-free medium to create starvation conditions. After 24 h, cells were treated with 50 and 100 µM of 6 for 1 h, and the cells were exposed to 20 ng/mL TNF-α for 6 h. Protein expressions of p-Akt, Akt, COX-2, HO-1, and GAPDH were determined with Western blotting. Results of each protein expression was presented as percent of the vehicle control. The results were obtained through three replicate experiments, and the graphs are represented as mean ± SEM. # p < 0.05 and ## p < 0.01 vs. vehicle control. * p < 0.05 vs. TNF-α stimulated control.
An increase in HO-1 through the transcriptional activation of Nrf2 suppresses free radical production, thereby preventing inflammatory damage and apoptosis in human skin cells [47,48]. In Figure 6, exposure of 20 ng/mL TNF-α did not show alteration of HO-1 expression in NHDFs compared to that in the vehicle control; however, this was increased by treatment with 6. HO-1 expression increased by 1.68 ± 0.29-fold (50 µM) and 3.03 ± 0.36-fold (100 µM, p < 0.05) of the control upon treatment with 6. These results suggest that 3,5,7-trimethoxyflavone (6) inhibits the ROS accumulation induced by TNF-α stimulation through free radical trapping by HO-1.

Effect of 3,5,7-Trimethoxyflavone on MMP-1 and Pro-Collagen I α1 mRNA and Protein Expression in TNF-α-Stimulated NHDFs
UV radiation induces ROS accumulation, which alters the structure and function of genes and proteins, ultimately causing skin damage. It increases the expression and secretion of MMP-1 collagenase and induces collagen degradation. Eventually, the skin ECM disintegrates, leading to skin damage such as wrinkles [49,50]. Therefore, substances that inhibit collagenase activity are considered as potential therapeutic candidates for skin damage [51]. In Figure 7A, exposure of 20 ng/mL TNF-α did show increased MMP-1 expression in NHDFs compared to that in the vehicle control; this was reduced by 3,5,7-trimethoxyflavone (6) treatment. TNF-α stimulation also increased MMP-1 mRNA expression by 3.76 ± 0.03-fold, whereas treatment with 6 suppressed it by 2.53 ± 0.13-fold (50 µM) and 1.70 ± 0.23-fold (100 µM, p < 0.05) compared with that in the control. As shown in Figure 3, MMP-1 secretion was induced by TNF-α-stimulation, whereas treatment with 6 suppressed it. These results indicate that 3,5,7-trimethoxyflavone (6) inhibits MMP-1 expression induced by TNF-α stimulation. . We seeded NHDFs at 3 × 10 5 cells/well in 6-well plates, incubated them for 24 h, and then replaced the medium with a serum-free medium to create starvation conditions. After 24 h, cells were treated with 50 and 100 μM of 6 for 1 h, and the cells were exposed to 20 ng/mL TNF-α for 12 h. The mRNA expressions of MMP-1 and COLIA1 were determined with RT-qPCR. Results of each mRNA expression was presented as percent of the vehicle control. (B) The protein secretion of COLIA1. We seeded NHDFs at or 2 × 10 4 cells/well in 48-well plates, incubated them for 24 h, and then replaced the medium with a serum-free medium to create starvation conditions. After 24 h, NHDFs were exposed to 20 ng/mL TNF-α in the presence or absence of 6 for 24 h. The protein secretion of COLIA1 were determined with an ELISA kit. The results were obtained through three replicate experiments, and the graphs are represented as mean ± SEM. ## p < 0.01 and ### p < 0.001 vs. vehicle control. * p < 0.05 and ** p < 0.01 vs. TNF-α stimulated control.
In Figure 8A, exposure of 20 ng/mL TNF-α did show an increase in IL-1β, IL-6, and . We seeded NHDFs at 3 × 10 5 cells/well in 6-well plates, incubated them for 24 h, and then replaced the medium with a serum-free medium to create starvation conditions. After 24 h, cells were treated with 50 and 100 µM of 6 for 1 h, and the cells were exposed to 20 ng/mL TNF-α for 12 h. The mRNA expressions of MMP-1 and COLIA1 were determined with RT-qPCR. Results of each mRNA expression was presented as percent of the vehicle control. (B) The protein secretion of COLIA1. We seeded NHDFs at or 2 × 10 4 cells/well in 48-well plates, incubated them for 24 h, and then replaced the medium with a serum-free medium to create starvation conditions. After 24 h, NHDFs were exposed to 20 ng/mL TNF-α in the presence or absence of 6 for 24 h. The protein secretion of COLIA1 were determined with an ELISA kit. The results were obtained through three replicate experiments, and the graphs are represented as mean ± SEM. ## p < 0.01 and ### p < 0.001 vs. vehicle control. * p < 0.05 and ** p < 0.01 vs. TNF-α stimulated control.
