Melanin content of cultured human melanocytes and UV-induced cytotoxicity.

Cultured melanocytes originating from persons with different skin phototypes were utilized for measurement of endonuclease sensitive sites induced by UVB and the determination of cell survival after UVA or UVB irradiation. During culture, the melanocytes largely maintained their phenotypic characteristics according to their original skin phototype. Total melanin concentrations were 4.9 times higher in the darker skin phototype (IV-VI) melanocytes when compared to the cells from lighter skin phototypes (I-III). Also phaeomelanin contents were higher (2.5 times) in the skin phototype (IV-VI) melanocytes which implies that the cells from light skin types contain less melanin, but a relatively high proportion of phaeomelanin. After UVB irradiation a stronger induction of endonuclease sensitive sites was found for melanocytes with a lower level of total melanin and a high content of pheomelanin. By measuring the clone forming ability in different melanocyte cultures after UVB irradiation, significant better survival was found in case of the cells with the higher melanin content. Despite the large variations in melanin content, no significant difference in survival after UVA irradiation could be demonstrated in this way. Our results suggest a protective effect of melanin for UVB and indicate the importance of the measurements of melanin content and composition when different parameters of UV-induced damage are studied in melanin producing cells.

Low doses of UVB or UVA induce chromosomal aberrations in cultured human skin cells.

Chromosomal defects are frequently present in malignant and premalignant skin disorders; however, it is not known whether ultraviolet radiation from sunlight plays a role in their induction. To obtain information on the ability of ultraviolet A and ultraviolet B to induce chromosomal aberrations, cultured melanocytes and fibroblasts were exposed to physiologic doses of ultraviolet A or ultraviolet B and, for comparison, to gamma rays. As a measure of chromosomal aberrations, the formation of micronuclei was determined. To obtain sufficient statistical data on induced micronuclei and cell kinetics, a flow cytometry method has been modified and applied. The flow cytometry method analysis is based on staining the DNA with ethidium bromide and the cell membranes with 1,6-diphenyl-1,3,5,-hexatriene. We observed dose-dependent micronuclei formation after gamma or ultraviolet B irradiation in both cell types and also for ultraviolet A in fibroblasts. The yield of micronuclei induced in fibroblasts by ultraviolet A was only a factor 15 smaller than that induced by ultraviolet B (313 nm). The results indicate that 10 kJ per m2 (equivalent to 1 minimal erythema dose) of ultraviolet B and 150 kJ per m2 of ultraviolet A (0.2 minimal erythema dose) can induce 1% of micronuclei in fibroblasts, equivalent to the induction due to 0.6 Gy of gamma radiation. In conclusion, physiologic doses of sunlight can induce chromosomal aberrations at a level comparable with that observed after exposure to approximately 1 Gy of ionizing radiation. Therefore, sunlight can be considered a potential inducer of chromosomal aberrations in skin cells, which may contribute to skin carcinogenesis.

(Pheo)melanin photosensitizes UVA-induced DNA damage in cultured human melanocytes.

The question of whether melanins are photoprotecting and/or photosensitizing in human skin cells continues to be debated. To evaluate the role of melanin upon UVA irradiation, DNA single-strand breaks (ssb) were measured in human melanocytes differing only in the amount of pigment produced by culturing at two different concentrations, basic (0.01 mM) or high (0.2 mM), of L-tyrosine, the main precursor of melanin. In parallel, pheo- and total melanin contents of the cells were determined. Identical experiments were performed with two melanocyte cultures derived from a skin type I and a skin type VI individual. For the first time the correlation between UVA-induced genotoxicity and pheo-/total melanin content has been investigated. We observed that cultured in basic medium, the skin type VI melanocytes contained 10 times more total melanin and about seven times more pheomelanin than the skin type I melanocytes. Elevation of tyrosine level in the culture medium resulted in an increase of both pheo- and total melanin levels in both melanocyte cultures; however, the melanin composition of skin type I melanocytes became more pheomelanogenic, whereas that of skin type VI melanocytes remained the same. The skin type VI melanocytes cultured in basic medium demonstrated a very high sensitivity (1.18 ssb per 10(10) Da per kJ per m2) toward UVA that is probably related to their high pheo- and total melanin content. Their UVA sensitivity, however, did not change after increasing their melanin content by culturing at high tyrosine concentration. In contrast, the skin type I melanocytes demonstrated a low sensitivity (0.04 ssb per 10(10) Da per kJ per m2) toward UVA when cultured in basic medium, but increasing their melanin content resulted in a 3-fold increase in their UVA sensitivity (0.13 ssb per 10(10) Da per kJ per m2). These results demonstrate that UVA-irradiated cultured human melanocytes are photosensitized by their own synthesized chromophores, most likely pheomelanin and/or melanin intermediates.

