3,5,6,7,8,3′,4′-Heptamethoxyflavone, a Citrus Flavonoid, Inhibits Collagenase Activity and Induces Type I Procollagen Synthesis in HDFn Cells

Citrus fruits contain various types of flavonoids with powerful anti-aging and photoprotective effects on the skin, and have thus been attracting attention as potential, efficacious skincare agents. Here, we aimed to investigate the chemical composition of Citrus unshiu and its protective effects on photoaging. We isolated and identified a bioactive compound, 3,5,6,7,8,3′,4′-heptamethoxyflavone (HMF), from C. unshiu peels using ethanol extraction and hexane fractionation. HMF inhibited collagenase activity and increased type I procollagen content in UV-induced human dermal fibroblast neonatal (HDFn) cells. HMF also suppressed the expression of matrix metalloproteinases 1 (MMP-1) and induced the expression of type I procollagen protein in UV-induced HDFn cells. Additionally, HMF inhibited ultraviolet B (UVB)-induced phosphorylation of the mitogen-activated protein kinases (MAPK) cascade signaling components—ERK, JNK, and c-Jun—which are involved in the induction of MMP-1 expression. Furthermore, HMF affected the TGF-β/Smad signaling pathway, which is involved in the regulation of type I procollagen expression. In particular, HMF induced Smad3 protein expression and suppressed Smad7 protein expression in UV-induced HDFn cells in a dose-dependent manner. These findings suggest a role for Citrus unshiu in the preparation of skincare products in future.

As described above, flavonoids are powerful agents with antioxidant and anti-skin-aging effects, especially against photoaging. However, some flavonoids have also been found to be mutagenic in vitro through pro-oxidant, rather than antioxidant, actions of these compounds. Therefore, it is necessary to investigate their toxic effects before use in humans [30]. In this study, we isolated and identified 3,5,6,7,8,3 ,4 -heptamethoxyflavone (HMF, methylated flavonoid) from the Citrus unshiu peel. HMF inhibited MMP-1 expression and collagen degradation through suppression of the MAPK signaling pathway in UVB-induced human dermal fibroblasts (HDFn) at noncytotoxic concentrations. These findings suggest that HMF may be beneficial in preventing UVB-induced oxygen free radical generation and is a useful agent for preventing skin damage and photoaging.

Chemical Structure and Cytotoxicity of HMF on HDFn Cells
Flavonoids are major bioactive compounds in Citrus fruits. There are six classes of flavonoids: anthocyanins, flavanols (or catechins), flavanones, flavones, flavonols, and isoflavones [31]. Initially, a 50% ethanol extract of dried C. unshiu peel was suspended in water and partitioned successively with hexane. The bioactive compound was identified as HMF by comparing its spectroscopic nuclear magnetic resonance (NMR) data with those previously reported in the literature (Figures 1A, S1 and S2, see also Materials and Methods Section 3.2) [32][33][34]. Values are means ± standard deviation (SD) of three independent experiments and relative to percentages of control cells. Statistical significance of differences was evaluated using one-way analysis of variance (ANOVA) followed by Tukey's test. *** p < 0.001 versus HDFn cells without HMF treatment.
Although they have been widely reported to have beneficial effects on health, several studies suggest that flavonoids also have mutagenic effects due to their pro-oxidant activities [30,35,36]. Moreover, high intakes of flavonoids may affect the activity of key enzymes in hormone metabolism due to their diverse properties [30]. For example, myricetin, quercetagetin, and quercetin caused respiratory bursts in mitochondria and underwent autoxidation resulting in ROS (hydrogen peroxide, hydroxyl radical, and superoxide) formation [37]. It has also been reported that daidzein, genistein, kaempferol, naringenin, and quercetin inhibit thyroxine synthesis and thyroid peroxidase (TPO), which play important roles in thyroid hormone synthesis [38]. In addition, Matsuo et al. [39] indicated that flavonoids can exert toxic effects on human lung embryonic fibroblasts (TIG-1) and human umbilical vein endothelial cells (HUVECs) at relatively high concentrations, even though they showed beneficial effects at relatively low concentrations. For this reason, we examined the cytotoxic effects of HMF on HDFn cells treated with the indicated concentrations (50,100,200, and 400 μg/mL) for 24 h. As shown in Figure 1B, HMF showed no cytotoxic effects on HDFn cells up to 200 μg/mL. However, cell viability was significantly affected by HMF at 400 μg/mL. This concentration resulted in approximately 50% HDFn cell death by when compared with untreated cells. Therefore, HDFn cells were treated with HMF at concentrations of 50, 100, and 200 μg/mL in subsequent experiments.

