Introduction

Sunbeds emit considerable amounts of UVA radiation. The full health effects of sunbed use are not yet known, but repetitive exposure of the skin to solar UVA radiation results in premature skin aging, also termed photoaging (Krutmann and Gilchrest, 2006). Intentional exposure to UV radiation from sunbeds may have similar detrimental consequences, but evidence for this assumption has so far been elusive (International Agency for Research on Cancer, 1997; Westerdahl et al., 2000; Diffey, 2003; Young, 2004; Autier, 2005; Gallagher, 2005; Scientific Committee on Consumer Products, 2006).

We have previously shown that repetitive UVA irradiation causes mtDNA mutagenesis in vivo and in vitro in human skin cells (Berneburg et al., 2004) and that it can lead to the generation of large-scale deletions of mtDNA known as the Common Deletion (Pang et al., 1994; Yang et al., 1995; Berneburg et al., 1997; Birch-Machin et al., 1998). More recently, we also provided evidence that the UVA-induced Common Deletion is of pathogenetic relevance for photoaging of human skin (Berneburg et al., 2005).

The main aim of this study was to assess whether sunbed use leads to mtDNA mutagenesis in human skin. Therefore, we determined the content of the Common Deletion before and after sunbed use in neck and buttock skin samples of people voluntarily starting to use sunbeds.

Results

Participation

One hundred and seven participants were enrolled in the study. Fifty-nine of them had voluntarily planned to use sunbeds and provided two pairs of biopsies, one before sunbed use and the other thereafter. As was planned a priori, they were divided into three groups according to their age and prior sunbed exposure. Group 1 had a mean age of 224 years and had never used a sunbed before. People from Group 2 had used a sunbed before. They had about the same age range (277 years). Group 3 were older participants (5012 years) with prior sunbed use. Two additional groups without sunbed use during the study were investigated: Group 4 had constantly used sunbeds for at least 10 years once weekly; Group 5 never used sunbeds on a regular basis (both had about the same age range as Group 3).

Study subjects

Sixty-three (59%) of the 107 participants were male, considerably less in Group 3 (5/19). Twenty-seven percent were smokers and 80% belonged to skin type III/IV, according to Fitzpatrick, with no differences between the groups. The seasonal distribution of times when biopsies were taken did not differ between Groups 1, 2, and 3. Groups 4 and 5 were mainly investigated in autumn and winter (Table S1).

Common deletion before sunbed use

The content of the Common Deletion of mtDNA was measured as ppm, that is, parts of mtDNA molecules per million total mtDNA molecules. The mean content before sunbed use was higher in neck skin than in buttock skin, and both were higher than in the blood samples. The values were higher in older people than in younger ones, and young people never having used a sunbed before had lower contents of the Common Deletion in buttock skin than those with prior sunbed use. In buttock skin, highest values showed in Group 4, with the most frequent sunbed use throughout life (at least once weekly for a minimum of about 10 years). The neck-to-buttock ratio (NBR) was defined as the individual ratio of ppm Common Deletion (neck/buttock). It was especially low in the "sunbed group" (Group 4) and in the younger participants with prior sunbed use (Group 2) (Table S2).

Multiple linear regression revealed that age had the strongest influence on the Common Deletion content (Table S3). In neck skin, the Common Deletion nearly doubled per 10 years, whereas buttock skin showed weaker increases. Interestingly, when calculating age-associated increases separately for the single groups, the "sunbed group" (Group 4) showed no increase of the ppm Common Deletion (mean ratio: 1.08 (0.68–1.71) per 10 years). These participants had already high ppm Common Deletion at a young age and showed no further increase with age. The NBR was significantly lower in men than in women, perhaps due to more sunlight exposure of the neck in women. This ratio was also significantly lower in those already having used sunbeds before the study, due to more exposure of buttock skin.

Induction of the Common Deletion

The induction of the Common Deletion of mtDNA was measured as the quotient of the ppm Common Deletion after/before sunbed use and the significance was tested by paired t-test. Three months of regular sunbed use induced the Common Deletion in the skin of participants without prior sunbed use. Neck skin showed a mean 2.56-fold increase (95% CI: 0.97–6.78; P<0.1). The mean increase in buttock skin was 3.58 (95% CI: 1.44–8.89; P<0.01) (Table 1). Participants with prior sunbed use showed no significant increase of the Common Deletion after sunbed use. In blood, the content of Common Deletion of mtDNA was reduced after sunbed use.

Table 1 - Induction of the Common Deletion of mtDNA in buttock skin, neck skin, and blood samples by 3 months of sunbed use: ratio of ppm. Common Deletion of mtDNA after/before sunbed use.

