Lutein, a Nonprovitamin A, Activates the Retinoic Acid Receptor to Induce HAS3-Dependent Hyaluronan Synthesis in Keratinocytes

Biosci. Biotechnol. Biochem., 77 (6), 1282–1286, 2013

y Tetsuya SAYO, Yoshinori SUGIYAMA, and Shintaro INOUE

Innovative Beauty Science Laboratory, Kanebo Cosmetics Inc., Kotobuki-cho, Odawara 250-0002, Japan

Received February 15, 2013; Accepted March 13, 2013; Online Publication, June 7, 2013 [doi:10.1271/bbb.130124]

Carotenoids have been reported to have potent -Cryptoxanthin, which has a structure similar to - antioxidant activities and to protect tissues and cells carotene, can also be cleaved by -carotene 15, 150- from certain diseases and environmental insults. The monooxygenase (BCO1) to produce retinal and retinoic molecular mechanism of the action of provitamin A acid.7) It has been reported that high doses of -carotene carotenoids such as -carotene and -cryptoxanthin is substantially reduced the malignant conversion of mediated in part by retinoic acid, an active form of papillomae to carcinomas as well as retinoic acid.8,9) provitamin A, but the molecular basis of the biological Because both provitamin A carotenoids and retinoids are activities of non-provitamin A carotenoids such as used as potential protective agents in tumor formation, lutein, zeaxanthin, and astaxanthin is not fully under- the molecular mechanisms of the action of carotenoids is stood. In this study, we investigated to determine probably mediated at least in part by retinoic acid. whether the actions of non-provitamin A carotenoids Topical treatments of retinoic acid have been found are mediated via retinoid signaling by monitoring to be beneficial to photodamaged skin.10,11) Retinoic retinoic acid receptor (RAR)-dependent hyaluronan acid inhibits UV-caused loss of procollagen synthe- production in cultured human keratinocytes. Not only sis,12) the induction of matrix metalloproteinases,13) and -carotene and -cryptoxanthin, but also lutein, zeax- the tortuosity of elastic fibers.14) In addition to these anthin, and astaxanthin, upregulated HAS3 gene ex- effects, it increases the synthesis of epidermal hyalur- pression and were followed by hyaluronan synthesis. We onan,15,16) a high molecular mass glycosaminoglycan found that LE540, an antagonist of retinoic acid that might support epidermal homeositasis by main- receptors, abolished lutein dependent hyaluronan syn- taining the extracellular space, facilitating the exchange thesis and that lutein significantly increased retinoic of nutrients and waste products.17) This suggests the acid responsive element (RARE)-driven transcript aci- possible involvement of hyaluronan in retinoic acid- tivity. In addition, we found that citral, an inhibitor of induced repair, because hyaluronan has been reported retinal dehydrogenases, decreased lutein-stimulated to be closely involved in keratinocyte proliferation and hyaluronan synthesis, indicating that lutein metabolites differentiation.18–20) rather than lutein itself act as an RAR ligand in RAR- Hyaluronan is synthesized by hyaluronan synthase mediated transcription activity in keratinocytes. A proteins (HAS1, HAS2, and HAS3) at the inner surface series of non-provitamin A can be substituted for of the plasma membrane.21) In human epidermal retinoids and should be considered as a potential means keratinocytes, HAS3 mRNA is dominantly expressed of improving skin health. and primarily regulates hyaluronan synthesis.22–24) In- duction of HAS3 gene expression by -carotene stim- Key words: provitamin A; non-provitamin A; lutein; ulation was also found in keratinocytes.25) Since the retinoic acid receptor; hyaluronan expression of BCO17,26) and the conversion from retinal to retinoic acid have been confirmed in the skin,27) the Carotenoids are plant pigments with roles in photo- molecular mechanism of action of provitamin A is synthesis and photoprotection. It has been found that probably mediated by retinoic acid receptors (RARs) in carotenoids have antitumor,1) photoprotective,2) radical the keratinocytes. quenching,3) and antioxidant4) activities. They display On the other hand, non-provitamin A carotenoids these beneficial effects, as well as the ability to act as have also been found to have physiological activities. provitamin A and precursors of retinoic acid. The major Suppression of mouse skin papilloma28) and morpho- provitamin A carotenoids in mammals are -carotene, - logical changes in human mammary cells29) were cryptoxanthin, and -carotene. In contrast, it is currently observed in response not only to -carotene, but also thought that non-provitamin A carotenoids including to canthaxanthin, a non-provitamin A carotenoid. These lutein, zeaxanthin, and astaxanthin, are not precursors of effects have been observed in response to both provi- retinoic acid in mammals.5) tamin A and non-provitamin A, suggesting that retinoid- -Carotene, a major dietary source of vitamin A, is like metabolites converted from non-provitamin A act in converted to retinoic acid, an active form of vitamin A, these biological effects. Recently, Sangeetha reported by a series of enzymatic steps in the intestinal mucosa.6) that astaxanthin, a non-provitamin A carotenoid, is

