Tanning salon exposure and molecular alterations

Tanning salon exposure and molecular alterations S. Elizabeth Whitmore, MD,a Warwick L. Morison, MD,a Christopher S. Potten, DSc,b and Caroline Chadwickb Baltimore, Maryland, and Manchester, United Kingdom Background: Human studies of the short-term cellular effects of tanning salon exposures are lacking. Findings of such studies may prove extremely helpful in educating consumers considering or currently attending tanning salons. Objective: Our purpose was to determine whether tanning salon exposure causes DNA alterations and p53 protein expression in epidermal keratinocytes and/or circulating peripheral lymphocytes. Methods: Eleven subjects received 10 full-body tanning salon exposures over a 2-week period. UV-induced DNA cyclobutane pyrimidine dimers and p53 protein expression were examined, comparing pretreatment peripheral blood lymphocytes and epidermal biopsy specimens with analogous specimens obtained after the 10 tanning salon exposures. Results: Cyclobutane pyrimidine dimers in DNA and p53 protein expression were detected in epidermal keratinocytes, but were absent in lymphocytes. Conclusion: Similar to outdoor sun exposure, short-term recreational tanning salon exposure causes molecular alterations believed essential in the development of skin cancer. (J Am Acad Dermatol 2001;44:775-80.)

D

espite years of concern about the adverse effects of ultraviolet (UV) radiation and tanning, debate about this issue continues. In 1998, the National Institutes of Health (NIH), Centers for Disease Control and Prevention, and the Food and Drug Administration (FDA) sponsored a 2day seminar entitled “Research Workshop on the Risks and Benefits of Exposure to Ultraviolet Radiation and Tanning” (NIH, Bethesda, Md, Sept 1618, 1998). Here, tanning enthusiasts and tanning salon industry representatives presented tanning as safe and healthful. Although most scientists attending the seminar provided evidence of the hazards of tanning, a small minority reported theories on the

From the Department of Dermatology, Johns Hopkins University School of Medicine, Baltimorea; and Paterson Institute for Cancer Research, The University of Manchester.b Supported by the Dermatology Foundation Clinical Career Development Award sponsored by Ortho Dermatological and by the American Cancer Society Institutional Research Grant, #1RG11-35. Accepted for publication Aug 28, 2000. Reprint requests: S. Elizabeth Whitmore, MD, Department of Dermatology, 550 N Broadway, Suite 1002, Baltimore, MD 21205. Copyright © 2001 by the American Academy of Dermatology. 0190-9622/2001/$35.00 + 0 16/1/112581 doi:10.1067/mjd.2001.112581

benefits of UV radiation. The tanning industry is extremely large in America and Northern Europe. On an average day in the United States alone, more than 1 million people invest both time and money to attend tanning salons.1 The yearly world market for tanning salon lamp sales is estimated to be 13 million lamps. At 40 lamps per tanning unit and an 800-hour life per lamp, this number of lamps allows 125,000 units to run without interruption for 40 hours per week for 52 weeks. A recent study from Sweden involving adolescents who attend tanning salons estimated that yearly UV radiation exposure was easily doubled in salon users when taking into account body surface area irradiated and dose. Remarkably, the effect was similar to that which would be experienced with a 10% depletion of atmospheric ozone.2 Because the tanning industry is not responsible for providing data on the carcinogenic effects of tanning salons, there are few studies of the biologic effects of tanning salons.3 This is disconcerting at best. Even if tanning salon exposure increases the risk of skin cancer by a small amount, it is a significant potential public health hazard given the number of people attending salons. Because of these concerns, we studied the acute effects of tanning salon exposure on 2 factors that are associated with cell damage or are believed to be 775

776 Whitmore et al

J AM ACAD DERMATOL MAY 2001

Table I. Cyclobutane pyrimidine dimer quantification: Integrated optical density measurement (see text)* 1 Exposure Subject No.

