Biosci. Biotechnol. Biochem., 76 (12), 2317–2320, 2012 Note Screening of Medicinal and Edible in Okinawa, Japan, for Enhanced Proliferative and Collagen Synthesis Activities in NB1RGB Human Skin Fibroblast Cells

y Makoto TAKAHASHI,1 Yonathan ASIKIN,2 Kensaku TAKARA,1 and Koji WADA1;

1Department of Bioscience and Biotechnology, Faculty of Agriculture, University of the Ryukyus, Okinawa 903-0213, Japan 2Course of Biochemistry and Applied Bioscience, United Graduate School of Agricultural Science, Kagoshima University, Kagoshima 890-0065, Japan

Received June 21, 2012; Accepted August 21, 2012; Online Publication, December 7, 2012 [doi:10.1271/bbb.120478]

To identify plants with bioactive potential for skin survey the additional medicinal characteristics of these care, methanol extracts of 56 parts from 47 plants, we evaluated the effects of the Okinawan plant medical and edible plants cultivated in Okinawa were extracts on cell proliferation and type I collagen syn- tested for their proliferative effects on NB1RGB skin thesis in human skin cell lines. fibroblast cells. Extracts from six plants, Bischofia A total of 56 plant parts from 47 species collected javanica, Colocasia esculenta, Melaleuca alternifolia, from the University of Ryukyus campus (Okinawa, Piper angustifolia, Jasminum sambac, and Curcuma Japan) are listed in Table 1. Each dried plant part (2 g) longa, showed higher NB1RGB cell proliferation activ- was ground using a mini-grinder (Model DF-15, ity (>10%) than the control, at various concentrations. Shenzhen Laitong Co., Shenzhen, China), and then Among the six extracts, only the C. longa extract caused soaked in 20 mL of methanol at room temperature for an increase in collagen synthesis in NB1RGB cells, as 24 h. The methanol solution was then filtered and compared to treatment with the positive control, concentrated under reduced pressure, and the obtained ascorbic acid (AsA). Expression of the collagen synthesis extract was used in the subsequent experiments, unless marker, transforming growth factor-1, was higher otherwise indicated. after treatment with the C. longa extract than with AsA. The NB1RGB cell line used in the assay for proliferation and collagen synthesis were normal human Key words: Okinawan plants; skin care; fibroblast pro- skin fibroblasts, purchased from the Riken Cell Bank liferation; collagen synthesis; transforming (Ibaraki, Japan). NB1RGB cells were cultivated in - growth factor-1 (TGF-1) MEM containing 10% fetal bovine serum (FBS) and were maintained at 37 C and 5% CO2. For the assay on Type I collagen is the predominant fibrous protein of proliferation, the cells were seeded on 96-well micro- the extracellular matrix, and it serves as the major plates (5 103 cells/well) and were cultivated as pre- protein constituent in human connective tissues.1) viously mentioned. After 24 h, the medium was changed Approximately 3–6% of the tissues in the body consist to 200 mLof-MEM containing 10% FBS supple- of collagen, and the functional properties of the skin mented with a plant extract or L-ascorbic acid (AsA), depend on the integrity of collagen in the dermis. The and the cells were incubated for 72 h. In this assay, the deposition of collagen is intricately controlled and is concentrations of the test compounds were 0.0032, dependent on the physiological status of the body. 0.016, 0.08, 0.4, 2, 4, 10, and 50 mg/mL. The control Changes in the rate of collagen deposition occur during cells were cultivated in the medium without the plant wound-healing,2) bone development,3) and aging.4) extract or AsA. AsA was used as a positive control in Therefore, the control of skin fibroblast proliferation this study because it has been identified as an inducer of and collagen metabolism may be useful in a variety of proliferation and type I collagen synthesis in fibro- therapeutic and cosmetic applications. blasts.5) After 72 h of exposure to the plant extract or The Okinawa area is located at the southern region of AsA, cell proliferation was determined using the MTS Japan and consists of more than 100 islands. Okinawa is assay, as described elsewhere.6) Data were expressed in the only region in Japan that has a subtropical climate, terms of cell proliferation rate (%), which was calculated allowing the growth and cultivation of both tropical and by the formula [(absorbance with the supplemented temperate plants. The inhabitants of Okinawa have medium absorbance with the supplemented medium traditionally utilized these plants as medicinal herbs and without cells)/(absorbance with the control medium food. In this study, we collected a large variety of absorbance of the medium itself)] 100. Table 1 shows tropical and subtropical plants, most of which have been the maximum NB1RGB cell proliferation rate in the used as traditional medicinal herbs in Okinawa. To presence of various test compound concentrations. Only

