Oddioside A, a New Phenolic Glycoside Isolated from the Fruits of Morus alba (Mulberry), Protects TNF-α-Induced Human Dermal Fibroblast Damage

In our preliminary study, a hot water extract from the fruits of Morus alba (mulberry) inhibited the secretion of metalloproteinase-1 (MMP-1) against tumor necrosis factor-α (TNF-α)-stimulated human dermal fibroblasts (HDFs), and therefore we researched its active compounds. In the present study, a new phenolic glycoside (oddioside A, 1) and 21 known compounds (2−22) were isolated from the hot water extract from the fruits of M. alba by repeated chromatography. The chemical structure of the new compound 1 was elucidated by its spectroscopic data (1D− and 2D−NMR and HRMS) measurement and by acidic hydrolysis. The presence of sargentodoside E (2), eugenyl glucoside (6), 2-O-β-d-glucopyranosyl-4,6-dihydroxybenzaldehyde (7), 7S,8R-erythro-7,9,9’-trihydroxy-3,3’-dimethoxy-8-O-4’-neolignan-4-O-β-d-glucopyranoside (11), pinoresinol-4-O-β-d-glucopyranoside (12), taxifolin-7-O-β-d-glucopyranoside (20), and pinellic acid (21) were reported from M. alba for the first time in this study. The new compound oddioside A (1) suppressed the secretion of MMP-1 and increased collagen in TNF-α-stimulated HDFs. In addition, the phosphorylation of mitogen-activated protein kinases (MAPKs) was inhibited by oddioside A. In conclusion, the extract from fruits of M. alba and its constituent oddioside A may be a potential agent to prevent inflammation-related skin aging and other skin disorders.


Introduction
The skin is an organ that can protect the body from water loss or microbial infection, and is directly affected by external environmental factors such as air pollution and UV irradiations [1]. Continuous exposure to these external factors accelerates skin aging such as deep and shallow wrinkles, sagging, rough and dry skin, and pigmentation. Regarding skin aging, many people spend a lot on cosmetics or therapeutics that prevent or improve aging. This demand for cosmetics promotes research on skin aging [2].
Aging is largely divided into intrinsic aging and extrinsic aging. Intrinsic aging is a process that changes physiologically over time, and extrinsic aging is caused by external environmental factors including tobacco, air pollution, and UV irradiations [3]. In photoaging, symptoms such as wrinkle formation, low elasticity, and rough skin appear due to structural changes in the dermal connective tissue [4].

Acidic Hydrolysis of 1 and Sugar Identification
Compound 1 (1.0 mg) was subjected to an acid hydrolysis and the absolute configuration of glucose and rhamnose in 1 was confirmed as D and L, respectively, by the method from Tanaka et al. [18].

Cell Viability
HDFs were seeded in 96-well plates at 1 × 10 4 cells/well and cultured for 24 h. Then, cell medium was replaced with a serum-free condition and incubated overnight. Cells were treated with each concentration (µg/mL or µM) of samples and incubated for 24 h. Then, to measure the cell viability, the supernatant was removed, and 100 µL of 10% EZ-Cytox solution (DoGenBio, Seoul, Korea) in serum-free DMEM was put into each well and incubated for 1 h. Absorbance was measured with a microplate reader (SPARK 10M; Tecan, Männedorf, Switzerland) using a wavelength of 450/600 nm.

Enzyme-Linked Immunosorbent Assay (ELISA)
HDFs were seeded in 48-well plates at 2 × 10 4 cells/well and incubated for 24 h. Then, cell medium was replaced with a serum-free condition and incubated overnight. After 24 h, non-toxic concentrations of samples were pretreated with HDFs for 1 h, followed by treatment with 20 ng/mL TNF-α for 15 min. Absorbance was measured with a microplate reader (SPARK 10M) using a wavelength of 450/600 nm.

