Photodermatol Photoimmunol Photomed 2001; 17: 266–271 Copyright C Munksgaard 2001Printed in Denmark ¡ All rights reservedMunksgaard ¡ Copenhagen

ISSN 0905-4383

Single dose radiation is more effective for the UV-induced activation andproliferation of melanocytes than fractionated dose radiation

H. T. An, J. Y. Yoo, M. K. Lee, M. H. Shin, G. E. Rhie, J. Y. Seo, J. H. Chung, H. C. Eun, K. H. ChoDepartment of Dermatology, Seoul National University College of Medicine, and Laboratory for Cutaneous Aging Research, Clinical Research

Institute, Seoul National University Hospital, Seoul, Korea

Purpose: To establish whether the effect of fractionat-ing radiation modifies the effects of ultraviolet (UV)radiation on epidermal melanocytes, we compared theclinical and histological effects of single high doseradiation against repeated intermediate to low doseradiation on epidermal melanocytes.Methods: Three minimal erythema UV doses (MED)were administered to three sites on the buttocks ofhealthy volunteers. One site was irradiated with 0.5MED UV every day for 6 consecutive days, anothersite was irradiated with 1 MED UV every second day,and a third site received a single dose of radiation with3 MED UV. The treatment was replicated on the otherbuttock. For the evaluation of UV-induced erythemaand pigmentation, erythema and melanin indices weremeasured at 2 and 14 days post-irradiation. For pur-poses of histological evaluation, tissue specimenstaken from each irradiated site at 2 and 14 days post-irradiation and were stained with monoclonal anti-

The obvious clinical changes due to UV radiation areerythema and pigmentation. Exposure of the skin to

adequate doses of UV results in erythema (inflammation)and this response is followed by hyperpigmentation (tan-ning), with maximal intensity shown between 10 days and3 weeks (1). Generally, the degree of erythema after UVradiation is determined by the total energy administered.As a consequence, doubling the irradiance and halvingthe irradiation time will lead to the same degree of ery-thema (2). Similarly, the degree of UV-induced pigmen-tation correlates with the severity of the prior erythemainduced by UV radiation (3). Thus, the same degree ofpigmentation can be expected as long as the total doseadministered is identical.

In this study, we examined whether single high dose,and repeated intermediate to low dose, UV irradiation hasthe same effects on epidermal melanocytes if the total ad-

266

bodies against Mel-5, HMB-45 and tyrosinase. Fon-tana-Masson silver staining, DOPA staining and splitDOPA reactions were also performed.Results: At 14 days post-irradiation, UV radiation in-duced melanocyte activation, proliferation and mel-anogenesis in proportion to the radiation dose ad-ministered to each fraction. The most prominent re-sponses were observed after single high doses ofradiation.Conclusion: When the total administered dose is iden-tical, fractionation of radiation dose diminishes theeffects of UV radiation on epidermal melanocytes.Furthermore, long, uninterrupted doses of UV radi-ation were found to more effective in inducing mel-anogenesis and melanocyte activation.

Key words: fractionation; melanocytes; ultravioletradiation.

ministered dose is identical. For this purpose, we exploredthe degree of melanocyte activation, proliferation andmelanogenesis after 3 MED UV radiation with single ormultiple fractions.

Material and methodsSubjects and UV sourceThree healthy male Koreans aged 20–26 years, with skinphototypes varying from IV to V according to the Fitzpa-trick-Pathak classification (4), participated in this study.None of them had a skin disease or a history of photosen-sitivity. After informed consent had been obtained, 12 dif-ferent doses of UV (40–140 mJ/cm2) were administered intwo parallel rows on the lower backs of the subjects withan irradiance of 1 mW/cm2. UV 800K (Waldmann,Schwenningen, Germany) with four unfiltered Philips TL

Single dose vs. fractionated radiation effects on melanocytes

20 W/12 lamps emitting UV radiation ranging from 285nm to 350 nm with maximum spectrum between 310–315nm were used as optical UV sources. Sites of MED wereidentified at 24 h post-irradiation.

