Advertisement

Oral Supplementation with Cocoa Extract Reduces UVB-Induced Wrinkles in Hairless Mouse Skin

  • Author Footnotes
    11 These authors contributed equally to this work.
    Jong-Eun Kim
    Footnotes
    11 These authors contributed equally to this work.
    Affiliations
    WCU Biomodulation Major, Center for Food and Bioconvergence, Department of Agricultural Biotechnology, Seoul National University, Seoul, Republic of Korea

    Advanced Institutes of Convergence Technology, Seoul National University, Suwon, Republic of Korea
    Search for articles by this author
  • Author Footnotes
    11 These authors contributed equally to this work.
    Dasom Song
    Footnotes
    11 These authors contributed equally to this work.
    Affiliations
    WCU Biomodulation Major, Center for Food and Bioconvergence, Department of Agricultural Biotechnology, Seoul National University, Seoul, Republic of Korea

    Advanced Institutes of Convergence Technology, Seoul National University, Suwon, Republic of Korea
    Search for articles by this author
  • Author Footnotes
    11 These authors contributed equally to this work.
    Junil Kim
    Footnotes
    11 These authors contributed equally to this work.
    Affiliations
    CHA Cancer Institute, CHA University, Seongnam, Republic of Korea
    Search for articles by this author
  • Jina Choi
    Affiliations
    CHA Cancer Institute, CHA University, Seongnam, Republic of Korea

    Department of Biochemistry, School of Medicine, CHA University, Seongnam, Republic of Korea
    Search for articles by this author
  • Jong Rhan Kim
    Affiliations
    WCU Biomodulation Major, Center for Food and Bioconvergence, Department of Agricultural Biotechnology, Seoul National University, Seoul, Republic of Korea

    Advanced Institutes of Convergence Technology, Seoul National University, Suwon, Republic of Korea
    Search for articles by this author
  • Hyun-Sun Yoon
    Affiliations
    Department of Dermatology, Seoul National University College of Medicine, Seoul, Republic of Korea

    Institute of Human-Environment Interface Biology, Seoul National University, Seoul, Republic of Korea

    Department of Dermatology, Seoul National University Boramae Hospital, Seoul, Republic of Korea
    Search for articles by this author
  • Jung-Soo Bae
    Affiliations
    Department of Dermatology, Seoul National University College of Medicine, Seoul, Republic of Korea

    Institute of Human-Environment Interface Biology, Seoul National University, Seoul, Republic of Korea
    Search for articles by this author
  • Mira Han
    Affiliations
    Department of Dermatology, Seoul National University College of Medicine, Seoul, Republic of Korea

    Institute of Human-Environment Interface Biology, Seoul National University, Seoul, Republic of Korea
    Search for articles by this author
  • Sein Lee
    Affiliations
    WCU Biomodulation Major, Center for Food and Bioconvergence, Department of Agricultural Biotechnology, Seoul National University, Seoul, Republic of Korea

    Advanced Institutes of Convergence Technology, Seoul National University, Suwon, Republic of Korea
    Search for articles by this author
  • Ji Sun Hong
    Affiliations
    WCU Biomodulation Major, Center for Food and Bioconvergence, Department of Agricultural Biotechnology, Seoul National University, Seoul, Republic of Korea

    Advanced Institutes of Convergence Technology, Seoul National University, Suwon, Republic of Korea
    Search for articles by this author
  • Dayoung Song
    Affiliations
    CHA Cancer Institute, CHA University, Seongnam, Republic of Korea

    Department of Biochemistry, School of Medicine, CHA University, Seongnam, Republic of Korea
    Search for articles by this author
  • Seong-Jin Kim
    Affiliations
    CHA Cancer Institute, CHA University, Seongnam, Republic of Korea
    Search for articles by this author
  • Myoung-Jin Son
    Affiliations
    Health Foods Team, Lotte R&D Center, Seoul, Republic of Korea
    Search for articles by this author
  • Sang-Woon Choi
    Affiliations
    CHAUM Life Center, School of Medicine, CHA University, Seoul, Republic of Korea
    Search for articles by this author
  • Jin Ho Chung
    Affiliations
    Department of Dermatology, Seoul National University College of Medicine, Seoul, Republic of Korea

    Institute of Human-Environment Interface Biology, Seoul National University, Seoul, Republic of Korea

    Institute on Aging, Seoul National University, Seoul, Republic of Korea
    Search for articles by this author
  • Tae-Aug Kim
    Correspondence
    Tae-Aug Kim, CHA Cancer Institute, Department of Biochemistry, School of Medicine, CHA University, Seongnam 463-400, Republic of Korea.
    Affiliations
    CHA Cancer Institute, CHA University, Seongnam, Republic of Korea

    Department of Biochemistry, School of Medicine, CHA University, Seongnam, Republic of Korea
    Search for articles by this author
  • Ki Won Lee
    Correspondence
    Correspondence: Ki Won Lee, Department of Agricultural Biotechnology, Seoul National University, Seoul 151-921, Republic of Korea.
    Affiliations
    WCU Biomodulation Major, Center for Food and Bioconvergence, Department of Agricultural Biotechnology, Seoul National University, Seoul, Republic of Korea

    Advanced Institutes of Convergence Technology, Seoul National University, Suwon, Republic of Korea

    Institute on Aging, Seoul National University, Seoul, Republic of Korea
    Search for articles by this author
  • Author Footnotes
    11 These authors contributed equally to this work.
Open ArchivePublished:February 05, 2016DOI:https://doi.org/10.1016/j.jid.2015.11.032
      Cacao beans contain various bioactive phytochemicals that could modify the pathogeneses of certain diseases. Here, we report that oral administration of cacao powder (CP) attenuates UVB-induced skin wrinkling by the regulation of genes involved in dermal matrix production and maintenance. Transcriptome analysis revealed that 788 genes are down- or upregulated in the CP supplemented group, compared with the UVB-irradiated mouse skin controls. Among the differentially expressed genes, cathepsin G and serpin B6c play important roles in UVB-induced skin wrinkle formation. Gene regulatory network analysis also identified several candidate regulators responsible for the protective effects of CP supplementation against UVB-induced skin damage. CP also elicited antiwrinkle effects via inhibition of UVB-induced matrix metalloproteinases-1 expression in both the human skin equivalent model and human dermal fibroblasts. Inhibition of UVB-induced activator protein-1 via CP supplementation is likely to affect the expression of matrix metalloproteinases-1. CP supplementation also downregulates the expression of cathepsin G in human dermal fibroblasts. 5-(3′,4′-Dihydroxyphenyl)-γ-valerolactone, a major in vivo metabolite of CP, showed effects similar to CP supplementation. These results suggest that cacao extract may offer a protective effect against photoaging by inhibiting the breakdown of dermal matrix, which leads to an overall reduction in wrinkle formation.

      Abbreviations:

      AP-1 (activator protein), CP (cacao powder), DEGs (differentially expressed genes), DHPV (5-(3′,4′-dihydroxyphenyl)-γ-valerolactone), ECM (extracellular matrix), GRN (Gene Regulatory Network), HDF (human dermal fibroblasts), HSE (human skin equivalent), MEdD (minimal edema dose), MMP (matrix metalloproteinases), Nrf2 (NF-E2-related factor 2), TFs (transcription factors)