ROS Assay
HDFs were seeded in 96-well plates at 1 × 10 4 cells/well and incubated for 24 h. Then, cell medium was replaced with a serum-free condition and incubated overnight. After 24 h, non-toxic concentrations of sample were pretreated with HDFs for 1 h, followed by treatment with 20 ng/mL TNF-α and 10 µM dichlorofluorescein diacetate (DCFDA; Sigma-Aldrich) for 15 min. After 15 min of incubation, washing with Dulbecco's phosphatebuffered salines (DPBS; Welgene, Gyeongsangbuk-do, Korea) and fluorescence was measured with a microplate reader (SPARK 10M) using a wavelength of excitation and emission 485/535 nm.

Quantitative Real-Time Polymerase Chain Reaction (qRT-PCR)
HDFs were seeded in 48-well plates at 3 × 10 5 cells/well and incubated for 24 h. Then, cell medium was replaced with a serum-free condition and incubated overnight. After 24 h, non-toxic concentrations of sample were pretreated with HDFs for 1 h, followed by treatment with 20 ng/mL TNF-α for 24 h. The RNeasy Mini Kit (Qiagen, Germantown, MD, USA) was used to extract total RNA, and the RevertAid First Strand cDNA Synthesis Kit (Thermo Fisher Scientific, Eugene, OR 97402, USA) was used to reverse transcribing the RNA into cDNA. cDNA was amplified using AccuPower ® 2X GreenStar™ qPCR Master Mix (Bioneer, Daejeon, Korea), the primers in Table 1, and the QuantStudio 3 real-time PCR system (Applied Biosystems, Foster City, CA, USA). qPCR amplification conditions were as follows: 50 • C 2 min; 95 • C for 10 min; followed by 40 cycles of 95 • C for 15 s; 60 • C for 1 min; and 95 • C for 15 s; 60 • C for 1 min; 95 • C for 15 s. The primer sequences were shown Table 1.

Western Blotting
HDFs were seeded in 6-well plates at a density of 3 × 10 5 cells/well and cultured for 24 h. Then, cell medium was replaced with a serum-free condition and incubated overnight. After 24 h, 5, 10 and 50 µM compounds were pretreated with HDFs for 1 h, followed by 20 ng/mL TNF-α for 15 min or 6 h. Protein expression levels of p-ERK, ERK, p-JNK, JNK, p-p38, p-38, p-NF-κB, NF-κB and GAPDH by treatment with TNF-α for 15 min were determined, and COX-2 and GAPDH were detected by treatment for 6 h. Before being lysed with 1× radioimmunoprecipitation assay (RIPA) buffer (Tech & Innovation, Gangwon, Korea), cells were washed with DPBS. After centrifugation at 13,000 rpm, 4 • C, the supernatant was used to detect the protein concentration through the BCA Protein Assay Kit (Thermo scientific, Waltham, MA, USA). The same amount of protein was separated using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to a polyvinylidene difluoride (PVDF) membrane. Membrane was blocked with 5% skim milk in TBS-Tween20 (TBS-T; Thermo Fisher Scientific). The primary and secondary antibodies were diluted in 1% BSA solution and each reacted at 4 • C overnight and at room temperature for 2 h. To measure the protein bands, SuperSignal ® West Femto Maximum Sensitivity Chemiluminescent Substrate (Thermo Fisher Scientific) and Fusion Solo Chemiluminescence System (PEQLAB Biotechnologie GmbH, Erlangen, Germany) were used and quantified using Image J program (TotalLab, Newcastle, UK).

Statistical Analyses
Experimental results were analyzed with GraphPad Prism version 8.0.0 (GraphPad Software Inc., La Jolla, CA, USA) statistical program and expressed as mean ± standard deviation of the mean (SEM). The statistical significance of each group was evaluated by Tukey's test at the level of p < 0.05 after analysis by one-way ANOVA.