1 2 3 4 5 6 7 8 9 10 11 x– SD SE 95% CI

Immediately

24 h later

10 Exposures Immediately

24 h later

193 127 73 159 93 71 252 87 0 0 20 27 4 0 9 11 0 7 15 2 16 –1 54 -2 55 40 77 51 0 193 95 28 –19 7 121 100 43 7 0 11 0 12 28 4 35.0 42.1 67.6 43.5 61.6 63.9 72.5 51.4 18 19.3 21.9 13.1 [-0.3, 70.3] [4.3, 79.9]† [24.7, 110.5]† [17.3,69.2]†

CI, Confidence interval; SD, standard deviation; SE, standard error. *Values are those obtained minus CPD value obtained from unexposed control. †Statistically significant, P <.05.

important in carcinogenesis: molecular alterations in the form of cyclobutane pyrimidine dimer (CPD) photoproduct formation in DNA and p53 protein expression.4

MATERIAL AND METHODS To avoid the potential confounding effects of casual outdoor sunlight exposure, this investigation was performed in the spring. Subject inclusion criteria included age between 18 and 50 years, good general health, Fitzpatrick skin type II or III, nongravid and non-breast-feeding state, and willingness to use contraception throughout the study. Exclusion criteria included dysplastic nevi, history of malignancy, connective tissue disease or any type of photosensitivity, use of systemic or topical photosensitizing medications, exogenous or endogenous immunosuppression due to immunosuppressive medication or medical illness, and outdoor or indoor suntanning within the previous 30 days. The protocol was approved by the Johns Hopkins Joint Committee on Clinical Investigation and written informed consent was obtained from all participants. Tanning salon exposures Because the most common UV sources used by tanning salons in the United States emit 4.6% UVB,1 we chose a similar source. A bank of F72T12BLHO

bulbs (UV Resources, Cleveland, Ohio), the most popular bulbs used by salons in this country according to the distributor, was used to deliver the 10 UV treatments over a 2-week period. The manufacturers’ specifications state that these lamps emit 5% UVB and 95% UVA. With the use of a research radiometer (model IL700A; International Light, Danvers, Mass), our measurement of fluence at 20 cm with a UVB probe SEE 240 was 0.6 mW/cm2 and with a UVA probe SEE DO15 was 9.4 mW/cm2. Subjects wore goggles, were positioned at a distance of 20 cm, and received full body exposures (ventral followed by dorsal surface exposure), excluding a 4 × 4 cm square on the buttock (shield provided by a double, 2-ply, 4- × 4-cm gauze patch). On the final (10th) exposure, half of this shielded square was also exposed, thereby providing an area for sampling after a single UV exposure. The first UV exposure was begun at 70% of a minimal erythema dose, as determined in one person with skin type II. Each dose was increased by 20% with the following exceptions: (1) if the treatment 24 hours earlier caused mild erythema, the dose was not increased; (2) if it caused moderate or severe erythema, the treatment was not given. No make-up treatments were given for those treatments missed because of moderate or severe erythema. Skin biopsies Within 3 minutes of the final (10th) exposure, 6mm epidermal shave biopsy specimens were taken from 3 sites on the buttock: the shielded control site, the adjacent site shielded throughout the study with the exception of the final UV exposure to which it was exposed, and the adjacent buttock skin that had been exposed to all 10 UV tanning exposures over the preceding 2 weeks. Approximately 24 hours later, epidermal biopsy specimens were again taken from the latter 2 sites (ie, the site exposed to UV on one occasion and the site exposed to UV on 10 occasions). Before biopsy, sites were anesthetized with intradermal injection of 1% lidocaine with 1:100,000 dilution epinephrine. Each biopsy specimen was bisected and half was processed for CPD analysis. As previously described, tissue was sectioned into 1-mm wide strips and fixed in formalin at 4oC for 18 hours, transferred to ethanol for further fixation, and paraffin embedded. Three-micrometer tissue sections were stained with the monoclonal antibody TDM-1, which recognizes thymine dimers.5,6 The other half of the biopsy specimen was routinely processed with formalin fixation and paraffin embedding, and 3-µm sections were studied by means of immunohistochemistry (avidinbiotin peroxidase technique), staining for p53 protein antibody (steam pretreatment: 1/250 dilution; Dako Corp, Carpinteria, Calif).