y To whom correspondence should be addressed. Tel: +81-465-34-6116; Fax: +81-465-34-3037; E-mail: [email protected] Abbreviations: HAS, hyaluronan synthase; RA, retinoic acid; RAL, retinal; RAR, retinoic acid receptor; RARE, retinoic acid responsive element; SEAP, secreted alkaline phosphatase; BCO1, -carotene 15, 150-monooxygenase; RALDH, retinal dehydrogenase; BPE, bovine pituitary extracts Lutein Activates the Retinoic Acid Receptor in Keratinocytes 1283 converted to -carotene and retinol in retinol-deficient promoter (Clontech, Palo Alto, CA). A pSEAP2-Basic vector rats,30) but the details as to where and how astaxanthin is (Clontech) containing a SEAP gene without a promoter was used as converted to retinoids have remained obscure. negative control to measure the background signal associated with the cell culture media. Keratinocytes in a 12-well plate were transfected In view of the above, we hypothesized that non- with pRARE-TA-SEAP or pSEAP2-Basic vector with Fugene 6 provitamin A can be converted to carboxylic acids transfection reagent (Roche Diagnostics, Mannheim, Germany), as corresponding to retinoic acid in the skin, and have indicated in the manual. After transfection for 24 h, the cells were similar functions. treated with lutein and incubated for 24 h. SEAP activities in the In the present study, we examined the effects of non- conditioned media were measured by the EscApe SEAP chemilumi- provitamin A carotenoids on hyaluronan production in nescence detection system (Clontech) and Luminescencer PSN human keratinocytes. Furthermore, we sought to deter- AB2200 (ATTO, Tokyo). mine whether the physiological effects of non-provita- Statistical analysis. Results were expressed as means SD. All min A carotenoids in cultured keratinocytes is due to statistical calculations were performed using SAS software. The data RAR activation. were analyzed using Dunnett’s multiple comparison test. Dose dependency was determined by the Williams test. A p value of less Materials and Methods than 0.05 was considered to be statistically significant. Chemicals and materials. Human foreskin keratinocytes and bovine Results pituitary extracts (BPE) were purchased from Kurabo (Osaka, Japan). MCDB153 medium, LE540, and Am80 were from Wako Pure Chemical (Osaka, Japan). -Cryptoxanthin, lutein, zeaxanthin, Effects of carotenoids on HAS3-mediated hyaluronan astaxanthin, lycopene, -ionone, -ionone, and citral were from synthesis in keratinocytes Extrasynthese (Genay, France). Phytoene was from Carotenature We evaluated hyaluronan synthesis in human kerati- (Lupsingen, Switzerland). All-trans retinoic acid, all-trans retinal, nocytes by various carotenoids, including -carotene as and -carotene were from Sigma (St. Louis, MO). Ch55 was from positive control,25) and related compounds to investigate Tocris Bioscience (Bristol, UK). structure-activity relationships (Fig. 1). The structure of Cell culture. Human epidermal keratinocytes were grown routinely each of the compounds investigated is shown in Fig. 2. in MCDB153 (0.1 mM Ca2þ) supplemented with 5 mg/L of insulin, Hyaluronan increased significantly when keratino- 180 mg/L hydrocortisone, 14.1 mg/L O-phosphorylethanolamine, cytes were exposed not only to the provitamin A 6.1 mg/L 2-aminoethanol, 100 ng/L EGF, and 0.4% (vol/vol)BPE. carotenoids -carotene and -cryptoxanthin, but also to When the cells reached confluence, they were subcultured in either 24- non-provitamin A carotenoids lutein, zeaxanthin, and well plates or 12-well plates in the same medium. After subculturing, astaxanthin (Fig. 1a). In contrast, lycopene, phytoene, the cells were confluent and the medium was exchanged for an identical medium, except that hydrocortisone and EGF were depleted -ionone, and -ionone did not exhibit significant and BPE was reduced to a concentration of 0.04%. Agents were added to the medium and cultured for the time of indicated in the legends to the figures. The conditioned media of the 24-well plates were assayed a b for hyaluronan production, and the cells were used in RT-PCR HAS3 analyses. The cells of the 12-well plates were used in secreted alkaline HAS3 phosphatase (SEAP) reporter assays. G3PDH G3PDH RT-PCR. Total RNA from the cultured cells was extracted using ** 1200 1200 TRIzol reagent (Invitrogen, Carlsbad, CA) folloing the manufacturer’s *** ** ** recommended protocol. The reaction was carried out with a Tran- 800 scriptor First Strand cDNA Synthesis Kit (Roche Diagnostics, 800 *** *** Mannheim, Germany). The following primer sets were used in PCR *** analysis: HAS3 sense: 50-CTGCACCTGCTCATTCAGAG-30, HAS3 * 400 0 0 0 anti-sense: 5 -GAGTCGCACACCTGGATGTA-3 , G3PDH sense: 5 - 400 0 0 0 ACCACAGTCCATGCCATCAC-3 , G3PDH anti-sense: 5 -TCCACC- content (ng/well) Hyaluronan 0 0.3 1 3 5 ACCCTGTTGCTGTA-30. These primers were designed to span exons Hyaluronan content (ng/well) Hyaluronan Lutein (µM) using Primer3. The PCR cycles and amplified product sizes were as 0 follows: HAS3, 25 cycles, 488 bp, and G3PDH, 22 cycles, 452 bp. c Amplified cDNAs were analyzed on a 2% agarose gel containing 0 2 8 24 (h) ethidium bromide. A 100 bp-DNA ladder standard was used as size HAS3 marker. G3PDH