Effects of HMF on Collagenase Activity and Type I Procollagen Contents of HDFn Cells
Type I collagen, the most abundant fibrillar type collagen, is the main structural component of the extracellular matrix (ECM). Type I procollagen, a soluble precursor of type I collagen, is synthesized and secreted from fibroblasts then proteolytically cleaved by procollagen N-and Cproteases to form collagen fibrils [40]. Matrix metalloproteases (MMPs)-homogeneous zincdependent endopeptidases-can degrade all components of the ECM [9]. Among the MMP family, including gelatinases, matrilysins, membrane-type MMPs, stromelysins, and nonclassified MMPs, collagenases are capable of degrading native collagen fibrils in the ECM. Thus, collagenase is a key enzyme in the degradation of collagen in normal connective tissue [41]. For this reason, natural compounds that inhibit collagenase activity and consequently prevent collagen degradation may be potential candidates for preventing skin photoaging and photodamage. Flavonoids are known to have powerful anti-aging and photoprotective effects on the skin. Over 60 types of flavonoids including anthocyanins, flavanones, flavans, flavones, and flavonols have been identified from Citrus fruits [31]. A direct inhibitory effect of flavonoids on collagenase has rarely been demonstrated, despite the importance of collagenase and collagen breakdown in photoaging and photodamage. It Values are means ± standard deviation (SD) of three independent experiments and relative to percentages of control cells. Statistical significance of differences was evaluated using one-way analysis of variance (ANOVA) followed by Tukey's test. *** p < 0.001 versus HDFn cells without HMF treatment.
Although they have been widely reported to have beneficial effects on health, several studies suggest that flavonoids also have mutagenic effects due to their pro-oxidant activities [30,35,36]. Moreover, high intakes of flavonoids may affect the activity of key enzymes in hormone metabolism due to their diverse properties [30]. For example, myricetin, quercetagetin, and quercetin caused respiratory bursts in mitochondria and underwent autoxidation resulting in ROS (hydrogen peroxide, hydroxyl radical, and superoxide) formation [37]. It has also been reported that daidzein, genistein, kaempferol, naringenin, and quercetin inhibit thyroxine synthesis and thyroid peroxidase (TPO), which play important roles in thyroid hormone synthesis [38]. In addition, Matsuo et al. [39] indicated that flavonoids can exert toxic effects on human lung embryonic fibroblasts (TIG-1) and human umbilical vein endothelial cells (HUVECs) at relatively high concentrations, even though they showed beneficial effects at relatively low concentrations. For this reason, we examined the cytotoxic effects of HMF on HDFn cells treated with the indicated concentrations (50, 100, 200, and 400 µg/mL) for 24 h. As shown in Figure 1B, HMF showed no cytotoxic effects on HDFn cells up to 200 µg/mL. However, cell viability was significantly affected by HMF at 400 µg/mL. This concentration resulted in approximately 50% HDFn cell death by when compared with untreated cells. Therefore, HDFn cells were treated with HMF at concentrations of 50, 100, and 200 µg/mL in subsequent experiments.