Full table

Linear regression demonstrates that neither age nor gender, smoking, skin type, or season when biopsies were taken influenced the induction significantly. The mean induction in buttock skin was significantly different between Groups 2 and 3 and Group 1 (Table S4). Generally, groups with highest initial values showed the smallest induction.

Discussion

This study implies that sunbed use induces photoaging. Thus, our results indicate that under cosmetical considerations the use of sunbeds is paradoxical and should be discouraged from a medical point of view.

Mean initial contents of the Common Deletion were highest in neck skin, which is more exposed to uncontrolled daily ambient UV irradiation than buttock skin, and its contents in blood were lowest. This confirms previous studies, where we observed the same order of mean values (Berneburg et al., 1997).

The "sunbed group" showed high ppm Common Deletion values already in young age; in this group, no further increase of the ppm Common Deletion with age could be shown. This corroborates our previous notion that the Common Deletion represents a long-term biomarker for actinic damage previously afflicted to human skin (Berneburg et al., 2004).

The most sensitive marker for prior sunbed use was the NBR: a quotient near to 1 indicated prior sunbed use. The NBR showed that younger people were clearly affected by their prior sunbed use, as in this group the NBR did not differ from 1, whereas it was significantly greater than 1 in Group 1 (without prior sunbed use). After sunbed use, the quotient of the young group without prior sunbed use changed toward 1.

As we have shown that sunbed use causes mtDNA mutagenesis in previously unirradiated skin, it is unlikely that the lack of mtDNA mutagenesis in Groups 2 and 3 is due to a protective effect of sunbeds. Instead, we believe that sunbed use seems to be important only for previously unexposed skin, and that in healthy skin, the Common Deletion is tolerated only up to a certain threshold, which may vary with age. mtDNA mutagenesis is induced by reactive oxygen species (Berneburg et al., 1999), which may induce antioxidant defense mechanisms and prevent further mutations. But the existence of an upper limit of mtDNA mutations in the skin would not necessarily mean that other end points of skin aging also show no further increase. The existence of an upper threshold is further supported by two observations: (1) participants with very frequent sunbed visits (Group 4) had already high values in young age and showed no further increase with age and (2) the UVA radiation-induced generation of the Common Deletion in cultured human skin fibroblasts is limited, that is, after reaching a maximum, Common Deletion bearing cells are eliminated (Berneburg et al., 1999). In this context, it is remarkable that blood cells showed a reduction of the Common Deletion content after sunbed use. Possibly, sunbed use leads to a higher peripheral blood flow, which may increase the turnover of blood cells and thereby facilitate a negative selection process directed against cells harboring the Common Deletion, as described in patients with clonal bone marrow disorders (Gattermann et al., 1995).

One limitation of our study is that, due to ethical considerations, sunbed exposure was neither standardized with regard to overall sunbed intensity or exposure time, nor with respect to UV emission spectrum (Gerber et al., 2002) or number, age, or temperature of the lamps. Even in one sunbed, there may exist variations in radiation, although not systematically (Gerber et al., 2002). Differences in sunbed use may therefore be responsible for the huge variations in mtDNA mutagenesis. Nevertheless, the effect of sunbed use on young and previously unirradiated skin was still big enough to be significant in spite of the high random variation.

Our study was observational and not experimental. Therefore, it has to be considered whether unforeseen trends might be able to explain our result. This, however, seems highly implausible for the following arguments: we investigated the induction of the Common Deletion of mtDNA in three different groups with identical time frame. The effect of unforeseen trends should be identical in all three groups. A significant induction of the Common Deletion, however, could be observed only in one group, that is, in those with no sunbed exposure before the study. The adjusted difference in induction in buttock skin was different between Group 1 and the other groups.

We conclude that sunbed use induces mtDNA mutagenesis in human skin. The Common Deletion was strongly elevated only after 3 months of sunbed use in formerly unirradiated skin. This is especially troublesome because primarily young people start to use sunbeds and as a consequence they will have elevated levels of mtDNA mutations in their skin for a very long time, if not for the rest of their lives. Our results support regulatory attempts to limit sunbed use by excluding underage people (Scientific Committee on Consumer Products, 2006). Future studies will have to clarify whether and how increased levels of mtDNA mutations form the molecular basis for a chronic path to mechanisms responsible for the later development of structural and functional alterations in human skin, which characterize photoaging (Krutmann and Gilchrest, 2006).

