Xeroderma Pigmentosum A-Deficient, Stem Cell Factor-Transgenic

Home , Xeroderma pigmentosum

Development of a New Mouse Model (Xeroderma Pigmentosum A-Deﬁcient, Stem Cell Factor-Transgenic) of Ultraviolet B-Induced Melanoma

Fumikazu Yamazaki,Ã Hiroyuki Okamoto,Ã Yasuhiro Matsumura,Ã Kiyoji Tanaka,w Takahiro Kunisada,z and Takeshi HorioÃ ÃDepartment of Dermatology, Kansai Medical University, Moriguchi, Osaka, Japan; wInstitute for Molecular and Cellular Biology, Graduate School of Frontier Biosciences, Osaka University, CREST, Japan Science and Technology Agency, Suita, Osaka, Japan; zStructure and Organ Formation Control, Regeneration Medicine and Bioethics, Gifu University, Gifu, Japan

It is well established that exposure to sunlight or ultraviolet radiation (UVR) is the major environmental risk factor for the development of skin neoplasms. To date, however, there have been few appropriate mouse models available for studying the role of UVR in melanoma carcinogenesis, mainly because of the murine lack of the epidermal melanocyte, which is a major source of origin of human melanoma. In this study, we established xeroderma pigmentosum group A gene-deﬁcient, stem cell factor-transgenic mice, which are defective in the repair of damaged DNA and do have epidermal melanocytes. The mice were exposed to UVR three times a week for 10 wk. More than 30% of the irradiated mice developed tumors of melanocyte origin that metastasized to the lymph nodes. Histologically, proliferated cells exhibited lentigo maligna melanoma or nodular melanoma. Immunohistochemistry conﬁrmed that the tumor cells were characteristic of melanoma. Non-irradiated mice did not develop skin tumors spontaneously. The newly generated model mouse might be useful for studying the photobiological aspects of human melanoma, because the mice developed melanoma from epidermal melanocytes only after UVR exposures. Key words: lentigo maligna/melanocyte/melanoma/stem cell factor/UV-carcinogenesis/xeroderma pigmentosum A J Invest Dermatol 125:521 –525, 2005

Melanoma, which originates from skin melanocytes, is one excision repair (Kraemer et al, 1994). We have previously of the most morbid and lethal diseases among the skin reported that group-A XP gene-deﬁcient, XPA (/), mice disorders. The incidence rates of melanoma have steadily are highly susceptible to UVR-induced skin carcinogenesis increased for the past several decades all over the world (Nakane et al, 1995). These mice easily and rapidly develop (Serraino et al, 1998). The rising incidence may be due in squamous cell carcinoma (SCC) of keratinocyte origin, but part to increased recreational exposure to sunlight (Langley do not develop melanoma, presumably because of a lack in et al, 1997). Elwood et al (1997) reviewed that a number of epidermal melanocytes. The growth and differentiation of retrospective epidemiological studies strongly suggest that melanocytes require stem cell factor (SCF), the ligand for sunlight plays a critical role in the induction of cutaneous the c-Kit receptor tyrosine kinase. The lack of SCF expres- melanoma. Investigations to clarify the functional role of ul- sion in keratinocytes is thought to result in the amelanocytic traviolet radiation (UVR) in melanoma carcinogenesis, how- skin of the mouse except for the hair follicle. In contrast, the ever, have been limited because of the lack of experimental continuous expression of SCF in human keratinocytes animals in which to induce experimental melanomas with maintains melanocytes in the skin (Yoshida et al, 1996; UVR. In contrast, the UVR mechanisms of action in the de- Kunisada et al, 1998). Targeting the expression of SCF- velopment of non-melanoma skin cancers have been ex- transgenic (SCF-Tg) to epidermal keratinocytes in mice with tensively studied in mammals (Blum, 1959; Fisher et al, transgenes under the control of the human keratin 14 pro- 1987). It is now well established that DNA is a major chro- moter resulted in epidermal melanocytosis, which produces mophore for UVR-induced cancers. Xeroderma pigmento- and transfers melanins to keratinocytes (Kunisada et al, sum (XP) is characterized by a high frequency of skin 1998). We crossed XPA (/) mice and SCF-Tg mice, and cancers, including melanoma, on sun-exposed areas of the generated XPA (/), SCF-Tg mice (Yamazaki et al, 2004). body as a result of a defect in the early steps of nucleotide- These mice possess a hairless and deeply black skin. The epidermal melanin in the mice revealed a high inhibitory activity against the enhanced photosensitivity of XPA (/) Abbreviations: HE, hematoxylin and eosin; LMM, lentigo maligna mice. Although the epidermal melanocytes of XPA (/), melanoma; NM, nodular melanoma; SCC, squamous cell carcinoma; SCF, stem cell factor; UVR, ultraviolet radiation; XP, xeroderma SCF-Tg mice can sufﬁciently protect keratinocytes against pigmentosum UVR; melanocytes in these mice may be more sensitive