      Introduction

      Many natural products are known to influence the development of skin structures and its biological functions. Cacao beans have the antioxidant capacity higher than the capacity provided by green teas and red wine (
      • Lee K.W.
      • Kim Y.J.
      • Lee H.J.
      • Lee C.Y.
      Cocoa has more phenolic phytochemicals and a higher antioxidant capacity than teas and red wine.
      ,
      • Subhashini R.
      • Mahadeva Rao U.S.
      • Sumathi P.
      • Gunalan G.A.
      Comparative phytochemical analysis of cocoa and green tea.
      ). The antioxidant activity of cacao can modify the pathogeneses of a different spectrum of diseases, including the cardiovascular diseases, cancer, and other chronic conditions (
      • Park H.S.
      • Jin S.P.
      • Lee Y.
      • Oh I.G.
      • Lee S.
      • Kim J.H.
      • et al.
      Toll-like receptor 2 mediates a cutaneous reaction induced by repetitive ultraviolet B irradiation in C57/BL6 mice in vivo.
      ). Recent studies have demonstrated the beneficial effects of cacao consumption are associated with human health, especially with the improved condition of the skin (
      • Park H.S.
      • Jin S.P.
      • Lee Y.
      • Oh I.G.
      • Lee S.
      • Kim J.H.
      • et al.
      Toll-like receptor 2 mediates a cutaneous reaction induced by repetitive ultraviolet B irradiation in C57/BL6 mice in vivo.
      ,
      • Scapagnini G.
      • Davinelli S.
      • Di Renzo L.
      • De Lorenzo A.
      • Olarte H.H.
      • Micali G.
      • et al.
      Cocoa bioactive compounds: significance and potential for the maintenance of skin health.
      ). Cacao provides positive effects on the skin structure and the dermal microcirculation (
      • Katz D.L.
      • Doughty K.
      • Ali A.
      Cocoa and chocolate in human health and disease.
      ,
      • Neukam K.
      • Stahl W.
      • Tronnier H.
      • Sies H.
      • Heinrich U.
      Consumption of flavanol-rich cocoa acutely increases microcirculation in human skin.
      ), and its topical preparations are able to protect the skin from the oxidative damages arising from UV radiation (
      • Katz D.L.
      • Doughty K.
      • Ali A.
      Cocoa and chocolate in human health and disease.
      ). Besides the photoprotection against UVB-induced erythema, long-term ingestion of cacao also ameliorates aberrant skin conditions by increasing the blood flow to the cutaneous and subcutaneous tissues to increase the skin density and hydration (
      • Heinrich U.
      • Neukam K.
      • Tronnier H.
      • Sies H.
      • Stahl W.
      Long-term ingestion of high flavanol cocoa provides photoprotection against UV-induced erythema and improves skin condition in women.
      ). Our previous studies have shown that cacao inhibits skin cancer growth and skin inflammation both in vitro and in vivo (
      • Kang N.J.
      • Lee K.W.
      • Lee D.E.
      • Rogozin E.A.
      • Bode A.M.
      • Lee H.J.
      • et al.
      Cocoa procyanidins suppress transformation by inhibiting mitogen-activated protein kinase kinase.
      ,
      • Kim J.E.
      • Son J.E.
      • Jung S.K.
      • Kang N.J.
      • Lee C.Y.
      • Lee K.W.
      • et al.
      Cocoa polyphenols suppress TNF-alpha-induced vascular endothelial growth factor expression by inhibiting phosphoinositide 3-kinase (PI3K) and mitogen-activated protein kinase kinase-1 (MEK1) activities in mouse epidermal cells.
      ,
      • Lee K.W.
      • Kundu J.K.
      • Kim S.O.
      • Chun K.S.
      • Lee H.J.
      • Surh Y.J.
      Cocoa polyphenols inhibit phorbol ester-induced superoxide anion formation in cultured HL-60 cells and expression of cyclooxygenase-2 and activation of NF-kappaB and MAPKs in mouse skin in vivo.
      ). Although there is accumulating evidence that cacao consumption can improve the skin health, the molecular mechanisms responsible for these beneficial effects have not been thoroughly investigated.
      Skin wrinkling is a typical characteristic of photoaging that results from chronic exposure to solar UV radiation. Repeated exposure to UV light decreases procollagen production and breaks down collagen fibers. The process is partially due to the overexpression of matrix metalloproteinases (MMP) (
      • Fisher G.J.
      • Wang Z.Q.
      • Datta S.C.
      • Varani J.
      • Kang S.
      • Voorhees J.J.
      Pathophysiology of premature skin aging induced by ultraviolet light.
      ,
      • Ichihashi M.
      • Ando H.
      • Yoshida M.
      • Niki Y.
      • Matsui M.
      Photoaging of the skin.
      ,
      • Xu Y.
      • Fisher G.J.
      Ultraviolet (UV) light irradiation induced signal transduction in skin photoaging.
      ). Our previous studies demonstrated that cathepsin G regulates MMP expression and UVB-induced skin photoaging (
      • Son E.D.
      • Kim H.
      • Choi H.
      • Lee S.H.
      • Lee J.Y.
      • Kim S.
      • et al.
      Cathepsin G increases MMP expression in normal human fibroblasts through fibronectin fragmentation, and induces the conversion of proMMP-1 to active MMP-1.
      ,
      • Son E.D.
      • Shim J.H.
      • Choi H.
      • Kim H.
      • Lim K.M.
      • Chung J.H.
      • et al.
      Cathepsin G inhibitor prevents ultraviolet B-induced photoaging in hairless mice via inhibition of fibronectin fragmentation.
      ). Cathepsins comprise a family of serine proteases whose members are classified into A, B, C, D, E, G, H, and L groups, according to their substrate specificities (
      • Son E.D.
      • Kim H.
      • Choi H.
      • Lee S.H.
      • Lee J.Y.
      • Kim S.
      • et al.
      Cathepsin G increases MMP expression in normal human fibroblasts through fibronectin fragmentation, and induces the conversion of proMMP-1 to active MMP-1.
      ). Cathepsins B, D, K, and G may act as biomarkers in photoaged human skin (
      • Zheng Y.
      • Lai W.
      • Wan M.
      • Maibach H.I.
      Expression of cathepsins in human skin photoaging.
      ). Cathepsin G is a single 30-kDa polypeptide released by the neutrophils and the UVA-irradiated normal human fibroblasts (
      • Son E.D.
      • Kim H.
      • Choi H.
      • Lee S.H.
      • Lee J.Y.
      • Kim S.
      • et al.
      Cathepsin G increases MMP expression in normal human fibroblasts through fibronectin fragmentation, and induces the conversion of proMMP-1 to active MMP-1.
      ). Inhibitors of cathepsin G may be useful for the prevention of UVB-induced photoaging because they could ameliorate the extracellular matrix (ECM) damage and MMP upregulation (
      • Son E.D.
      • Shim J.H.
      • Choi H.
      • Kim H.
      • Lim K.M.
      • Chung J.H.
      • et al.
      Cathepsin G inhibitor prevents ultraviolet B-induced photoaging in hairless mice via inhibition of fibronectin fragmentation.
      ). Serpin b6 is a member of the superfamily of serine protease inhibitors known as serpins. Serpins bind with serine proteases involved in inflammatory processes, coagulation, fibrinolysis, tumorigenesis, and apoptosis. The association of serpin b6 with cathepsin G has been postulated to inhibit cathepsin G activity (
      • Scott F.L.
      • Sun J.
      • Whisstock J.C.
      • Kato K.
      • Bird P.I.
      SerpinB6 is an inhibitor of kallikrein-8 in keratinocytes.
      ).
      In this study, we first examined the protective effect of CP on UVB-induced wrinkle formation in hairless mice, and then we showed gene expression profiles using RNA sequencing analysis in comparison with several other well-known food materials used to modify skin health (
      • Cho H.S.
      • Lee M.H.
      • Lee J.W.
      • No K.O.
      • Park S.K.
      • Lee H.S.
      • et al.
      Anti-wrinkling effects of the mixture of vitamin C, vitamin E, pycnogenol and evening primrose oil, and molecular mechanisms on hairless mouse skin caused by chronic ultraviolet B irradiation.
      ,
      • Marini A.
      • Grether-Beck S.
      • Jaenicke T.
      • Weber M.
      • Burki C.
      • Formann P.
      • et al.
      Pycnogenol(R) effects on skin elasticity and hydration coincide with increased gene expressions of collagen type I and hyaluronic acid synthase in women.
      ). To better investigate the antiphotoaging effects of CP and the implications for clinical settings, we measured its effects using a human dermal fibroblasts (HDF) and human skin equivalent (HSE) model.

