Human photoreactivating enzyme action spectrum and safelight conditions

Repair of cyclobutyl pyrimidine dimers in human skin: variability among normal humans in nucleotide excision and in photorepair

BACKGROUND/AIMS

Photoreactivation (PR) of cyclobutyl pyrimidine dimers (CPD) in human skin remains controversial. Recently Whitmore et al. (1) reported negative results of experiments using two photorepair light (PRL) sources on UV-irradiated skin of volunteers. However, their PRL sources induced substantial levels of dimers in skin, suggesting that the additional dimers formed could have obscured PR. We met a similar problem of dimer induction by a PRL source. We designed and validated a PRL source of sufficient intensity to catalyse PR, but that did not induce CPD, and used it to measure photorepair in human skin.

METHODS AND RESULTS

Using a solar simulator filtered with three types of UV-filters, we found significant dimer formation in skin, quantified by number average length analysis using electrophoretic gels of isolated skin DNA. To prevent scattered UV from reaching the skin, we interposed shields between the filters and skin, and showed that the UV-filtered/shielded solar simulator system did not induce damage in isolated DNA or in human skin. We exposed skin of seven healthy human volunteers to 302 nm radiation, then to the improved PRL source (control skin areas were kept in the dark for measurement of excision repair).

CONCLUSIONS

Using a high intensity PRL source that did not induce dimers in skin, we found that three of seven subjects carried out rapid photorepair of dimers; two carried out moderate or slow dimer photorepair, and three did not show detectable photorepair. Excision repair was similarly variable in these volunteers. Subjects with slower excision repair showed rapid photorepair, whereas those with rapid excision generally showed little or no photoreactivation.

An enzyme similar to animal type II photolyases mediates photoreactivation in Arabidopsis

The important issue of photoreactivation DNA repair in plants has become even more interesting in recent years because a family of genes that are highly homologous to photoreactivating DNA repair enzymes but that function as blue light photoreceptors has been isolated. Here, we report the isolation of a novel photolyase-like sequence from Arabidopsis designated PHR1 (for photoreactivating enzyme). It shares little sequence similarity with either type I photolyases or the cryptochrome family of blue light photoreceptors. Instead, the PHR1 gene encodes an amino acid sequence with significant homology to the recently characterized type II photolyases identified in a number of prokaryotic and animal systems. PHR1 is a single-copy gene and is not expressed in dark-grown etiolated seedlings: the message is light inducible, which is similar to the expression profile for photoreactivation activity in plants. The PHR1 protein complements a photolyase-deficient mutant of Escherichia coli and thus confers photoreactivation activity. In addition, an Arabidopsis mutant that is entirely lacking in photolyase activity has been found to contain a lesion within this Arabidopsis type II photolyase sequence. We conclude that PHR1 represents a genuine plant photolyase gene and that the plant genes with homology to type I photolyases (the cryptochrome family of blue light photoreceptors) do not contribute to photoreactivation repair, at least in the case of Arabidopsis.

Human white blood cells contain cyclobutyl pyrimidine dimer photolyase

Although enzymatic photoreactivation of cyclobutyl pyrimidine dimers in DNA is present in almost all organisms, its presence in placental mammals is controversial. We tested human white blood cells for photolyase by using three defined DNAs (supercoiled pET-2, nonsupercoiled bacteriophage lambda, and a defined-sequence 287-bp oligonucleotide), two dimer-specific endonucleases (T4 endonuclease V and UV endonuclease from Micrococcus luteus), and three assay methods. We show that human white blood cells contain photolyase that can photorepair pyrimidine dimers in defined supercoiled and linear DNAs and in a 287-bp oligonucleotide and that human photolyase is active on genomic DNA in intact human cells.

UVA does not photoreactivate pyrimidine dimers in cultured human fibroblasts

Pyrimidine dimers were induced in duplicates of cultured human skin fibroblasts by irradiation with various doses of UVB radiation. Subsequently, one set of cells was further exposed to either 5 or 10 J/cm2 of UVA radiation to assess the photoreactivating activity of this spectral range in a human cell system. Following irradiation, pyrimidine dimers were quantified in all cells by determining the number of endonuclease-sensitive sites (ESS). No difference in the yield of ESS was observed between cells which had been irradiated with UVB only as compared to cells which subsequently had been exposed to 5 or 10 J/cm2 UVA. In contrast, subsequent exposure of UVB-irradiated cells of Monodelphis domestica to 10 J/cm2 UVA resulted in an almost 50% reduction of UVB-induced pyrimidine dimers. These data indicate that UVA does not induce photoenzymatic repair in human fibroblasts.