Copyright r 2005 by The Society for Investigative Dermatology, Inc. 521 522 YAMAZAKI ET AL THE JOURNAL OF INVESTIGATIVE DERMATOLOGY than those in XPA ( þ / þ ) mice because of a defect in the nucleotide excision repair of UVR-induced DNA damage. In this study, an effort was made to induce cutaneous melanoma by repeated exposures of a higher dose of UVR onto XPA (/), SCF-Tg mice.1

Results and Discussion

Non-treated XPA (/), SCF-Tg mice, and XPA-normal, SCF-Tg mice had black skin and hairs (Fig 1A, B). The epidermal cells consisted of keratinocytes and clear cells with abundant melanin in hematoxylin and eosin (HE)-stained sections (Yamazaki et al, 2004). The latter were positive to anti-S100 antibody immunohistochemically, thus consistent with melanocytes or Langerhans cells (Fig 1C). No cells in the epidermis or dermis are positive to anti-melanoma an- Figure 1 Clinical and histological features of non-treated xeroderma tibodies (anti-HMB45 antibody and anti-MART-1 or Melan A pigmentosum (XP)A (/), stem cell factor-transgenic (SCF)-Tg antibody) on immunohistochemical study (Fig 1D). mouse and XPA normal ( þ / þ ), SCF-Tg mouse. XPA (/); SCF-Tg The mice at 8–10 wk of age were irradiated with UVB at a mouse has uniformly black skin (A) and XPA ( þ / þ ), SCF-Tg mouse 2 has black hair and skin (B). Keratinocytes contain abundant melanin in dose of 5 J per cm on the back three times a week for 10 hematoxylin and eosin staining (Yamazaki et al, 2004), and clear cells 2 wk (total dose: 150 J per cm ). At the termination of the are positive for anti-S100 antibody in the immunohistochemical staining exposure protocol, scaling, crusting, and/or ulceration was (C). In non-treated skin, neither epidermal nor dermal cells are positive observed on the irradiated back areas. At 4 mo after the for anti-melanoma antibody (D). C and D are of the same magnification, and the bar in D indicates 200 mm. termination of UVR exposure, an unevenly pigmented macule with an irregular border appeared on the depigmented LMM were induced on the re-epithelized skin after severe skin that had re-epithelized after ulceration (Fig 2A). Histo- burning, whereas NM appeared on the skin without pre- logically, a proliferation of large, pleomorphic melanocytes ceding ulceration. Six of eight mice with NM developed was found at the epidermal–dermal junction, revealing the multiple tumors, and the remaining two bore a solitary findings of lentigo maligna melanoma (LMM) in the HE- tumor. Four mice developed skin tumors that were histo- stained sections (Fig 2B). When these skin lesions were left logically confirmed to be SCC revealing proliferation of untreated, the black macules became elevated tumors with atypical keratinocytes with frequent mitotic figures, during a ulceration and an irregular border (Fig 2C, white arrow). On period of 4–8 mo after the discontinuation of UVR exposure histology, tumor cells with abundant melanin were found to (Fig 2E, dotted white arrow, Fig 2G). Three of these also have invaded vertically into the deep dermis (Fig 2D). At developed NM associated with SCC, but the remaining 6 mo, some of the other mice developed black nodules on mouse had NM alone. No tumor was induced in the non- the back (Fig 2E, white arrow) and ulcerated tumors ap- irradiated XPA (/), SCF-Tg mice, or the irradiated SCF-Tg peared (Fig 2E, dotted white arrow). Histology of the black (XPA normal ( þ / þ )) mice. Animals with skin melanomas nodules revealed the findings of nodular melanoma (NM), were euthanized at various time points. The axillar and in- with a proliferation of atypical melanocytes in the dermis in guinal lymph nodes were collected for histological exami- connection with the epidermis (Fig 2F). Tumor cells were so nation. Melanoma metastasis was found in four of 12 mice heavily pigmented that bleaching was required for with LMM and in