      Results

      Oral administration of CP reduces UVB-induced wrinkle formation and prevents UVB-induced collagen degradation

      To investigate the effect of CP on wrinkle formation, the dorsal skins of hairless mice were exposed to UVB with low and high concentrations of CP (CL, 39.1 mg/kg, CH, 156.3 mg/kg) and pycnogenol (Pyc, 625 mg/kg) for 8 weeks as described (Figure 1a). UVB-induced wrinkle formation was markedly reduced in the CP-administered groups (Figure 1b). Quantification of skin wrinkle severity through the assessment of the area of wrinkling (Figure 1c) and visual wrinkle grade (Figure 1d) confirmed a significant decrease in wrinkle formation in the CP groups. We then stained the skin samples of the mice with Masson’s trichrome staining to observe the effect of CP on amorphous collagens of the skin (Figure 1e). Collagen levels gradually recovered in the CP groups to an extent greater than the UVB-irradiated group (Figure 1e, Supplementary Figure S1 online). The physical aesthetics of the CP groups were similar or superior to those of the pycnogenol-treated group (Figure 1c–e). Taken together, these results suggest that oral administration of CP reduces UVB-induced wrinkle formation and prevents UVB-induced collagen degradation.
      Figure 1
      Figure 1Effect of cacao powder (CP) on UVB-induced wrinkle formation. (a) A schematic diagram of the animal experiment. (b) Back skins of hairless mice (8–10 mice per each group) were exposed to UVB for 8 weeks as scheduled in (a). Bottom figures are replica from back of mice as developed in the Materials and Methods section. (c) Skin impressions were analyzed by Skin-Visiometer software after 8 weeks of treatment. (d) The severity of skin wrinkling was visually graded as described in the Materials and Methods section after 8 weeks of treatment. Data represent the means ± SEM (n = 8–10). Means with letters (a–c) within a graph are significantly different from each other at P < 0.05. (e) Masson’s trichrome staining for collagen fibers. Collagen fibers appear blue. Scale bar = 100 μm. (a) The vehicle was 0.5% sodium carboxymethylcellulose. (b) UVB, (c) UVB+ CL (low concentration of CP), (d) UVB+CH (high concentration of CP), and (e) UVB+Pyc (pycnogenol). (f) Expression profile of differentially expressed genes (DEGs) mediated by CP in mouse skin tissue. Heat map of DEGs mediated by low or high CP concentration. Expression of genes represent the log2 ratio of the indicated group. (g) Heat map of DEGs upregulated by UVB and downregulated by high concentrations of CP. (h) Heat map of DEGs downregulated by UVB and upregulated by high concentrations of CP. Heat maps (g, h) are drawn based on log2 FPKM of each group. Blue color indicates low expression of genes in each group, whereas red color represents high level of gene expression. SEM, standard error of the mean.

      Expression profiling of differentially expressed genes (DEGs) mediated by CP supplementation and/or UVB-irradiation of mouse skin tissue

      To identify genes associated with the UVB-protective effect of CP supplementation in skin, we systematically analyzed the transcriptome from the mice exposed to UVB-irradiation and/or administrated with CP and pycnogenol. The heat map of DEGs in the UVB-irradiated mice indicated that 788 genes were up- or downregulated by at least one concentration of CP supplementation (Figure 1f). Among the 788 DEGs, 156 genes were upregulated by UVB compared with control and downregulated by CH compared with UVB-irradiation (Figure 1g), and 199 genes were downregulated by UVB-IR compared with the controls and upregulated by CH compared with the UVB irradiated group (Figure 1h). Supplementation with CH elicited transcriptomic recovery on the up- and downregulated genes after UVB-IR (Figure 1g and h). Furthermore, CP administration shows more impact on transcriptomic recovery than the recovery induced by pycnogenol (Figure 1g and h), suggesting that CP may be a more potent antiphotoaging agent than pycnogenol.

      Expression patterns of genes associated with antiphotoaging

      To further characterize gene expression patterns, we identified the genes associated with antiphotoaging effects using the related gene ontology terms including extracellular matrix disassembly (Figure 2a), cell adhesion (Figure 2b), lipid metabolic process (Figure 2c), and proteinaceous extracellular matrix (Figure 2d and Supplementary Table S1 online). The gel-like ECM is the largest component of the dermal skin layer and comprises a variety of fibrous structural proteins, including collagens, elastin, laminin, and proteoglycans such as dermatan sulfate and hyaluronan (
      • Bradley E.J.
      • Griffiths C.E.
      • Sherratt M.J.
      • Bell M.
      • Watson R.E.
      Over-the-counter anti-ageing topical agents and their ability to protect and repair photoaged skin.
      ). Differentially expressed genes in ECM disassembly indicated that CP-fed mice had markedly inverted changes in their UVB-mediated transcriptomes (Figure 2a). CP significantly diminished UVB-induced cathepsin G (Ctsg) expression. Interestingly, the effect of CP on these expression patterns was more significant than that of pycnogenol (Figure 3a). Among the various serpin b6 genes, CP supplementation specifically enhanced the expression of serpin b6c (Figure 3b). These findings suggest that both inhibition of cathepsin G and induction of serpin B6c by CP supplementation may contribute to a protective effect against UVB-induced wrinkle formation. To identify potential mediators of the changes in transcriptome expression patterns, we constructed a Gene Regulatory Network (GRN) analysis composed of the DEGs in Figure 3c, and significantly enriched transcription factors (TFs) obtained from the TF-target relationships derived from the Encyclopedia of DNA Element data (
      Consortium EP
      An integrated encyclopedia of DNA elements in the human genome.
      ,
      • Gerstein M.B.
      • Kundaje A.
      • Hariharan M.
      • Landt S.G.
      • Yan K.K.
      • Cheng C.
      • et al.
      Architecture of the human regulatory network derived from ENCODE data.
      ) and SignaLink database (
      • Fazekas D.
      • Koltai M.
      • Turei D.
      • Modos D.
      • Palfy M.
      • Dul Z.
      • et al.
      SignaLink 2—a signaling pathway resource with multi-layered regulatory networks.
      ) (Figure 3d and e). Thus, GRN analysis identifies the mediators involved in the antiphotoaging effects of CP.
      Figure 2
      Figure 2Expression of genes involved in dermal matrix formation, following GO terms. Expression profiles of DEGs were classified by GO terms as follows: (a) Extracellular matrix disassembly, (b) cell adhesion, (c) lipid metabolic process, and (d) proteinaceous extracellular matrix. The bar graph shows the relative log2 mRNA ratio of the indicated targets from RNA sequence data of each group with statistically significant P < 0.05. DEGs, differentially expressed genes; GO, gene ontology.
      Figure 3
      Figure 3CP (cocoa powder) regulates the expression of extracellular matrix (ECM) genes. (a) CP supplementation significantly inhibits expression of cathepsin G in UVB-irradiated skin tissue. (b) The expression of Serpin B6c is markedly enhanced by CP supplementation. The graph is representative of the RNA sequence analyses from five mice. Bars followed by the same letter do not differ significantly (P < 0.05). (c) The largest connected component of CP-mediated skin response Gene Regulatory Network (GRN). (d) The second largest connected component of the GRN in response to UVB and CP administration. GRNs were constructed as described in the Materials and Methods section. (e) Candidates of transcription factors (TFs) that regulate the specific target genes such as cathepsin G, serpin B6c, collagen 25A1 (COL25A1), and fibronectin 1 (FN1). Hexagon nodes with green borders denote TFs. Red and blue nodes denote upregulated and downregulated genes by high concentrations of CP, respectively.
      Figure 3
      Figure 3CP (cocoa powder) regulates the expression of extracellular matrix (ECM) genes. (a) CP supplementation significantly inhibits expression of cathepsin G in UVB-irradiated skin tissue. (b) The expression of Serpin B6c is markedly enhanced by CP supplementation. The graph is representative of the RNA sequence analyses from five mice. Bars followed by the same letter do not differ significantly (P < 0.05). (c) The largest connected component of CP-mediated skin response Gene Regulatory Network (GRN). (d) The second largest connected component of the GRN in response to UVB and CP administration. GRNs were constructed as described in the Materials and Methods section. (e) Candidates of transcription factors (TFs) that regulate the specific target genes such as cathepsin G, serpin B6c, collagen 25A1 (COL25A1), and fibronectin 1 (FN1). Hexagon nodes with green borders denote TFs. Red and blue nodes denote upregulated and downregulated genes by high concentrations of CP, respectively.