Effects of HMF on Collagenase Activity and Type I Procollagen Contents of HDFn Cells
Type I collagen, the most abundant fibrillar type collagen, is the main structural component of the extracellular matrix (ECM). Type I procollagen, a soluble precursor of type I collagen, is synthesized and secreted from fibroblasts then proteolytically cleaved by procollagen N-and C-proteases to form collagen fibrils [40]. Matrix metalloproteases (MMPs)-homogeneous zinc-dependent endopeptidases-can degrade all components of the ECM [9]. Among the MMP family, including gelatinases, matrilysins, membrane-type MMPs, stromelysins, and nonclassified MMPs, collagenases are capable of degrading native collagen fibrils in the ECM. Thus, collagenase is a key enzyme in the degradation of collagen in normal connective tissue [41]. For this reason, natural compounds that inhibit collagenase activity and consequently prevent collagen degradation may be potential candidates for preventing skin photoaging and photodamage. Flavonoids are known to have powerful anti-aging and photoprotective effects on the skin. Over 60 types of flavonoids including anthocyanins, flavanones, flavans, flavones, and flavonols have been identified from Citrus fruits [31]. A direct inhibitory effect of flavonoids on collagenase has rarely been demonstrated, despite the importance of collagenase and collagen breakdown in photoaging and photodamage. It has only been described that epicatechin gallate (ECG) and epigallocatechin gallate (EGCG), isolated from tea and Citrus flavonoids including hesperidin, hesperetin, and naringenin, inhibit collagenase [42,43]. Moreover, Citrus peel and seeds are very rich in flavonoids, particularly the peel [44]. We therefore investigated the inhibitory activity of HMF, isolated from C. unshiu peel, on cellular collagenase activity in HDFn cells.
Comparing the collagenase inhibitory activities of EGCG and HMF, treatment with HMF showed inhibitory activity approximately two times higher than that of EGCG at the same concentrations ( Figure 2A). In addition, we examined the effects of HMF on collagen degradation, which was shown to decrease in UVB-induced HDFn cells after treatment with HMF ( Figure 2B). Treatment with HMF significantly increased the type I procollagen content of HDFn cells in a dose-dependent manner. The amount of the type I procollagen in UVB-irradiated HDFn cells was increased to 0.46 ± 0.01, 0.56 ± 0.04, and 0.61 ± 0.01 µg/mL by pretreatment with HMF at 50, 100, and 200 µg/mL concentrations, respectively ( Figure 2B). These findings indicate that HMF significantly inhibited collagenase activity and consequently prevented collagen degradation in UVB-induced HDFn cells by 87% (at 200 µg/mL of HMF) suggesting that HMF may protect against skin photoaging and photodamage ( Figure 2B). has only been described that epicatechin gallate (ECG) and epigallocatechin gallate (EGCG), isolated from tea and Citrus flavonoids including hesperidin, hesperetin, and naringenin, inhibit collagenase [42,43]. Moreover, Citrus peel and seeds are very rich in flavonoids, particularly the peel [44]. We therefore investigated the inhibitory activity of HMF, isolated from C. unshiu peel, on cellular collagenase activity in HDFn cells.
Comparing the collagenase inhibitory activities of EGCG and HMF, treatment with HMF showed inhibitory activity approximately two times higher than that of EGCG at the same concentrations ( Figure 2A). In addition, we examined the effects of HMF on collagen degradation, which was shown to decrease in UVB-induced HDFn cells after treatment with HMF ( Figure 2B). Treatment with HMF significantly increased the type I procollagen content of HDFn cells in a dosedependent manner. The amount of the type I procollagen in UVB-irradiated HDFn cells was increased to 0.46 ± 0.01, 0.56 ± 0.04, and 0.61 ± 0.01 μg/mL by pretreatment with HMF at 50, 100, and 200 μg/mL concentrations, respectively ( Figure 2B). These findings indicate that HMF significantly inhibited collagenase activity and consequently prevented collagen degradation in UVB-induced HDFn cells by 87% (at 200 μg/mL of HMF) suggesting that HMF may protect against skin photoaging and photodamage ( Figure 2B). Values are means ± standard deviation (SD) of three independent experiments. Statistical significance of differences was evaluated using one-way analysis of variance (ANOVA) followed by Tukey's test. *** p < 0.001 versus HDFn cells without HMF treatment.

Effects of HMF on Cellular MMP-1 and Type I Procollagen Expression in HDFn Cells
Among the MMPs, MMP-1 (collagenase-1) was the first identified enzyme primarily responsible for degradation of the extracellular matrix (ECM) in UV-irradiated dermal tissues [6,9,10]. Active compounds possessing the ability to inhibit the collagen degradation enzymes (such as MMP-1) and promote collagen synthesis have possible uses in cosmetic and therapeutic agents to prevent photoaging and photodamage. It was previously reported that several bioactive compounds such as apigenin, luteolin, nobiletin, quercetin, and wogonin from Citrus inhibit MMP-1 expression in UVinduced human keratinocytes or human dermal fibroblasts [17,[25][26][27][28]. Therefore, in the present study, the effects of HMF on MMP-1 and type I procollagen expression in UV-induced HDFn cells were also examined.
Western blot analysis showed that HMF inhibited the expression of MMP-1 in UV-induced HDFn cells (Figure 3). The induction of this protein by UVB (20 mJ/cm 2 ) was significantly suppressed by pretreatment with HMF at a concentration of 200 μg/mL ( Figure 3B). Our result also showed that the expression of type I procollagen was markedly increased by pretreatment with HMF in the UVinduced HDFn cells in a dose-dependent manner ( Figure 3C). The relative expression of type I procollagen after pretreatment with 200 μg/mL of HMF was about 5.5 times higher than that of Values are means ± standard deviation (SD) of three independent experiments. Statistical significance of differences was evaluated using one-way analysis of variance (ANOVA) followed by Tukey's test. *** p < 0.001 versus HDFn cells without HMF treatment.