seven of eight with NM (Fig 4A and B). immunohistochemical observations (Fig 3D–F). Monoclon- There were possible melanocytes bearing melanin granules al antibodies revealed that they were positive for the S-100 in lymph nodes in the HE-stained section from non-irradi- protein, melanoma antibodies (a mixture of anti-HMB-45 ated mice. But they were negative immunohistochemically antibody and MART-1 or Melan A antibody), and the Ki-67 for anti-melanoma antibodies. Two pathologists experi- antibody (Fig 3D–F). The Ki-67 positive ratio was 18.1% in enced with melanoma evaluated the HE and immunohisto- NM and 9.8% in LMM. Basal melanocytes of the non-ex- chemically stained sections. posed XPA (/), SCF-Tg mouse were positive only for the Walker reviewed that the development of animal models S-100 protein. The number of mice bearing LMM or NM of melanoma has been undertaken by a number of inves- gradually increased over time. At the end of the observation tigators (Walker et al, 2002). There have been few previous period of 58 wk after the start of UVR exposure, 12 and reports of melanoma induction in experimental animals by eight of 61 irradiated mice had developed LMM and NM, UVR alone. Melanoma was induced in platyfish–swordtail respectively. Twenty-three mice died before developing hybrid fish by UVA and a visible range of light, which are not tumors because of persistent severe sunburn. Consequent- absorbed directly in DNA (Setlow et al, 1993). The question ly, 33% (20 of 61) of irradiated mice, and 53% (20 of 38) of remains, however, whether these experimental data are di- the survivors developed melanoma. Interestingly, individual rectly applicable to human melanoma since there are great mice developed either LMM or NM, but not both. All of the differences in the structure of the skin and in the environ- ment between fish and humans. In terms of a mammalian 1Yamazaki F, Okamoto H, Matsumura Y, Tanaka K, Kunisada T, Horio T: Development of a new mouse model of UVB-induced model, the South American opossum, Monodelphis domes- melanoma. Pigment Cell Res 17: 684–685, 2004 (abstr). tica, was exposed to UVR three times a week for 70 wk, and 125 : 3 SEPTEMBER 2005 NEW MOUSE MODEL OF UVB-INDUCED MELANOMA 523 developed melanotic tumors in the dermis at 100 wk after the first exposure (Ley et al, 1989). Since this animal possesses a photoreactivation repair pathway for DNA damage, the induction of melanocyte hyperplasia was sup- pressed by the exposure of photoreactivating light (4320 nm) after UVR, indicating that DNA damage by UVR is linked to the melanoma development. The mouse has been used as a useful and convenient experimental animal to study the effect of UVR on non-melanoma skin cancers. To investigate the role of UVR on melanomagenicity, it is help- ful to generate an animal model that develops melanoma after UVR exposure. The wild-type mouse, however, has not been a suitable model for melanoma investigation, since the mouse does not have epidermal melanocytes, although the colored mouse has melanocytes in the dermis and hair follicle. In inbred mice, melanoma has been induced by UVR in combination with chemical carcinogens (Romerdahl et al, Figure 3 1989; Hussain et al, 1991). Recently, genetically engineered Immunohistochemical staining of skin tumor specimens after a mouse models have been established for the study of me- bleaching treatment of melanin. Lentigo maligna melanoma (LMM) (A–C) and nodular melanoma (NM) (D–F) were positive for the anti-S100 lanomagenesis. In some of them, melanomas developed antibody, anti-melanoma antibodies, and the anti-Ki67 antibody show- ing brownish staining. Squamous cell carcinoma (SCC) (G–I) was negative for the anti-S100 and anti-melanoma antibodies, and positive for the anti-Ki67 antibody. All photographs are of the same magnification, and the bar in Fig 1 indicates 200 mm.

