Blue light induces mitochondrial DNA damage and free radical production in epithelial cells

Exposure of biological chromophores to ultraviolet radiation can lead to photochemical damage. However, the role of visible light, particularly in the blue region of the spectrum, has been largely ignored. To test the hypothesis that blue light is toxic to non-pigmented epithelial cells, confluent cultures of human primary retinal epithelial cells were exposed to visible light (390-550 nm at 2.8 milliwatts/cm2) for up to 6 h. A small loss of mitochondrial respiratory activity was observed at 6 h compared with dark-maintained cells, and this loss became greater with increasing time. To investigate the mechanism of cell loss, the damage to mitochondrial and nuclear genes was assessed using the quantitative PCR. Light exposure significantly damaged mitochondrial DNA at 3 h (0.7 lesion/10 kb DNA) compared with dark-maintained controls. However, by 6 h of light exposure, the number of lesions was decreased in the surviving cells, indicating DNA repair. Isolated mitochondria exposed to light generated singlet oxygen, superoxide anion, and the hydroxyl radical. Antioxidants confirmed the superoxide anion to be the primary species responsible for the mitochondrial DNA lesions. The effect of lipofuscin, a photoinducible intracellular generator of reactive oxygen intermediates, was investigated for comparison. Exposure of lipofuscin-containing cells to visible light caused an increase in both mitochondrial and nuclear DNA lesions compared with non-pigmented cells. We conclude that visible light can cause cell dysfunction through the action of reactive oxygen species on DNA and that this may contribute to cellular aging, age-related pathologies, and tumorigenesis.     

Photoreactivation and repair replication in rat kangaroo cells

Potorous tridactylis cells can perform photoreactivation, i.e., the visible light- catalyzed reversal of UV-induced pyrimidine dimers in DNA. UV-induced inhibitions of total RNA and DNA synthesis can also be partially reversed by exposure to visible light. P. tridactylis cells can also perform repair replication, but the extent of the latter is reduced if the cells are exposed to visible light (VL). None of these effects are observed in mouse L cells, which cannot perform photoreactivation. The results are consistent with the concept of pyrimidine dimers are one of the main substrates for repair replication.     

Expression of a mammalian DNA photolyase confers light-dependent repair activity and reduces mutations of UV-irradiated shuttle vectors in xeroderma pigmentosum cells

Photoreactivation is one of the DNA repair mechanisms to remove UV lesions from cellular DNA with a function of the DNA photolyase and visible light. Two types of photolyase specific for cyclobutane pyrimidine dimers (CPD) and for pyrimidine (6-4) pyrimidones (6-4PD) are found in nature, but neither is present in cells from placental mammals. To investigate the effect of the CPD-specific photolyase on killing and mutations induced by UV, we expressed a marsupial DNA photolyase in DNA repair-deficient group A xeroderma pigmentosum (XP-A) cells. Expression of the photolyase and visible light irradiation removed CPD from cellular DNA and elevated survival of the UV-irradiated XP-A cells, and also reduced mutation frequencies of UV-irradiated shuttle vector plasmids replicating in XP-A cells. The survival of UV-irradiated cells and mutation frequencies of UV-irradiated plasmids were not completely restored to the unirradiated levels by the removal of CPD. These results suggest that both CPD and other UV damage, probably 6-4PD, can lead to cell killing and mutations.     

Visible light-inducible photolyase gene from the goldfish Carassius auratus.

By introducing and expressing a cDNA library constructed from mRNA of the cultured goldfish Carassius auratus cells in Escherichia coli, a gene encoding photolyase of the vertebrate was isolated, the first example from metazoa. The amino acid sequence deduced from the nucleotide sequence differs significantly from those of microorganisms. Five out of 6 tryptophan residues strictly conserved in photolyases from microorganisms and thought to play important roles in DNA and chromophore binding of the enzyme are substituted by other residues of different characteristics. By Northern analysis the expression of the photolyase gene was found to be induced more than 10 times by exposure of the cells to visible light. These results indicate a unique evolution of the photolyase gene and a novel mechanism of gene regulation, in which visible light triggers the production of the light-dependent enzyme for repair of DNA damages induced by harmful ultraviolet part of sunlight.     

Fate during cell growth of yeast mitochondrial and nuclear DNA after photolytic attachment of the monoazide analog of ethidium.

