Spectral Responses of Melanin to Ultraviolet A Irradiation

Spectral Responses of Melanin to Ultraviolet A Irradiation Hao Ou-Yang, Georgios Stamatas, and Nikiforos Kollias Models and Methods Development, Johnson and Johnson Consumer and Personal Products Worldwide, Skillman, New Jersey, USA The purpose of this investigation was to study the ultraviolet A-induced effects on melanin pigmentation both in an in vitro model system and in vivo. Ultraviolet-Vis absorbance spectra of L-3,4-dihydroxyphenylalanine-melanin solutions at different concentrations were measured before and after ultraviolet A exposure (10–120 J per cm2). The difference spectra reveal that following ultraviolet A exposure the absorbance increases exponentially from 800 nm to 450 nm accompanied by a prominent decrease of absorbance in the ultraviolet A range. This change of spectral features depends on both ultraviolet A doses and melanin concentrations. The photo-bleaching effect observed in the ultraviolet A range also depends on oxygen. Human subjects were irradiated with ultraviolet A (40–80 J per cm2) on their back and diffuse reflectance spectra were collected at both irradiated and untreated sites. The absorption spectra of ultraviolet A-induced pigment were calculated as the difference of the two. The ultraviolet A-induced pigment in vivo has similar spectral characteristics and dose dependency as the in vitro system. Photo-oxidation of pheomelanin solutions presents distinctively different spectral and dose–response characteristics from eumelanin. After ultraviolet A irradiation pheomelanin absorbance decreases both in the visible and the ultraviolet A range. We conclude that irradiation with ultraviolet A induces significant photochemical alterations in the skin witnessed by increased photoprotection in the visible spectral range and reduced protection in the ultraviolet A range. We suggest that soluble melanin plays an important part in ultraviolet A-induced pigment in skin and two distinct absorption mechanisms of melanin may be involved in ultraviolet A photo-oxidation. We also propose that eumelanin and pheomelanin could be differentiated according to their spectral responses to ultraviolet A irradiation. Key words: eumelanin/immediate pigment darkening/pheomelanin/pigmentation/spectroscopy. J Invest Dermatol 122:492 – 496, 2004 increasing UVA doses (41 J per cm2), it can last longer and eventually turns to persistent pigment darkening. Spectroscopic evidence has shown that UVA-induced pigment (including IPD) is different from UVB-induced pigment or native epidermal melanin (Kollias and Baqer, 1987; Kollias, 1992, 1995). Furthermore, UVA-induced pigment has been shown to provide no protection from UVB radiation (Honigsmann et al, 1986; Gange, 1988; Kollias et al, 1991). IPD is an interesting and unique part of the UVA-induced pigment reaction in human skin. The transient nature of IPD makes it difficult to characterize and is inconsistent with the neomelanogenesis mechanism. Honigsmann et al (1986) studied the biochemistry of this phenomenon and concluded that: (1) the production of IPD does not depend on the structural and functional integrity of the melanocyte cytoskeletal apparatus; (2) it does not lead to an improved melanosome transfer to keratinocytes; and (3) it is not confined to viable skin. Therefore, it was suggested that IPD is formed through oxidation or polymerization of melanin precursors through an ‘‘acellular’’ (meaning that it does not require activation of any cellular process) photochemical reaction. It is important to study the difference between UVA pigment and UVB pigment in vivo in a noninvasive manner. Previous spectroscopic studies (Kollias and Baqer, 1987) concluded that there are two types of absorption mechanisms for human epidermal melanin throughout the UVvisible wavelength range. One mechanism is emphasized Native human epidermal melanin includes eumelanin and sulfur-containing pheomelanin. The exact chemical structures of these two types of melanin are still unknown, probably due to the complication of copolymerization and numerous postpolymerization modifications (Ito, 1986; Kollias et al, 1991). Whereas eumelanin has been found in almost every type of human skin, pheomelanin is found to be abundant mostly in fair-skinned persons with red hair (Hunt et al, 1995; Vincensi et al, 1998). The photoprotective properties of human epidermal melanin in skin have been well documented (Kollias et