y To whom correspondence should be addressed. Tel: +81-98-895-8809; Fax: +81-98-895-8734; E-mail: [email protected] Abbreviations: FBS, fetal bovine serum; -MEM, alpha modified eagle minimum essential medium; AsA, L-ascorbic acid; MTS, 3-(4,5- dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium; ELISA, enzyme-linked immunoassay; TGF-1, transforming growth factor-1; SDS, sodium dodecyl sulfate 2318 M. TAKAHASHI et al. Table 1. NB1RGB Cell Proliferation in the Presence of Plant Extracts from Okinawa, Japan

Maximum Concentrationc # Scientific name Family Plant organ Proliferationa,b (mg/mL) (% of control) 1 Bischofia javanica Euphorbiaceae Twig 112.0 0.0032 2 Leaf <100 — 3 Oenothera rosea Onagraceae Stem-leaf 109.6 10 4 Acalypha hispaniolae Euphorbiaceae Twig <100 — 5 Leaf <100 — 6 Wedelia trilobata Asteraceae Leaf <100 — 7 Geranium carolinianum Geraniaceae Leaf <100 — 8 Allamanda cathartica Apocynaceae Leaf 107.6 0.0032 9 Ficus elastica Moraceae Twig <100 — 10 Ficus religiosa Moraceae Twig 109.4 0.0032 11 Leaf <100 — 12 Foeniculum vulgare Apiaceae Stem-leaf <100 — 13 Ficus septica Moraceae Leaf <100 — 14 Syzygium samarangense Myrtaceae Leaf <100 — 15 Fruit <100 — 16 Pongamia pinnata Leguminosae Leaf <100 — 17 Murraya paniculata Rutaceae Leaf <100 — 18 Heritiera littoralis Sterculiaceae Leaf <100 — 19 Colocasia esculenta Araceae Leaf 112.5 0.4 20 Rhaphiolepis indica Rosaceae Twig 101.9 0.0032 21 Leaf 105.6 10 22 Eugenia jambolana Myrtaceae Twig <100 — 23 Leaf 109.3 0.8 24 Melaleuca alternifolia Asteraceae Twig-leaf 126.2 0.016 25 Ixeris dentata Leaf <100 — 26 Hibiscus rosa-sinensis Malvaceae Twig <100 — 27 Leaf 109.6 10 28 Zanthoxylumbeecheyanum Rutaceae Leaf 108.6 0.0032 29 Beta vulgaris Chenopodiaceae Aerial part <100 — 30 Callistemon linearis Myrtaceae Stem-leaf 107.8 0.016 31 Peucedanum japonicum Apiaceae Whole body <100 — 32 Piper angustifolia Piperaceae Leaf 115.3 10 33 Jasminum sambac Oleaceae Twig 104.7 10 34 Leaf 113.9 0.0032 35 Bauhinia purpurea Leguminosae Twig 104.1 0.0032 36 Leaf <100 — 37 Heliotropium foertherianum Boraginaceae Leaf <100 — 38 Cinnamomum tenuifolium Lauraceae Leaf 109.2 2 39 Daucus carota Apiaceae Rhizome <100 — 40 Gynura bicolor Asteraceae Leaf 101.1 0.0032 41 Tradescantia ohiensis Commelinaceae Leaf 106.6 0.0032 42 Artemisia ndica Asteraceae Leaf 107.9 50 43 Anredera cordifolia Basellaceae Leaf 101.7 10 44 Viola mandshurica Violaceae Leaf 107.5 50 45 Psidium guajava Myrtaceae Leaf 109.6 0.016 46 Cymbopogon citratus Poaceae Leaf 103.5 10 47 Malpighia glabra Malpigiaceae Leaf 101.4 50 48 Myrciaria cauliflora Myrtaceae Leaf 105.5 0.0032 49 Hibiscus sabdariffa Malvaceae Leaf 106.0 0.0032 50 Crassocephalum crepidioides Asteraceae Leaf <100 — 51 Momordica charantia Cucurbitaceae Fruit 100.2 0.0032 52 Caulerpaceae Leaf 100.1 0.016 53 Curcuma longa Zingiberaceae Rhizome 112.0 4 54 Citrus depressa Rutaceae Rhizome 107.1 50 55 Carica papaya Caricaceae Rhizome 106.8 50 56 Averrhoa carambola Oxalidaceae Fruit 106.8 50 Ascorbic acid 143.8 4