Ultraviolet irradiation corrects vitamin D deficiency and suppresses secondary hyperparathyroidism in the elderly.

The objective of this study was to compare the effect of ultraviolet radiation (UV) and oral vitamin D3 on the vitamin D status and parathyroid hormone (PTH) concentration in elderly nursing home patients. The design of the study was a randomized clinical trial. The setting was a psychogeriatric nursing home. Subjects included 45 female psychogeriatric patients with a mean age of 85 years. Exclusion criteria were going outdoors more than once a week and the presence of actinic or cancer skin lesions. Intervention was random allocation of UV-B irradiation at half the minimal erythemal dose of the lower back, three times per week during 12 weeks (UV-B), or oral vitamin D3 400 IU/day during 12 weeks (VIT-D), or no treatment (CONTR). Main outcome measures were change in fasting serum levels of vitamin D metabolites at 0, 2, 4, 8, and 12 weeks in the treatment groups, compared with the control group. PTH(1-84) was measured at 0 and 12 weeks. Baseline serum 25-hydroxyvitamin D (25(OH)D) was lower than 30 nmol/l in 95% of the participants. It increased to a median value of around 60 nmol/l after 12 weeks both in the UV-B and VIT-D groups, whereas there was no change in the CONTR group. Serum 1,25-dihydroxyvitamin D increased significantly in the UV-B group. Serum calcium increased significantly in both treatment groups. Serum PTH decreased more than 30% in both treatment groups (p < 0.001), whereas there was no significant change in the control group. Irradiation with UV-B in the very elderly for a few minutes per day leads to adequate improvement of the vitamin D status. It is as effective as oral vitamin D3 in increasing serum 25(OH)D and suppressing secondary hyperparathyroidism.

Determination of pheomelanin by measurement of aminohydroxyphenylalanine isomers with high-performance liquid chromatography.

We describe an improved method for the analysis of pheomelanin in biological samples. The method is based on a chemical degradation of the melanin polymer and HPLC analysis of specific degradation products. Hydriodic hydrolysis provides 4-amino-3-hydroxyphenylalanine (AHP) and 3-amino-l-tyrosine (AT) which are detected with an electrochemical detector. We have examined each step of the analysis and the results are presented in this paper. First the samples are hydrolyzed for 16 h. AT and AHP are then isolated from the hydrolysates by ion-exchange chromatography and then separated and quantitated by HPLC and electrochemical detection. The method shows good reproducibility with a total imprecision below 5.6%. The linearity of the method was shown from 0 to 490 ng AT and 0 to 850 ng AHP per sample, using a melanoma cell suspension (27 mg protein/ml) with up to 24-fold dilutions of the original sample. For cultured "normal" human melanocytes a minimal amount of 0.1 mg protein is sufficient for analysis of pheomelanin in the samples. This method provides the opportunity to study the composition of the formed melanin in cell lines, cultured in different growth media.

Cytotoxic interactions of Zn2+ in vitro: melanoma cells are more susceptible than melanocytes.