Effects of the Hot Water Extract on the Viability and MMP-1 Secretion of HDFs
Prior to analyze the anti-skin aging effect of the hot water extract from the fruits of M. alba, we investigated the non-toxic concentration of extract on HDFs. As shown in Figure 1A, the extract was not toxic at 12.5~100 µg/mL. Subsequently, we investigated whether the extract could prevent skin aging in TNF-a-stimulated HDFs. The level of MMP-1 secretion to the extract was evaluated at concentrations of 50 µg/mL or less. In Figure 1B, TNF-α treatment group significantly increased the secretion of MMP-1 by 1.70 ± 0.07-folds (p < 0.01) compared to the non-treatment group. The extract from the fruits of M. alba exhibited the inhibitory effects on the increase in MMP-1 induced by TNF-α.
Kit (Thermo scientific, Waltham, MA, USA). The same amount of protein was separated using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to a polyvinylidene difluoride (PVDF) membrane. Membrane was blocked with 5% skim milk in TBS-Tween20 (TBS-T; Thermo Fisher Scientific). The primary and secondary antibodies were diluted in 1% BSA solution and each reacted at 4 °C overnight and at room temperature for 2 h. To measure the protein bands, SuperSignal ® West Femto Maximum Sensitivity Chemiluminescent Substrate (Thermo Fisher Scientific) and Fusion Solo Chemiluminescence System (PEQLAB Biotechnologie GmbH, Erlangen, Germany) were used and quantified using Image J program (TotalLab, Newcastle, UK).

Statistical Analyses
Experimental results were analyzed with GraphPad Prism version 8.0.0 (GraphPad Software Inc., La Jolla, CA, USA) statistical program and expressed as mean ± standard deviation of the mean (SEM). The statistical significance of each group was evaluated by Tukey's test at the level of p < 0.05 after analysis by one-way ANOVA.

Effects of the Hot Water Extract on the Viability and MMP-1 Secretion of HDFs
Prior to analyze the anti-skin aging effect of the hot water extract from the fruits of M. alba, we investigated the non-toxic concentration of extract on HDFs. As shown in Figure 1A, the extract was not toxic at 12.5~100 μg/mL. Subsequently, we investigated whether the extract could prevent skin aging in TNF-a-stimulated HDFs. The level of MMP-1 secretion to the extract was evaluated at concentrations of 50 μg/mL or less. In Figure 1B, TNF-α treatment group significantly increased the secretion of MMP-1 by 1.70 ± 0.07-folds (p < 0.01) compared to the non-treatment group. The extract from the fruits of M. alba exhibited the inhibitory effects on the increase in MMP-1 induced by TNF-α. The cells were seeded on 96-well plate with the density of 1 × 10 4 cells/well and incubated for 24h. Next, the cells were treated with indicated concentrations of sample for 24 h. The Ez-Cytox kit was used to assess the viability of the cells. (B) The cells were seeded on 48-well plate with the density of 2 × 10 4 cells/well and starved with non-serum media for 24 h. Next, before being exposed to 20 ng/mL TNF-α for 24 h, the cells were first given the relevant sample concentrations to use for 1 h. The MMP-1 secretion in supernatants were determined using ELISA kit. The data were described as mean ± SEM. # p < 0.05 and ## p < 0.01 non-treatment group versus TNF-α treatment group. * p < 0.05 and *** p < 0.001 extract treatment group versus TNF-α treatment group.

Stucture Elucidation of Compound 1 and Identification of the Isolates
In the present study, a new phenolic glycoside (1) and 21 known compounds  were isolated from the fruits of M. alba ( Figure 2). The cells were seeded on 96-well plate with the density of 1 × 10 4 cells/well and incubated for 24 h. Next, the cells were treated with indicated concentrations of sample for 24 h. The Ez-Cytox kit was used to assess the viability of the cells. (B) The cells were seeded on 48-well plate with the density of 2 × 10 4 cells/well and starved with non-serum media for 24 h. Next, before being exposed to 20 ng/mL TNF-α for 24 h, the cells were first given the relevant sample concentrations to use for 1 h. The MMP-1 secretion in supernatants were determined using ELISA kit. The data were described as mean ± SEM. # p < 0.05 and ## p < 0.01 non-treatment group versus TNF-α treatment group. * p < 0.05 and *** p < 0.001 extract treatment group versus TNF-α treatment group.

Effects of Compounds Isolated from the Fruits of M. alba on the Viability of HDFs
) correlations of compound 1.