Effects of HMF on Cellular MMP-1 and Type I Procollagen Expression in HDFn Cells
Among the MMPs, MMP-1 (collagenase-1) was the first identified enzyme primarily responsible for degradation of the extracellular matrix (ECM) in UV-irradiated dermal tissues [6,9,10]. Active compounds possessing the ability to inhibit the collagen degradation enzymes (such as MMP-1) and promote collagen synthesis have possible uses in cosmetic and therapeutic agents to prevent photoaging and photodamage. It was previously reported that several bioactive compounds such as apigenin, luteolin, nobiletin, quercetin, and wogonin from Citrus inhibit MMP-1 expression in UV-induced human keratinocytes or human dermal fibroblasts [17,[25][26][27][28]. Therefore, in the present study, the effects of HMF on MMP-1 and type I procollagen expression in UV-induced HDFn cells were also examined.
Western blot analysis showed that HMF inhibited the expression of MMP-1 in UV-induced HDFn cells (Figure 3). The induction of this protein by UVB (20 mJ/cm 2 ) was significantly suppressed by pretreatment with HMF at a concentration of 200 µg/mL ( Figure 3B). Our result also showed that the expression of type I procollagen was markedly increased by pretreatment with HMF in the UV-induced HDFn cells in a dose-dependent manner ( Figure 3C). The relative expression of type I procollagen after pretreatment with 200 µg/mL of HMF was about 5.5 times higher than that of nontreated UV-induced HDFn cells ( Figure 3C). These findings indicate that HMF significantly downregulates MMP-1 expression, whereas it upregulates type I procollagen expression in UV-induced HDFn.
nontreated UV-induced HDFn cells ( Figure 3C). These findings indicate that HMF significantly downregulates MMP-1 expression, whereas it upregulates type I procollagen expression in UVinduced HDFn. Values are means ± standard deviation (SD) of three independent experiments. Statistical significance of differences was evaluated using oneway analysis of variance (ANOVA) followed by Tukey's test. *** p < 0.001 versus HDFn cells without HMF treatment.

Effects of HMF on MAPK Signaling Pathways in HDFn Cells
Degradation of the extracellular matrix (ECM) by UV irradiation has been well studied in human skin tissue [3][4][5]. UV irradiation produces reactive oxygen species (ROS), which converge to stimulate the activator protein-1 (AP-1) through mitogen-activated protein kinases (MAPKs) such as extracellular signal-regulated kinase (ERK), p38 kinase, and c-Jun N-terminal kinase (JNK) [3][4][5]. In addition, the degree of c-Fos expression and c-Jun phosphorylation affect the activity of AP-1, which triggers the production of matrix metalloproteinases (MMPs) [6,7]. It was also previously reported that Citrus flavonoids such as apigenin, luteolin, nobiletin, and quercetin suppress the phosphorylation of the MAPK pathway components in UV-induced human keratinocytes or human dermal fibroblasts [17,[25][26][27][28]. Activation of ERK and JNK have been reported to phosphorylate c-Jun, suggesting cross-talk between ERK and JNK in the regulation of c-Jun activity [6,45,46]. It has been reported that p38 contributes to AP-1 activity by activating transcription factors such as transcription factor 2 (ATF2), Ets-like transcription factor 1 (Elk1), and SRF accessory protein 1 (SAP1), which upregulate c-Jun and c-Fos promoter activity [47]. In addition, AP-1 activity is dependent on the abundance of c-Jun and c-Fos as well as their degree of phosphorylation [6,7].
In this study, we evaluated the phosphorylation levels of MAP kinases including ERK, p38, JNK, and c-Jun in UV-induced HDFn cells pretreated with HMF. As shown in Figure 4, the phosphorylation of ERK and JNK proteins were significantly decreased by treatment with HMF at all concentrations. The phosphorylation of c-Jun protein was also significantly decreased by treatment with HMF at concentrations of 100 and 200 μg/mL. Our results also show that neither phosphorylation nor expression of the p38 protein is affected by HMF at all concentrations ( Figure 4). However, the expression of c-Fos was significantly inhibited by treatment with HMF at concentrations of 50, 100, and 200 μg/mL. These results indicate that the inhibition of MMP-1 expression by HMF is associated with MAPK family proteins including ERK, JNK, and c-Jun. In particular, HMF induced the phosphorylation of ERK, JNK, and c-Jun. Although the phosphorylation of the p38 protein was not inhibited, HMF suppressed the expression of the c-Fos protein.