al, 1991). It absorbs both ultraviolet (UV) and visible light. The absorption increases linearly in the range of 720 to 620 nm and then exponentially toward shorter wavelengths (300– 600 nm). Upon UVB exposure, delayed pigment darkening appears in the skin. This pigment indicates neomelanogenesis in the epidermis and it has the same spectral characteristics as the native epidermal melanin. In contrast, following low-dose (o10 J per cm2) of UVA exposure the skin undergoes an immediate pigment darkening (IPD) reaction (Hausser and Vahle, 1927; Pathak et al, 1962). This pigment is transient and fades away in a few minutes. With Abbreviations: DOPA, L-3,4-dihydroxyphenylalanine; IPD, immediate pigment darkening; UVA, 320–400 nm part of the ultraviolet radiation; ultraviolet-Vis, ultraviolet and visible light. Copyright r 2004 by The Society for Investigative Dermatology, Inc. 492 122 : 2 FEBRUARY 2004 SPECTRAL RESPONSES OF MELANIN TO UVA Results The typical absorbance spectra of soluble eumelanin at three different dilutions (1
, 3
, 30
) and the corresponding picture are shown in Fig 1. The features of the spectrum include a linear increase of absorbance from 800 to 600 nm and an exponential increase of absorbance from 600 to 300 nm. The spectrum of soluble melanin resembles the in vivo epidermal melanin spectra. The relative amount of melanin, as demonstrated previously (Kollias and Baqer, 1985, 1986), can be determined using the slope in the 620 to 750 nm range. The difference of absorbance spectra measured before and immediately after UVA exposure (termed here ‘‘difference spectra’’) is shown in Fig 2(a) to illustrate the effect of UVA exposure on the soluble eumelanin. The spectra exhibit an exponential increase in absorbance from 800 to 450 nm accompanied by a prominent decrease of absorbance in the UVA range for all three different dilutions after 70 J per cm2 of UVA. Similar behavior in difference absorbance spectra was also observed for the same eumelanin concentration exposed to increasing UVA doses, as evident in Fig 2(b). The degree of absorbance increase in the visible wavelength range and the level of sharp absorbance decrease in the UVA are dependent on both UVA dose and soluble a b 5.0 4.5 4.0 Dilution 30x 3x 1x 3.5 Absorbance more in the visible and the other more in the UV wavelength range. Change in melanin content or structure can directly affect the absorption spectra. Therefore, the spectroscopic features of melanin absorption may be able to provide useful information about the uniqueness of UVA-induced pigment in skin. The spectral characteristics of pigment induced by UVA exposure have been measured in vivo and reported previously (Kollias and Baqer, 1987; Kollias, 1995). The absorbance increases exponentially from 800 nm to 450 nm accompanied by a prominent decrease of absorbance in the UVA range. As the formation of UVA-induced pigment is a passive process (at least in the beginning) and IPD reaction is very unstable, weak, and difficult to measure in vivo, the question arises if we can simulate the UVAinduced pigment formation in an in vitro model system and study the spectroscopic features of this reaction in a controlled environment. As a heterogeneous biopolymer, human epidermal melanin exists in vivo both in particulate and in soluble forms. Comparison of the absorbance spectra of particulate melanin with that of soluble melanin suggested the two forms have different absorption mechanisms. Melanin powder has a higher absorbance in the visible than soluble melanin. The absorbance of particulate melanin increases almost linearly from 800 to 400 nm, whereas the absorbance of soluble melanin shows an exponential increase from 600 to 300 nm (Kollias and Baqer, 1987). The spectrum of soluble melanin resembles more closely the spectrum of native melanin in skin. Therefore, the study of the response of soluble melanin to UVA in vitro may shed some light on the in vivo UVA-induced pigment reaction. Pheomelanin can be decomposed by UV light in vitro (Chedekel et al, 1977; Poh Agin et al, 1980). This may account for the severity of skin reactions to UV radiation of light-skinned, red-haired persons with freckles, and poor tanning ability. It is of interest to investigate the photostability of pheomelanin to UVA irradiation in the model system of pheomelanin in solution, as it is not clear whether pheomelanin contributes to the UVA-induced pigment reaction in skin. Furthermore, it remains unknown how the absorption mechanisms of eumelanin and pheomelanin change after UVA irradiation since the photodegradation of pheomelanin was studied only in the UV part of the spectrum in previous reports. 