Topic Article:
Narrow-Band UVB Induces More Carcinogenic Skin Tumors than Broad-Band UVB through the Formation of Cyclobutane Pyrimidine Dimer
Makoto Kunisada, Hiroshi Kumimoto, Kanji Ishizaki, Kunihiko Sakumi, Yusaku Nakabeppu and Chikako Nishigori
Journal of Investigative Dermatology (2007) 127, 2865–2871; doi:10.1038/sj.jid.5701001

Insight into Photocarcinogenesis

Marianna L. Blyumin¹, Shasa Hu¹ and Robert S. Kirsner¹

Journal of Investigative Dermatology (2007) 127, 2700; doi:10.1038/sj.jid.5701158

Advances in phototherapy have led to significant improvement in the care of patients with a variety of skin disorders, ranging from psoriasis to vitiligo. The use of narrow-band ultraviolet radiation B (NB-UVB), with peak fluence at 311 nm, has been one such advance. Superior in efficacy to broad-band UVB (BB-UVB), it combines added efficacy similar to that of photochemotherapy (PUVA), with less inconvenience (van Weelden et al., 1998; Coven et al., 1997; Gathers et al., 2002). However, experimental data suggest that the use of NB-UVB is not without cost. Several investigative studies have determined that NB-UVB possesses greater potential for skin carcinogenesis (Flindt-Hansen et al., 1991; Wulf et al., 1994; Gibbs et al., 1995) than BB-UVB, already a known carcinogen (Kraemer, 1997).

In a follow-up to recent work linking BB-UVB with oxidative DNA damage and tumor formation (Kunisada et al., 2005), Kunisada et al. (2007) pursue the question of photocarcinogenesis by NB-UVB in greater detail. The investigators employed comparable (in terms of MED) doses of NB-UVB and BB-UVB in three susceptible mouse strains, examining tumor formation and three types of DNA damage. In two of the three strains of mice, they found a higher number of malignant tumors following exposure to NB-UVB. This was associated with increased cyclobutane pyrimidine dimer formation and decreased formation of 6-4 photoproducts and 8-oxoguanine. Only in Ogg1 knockout mice, which lack the gene that codes for 8-oxoG–DNA glycosylase, a repair enzyme, did BB-UVB exposure lead to similar numbers of malignant tumors. These results suggest that photocarcinogenesis is greater with NB-UVB and that it is related to cyclobutane pyrimidine dimer formation. Through the following questions we will delve into this article in greater detail.

QUESTIONS

1. What are the major findings of the study?

2. What is the benefit of using different strains of mice in this study?

3. Why would NB-UVB be more carcinogenic than BB-UVB, considering that BB-UVB encompasses a wider spectrum of wavelengths?

4. Was the choice of dose (energy) important in the outcome of the study?

5. Because sarcomas may develop in mice after UV exposure, does UV play a role in sarcoma development in humans?

6. What are the clinical implications of this article?

ANSWERS

1. In the article by Kunisada et al. (2007), the major findings were that NB-UVB led to a greater frequency of malignant tumors than did BB-UVB and that they occurred by way of different mechanisms. Following up on previous work in which BB-UVB induced oxidative DNA damage (with a resultant increase in 8-oxoguanine (8-oxoG), a DNA photoproduct) (Kunisada et al., 2005), the authors sought to characterize the mechanisms by which NB-UVB caused malignant tumor formation. To do this, they examined the generation of three specific DNA photoproducts: cyclobutane pyrimidine dimers (CPDs), 6-4 products, and 8-oxoG. Ogg1 knockout mice, which possess a mutation in 8-oxoG-DNA glycosylase, which normally excises oxidized bases from DNA, and its otherwise identical wild-type strain, C57BL/GJ, were employed.

If the increase in malignant tumor formation in NB-UVB-irradiated mice was due to oxidative DNA damage, one would expect increased amounts of 8-oxo G after NB-UVB irradiation in wild-type mice (C57BL/GJ), with even greater increases in Ogg1 knockout mice, in which one would also expect to observe more malignant tumors. However, this was not the case. The authors found that other DNA photoproducts, CPDs, and 6-4 photoproducts (not 8-oxo G) were increased in NB-UVB-irradiated mice compared with those irradiated with BB-UVB. In contrast to other mouse strains studied, Ogg1 knockout mice irradiated with NB-UVB did not have more malignant tumors than those irradiated with BB-UVB.