Measurements of hyaluronan production. Hyaluronan levels were Fig. 1. Effects of Carotenoids on Hyaluronan Synthesis in Keratino- was measured by sandwich assay, as described previously.31) Briefly, cytes. 96-well plates were precoated with hyaluronan binding protein a, Confluent human keratinocytes were incubated for 48 h in the (HABP) (Seikagaku, Tokyo) to catch hyaluronan from the samples absence and the presence of -carotene, -cryptoxanthin, lutein, and standards. After washing, HABP-labeled with an EZ-Link zeaxanthin, and astaxanthin at a concentration of 5 mM. Hyaluronan Maleimide Activated Horseradish Peroxidase Kit (Pierce, Rockford, levels were measured as described in ‘‘Materials and Methods.’’ IL) was added to detect bound hyaluronan for spectrophotometric Values represent mean SD for three replicate assays. Significantly 0 0 quantification at 420 nm with a 3,3 ,5,5 -tetramethylbenzidine (TMB) different from the control value: p < 0:05, p < 0:001 (Dunnett’s substrate solution (Moss, Pasadena, MD). test). Total RNA was extracted and subjected to RT-PCR. The PCR products were visualized in a 2% agarose gel stained with ethidium Assay for DNA. DNA was measured using fluorescent reagent 3, 5- bromide. b, Human keratinocytes were cultured with various 32) diaminobenzoic acid, as described by Johnson-Wint and Hollis. concentrations of lutein (0.3–5 mM) for 48 h. Hyaluronan assay and RT-PCR were carried out as described above. Significantly different SEAP reporter assay. The RARE reporter construct pRARE-TA- from the control value: p < 0:01 (Williams test). c, Human SEAP contains 2 copies of a RARE sequence upstream of the minimal keratinocytes were cultured with 5 mM lutein for various times. Total TA promoter, and contains a SEAP promoter gene driven by an SV40 RNA was analyzed as described above. 1284 T. SAYO et al.