Effects of Compounds Isolated from the Fruits of M. alba on MMP-1 Secretion in TNF-α Induced HDFs
Next, we investigated whether the isolates (1-22) can prevent skin aging in TNF-αstimulated HDFs. Base on the cell viability results, the level of MMP-1 secretion to the compounds was screened at concentrations of 50 µM or less.

Effects of Compounds 1 on COLIA1 Protein Expression and ROS Production in TNF-α Induced HDFs
MMP-1 is a collagenase that plays an important role in the degradation of collagen, and MMP-1 inhibitors can improve collagen reduction caused by external factors such as UV. Therefore, we measured the levels of the procollagen COLIA1 to investigate the effect of compound 1 on collagen reduction. In Figure 6A, TNF-α treatment significantly decreased the protein secretion of COLIA1 compared to the control group (0.12 ± 0.00-folds, p < 0.001). The secretion of COLIA1 was significantly increased at the 50 µM of compound 1 (0.50 ± 0.03-folds, p < 0.01) compared to the TNF-α-treatment group. Antioxidants 2022, 11, x FOR PEER REVIEW 10 of 18 HDFs. The cells were seeded on 48-well plate with the density of 2 × 10 4 cells/well and starved with non-serum media for 24 h. Next, before being exposed to 20 ng/mL TNF-α for 24 h, the cells were first given the relevant sample concentrations to use for 1 h. The MMP-1 secretion in supernatants were determined using ELISA kit. The data were described as mean ± SEM. # p < 0.05, ## p < 0.01 and ### p < 0.001 non-treatment group versus TNF-α treatment group. * p < 0.05, ** p < 0.01 and *** p < 0.001 compound treatment group versus TNF-α treatment group. The cells were seeded on 48-well plate with the density of 2 × 10 4 cells/well and starved with non-serum media for 24 h. Next, before being exposed to 20 ng/mL TNF-α for 24 h, the cells were first given the relevant sample concentrations to use for 1 h. The COLIA1 secretion in supernatants were determined using ELISA kit. (B) The cells were seeded on 96-well plate with the density of 1 × 10 4 cells/well and starved with non-serum media for 24 h. Next, the cells were treated with indicated concentration of samples for 1 h before exposure to 20 ng/mL TNF-α and 10 µM DCFDA. MMP-1 and COLIA1 mRNA was assessed using qRT-PCR analysis. The data were described as mean ± SEM. ### p < 0.001 non-treatment group versus TNF-α treatment group. ** p < 0.01 and *** p < 0.001 sample treatment group versus TNF-α treatment group.
Excessive UV and TNF-α cause oxidative stress such as ROS overproduction, promotes synthesis of MMPs and leads to collagen degradation. Because compound 1 reversed the reduction of collagen by TNF-α, we investigated the inhibitory effect of ROS by compound 1. In Figure 6B, the TNF-α-treatment group significantly increased ROS production by 1.52 ± 0.00-folds (p < 0.001) compared to the control group. Compound 1 showed a significant reduction in a concentration-dependent manner (10 µM: 1.27 ± 0.03-folds, p < 0.01; 50 µM: 1.17 ± 0.03-folds, p < 0.001).