Evidence for lack of DNA photoreactivating enzyme in humans

Photoreactivating enzyme (DNA photolyase; deoxyribocyclobutadipyrimidine pyrimidine-lyase, EC 4.1.99.3) repairs UV damage to DNA by utilizing the energy of near-UV/visible light to split pyrimidine dimers into monomers. The enzyme is widespread in nature but is absent in certain species in a seemingly unpredictable manner. Its presence in humans has been a source of considerable controversy. To help resolve the issue we used a very specific and sensitive assay to compare photoreactivation activity in human, rattlesnake, yeast, and Escherichia coli cells. Photolyase was easily detectable in E. coli, yeast, and rattlesnake cell-free extracts but none was detected in cell-free extracts from HeLa cells or human white blood cells with an assay capable of detecting 10 molecules per cell. We conclude that humans most likely do not have DNA photolyase.

Does exposure of human skin in situ to 385 or 405 nm UV induce pyrimidine dimers in DNA?

A previous report [Freeman et al. (1986) Photochem. Photobiol. 43S, 93S] indicated that irradiation of human skin in situ with 385 or 405 nm radiation produced detectable levels of pyrimidine dimers in DNA. Since these wavelengths are absorbed poorly by DNA, these results suggested that DNA damage was sensitized by other absorbing molecules present in skin. Examination of two experimental aspects of the previous work indicates that (1) the static gel electrophoresis method for DNA dispersion used in lesion determination gave accurate values of the levels of induced dimers, and (2) the DNA damage apparently induced by 385 nm was actually induced by shorter wavelength UV present in the 20 nm bandpass beam of the monochromator. The current results indicate that monochromatic 385 and 405 nm radiation are ineffective in dimer production in human skin in situ.

DNA photoreactivating enzyme from human tissues

Photoreactivating enzyme activity has been quantitated in human fetal skin, kidney, lung, liver, brain and intestine, and in neonatal human foreskin. In all the tissues examined there were at least two activities: one nominally greater than 10,000 Da, and one nominally less than 10,000 Da. Both can photolyze pyrimidine dimers in DNA using only light of wavelengths greater than 320 nm, thus excluding tryptophan-mediated dimer splitting as an important mechanism for these activities. The activities are inactivated by digestion with trypsin or pronase, and decreased partially or totally by heating to 65 degrees C. The activities from all six tissues, as well as that from neonatal foreskin, act catalytically in dimer photolysis. The properties of macromolecular size, heat lability, protease sensitivity and catalytic pyrimidine dimer photolysis by a non-tryptophan-mediated mechanism correspond to those of a true photoreactivating enzyme.

Variations in excision repair of UVB-induced pyrimidine dimers in DNA of human skin in situ

The excision repair kinetics of UVB (280-320 nm)*-induced pyrimidine dimers in DNA of human skin in situ was determined for seventeen volunteers using a dimer-specific endonuclease from Micrococcus luteus in conjunction with agarose gel electrophoresis. Removal of pyrimidine dimers from human skin could be detected within 6 h after irradiation and the average half-life for removal of pyrimidine dimers was 11.0 h (+/- 4.3 h). However, there was significant inter-individual variability of repair as indicated by a half-life coefficient of variation of 38%.

Side effects in excimer corneal surgery. DNA damage as a result of 193 nm excimer laser radiation

UV radiation is known to cause actinic damage to the DNA. Excimer laser light, possibly used for keratorefractive surgery, should not produce this damage, as the penetration depth is far less than the diameter of a cell. However, photoreactivation experiments with yeast cells show a significant amount of DNA repair after excimer irradiation. The zone of influence of a small slitlike exposure has a diameter of 2 cm. Consequently, the limbus, the critical location of epithelial neoplasia, always lies within the sphere of actinic damage. Radiation damage is induced by secondary radiation rather than by direct interaction.