β -Carotene Lycopene

β-Cryptoxanthin Phytoene

Lutein α-Ionone

Zeaxanthin β-Ionone

Astaxanthin

Fig. 2. Chemical Structure of the Compounds Examined in This Study. The carotenoids used in this study were in the all-trans conﬁguration.

8,000 activity (254, 183, 274, and 274 ng/well respectively ** *** as compared to the control, 288 ng/well), suggesting 300 ab*** ** that both cyclohexene rings and polyene moieties are 6,000 necessary to induce hyaluronan synthesis. The stimula- 200 ** 4,000 tion of hyaluronan synthesis by carotenoids was accom- SEAP activity panied by an increase in the level of HAS3 mRNA 100 2,000 (Fig. 1a). (arbitrary chemiluminescence units) Hyaluronan content (ng/well) Hyaluronan Among the non-provitamin A carotenoids tested, we 0 0 Control β-Carotene Lutein 02510 focused on lutein to investigate in detail how non- µ provitamin A can stimulate hyaluronan synthesis via Lutein ( M) upregulation of HAS3 gene expression. Lutein showed Fig. 3. Effect of Lutein on RAR Activation. a dose-dependent increase in hyaluronan production a, Following 30 min of incubation of human keratinocytes in the with a corresponding increase in HAS3 gene expression presence (open columns) and the absence (closed columns) of 0.2 mM LE540, -carotene or lutein was added to the culture at a at a concentration of 1 to 5 mM (Fig. 1b). Under these concentration of 5 mM and this was incubated for 24 h. Hyaluronan conditions, the addition of lutein did not induce notable content was measured as described in ‘‘Materials and Methods.’’ alterations in DNA content (Supplemental Fig. 1; see Values represent mean SD for three replicate assays. Significantly Biosci. Biotechnol. Biochem. Web site), indicating that different from the value of LE540 untreated cells: p < 0:01, the effects of lutein on hyaluronan synthesis were p < 0:001 (Dunnett’s test). b, Keratinocytes transfected with independent of cell proliferation. Upregulation of HAS3 pRARE-TA-SEAP reporter vector and an internal control vector were treated with the indicated concentrations of lutein (2–10 mM) mRNA expression was observed at 2 h after the for 24 h. SEAP activity was measured by the EscApe SEAP addition of lutein and increased time-dependently Chemiluminescence system. Values represent mean SD for three (Fig. 1c). replicate assays. Significantly different from the control value: p < 0:01 (Williams test). Effects of lutein on RAR activation To assess whether hyaluronan synthesis is induced or their metabolites serve as a ligand for RARs in human by lutein via RARs, the effect of LE540, a pan- keratinocytes. antagonist for RARs,33) was investigated. Hyaluronan production induced by lutein as well as -carotene was Effects of citral on hyaluronan synthesis induced by completely abolished by LE540 treatment, although lutein basal hyaluronan production was also partially reduced Keratinocytes can convert retinal into retinoic acid.27) (Fig. 3a). Additionally, the capacity of lutein to We treated cells with lutein in the presence and the activate RARs was investigated by SEAP reporter absence of citral,34) an inhibitor of retinal dehydrogen- assay. Lutein significantly increased RARE-driven ases (RALDHs) that converts retinal to retinoic acid, and SEAP activity in a reasonable range of concentrations estimated hyaluronan production. Citral did not affect to stimulate hyaluronan production in a dose-dependent retinoic acid-induced hyaluronan production, whereas it manner (Fig. 3b). These results indicate that lutein or inhibited the effects of lutein as well as retinal (Fig. 4). its metabolites serve as a ligand for RARs in This suggests that a non-provitamin A carotenoid, lutein, keratinocytes. We have confirmed that zeaxanthin, a is oxidatively cleaved to aldehydes, followed by non-provitamin A carotenoid, and -carotene, a provi- RALDH-dependent oxidative formation of carboxylic tamin A carotenoid, also increased SEAP activity acids corresponding to retinoic acid, a metabolic process (Supplemental Fig. 2). These results suggest that non- similar to that of provitamin A carotenoids. provitamin A carotenoids such as lutein and zeaxanthin Lutein Activates the Retinoic Acid Receptor in Keratinocytes 1285

800 n.s. Lutein ** Monooxygenase? *** 600 Aldehydes n.s. Aldehyde 400 dehydrogenase Citral

