July 1998

Ultraviolet-Induced Acute Histological Changes in Irradiated Nevi Are Not Associated With Allelic Loss

Author Affiliations

From the Department of Dermatology, University Hospital, Zurich, Switzerland (Drs Böni, Matt, and Burg); the Department of Dermatology, Medical University of Lübeck, Lübeck, Germany (Dr Tronnier); and the Department of Pathology, National Cancer Institute, National Institutes of Health, Bethesda, Md (Drs Vortmeyer and Zhuang).


Copyright 1998 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.1998

Arch Dermatol. 1998;134(7):853-856. doi:10.1001/archderm.134.7.853

Background  Transformed melanocytes in atypical nevi, which are thought to be precursors of melanoma, are frequently deleted on chromosomes 1p, 9q, and 9p21 (p16 locus). Single UV irradiation induces histological changes that are similar to those of atypical nevi and, in part, of melanoma in situ.

Objective  To determine the effects of UV irradiation on benign melanocytic nevi in vivo.

Design  We investigated one half of a symmetric nevus 1 week after a single UV exposure with 4 times the patient's minimal erythema dose and compared it with the nonirradiated, shielded half of the same nevus. Two to 3 areas containing 5 to 30 melanocytes in 7 nevi were microdissected (a total of 18 areas in each nonirradiated and irradiated part), followed by a single-step DNA extraction. Extracted genomic DNA was amplified using a polymerase chain reaction with polymorphic markers D1S450 (1p), D9S12 (9q), IFNA, and D9S171 (9p21) and subjected to autoradiography.

Observations  Two, 3, 2, and 2 of 18 areas were homozygous for D1S450, D9S12, IFNA, and D9S171, respectively. No allelic loss could be demonstrated in either nonirradiated or irradiated nevi.

Conclusions  Acute histological changes demonstrated in melanocytic nevi after UV irradiation are not followed by allelic loss on identical chromosomal areas found in dysplastic melanocytes of atypical nevi. This finding supports the hypothesis that initial nonspecific genetic events may occur after UV irradiation, followed by an increase in various repair mechanisms potentially leading to specific genetic damage and loss of heterozygosity; however, loss of heterozygosity is not detectable at an early stage.

ULTRAVIOLET irradiation influences the proliferative and metabolic activity of epidermal melanocytes and has been shown to induce histological changes that in part are similar to those of melanoma in situ.13 Histological changes induced by UV irradiation include an increased number of melanocytes above the basal lamina that are larger in size and have enhanced dendricity.1 These histological patterns are also found in acquired atypical nevi, which are believed to be precursors of melanoma.4,5 Ultraviolet irradiation induces a genetic program designed to repair DNA damage.6 During this response, however, mutations occur that could potentially lead to loss of heterozygosity.

Using a set of microsatellite markers on chromosomes 1p (D1S450), 9q (D9S12), and 9p21 (IFNA and D9S171 ), we demonstrated frequent loss of heterozygosity in microdissected areas containing dysplastic melanocytes, thus suggesting that atypical nevi may potentially be premalignant melanoma lesions.7

We therefore searched for loss of heterozygosity in areas containing altered melanocytes induced by UV irradiation in nevi and compared the frequency of allelic deletion with the nonirradiated halves of the same lesions. The sections of partially irradiated melanocytic nevi that were used in this study were fixed with formalin and embedded in paraffin. In these nevi, acute changes had been induced by UV irradiation. The histopathological changes in these nevi, including an increase in the number of melanocytes above the basal lamina, enlargement of both the cytoplasm and the size of the nuclei, as well as dendricity of these cells, have been previously characterized.1


Patients with melanocytic nevi larger than 5 mm were selected; 7 patients entered the study. All nevi were located on the patients' trunks. The nevi displayed no signs of malignancy either clinically or by surface microscopy.

Irradiation was performed with a high-pressure mercury lamp (Blue-Point, Dr Hönle, Lübeck, Germany) with an emission ranging from 295 to 400 nm. In the UV-B range, the intensity of the lamp was 3.5 mJ/cm2, and in the UV-A range, the intensity was 24 mJ/cm2 (Centra, Osram, Germany). The minimal erythema dose was determined on the back of each patient and ranged between 13 and 33 mJ/cm2 in the UV-B range (4-10 seconds of irradiation time). One half of each nevus was irradiated with 4 times the minimal erythema dose, and the other half was shielded with black tape. The nevi were excised 1 week after the single irradiation. During excision, the lines between irradiated and nonirradiated parts were marked with sutures.


The tissue was fixed in 5% buffered formalin, divided into nonirradiated and irradiated parts, and separately embedded in paraffin. Paraffin sections from both parts of the nevus were cut from the center toward the margin. Sections were stained with hematoxylin-eosin. All lesions were melanocytic nevi of junctional or compound type.