3.0 2.5 2.0 1.5 1.0 0.5 0.0 400 500 600 700 Wavelength (nm) 30x 3x 1x b 0.04 0.04 0.02 0.02 Difference Abs. Difference Abs. 800 Figure 1 In vitro synthesized eumelanin solutions at different concentrations. Soluble eumelanin was synthesized and diluted according to text. (a,b) The absorption spectra and corresponding color pictures. a Figure 2 The effect of UVA irradiation on eumelanin solutions: concentration and dose dependency. (a) Shows the difference spectra of three eumelanin solutions a different dilutions but subjected to the same UVA dose (70 J per cm2). (b) Shows the difference spectra of three eumelanin solutions of the same concentration (30
dilution) but subjected to different UVA doses. 493 0.00 −0.02 Dilution 30x 3x 1x 0.00 UVA dose 2 10J/cm 2 20J/cm 2 40J/cm 2 70J/cm −0.02 −0.04 −0.04 300 400 500 600 700 Wavelength (nm) 800 300 400 500 600 700 Wavelength (nm) 800 494 OU-YANG ET AL THE JOURNAL OF INVESTIGATIVE DERMATOLOGY 0.12 UVA dose 0.20 0.10 36J/cm2 59J/cm2 82J/cm2 Sealed Exposed to air 0.08 Apparent Abs 0.15 Diff Abs. 0.06 0.04 0.10 0.02 0.05 0.00 -0.02 -0.04 300 0.00 300 400 500 600 Wavelength (nm) 700 800 Figure 3 The effects of UVA on eumelanin solution: oxygen dependency. The difference spectra of two eumelanin solutions of same concentration and subject to the same UVA dose (30 J per cm2) are compared. During the irradiation, the oxygen supply to one was limited (sealed), whereas the other one was exposed to air. eumelanin concentration. Interestingly, the photo-bleaching of the pigment in the UVA range accompanies the increase in the visible wavelength range in difference spectra. In one experiment, the eumelanin solution was sealed during the irradiation. The resulting difference spectrum for this sealed eumelanin solution is shown in Fig 3 and compared with the difference spectrum of another eumelanin solution of the same concentration and subjected to the same UVA dose with access to oxygen during the irradiation (for 3 h). The difference is evident showing more photo-bleaching with the melanin solution exposed to air. This result implies that the formation of UVA pigment may be a photo-oxidation reaction and may involve oxygen. The UVA-induced pigments in human skin were measured by diffuse reflectance spectroscopy before and immediately after UVA exposure following the published procedure (Kollias and Baqer, 1987). UVA doses of 30 to 80 J per cm2 were used to generate measurable and reproducible pigment in skin. Erythema was observed for some panelists immediately following exposure. The induced pigment persisted in skin at least for 1 wk. Apparent absorbance spectra, which are defined as the logarithm of the ratio between remittance spectra measured before and immediately after UVA exposure, emphasize the pigment changes caused by UVA. Representative spectra are shown in Fig 4. The in vivo apparent absorbance spectra of UVAinduced pigment are similar to the difference spectra of the in vitro eumelanin solution following exposure to UV (Fig 2b). The dose–response of UVA-induced pigment also shows a similar trend as the in vitro system. Note that the absorption bands at about 540 nm and 575 nm are due to oxyhemoglobin indicating the existence of slight erythema. Pheomelanin solutions were also prepared by adding cystein to the system. The weight percentage of sulfur was checked before irradiation by elemental analysis and was 400 500 600 Wavelength (nm) 700 Figure 4 Diffuse reflectance spectra of in vivo UVA-induced pigment. Apparent absorbance spectra of skin pigment as a result of UVA exposure were shown for three doses. These spectra were measured on the back of a skin type III person. Hemoglobin bands (500–600 nm) indicating slight erythema can be observed. Marked similarities exist between this figure and Fig 2(b). determined to be about 7%. The absorbance spectra of pheomelanin are quite similar to those of eumelanin (Menon et al, 1983). After being subjected to the same UVA irradiation procedure, however, the pheomelanin solution shows a dramatic difference compared with the eumelanin solution. As shown in Fig 5(a) for pheomelanin solutions of different concentrations exposed to same dose and in Fig 5(b) for pheomelanin solutions of the same concentration