aThe mean values are maximum NB1RGB cell proliferation rate in the presence of various concentrations of 0.0032 to 50 mg/mL. bValues are expressed as percentage of control, which was set at 100%. cValues are concentrations of each test compound that indicated maximum cell proliferation. six plant extracts obtained from Bischofia javanica tions, whereas their activities were lower than that of (twig), Colocasia esculenta (leaf), Melaleuca alternifo- AsA (maximum cell proliferation: 143.8% at 4 mg/mL). lia (twig and leaf), Piper angustifolia (leaf), Jasminum NB1RGB cell proliferation evaluated in vitro has been sambac (leaf), and Curcuma longa (rhizome) showed reported to be an important assay for studies on collagen >10% increase in proliferation at different concentra- synthesis,7) showing a positive correlation with collagen Screening of Skin Care Ability of Okinawan Plants 2319

250 *

200

150

100

50

0 Collagen production rate (% of control) Collagen production rate

Fig. 1. Effects of Six Plants Extracts on Type I Collagen Synthesis in Normal Human Skin Fibroblasts. The final concentration of each extract was fixed at the concentrations indicating maximum cell proliferation as follows: Bischofia javanica (twig), 0.0032 mg/mL; Colocasia esculenta (leaf), 0.4 mg/mL; Melaleuca alternifolia (twig and leaf), 0.016 mg/mL; Piper angustifolia (leaf), 10 mg/mL; Jasminum sambac (leaf), 0.0032 mg/mL; Curcuma longa (rhizome), 4 mg/mL; and ascorbic acid, 4 mg/mL. Each value represents the mean SD (n ¼ 3). Statistical analysis was performed using Dunnet’s test. The value marked by an asterisk was significantly different from that of the control at p < 0:05 (). synthesis. Hence, the six plant extracts were extensively used in the experiment of type I collagen synthesis in TGF-β1 fibroblasts, with each concentration indicating maxi- mum cell proliferation. For the assay on collagen synthesis, NB1RGB cells were seeded at a concentration of 5 104 cells/dish (’: β-Actin 60 mm, height: 15 mm) and were cultivated as described above. After 24 h, the medium was changed to -MEM that contained 10% FBS supplemented with each of the Control As A C. longa six plant extracts or AsA, and this was maintained for Fig. 2. Effect of Ascorbic Acid and Curcuma longa Extract on the 72 h. Cell viability was then measured using a blood Expression of TGF-1 in Normal Human Skin Fibroblasts. corpuscle counting chamber. After a test confirming the The cell lysates were separated by SDS–PAGE, and the signals of non-toxicity of the extracts (cell viability > 95%), the TGF-1 and -actin as the controls were detected by immunoblot amount of type I collagen in the culture medium was analysis. also measured using an enzyme-linked immunoassay (ELISA) kit (human collagen type I ELISA kit, Cosmo NB1RGB cells was determined using Western blot Bio Co., Ltd., Tokyo, Japan), according to the manu- analysis. NB1RGB cells treated with the C. longa facturer’s instructions. Data were expressed in terms of extract or AsA were prepared using the same conditions collagen production rate (%), as compared to the as described above for the collagen synthesis assay. control, in micrograms per live cell. All data were After cultivation with the plant extract or control, the expressed as mean SD. The data were analyzed using medium was removed, and the cells were then scraped one-way analysis of variance with Dunnett’s multiple into a lysis buffer (60 mM Tris–HCl [pH 7.4], 5% comparison test. Differences were considered significant 2-mercaptoethanol, 2% sodium dodecyl sulfate [SDS], at p < 0:05. Figure 1 shows the production rate for type and 5% sucrose), sonicated for 15 s, heated at 95 C I collagen in NB1RGB cells. The results indicated that for 5 min, collected by centrifugation, and stored at the production rate of collagen was higher in the 80 C. The lysates were subjected to SDS-polyacryl- presence of the C. longa (rhizome) extract than in the amide gel electrophoresis, and the proteins were