From each specimen, a 4-µm tissue section was obtained for hematoxylin-eosin staining and microdissection. In nonirradiated and irradiated parts of 7 nevi, 2 to 3 areas containing 5 to 30 melanocytes were sampled (a total of 18 areas from both nonirradiated and irradiated parts). Samples from normal cell structures other than melanocytes, eg, sebaceous glands or inflammatory cells, were also obtained in each case from the same slide and processed the same as the dissected melanocytes. Microdissection was performed under light microscope visualization (magnification×200) using a 30-gauge needle.


Procured cells were immediately suspended in 10 µL of buffer with a pH of 8.0 containing 0.05-mol/L Tris–hydrochloric acid, 0.001-mol/L EDTA, 1% polysorbate 20, and 0.1-g/L proteinase K, and were incubated overnight at 37°C. The mixture was boiled for 10 minutes to inactivate proteinase K, and 1.5 µL of this solution was used for polymerase chain reaction (PCR).


The following markers were used to amplify extracted genomic DNA: 1p (D1S450), 1 marker at chromosome 9q (D9S12), and 2 markers at 9p21 (IFNA and D9S171) (Research Genetics, Huntsville, Ala). Polymerase chain reaction was performed in 10-µL volumens containing l µL of×10 PCR buffer (Perkin Elmer, Zurich, Switzerland); 1.5 µL of template DNA; 50 pmol of each primer per liter; 20-nmol/L cytosine triphosphate, guanosine triphosphate, thyrosine triphosphate, and adenosine triphosphate; 0.2 µL of cytosine triphosphate tagged with phosphorus 32 (7.4 Mbq/mL); and 0.1 U of Taq DNA polymerase. After initial heating for 60 seconds, reactions were cycled in a thermal cycler (Gene Amp PCR System 9600, Perkin Elmer) as follows: D1S450, 55°C for 60 seconds and 72°C for 60 seconds for a total of 35 cycles; D9S12, 61°C for 45 seconds and 72°C for 45 seconds for a total of 35 cycles; and IFNA and D9S171, 57°C for 60 seconds and 72°C for 60 seconds for a total of 35 cycles.

Labeled amplified DNA was mixed with an equal volume of formamide loading dye (95% formamide, 20-mmol/L EDTA, 0.05% bromophenol blue, and 0.05% xylene cyanol FF). The samples were denatured for 5 minutes at 94°C and loaded onto a gel consisting of 6% acrylamide (acrylamide-bisacrylamide, 49:1). Electrophoresis was performed on samples at 1800 V for 1 to 2 hours. Gels were transferred to 3 mm of Whatman paper (Merck, Zürich, Switzerland) and dried, and autoradiography was performed with x-ray film (Typon DX-41, Typon, Burgdorf, Switzerland). Loss of heterozygosity was defined as the absence of 1 allele in the tumor DNA evaluated by direct visualization.


In the analyzed nevi, morphologic changes were noted after UV irradiation. In brief, an increased number of melanocytes above the basal lamina, enlargement of both the cytoplasm and the size of the nuclei, as well as dendricity of these cells were found. In this study, cells from both nonirradiated and irradiated areas from the same nevus were selectively procured and genetically analyzed. From irradiated nevi, areas containing atypical melanocytes were procured. No allelic loss could be demonstrated in either nonirradiated or irradiated nevi using D1S450, D9S12 (Figure 1 and Figure 2), IFNA, and D9S171 . Two, 3, 4, and 2 areas were homozygous (noninformative) for D1S450, D9S12, IFNA, and D9S171, respectively (Table 1).

Figure 1.
Image not available

Comparison of the UV-exposed half and the nonexposed covered half of a symmetrical melanocytic nevus of a 58-year-old patient, located on the trunk. A, Half of the nevus protected from irradiation before microdissection. Grouped melanocytes or nevus cells are located mainly in the basal layer of the epidermis. B, Half of the nevus protected from irradiation after microdissection. C, Half of the nevus 1 week after irradiation with double the patient's minimal erythema dose before microdissection. An increased number of melanocytes or nevus cells with enlargement of the cytoplasm and the size of nuclei are found. D, Half of the nevus 1 week after irradiation with double the patient's minimal erythema dose after microdissection.

Figure 2.
Image not available

Results of loss of heterozygosity analysis in the half of the nevus protected from irradiation (A) and the half of the nevus 1 week after irradiation with double the patient's minimal erythema dose (B). The results correspond to the histological section shown in Figure 1. No loss of heterozygosity could be found using markers D1S450, D9S12, IFNA, and D9S171 for either the protected or irradiated halves of the nevus. The marker D9S12 was homozygous (noninformative) in this section and shows only 1 band.

Image not available
Comparison of Allelic Loss in Nonirradiated and Irradiated Parts of 7 Melanocytic Nevi Using Different Markers

Clinical, epidemiological, and biological observations showing cellular and molecular alterations have supported the hypothesis that UV light is an etiologic agent for the development of cutaneous melanoma.810 The role of UV irradiation in the production of epidermal melanocytes was further emphasized by the observation that melanocytes from a transgenic cell line became anchorage independent, formed foci, and yielded malignant melanomas in graft hosts after exposure to a single low–UV-B dose.11 It has been documented that UV irradiation may also alter the morphologic characteristics of melanocytes, producing a higher melanin content, enhanced dendricity, and prominent perikaryons both in vivo and in vitro.1,2,12 An increased frequency of tumor suppressor gene deletion has been shown to be associated with the process of melanocytic transformation in clinically atypical nevi13 that show similar morphologic changes compared with nevi after UV irradiation.