      CP prevents UVB-induced MMP-1 upregulation in HDF and in HSE layers

      To better understand whether the antiwrinkling effects of CP in mice could be relevant for clinical settings, we examined the effect of CP on collagenase (MMP-1) in HDF and HSE layers. CP treatment elicited a decrease in MMP-1 protein expression in a concentration-dependent manner and significantly suppressed the mRNA levels of UVB-induced MMP-1 (Figure 4a and b). Furthermore, CP inhibited UVB-induced activator protein (AP-1) transactivation (Figure 4c). These inhibitory effects arose within a concentration range that did not significantly affect cell viability in the presence of UVB irradiation (Figure 4d). These results suggest that CP may downregulate both UVB-induced MMP-1 protein and gene expression through the suppression of UVB-induced AP-1 transcriptional activity in HDF. To confirm the effects of CP on cathepsin G expression, we measured the expression of cathepsin G in HDF. Cathepsin G expression is decreased by CP treatment in HDF (Figure 4e). Next, to verify whether the antiwrinkle effect of CP in mice and in vitro could be applied to humans, we examined the effect of CP on collagenase (MMP-1) in HSE as described in Supplementary Figure S2 online. Immunohistochemical staining showed that CP markedly inhibited UVB-induced MMP-1 levels with increasing levels of CP (Figure 4f).
      Figure 4
      Figure 4CP inhibits UVB-induced MMP-1 expression in human dermal fibroblast (HDF) and human skin equivalent (HSE). (a) CP inhibits the expression of UVB-induced MMP-1. MMP-2 is used as a loading control. (b) MMP-1 mRNA levels for the CP group are analyzed by real-time quantitative PCR (RT-qPCR). Data (n = 3) represent the means ± SD. (c) Regulation of AP-1 transcriptional activity by CP. A luciferase reporter gene assay was performed in HDF as described in the Materials and Methods section. (d) Cell viability after CP treatment. Cell viability is measured using the CellTiter 96 AQueous One Solution Cell Proliferation Assay. (a–d) HDF cells are pretreated with CP at the indicated concentrations for 1 hour, and then further treated with 0.02 J/cm2 UVB for 48 hours at 37 °C. Data (n = 4) represent the means ± SD. Means with letters (a–e) within a graph are significantly different from each other at P < 0.05. (e) Cathepsin G expression is inhibited by CP. Treatment of CP and UVB is the same as described in (a)–(d). (f) CP inhibits UVB-induced MMP-1 protein expression in HSE. HSE was developed as described in the schematic diagram of the HSE system ( online). HSE serial sections from the human skin equivalent were mounted onto silane-coated slides and subjected to immunohistochemical staining using the anti-MMP-1 antibody as described in the Materials and Methods section. Scale bar = 50 μm. AP-1, activator protein-1; CP, cocoa powder; MMP, matrix metalloproteinases; SD, standard deviation.

      DHPV significantly decreased UVB-induced MMP-1 protein expression, gene transcription, and AP-1 transactivation in HDF

      To investigate the metabolite effect of CP on UVB-induced wrinkle formation, 5-(3′,4′-dihydroxyphenyl)-γ-valerolactone (DHPV) (Figure 5a), a major metabolite form of CP in the body, was used as shown in the previous study (
      • Urpi-Sarda M.
      • Monagas M.
      • Khan N.
      • Llorach R.
      • Lamuela-Raventos R.M.
      • Jauregui O.
      • et al.
      Targeted metabolic profiling of phenolics in urine and plasma after regular consumption of cocoa by liquid chromatography-tandem mass spectrometry.
      ). The effect of DHPV on UVB-induced MMP-1 protein and gene expression in vitro has been measured (Figure 5b and c). DHPV decreased MMP-1 protein expression (Figure 5b) and significantly suppressed UVB-induced MMP-1 mRNA level (Figure 5c), compared with those of the UVB-irradiated cells. We also examined the effect of DHPV on AP-1 transcriptional activity induced by UVB irradiation, showing that DHPV suppressed UVB-induced AP-1 transactivation in HDF (Figure 5d). DHPV inhibits capthesin G expression similar to that of CP (Figure 5e). The concentrations of DHPV used in this experiment were not toxic to proliferation of HDF in the presence of UVB irradiation (Figure 5f). These results indicated that CP metabolite DHPV may act as a driver to inhibit UVB-induced wrinkle formation by suppressing MMP-1 protein expression and gene transcription by inhibiting AP-1 activity in HDF.
      Figure 5
      Figure 5A major metabolite, 5-(3′,4′-dihydroxyphenyl)-γ-valerolactone (DHPV) inhibits UVB-induced MMP-1 protein expression, gene transcription, and AP-1 transactivation in HDF. (a) Chemical structure of DHPV. (b) Expression of MMP-1 was inhibited by DHVP under UVB-IR in HDF. MMP-2 was used as a loading control. (c) Suppression of MMP-1 mRNA expression by DHVP was analyzed by RT-qPCR. Triplicate samples were used and experiments were repeated three times; the mean ± SD. (d) The effect of DHVP on AP-1 transcriptional activity in HDF. AP-1 transactivation ability was measured by a luciferase reporter gene assay. Data (n = 3) represent the mean ± SD. (c, d) Means with letter (a–c) within a graph are significantly different from each other at P < 0.05. (e) Inhibition of UVB-induced cathepsin G expression by DHPV. (f) DHPV did not affect cell viability. Data (n = 8) represent the mean ± SD. Means with letter (a) within a graph are no significantly different from each other at P < 0.05. (b–f) Conditions of treatment of DHPV and UVB in HDF are the same as described in . AP-1, activator protein-1; MMP, matrix metalloproteinases; RT-qPCR, real-time quantitative PCR; SD, standard deviation.