Previous studies have shown that sensitivity to high extracellular levels of Zn2+ is a general feature of cells in vitro and that a prerequisite of the toxic action of zinc is entry into cells via channels that are shared with iron or calcium. As the biochemical and toxicological behaviour of zinc chelate complexes could be different from that of free Zn2+, the effect of chelating agents on zinc transport into human melanoma cell lines was tested. EDTAcal and tetracycline reduced the toxic action of zinc ions in vitro, whereas phenytoin and diethyldithiocarbamate potentiated its effects. D-penicillamine, an effective chelator of zinc in vivo, also exerted a protective action in vitro. Comparison of sensitivity to Zn2+ in vitro between human melanoma lines and several lines of pigment cells from skin of various origins demonstrated that melanoma cells are killed by zinc ions at concentrations which are only partially toxic for normal pigment cells. This is consistent with the repeatedly observed high uptake of 65Zn by melanoma cells.

Physiological doses of ultraviolet irradiation induce DNA strand breaks in cultured human melanocytes, as detected by means of an immunochemical assay.

An immunochemical assay, i.e. sandwich enzyme-linked immunosorbent assay, has been modified to detect UV-induced damage in cellular DNA of monolayer-grown human melanocytes. The method is based on the binding of a monoclonal antibody to single-stranded DNA. The melanocytes derived from human foreskin of skin type II individuals were suspended and exposed to UVA, UVB, solar-simulated light or gamma-rays. Following physiological doses of UVA, UVB or solar-simulated light, a dose-related DNA unwinding comprising a considerable number of single-strand breaks (ssb) was observed. No correlation was found between different seeded cell densities or different culturing periods and the UVA sensitivity of the cells. After UVA irradiation, 0.07 ssb/10(10) Da/kJ/m2 were detected and after UVB irradiation 1.9 ssb/10(10) Da/kJ/m2 were seen. One minimal erythema dose of solar-simulated light induced 2.25 ssb/10(10) Da. Our results from melanocytes expressed in ssb/Da DNA are comparable and have the same sensitivity toward UVA as well as toward UVB as nonpigmented skin cells. As low doses of UVA have already been shown to induce detectable numbers of ssb, this assay is of great interest for further investigations about the photoprotecting and/or photosensitizing effects of melanins in human melanocytes derived from different skin types.

Comet assay demonstrates a higher ultraviolet B sensitivity to DNA damage in dysplastic nevus cells than in common melanocytic nevus cells and foreskin melanocytes.

We used the single cell gel electrophoresis assay (comet assay) to study ultraviolet B (UVB)-induced DNA damage in pigment cells. This assay detects DNA damage, mainly DNA strand breaks and alkali labile sites in the DNA molecule. We studied the effect of biologically relevant doses (comparable to 2-3 MED (minimal erythemal dose) for in vivo irradiated full-thickness skin) of monochromatic UVB light of 302 nm on cultured melanocytes derived from foreskin, common melanocytic nevi, and dysplastic nevi. We were able to demonstrate a linear dose-response relationship between UV dose and the migration coefficient of the comet tail in all three types of pigment cells. Nevus cells originating from dysplastic nevi showed the highest sensitivity to UVB irradiation: 65% higher induction of DNA damage compared to the induction in foreskin melanocytes. Common melanocytic nevus cells were most resistant and showed a 30% lower induction of DNA damage in comparison to foreskin melanocytes. Differences in chromatin structure and cell cycle profile may influence the results of the comet assay. Control experiments with x-ray irradiation, which is well known to produce direct DNA strand breaks via radical formation, revealed only small differences between the three types of melanocytic cells. It is unlikely, therefore, that intrinsic nuclear characteristics may account for the observed differences.

Comparison of melanocytes and keratinocytes in ultraviolet-induced DNA damage per minimum erythema dose sunlight: applicability of ultraviolet action spectra for risk estimates.