In contrast to BB-UVB, mechanisms other than oxidative DNA damage accounted for the increase in malignant tumor formation after NB-UVB irradiation. The malignant tumor rate was 91% for NB-UVB-irradiated C57BL/6J mice and 50% for BB-UVB-irradiated mice. In Ogg1 knockout mice, increased 8-oxoG was found after BB-UVB irradiation compared with NB-UVB, supporting earlier research that suggested a role for oxidative DNA damage after BB-UVB exposure; it also supports the idea that the increase in malignant tumor formation after NB-UVB exposure is not mediated by oxidative DNA damage but rather by other mechanisms.

2. In addition to the two strains of mice described above, a third strain, the albino hairless mouse, was employed. This strain reliably develops skin cancer at relatively low doses and short durations of UV exposure, and there is significant experimental experience with these mice in photocarcinogenesis studies (de Gruijl and Forbes, 1995). However, to generalize the observations to more than one strain, the authors included in the studies C57BL/GJ mice, the background strain of the Ogg1 knockout mice. If both mouse strains behaved in a similar fashion, it would indicate that the observations were not unique to albino hairless mice. Studying several strains of mice therefore provided more reliable and more reproducible results.

As noted above, Ogg1 knockout mice allowed the researchers to determine whether a specific type of DNA damage and its associated repair mechanism (mutation of 8-oxoG-DNA glycosylase which excises oxidized base from DNA) is affected by NB- or BB-UVB. A previous study by this group demonstrated that BB-UVB is carcinogenic in Ogg1 mutant mice (Kunisada et al., 2005), suggesting a role for the enzyme 8-oxoG-DNA glycosylase in DNA repair and an important role for mutations of this enzyme in BB-UVB photocarcinogenesis. These investigators then sought to study the effect of NB-UVB on Ogg1 mice, which had not been previously studied in this fashion. The investigators’ finding that NB-UVB photocarcinogenesis is not related to Ogg1 mutation susceptibility suggests that other mechanisms are involved. Perhaps the authors missed an opportunity to broaden the investigation by choosing not to study other mutant mice. Future studies might compare the effects of NB- and BB-UVB on DNA damage markers in mice with corresponding mutations, CPD excision mutant mice (e.g., Xpa-knockout mice) (Ikehata, 2007), and 6-4 photoproduct excision mutant mice.

3. This study did not address whether NB-UVB wavelengths are more carcinogenic than BB-UVB wavelengths. What probably caused the increase in malignant tumors following NB-UVB exposure was the greater amount of energy required to produce clinical effects, such as erythema, using NB-UVB compared with BB-UBB. The study did not evaluate effects using similar amounts of energy. Based on the greater ratio of malignant to benign tumor development in NB-UVB-exposed mice and increased CPD associated with NB-UVB exposure, the researchers propose that NB-UVB generates larger accumulations of CPD in the skin, leading to more p53 tumor-suppressor gene mutations, immunosuppression, and a higher rate of malignant tumor formation. Other reports indicate that NB-UVB has greater immune suppressive effects than BB-UVB on systemic immune responses, as judged by natural killer cell activity, lymphoproliferation, and cytokine responses (el-Ghorr and Norval, 1997). This might be important, as UV-induced immune suppression is another mechanism by which UV (in susceptible persons) leads to skin cancer formation.

Of note, the BB-UVB utilized in this study was not “pure” UVB; it included wavelengths from 275 to 390 nm, including 320–400 nm (UVA light). It is well known that UVA plays a significant role in generating DNA damage markers and skin carcinogenesis (Mouret et al., 2006). The extent to which this UVA contamination may have influenced the results of this study is not known.

4. Two types of fluorescent lamps were used in this study: TL 20W/01RS lamps emitting NB-UVB at 311 nm with irradiance of 7.6 J/m2 per second and TL20W/12RS lamps emitting 65% of UVB at 275–390 nm (peak 313 nm) with irradiance of 3.8 J/m2 per second. The irradiance exposure energies were determined on the basis of minimal erythema dose (MED). As expected, the MEDs varied with light source and with mouse strain. The MED with NB-UVB exposure was 370 mJ/cm2 for albino mice and 850 mJ/cm2 for C57BL/6J mice. The MED with BB-UVB exposure was 170 mJ/cm2 for albino mice and 250 mJ/cm2 for C57BL6J mice.

Therefore, the ratio of BB- to NB-UVB MED was 1:2.2 for albino hairless mice and 1:3.4. for C57BL/6J mice. The exact mechanism by which these ratios were determined was not explained, other than that it was similar to clinical experience in patients. The rate of malignant tumors for C57BL/6J mice was higher than that for albino hairless mice (91% and 80%, respectively). The authors noted, however, that one rationale for a higher ratio of malignant tumor formation in C57BL/6J mice is the higher ratio of BB- to NB-UVB MED in these mice, suggesting that higher energy influenced tumor formation and, perhaps, the outcome of the study. We are left to consider whether an equal MED ratio (energy) in each mouse group might have generated different results and answered the question of whether NB-UVB wavelengths are truly more carcinogenic.