but exposed to different UVA doses, the difference spectra have two regions where the absorbance decreases, one in the UVA and the other in the visible wavelength range (500–600 nm). UVA seems to be able to photodegrade both absorption mechanisms in pheomelanin solution and it is consistent with the poor ability of photoprotection observed for people with light skin and red hair (more pheomelanin in skin) (Azizi et al, 1988). Discussion It has been proposed that there are two distinct absorption mechanisms for skin melanin (Kollias and Baqer, 1987) and this provides a reasonable interpretation for the diffuse reflectance spectra recorded for UVB-induced pigment and UVA-induced pigment (including IPD). The two mechanisms change independently of each other (Kollias and Baqer, 1987). The UVB-induced pigment, as well as native epidermal pigment, involves increases of both mechanisms, whereas the UVA-induced pigment appears as the increase of one mechanism (in the visible wavelengths) at the expense of the other (in the UVA wavelengths). The consistency between the in vivo result and in vitro measurement highlights that the soluble melanin is an important contributor to the UVA-induced pigment. The decrease of absorbance in the UV wavelength range 122 : 2 FEBRUARY 2004 SPECTRAL RESPONSES OF MELANIN TO UVA a b 0.04 Dilution 30x 3x 1x 0.02 UVA dose 2 10J/cm 2 30J/cm 2 60J/cm 0.02 Difference Abs. 0.00 −0.02 −0.04 0.00 −0.02 −0.06 −0.08 −0.10 300 −0.04 400 500 600 700 Wavelength (nm) following UVA exposure offers a direct explanation on the lack of any UV photoprotection of UVA-induced pigment. At the present time, there is no information concerning which molecules or chromophores these two absorption mechanisms may represent; however, it has been shown that the absorption mechanism in the visible wavelength range corresponds to the skin melanin concentration (Kollias and Baqer, 1986). Particulate melanin shows a significant higher absorbance in the visible range (Kollias and Baqer, 1986). Histologic staining for melanin applies only to the insoluble melanin and correlates strongly with epidermal melanin content measured by high-performance liquid chromatography (Tadokoro et al, 2003). Therefore, it has been suggested that the absorption mechanism in the visible wavelength range may be related to the insoluble form of melanin (Kollias and Baqer, 1986; Simon 2000; Nofsinger et al, 2002a, b). This form of melanin has high molecular weight and is the major contributor to the pigment of native skin. The absorption mechanism in the visible wavelength range, however, is not limited to insoluble melanin. We report in this study that difference absorption spectra of soluble melanin to UVA irradiation demonstrate a slight increase of absorbance in the visible wavelength range (Fig 2), indicating that soluble melanin may also contribute to the absorption in this wavelength range. In addition, the absorbance spectra of two eumelanin solutions prepared by centrifugal filtering of the original solution through a molecular weight based membrane (MW limit ¼ 10,000) are shown in Fig 6. Both spectra were normalized to 400 nm to compensate the differences due to concentration. Melanin solution with higher molecular weight has a higher absorbance in the visible wavelengths and this result is consistent with a previous report (Simon, 2000) and the above suggestion. It is not clear what is the other form of absorption, which accounts for the absorption characteristics in the UVA range. A previous study has shown that this form contributes only a small part to the observed epidermal pigment in skin (Kollias and Baqer, 1987). The photooxidation and polymerization of melanin were well documented in the interaction between UVA and skin pigment (Sarna et al, 1980) (Fig 3). It is possible that one form of melanin with low molecular weight or a melanin polymer 800 300 400 500 600 700 Wavelength (nm) 800 precursor transforms itself to the other form (with higher molecular weight). IPD reaction is reversible up to some extent, but the reversibility of the IPD reaction may require the degradation of polymers, which may consume energy and cellular enzyme activities as shown by Honigsmann et al (1986). When the UVA dose surpasses the threshold, the pigment becomes irreversible and is still different from native or UVB-induced pigment at least for the time period immediately following the exposure. Preliminary data show that this pigment can