Effects of HMF on MAPK Signaling Pathways in HDFn Cells
Degradation of the extracellular matrix (ECM) by UV irradiation has been well studied in human skin tissue [3][4][5]. UV irradiation produces reactive oxygen species (ROS), which converge to stimulate the activator protein-1 (AP-1) through mitogen-activated protein kinases (MAPKs) such as extracellular signal-regulated kinase (ERK), p38 kinase, and c-Jun N-terminal kinase (JNK) [3][4][5]. In addition, the degree of c-Fos expression and c-Jun phosphorylation affect the activity of AP-1, which triggers the production of matrix metalloproteinases (MMPs) [6,7]. It was also previously reported that Citrus flavonoids such as apigenin, luteolin, nobiletin, and quercetin suppress the phosphorylation of the MAPK pathway components in UV-induced human keratinocytes or human dermal fibroblasts [17,[25][26][27][28]. Activation of ERK and JNK have been reported to phosphorylate c-Jun, suggesting cross-talk between ERK and JNK in the regulation of c-Jun activity [6,45,46]. It has been reported that p38 contributes to AP-1 activity by activating transcription factors such as transcription factor 2 (ATF2), Ets-like transcription factor 1 (Elk1), and SRF accessory protein 1 (SAP1), which upregulate c-Jun and c-Fos promoter activity [47]. In addition, AP-1 activity is dependent on the abundance of c-Jun and c-Fos as well as their degree of phosphorylation [6,7].
In this study, we evaluated the phosphorylation levels of MAP kinases including ERK, p38, JNK, and c-Jun in UV-induced HDFn cells pretreated with HMF. As shown in Figure 4, the phosphorylation of ERK and JNK proteins were significantly decreased by treatment with HMF at all concentrations. The phosphorylation of c-Jun protein was also significantly decreased by treatment with HMF at concentrations of 100 and 200 µg/mL. Our results also show that neither phosphorylation nor expression of the p38 protein is affected by HMF at all concentrations ( Figure 4). However, the expression of c-Fos was significantly inhibited by treatment with HMF at concentrations of 50, 100, and 200 µg/mL. These results indicate that the inhibition of MMP-1 expression by HMF is associated with MAPK family proteins including ERK, JNK, and c-Jun. In particular, HMF induced the phosphorylation of ERK, JNK, and c-Jun. Although the phosphorylation of the p38 protein was not inhibited, HMF suppressed the expression of the c-Fos protein. Values are means ± standard deviation (SD) of three independent experiments. Statistical significance of differences was evaluated using one-way analysis of variance (ANOVA) followed by Tukey's test. ** p < 0.01 and *** p < 0.001 versus HDFn cells without HMF treatment.