Procollagen is secreted out of the cell, a part of the N-terminal peptide is cleaved by enzymes, and the cleaved procollagen is combined with the help of oxygen, iron ions, and ascorbic acid to form collagen in the form of fibers [52]. Therefore, the expression of procollagen COLIA1 was examined to indirectly evaluate collagen synthesis. In the Figure 7A, exposure of 20 ng/mL TNF-α did show a decrease in COLIA1 expression in NHDFs compared to that in the vehicle control. TNF-α stimulation decreased COL1A1 mRNA expression to 0.38 ± 0.06-fold, but this remained unchanged upon treatment with 3,5,7-trimethoxyflavone (6) at 50 and 100 µM. In Figure 7B, COLIA1 secretion induced by TNF-α stimulation decreased to 12.59 ± 1.24 ng/mL, compared with that in the vehicle control (21.13 ± 1.86 ng/mL). Treatment with 100 µM of 6 increased COLIA1 to 15.65 ± 0.46 ng/mL; however, this difference was not significant. Despite the lack of significance, this result indicated that 3,5,7-trimethoxyflavone (6) can reverse the decrease in COLIA1. In summary, these results suggest that 3,5,7-trimethoxyflavone (6) has the potential to suppress the advanced cutaneous ECM degradation induced by oxidative stress.
These results indicate that 3,5,7-trimethoxyflavone (6) suppresses TNF-α stimulation-induced proinflammatory cytokines IL-1β, IL-6, and IL-8. Therefore, 3,5,7-trimethoxyflavone (6) has the potential to suppress inflammatory responses in the skin caused by proinflammatory cytokines. The mRNA expressions of IL-1β, IL-6, and IL-8. We seeded NHDFs at 3 × 10 5 cells/well in 6-well plates, incubated them for 24 h, and then replaced the medium with a serum-free medium to create starvation conditions. After 24 h, NHDFs were exposed to 20 ng/mL TNF-α in the presence or absence of 6 for 4 h. The mRNA expressions of IL-1β, IL-6, and IL-8 were determined with RT-qPCR. Results of each mRNA expression was presented as percent of the vehicle control. (B) The protein secretion of IL-1β, IL-6, and IL-8. We seeded NHDFs at or 2 × 10 4 cells/well in 48-well plates, and incubated them for 24 h, and then replaced the medium with a serum-free medium to create starvation conditions. After 24 h, cells were treated with 3,5,7-trimethoxyflavone for 24 h. The cells were then exposed to 20 ng/mL TNF-α for 12 h. The protein secretion of IL-1β, IL-6, and IL-8 were determined with an The mRNA expressions of IL-1β, IL-6, and IL-8. We seeded NHDFs at 3 × 10 5 cells/well in 6-well plates, incubated them for 24 h, and then replaced the medium with a serum-free medium to create starvation conditions. After 24 h, NHDFs were exposed to 20 ng/mL TNF-α in the presence or absence of 6 for 4 h. The mRNA expressions of IL-1β, IL-6, and IL-8 were determined with RT-qPCR. Results of each mRNA expression was presented as percent of the vehicle control. (B) The protein secretion of IL-1β, IL-6, and IL-8. We seeded NHDFs at or 2 × 10 4 cells/well in 48-well plates, and incubated them for 24 h, and then replaced the medium with a serum-free medium to create starvation conditions. After 24 h, cells were treated with 3,5,7-trimethoxyflavone for 24 h. The cells were then exposed to 20 ng/mL TNF-α for 12 h. The protein secretion of IL-1β, IL-6, and IL-8 were determined with an ELISA kit. The experiments were performed in triplicate, and the graphs are represented as mean ± standard error of the mean (SEM). # p < 0.05 and ## p < 0.01 vs. vehicle control. * p < 0.05 and ** p < 0.01 vs. TNF-α stimulated control.

Discussion
The skin is the largest organ of the body. It acts as a barrier against chemical and physical pollutants and protects internal organs. Proteins of extracellular matrix are generated by fibroblasts present in the dermis and are accountable for the elasticity and strength of the skin. Breakdown of ECM from the dermis results in skin aging. There are many types of collagens in the dermal ECM (III, V, VII); however, type I collagen is the most important structural protein. The main cause of aged skin that appears thin, smooth, dry, and inelastic is a decrease in collagen fibers and type I collagen production. In the process of skin aging, the collagenase MMP-1 destroys collagen fibrils. It has been shown that with aging, MMP-1 levels increase and collagen expression in human skin decreases. Furthermore, excessive ECM degradation of the dermis by MMP-1 contributes to inflammation of connective tissue. Therefore, the balance between MMP 1 and type I collagen expression plays an important role in skin aging [55].