Whitmore et al 777

J AM ACAD DERMATOL VOLUME 44, NUMBER 5

Table II. Epidermal p53 protein expression: Variation with number of exposures and time after exposure* p53-positive cells/HPF (No.) Subject No.

No UV “control”

Immediately after 1st UV exposure

24 h after 1st UV exposure

Immediately after 10th UV exposure

24 h after 10th UV exposure

1 2 3 4 5 6 –x (SD) SE 95% CI

0 0 2 3 1 0 1 (1.3) 0.53 [–0.04, 2.04]

0 0 2 5 12 0 3.2 (4.8) 2.0 [–0.6, 7.0]

110 102 115 60 20 50 76.2 (38.5)† 15.7 [45.4, 107.0]†

39 48 0 60 62 12 36.8 (25.6)† 10.5 [16.3, 57.3]†

49 42 42 52 25 10 36.7 (16.1)† 6.6 [23.8, 49.6]†

*Specimens from subjects 7-11 saturated with stain and showed 100% positivity on all control and UV slides; data excluded. †Statistically significant difference from that of no UV “control.”

Sections stained for CPD were analyzed by means of the Discovery Automated Image Analysis System (Becton Dickinson), as previously described. This system measures an integrated optical density that quantifies the number of cells demonstrating staining (ie, binding by the CPD monoclonal antibody).5,6 Sections stained for p53 protein were examined with light microscopy, visually quantifying the number of cells showing p53 positivity per high power field (HPF). For each slide, 3 HPFs were quantified and averaged. Peripheral blood lymphocyte analysis At baseline and again immediately on completion of the 10th UV exposure, blood was drawn and peripheral blood lymphocytes (PBLs) were examined for evidence of p53 protein expression and CPD formation. Lymphocyte separation was performed with lymphocyte separation medium (Organon Teknika Corp, Durham, NC) with the use of the methods recommended by the manufacturer (modified Böyum procedure).7 On lymphocyte separation, two cytocentrifuge (Cytospin; Thermo Shandon, Pittsburgh, Pa) slide preparations were made for each specimen; one slide was stained with p53 protein antibody and the other with CPD antibody.5,6

RESULTS Eleven subjects were recruited and completed the study as planned. The average age of subjects was 38 years (standard deviation [SD], 8.2 years); 4 were males; 10 had Fitzpatrick skin type II and one had skin type III. As outlined in the Methods section, the UV dose was increased by 20% per treatment to a maximum of 2 times the initial dose. Although no subjects had erythema in the 24 hours after the first exposure, on the subsequent 9 exposures all sub-

jects had at least one episode of mild erythema (average number of instances/subject, 1.8), and 2 subjects developed moderate erythema. As seen in Table I, epidermal CPD quantity immediately and 24 hours after both 1 and 10 exposures varied significantly among subjects. Relative to the unexposed control skin, a statistically significant increase in CPD was found in specimens taken immediately after the 10th UV exposure. Repeat biopsy specimens taken from this same site 24 hours later showed a nonsignificant decrease in CPD quantity (ie, a statistically nonsignificant quantitative measure of CPD repair occurred in the first 24 hours after completion of the UV exposures). In specimens taken both immediately and again 24 hours after a single UV exposure, an increase in CPD quantity relative to an unexposed control site was found, but this quantity was only statistically significant in specimens obtained 24 hours after exposure. Finally, no statistically significant difference in the CPD quantity was found 24 hours after a single UV exposure compared with 10 UV exposures. The p53 antibody– stained slides on the first 6 subjects were read blindly with resultant data reported in Table II. Epidermal expression of p53 protein was not present immediately after a single UV exposure, but was present in all layers of the epidermis 24 hours later. Relative to the result after just one exposure, p53 expression after the 10th UV exposure was approximately 50% less (p53-positive _keratinocytes/HPF: 24 hours after a single exposure: x = 76.2; _ SD = 38.5 vs 24 hours after the 10th exposure: x = 36.7; SD = 16.1 [Fig 1]; difference in p53-positive cell numbers: after single _ vs 10 exposures: x = 39.5; standard error [SE] = 12.9; 95% confidence interval, 14.2, 64.8). For reasons that are unclear, difficulty was encountered on p53 staining in the 5 remaining subjects’ samples; all