Effects of HMF on Smad3 and Smad7 Expression in HDFn Cells
UV irradiation influences not only the degradation of the extracellular matrix (ECM) through MAP kinase signaling but also the inhibition of collagen synthesis by the TGF-β/Smad pathway [3][4][5]40]. Once TGF-β binds to the cell surface receptor (TβRI and TβII), the TGF-β receptor complex initiates its cellular action by transiently interacting with the receptor-regulated Smad proteins resulting in their phosphorylation. Phosphorylated Smad2/Smad3 combine with Smad4 and translocate into the nucleus to regulate transcription of TGF-β-regulated genes including type I procollagen [48]. Decreased expression of the type I procollagen gene was observed in skin fibroblasts, which interfered with TGF-β, Smad3, or Smad4 expression [49][50][51][52][53]. Smad7 interacts with TβRI to prevent phosphorylation and activation of Smad2 and Smad3 which interrupt TGF-β signaling [15,16]. Therefore, we hypothesized that HMFinduced type I procollagen upregulation is regulated by the TGF-β/Smad signal transduction cascade. As expected, the expression of Smad3 protein was dose-dependently increased in the UV-induced HDFn cells by pretreatment with HMF ( Figure 5). In addition, Smad7 protein expression was significantly decreased by pretreatment with HMF at concentrations of 100 and 200 μg/mL ( Figure 5C). Although the TGF-β expression level was not determined in the present study, these results suggest that HMF contributed to the induction of type I procollagen expression through the TGF-β/Smad pathway.  Values are means ± standard deviation (SD) of three independent experiments. Statistical significance of differences was evaluated using one-way analysis of variance (ANOVA) followed by Tukey's test. ** p < 0.01 and *** p < 0.001 versus HDFn cells without HMF treatment.

Effects of HMF on Smad3 and Smad7 Expression in HDFn Cells
UV irradiation influences not only the degradation of the extracellular matrix (ECM) through MAP kinase signaling but also the inhibition of collagen synthesis by the TGF-β/Smad pathway [3][4][5]40]. Once TGF-β binds to the cell surface receptor (TβRI and TβII), the TGF-β receptor complex initiates its cellular action by transiently interacting with the receptor-regulated Smad proteins resulting in their phosphorylation. Phosphorylated Smad2/Smad3 combine with Smad4 and translocate into the nucleus to regulate transcription of TGF-β-regulated genes including type I procollagen [48]. Decreased expression of the type I procollagen gene was observed in skin fibroblasts, which interfered with TGF-β, Smad3, or Smad4 expression [49][50][51][52][53]. Smad7 interacts with TβRI to prevent phosphorylation and activation of Smad2 and Smad3 which interrupt TGF-β signaling [15,16]. Therefore, we hypothesized that HMF-induced type I procollagen upregulation is regulated by the TGF-β/Smad signal transduction cascade. As expected, the expression of Smad3 protein was dose-dependently increased in the UV-induced HDFn cells by pretreatment with HMF ( Figure 5). In addition, Smad7 protein expression was significantly decreased by pretreatment with HMF at concentrations of 100 and 200 µg/mL ( Figure 5C). Although the TGF-β expression level was not determined in the present study, these results suggest that HMF contributed to the induction of type I procollagen expression through the TGF-β/Smad pathway.  Values are means ± standard deviation (SD) of three independent experiments. Statistical significance of differences was evaluated using one-way analysis of variance (ANOVA) followed by Tukey's test. ** p < 0.01 and *** p < 0.001 versus HDFn cells without HMF treatment.

Effects of HMF on Smad3 and Smad7 Expression in HDFn Cells
UV irradiation influences not only the degradation of the extracellular matrix (ECM) through MAP kinase signaling but also the inhibition of collagen synthesis by the TGF-β/Smad pathway [3][4][5]40]. Once TGF-β binds to the cell surface receptor (TβRI and TβII), the TGF-β receptor complex initiates its cellular action by transiently interacting with the receptor-regulated Smad proteins resulting in their phosphorylation. Phosphorylated Smad2/Smad3 combine with Smad4 and translocate into the nucleus to regulate transcription of TGF-β-regulated genes including type I procollagen [48]. Decreased expression of the type I procollagen gene was observed in skin fibroblasts, which interfered with TGF-β, Smad3, or Smad4 expression [49][50][51][52][53]. Smad7 interacts with TβRI to prevent phosphorylation and activation of Smad2 and Smad3 which interrupt TGF-β signaling [15,16]. Therefore, we hypothesized that HMFinduced type I procollagen upregulation is regulated by the TGF-β/Smad signal transduction cascade. As expected, the expression of Smad3 protein was dose-dependently increased in the UV-induced HDFn cells by pretreatment with HMF ( Figure 5). In addition, Smad7 protein expression was significantly decreased by pretreatment with HMF at concentrations of 100 and 200 μg/mL ( Figure 5C). Although the TGF-β expression level was not determined in the present study, these results suggest that HMF contributed to the induction of type I procollagen expression through the TGF-β/Smad pathway.  Values are means ± standard deviation (SD) of three independent experiments. Statistical significance of differences was evaluated using one-way analysis of variance (ANOVA) followed by Tukey's test. ** p < 0.01 and *** p < 0.001 versus HDFn cells without HMF treatment. 5% CO 2 incubator. After incubation, the cells (~90% confluency) were harvested and washed using phosphate-buffered saline (PBS, pH 7.4). The cell pellets were suspended in 0.1 M Tris-HCl (pH 7.4) containing 0.1% Triton X-100 and 1 mM phenylmethylsulfonyl fluoride (PMSF), then disrupted by sonication for 5 min. Each tube contained HMF (100 and 200 µg/mL), 125 µL of 0.1 M Tris-HCl (pH 7.5) including 4 mM CaCl 2 , 4-phenylazo-benzyloxy-carbonyl-Pro-Leu-Gly-Pro-D-Arg (Pz-peptide), and 75 µL of the cell-free supernatant. The tube was incubated at 37 • C for 20 min, and the reaction was stopped by addition of 250 µL stop solution (6% citric acid). The reaction tube was mixed with 750 µL ethyl acetate and the absorbance of the supernatant (200 µL) was measured at 320 nm using a microplate reader. Collagenase activity was measured according to the following formula: collagenase inhibitory activity (%) = {1 − (OD 320 of sample/OD 320 of control)} × 100.