spontaneously, but were enhanced by UVR (hepatocyte growth factor/scatter factor (HGF/SF mouse)) (Noonan et al, 2000), and in other mice melanomas occurred only when chronically exposed to UVR, but from dermal melanocytes (Kelsall et al, 1998; Powell et al, 1999). Until recently, the responsible spectrum of UVR for melanomagenesis has not been experimentally identiﬁed. De Fabo revealed that UVB but not UVA radiation-initiated melanomas in HGF/SF mouse (De Fabo et al, 2004). In this experiment, the mice were exposed mainly to UVB, since XPA (/) mice have a defect in repair of DNA damage of cyclobutane pyrimidine dimmer-type that is induced by UVB. The melanin pigment

Figure 2 Lentigo maligna melanoma (LMM), nodular melanoma (NM), and squamous cell carcinoma (SCC) developed on the irradiated skin of the xeroderma pigmentosum (XP)A (/), stem cell factor- transgenic (SCF)-Tg mouse. Black macules with irregular border (arrow) appeared on the re-epithelized, depigmented back skin of the ultraviolet (UV)-irradiated mouse at 4 mo after termination of UVR exposure (a total amount of 150 J per cm2 UVB) (A), and histology revealed the findings of LMM with large and atypical cells with abundant melanin mainly in the epidermal–dermal junction on hematoxylin and eosin (HE) staining (B). Elevated tumors with ulceration and an irregular border developed on the black macule (C, white arrow). In histology, tumor cells with abundant melanin invaded vertically into the deep dermis (D). In other mice, nodules developed on the irradiated skin without preceding ulcer and depigmentation (white arrow) and an ulcerated tumor appeared after 6 mo of UVR (E, dotted white arrow). Histological findings of the solid black tumors were compatible with those of NM, revealing a proliferation of melanocytes with massive melanin down into the dermis in connection with the epidermis on HE staining (F). The ulcerated tumor was histologically diagnosed as SCC composed of atypical keratinocytes without melanin (G). (H) Melanoma-occurrence percentage of the mice. The x-axis indicates the time course after the termination of UVB exposure. The y-axis indicates the percentage of melanoma-developed mice. About 55% of UV-treated XPA (/), SCF- Tg mice developed melanoma at 70 wk after UVB radiation. UV-treated XPA-normal, SCF-Tg mice and non-treated XPA (/), SCF-Tg mice did not develop melanoma. The bar in B indicates 200 mm. D and F are of the same magnification, and bars indicate 1.5 mm. The bar in G indicates 100 mm. 524 YAMAZAKI ET AL THE JOURNAL OF INVESTIGATIVE DERMATOLOGY