778 Whitmore et al

J AM ACAD DERMATOL MAY 2001

A

B

C Fig 1. Immunohistochemistry with p53 protein antibody demonstrated in subject 1. A, Immediately after UV exposure (0 ⊕ cells/HPF). B, Twenty-four hours after one UV exposure (110 ⊕ cells/HPF). C, Twenty-four hours after the 10th of 10 UV exposures (49 ⊕ cells/HPF).

specimens including those from unirradiated skin showed 100% staining of the epidermis and therefore these data were not included in the analysis. Cytospin slide preparations of PBLs stained with antip53 and CPD antibodies were both negative in samples obtained at baseline and immediately after the 10th UV exposure 2 weeks later.

DISCUSSION Epidemiologic studies and case reports have cited an increased risk of potentially fatal melanoma in persons frequenting tanning salons.8-16 Experimental animal studies have supported the role of tanning salon–type radiation sources in cutaneous carcinogenesis.17-19 Despite these studies and concerns regarding other tanning salon–associated morbidities, regulation of this $2 billion industry in the United States is spotty, with only 21 states having regulatory laws in place.1,20,21 The objective of our study was to assess whether tanning salon exposure results in cell damage within the skin and peripheral blood, specifically exam-

ining for UV-induced DNA alterations and p53 protein expression in keratinocytes and circulating lymphocytes. Although we used a tanning schedule (10 exposures over 2 weeks) and dose of exposure (borderline erythemogenic with erythema occurring on at least 1 occasion) more intense than is recommended by the FDA, the schedule we used was probably not unlike that used by teenagers preparing for the prom or fair-skinned travelers readying for a tropical vacation. In a recent presentation by Hornung at the Society of Investigative Dermatology, it was cited that in North Carolina, a state in which salons are overseen by state inspectors, at least 95% of salon patrons do not follow the FDA-recommended tanning exposure schedules. This misuse may help explain the Centers for Disease Control and Prevention’s reported estimate of 700 emergency department visits per year related to tanning salon exposure.20 Overexposure to UV radiation in tanning salons is not unique to the United States. A study performed in Italy found that 25% of patrons interviewed have experienced

J AM ACAD DERMATOL VOLUME 44, NUMBER 5

salon-induced sunburns, and only 60% suspended sessions after burning.22 In our investigation, we intentionally selected a UV source emitting 5% UVB because this is what is commonly used in the United States. For this reason, it could be anticipated that changes primarily due to UVB would be evident. Our data confirmed this. Tanning salon exposures ranging from a single nonerythemogenic exposure to 10 exposures given over 2 weeks caused epidermal DNA pyrimidine dimers. Notably, the quantity detected was similar after just 1 exposure versus 10 exposures, supporting the idea that a steady state of pyrimidine dimer formation and repair seems to be achieved with exposure at a constant frequency. With regard to epidermal p53 protein, as expected, expression was absent immediately after the initial UV exposure and markedly expressed uniformly throughout the epidermis 24 hours after a single exposure. However, we did not anticipate that expression 24 hours after the last of 10 exposures would be approximately 50% of that found 24 hours after a single exposure. Although there are no similar investigations of p53 protein expression after tanning salon exposures, previous studies have shown that UVB causes p53 protein expression throughout the epidermis, UVA causes p53 protein expression predominantly in the basal layer, and UVC causes p53 protein expression predominantly in the granular and spinous layers.23 Solar simulator exposure with epidermal examination at 4, 24, 48, and 120 hours has shown that p53 protein peaks in suprabasal cells at 4 hours and in basal cells at 48 hours, with return to baseline by 120 hours.24 These two studies indicate that UVA-induced p53 protein expression is both more localized (basal layer) and delayed (peaking at 48 hours) than is UVBinduced p53 protein expression. (Interestingly, this is in contrast to UVA- vs UVB-induced apoptosis, in which the former temporally precedes the latter.25) As expected, we found that a single tanning salon exposure causes a pattern of epidermal p53 protein expression similar to that caused by UVB or solar simulator exposure, being throughout all epidermal layers. However, we were surprised by the reduction in p53 protein on multiple exposures, particularly because this contrasted with the fairly constant quantity of CPD after multiple exposures. It is known that a single UV exposure causes p53 protein expression without an increase in p53 messenger RNA. This fact may explain the greater amount of p53 protein being released after the first UV exposure.26 Although speculative, it may be that multiple exposures lead to upregulation of DNA repair, thereby leading to a lesser functional need for p53 protein expression to slow the cell cycle or cause apoptosis.