Type I Procollagen Assay
Type I procollagen content was determined by Procollagen Type I C-Peptide EIA Kit (Takara, Seoul, Korea). HDFn cells were seeded in 6-well plates at 7 × 10 5 cells/well and allowed to attach for 24 h. HDFn cells were incubated with various concentrations of the HMF (50, 100, and 200 µg/mL) for 24 h in FBS-free media. Cells washed with phosphate-buffered saline (PBS, pH 7.4) were irradiated with UVB (20 mJ/cm 2 , Vilber Lourmat, Marne La Vallée, France) and then further incubated for 24 h at 37 • C in a humidified 5% CO 2 incubator. After centrifugation at 10,000× g for 5 min, the 1/5 diluted cell-free supernatant was mixed with 100 µL antibody-horseradish peroxidase (POD) conjugate solution and then incubated for 3 h at 37 • C in an antibody-coated 96-well microtiter plate (Procollagen Type I C-Peptide EIA Kit, Takara). The cells were rinsed four times with phosphate-buffered saline (PBS, pH 7.4) and incubated for 15 min in the dark in substrate solution (H 2 O 2 and tetramethylbenzidine) (Procollagen Type I C-Peptide EIA Kit, Takara). The reaction was stopped with 100 µL H 2 SO 4 (1 N) and absorbance was measured at 450 nm. A standard was prepared using the serum concentration of procollagen type I C-peptide and the type I procollagen content was expressed as µg/mL.

Statistical Analysis
All data are presented as mean ± standard deviation (SD) of three independent experiments. Statistical significance between groups was evaluated using a one-way analysis of variance (ANOVA) followed by Tukey's test using Prism (GraphPad Software Inc., La Jolla, CA, USA).

Conclusions
The results of the present study suggest that HMF, a flavonoid compound isolated from C. unshiu peel, effectively protected against UV-induced photoaging of HDFn cells. We found that HMF inhibited collagenase activity and increased the type I procollagen content in UV-induced HDFn cells. It also suppressed MMP-1 and induced type I procollagen expression. Moreover, HMF affected the MAPK signaling pathway, which may contribute to the reduction in MMP-1 expression. UV irradiation influenced not only the degradation of the extracellular matrix (ECM) through MAP kinase signaling but also the inhibition of collagen synthesis through the TGF-β/Smad pathway. In the UV-induced HDFn cells, HMF affected the TGF-β/Smad signaling pathway. HMF induced Smad3 expression and suppressed Smad7 expression. This is the first report to show that HMF has a photoprotective effect in UV-induced HDFn cells through the phosphorylation of MAPK signals, such as ERK, JNK, and c-Jun, followed by the suppression of MMP-1 expression. Although further studies are required to elucidate an anti-photoaging effect of HMF in HDFn cells irradiated with different light sources including solar and UVA, this bioactive compound warrants further investigation for development as a potential agent in cosmetic products.