melanoma. The dose of UV was very high compared with models that have been established by other investigators. This discrepancy seems to be mainly because of extremely dark skin with heavily pigmented epidermal keratinocytes and melanocytes in the present mice. The severe inflammation, which was induced by repeated exposures to high doses of UVR, may be partly involved in melanoma progression, as it is widely accepted that inflammation is a critical factor in tumor progression (Coussens et al, 2002). We rarely observed SCC in DNA repair-deficient mice even after exposures to a huge amount of UVB, indicating that melanocytes effectively protect keratinocytes against malignant transformation. The mouse generated in this study is a new animal model of melanoma in that melanoma occurred from epidermal melanocytes only after and as a result of experimental UVR exposure. Although melanoma was experimentally induced in the mice by exposure to an unusually high dose of UVB, we are currently generating XPA (/), SCF-Tg mice with a lighter skin color for clin- ically relevant investigations. This is the initial step toward establishing a model of melanoma that will allow close investigation of the development of the pathology in connection to light exposure. There have been several reports that mutation patterns were different (deletion, insertion, and point mutation) among clinical stages of melanomas in the tumor suppressor genes (Kannan et al, 2003). We are going to analyze the mutation of tumor suppressor genes in melanomas experimentally developed in XPA (/), SCF-Tg mice. In conclusion, we have generated a mammalian model that can develop melanocyte tumor after UVR exposure with proliferating and metastasizable ability.

Figure 4 Materials and Methods Histology of a metastatic lesion in the lymph nodes. (A) Atypical large cells with abundant melanin were observed in a hematoxylin and Mice Hairy XPA (/) mice with a CBA, C57BL/6, and CD-1 chi- eosin-stained section of an axillar lymph node from the mice that had developed nodular melanoma. (B) The atypical cells were positive for meric genetic background (Nakane et al, 1995) were backcrossed the melanoma antibodies on immunohistochemical staining after a with hairless albino mice of the inbred strains Hos/HR-1, and the bleaching treatment of melanin. All photographs are of the same mag- resultant hairless XPA (/). A hairy SCF-Tg mouse has a C57BL/6 nification, and the bar in Fig 1 indicates 200 mm. and SLJ background (Kunisada et al, 1998). XPA (/) mice and SCF-Tg mice were crossed and the resultant hairy XPA ( þ /), SCF-Tg mice were generated. Thereafter, offspring of XPA ( þ /), is classified into two types: eumelanin and pheomelanin. SCF-Tg mice, and hairless XPA (/) mice were backcrossed, re- Some in vitro investigations indicate that pheomelanin has sulting in hairless XPA (/), SCF-Tg mice (Yamazaki et al, 2004). cytotoxicity by UVA radiation and relate to UVA carcino- We used 61 XPA (/), SCF-Tg mice in this experiment. As control genesis of melanoma (Jimbow et al, 1991). We have not mouse, we used 20 XPA normal, SCF-Tg mice (Kunisada et al, analyzed the type of melanin in this experiment. 