Whitmore et al 779

We did not find any evidence of PBL molecular alterations with tanning salon exposures. In that we examined a 20-mL volume of blood, representing approximately 0.3% to 0.4% of the total blood volume, we can conclude that if 10 tanning salon exposures administered over 2 weeks cause CPD or p53 protein expression in PBLs, it occurs in less than 0.3% to 0.4% of PBLs. Despite the lack of these specific molecular changes detected, it is notable that we27 and others28 have reported an increase in circulating T suppressor cell lymphocytes and inhibition of phytohemagglutinin-induced mitogenesis after tanning salon exposures.29 In summary, similar to sunlight, tanning salon exposure as commonly used in this country causes DNA alterations and p53 protein expression in the skin. We believe that entrepreneurs providing this service should be required to inform clients of these “molecular changes” and the potential that they may increase a person’s risk of skin cancer. REFERENCES 1. Spencer JM, Amonette RA. Indoor tanning: risks, benefits, and future trends. J Am Acad Dermatol 1995;33:288-98. 2. Wester U, Boldemann C, Jansson B, Ullén H. Population UV-dose and skin area–do sunbeds rival the sun? Health Phys 1999; 77:436-40. 3. Woollons A, Kipp C, Young R, Petit-Frere C, Arlett CF, Green MH, et al. The 0.8% ultraviolet B content of an ultraviolet A sunlamp induces 75% of cyclobutane pyrimidine dimers in human keratinocytes in vitro. Br J Dermatol 1999;140:1023-30. 4. Brash DE, Ziefler A, Jonason AS, Simon JA, Kunala S, Leffell DJ. Sunlight and sunburn in human skin cancer: p53, apoptosis, and tumor promotion. J Invest Dermatol Symp Proc 1996;1: 136-42. 5. Chadwick CA, Potten CS, Nikaido O, Matsunaga T, Proby C, Young AR. The detection of cyclobutane thymine dimers, (6-4) photolesions and the Dewar photoisomers in sections of UVirradiated human skin using specific antibodies, and the demonstration of depth penetration effects. J Photochem Photobiol 1995;28:163-70. 6. Potten CS, Chadwick CA, Cohen AJ, Nikaido O, Matsunaga T, Schipper NW, Young AR. DNA damage in UV-irradiated human skin in vivo: automated direct measurement by image analysis (thymine dimers) compared with indirect measurement (unscheduled DNA synthesis) and protection by 5-methoxypsoralen. Int J Radiat Biol 1993;63:313-24. 7. Böyum A. Isolation of mononuclear cells and granulocytes from human blood. Scan J Clin Lab Invest 1968;77(Suppl 97):21. 8. Autier P, Doré J-F, Lejeune F, Koelmel KF, Geffeler O, Hille P, et al. Cutaneous malignant melanoma and exposure to sunlamps or sunbeds: an EORTC multicenter case-control study in Belgium, France, and Germany. Int J Can 1994;58:809-13. 9. Autier P, Joarlette M, Lejeune F, Liénard D, André J, Achten G. Cutaneous malignant melanoma and exposure to sunlamps and sunbeds: a descriptive study in Belgium. Melanoma Res 1991;1:69-74. 10. Brodthagen H. Malignant melanoma caused by UV-A suntan bed? Acta Derm Venereol (Stockh) 1982;62:356. 11. Jones SK, Moseley H, MacKie RM. UVA induced melanocytic lesions. Br J Dermatol 1987;117:111-5. 12. Retsas S. Sun beds and melanoma. BMJ 1983;286:892.