UV-induced unscheduled DNA synthesis in human skin: dose response, correlation with erythema, time course and split dose exposure in vivo

Unscheduled DNA synthesis (UDS) has been shown to be saturated above a threshold dose of UV-C in human fibroblasts in vitro. We have investigated by autoradiography whether a similar saturation occurs in human skin in vivo with UV-B and whether this phenomenon correlates with the erythemal response. In addition, we determined the time course of UDS at 24 h after exposure and the effect of dual exposures separated by 24 h. The dose-response curve was established by exposure to 1/16, 1/8, 1/4, 1/2, 1, 2, 3, 4 and 6 MEDs UV-B. For the time-course study, areas exposed to 1/2 and 2 MEDs were biopsied after 1, 3, 6, 12 and 24 h. Autoradiography was performed in vitro. The dose-response curve showed a significant increase in UDS from 1/16 to 1 minimal erythema dose (MED), whereas no significant difference was observed between 1 MED and the higher UV-B doses tested. The 24 h time sequence revealed a gradual decrease in UDS activity. The 1/2 MED curve declined more rapidly and reached the zero-level between 12 h and 24 h, whereas about 50% of the initial UDS value was still retained 24 h after 2 MEDs. The dual-dose study revealed that a second hit of fractions of the MED resulted in lower levels of UDS than induced by these fractions alone in previously untreated areas. UDS increases with the erythemal dose between 1/16 and 1 MED. It reaches a plateau after 1 MED and cannot be increased by doses up to 6 MEDs, suggesting a saturation of excision repair in vivo. Time course studies support such a saturation phenomenon. The failure to increase significantly UDS by a second irradiation 24 h after the first exposure needs further clarification. Since persistence of DNA lesions may lead to an accumulation after repeated exposures, additional mechanisms other than excision repair may protect human skin by error-free removal of possibly mutagenic sites. Photoreactivation may be important in this respect.

Production of pyrimidine dimers in DNA of human skin exposed in situ to UVA radiation

Cyclobutyl pyrimidine dimers, measured as sites recognized by the dimer-specific ultraviolet (UV) endonuclease from Micrococcus luteus, were produced in DNA of human skin exposed in situ to UVA (320-400 nm) radiation. The dimer yields produced by a broadband UVA source, by broadband UVA filtered to remove all light of wavelength less than 340 nm, and by narrow band radiation centered at 365 nm were similar, indicating that UVA radiation, and not stray shorter wavelength radiation, was responsible for dimer production. The identity of the UVA-induced DNA lesions was confirmed as pyrimidine dimers by photoreactivation of approximately 100% of the endonuclease-sensitive sites in vitro with the 40,000 dalton Escherichia coli photoreactivating enzyme.

Photoreactivation of ultraviolet-irradiated, plasmid-bearing, and plasmid-free strains of Bacillus anthracis

The effects of toxin- and capsule-encoding plasmids on the kinetics of UV inactivation of various strains of Bacillus anthracis were investigated. Plasmids pXO1 and pXO2 had no effect on bacterial UV sensitivity or photoreactivation. Vegetative cells were capable of photoreactivation, but photo-induced repair of UV damage was absent in B. anthracis Sterne spores.

Higher pyrimidine dimer yields in skin of normal humans with higher UVB sensitivity

We have measured UVB (280-320 nm)-induced DNA damage in skin of individuals with different sensitivities to UVB irradiation as measured by minimal erythema dose (MED). The DNA damage was susceptible to cleavage by Micrococcus luteus UV endonuclease, which recognizes pyrimidine dimers in DNA. An alkaline agarose gel electrophoresis method was used to quantitate the number of M. luteus UV endonuclease-sensitive sites in nonradioactive DNA from skin biopsies of 7 individuals irradiated with UVB (0-180 mJ X cm-2). The production of sites correlated well with MED (correlation coefficient = 0.78). The slope of the dose response curve for the most UVB-sensitive individual (MED = 24 mJ X cm-2) and for the least UVB-sensitive individual (MED = 146 mJ X cm-2) were 11.5 X 10(-4) and 2.6 X 10(-4) sites per 1000 bases per mJ X cm-2, respectively. The UVB-induced DNA damage was determined to be pyrimidine dimers by its susceptibility to cleavage by M. luteus UV endonuclease and its photoreactivability by Escherichia coli photoreactivating enzyme.