      Discussion

      UVB is the major etiological factor of skin photoaging and carcinogenesis. In our previous studies, we measured the minimal UVB dose on the dorsal skin of the mice as the minimal edema dose (MEdD), setting MEdD as 100 mJ/cm2. According to study of Bernerd et al., the daily dose of UV on earth is 100–200 J/cm2 and the average ratio of UVA/UVB is 27.3 (
      • Marionnet C.
      • Tricaud C.
      • Bernerd F.
      Exposure to non-extreme solar UV daylight: spectral characterization, effects on skin and photoprotection.
      ). Based on the calculation of physiological UVB dose, 200 mJ/cm2 of UVB was used in this study: Average UVB dose of about 2 hours activity in outside (37.75 J/cm2 [UV dose of New York]/28.3[UVB/UV]/14[day time] = 95 mJ/h cm2).
      CP-supplemented diets have been suggested to elicit many beneficial effects, particularly for skin health (
      • Park H.S.
      • Jin S.P.
      • Lee Y.
      • Oh I.G.
      • Lee S.
      • Kim J.H.
      • et al.
      Toll-like receptor 2 mediates a cutaneous reaction induced by repetitive ultraviolet B irradiation in C57/BL6 mice in vivo.
      ). In the present study, we investigated the antiwrinkle effects of CP in vivo. Oral administration of CP reduced UVB-induced wrinkle formation and prevented UVB-induced collagen degradation in hairless mice. We also used a human equivalent skin model and primary human skin fibroblasts, and found that CP inhibits UVB-induced MMP-1 expression in both models. In addition, our clinical study shows that CP (4 g/day for 24 weeks) significantly reduces wrinkle formation without side effects (data not shown). To investigate the antiwrinkling mechanisms of CP, we performed an RNA SEQ array. A total of 788 genes were found to be up- or downregulated by CP treatment in UVB-irradiated skin tissues. Such significant changes in transcriptome may imply the existence of signature molecules to regulate UV-induced skin aging. Of particular note, cathepsin G was significantly inhibited whereas serpin b6c was upregulated in the presence of CP. Cathepsin G is known to induce fibronectin fragmentation (
      • Son E.D.
      • Kim H.
      • Choi H.
      • Lee S.H.
      • Lee J.Y.
      • Kim S.
      • et al.
      Cathepsin G increases MMP expression in normal human fibroblasts through fibronectin fragmentation, and induces the conversion of proMMP-1 to active MMP-1.
      ). It has previously been reported that serpin b6 is a potent inhibitor of cathepsin G (
      • Scott F.L.
      • Hirst C.E.
      • Sun J.
      • Bird C.H.
      • Bottomley S.P.
      • Bird P.I.
      The intracellular serpin proteinase inhibitor 6 is expressed in monocytes and granulocytes and is a potent inhibitor of the azurophilic granule protease, cathepsin G.
      ). Although the detailed molecular relationship between skin wrinkling and cathepsin G activity has not been clearly elucidated, cathepsin G activity has in the past been linked to skin wrinkling (
      • Son E.D.
      • Kim H.
      • Choi H.
      • Lee S.H.
      • Lee J.Y.
      • Kim S.
      • et al.
      Cathepsin G increases MMP expression in normal human fibroblasts through fibronectin fragmentation, and induces the conversion of proMMP-1 to active MMP-1.
      ), and is known to regulate MMP-1 mRNA expression (
      • Son E.D.
      • Shim J.H.
      • Choi H.
      • Kim H.
      • Lim K.M.
      • Chung J.H.
      • et al.
      Cathepsin G inhibitor prevents ultraviolet B-induced photoaging in hairless mice via inhibition of fibronectin fragmentation.
      ). Our in vivo study shows that CP supplementation inhibits UVB-induced skin wrinkling concurrent with the inhibition of cathepsin G and upregulation of serpin b6c. Cathepsin G expression was similarly inhibited by CP treatment in HDFs.
      Intriguingly, we found that the CP-mediated skin response GRN included critical regulators of photoaging in skin such as NFE2L2 (
      • Kawachi Y.
      • Xu X.
      • Taguchi S.
      • Sakurai H.
      • Nakamura Y.
      • Ishii Y.
      • et al.
      Attenuation of UVB-induced sunburn reaction and oxidative DNA damage with no alterations in UVB-induced skin carcinogenesis in Nrf2 gene-deficient mice.
      ,
      • Tian F.F.
      • Zhang F.F.
      • Lai X.D.
      • Wang L.J.
      • Yang L.
      • Wang X.
      • et al.
      Nrf2-mediated protection against UVA radiation in human skin keratinocytes.
      ), peroxisome proliferator-activated receptor γ and TP53 (
      • El-Domyati M.
      • El-Ammawi T.S.
      • Medhat W.
      • Moawad O.
      • Mahoney M.G.
      • Uitto J.
      Expression of p53 protein after nonablative rejuvenation: the other side of the coin.
      ,
      • Lee Y.K.
      • Cha H.J.
      • Hong M.
      • Yoon Y.
      • Lee H.
      • An S.
      Role of NF-kappaB-p53 crosstalk in ultraviolet A-induced cell death and G1 arrest in human dermal fibroblasts.
      ) (Figure 3c). NFE2L2 is known as NF-E2-related factor 2 (Nrf2) and a transcription activator that binds to antioxidant response elements and Maf recognition elements in the promoter regions of target genes. Nrf2 is also important for the coordinated responses to oxidative stress (
      • Itoh K.
      • Mimura J.
      • Yamamoto M.
      Discovery of the negative regulator of Nrf2, Keap1: a historical overview.
      ,
      • Sykiotis G.P.
      • Bohmann D.
      Stress-activated cap'n'collar transcription factors in aging and human disease.
      ). UVA strongly induces Nrf2 expression in human skin fibroblasts but is weakly induced in skin keratinocytes. Knockdown of Nrf2 has been shown to markedly increase cell damage by UVA irradiation in skin keratinocytes, suggesting that Nrf2 may protect human skin keratinocytes from UVA radiation-induced damage (
      • Tian F.F.
      • Zhang F.F.
      • Lai X.D.
      • Wang L.J.
      • Yang L.
      • Wang X.
      • et al.
      Nrf2-mediated protection against UVA radiation in human skin keratinocytes.
      ). Furthermore, UVB-induced sunburn reactions and oxidative DNA damage have been observed to be more prominent in nrf2−/− mice (
      • Kawachi Y.
      • Xu X.
      • Taguchi S.
      • Sakurai H.
      • Nakamura Y.
      • Ishii Y.
      • et al.
      Attenuation of UVB-induced sunburn reaction and oxidative DNA damage with no alterations in UVB-induced skin carcinogenesis in Nrf2 gene-deficient mice.
      ). The photoprotective effect of Nrf2 is closely related to the inhibition of ECM degradation and inflammation (
      • Saw C.L.
      • Yang A.Y.
      • Huang M.T.
      • Liu Y.
      • Lee J.H.
      • Khor T.O.
      • et al.
      Nrf2 null enhances UVB-induced skin inflammation and extracellular matrix damages.
      ). Enhanced Nrf2 activity in keratinocytes has also been associated with epidermal barrier function and antioxidant defense (
      • Schafer M.
      • Farwanah H.
      • Willrodt A.H.
      • Huebner A.J.
      • Sandhoff K.
      • Roop D.
      • et al.
      Nrf2 links epidermal barrier function with antioxidant defense.
      ). Peroxisome proliferator-activated receptor are a family of nuclear hormone receptors and play key roles in lipid metabolism and glucose homeostasis (
      • Kota B.P.
      • Huang T.H.
      • Roufogalis B.D.
      An overview on biological mechanisms of PPARs.
      ,
      • Kuenzli S.
      • Saurat J.H.
      Effect of topical PPARbeta/delta and PPARgamma agonists on plaque psoriasis. A pilot study.
      ,
      • Lalloyer F.
      • Wouters K.
      • Baron M.
      • Caron S.
      • Vallez E.
      • Vanhoutte J.
      • et al.
      Peroxisome proliferator-activated receptor-alpha gene level differently affects lipid metabolism and inflammation in apolipoprotein E2 knock-in mice.
      ,
      • Varga T.
      • Czimmerer Z.
      • Nagy L.
      PPARs are a unique set of fatty acid regulated transcription factors controlling both lipid metabolism and inflammation.
      ). Evidence suggests that peroxisome proliferator-activated receptor α/γ regulated gene responses have an effect on age-related inflammatory and photoaging mediators such as cytokines, MMPs, and AP-1, in NF-κB signaling (
      • Chung J.H.
      • Seo A.Y.
      • Chung S.W.
      • Kim M.K.
      • Leeuwenburgh C.
      • Yu B.P.
      • et al.
      Molecular mechanism of PPAR in the regulation of age-related inflammation.
      ,
      • Kim J.K.
      • Mun S.
      • Kim M.S.
      • Kim M.B.
      • Sa B.K.
      • Hwang J.K.
      5,7-Dimethoxyflavone, an activator of PPARalpha/gamma, inhibits UVB-induced MMP expression in human skin fibroblast cells.
      ,
      • Michalik L.
      • Wahli W.
      Peroxisome proliferator-activated receptors (PPARs) in skin health, repair and disease.
      ). In addition, the tumor suppressor gene p53 plays an important role in protecting cells against DNA damage from sources of extrinsic stress (
      • Nelson W.G.
      • Kastan M.B.
      DNA strand breaks: the DNA template alterations that trigger p53-dependent DNA damage response pathways.
      ). An earlier study has examined the significant impact of UVB on p53 (
      • van Kranen H.J.
      • de Laat A.
      • van de Ven J.
      • Wester P.W.
      • de Vries A.
      • Berg R.J.
      • et al.
      Low incidence of p53 mutations in UVA (365-nm)-induced skin tumors in hairless mice.
      ). Moreover, p53 is known to be activated by DNA damage, oxidative stress, and inflammation (
      • Ak P.
      • Levine A.J.
      p53 and NF-kappaB: different strategies for responding to stress lead to a functional antagonism.
      ,
      • Han E.S.
      • Muller F.L.
      • Perez V.I.
      • Qi W.
      • Liang H.
      • Xi L.
      • et al.
      The in vivo gene expression signature of oxidative stress.
      ,
      • Nelson W.G.
      • Kastan M.B.
      DNA strand breaks: the DNA template alterations that trigger p53-dependent DNA damage response pathways.
      ,
      • Reuter S.
      • Gupta S.C.
      • Chaturvedi M.M.
      • Aggarwal B.B.
      Oxidative stress, inflammation, and cancer: how are they linked?.
      ), and has recently been identified as a UV target gene that associates with V600EBRAF to induce melanoma formation (
      • Viros A.
      • Sanchez-Laorden B.
      • Pedersen M.
      • Furney S.J.
      • Rae J.
      • Hogan K.
      • et al.
      Ultraviolet radiation accelerates BRAF-driven melanomagenesis by targeting TP53.
      ). These genes could be novel candidates responsible for the CP-mediated UVB-protective effects observed. We also constructed a GRN composed of possible candidate TFs regulating cathepsin G and serpin b6c based on the unfiltered TF-target relationships in the Encyclopedia of DNA Element data. This GRN analysis suggested that the expression of these genes may be regulated by other TFs such as SPI1 and MAFK.
      When human beings take in CP orally, it is then metabolized in the body. Namely, CP is changed into its metabolite and interacts with skin cells in the form of metabolite. Various metabolites occur at digestion and absorption by CP consumption. According to the previous study (
      • Urpi-Sarda M.
      • Monagas M.
      • Khan N.
      • Llorach R.
      • Lamuela-Raventos R.M.
      • Jauregui O.
      • et al.
      Targeted metabolic profiling of phenolics in urine and plasma after regular consumption of cocoa by liquid chromatography-tandem mass spectrometry.
      ), DHPV is mainly produced in plasma and appears the biggest variation after regular consumption of CP compared with before consumption of CP. Epicatechin and procyanidins, which are major constituent in CP, are metabolized into DHPV (
      • Urpi-Sarda M.
      • Monagas M.
      • Khan N.
      • Llorach R.
      • Lamuela-Raventos R.M.
      • Jauregui O.
      • et al.
      Targeted metabolic profiling of phenolics in urine and plasma after regular consumption of cocoa by liquid chromatography-tandem mass spectrometry.
      ). We, then, hypothesized that the antiwrinkle effect of CP may be derived from DHPV. As a result, DHPV suppressed UVB-induced MMP-1 protein expression and gene transcription by inhibiting AP-1 activity as the same as the effect of CP on those in HDF. DHPV, therefore, was considered to be an active compound based on these results. Further studies should validate the various metabolites of CP of their effects on skin structure and developments.
      Taken together, our studies indicate that CP supplementation contributes to a reduction in wrinkle formation and collagen degradation. Transcriptomic changes in response to UVB-irradiation in CP-supplemented mice provide evidence for an antiphotoaging effect of CP extract. Oral treatment of CP significantly downregulates cathepsin G while upregulating serpin b6c, which itself is known to inactivate cathepsin G. Therefore, CP supplementation may prevent breakdown of the dermal matrix. For clinical application, we examined the effects of CP and its major metabolites DHPV on photoaging in HDF. These results underline the potential for CP extracts to be further developed as antiphotoaging agents.