1998) and unexposed 20 XPA (/), SCF-Tg mice. All mice were We previously reported that a defect in the nucleotide 8–10 wk of age at the beginning of each experiment. The maximum observation interval was 2 y. Mouse experiments were approved excision repair of damaged DNA in XPA (/) mice causes by the Kansai Medical University Subcommittee on Research an extreme susceptibility to UVR-induced skin cancers Animal Care. (Nakane et al, 1995), enhanced immunosuppression (Mi- yauchi-Hashimoto et al, 1996; Kuwamoto et al, 2000; Mi- Induction of skin tumors by UV irradiation The UVB source was yauchi-Hashimoto et al, 2001), and inactivation of natural a bank of fluorescent sunlamps (FL.20SE.30; Toshiba Medical Supply, Tokyo, Japan) that emit approximately 55% of their radi- killer cells (Miyauchi-Hashimoto et al, 1999). The cancers ation within the UVB range peaking at 305 nm, 25% and less than induced were SCC that developed from epidermal keratin- 1%, within UVA and UVC, respectively. The irradiance of UVB was ocytes. XPA (/), SCF-Tg mice with epidermal me- measured by a radiometer (UVR-305/365D (II); Toshiba Medical lanocytes were highly resistant to UVR, and developed Supply). XPA (/) SCF-Tg mice, XPA normal, and SCF-Tg mice neither SCC nor MM even after repeated exposures to UVB were irradiated with 5 J per cm2 of UVB on the back three times a for 30 wk at a total dose of 72 J per cm2 (Yamazaki et al, week. 2004). In this study, the mice were repeatedly exposed to Histological examination of tumors Skin samples and tumor 2 2 5 J per cm of UVB at a total dose of 150 J per cm over samples were prepared from mice. A 5 mm section was stained 10 wk and resulted in the frequent development of with HE. For the immunohistochemical analysis, melanin in the 125 : 3 SEPTEMBER 2005 NEW MOUSE MODEL OF UVB-INDUCED MELANOMA 525 deparaffinized sections was bleached by 0.25% potassium per- Kelsall SR, Mintz B: Metastatic cutaneous melanoma promoted by ultraviolet manganate for 1 h, and 2% oxalic acid for 1 min. The specimens radiation in mice with transgene-initiated low melanoma susceptibility. were reincubated for 10 min in 10% normal goat serum, and treat- Cancer Res 58:4061–4065, 1998 ed with target retrieval solution (DAKO, Carpinteria, California) and Kraemer KH, Lee MM, Andrews AD, Lambert WC: The role of sunlight and DNA incubated for 12 h at 41C with monoclonal antibodies including repair in melanoma and nonmelanoma skin cancer. The xeroderma pig- anti-melanoma antibody (anti-HMB45 antibody and anti-MART-1 mentosum paradigm. Arch Dermatol 130:1018–1021, 1994 or Melan A antibody, ab732, abcam, Cambridge, UK), anti-S100 Kunisada T, Lu SZ, Yoshida H, et al: Murine cutaneous mastocytosis and epidermal melanocytosis induce by keratinocyte expression of transgenic antibody (ab4066, abcam, Cambridge, UK), and anti-Ki-67 anti- stem cell factor. J Exp Med 187:1565–1573, 1998 body (NCL-Ki67-MM1, Novocastra Laboratories, Newcastle, UK). Kuwamoto K, Miyauchi-Hashimoto H, Tanaka K, Eguchi N, Inui T, Urade Y, Horio