transform itself to native epidermal pigment indicated by the increase of UV absorbance from negative to positive in the apparent absorbance spectra (data not shown). It remains to be answered whether the same recovery process (as in IPD) or a different biochemical process is involved for that transformation. In summary, we studied the spectral responses of melanin to UVA exposure both in an in vitro solution system and in vivo. We found that UVA induced significant 4.0 3.5 Absorbance (Normalized to 400nm) Difference Abs. Figure 5 The effect of UVA irradiation on pheomelanin solutions: concentration and dose dependency. (a) Shows the difference spectra of three pheomelanin solutions with different dilutions but subjected to the same UVA dose (60 J per cm2). (b) Shows the difference spectra of three pheomelanin solutions of the same concentration (30
dilution) but subjected to different UVA doses. Marked differences exist between this figure and Fig 2. 495 MW<10K MW>10K 3.0 2.5 2.0 1.5 1.0 0.5 0.0 300 400 500 600 Wavelength (nm) 700 800 Figure 6 The absorbance spectra of two eumelanin solutions with different molecular weights. Solid line represents eumelanin solution with MWo10,000, whereas the dotted line represents eumelanin solution with MW410,000. 496 OU-YANG ET AL photochemical alterations in skin pigment and we conclude that soluble melanin is an important contributor to the UVAinduced pigment in skin. The melanin solution can be used as a model to study the nonenzymatic interaction between UVA and skin pigment. The interaction depends on oxygen and involves two distinct absorption mechanisms. The UVA-induced pigment is spectrally different from native or UVB-induced pigment. It is characterized by an increase in absorption in the visible and a decrease in the UVA range. We proposed that these two absorption mechanisms correspond to melanin polymers of high molecular weight existing (both in a particulate form and a soluble form) and low molecular weight (in a soluble form). Finally, the photostability of pheomelanin is different from that of eumelanin. Both absorption mechanisms tend to drop following UVA exposure for pheomelanin, indicating a different mechanism of photodegradation. We therefore suggest that eumelanin and pheomelanin could be differentiated according to their spectral responses to UVA irradiation. Materials and Methods Chemicals All chemicals were from Sigma (St Louis, Missouri). Light source The UVA source used in this study was a Hamamatsu LightingCure 200 unit (Hamamatsu, Japan) with a 150 W Xenon lamp and a UV reflector (300–450 nm with maximum at 365 nm). The source was further filtered with Schott (Yonkers, New York) UG11 1 mm and WG335 (3 mm) glass filters. The spectral distribution of this source was characterized by a spectroradiometer (Optronic 752, Optronic Laboratories, Orlando, Florida) with the maximum output at 365 nm. The output of this source was routinely measured with a thermopile (Oriel 71140, calibrated by The Eppley Laboratory, Inc. Newport, RI). Less than 10% of the lamp output was below 340 nm. Synthesis of melanin Eumelanin and pheomelanin were synthesized in vitro following the standard method (Ito et al, 1980; Ito, 1989). Briefly, 20 mg L-3,4-dihydroxyphenylalanin (DOPA) was dissolved in 40 mL phosphate-buffered saline buffer at pH 7.4 (equivalent moles of cystein were added into the solution for pheomelanin). Mushroom tyrosinase (2 mg) was added and then the solution was exposed to air for 5 d to complete the oxidation. The final oxidized products were centrifuged to remove the insoluble solid. Some of the soluble melanin were acidified to pH 3 for precipitation and then centrifuged and lyophilized for element analysis (Robertson Microlit Laboratories, Inc., Madison, New Jersey). The melanin solution was further separated by their molecular weights using centrifugal filter devices (Millipore, Bedford, Massachusetts; MW threshold ¼ 10,000). UVA irradiation The melanin solutions were diluted (two to 30 times) exposed to UVA radiation (10–120 J per cm2 with a fluence rate of 3 mW per cm2) in UV-transparent cuvettes. For the oxygen dependency experiment, the cuvette was covered with a plastic top during the exposure to reduce the oxygen supply from the environment. The solution was examined before and after UVA exposure with a Cary-3 UV-Vis spectrophotometer (Varian, Walnut Creek, California). The experiments were repeated three times for different batches