Comparative protection efficiency of UVA- and UVB-induced tans against erythema and formation of endonuclease-sensitive sites in DNA by UVB in human skin

UVA- and UVB-induced tans which were visually identical with each other were induced in separate sites on the lower back of 5 normal human volunteers of good tanning ability. Tanning was achieved by 4 exposures to UVA and UVB administered over an 8-day period. One week after the last exposure the protection afforded by the two types of tan against UVB-induced erythema and against UVB-induced DNA damage was measured. Protection against erythema was measured by comparison of the minimal erythema doses of UVB in tanned and untanned skin. Protection against DNA damage was assessed by comparing the numbers of endonuclease-sensitive sites in epidermal DNA extracted from biopsies taken from tanned and untanned sites exposed to the same dose of UVB. The UVB tans conferred significant protection (mean 2.98-fold) against UVB-induced erythema. UVA tans were not associated with significant protection (mean 1.4-fold). In contrast, both UVA- and UVB-induced tans were associated with a similar reduction in yield of endonuclease-sensitive sites in epidermal DNA (in UVA tan to 47% and in UVB tan to 45% of the yield in untanned skin). Protection conferred by the tans against erythema was therefore not paralleled by protection against DNA damage.

Photoreactivation in bacteria and in skin

In many procaryotic and eucaryotic cells, photoreactivating enzyme mediates light-dependent repair of UV-induced damage: the enzyme binds to a pyrimidine dimer in DNA, and, on absorption of a photon (300-600 nm), specifically monomerizes the dimer, thus repairing the DNA. Photoreactivating enzyme has been found in human tissues and human cells in culture; human cells in culture can photoreactivate cellular dimers, and can mediate photoreactivation of Herpes (human fibroblasts) and Epstein-Barr virus (human leukocytes). Measurements of pyrimidine dimer formation and repair in human skin indicate that detectable numbers of dimers are formed at 1 minimal erythemal dose, that the dimers are rapidly removed in skin kept in the absence of light, and they are more rapidly removed when the skin is exposed to visible light.

Damage and repair in mammalian cells after exposure to non-ionizing radiations. II. Photoreactivation and killing of rat kangaroo cells (Potorous tridactylus) and Herpes simplex virus-1 by exposure to fluorescent "white" light or sunlight

Photoreactivation (PR) of ultraviolet (254 nm)-inactivated cornea cells of the potoroo (or rat kangaroo; Potorous tridacylus) has been studied at wavelengths greater than 375 nm from either fluorescent "white" light or sunlight. In both cases the PR kinetics curves pass through maxima, which most likely result from the superposition of concomitant inactivation by the photoreactivating light. The inactivating effect of light was directly demonstrated for non-UV-irradiated cells, permitting correction of the PR curves. Wavelengths greater than 475 nm, and even greater than 560 nm, which do not noticeably damage cells, still photoreactivate, though less effectively than shorter wavelengths. Light treatment of UV-inactivated Herpes simplex Virus-1 (HSV-1) after infection leads to PR effects resembling those observed for cells, while light treatment of unirradiated virus after infection likewise causes inactivation. The "fluence-reduction factor" of PR, which is greater than 3 for the virus, exceeds that for the cells, where it decreases with increasing UV fluence. In vitro tests have indicated that sunlight greater than 375 nm causes photorepairable DNA lesions which are virtually fully repaired by the same light. Thus cell inactivation resulting from these solar wavelengths must be due to non-photorepairable damage.

Deoxyribonucleic acid repair capacities of Neisseria gonorrhoeae: absence of photoreactivation

No difference in survival was observed when ultraviolet-irradiated gonococcal cells were subsequently incubated in the dark or exposed to photoreactivating light. This observation indicates that photoreactivation is absent in Neisseria gonorrhoeae.

Culture conditions affect photoreactivating enzyme levels in human fibroblasts

Photoreactivation of pyrimidine dimers in mammalian cells occurs under our experimental conditions but has not been observed under conditions used by others. We have tested three possible differences in experimental procedures including dimer separation and analysis methods, illumination conditions and cell culture techniques. We show that out methods of dimer separation and analysis indeed measure cis-syn pyrimidine dimers and give results in quantitative agreement with the methods of others. We find that while light pre-illumination of fibroblasts from the xeroderma pigmentosum line XP12BE or of normal cells does not affect the cellular capacity for dimer photoreactivation. However, we show that cell culture conditions can affect photoreactivating enzyme levels and thus, cellular dimer photoreactivation capacity. Cells grown in Eagle's minimal essential medium (supplemented with 15% fetal bovine serum) contain very low levels of photoreactivating enzyme and cannot photoreactivate dimers in their DNA; however, companion cultures maintained in Dulbecco's modified Eagle's minimal medium do contain photoreactivating enzyme and can photoreactivate cellular dimers.