Action spectra are being used in risk estimates for ultraviolet (UV) damage. The purpose of our investigation was to compare the susceptibility of cultured melanocytes and keratinocytes to UV-induced DNA damage per minimum erythema dose (MED) and to determine whether the predictions made with action spectra agree with the damage actually induced. Genetic damage was measured as the number of T4-endonuclease V-sensitive sites (ESS). Predictions made with the action spectrum for the induction of DNA damage in melanocytes after irradiation with sunlight and a solar simulator were 15.9 and 13.2 ESS per 10(8) daltons per MED, respectively; with the action spectrum for the induction of DNA damage in keratinocytes the predictions were 12.1 and 9.8 ESS per 10(8) daltons per MED, respectively. To determine the actual damage per MED, cultured cells were irradiated with sunlight or a solar simulator, and MED was determined with an erythema UV meter. The induction of DNA damage in melanocytes after sunlight and solar simulator irradiation was 8.01 and 6.7 ESS per 10(8) daltons per MED, respectively, and in keratinocytes 7.49 and 7.12 ESS per 10(8) daltons per MED, respectively. This was considered to be in agreement with the predicted data. The use of action spectra for risk estimates in melanocytes appears justified.

UV induction of cyclobutane thymine dimers in the DNA of cultured melanocytes from foreskin, common melanocytic nevi and dysplastic nevi.

We compared the induction of cyclobutane thymine dimers after exposure to 302 nm UV in foreskin-derived melanocytes and melanocytes from nevocellular nevi, as well as in melanocytes cultured from dysplastic nevi, precursor lesions of melanoma, derived from four, three and four individuals, respectively. Cyclobutane thymine dimers were quantified in situ by means of an immunofluorescence assay with a specific monoclonal antibody. A method was developed to compare separately performed experiments in a standardized manner. For melanocytes from each source, we demonstrated a linear relationship between UV dose and immunofluorescence. In nevocellular and dysplastic nevi, two subpopulations could be detected, distinguished by their nuclear size. Large nucleated nevocellular nevus cells were most susceptible to the induction of thymine dimers (49% higher induction compared to induction in foreskin melanocytes), while in normal-sized nuclei of these nevus cells the same induction of thymine dimers was found as in nuclei from foreskin melanocytes. In contrast, large nucleated dysplastic nevus melanocytes did not differ from the foreskin melanocytes, while normal-sized nuclei of dysplastic nevus cells showed a lower induction (32% lower induction than in foreskin melanocytes).

The UV action spectra for the clone-forming ability of cultured human melanocytes and keratinocytes.

Melanocytes (skin type 2) and keratinocytes were irradiated with UV light of 254, 297, 302, 312 and 365 nm and the survival was measured. Clone-forming ability was chosen as the parameter for cell survival. Melanocytes were found to be less sensitive to UV light than keratinocytes (a difference of a factor 1.22-1.92 for the UV-C and UV-B wavelengths (254, 297, 301 and 312 nm) and a factor 6.71 for the UV-A wavelength (365 nm). Because melanin does not appear to protect against the induction of pyrimidine dimers the difference between melanocytes and keratinocytes in the UV-C and UV-B region could not be explained by the presence of melanin in the melanocytes. The relatively small difference can be explained by the longer cell cycle of melanocytes, which provides more time for the melanocytes to repair UV damage. In the UV-A region the difference between melanocytes and keratinocytes was much larger, suggesting that besides the longer cell cycle some additional factors must be involved in protection against UV-A light.

Pyrimidine dimer induction and repair in cultured human skin keratinocytes or melanocytes after irradiation with monochromatic ultraviolet radiation.