Effects of Compound 1 on Phosphorylation of MAPKs in TNF-α Induced HDFs
Next, the effects of compound 1 on phosphorylation of MAPKs in TNF-α-induced HDFs were investigated. The expression of MAPKs was determined using Western blotting. In Figure 7, the ERK phosphorylation of ERK was increased by 2.02 ± 0.04-folds (p < 0.01) in the TNF-α treatment group compared to the control group and was inhibited by 1.27 ± 0.21-folds at 50 µM in the compound 1 treatment group. The phosphorylation of p38 was increased by 6.38 ± 0.06-folds (p < 0.001) in the TNF-α treatment group compared to the control group and was inhibited by 5.05 ± 0.10-folds at 50 µM in the compound 1 treatment group. The phosphorylation of JNK was increased by 2.60 ± 0.26-folds in the TNF-α treatment group compared to the control group but was not inhibited at 50 µM in the compound 1 treatment group. The TNF-α treatment group induced phosphorylation of ERK, JNK and p38 compared to the control group, and the compound 1 treatment group showed a tendency to inhibit phosphorylation of ERK and p38 at a concentration of 50 µM. TNF-α treatment group compared to the control group but was not inhibited at 50 μM in the compound 1 treatment group. The TNF-α treatment group induced phosphorylation of ERK, JNK and p38 compared to the control group, and the compound 1 treatment group showed a tendency to inhibit phosphorylation of ERK and p38 at a concentration of 50 μM. Figure 7. The effects of compound 1 on phosphorylation of MAPKs in TNF-α induced HDFs. The cells were seeded on 6-well plate with the density of 3 × 10 5 cells/well and starved with non-serum media for 24h. Next, before being exposed to 20 ng/mL TNF-α for 15 min, the cells were treated with 5, 10 and 50 M of compound 1 for 1 h. Expression of p-ERK, ERK, p-p38, p38, p-JNK, JNK and GAPDH were determined using Western blotting. The data were described as mean ± SEM. # p < 0.05, ## p < 0.01 and ### p < 0.001 non-treatment group versus TNF-α treatment group. * p < 0.05 sample treatment group versus TNF-α treatment group.

Effects of Compound 1 on NF-κB and COX-2 in TNF-α Treated HDFs
Next, the effects of compound 1 on NF-κB and COX-2 in TNF-α-treated HDFs were investigated. The expression of NF-κB and COX-2 was measured using Western blotting. In Figure 8, the phosphorylation of NF-κB was increased by 2.32 ± 0.114-folds in the TNFα treatment group compared to the control group and was inhibited by 1.50 ± 0.22-folds at 50 μM in the compound 1 treated roup. The expression of COX-2 was increased by 2.94 ± 0.10-folds (p < 0.05) in the TNF-α treatment group compared to the control group and was inhibited by 2.25 ± 0.08-folds (p < 0.05) at 50 μM in the compound 1 treatment group. The cells were seeded on 6-well plate with the density of 3 × 10 5 cells/well and starved with non-serum media for 24 h. Next, before being exposed to 20 ng/mL TNF-α for 15 min, the cells were treated with 5, 10 and 50 M of compound 1 for 1 h. Expression of p-ERK, ERK, p-p38, p38, p-JNK, JNK and GAPDH were determined using Western blotting. The data were described as mean ± SEM. # p < 0.05, ## p < 0.01 and ### p < 0.001 non-treatment group versus TNF-α treatment group. * p < 0.05 sample treatment group versus TNF-α treatment group.

Effects of Compound 1 on NF-κB and COX-2 in TNF-α Treated HDFs
Next, the effects of compound 1 on NF-κB and COX-2 in TNF-α-treated HDFs were investigated. The expression of NF-κB and COX-2 was measured using Western blotting. In Figure 8, the phosphorylation of NF-κB was increased by 2.32 ± 0.114-folds in the TNF-α treatment group compared to the control group and was inhibited by 1.50 ± 0.22-folds at 50 µM in the compound 1 treated roup. The expression of COX-2 was increased by 2.94 ± 0.10-folds (p < 0.05) in the TNF-α treatment group compared to the control group and was inhibited by 2.25 ± 0.08-folds (p < 0.05) at 50 µM in the compound 1 treatment group. Figure 8. The effects of compound 1 on NF-κB and COX-2 in TNF-α-treated HDFs. The cells were seeded on 6-well plate with the density of 3 × 10 5 cells/well and starved with non-serum media for 24h. Next, before being exposed to 20 ng/mL TNF-α for 15 min and 6 h, the cells were treated with 5, 10 and 50 M of compound 1 for 1 h. Expression of NF-κB, COX-2 and GAPDH were determined using Western blotting. The data were described as mean ± SEM. # p < 0.05 and ## p < 0.01 and nontreatment group versus TNF-α treatment group. * p < 0.05 sample treatment group versus TNF-α treatment group.