      Materials and Methods

      Preparation of CP

      CP was provided by Barry Callebaut (Lebbeke-Wieze, Belgium). Cacao beans were roasted and ground to make cacao liquor, which was separated with cacao butter, to produce cacao cakes. CP was produced by grinding the cacao cakes. The flavanol content in the resultant CP was 71.5 mg/g, which was determined independently by the Korea Health Supplement Institute (Gyeonggi-do, Korea). CP was dissolved in 0.5% sodium carboxymethylcellulose for animal treatment and in 50% ethanol for cell treatment.

      Animals and treatments

      Six-week-old female albino hairless mice (Skh-1) were obtained from Bio Genomics (Seoul, Korea). All experimental protocols were approved by the Institutional Animal Care and Use Committee (Case Number: 14-0008-S1A0) of the Biomedical Research Institute at Seoul National University Hospital. Groups of 8–10 mice were allocated to receive one of six treatment types. CP or a positive control were administered to the mice according to the following treatments groups: CP low (39.1 mg/kg of CP), CP high (156.3 mg/kg of CP), and pycnogenol (positive control, 625 mg/kg of pycnogenol). CP and vehicle (0.5% sodium carboxymethylcellulose) were orally administered for 8 weeks, and body weight and food intake were monitored on a weekly basis. A photoaging experiment was also performed, as described previously (
      • Kim H.H.
      • Lee M.J.
      • Lee S.R.
      • Kim K.H.
      • Cho K.H.
      • Eun H.C.
      • et al.
      Augmentation of UV-induced skin wrinkling by infrared irradiation in hairless mice.
      ). A UVB irradiation device containing TL20W/12RS UV lamps (Philips, Eindhoven, The Netherlands) with an emission spectrum between 275 and 380 nm (peak, 310–315 nm) served as the UV source. A Kodacel filter (TA401/407; Eastman Kodak, Rochester, NY) was mounted 2 cm in front of the UV lamp to remove wavelengths of less than 290 nm (UVC) (
      • Seo J.Y.
      • Lee S.H.
      • Youn C.S.
      • Choi H.R.
      • Rhie G.E.
      • Cho K.H.
      • et al.
      Ultraviolet radiation increases tropoelastin mRNA expression in the epidermis of human skin in vivo.
      ). Initially, we measured the minimal UVB dose on the dorsal skin of the mice as the MEdD comparable with a minimal erythema dose in human skin. In contrast to human skin, mouse skin showed peak responses to UVB primarily as edema, manifesting as an increased thickness of dorsal skin at 48 hours after UVB irradiation (
      • Benavides F.
      • Oberyszyn T.M.
      • VanBuskirk A.M.
      • Reeve V.E.
      • Kusewitt D.F.
      The hairless mouse in skin research.
      ,
      • Learn D.B.
      • Beasley D.G.
      • Giddens L.D.
      • Beard J.
      • Stanfield J.W.
      • Roberts L.K.
      Minimum doses of ultraviolet radiation required to induce murine skin edema and immunosuppression are different and depend on the ultraviolet emission spectrum of the source.
      ). The irradiation doses were increased weekly in increments of 0.5 MEdD (1 MEdD = 100 mJ/cm2) up to 2 MEdD and then maintained at 2 MEdD thereafter. UVB irradiation was stopped after 8 weeks (Figure 1a) (
      • Jin X.J.
      • Kim E.J.
      • Oh I.K.
      • Kim Y.K.
      • Park C.H.
      • Chung J.H.
      Prevention of UV-induced skin damages by 11,14,17-eicosatrienoic acid in hairless mice in vivo.
      ,
      • Kim M.K.
      • Lee S.
      • Kim E.J.
      • Kong K.H.
      • Lee D.H.
      • Chung J.H.
      Topical application of anacardic acid (6-nonadecyl salicylic acid) reduces UV-induced histone modification, MMP-13, MMP-9, COX-2 and TNF-alpha expressions in hairless mice skin.
      ,
      • Park H.S.
      • Jin S.P.
      • Lee Y.
      • Oh I.G.
      • Lee S.
      • Kim J.H.
      • et al.
      Toll-like receptor 2 mediates a cutaneous reaction induced by repetitive ultraviolet B irradiation in C57/BL6 mice in vivo.
      ,
      • Yoon H.S.
      • Shin C.Y.
      • Kim Y.K.
      • Lee S.R.
      • Chung J.H.
      Endogenous estrogen exacerbates UV-induced inflammation and photoaging in mice.
      ).

      RNA sequencing analysis

      For the transcriptome analysis, frozen skin tissue was pulverized in liquid nitrogen, and total RNA was isolated using an RNeasy Plus Mini Kit (Qiagen, Valencia, CA) according to the manufacturer’s instructions. The integrity of the RNA was assessed by 1% agarose gel electrophoresis and visualization of the 18S and 28S RNA species after ethidium bromide staining. RNA sequencing was performed by TheragenEtex. DEGs in CL- and CH-fed mice skin irradiated with UVB were identified by comparing with the normal diet-fed mice irradiated with UVB using both twofold criteria and P < 0.05 value cutoffs obtained from Student’s t-test. We then amalgamated the two DEG groups (in CL- and CH-fed mice) into CP-mediated DEGs.

      GRN analysis

      We constructed the CP-mediated skin response GRN by retrieving a reliable human GRN from the study by
      • Gerstein M.B.
      • Kundaje A.
      • Hariharan M.
      • Landt S.G.
      • Yan K.K.
      • Cheng C.
      • et al.
      Architecture of the human regulatory network derived from ENCODE data.
      and SignaLink Version 2 (
      • Fazekas D.
      • Koltai M.
      • Turei D.
      • Modos D.
      • Palfy M.
      • Dul Z.
      • et al.
      SignaLink 2—a signaling pathway resource with multi-layered regulatory networks.
      ) as follows: the significant TFs for the human homologs of the CP-mediated DEGs were obtained by performing hypergeometric tests between the DEGs and a target gene group for a TF using both the P < 0.05 criteria and the size of the intersection between the DEG group and a target gene group > 1 criteria. We then amalgamated the significant TFs and their target genes into the CP-mediated skin response GRN. The CP-mediated skin response GRN was composed of two isolated connected components (Figure 3c and d).

      Cell culture and treatments

      Primary human dermal fibroblasts were isolated from the outgrowth of foreskin obtained healthy 12-year-old volunteers with the approval of the Institutional Review Board of Seoul National University Hospital (Approval No. H-1101-116-353) and Seoul National University (No. E1408/001-002). HDFs were cultured in DMEM with 10% (v/v) fetal bovine serum and 1% (v/v) penicillin/streptomycin at 37 °C and 5% CO2. Serum-starved monolayer cultures of HDF were exposed to UVB irradiation and treated with CP dissolved in 50% ethanol. HDFs were exposed to UVB at a dose of 0.02 J/cm2 using a UVB source (Bio-Link crosslinker, VilberLourmat, Cedex 1, France) with a spectral peak set at 312 nm.

      Preparation of human skin equivalent model

      Neoderm-ED, an HSE model, was purchased from TEGO Science (Seoul, Korea). Briefly, HDFs were cultured onto a collagen matrix for 1 day, before keratinocytes were seeded on top of the matrix and cocultured for 4 days. Next, the keratinocytes and HDF block were raised for exposure to the air. CP was treated for 1 hour after 12 days of air exposure. The HSE layer was then irradiated with 0.05 J/cm2 UVB twice per day for 8 days. Media was changed every 2 days, and the layer was incubated at 37 °C with 5% CO2.