UV exposure schedulesThree sites were irradiated with a total of 3 MED UV,each site being 1¿1 cm in size, in a single line on thebuttock of the three subjects. The irradiance was 1 mW/cm2. One site was irradiated every day with a low doseUV (0.5 MED) for a period of 6 days. A second site wasirradiated with an intermediate dose of UV (1 MED)three times on alternate days of a 6-day period. The thirdsite received a single high dose of UV radiation (3 MED)(Fig. 1). The procedure was replicated on the other but-tock. Each subject was thus irradiated on six differentsites, three on each buttock.

Measuring erythema index and melanin indexTwo and 14 days after the last irradiation session, onceUV-induced inflammation and pigmentation was at itsmaximum, the degree of erythema and pigmentation of

Fig. 1. Schematic summary of UV radiation schedules.UV was exposed to three sites on the buttock of eachsubject. One site received six consecutive daily appli-cations of 0.5 MED UV, another site received three appli-cations of 1 MED UV on alternate days, and the thirdsite received a single dose of 3 MED UV. The experimentwas replicated on the other buttock for the evaluation ofthe effects of UV on melanocytes at 2 and 14 days post-irradiation.

267

the UV-irradiated and unradiated control skin were meas-ured using a Derma-SpectrometerA (Cortex technology,Hadsund, Denmark). This instrument measures both theerythema and melanin indices by indicating the relativecontent of red blood cells in the upper dermis and melaninin the epidermis, respectively (5, 6). Each value was anaverage of three recordings and expressed as an increaseof value after radiation relative to that of unradiated skin.

The relationship between the degree of erythema at 2days post-irradiation and pigmentation at 14 days post-irradiation was assessed using Pearson’s correlation coef-ficient. A value of P,0.01 was considered indicative ofstatistical significance.

Tissue collectionTwo days after the last radiation event, biopsy specimensof irradiated and unradiated skin samples were takenfrom one side of the buttocks with an 8 mm punch. Four-teen days after the last radiation event, biopsy specimensof irradiated skin were taken from the other side of thebuttocks. Each specimen was divided into two pieces anda piece of each specimen was embedded immediately inTissue Tek (Miles, Torrance, CA, USA) and frozen inliquid nitrogen for histological evaluation. These speci-mens were stored at ª20 æC until they were used. Theother half of the specimens were used in split DOPA reac-tions.

Histology and histochemistryVertical frozen sections were prepared for histologicalevaluation. Frozen sections 5 mm in thickness were stainedwith hematoxylin and eosin. Melanin and melanocyteswere observed using the Fontana-Masson method and byincubating samples in L-3,4-dihydroxyphenylalanine(DOPA), respectively.

ImmunohistochemistryImmunohistochemical studies were performed on 5 mmfrozen sections using labeled streptavidin-biotin (DAKOLSAB System, Glostrup, Denmark). The following pri-mary antibodies were used:O monoclonal antibody against Mel-5 (Signet, Dedham,

MA, USA), recognizing a 75 kDa glycoprotein usuallyexpressed in normal melanocytes (7);

O monoclonal antibody against HMB-45 (Enzo, NewYork, NY, USA), which appears on the immature mel-anosomes during melanogenesis (8);

O anti-tyrosinase monoclonal antibody (Neomarkers,Fremont, CA, USA).

Immunolocalization of these primary antibodies was per-formed with 3-amino-9-ethylcarbazole (ACE) as a sub-strate. Counterstaining was performed using Mayer’shematoxylin.

An et al.

Assessment of immunohistochemical studiesThe number of cells expressing HMB-45 and tyrosinase/100 basal cells was measured using a color image analysissystem (BMI PLUSA image analysis system equippedwith a Toshiba 3 CCD color video camera coupled to aOlympus BH-2 microscope). An average of five recordingsfrom different fields was calculated as a multiple relativeto the value obtained for unradiated control skin samples.

Split DOPA reactionThe number of epidermal melanocytes induced by singleor repetitive fractionated UV irradiation events wasstudied using the combined skin splitting technique andDOPA incubation first introduced by Staricco & Pinkus(9). The epidermis was separated from the dermis after a2-h incubation at 37 æC in a 2 N sodium bromide solution.The epidermis was then incubated in a 0.1% DOPA solu-tion for 6 h at 37 æC. The DOPA-reagent was replacedafter the first hour. After incubation, the epidermis wasfixed in 10% formalin for 1 h and then mounted (dermalside up). The number of melanocytes was estimated basedon counts from five ocular fields at ¿100 magnification.Each value was calculated as a multiple relative to thevalue obtained from the unradiated control skin sample.