Discussion
The skin is an organ that is greatly affected by external environmental factors such as ultraviolet rays, stress, and chemicals, and aging can be visually observed [8]. The skin is composed of the epidermis and the dermis, which communicate in various ways to establish, maintain, or restore tissue homeostasis. Dermal tensile strength and elasticity are defined as the properties of the extracellular matrix (ECM) with type I and type III collagen fibrils, microfibrils and elastic fibers [37].
Aging includes intrinsic aging and extrinsic aging, and UV, one of the various causes of extrinsic aging, causes photoaging of the skin. Photoaging induces ROS and pro-inflammatory cytokines such as TNF-α in skin cells and promotes the production of collagen-degrading enzymes, MMPs [38]. MMPs degrade ECM proteins including type 1 collagen, elastin, and fibronectin [39]. ROS and TNF-α activate the phosphorylation of ERK, p38 and JNK and activate the phosphorylation of two subunits of activator protein-1 (AP-1), c-Fos and c-Jun. It also activates and induces phosphorylation and translocation of nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) [40,41].
The role of natural products as therapeutic agents has been recognized since ancient times and contributed greatly to various therapeutics, including anti-inflammatory, anticancer and anti-diabetic [42]. Among the natural compounds, flavonoids and phenolic compounds have been reported in many reports to contribute to antioxidant and antiinflammatory activities [43][44][45][46][47]. The fruits of M. alba contain a variety of secondary metabolites such as flavonoids, lignans, coumarins, phenolic compounds and other compounds [16]. Among these, the constituents such as catechol, rutin and quercetin have been reported to have antioxidant effects [48,49]. Therefore, the ameliorating effect of the Figure 8. The effects of compound 1 on NF-κB and COX-2 in TNF-α-treated HDFs. The cells were seeded on 6-well plate with the density of 3 × 10 5 cells/well and starved with non-serum media for 24 h. Next, before being exposed to 20 ng/mL TNF-α for 15 min and 6 h, the cells were treated with 5, 10 and 50 M of compound 1 for 1 h. Expression of NF-κB, COX-2 and GAPDH were determined using Western blotting. The data were described as mean ± SEM. # p < 0.05 and ## p < 0.01 and non-treatment group versus TNF-α treatment group. * p < 0.05 sample treatment group versus TNF-α treatment group.