200 Carboxylic acids

Hyaluronan content (ng/well) Hyaluronan LE540

0 RARE transactivation Control RA RAL Lutein

Fig. 4. Citral Inhibited the Effect of Lutein on Hyaluronan Synthesis. HAS3 mRNA induction Following 30 min of incubation of human keratinocytes in the presence (open columns) and the absence (closed columns) of 20 mM Hyaluronan production citral, 10 nM retinoic acid (RA), 50 nM retinal (RAL), and 10 mM lutein were added to the culture and this was incubated for 48 h. Fig. 5. Possible Metabolic Transformation of Lutein in Keratinocytes. Hyaluronan content was measured as described in ‘‘Materials and Methods.’’ Values represent mean SD for three replicate assays. stimulate RAR-mediated transcription activity in the Significantly different from the value of cells not treated with lutein: p < 0:01, p < 0:001 (Dunnett’s test). keratinocytes. The conversion of astaxanthin to retinol in a murine model39) has been reported. More recently, Sangeetha Discussion et al. also reported that astaxanthin was converted to -carotene and retinol in retinol-deficient rats.30) This The present study provides for the first time evidence indicates that non-provitamin A taken orally can be that lutein, a non-provitamin A carotenoid, activates converted to retinoids, although the possibility of RARs in keratinocytes. We investigated the RAR intestinal bacteria contributing to conversion cannot be stimulation activity of non-provitamin A by monitoring excluded. We found that human epidermal keratinocytes keratinocyte hyaluronan production, because retinoic can convert lutein oxidatively to carboxylic acids acid can induce HAS3 mRNA followed by hyaluronan corresponding to retinoic acid, and that the metabolites production in human keratinocytes.25) Hyaluronan syn- act as an RAR ligand. The structure of lutein metabolites thesis induced by -carotene, a provitamin A, was as RAR ligands requires further investigation. inhibited by LE540, an RAR pan-antagonist. Since Of RAR families, it has been reported that RAR is RAR-isoforms are expressed in keratinocytes,35,36) all- predominantly expressed in the epidermis.36) Am80, a trans retinoic acid converted from -carotene in cells selective agonist for RAR and RAR ,40) induced can act as an RAR ligand to activate transcription. This hyaluronan synthesis in the same way as Ch55, a pan- indicates that -carotene involves retinoid signaling and agonist of RAR (Supplemental Fig. 3), suggesting that is metabolized in keratinocytes to retinoids with bio- RAR is not the only receptor responsible for all-trans logical activity. retinoic acid in keratinocytes. Under the same conditions, we found that LE540 LE540 reduced basal hyaluronan synthesis in the completely abolished lutein-dependent hyaluronan syn- keratinocytes. Bovine pituitary extract (BPE), a supple- thesis, to the control level. Furthermore, treatment with ment for keratinocyte growth, might influence RAR- lutein significantly increased RARE-driven transcript dependent hyaluronan synthesis because retinoic acid is activity, suggesting that lutein or its metabolites serve as a regulatory factor in pituitary function and is generated a ligand for RARs in keratinocytes. in the anterior pituitary gland.41) Matsumoto et al. have reported that lutein bound to Lutein and zeaxanthin are the predominant carote- RAR in a cell-free CoA-BAP system,37) and it seems noids found in the human retina, and they are known to likely that lutein itself can