The ¹⁴C-labeled photosensitive monoazide analog of ethidium, 3-amino-8-azido-5-ethyl-6-phenylphenanthridinium chloride, produced covalent adducts in yeast cells with both nuclear and mitochondrial DNA on photolysis by visible light. With subsequent cultivation in nutrient medium, drug molecules on mitochondrial DNA were removed only through extensive mitochondrial DNA degradation. In contrast, drug attached to nuclear DNA was eliminated with conservation of DNA, presumably through a repair process.     

Excision repair does not protect Bacillus subtilis cells from inactivation by sunlight]

Exponentially growing Bacillus subtilis cells are highly sensitive to inactivating action of sunlight, the strain deficient in excision repair being every more sensitive than the uvr1 mutant. The inactivating effect is connected with the action of irradiation located in the left part of the spectrum (the whole UV region and some zones of the visible one). Increased sensitivity to sunlight disappeared when cells were exposed to sunlight in a liquid medium with casaminoacids (2 g/l). The inactivating effect was probably of photodynamic nature and was caused either by DNA lesions that were not removed by excision repair or by the damage which arose not in DNA.     

Photodynamic damage of yeast subcellular fraction induced by elevated levels of endogenous protoporphyrin IX

The 2,2'-dipyridyl-induced accumulation of protoporphyrin IX in Saccharomyces cerevisiae cells was shown to be accompanied by the photoinhibition of cell respiration and the enhancement of the photoinduced permeability of plasma membranes to the fluorescent dye primuline. The visible-light illumination (at 400-600 nm) of the mitochondria and plasma membranes isolated from yeast cells with a high level of endogenous protoporphyrin IX intensified lipid peroxidation in these subcellular organelles. Comparative studies showed that the rad 52 mutant cells, which are deficient in the postreplicative recombinational DNA repair system, are considerably more sensitive to the inactivating action of visible light than are the wild-type cells and the rad 3 mutant cells, which are deficient in the excision DNA repair system. The contribution of photodynamic damage to the yeast subcellular organelles to the lethal photodynamic effect is discussed.     

Excision repair of uracil in higher plant cells: Uracil-DNA glycosylase and sister-chromatid exchanges

Uracil is not a normal constituent of DNA. Under natural conditions, it may appear either by deamination of cytosine residues or by incorporation of deoxyuridine monophosphate (dUMP). Visible light irradiation of BrdUrd-treated cells efficiently leads, under experimental situations, to the formation of dUMP residues in DNA. Plant cells, like other living organisms, can eliminate this potentially harmful base from DNA by an excision repair pathway, uracil-DNA glycosylase being the first enzyme acting during the incision process. Purified plant uracil-DNA glycosylase is a low molecular weight enzyme (27–29.5 kD) that specifically releases uracil present in DNA by splitting off the sugar-base bond. This enzyme is non-competitively inhibited by uracil and 6-aminouracil, but not by thymine, both in vitro and in vivo. However, other structurally related compounds do not show any inhibitory effect. This characteristic poses a number of unaswered questions regarding its mechanism of action. At the chromosome level, dUMP residues appear to be sister-chromatid exchange (SCE)-initiating events. This has been demonstrated for dUMP residue introduced either by visible light exposure of BrdUrd-treated cells or by dUMP mis-incorporation instead of dTMP in cells treated with inhibitors of thymidylate synthetase. The excision repair of uracil in plants appears to be finely regulated in different cell types depending on their proliferation rate and their development stage. Thus, high levels of uracil-DNA glycosylase do not seem to be necessarily associated with DNA replication, since non-proliferating cells, natural constituents of dormant meristems, contain enzyme levels comparable to those found in proliferating tissues, where it is modulated: the higher the cell cycle rate (and the DNA replication rate) the higher the uracil-DNA glycosylase activity. Finally, this excision repair enzyme seems to be turned off as cells enter their differentiated state.     

Postreplication repair of DNA in chick cells: studies using photoreactivation.

During replication of DNA after ultraviolet irradiation, gaps are left in the newly-synthesized DNA strands in both bacterial and animal cells and these gaps are subsequently sealed by a process known as postreplication repair. In order to test whether it is the ultraviolet-induced pyrimidine dimers which are responsible for the production of these daughter-strand gaps in animal cells, we have used chick embryo fibroblasts. In these cells the pyrimidine dimers are photoreactivable, i.e. they can be split by an enzymatic process dependent on visible or near ultraviolet light. Our results indicate that chick cells possess a postreplication repair system similar to that in mammalian cells; gaps are produced in the newly-synthesized strands and then filled in. If the ultraviolet-irradiated cells are first photoreactivated to remove most of the dimers, the number of daughter-strand gaps produced is much less than without photoreactivation. This suggests that the dimers are indeed responsible for the formation of many of the gaps in the newly-synthesized DNA. Ultraviolet light also inhibits the overall rate of DNA synthesis. This inhibition is, however, only partly overcome by photoreactivation.     