ResultsErythema index and melanin indexTwo days after the last UV radiation had been adminis-tered, erythema indices increased in proportion to thefraction size, the dose of radiation administered per frac-

Fig. 2. Increase of erythema index and melanin index 2and 14 days after UV radiation, respectively. Data are ex-pressed as mean∫standard error. The degree of erythemaat 2 days post-irradiation linearly correlated with the de-gree of pigmentation at 14 days post-irradiation. Theircorrelation coefficient was over 0.99 (P,0.01).

268

tion, with the most prominent erythema observed aftera single radiation with high dose UV (Fig. 2). Whetherfractionated or not, UV radiation caused no discerniblechanges in the degree of pigmentation at 2 days post-irradiation. At 14 days post-irradiation, melanin indicesincreased in proportion to the fraction size. The scorecurve for the melanin index of 14 days post-irradiationwas similar to that of the erythema index at 2 days post-irradiation with a smaller magnitude of response (Fig. 2).The relationship between the degree of erythema at 2 dayspost-irradiation and the pigmentation at 14 days post-irradiation was linear, with a correlation coefficient ex-ceeding 0.99 (P,0.01).

Histology and histochemistryTwo days after the last UV exposure, repeated interme-diate and low dose radiation sessions (six sessions with0.5 MED and three sessions with 1 MED UV) followedby DOPA staining revealed that the number of melano-cytes had increased markedly. Single high dose radiationresulted in a slight increase in the number of melanocytesobserved using DOPA staining. No discernible changeswere observed with Fontana-Masson stain. At 14 dayspost-irradiation, DOPA staining revealed that fractionsize was positively correlated with the occurrence of mel-anocytes with tortuous dendrites and abundant cyto-plasm. The Fontana-Masson staining revealed that themost prominent melanin depositions in the basal andhorny layer of the epidermis were observed after a singlehigh dose of radiation.

ImmunohistochemistryTwo days after the last UV exposure, repeated intermedi-ate to low dose sessions markedly increased the numberof melanocytes expressing Mel-5, HMB-45 (Fig. 3a) andtyrosinase (Fig. 3b). Single high dose radiation slightlyincreased the number of melanocytes expressing thesemarkers. At 14 days post-irradiation, the number of mel-anocytes expressing HMB-45 and tyrosinase increased inproportion to the fraction size. The melanocytes positivelystained with anti-Mel-5 could not be counted accuratelybecause of the increased dendricity and overlapping ofmelanocytes.

Split DOPA reactionAt 2 days post-irradiation, repeated intermediate to lowdose irradiation increased the number of melanocytesmore effectively than single high dose radiation. However,at 14 days post-irradiation, the number of melanocytesincreased in proportion to the fraction size and single highdose radiation caused the most prominent increase in thenumber of melanocytes and the degree of dendricity (Fig.4).

Single dose vs. fractionated radiation effects on melanocytes

Fig. 3. Increase in the number of melanocytes expressingHMB-45 (a) and tyrosinase (b) after UV radiation. Num-ber of melanocytes expressing HMB-45 or tyrosinase/100basal cells was measured with image analysis system. Anaverage of five recordings from different fields was calcu-lated as a multiple relative to the value obtained for un-radiated control skin samples. At 2 days post-irradiation,fractionated radiation was more effective in inducing mel-anocytes expressing HMB-45 and tyrosinase than singledose radiation. However, at 14 days post-irradiation, mel-anocytes expressing HMB-45 and tyrosinase increased inproportion to the radiation dose administered per eachfraction.

DiscussionOur experiment showed that at 14 days post-irradiation,when UV-induced pigmentation is maximal, single highdose radiation is more effective for inducing melanocytesexpressing HMB-45, tyrosinase and Mel-5 than are re-peated intermediate to low dose sessions. Moreover, thedegree of positivity with Fontana-Masson and DOPAstaining, and the increase in the number of melanocytesin split DOPA reaction, were also maximal after singlehigh dose radiation. However, at 2 days post-irradiation,when UV-induced erythema was maximal, repeated in-termediate to low dose irradiation sessions had a greatereffect on epidermal melanocytes than did a single highdose of radiation. HMB-45, known as a 10 kDa onco-fetal protein (10), is an antigen that appears mainly on