We compared the susceptibilities of cultured melanocytes and keratinocytes to dimer induction in DNA by monochromatic ultraviolet (UV) radiation. Keratinocytes as well as melanocytes were derived from human foreskin, grown as a monolayer in petri dishes, covered with phosphate-buffered saline containing 0.1% glucose, and irradiated. UV irradiation was carried out at 254, 297, and 302 nm as well as with a light source emitting predominantly 312 nm. The induction of pyrmidine dimers was assessed by determination of the number of T4 endonuclease V-sensitive sites (ESS). We found a slightly higher response for dimer induction in melanocytes at 254, 297, and 302 nm; this difference was only significant at the 297-nm wavelength. Action spectra for pyrimidine dimer induction were derived from the exposure-response data obtained. The action spectra mimic to a large degree the action spectra for dimer induction in other cultured mammalian cells. The repair rate during a post-irradiation period lasting up to 24 h was substantially the same for the two cell types. The percentage of T4 endonuclease V-sensitive sites (ESS) remaining 9 and 24 h after irradiation was 45% and 30%, respectively.

'High single-dose' European PUVA regimen also causes an excess of non-melanoma skin cancer.

We report the results of a long-term (12.8 years) follow-up study of the detection of malignant and benign skin tumours in patients with psoriasis, who were treated with PUVA according to the European, 'high single-dose' regimen. A total of 13 squamous cell carcinomas (SCC) and 24 basal cell carcinomas (BCC) were diagnosed in 11 of 260 patients. The incidence of both SCC and BCC was increased in comparison with the general Dutch population. The ratio of SCC to BCC in the general population was 1:8 but was 1:2.5 in our study group. A positive correlation was observed between the development of SCC and the total UVA dosage, the age of the patient at the start of the PUVA treatment and a history of arsenic use. This dose-related increase in the incidence of SCC, reported in studies from the U.S.A., has not been found in earlier European studies. The average time period between the start of PUVA therapy and the diagnosis of the first malignant skin tumour was 6.0 years for SCC and 4.7 years for BCC. Among the 49 benign skin tumours were actinic keratoses, a keratoacanthoma and 'PUVA keratoses', a newly described hyperkeratotic lesion, especially found in PUVA-treated patients.

UVA irradiation induces collagenase in human dermal fibroblasts in vitro and in vivo.

We report the effect of UVA irradiation on collagen metabolism of fibroblasts, including both synthesis of the collagen degrading enzyme collagenase and de novo synthesis of type I collagen as the major structural component of the dermis. For this purpose confluent fibroblast monolayers were irradiated under standardized conditions (5, 15, 35, 60 J/cm2 using UVASUN 3000, Mutzhas, Munich, FRG, and UV source Sellas sunlight type 2.001, Sellas, Gevelsberg, FRG). Subsequently, total RNA was isolated and subjected to dot blot and northern blot analysis using oligolabelled cDNA clones for human type I collagen, collagenase and beta-actin. Collagen type I and beta-actin mRNA levels remained unaltered following irradiation, suggesting that the synthetic pathway of collagen metabolism at the pretranslational level is not affected by short-term UVA irradiation. However, collagenase mRNA was found to be dose-dependently induced in fibroblasts after irradiation, thus probably contributing to the actinic damage to the dermis. These in vitro data were confirmed in vivo using in situ hybridization on frozen sections of biopsy material obtained from UVA irradiated patients.

Differential immune response of congenic mice to ultraviolet-treated major histocompatibility complex class II-incompatible skin grafts.

The influence of ultraviolet (UVB) irradiation on the survival of H-2 class II-disparate skin grafts was studied in congenic mouse strains. Isolated skin was UVB irradiated in vitro at a dose of 40 mJ/cm2 from both sides to remove Ia immunogenicity. Immediately after irradiation the skin was transplanted onto the flank of allogeneic mice. When B10.AQR grafts were transplanted onto B10.T(6R) recipients, a significant prolongation of the survival time was observed, while 50% of the UVB-treated grafts were not rejected at all. However, in the opposite direction--i.e., B10.T(6R) grafts onto B10.AQR recipients, no significant prolongation of the survival was observed. To test whether this effect was due to a difference in susceptibility of the donor skin to UVB irradiation or to a different immune response in the recipients, (B10.T(6R) x B10.AQR) grafts were transplanted onto the parent strains. Similar results were obtained, in that UVB-treated grafts did not show a prolonged survival in B10.AQR recipients, whereas a significant prolongation (50% of the grafts survived more than 100 days) was observed in B10.T(6R) recipients. UVB-treated (B10.T(6R) x B10.AQR)F1 grafts were also transplanted onto (B10.T(6R) x C57B1/10)F1, (B10.AQR x C57B1/10)F1, (B10.T(6R) x Balb/c)F1 and (B10.AQR x Balb/c)F1 recipients--but in none of these combinations was a prolonged survival time observed. These data suggest that, in contrast to all in vitro experiments, the abrogation of the immune response by UVB treatment of the stimulator cells is, in vivo, not a general phenomenon. The genetic constitution of the responder mice seems to play an important role in determining whether or not an immune response takes place.