substitute for retinoic acid. In be essential dietary nutrients for macular pigment.42) order to ascertain which of lutein and its metabolites is They have also been to accumulate in the skin,43) and functioning as a ligand for RARs in keratinocytes, we have superior antioxidant properties. They have been estimated the effect of citral, an inhibitor of RALDH, on found to inhibit the auto-oxidation of cellular lipids44) lutein-dependent action. The RALDH protein has been and to protect cells from oxidant-induced damage.45) In observed throughout the epidermis with equivalent in vivo studies, oral intake of lutein and zeaxanthin had expression in all keratinocyte layers,38) and we con- protective effects against epidermal inflammation,46) firmed that RALDH3 mRNA was predominantly ex- photoaging, and photocarcinogenesis47) under UVB pressed among the isoforms of RALDHs in cultured exposure. Moreover, topical administration of these keratinocytes (data not shown). In the present study, we two carotenoids provided significant effects in terms of found that citral significantly decreased lutein-dependent skin elasticity and hydration, and induced reductions in hyaluronan synthesis. This suggests that lutein is skin lipid peroxidation.48) Taken together with our oxidatively cleaved to aldehydes by BCO1 or similar findings, it appears that not only the antioxidant activity monooxygenases, followed by RALDH-dependent oxi- of lutein and zeaxanthin but also RAR activation by dative formation of carboxylic acids corresponding to their metabolites might contribute to the protective retinoic acid (Fig. 5), a metabolic process similar to that process of various clinical effects. of provitamin A carotenoids. In sum, lutein metabolites In conclusion, the present study indicates that metab- rather than lutein itself act as an RAR ligand and olites of lutein in keratinocytes exhibit agonist activity 1286 T. SAYO et al. for RARs, inducing hyaluronan synthesis. A series of 23) Kakizaki I, Itano N, Kimata K, Hanada K, Kon A, Yamaguchi non-provitamin As can be substituted for retinoids, and M, Takahashi T, and Takagaki K, Arch. Biochem. Biophys., should be considered as potential candidates to improve 471, 85–93 (2008). 24) Ohtani T, Memezawa A, Okuyama R, Sayo T, Sugiyama Y, skin health. Inoue S, and Aiba S, J. Invest. Dermatol., 129, 1412–1420 (2009). Acknowledgments 25) Sayo T, Sakai S, and Inoue S, Skin Pharmacol. Physiol., 17, 77– 83 (2004). We thank Ms. Y. Ota for skillful technical assistance. 26) Takeda A, Morinobu T, Takitani K, Kimura M, and Tamai H, J. Nutr. Sci. Vitaminol. (Tokyo), 49, 69–72 (2003). References 27) Siegenthaler G, Saurat JH, and Ponec M, Biochem. J., 268, 371– 378 (1990). 1) Muto Y and Moriwaki H, J. Natl. Cancer Inst., 73, 1389–1393 28) Katsumura N, Okuno M, Onogi N, Moriwaki H, Muto Y, and (1984). Kojima S, Nutr. Cancer, 26, 203–208 (1996). 2) Mathews-Roth MM, Ann. NY Acad. Sci., 691, 127–138 (1993). 29) Rock CL, Kusluski RA, Galvez MM, and Ethier SP, Nutr. 3) Burton GW and Ingold KU, Science, 224, 569–573 (1984). Cancer, 23, 319–333 (1995). 4) Olson JA, J. Nutr. Sci. Vitaminol. (Tokyo), 39 Suppl, S57–65 30) Sangeetha RK and Baskaran V, Can. J. Physiol. Pharmacol., (1993). 88, 977–985 (2010). 5) Bendich A and Olson JA, FASEB J., 3, 1927–1932 (1989). 31) Rilla K, Pasonen-Seppanen S, Rieppo J, Tammi M, and Tammi 6) Olson JA, J. Nutr., 119, 105–108 (1989). R, J. Invest. Dermatol., 123, 708–714 (2004). 