After three washes with phosphate-buffered saline (PBS) plus T: Possible involvement of enhanced prostaglandin E2 production in the 0.5% Tween, the sections were incubated for 15 min with a biotin- photosensitivity in xeroderma pigmentosum group A model mice. J Invest conjugated secondary antibody (Vector Laboratories, Burlingame, Dermatol 114:241–246, 2000 California). After three washes with PBS plus 0.5% Tween, the Langley RGB, Sober AJA: A clinical review of the evidence for the role of ultra- slides were treated with diaminobenzidine (Vectastain Elite ABC kit, violet radiation in the etiology of cutaneous melanoma. Cancer Invest Vector Laboratories) as recommended by the manufacturer. Coun- 15:561–567, 1997 terstaining was performed with hematoxylin. As a negative control, Ley RD, Applegate LA, Padilla RS, Stuart TD: Ultraviolet radiation-induced malignant melanoma in Monodelphis domestica. Photochem Photobiol 50: tissue sections were stained with the secondary antibody only. 1–5, 1989 Positive staining is brownish for all antibodies in immunohisto- Miyauchi-Hashimoto H, Kuwamoto K, Urade Y, Tanaka K, Horio T: Carcinogen- chemistry. In each case, a minimum of 1000 tumor cells were ob- induced inflammation and immunosuppression are enhanced in xero- served, and the results were expressed as the percentage of Ki-67 derma pigmentosum group A model mice associated with hyperproduc- positive cells. tion of prostaglandin E2. J Immunol 166:5782–5791, 2001 Miyauchi-Hashimoto H, Okamoto H, Tanaka K, Horio T: Ultraviolet radiation-in- DOI: 10.1111/j.0022-202X.2005.23753.x duced suppression of natural killer cell activity is enhanced in xeroderma pigmentosum group A (XPA) model mice. J Invest Dermatol 112:965–970, Manuscript received November 17, 2004; revised February 10, 2005; 1999 accepted for publication February 15, 2005 Miyauchi-Hashimoto H, Tanaka K, Horio T: Enhanced inflammation and immunosuppression by ultraviolet radiation in xeroderma pigmentosum group A Address correspondence to: Fumikazu Yamazaki, MD, Department of (XPA) model mice. J Invest Dermatol 107:343–348, 1996 Dermatology, Kansai Medical University, Fumizono 10-15, Moriguchi, Nakane H, Takeuchi S, Yuba S, et al: High incidence of ultraviolet-B or chemical Osaka 570-8507, Japan. Email: [email protected] carcinogen-induced skin tumors in mice lacking the xeroderma pigmentosum group A gene. Nature 377:165–168, 1995 Noonan FP, Otsuka T, Bang S, Anver MR, Merlino G: Accelerated ultraviolet radiation-induced carcinogenesis in hepotocyte growth factor/scatter References factor transgenic mice. Cancer Res 60:3738–3743, 2000 Powell MB, Gause PR, Gregus PHJ, Nagle MLR, Bowden GT: Induction of Blum HF: Carcinogenesis by Ultraviolet Light. Princeton, NJ: Princeton University melanoma in TPras transgenic mice. Carcinogenesis 20:1747–1753, Press, 1959 1999 Coussens LM, Werb Z: Inflammation and cancer. Nature 420:860–867, 2002 Romerdahl C, Stephens ALC, Bucana C, Kripke ML: The role of ultraviolet ra- De Fabo EC, Noonan FP, Fears T, Merlino G: Ultraviolet B but not ultraviolet A diation in the induction of melanocytic skin tumors in inbred mice. Cancer radiation initiates melanoma. Cancer Res 64:6372–6376, 2004 Commun 49:209–216, 1989 Elwood JM, Jopson J: Melanoma and sun exposure: An overview of published Serraino D, Fratino L, Gianni W, Campisi C, Pietropaolo M, Trimarco G, Marigliano studies. Int J Cancer 73:198–203, 1997 V: Epidemiological aspects of cutaneous malignant melanoma (review). Fisher MS, Kripke ML: Systemic alteration induced in mice by ultraviolet light Oncol Rep 5:905–909, 1998 irradiation and its relationship to ultraviolet carcinogenesis. Proc Natl Setlow RB, Grist E, Thompson K, Woodhead AD: Wavelengths effective in in- Acad Sci USA 74:1688–1692, 1987 duction of malignant melanoma. Proc Natl Acad Sci USA 90:6666–6670, Hussain Z, Pathak MA, Flotte T, Wick MM: Role of ultraviolet radiation in the 1993 induction of melanocytic tumors in hairless mice following 7, 12-dime- Walker G.J, Hayward NK: Pathways to melanoma development: Lesions from the thylbenz(a)anthracene application and ultraviolet radiation. Cancer Res mouse. J Invest Dermatol 119:783–792, 2002 51:4964–4970, 1991 Yamazaki F, Okamoto H, Miyauchi-Hashimoto H, et al: XPA gene-deficient, SCF- Jimbow K, Salopek TG, Dixon WT, Seareles GE, Yamada K: The epidermal mel- transgenic mice with epidermal melanin are resistant to UV-induced car- anin unit in the pathophysiology of malignant melanoma. Am J De- cinogenesis. J Invest Dermatol 123:220–228, 2004 rmatopathol 13:179–188, 1991 Yoshida H, Hayashi S, Shultz LD, Yamamura K, Nishikawa S, Nishikawa S, Kuni- Kannan K, Sharpless NE, Xu J, O’Hagan RC, Bosenberg M, Chin L: Components sada T: Neural and skin cell-specific expression pattern conferred by of the Rb pathway are critical targets of UV mutagenesis in a murine steel factor regulatory sequence in transgenic mice. Dev Dyn 207: melanoma model. Proc Natl Acad Sci 100:1221–1225, 2003 222–232, 1996