of synthetic melanin. Following written informed consent, five human subjects of skin types IV to V were irradiated with UVA (40–80 J per cm2 with fluence rate of 50 mW per cm2) on their backs. The diffuse reflectance spectra were collected for both the irradiated sites and untreated sites immediately following exposure and apparent absorbance spectra were calculated according to the published procedure (Kollias and Baqer, 1985, 1986). The in vivo experiments were conducted at Al Sabah THE JOURNAL OF INVESTIGATIVE DERMATOLOGY Hospital, Ministry of Public Health, Kuwait, under the supervision of MM Selim, MD. DOI: 10.1046/j.0022-202X.2004.22247.x Manuscript received August 6, 2003; accepted for publication September 12, 2003 Address correspondence to: Georgios Stamatas, 199 Grandview Road, Skillman, NJ 08558, USA. Email: [email protected] References Azizi E, Lusky A, Kushelevsky AP, Schewach-Millet M: Skin type, hair color, and freckles are predictors of decreased minimal erythema ultraviolet radiation dose. J Am Acad Dermatol 19:32–38, 1988 Chedekel MR, Post PW, Deibel RM, Klause M: Photodestruction of pheomelanin. Photochem Photobiol 26:651–653, 1977 Gange RW: Comparison of pigment responses in human skin to UVB and UVA radiation. Prog Clin Biol Res 256:475–485, 1988 Hausser KW, Vahle W: Sonnenbrand und Sonnenbräunung. Wissenschaftliche Veröffentlichungen des Siemens Konzerns 6: 1927, pp 101–120 (English translation: Sunburn and suntanning. In: Urbach F (eds). The Biological Effects of Ultraviolet Radiation (with Emphasis on Skin). Oxford: Pergamon, 1969; p. 3–21 Honigsmann H, Schuler G, Aberer W, Romani N, Wolff K: Immediate pigment darkening phenomenon: A reevaluation of its mechanisms. J Invest Dermatol 87:648–652, 1986 Hunt G, Kyne S, Ito S, Wakamatsu K, Todd C, Thody A: Eumelanin and pheomelanin contents of human epidermis and cultured melanocytes. Pigment Cell Res 8:202–208, 1995 Ito S: Optimization of conditions for preparing synthetic pheomelanin. Pigment Cell Res 2:53–56, 1989 Ito S: Reexamination of the structure of eumelanin. Biochem Biophys Acta 883:155–161, 1986 Ito S, Novellino E, Chioccara F, Misuraca G, Prota G: Co-polymerization of dopa and cysteinyldopa in melanogenesis. Experimenta 36:822–823, 1980 Kollias N: The spectroscopy of human melanin pigmentation. In: Zeise L, Chedekel MR, Fitzpatrick TB (eds). Melanin: Its Role in Human Photoprotection. Valdenmar Publishing Co., Overland Park, Kansas, 1995; p. 31–38 Kollias N: UVA melanogenesis-spectral observations. In: Urbach F (ed). Biological Responses to UVA Radiation. Valdenmar Publishing Co., Overland Park, Kansas, 1992; p. 151–157 Kollias N, Baqer A: Spectroscopic characteristics of human melanin in vivo. J Invest Dermatol 85:38–42, 1985 Kollias N, Baqer A: On the assessment of melanin in human skin in vivo. Photochem Photobiol 43:49–54, 1986 Kollias N, Baqer AH: Absorption mechanism of human melanin in the visible, 400–720 nm. J Invest Dermatol 89:384–388, 1987 Kollias N, Sayer RM, Zeise L, Chedekel MR: Photoprotection by melanin. J Photochem Photobiol B Biol 9:135–160, 1991 Menon IA, Persad S, Haberman HF, Kurian CJ: A comparative study of the physical and chemical properties of melanins isolated for human black and red hair. J Invest Dermatol 80:202–206, 1983 Nofsinger JB, Weinert EE, Simon JD: Establishing structure-function relationships for eumelanin. Biospectroscopy 67:302–305, 2002a Nofsinger JB, Liu Y, Simon JD: Aggregation of eumelanin mitigates photogeneration of reactive oxygen species. Free Radic Bio Med 32:72–730, 2002b Pathak MA, Reily FJ, Fitzpatrick TB, Curwen WL: Melanin formation in human skin induced by long-wave ultraviolet and visible light. Nature 193: 148–150, 1962 Poh Agin P, Sayre RM, Chedekel M: Photodegradation of pheomelanin: An in vitro model. Photochem Photobiol 31:359–362, 1980 Sarna T, Duleba A, Korytowski W, Swartz H: Interaction of melanin with oxygen. Arch Biochem Biophys 200:140–148, 1980 Simon JD: Spectroscopic and dynamic studies of the epidermal chromophores trans-urocaic and eumelanin. Acc Chem Res 33:307–313, 2000 Tadokoro T, Kobayashi N, Zmudzka B, et al: UV-induced DNA damage and melanin content in human skin differing in racial/ethnic origin. FASEB J 17:1177–1179, 2003 Vincensi MR, d’Ischia M, Napolitano A, et al: Pheomelanin versus eumelanin as a chemical indicator of ultraviolet sensitivity in fair-skinned subjects at high risk for melanoma: A pilot study. Melanoma Res 8:53–58, 1998