      Assessment of wrinkle formation

      To determine the severity of wrinkling, each hairless mouse was anesthetized and their UVB-exposed dorsal skin (wrinkle formation area) was photographed. The severity of wrinkling was measured by four trained graders using Bissett’s visual wrinkle scale (
      • Bissett D.L.
      • Hannon D.P.
      • Orr T.V.
      An animal model of solar-aged skin: histological, physical, and visible changes in UV-irradiated hairless mouse skin.
      ). A skin wrinkle replica was made with silicone rubber (Silflo Dental Impression Materials, Potters Bar, UK) from the backs of untreated mice. This was photographed using a coupling charge system video camera. Wrinkle severity was assessed using a photographic scale (0, none; 1, minimal; 2, mild; 3, moderate; 4, severe; and 5, very severe) and analyzed by Skin-Visiometer SV 600 software (CK Electronic GmbH, Köln, Germany). The visiometer consists of a computerized instrument that creates a skin microrelief map from the replica using a light transmission method.
      General laboratory experiments such as reverse transcriptase-PCR, western blotting Masson’s trichrome staining, immunohistochemistry, and luciferase reporter gene assay are described in Supplementary Materials and Methods online.

      Statistical analysis

      Statistical analyses were performed using one-way analysis of variance followed by Duncan’s statistical range test. P-values of less than 0.05 were considered statistically significant.

      Conflict of Interest

      The authors state no conflict of interest.

      Acknowledgments

      This work was supported by the R&D program of MOTIE/KIAT (Establishment of Infra Structure for Anti-aging Industry Support, No. N0000697), and the Mid-career Researcher Program (2015R1A2A1A10053567) through NRF grant funded by the MEST (Ministry of Education, Science and Technology) of the Republic of Korea. We thank Jieun H. Kim for proofreading the manuscript and Dr. Akira Oshima for helping histological analysis.