The skin can age internally by a temporal aging process that affects all body organs or externally because of environmental factors, such as sun exposure and smoking. External and internal skin aging produces inflammatory cytokines, which are closely related to the skin inflammation caused by them. In the process of intrinsic aging and extrinsic aging, various cells of the skin produce the inflammatory cytokine TNF-α, which functions as a major regulator of cell metabolism and activity. UV irradiation stimulates cell surface receptors, which are the main contributors of skin aging. Of the ligands for these receptors, TNF-α induces increased MMP expression and decreased collagen expression, leading to ECM disruption. Moreover, TNF-α is considered an important modulator of inflammatory dermatological symptoms and diseases. TNF-α increases pro-inflammatory cytokines such as IL-1 and IL-6, which activate NF-κB [25,56,57]. Therefore, regulation of TNF-α activity can be an attractive strategic method for searching for candidate substances to prevent skin damage and aging [55,58].
COX-2, IL-1, and IL-6 are pro-inflammatory mediators that play a major role in causing skin aging and inflammatory skin symptoms. In skin aging, COX-2 acts as a mediator for the biosynthesis of prostaglandin E2, which contributes to the production of MMPs. IL-1 and IL-6 are regulators that contribute to skin wrinkle formation, including inhibition of collagen biosynthesis [59]. Luteolin, a flavonoid isolated and reported from various natural products, suppresses ROS and pro-inflammatory mediators and promotes photoaging of the epidermis and dermis to reduce IL-6, IL-22, IL-17, and COX-2 [46]. Similarly, 3,5,7trimethoxyflavone (6) significantly inhibits TNF-α-induced expression of IL-1β, IL-6, and COX-2 and is expected to have a protective effect on skin damage based on inhibition of inflammation-related reactions.
Oxidative stress is also an important factor related to skin aging. Reactive oxygen species generated by intrinsic (chronic aging) and total permeable aging (optical) stimulate MMP expression, suppress growth factor-beta signaling, suppress collagen degradation, and trigger collagen fibrous biosynthesis [60]. Therefore, development of skin antioxidants with natural products is one of the major strategies of skincare product development in the cosmetics industry [61]. According to a recent study by Li et al., orange lowtemperature crimping oil has a high content of polymethoxyflavones and was effective in preventing UVB-induced oxidative damage in mouse skin [62]. Similarly, in this study, 3,5,7trimethoxyflavone (6) had the effect of suppressing the reactive oxygen species produced by stimulation with TNF-α in NHDFs. The position and number of methoxy groups in the polymethoxyflavone structure related to the antioxidant properties, suggest a relevance for these pharmacological effects.
MAPK plays a central role in controlling skin damage in combination with two downstream pathways associated with oxidative stress and inflammatory responses, AP-1 and NF-κB [62,63]. In addition, activation of MAPK induces phosphorylation and rearrangement of NF-κB, producing MMP-1 and inflammatory cytokines [62,64]. Therefore, retrograde movement of this pathway plays an important role in delaying skin aging.
However, this study had some limitations. For example, the potential of 3,5,7trimethoxyflavone (6) to prevent skin aging has been studied only in HDFs. To completely understand the anti-aging effect of 3,5,7-trimethoxyflavone (6), research needs to be extended to many different cell lines such as melanocytes, keratinocytes, and organic 3D skin models.
Our previous study has shown that 5,7,4 trimethoxyflavone is effective at low concentrations of 6.25 and 12.5 µM [63]. On Because 3,5,7-trimethoxyflavone (6) compound 6 is effective at 50 or more, the question of weak activity may be raised. However, skin improvers directly penetrate the skin to show their effects. Therefore, the active compound must penetrate the skin barrier without causing skin irritation. Recent studies have reported that the microemulsion formulation quercetin increases skin penetration [64]. Therefore, it is believed that additional studies such as topical and transdermal delivery are needed to determine if a physiological approach is possible.

Conclusions
The results of this study indicate that 3,5,7-trimethoxyflavone (6) isolated from K. parviflora has inhibitory effects on TNF-α-induced MMP-1 in NHDFs. It was found that 3,5,7-trimethoxyflavone inhibited TNF-α-induced ROS, which plays a key role in the inflammatory response and ECM degradation that occurs during skin damage, such as skin aging and various cutaneous lesions. The changes by which 3,5,7-trimethoxyflavone (6) reduces TNF-α-stimulated responses in NHDFs are correlated with the suppression of Akt, COX-2, MAPK activation, and induction of HO-1. Moreover, 3,5,7-trimethoxyflavone reduced the expression of proinflammatory cytokine mediators induced by TNF-α stimulation, including IL-1β, IL-6, and IL-8. These findings provide evidence that 3,5,7-trimethoxyflavones may protect against skin damage caused by accumulation of oxidative stress. Additional experiments are required to fully understand the mechanism of 3,5,7-trimethoxyflavone activity; however, it is a potential substance for preventing skin damage, including skin aging and various skin lesions.