7) Redmond TM, Gentleman S, Duncan T, Yu S, Wiggert B, Gantt 32) Johnson-Wint B and Hollis S, Anal. Biochem., 122, 338–344 E, and Cunningham FX Jr, J. Biol. Chem., 276, 6560–6565 (1982). (2001). 33) Umemiya H, Fukasawa H, Ebisawa M, Eyrolles L, Kawachi E, 8) De Luca LM, Sly L, Jones CS, and Chen LC, Carcinogenesis, Eisenmann G, Gronemeyer H, Hashimoto Y, Shudo K, and 14, 539–542 (1993). Kagechika H, J. Med. Chem., 40, 4222–4234 (1997). 9) Chen LC, Sly L, Jones CS, Tarone R, and De Luca LM, 34) Kikonyogo A, Abriola DP, Dryjanski M, and Pietruszko R, Eur. Carcinogenesis, 14, 713–717 (1993). J. Biochem., 262, 704–712 (1999). 10) Kligman AM, Grove GL, Hirose R, and Leyden JJ, J. Am. Acad. 35) Darwiche N, Celli G, Tennenbaum T, Glick AB, Yuspa SH, and Dermatol., 15, 836–859 (1986). De Luca LM, Cancer Res., 55, 2774–2782 (1995). 11) Bryce GF, Bogdan NJ, and Brown CC, J. Invest. Dermatol., 91, 36) Fisher GJ and Voorhees JJ, FASEB J., 10, 1002–1013 (1996). 175–180 (1988). 37) Matsumoto A, Mizukami H, Mizuno S, Umegaki K, Nishikawa 12) Fisher GJ, Datta S, Wang Z, Li XY, Quan T, Chung JH, Kang S, J, Shudo K, Kagechika H, and Inoue M, Biochem. Pharmacol., and Voorhees JJ, J. Clin. Invest., 106, 663–670 (2000). 74, 256–264 (2007). 13) Fisher GJ, Datta SC, Talwar HS, Wang ZQ, Varani J, Kang S, 38) Cheung C, Smith CK, Hoog JO, and Hotchkiss SA, Biochem. and Voorhees JJ, Nature, 379, 335–339 (1996). Biophys. Res. Commun., 261, 100–107 (1999). 14) Tsukahara K, Takema Y, Fujimura T, Moriwaki S, Kitahara T, 39) Matsuno T, Pure Appl. Chem., 63, 81–88 (1991). Imayama S, and Imokawa G, Br. J. Dermatol., 140, 1048–1053 40) Fukasawa H, Iijima T, Kagechika H, Hashimoto Y, and Shudo (1999). K, Biol. Pharm. Bull., 16, 343–348 (1993). 15) Lundin A, Berne B, and Michaelsson G, Acta Derm. Venereol., 41) Fujiwara K, Kikuchi M, Takigami S, Kouki T, and Yashiro T, 72, 423–427 (1992). Cell Tissue Res., 329, 321–327 (2007). 16) Tammi R and Tammi M, J. Cell. Physiol., 126, 389–398 (1986). 42) Neuringer M, Sandstrom MM, Johnson EJ, and Snodderly DM, 17) Tammi R, MacCallum D, Hascall VC, Pienimaki JP, Hyttinen Invest. Ophthalmol. Vis. Sci., 45, 3234–3243 (2004). M, and Tammi M, J. Biol. Chem., 273, 28878–28888 (1998). 43) Wingerath T, Sies H, and Stahl W, Arch. Biochem. Biophys., 18) Pasonen-Seppanen S, Karvinen S, Torronen K, Hyttinen JM, 355, 271–274 (1998). Jokela T, Lammi MJ, Tammi MI, and Tammi R, J. Invest. 44) Zhang LX, Cooney RV, and Bertram JS, Carcinogenesis, 12, Dermatol., 120, 1038–1044 (2003). 2109–2114 (1991). 19) Bourguignon LY, Ramez M, Gilad E, Singleton PA, Man MQ, 45) Martin KR, Failla ML, and Smith JC Jr, J. Nutr., 126, 2098– Crumrine DA, Elias PM, and Feingold KR, J. Invest. Dermatol., 2106 (1996). 126, 1356–1365 (2006). 46) Gonzalez S, Astner S, An W, Goukassian D, and Pathak MA, 20) Tobiishi M, Sayo T, Yoshida H, Kusaka A, Kawabata K, J. Invest. Dermatol., 121, 399–405 (2003). Sugiyama Y, Ishikawa O, and Inoue S, J. Dermatol. Sci., 64, 47) Astner S, Wu A, Chen J, Philips N, Rius-Diaz F, Parrado C, 31–38 (2011). Mihm MC, Goukassian DA, Pathak MA, and Gonzalez S, Skin 21) Weigel PH, Hascall VC, and Tammi M, J. Biol. Chem., 272, Pharmacol. Physiol., 20, 283–291 (2007). 13997–14000 (1997). 48) Palombo P, Fabrizi G, Ruocco V, Ruocco E, Fluhr J, Roberts 22) Sayo T, Sugiyama Y, Takahashi Y, Ozawa N, Sakai S, Ishikawa R, and Morganti P, Skin Pharmacol. Physiol., 20, 199–210 O, Tamura M, and Inoue S, J. Invest. Dermatol., 118, 43–48 (2007). (2002).