5. In this study the investigators found a causal relationship between UVB exposure and subsequent sarcoma formation in all three mouse strains. UV-induced sarcomas (fibrosarcomas) have been seen in other murine and animal models (Stenback, 1978). Another report suggests a connection between UV light exposure and the development of classic-type Kaposi’s sarcoma in humans (Monfrecola et al., 1997). However, the influence of UV light on the generation of different types of sarcomas in humans is unknown.

6. Phototherapy is used routinely in dermatology practice in treating a variety of inflammatory, malignant, pigmentary, and pruritic disorders. The development of phototherapy units that deliver NB-UVB has been considered a therapeutic advance, and understanding the potential and relative side effects of different UV wavelengths is important. This study points to greater carcinogenesis in NB-UVB vs. BB-UVB therapy. The investigators propose that this malignant tumorigenesis is likely due to the greater amounts of CPD generated in the epidermis after NB-UVB exposure than after BB-UVB exposure. Whether the greater amount of CPD is responsible—a likely conclusion—or an epiphenomenon was not studied. Therefore, more work is needed to confirm this as a direct and causal mechanism in NB-UVB photocarcinogenesis. It remains possible that there are other carcinogenic mechanisms even beyond the DNA photoproducts that were studied (CPD, 6-4 products, and 8-oxoG). Previous research using mouse cell lines harboring mutations in photoproduct-specific photolyases and reporter genes found that CPDs rather than 6-4 photoproducts or other DNA lesions are responsible for the many UVB-induced mutations (Pfeifer et al., 2005), consistent with this study’s finding that NB-UVB appears to influence CPD formation.

Although murine models of carcinogenesis have previously correlated well with human models of carcinogenesis, the relationship between NB-UVB exposure and carcinogenesis remains to be proven with other supportive and replicative work. Interestingly, some reports in the dermatologic literature for humans found that NB- and BB-UVB cause similar amounts of skin photodamage and inflammation (Tjioe et al., 2003). The relative energy of the BB-UVB and NB-UVB delivered in these studies may be a key to these apparently disparate findings. Nevertheless, the study by Kunisada et al. (2007) provides the basis for important and useful information for the clinician.

REFERENCES

de Gruijl FR, Forbes PD (1995) UV-induced skin cancer in a hairless mouse model. Bioessays 17:651–60

el-Ghorr AA, Norval M (1997) Biological effects of narrow-band (311 nm TL01) UVB irradiation: a review. J Photochem Photobiol B 38:99–106

Ikehata H, Yanase F, Mori T, Nikaido O, Tanaka K, Ono T (2007) Mutation spectrum in UVB-exposed skin epidermis of Xpa-knockout mice: frequent recovery of triplet mutations. Environ Mol Mutagen 48:1–13

Kunisada M, Sakumi K, Tominaga Y, Budiyanto A, Ueda M, Ichihashi M et al. (2005) 8-Oxoguaninie formation induced by chronic UVB exposure makes Ogg1 knockout mice susceptible to skin cancer. Cancer Res 65:6006–10

Kunisada M, Kumimoto H, Ishizaki K, Sakumi K, Nakabeppu Y, Nishigori C (2007) Narrow-band UVB induces more carcinogenic skin tumors than broad-band UVB through the formation of cyclobutane pyrimidine dimer. J Invest Dermatol 127:2865–71

Monfrecola G, Casula L, Procaccini EM (1997) Solar ultraviolet irradiance over a Mediterranean area. J Eur Acad Dermatol Venereol 8:258–9

Mouret S, Baudouin C, Charveron M, Favier A, Cadet J, Douki T (2006) Cyclobutane pyrimidine dimers are predominant DNA lesions in whole human skin exposed to UVA radiation. Proc Natl Acad Sci USA 103:13765–70

Pfeifer GP, You YH, Besaratinia A (2005) Mutations induced by ultraviolet light. Mutat Res 571:19–31
Stenback F (1978) Life history and histopathology of ultraviolet light-induced skin tumors. Natl Cancer Inst Monogr 50:57–70

Tjioe M, Smits T, van de Kerkhof PC, Gerritsen MJ (2003) The differential effect of broad band vs. narrow band UVB with respect to photodamage and cutaneous inflammation. Exp Dermatol 12:729–33

¹Department of Dermatology and Cutaneous Surgery, University of Miami Miller School of Medicine, Miami, Florida, USA