In this study, lesions from sections of irradiated and nonirradiated symmetrical nevi were compared for allelic loss using a panel of different microsatellite markers. In the irradiated parts, sections showing transformed melanocytes on histological examination were selected. As previously described,1 we chose nevi that displayed histological changes after 1 week following irradiation. Since melanoma may originate from epidermal melanocytes as well as from melanocytes in the papillary and upper dermis,14 atypical melanocytes at these locations were microdissected. In general, UV-B penetrates up to 500 µm, while UV-A may penetrate up to 1000 µm,15 thus affecting the dissected levels.

Our results demonstrate that the acute changes in melanocytic nevi induced by UV irradiation, ie, an increased number of melanocytes, enlargement of both the cytoplasm and the size of the nuclei, as well as dendricity of these cells, are not accompanied by loss of heterozygosity determined by markers on chromosome 1p (D1S450), 9q (D9S12), and 9p21 (IFNA and D9S171 ). It is possible that the postirradiation interval of only 1 week was not sufficient for the development of atypia and also for the potential loss of heterozygosity. Furthermore, it cannot be ruled out that UV irradiation with 4 times the patient's minimal erythema dose could cause small mutations in these nevi, which would be undetectable with the method we used. Because the atypical changes described in irradiated nevi have been demonstrated to be transient,1 it is possible that the usual response to UV irradiation in mature nevi does not involve any permanent alteration of the DNA, and that the genetic events that lead to the transformation of some but not most nevi are rare events. Loss of heterozygosity may be an event that follows initiating genetic events or occurs de novo after the development of morphologic atypia but before the onset of frank malignancy.

Back to top
Article Information

Accepted for publication April 1, 1998.

This work was supported in part by the Cancer League of the Canton of Zürich, Zürich, Switzerland, and the Deutsche Krebshilfe, Germany.

Reprints: Roland Böni, MD, Department of Dermatology, University Hospital, Gloriastreet 31, 8091 Zürich, Switzerland.

Tronnier  MSmolle  JWolff  HH Ultraviolet irradiation induces acute changes in melanocytic nevi. J Invest Dermatol. 1995;104475- 478Article
Friedmann  PSGilchrest  BA Ultraviolet radiation directly induces pigment production by cultured human melanocytes. J Cell Physiol. 1987;13388- 94Article
Rosdahl  IK Local and systemic effects on the epidermal melanocyte population in UV-irradiated mouse skin. J Invest Dermatol. 1979;73306- 309Article
Swerdlow  AJEnglish  JMacKie  RM  et al.  Benign melanocytic naevi as a risk factor for malignant melanoma. Br Med J (Clin Res Ed). 1986;2921555- 1559Article
Nordlund  JJKirkwood  JForget  BM  et al.  Demographic study of clinically atypical (dysplastic) nevi in patients with melanoma and comparison subjects. Cancer Res. 1985;451855- 1861
Murli  SWalker  GC SOS mutagenesis. Curr Opin Genet Dev. 1993;3719- 725Article
Park  WSDuray  PHPack  S  et al.  Allelic deletion at chromosome 9p21(pl6) and 17p13(p53) in microdissected dysplastic nevus. Hum Pathol. 1998;29127- 130Article
Ambach  WBlumthaler  M Biological effectiveness of solar UV radiation in humans. Experientia. 1993;49747- 753Article
Elder  DE Human melanocytic neoplasms and their etiologic relationship with sunlight. J Invest Dermatol. 1989;92297S- 303SArticle
Kripke  ML Effects of UV radiation on tumor immunity. J Natl Cancer Inst. 1990;821392- 1396Article
Larue  LDougherty  NMintz  B Genetic predisposition of transgenic mouse melanocytes to melanoma results in malignant melanoma after exposure to a low ultraviolet B intensity nontumorigenic for normal melanocytes. Proc Natl Acad Sci U S A. 1992;899534- 9538Article
Rosdahl  IKSzabo  G Mitotic activity of epidermal melanocytes in UV-irradiated mouse skin. J Invest Dermatol. 1978;70143- 148Article
Böni  RZhuang  ZAlbuquerque  AVortmeyer  ADuray  P Loss of heterozygosity detected on 1p and 9q in microdissected atypical nevi. Arch Dermatol. 1998;134882- 883Article
Clark  WHElder  DGuerry  DEpstein  MGreene  MVan Horn  M A study of tumor progression: the precursor lesions of superficial spreading and nodular melanoma. Hum Pathol. 1984;151147- 1165Article
Potten  CJ Radiation and Skin.  London, England Taylor & Francis1985;41