Materials and Methods

Study design and study population

We invited German adults voluntarily planning to use sunbeds for at least 3 months to participate in our study. Persons eligible should have never used a sunbed before or should have stopped regular sunbed use at least 18 months ago. For comparison, we invited people with very frequent use and older people without prior sunbed use. At the initial visit, all participants filled out a detailed questionnaire to provide information about smoking, nutrition, school education, and sun protection behavior. All participants underwent a dermatological examination. Exclusion criteria were pregnancy, underage, Fitzpatrick skin type I, frequent sunburns during childhood, photosensitivity disorders, atypical or large congenital naevi, more than 40 naevi, personal or family history of skin malignancies, skin premalignancies, organ transplantation, continuous medication, and/or contraindications for surgery.

We took two 4 mm punch biopsies from buttock and neck skin (laterally) and a blood sample (4 ml) 1 day after written informed consent. Then, participants began to use sunbeds of their own choice on a regular basis. They were instructed to record the frequency and duration of their sunbed use in a diary. After 3 months, the second pair of 4 mm punch biopsies was taken from neck and buttock skin, directly adjacent to the first biopsy sites, and a blood sample was drawn.

The study was approved by the local ethics committee. The Declaration of Helsinki Principles was obeyed.

Laboratory analysis

Skin biopsies were stored in liquid nitrogen and blood samples at -20 °C. Total cellular DNA was extracted employing the QIAamp DNA Mini KIT (skin samples) and the QIAamp DNA Blood Midi KITs (blood samples; Qiagen, Hilden, Germany). After extraction, DNA (skin and blood samples) were kept at -20 °C.

The Common Deletion content (skin and blood samples) was determined by quantitative real-time PCR using an ABI Prism 7000 SDS thermal cycler (Applied Biosystems, Darmstadt, Germany) as described previously (Berneburg et al., 2004). Amplification reactions were performed as 25 l triplicates for skin and 25 l duplicates for blood samples. Each amplification reaction contained 10 l (100 ng) of target DNA and 15 l of qPCR SuperMix with ROX (Invitrogen, Karlsnuhe, Germany). Primer sequences and concentrations have been described previously (Koch et al., 2001). To quantify the content of deleted mtDNA molecules, standard plasmids for total mtDNA and deleted mtDNA were generated (Koch et al., 2001). The content of the Common Deletion is given in ppm mtDNA molecules with Common Deletion per 10⁶ total (=complete+deleted) mtDNA molecules (detection limit: 0.01 ppm).

Statistical analysis

After logarithmic transformation, Common Deletion values and ratios were normally distributed. The significance of the induction of ppm Common Deletion of mtDNA in neck or buttock (ratio after/before sunbed use) was tested by a paired t-test.

Multiple linear regression analysis was used to test whether "initial values" or "induction" was modified by age, gender, smoking, skin type, and sunbed use before the study. To ease interpretation, the regression coefficients were retransformed to mean ratios. These ratios are the estimated multiplicative change in the mean associated with a one-unit increase in the influencing variable.

The NBR is the individual ratio of ppm Common Deletion (neck/buttock) and measures the difference in the mtDNA content between neck and buttock skin. An NBR greater than 1 indicates that light-exposed skin (represented by neck skin samples) has a higher Common Deletion level than protected skin (represented by buttock skin samples). As buttock skin is not protected from irradiation when using a sunbed, this ratio will be smaller in persons having used sunbeds. Therefore, this variable should be especially useful to discriminate sunbed users from non-users.

Conflict of Interest

The authors state no conflict of interest.