780 Whitmore et al

13. Sorahan T, Grimley RP. The aetiological significance of sunlight and fluorescent lighting malignant melanoma: a case-control study. Br J Cancer 1985;52:765-9. 14. Swerdlow AJ, English JSC, MacKie RM, O’Doherty CJ, Hunter JAA, Clark J, et al. Fluorescent lights, ultraviolet lamps, and risk of cutaneous melanoma. BMJ 1988;297:647-58. 15. Walter SD, Marrett LD, From L, Hertzman C, Shannon HS, Roy P. The association of cutaneous malignant melanoma with the use of sunbeds and sun lamps. Am J Epidemiol 1990;131:23243. 16. Westerdahl J, Olsson H, Måsbäck A, Ingvar C, Jonsson N, Brandt L, et al. Use of sunbeds or sun lamps and malignant melanoma in southern Sweden. Am J Epidemiol 1994;140:691-9. 17. Bech-Thomsen N, Wulf HC, Poulsen T, Christensen FG, Lundgren K. Photocarcinogenesis in hairless mice induced by ultraviolet A tanning devices with or without subsequent solar-simulated ultraviolet irradiation. Photodermatol Photoimmunol Photomed 1991;8:139-45. 18. Pathak MA. Ultraviolet radiation and the development of nonmelanoma and melanoma skin cancer: clinical and experimental evidence. Skin Pharmacol 1991;4(Suppl 1):85-94. 19. van Weelden H, van der Putte SCJ, Toonstra J, van der Leun JC. UVA-induced tumours in pigmented hairless mice and the carcinogenic risks of tanning with UVA. Arch Dermatol Res 1990; 282:289-94. 20. Jancin B. Most tanning users exceed FDA exposure limits. Skin Allergy News 1999;8:41. 21. Update on state tanning parlor laws. Dermatology World 1997;7(4):1.

J AM ACAD DERMATOL MAY 2001

22. Monfrecola G, Fabbrocini G, Posteraro G, Pini D. What do young people think about the dangers of sunbathing, skin cancer and sunbeds? A questionnaire survey among Italians. Photodermatol Photoimmunol Photomed 2000;16:15-8. 23. Campbell C, Quinn AG, Angus B, Farr PM, Rees JL. Wavelength specific patterns of p53 induction in human skin following exposure to UV radiation. Cancer Res 1993;53:2697-9. 24. Pontén F, Berne B, Ren Z-P, Nister M, Pontén J. Ultraviolet light induces expression of p53 and p21 in human skin: effect of sunscreen and constitutive p21 expression in skin appendages. J Invest Dermatol 1995;105:402-6. 25. Godar DE. Preprogrammed and programmed cell death mechanisms of apoptosis: UV-induced immediate and delayed apoptosis. Photochem Photobiol 1996;63:825-30. 26. Healy E, Reynolds NJ, Smith MD, Campbell C, Farr PM, Rees JL. Dissociation of erythema and p53 expression in human skin following UVB irradiation, and induction of p53 protein and mRNA following application of skin irritants. J Invest Dermatol 1994;103:493-9. 27. Whitmore SE, Morison WL. The effect of tanning parlor exposure on delayed and contact hypersensitivity. Photochem Photobiol 2000;71:700-5. 28. Hersey P, Bradley M, Hasic E, Haran G, Edwards A, McCarthy WH. Immunological effects of solarium exposure. Lancet 1983;1: 545-8. 29. Larcom LL, Morris TE, Smith ME. Tanning salon exposure suppression of DNA repair capacity and mitogen-induced DNA synthesis. Photochem Photobiol 1991;53:511-6.