      Supplementary Material

      References

        • Ak P.
        • Levine A.J.
        p53 and NF-kappaB: different strategies for responding to stress lead to a functional antagonism.
        FASEB J. 2010; 24: 3643-3652
        • Benavides F.
        • Oberyszyn T.M.
        • VanBuskirk A.M.
        • Reeve V.E.
        • Kusewitt D.F.
        The hairless mouse in skin research.
        J Dermatol Sci. 2009; 53: 10-18
        • Bissett D.L.
        • Hannon D.P.
        • Orr T.V.
        An animal model of solar-aged skin: histological, physical, and visible changes in UV-irradiated hairless mouse skin.
        Photochem Photobiol. 1987; 46: 367-378
        • Bradley E.J.
        • Griffiths C.E.
        • Sherratt M.J.
        • Bell M.
        • Watson R.E.
        Over-the-counter anti-ageing topical agents and their ability to protect and repair photoaged skin.
        Maturitas. 2015; 80: 265-272
        • Cho H.S.
        • Lee M.H.
        • Lee J.W.
        • No K.O.
        • Park S.K.
        • Lee H.S.
        • et al.
        Anti-wrinkling effects of the mixture of vitamin C, vitamin E, pycnogenol and evening primrose oil, and molecular mechanisms on hairless mouse skin caused by chronic ultraviolet B irradiation.
        Photodermatol Photoimmunol Photomed. 2007; 23: 155-162
        • Chung J.H.
        • Seo A.Y.
        • Chung S.W.
        • Kim M.K.
        • Leeuwenburgh C.
        • Yu B.P.
        • et al.
        Molecular mechanism of PPAR in the regulation of age-related inflammation.
        Ageing Res Rev. 2008; 7: 126-136
        • Consortium EP
        An integrated encyclopedia of DNA elements in the human genome.
        Nature. 2012; 489: 57-74
        • El-Domyati M.
        • El-Ammawi T.S.
        • Medhat W.
        • Moawad O.
        • Mahoney M.G.
        • Uitto J.
        Expression of p53 protein after nonablative rejuvenation: the other side of the coin.
        Dermatol Surg. 2013; 39: 934-943
        • Fazekas D.
        • Koltai M.
        • Turei D.
        • Modos D.
        • Palfy M.
        • Dul Z.
        • et al.
        SignaLink 2—a signaling pathway resource with multi-layered regulatory networks.
        BMC Syst Biol. 2013; 7: 7
        • Fisher G.J.
        • Wang Z.Q.
        • Datta S.C.
        • Varani J.
        • Kang S.
        • Voorhees J.J.
        Pathophysiology of premature skin aging induced by ultraviolet light.
        N Engl J Med. 1997; 337: 1419-1428
        • Gerstein M.B.
        • Kundaje A.
        • Hariharan M.
        • Landt S.G.
        • Yan K.K.
        • Cheng C.
        • et al.
        Architecture of the human regulatory network derived from ENCODE data.
        Nature. 2012; 489: 91-100
        • Han E.S.
        • Muller F.L.
        • Perez V.I.
        • Qi W.
        • Liang H.
        • Xi L.
        • et al.
        The in vivo gene expression signature of oxidative stress.
        Physiol Genomics. 2008; 34: 112-126
        • Heinrich U.
        • Neukam K.
        • Tronnier H.
        • Sies H.
        • Stahl W.
        Long-term ingestion of high flavanol cocoa provides photoprotection against UV-induced erythema and improves skin condition in women.
        J Nutr. 2006; 136: 1565-1569
        • Ichihashi M.
        • Ando H.
        • Yoshida M.
        • Niki Y.
        • Matsui M.
        Photoaging of the skin.
        Anti-Aging Med. 2009; 6: 46-59
        • Itoh K.
        • Mimura J.
        • Yamamoto M.
        Discovery of the negative regulator of Nrf2, Keap1: a historical overview.
        Antioxid Redox Signal. 2010; 13: 1665-1678
        • Jin X.J.
        • Kim E.J.
        • Oh I.K.
        • Kim Y.K.
        • Park C.H.
        • Chung J.H.
        Prevention of UV-induced skin damages by 11,14,17-eicosatrienoic acid in hairless mice in vivo.
        J Korean Med Sci. 2010; 25: 930-937
        • Kang N.J.
        • Lee K.W.
        • Lee D.E.
        • Rogozin E.A.
        • Bode A.M.
        • Lee H.J.
        • et al.
        Cocoa procyanidins suppress transformation by inhibiting mitogen-activated protein kinase kinase.
        J Biol Chem. 2008; 283: 20664-20673
        • Katz D.L.
        • Doughty K.
        • Ali A.
        Cocoa and chocolate in human health and disease.
        Antioxid Redox Signal. 2011; 15: 2779-2811
        • Kawachi Y.
        • Xu X.
        • Taguchi S.
        • Sakurai H.
        • Nakamura Y.
        • Ishii Y.
        • et al.
        Attenuation of UVB-induced sunburn reaction and oxidative DNA damage with no alterations in UVB-induced skin carcinogenesis in Nrf2 gene-deficient mice.
        J Invest Dermatol. 2008; 128: 1773-1779
        • Kim H.H.
        • Lee M.J.
        • Lee S.R.
        • Kim K.H.
        • Cho K.H.
        • Eun H.C.
        • et al.
        Augmentation of UV-induced skin wrinkling by infrared irradiation in hairless mice.
        Mech Ageing Dev. 2005; 126: 1170-1177
        • Kim J.E.
        • Son J.E.
        • Jung S.K.
        • Kang N.J.
        • Lee C.Y.
        • Lee K.W.
        • et al.
        Cocoa polyphenols suppress TNF-alpha-induced vascular endothelial growth factor expression by inhibiting phosphoinositide 3-kinase (PI3K) and mitogen-activated protein kinase kinase-1 (MEK1) activities in mouse epidermal cells.
        Br J Nutr. 2010; 104: 957-964
        • Kim J.K.
        • Mun S.
        • Kim M.S.
        • Kim M.B.
        • Sa B.K.
        • Hwang J.K.
        5,7-Dimethoxyflavone, an activator of PPARalpha/gamma, inhibits UVB-induced MMP expression in human skin fibroblast cells.
        Exp Dermatol. 2012; 21: 211-216
        • Kim M.K.
        • Lee S.
        • Kim E.J.
        • Kong K.H.
        • Lee D.H.
        • Chung J.H.
        Topical application of anacardic acid (6-nonadecyl salicylic acid) reduces UV-induced histone modification, MMP-13, MMP-9, COX-2 and TNF-alpha expressions in hairless mice skin.
        J Dermatol Sci. 2013; 70: 64-67
        • Kota B.P.
        • Huang T.H.
        • Roufogalis B.D.
        An overview on biological mechanisms of PPARs.
        Pharmacol Res. 2005; 51: 85-94
        • Kuenzli S.
        • Saurat J.H.
        Effect of topical PPARbeta/delta and PPARgamma agonists on plaque psoriasis. A pilot study.
        Dermatology. 2003; 206: 252-256
        • Lalloyer F.
        • Wouters K.
        • Baron M.
        • Caron S.
        • Vallez E.
        • Vanhoutte J.
        • et al.
        Peroxisome proliferator-activated receptor-alpha gene level differently affects lipid metabolism and inflammation in apolipoprotein E2 knock-in mice.
        Arterioscler Thromb Vasc Biol. 2011; 31: 1573-1579
        • Learn D.B.
        • Beasley D.G.
        • Giddens L.D.
        • Beard J.
        • Stanfield J.W.
        • Roberts L.K.
        Minimum doses of ultraviolet radiation required to induce murine skin edema and immunosuppression are different and depend on the ultraviolet emission spectrum of the source.
        Photochem Photobiol. 1995; 62: 1066-1075
        • Lee K.W.
        • Kim Y.J.
        • Lee H.J.
        • Lee C.Y.
        Cocoa has more phenolic phytochemicals and a higher antioxidant capacity than teas and red wine.
        J Agric Food Chem. 2003; 51: 7292-7295
        • Lee K.W.
        • Kundu J.K.
        • Kim S.O.
        • Chun K.S.
        • Lee H.J.
        • Surh Y.J.
        Cocoa polyphenols inhibit phorbol ester-induced superoxide anion formation in cultured HL-60 cells and expression of cyclooxygenase-2 and activation of NF-kappaB and MAPKs in mouse skin in vivo.
        J Nutr. 2006; 136: 1150-1155
        • Lee Y.K.
        • Cha H.J.
        • Hong M.
        • Yoon Y.
        • Lee H.
        • An S.
        Role of NF-kappaB-p53 crosstalk in ultraviolet A-induced cell death and G1 arrest in human dermal fibroblasts.
        Arch Dermatol Res. 2012; 304: 73-79
        • Marini A.
        • Grether-Beck S.
        • Jaenicke T.
        • Weber M.
        • Burki C.
        • Formann P.
        • et al.
        Pycnogenol(R) effects on skin elasticity and hydration coincide with increased gene expressions of collagen type I and hyaluronic acid synthase in women.
        Skin Pharmacol Physiol. 2012; 25: 86-92
        • Marionnet C.
        • Tricaud C.
        • Bernerd F.
        Exposure to non-extreme solar UV daylight: spectral characterization, effects on skin and photoprotection.
        Int J Mol Sci. 2015; 16: 68-90
        • Michalik L.
        • Wahli W.
        Peroxisome proliferator-activated receptors (PPARs) in skin health, repair and disease.
        Biochim Biophys Acta. 2007; 1771: 991-998
        • Nelson W.G.
        • Kastan M.B.
        DNA strand breaks: the DNA template alterations that trigger p53-dependent DNA damage response pathways.
        Mol Cell Biol. 1994; 14: 1815-1823
        • Neukam K.
        • Stahl W.
        • Tronnier H.
        • Sies H.
        • Heinrich U.
        Consumption of flavanol-rich cocoa acutely increases microcirculation in human skin.
        Eur J Nutr. 2007; 46: 53-56
        • Park H.S.
        • Jin S.P.
        • Lee Y.
        • Oh I.G.
        • Lee S.
        • Kim J.H.
        • et al.
        Toll-like receptor 2 mediates a cutaneous reaction induced by repetitive ultraviolet B irradiation in C57/BL6 mice in vivo.
        Exp Dermatol. 2014; 23: 591-595
        • Reuter S.
        • Gupta S.C.
        • Chaturvedi M.M.
        • Aggarwal B.B.
        Oxidative stress, inflammation, and cancer: how are they linked?.
        Free Radic Biol Med. 2010; 49: 1603-1616
        • Saw C.L.
        • Yang A.Y.
        • Huang M.T.
        • Liu Y.
        • Lee J.H.
        • Khor T.O.
        • et al.
        Nrf2 null enhances UVB-induced skin inflammation and extracellular matrix damages.
        Cell Biosci. 2014; 4: 39
        • Scapagnini G.
        • Davinelli S.
        • Di Renzo L.
        • De Lorenzo A.
        • Olarte H.H.
        • Micali G.
        • et al.
        Cocoa bioactive compounds: significance and potential for the maintenance of skin health.
        Nutrients. 2014; 6: 3202-3213
        • Schafer M.
        • Farwanah H.
        • Willrodt A.H.
        • Huebner A.J.
        • Sandhoff K.
        • Roop D.
        • et al.
        Nrf2 links epidermal barrier function with antioxidant defense.
        EMBO Mol Med. 2012; 4: 364-379
        • Scott F.L.
        • Hirst C.E.
        • Sun J.
        • Bird C.H.
        • Bottomley S.P.
        • Bird P.I.
        The intracellular serpin proteinase inhibitor 6 is expressed in monocytes and granulocytes and is a potent inhibitor of the azurophilic granule protease, cathepsin G.
        Blood. 1999; 93: 2089-2097
        • Scott F.L.
        • Sun J.
        • Whisstock J.C.
        • Kato K.
        • Bird P.I.
        SerpinB6 is an inhibitor of kallikrein-8 in keratinocytes.
        J Biochem. 2007; 142: 435-442
        • Seo J.Y.
        • Lee S.H.
        • Youn C.S.
        • Choi H.R.
        • Rhie G.E.
        • Cho K.H.
        • et al.
        Ultraviolet radiation increases tropoelastin mRNA expression in the epidermis of human skin in vivo.
        J Invest Dermatol. 2001; 116: 915-919
        • Son E.D.
        • Kim H.
        • Choi H.
        • Lee S.H.
        • Lee J.Y.
        • Kim S.
        • et al.
        Cathepsin G increases MMP expression in normal human fibroblasts through fibronectin fragmentation, and induces the conversion of proMMP-1 to active MMP-1.
        J Dermatol Sci. 2009; 53: 150-152
        • Son E.D.
        • Shim J.H.
        • Choi H.
        • Kim H.
        • Lim K.M.
        • Chung J.H.
        • et al.
        Cathepsin G inhibitor prevents ultraviolet B-induced photoaging in hairless mice via inhibition of fibronectin fragmentation.
        Dermatology. 2012; 224: 352-360
        • Subhashini R.
        • Mahadeva Rao U.S.
        • Sumathi P.
        • Gunalan G.A.
        Comparative phytochemical analysis of cocoa and green tea.
        Indian J Sci Technol. 2010; 3: 188-192
        • Sykiotis G.P.
        • Bohmann D.
        Stress-activated cap'n'collar transcription factors in aging and human disease.
        Sci Signal. 2010; 3: re3
        • Tian F.F.
        • Zhang F.F.
        • Lai X.D.
        • Wang L.J.
        • Yang L.
        • Wang X.
        • et al.
        Nrf2-mediated protection against UVA radiation in human skin keratinocytes.
        Biosci Trends. 2011; 5: 23-29
        • Urpi-Sarda M.
        • Monagas M.
        • Khan N.
        • Llorach R.
        • Lamuela-Raventos R.M.
        • Jauregui O.
        • et al.
        Targeted metabolic profiling of phenolics in urine and plasma after regular consumption of cocoa by liquid chromatography-tandem mass spectrometry.
        J Chromatogr A. 2009; 1216: 7258-7267
        • van Kranen H.J.
        • de Laat A.
        • van de Ven J.
        • Wester P.W.
        • de Vries A.
        • Berg R.J.
        • et al.
        Low incidence of p53 mutations in UVA (365-nm)-induced skin tumors in hairless mice.
        Cancer Res. 1997; 57: 1238-1240
        • Varga T.
        • Czimmerer Z.
        • Nagy L.
        PPARs are a unique set of fatty acid regulated transcription factors controlling both lipid metabolism and inflammation.
        Biochim Biophys Acta. 2011; 1812: 1007-1022
        • Viros A.
        • Sanchez-Laorden B.
        • Pedersen M.
        • Furney S.J.
        • Rae J.
        • Hogan K.
        • et al.
        Ultraviolet radiation accelerates BRAF-driven melanomagenesis by targeting TP53.
        Nature. 2014; 511: 478-482
        • Xu Y.
        • Fisher G.J.
        Ultraviolet (UV) light irradiation induced signal transduction in skin photoaging.
        J Dermatol Sci. 2005; : S1-S8
        • Yoon H.S.
        • Shin C.Y.
        • Kim Y.K.
        • Lee S.R.
        • Chung J.H.
        Endogenous estrogen exacerbates UV-induced inflammation and photoaging in mice.
        J Invest Dermatol. 2014; 134: 2290-2293
        • Zheng Y.
        • Lai W.
        • Wan M.
        • Maibach H.I.
        Expression of cathepsins in human skin photoaging.
        Skin Pharmacol Physiol. 2011; 24: 10-21