References

1. Autier P (2005) Cutaneous malignant melanoma: facts about sunbeds and sunscreen. Expert Rev Anticancer Ther 5:821–833 | Article | PubMed |
2. Berneburg M, Gattermann N, Stege H, Grewe M, Vogelsang K, Ruzicka T et al. (1997) Chronically ultraviolet-exposed human skin shows a higher mutation frequency of mitochondrial DNA as compared to unexposed skin and the hematopoietic system. Photochem Photobiol 66:271–275 | Article | PubMed | ISI | ChemPort |
3. Berneburg M, Gremmel T, Kurten V, Schroeder P, Hertel I, von Mikecz A et al. (2005) Creatine supplementation normalizes mutagenesis of mitochondrial DNA as well as functional consequences. J Invest Dermatol 125:213–220 | PubMed | ISI | ChemPort |
4. Berneburg M, Grether-Beck S, Kurten V, Ruzicka T, Briviba K, Sies H et al. (1999) Singlet oxygen mediates the UVA-induced generation of the photoaging-associated mitochondrial Common Deletion. J Biol Chem 274:15345–15349 | Article | PubMed | ISI | ChemPort |
5. Berneburg M, Plettenberg H, Medve-Koenigs K, Pfahlberg A, Ruzicka T, Gers-Barlag H et al. (2004) Induction of the photoaging-associated mitochondrial common deletion in vivo in normal human skin by repetetive UVA exposure and persistance for 16 months. J Invest Dermatol 122:1277–1283 | Article | PubMed | ISI | ChemPort |
6. Birch-Machin MA, Tindall M, Turner R, Haldane F, Rees JL (1998) Mitochondrial DNA deletions in human skin reflect photo- rather than chronologic aging. J Invest Dermatol 110:149–152 | Article | PubMed | ISI | ChemPort |
7. Diffey BL (2003) A quantitative estimate of melanoma mortality from ultraviolet A sunbed use in the UK. Br J Dermatol 149:578–581 | Article | PubMed | ChemPort |
8. Gallagher R (2005) Sunbeds—do they increase risk of melanoma or not? Eur J Cancer 41:2038–2039 | Article | PubMed |
9. Gattermann N, Berneburg M, Heinisch J, Aul C, Schneider W (1995) Detection of the ageing-associated 5-Kb common deletion of mitochondrial DNA in blood and bone marrow of hematologically normal adults. Absence of the deletion in clonal bone marrow disorders. Leukemia 9:1704–1710 | PubMed | ISI | ChemPort |
10. Gerber B, Mathys P, Moser M, Bressoud D, Braun-Fahrlander C (2002) Ultraviolet emission spectra of sunbeds. Photochem Photobiol 76:664–668 | Article | PubMed | ChemPort |
11. International Agency for Research on Cancer (eds) (1997) IARC monographs on the evaluation of carcinogenic risks to humans. Solar and ultraviolet radiation. IARC Monogr Eval Carcinog Risks Hum, 55
12. Koch H, Wittern KP, Bergemann J (2001) In human keratinocytes the common deletion reflects donor variabilities rather than chronologic aging and can be induced by ultraviolet A irradiation. J Invest Dermatol 117:892–897 | Article | PubMed | ISI | ChemPort |
13. Krutmann J, Gilchrest BA (2006) Photoaging of skin. In: Skin aging (Gilchrest BA, Krutmann J, eds), New York: Springer, 33–42
14. Pang CY, Lee HC, Yang JH, Wei YH (1994) Human skin mitochondrial DNA deletions associated with light exposure. Arch Biochem Biophys 312:534–538 | Article | PubMed | ISI | ChemPort |
15. Scientific Committee on Consumer Products, European Commission (2006) Opinion on biological effects of ultraviolet radiation relevant to health with particular reference to sunbeds for cosmetic purposes. http://ec.europa.eu/health/ph_risk/
    committees/04_sccp/docs/sccp_o_031b.pdf (accessed 2007)
16. Westerdahl J, Ingvar C, Masbäck A, Jonsson N, Olsson H (2000) Risk of cutaneous malignant melanoma in relation to use of sunbeds: further evidence for UV-A carcinogenicity. Br J Cancer 82:1593–1599 | Article | PubMed | ISI | ChemPort |
17. Yang JH, Lee HC, Wei YH (1995) Photoageing-associated mitochondrial DNA length mutations in human skin. Arch Dermatol Res 287:641–648 | Article | PubMed | ISI | ChemPort |
18. Young AR (2004) Tanning devices—fast track to skin cancer? Pigment Cell Res 17:2–9 | Article | PubMed |

Acknowledgments

We thank M. Weber and U. Schubhardt, who arranged the study subjects' appointments and helped taking skin and blood samples. We thank I. Hertel and S. Wild, who supported the laboratory analyses, and E. Link, G. Seitner and T. Schikowski, who helped with the layout of the paper and submitting the drafts. This study has been supported by the Federal Ministry for the Environment, Nature Conservation and Nuclear Safety (BMU) and the Deutsche Forschungsgemeinschaft, SFB 503, Project B2, and GRK 1033.

SUPPLEMENTARY MATERIAL

Table S1. Characteristics of the study subjects

Table S2. Common Deletion content in buttock, neck, and blood, and neck-to-buttock ratio before sunbed use

Table S3. Influences on ppm Common Deletion of mtDNA in buttock and neck skin samples and on neck-to-buttock ratio (N=103)¹

Table S4. Influences on the induction (=quotient after/before 3 months sunbed use) of the ppm Common Deletion of mtDNA in buttock and neck skin samples (N=57)¹