Risk evaluation of UVB therapy for psoriasis: comparison of calculated risk for UVB therapy and observed risk in PUVA-treated patients.

A descriptive dose-response model is presented to evaluate the long-term risk with respect to non-melanoma skin cancer associated with UVB therapy. The model is based on the results of animal dose-response studies and epidemiological data. A number of factors that influence the risk associated with therapy are evaluated: annual dose applied, solar exposure, therapeutic period, age at start of therapy. Shielding from therapeutic exposures of the skin areas that normally receive the most sun exposure could effectively reduce the risk. The model calculations were compared with observational data, as far as available. The calculated risk corresponded within the limits of statistical error with observed long-term risk of UVB therapy in Sweden; as far as there was a deviation, the model calculations tended to over-estimate the risk. A comparison of the model prognoses for UVB therapy with the observed risk among PUVA-treated patients in the USA shows a much higher observed risk for squamous cell carcinoma among the PUVA-treated patients.

Plastic blankets and heat shields decrease transmission of phototherapy light.

An in vitro study showed that the use of transparent plastic insulation to decrease heat loss from newborn infants substantially decreases the transmission of phototherapy light. The decrease is proportional to the number of plastic layers between the light source and the baby because of reflection from the surface of the plastic. This increases if the surface is irregular. Thickness and the type of plastic used had no effect.

UVB doses in maintenance psoriasis phototherapy versus solar UVB exposure.

A possible increase in the risk of skin cancer in psoriatic patients treated with long-term maintenance UVB phototherapy was assessed by comparing the cumulative doses of UVB with the amount of UVB received from sunlight by normal healthy people. The biologically-effective UVB dose (termed UVB(EE) ) was measured using polysulphone film and worn as a badge by individuals with either an indoor or an outdoor occupation during 4 summer months of 1983 in The Netherlands (52 degrees N). The calculated mean annual UV-B(EE) doses were 5.9 J/cm2 for persons with an indoor occupation and 134 J/cm2 for those with an outdoor occupation. The UVB(EE) doses received by psoriasis patients during an initial course of phototherapy, as well as during maintenance treatment, were also estimated and gave a mean value of 22 J/cm2. Mean annual amounts of solar UVB(EE) exposure were calculated and compared with the administered doses of UVB(EE) during maintenance phototherapy. A dose-response model is described in order to estimate the increased incidence of non-melanoma skin cancer associated with such therapy. The cumulative incidence among patients who received maintenance phototherapy for several decades was calculated to be a factor of 2.5 to 7.5 higher than the incidence among individuals with an outdoor occupation.

Influence of ultraviolet radiation treatment on the survival of heterotopic skin grafts in the mouse.

Isolated mouse tail skin was UV-irradiated in vitro at a dose of 40 mJ/cm2 from both sides to remove the Ia immunogenicity. Immediately after irradiation the skin was transplanted onto the flank of allogeneic mice. When there was a total H-2 difference between donor and recipient, the UV-irradiated skin did not show a prolonged survival compared to control grafts. In the case of an I-region difference only, i.e., B10.AQR grafts onto B10.T (6R) recipients, a significant prolongation of the survival time was observed, whereas 50% of the UV-treated grafts were not rejected at all.