269

Fig. 4. Photomicrographs from DOPA-incubated splitskin preparations at 2 days post-irradiation (a–c) and 14days post-irradiation (d–f). Specimens were taken fromthe skin exposed to six doses of radiation with 0.5 MEDUV (a, d), exposed to three doses of radiation with 1MED UV (b, e), and exposed to a single dose of radiationwith 3 MED UV (c, f). At 14 days post-irradiation, thenumber of melanocytes was increased in proportion to theradiation dose administered per each fraction (originalmagnification, ¿200).

the immature melanosomes during melanogenesis inboth neoplastic and non-neoplastic melanocytes (8).While malignant transformation of melanocytes is oneexplanation for HMB-45 expression, epidermal melano-cytes in a lesion with an increased epithelial proliferationshowed enhanced expression of HMB-45 (11). The ex-pression of HMB-45 was also suggested as a reflectionof melanocytic activation, correlating with melanocyticproliferation and melanogenesis (12). The antibody totyrosinase recognizes the common epitope of three kindsof enzymes referred to as the tyrosinase gene family:tyrosinase, and tyrosinase-related protein (TRP)-1 and

An et al.

TRP-2 (13). These enzymes are related to the productionof melanosomes and their increased activity and amountafter UV radiation causes UV-induced pigmentation.Thus, our study proved that, although repeated inter-mediate to low dose radiation seems to be more mel-anogenic at 2 days post-irradiation, single high doseradiation is the most effective means by which mel-anogenesis, melanocyte activation and proliferation canbe induced at 14 days post-irradiation.

Like other photobiologic responses, the tanning re-sponse requires direct interaction of UV photons withmolecular targets in the skin. Mechanisms by which UVradiation stimulates melanogenesis include the effects ofUV photons on melanocyte DNA and the melanocytemembrane, and UV-stimulated production of numerouskeratinocyte-derived factors such as endothelin-1 nervegrowth factor (NGF). Eller et al. (14) showed that ex-cised product from photodamaged DNA, pyrimidinedimer, is one of the signals for melanogenesis followingUV radiation. Imokawa et al. (15) showed that UVBradiation upregulated the expression of endothelin-1 inhuman skin 5 days after radiation. Tron et al. (16) foundthat UV radiation upregulated the expression of mRNAof NGF, which prevented UV-induced melanocyte deathby upregulating the anti-apoptotic bcl-2 gene (17), fol-lowing the initial inhibition of the expression of kera-tinocyte NGF mRNA. These results indicate that singlehigh dose radiation, responsible for causing more in-flammation and damage to DNA, can be more mel-anogenic than repeated sessions with suberythemogenicUV. However, because there is a time delay between UVradiation and the expression of melanogenic and anti-apoptotic factors for melanocytes, repeated sessions withintermediate to low dose UV may be more melanogenicat 2 days post-irradiation than single high dose radiationon a transient basis. The administration of higher doseUV may have induced apoptosis in a significant numberof epidermal melanocytes within 2 days post-irradiation,resulting in fewer melanocytes being apparent at thattime, which at 14 days post-irradiation would have beenmore strongly induced to proliferate.

UV radiation events with larger doses per exposuremay be more effective in the treatment of vitiligo. Theeffects of psoralen plus UVA (PUVA) photo-chemotherapy are mediated through immunosuppressionof cytotoxic T lymphocytes and the stimulation of mel-anocytes. In a study of the mechanism of repigmentationin vitiligo, Morelli et al. (18) demonstrated that treat-ment with PUVA stimulated the inactive melanocytes inthe outer root sheath to divide, proliferate and migrateupward to the epidermis. Considering the effects of UVon melanocyte activation and proliferation, radiationwith PUVA may be more effective for the treatment of

270

vitiligo when irradiated with a smaller fraction numberand a larger fraction size.

In conclusion, fractionation of radiation dose diminish-es the effects of UV on epidermal melanocytes. UV radi-ation with smaller fraction numbers and a larger fractiondose is therefore more melanogenic, and more effective,for UV-induced melanocyte activation and proliferation.

References1. Gilchrest BA, Park HY, Eller MS, Yaar M. Mechanisms of ultra-

violet light-induced pigmentation. Photochem Photobiol 1996:63: 1–10.