trans- presence of the other five extracts. Collagen synthesis in ferred onto immobilon polyvinylidene fluoride mem- the presence of the C. longa extract was also higher than branes (Millipore, Billerica, MA). The blot was probed that observed with AsA, whereas the proliferation rate of using either anti-TGF-1 (Cosmo Bio) or anti--actin NB1RGB cells after the introduction of the extract was (Cosmo Bio), and immune complexes were detected lower than that shown for AsA. These results might be using horseradish peroxidase conjugated with specific attributable to differences in the cytological actions of secondary antiserum (GE Healthcare Ltd., Tokyo, the C. longa extract and AsA. Japan). As expected, the expression of TGF-1 was Transforming growth factor-1 (TGF-1) has been increased by the administration of the C. longa extract reported to increase collagen synthesis and to induce cell to the cells, despite the normal expression of -actin proliferation and renewal in various types of skin cells.8) (Fig. 2). Interestingly, the signal density of TGF-1in To clarify the mechanism by which the C. longa extract the cells treated with the C. longa extract was also promoted collagen synthesis, expression of TGF-1in higher than that of cells incubated with AsA as the 2320 M. TAKAHASHI et al. positive control. These results suggest the possible References involvement of TGF-1 in the mechanism for the promotion of collagen production by using the C. longa 1) Lee CH, Singla A, and Lee Y, Int. J. Pharm., 221, 1–22 (2001). extract. The rhizomes of C. longa are widely used for 2) Frazier WF, Robert FD, Bruce AM, and Kelman C, J. Pediatr. Surg., l27, 945–949 (1992). different therapeutic purposes in India and other Asian 3) Mundlos S, Engel H, Michel-Behnke I, and Zabel B, Bone, 11, 9) countries. A variety of food supplements containing 275–279 (1990). C. longa are also currently being commercially manu- 4) Wang SY, Merrill C, and Bell E, Mech. Ageing Dev., 44, 127– factured and marketed. Curcumin, a polyphenol and one 141 (1988). of the most important lipidic ingredients in C. longa, has 5) Shibayama H, Hisama M, Matsuda S, Ohtsuki M, and Iwaki M, been reported to be effective as a traditional skin remedy Biol. Pharm. Bull., 31, 563–568 (2008). 10,11) 6) Takahashi M, Morita T, Fukuoka T, Imura T, and Kitamoto D, with anti-oxidative and anti-inflammatory activi- J. Oleo Sci., 61, 457–464 (2012). 12,13) ties and capacity for suppression of matrix metal- 7) Takahashi M, Kitamoto D, Asikin Y, Takara K, and Wada K, 14) loproteinase-2 expression. In the present study, we J. Oleo Sci., 58, 643–650 (2009). found that extracts derived from C. longa promote type I 8) Koya-Miyata S, Okamoto I, Ushio S, Iwaki K, Ikeda M, and collagen synthesis and TGF-1 expression. However, Kurimoto M, Biosci. Biotechnol. Biochem., 68, 767–773 (2004). the active component in C. longa that is mainly 9) Goel A, Kunnumakkara AB, and Aggarwal BB, Biochem. responsible for these molecular activities remains Pharmacol., 75, 787–809 (2008). 10) Kapoor S and Priyadarsini KI, Biophys. Chem., 92, 119–126 unknown, and further studies are therefore required. (2001). In conclusion, the results of our study indicate that 11) Pulla-Reddy ACH and Lokesh BR, Mol. Cell. Biochem., 111, among the 47 medicinal and edible plant species 117–124 (1992). cultivated in Okinawa, six plant extracts derived from 12) Bharti AC, Donato N, and Aggarwal BB, J. Immunol., 171, B. javanica, C. esculenta, M. alternifolia, P. angustifo- 3863–3871 (2003). lia, J. sambac, and C. longa showed higher NB1RGB 13) Arora RB, Basu N, Kapoor V, and Jain A, Indian J. Med. Res., 59 cell proliferative activity. In addition, the C. longa , 1289–1295 (1971). 14) Banerji A, Chakrabarti J, Mitra A, and Chatterjee A, Cancer extract caused an increase in type I collagen production Lett., 211, 235–242 (2004). in NB1RGB cells by promoting TGF-1 expression, suggesting its potential in a variety of skin therapeutic and cosmetic applications.