Discussion
The skin is an organ that is greatly affected by external environmental factors such as ultraviolet rays, stress, and chemicals, and aging can be visually observed [8]. The skin is composed of the epidermis and the dermis, which communicate in various ways to establish, maintain, or restore tissue homeostasis. Dermal tensile strength and elasticity are defined as the properties of the extracellular matrix (ECM) with type I and type III collagen fibrils, microfibrils and elastic fibers [37].
Aging includes intrinsic aging and extrinsic aging, and UV, one of the various causes of extrinsic aging, causes photoaging of the skin. Photoaging induces ROS and proinflammatory cytokines such as TNF-α in skin cells and promotes the production of collagen-degrading enzymes, MMPs [38]. MMPs degrade ECM proteins including type 1 collagen, elastin, and fibronectin [39]. ROS and TNF-α activate the phosphorylation of ERK, p38 and JNK and activate the phosphorylation of two subunits of activator protein-1 (AP-1), c-Fos and c-Jun. It also activates and induces phosphorylation and translocation of nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) [40,41].
The role of natural products as therapeutic agents has been recognized since ancient times and contributed greatly to various therapeutics, including anti-inflammatory, anti-cancer and anti-diabetic [42]. Among the natural compounds, flavonoids and phenolic compounds have been reported in many reports to contribute to antioxidant and anti-inflammatory activities [43][44][45][46][47]. The fruits of M. alba contain a variety of secondary metabolites such as flavonoids, lignans, coumarins, phenolic compounds and other compounds [16]. Among these, the constituents such as catechol, rutin and quercetin have been reported to have antioxidant effects [48,49]. Therefore, the ameliorating effect of the fruits of M. alba on inflammation-related skin aging and other skin diseases was investigated in this study. The hot water extract showed an inhibitory effect on the se-cretion of MMP-1 on TNF-α-stimulated HDFs. Therefore, it was expected that the hot water extract from the fruits of M. alba would contain active constituents. A new phenolic glycoside (oddioside A, 1) and 21 known compounds (2−22) were isolated from the hot water extract from the fruits of M. alba by repeated chromatography in the present work. To the best of our knowledge, the presence of sargentodoside E (2), eugenyl glucoside (6), 2-O-β-D-glucopyranosyl-4,6-dihydroxybenzaldehyde (7), (7S,8R)-erythro-7,9,9'trihydroxy-3,3'-dimethoxy-8-O-4'-neolignan-4-O-β-D-glucopyranoside (11), pinoresinol-4-O-β-D-glucopyranoside (12), taxifolin-7-O-β-D-glucopyranoside (20) and pinellic acid (21) in M. alba is reported for the first time in this study.
Type 1 collagen is mainly distributed in the skin, and type 3 collagen is distributed throughout the body, and the expression level decreases as aging progresses [67,68]. Procollagen, a precursor of collagen, plays an important role in maintaining the elasticity and support structure of the skin by promoting the synthesis of collagen protein [40]. For this reason, the expression level of collagen can be predicted through the investigation of procollagen type 1. Therefore, the effect of compound 1 treatment on the amount of type 1 collagen that affects skin wrinkles was investigated. In Figure 6A, it was measured that the secretion of COLIA1 decreased due to TNF-α was restored with 50 µM of compound 1.
ROS is one of the well-known causes of inflammation in intrinsic and extrinsic aging, and the inflammatory response accelerates skin aging [69,70]. In addition, ROS promotes the synthesis of MMPs, leading to the breakdown of collagen. One of the MMPs enzymes, MMP-1, is consequently It causes anger, which leads to the formation of wrinkles, loss of elasticity, and sagging [71]. Therefore, the use of antioxidants, one of the many effects of natural products and medicinal plants in the cosmetic industry, is one of the main approaches in skin care product development [72,73]. In this paper, ROS generation was investigated to verify the antioxidant effect of compound 1, and as a result, compound 1 showed an inhibitory effect on TNF-α-induced ROS generation ( Figure 6B).
Many previous studies have revealed that the level of MMP plays an important role in regulating skin damage by NF-κB, AP-1 and MAPKs signaling pathways [74][75][76]. MAPKs signaling pathways include ERK, JNK and p38, and phosphorylation of MAPKs influences phosphorylation of NF-κB and AP-1. In this paper, phosphorylation of ERK, JNK and p38 was induced in the TNF-α-treatment group compared to the control group, and ERK and p38 phosphorylation was inhibited in the 50 µM of compound 1 treatment group (Figure 7). NF-κB is a protein complex that plays an important role in the immune response and can be activated by UV, ROS and pro-inflammatory cytokines [10,77]. In addition, activated NF-κB induces collagen degradation by promoting MMP expression [78]. Next, COX-2 converts arachidonic acid into prostaglandins such as prostaglandin E 2 (PGE 2 ) and is a pro-inflammatory mediator that induces skin aging and inflammatory skin diseases [79,80]. In the present work, it was investigated whether compound 1 had an anti-inflammatory effect. The results showed that phosphorylation of NF-κB and COX-2 was induced in the TNF-α treatment group compared to the control group, and phosphorylation of NF-κB and COX-2 decreased in the group treated with 50 µM compound 1 (Figure 8).

Conclusions
In this study, the effect of TNF-α induced photoaging in HDFs was investigated using 22 isolated compounds from the fruits of M. alba, which is known to have antioxidant effects. Among them, oddioside A (1), a new phenolic glycoside, inhibited the secretion of MMP-1 and increased the secretion of type 1 procollagen (COLIA1) compared to the TNF-α treatment group. Compound 1 showed a significant decrease when ROS produced during skin aging and skin damage was induced from HDFs to TNF-α. In addition, compound 1 showed a tendency to inhibit in TNF-α induced HDFs phosphorylation of ERK and p38. Moreover, compound 1 inhibited the phosphorylation of NF-κB and COX-2, which are important for immune and inflammatory responses. Although further experiments are needed to understand the mechanism on skin aging, the fruit of M. alba and compound 1 could be used as a natural material to prevent photoaging of the skin.