2. Anderson RR. A survey of the acute effects of UV lasers onhuman animal skin. In: Pratesi R, Sacchi A, eds. Lasers in photo-medicine and photobiology. Berlin: Springer-Verlag, 1980: 109.

3. Takiwaski H, Shirai S, Kohno H, Soh H, Arase S. The degree ofUVB-induced erythema and pigmentation correlate linearly andare reduced in a parallel manner by topical anti-inflammatoryagents. J Invest Dermatol 1994: 103: 642–646.

4. Pathak MA, Nghiem P, Fitzpatrick TB: Acute and chronic effectsof the sun. In: Freedberg IM, Eisen AZ, Wolff K, et al., eds.Dermatology in general medicine. 5th edn. New York: McGraw-Hill, 1999: 1598–1607.

5. Anderson PH, Bjerring P. Spectral reflectance of human skin invivo. Photodermatol Photoimmunol Photomed 1990: 7: 5–12.

6. Feather JW, Ellis DJ, Leslie G. A portable reflectometer for therapid quantification of cutaneous haemoglobin and melanin.Phys Med Biol 1988: 33: 711–722.

7. Bhawan J. Mel-5: A novel antibody for differential diagnosis ofepidermal pigmented lesions of the skin in paraffin-embeddedsection. Melanoma Res 1997: 7: 43–48.

8. Kikuchi A, Shimizu H, Nishikawa T. Expression and ultrastruc-tural localization of HMB-45 antigen. Br J Dermatol 1996: 135:400–405.

9. Starrico RJ, Pinkus H. Quantitative and qualitative data on thepigment cells of adult human epidermis. J Invest Dermatol 1957:28: 33–45.

10. Esclamado RM, Gown AM, Vogel AM. Unique proteins definedby monoclonal antibodies specific for melanoma: some potentialclinical applications. Am J Surg 1986: 152: 376–385.

11. Lommel K, Tronnier M, Wolff HH. Activation of epidermal mel-anocytes is independent of epithelial proliferation and vasculariz-ation. Immunohistologic study using HMB-45 antibody with re-view of the literature. Hautarzt 1996: 47: 616–623.

12. Skelton HG, Smith KJ, Barrett TL, Lupton GP, Graham JH.HMB-45 staining in benign and malignant melanocytic lesions.A reflection of cellular activation. Am J Dermatolpathol 1991:13: 543–550.

13. Jimbow K, Quevedo WC, Prota G, Fitzpatrick TB. Biology ofmelanocytes. In: Freedberg IM, Eisen AZ, Wolff K, et al., eds.Dermatology in general medicine. 5th edn. New York: McGraw-Hill, 1999: 192–220.

14. Eller MS, Yaar M, Gilchrest BA. DNA damage and melanogen-esis. Nature 1994: 372: 413–414.

15. Imokawa G, Miyagishi M, Yada Y. Endothelin-1 as a new mel-anogen: coordinated expression of its gene and the tyrosinasegene in UVB-exposed human epidermis. J Invest Dermatol 1995:105: 32–37.

16. Tron VA, Coughlin MD, Jang DE, Stanisz J, Sauder DN. Expres-sion and modulation of nerve growth factor in murine keratino-cytes (PAM 212). J Clin Invest 1990: 85: 1085–1089.

17. Allsopp TE, Wyatt S, Paterson HF, Davies AM. The proto-onco-gene Bcl-2 can selectively rescue neurotrophic factor-dependentneurons from apoptosis. Cell 1993; 73: 295–307.

18. Morelli JG, Kincannon J, Yohn JJ, Zekman T, Weston WL,Norris DA. Leukotriene C4 and TGF-alpha are stimulators of

Single dose vs. fractionated radiation effects on melanocytes

human melanocyte migration in vitro. J Invest Dermatol 1991:98: 290–295.

AcknowledgmentsThis publication was supported by grant ‘Life Phenom-ena & Function Research Group Program’ (2000) fromthe Ministry of Science and Technology, and by a researchagreement with Pacific Corporation.

Accepted for publication March 16, 2001

271

Corresponding author:Kwang-Hyun ChoProfessor of Department of DermatologySeoul National University College of Medicine28 Yongon-Dong, Chongno-GuSeoul, Korea 110–744Tel: π82 2 760 2412Fax: π82 2 742 7344e-mail: khcho/snu.ac.kr