Photodynamic Damage to Yeast Subcellular Organelles Induced by Elevated Levels of Endogenous Protoporphyrin IX

The 2,2"-dipyridyl-induced accumulation of protoporphyrin IX in Saccharomyces cerevisiae cells was shown to be accompanied by the photoinhibition of cell respiration and the enhancement of the photoinduced permeability of plasma membranes to the fluorescent dye primuline. The visible-light illumination (at 400–600 nm) of the mitochondria and plasma membranes isolated from yeast cells with a high level of endogenous protoporphyrin IX intensified lipid peroxidation in these subcellular organelles. Comparative studies showed that the rad 52 mutant cells, which are deficient in the postreplicative recombinational DNA repair system, are considerably more sensitive to the inactivating action of visible light than are the wild-type cells and the rad 3 mutant cells, which are deficient in the excision DNA repair system. The contribution of photodynamic damage to the yeast subcellular organelles to the lethal photodynamic effect is discussed.     

Factors affecting the photokilling of cultured Chinese hamster cells by phthalocyanines

Chloroaluminum phthalocyanine (CAPC) was recently shown to sensitize the inactivation of cultured Chinese hamster cells by visible light. Several factors affecting the photodynamic action of CAPC have been defined in the present study. Thus the photosensitized inactivation of Chinese hamster cells is not affected by superoxide dismutase, suggesting that O-2 radicals are not involved in the process. Postillumination treatments with D2O or heat (42 degrees C, 90 min) enhanced CAPC-induced photosensitivity, indicating the existence of a repair mechanism for photodamage. Preillumination treatments with sodium salicylate and 5-bromodeoxyuridine also enhanced photosensitivity. The later observation suggests that CAPC-induced DNA damage is potentially lethal. However, 3-aminobenzamide, a potent inhibitor of poly(ADP-ribose) synthesis which is involved in repair of DNA strand breakage, had no effect on the photosensitivity. Photosensitized inactivation by CAPC is dependent on the pH value of the medium during irradiation. Thus, in the range of pH values 6-8, the sensitivity was increased at the lower values.     

Aphidicolin inhibits repair of DNA in UV-irradiated human fibroblasts

Aphidicolin, a specific inhibitor of DNA polymerase α, is shown to inhibit DNA repair in human diploid fibroblasts. Although aphidicolin has no apparent effect on the DNA of unirradiated cells, it causes a large number of strand breaks to accumulate in UV-irradiated cellular DNA. The number of breaks is the same as the number observed following a similar dose of ultraviolet light when cells are treated with arabinofuranosyl cytosine (araC) and hydroxyurea (HU), known inhibitors of repair. Moreover, two-dimensional paper chromatography shows that aphidicolin completely blocks removal of pyrimidine dimers. These observations are discussed in light of the proposed roles of DNA polymerases α β in DNA replication and repair and the action of aphidicolin on polymerase α.     

Deficiency of DNA ligase activity in Fanconi's anemia

Summary A significant decrease in DNA ligase activity was observed in lymphocytes and fibroblasts of a patient with Fanconi's anemia (FA). This decrease is related to the observed DNA repair deficiency indicated by the delayed closing of repair DNA strands following UV irradiation. Other steps of DNA repair were analyzed in the FA fibroblasts, including endonucleolytic incision of DNA, repair DNA synthesis, and exonucleolytic removal of the photoproducts. No differences were found against control cells. The action of DNA ligase is delayed during replication in the FA cells, as seen by an accumulation of replicative intermediates.     

Photoactivation of DNA thiobases as a potential novel therapeutic option

The thiopurines, 6-thioguanine and 6-mercaptopurine, are antileukemic agents that are incorporated into DNA following retrieval by the purine salvage pathway (see [1] for a review). Their toxicity requires active DNA mismatch repair (MMR), and thiopurine resistance is an acknowledged phenotype of MMR-defective cells [2, 3]. In addition to these direct cytotoxic effects, DNA thiobases have distinctive photochemical properties [4], the therapeutic potential of which has not been extensively evaluated. We report here that the thiopyrimidine nucleoside 4-thiothymidine is incorporated into DNA. It does not induce MMR-related toxicity, but it interacts synergistically with UVA light and dramatically sensitizes cultured human cells to very low, nonlethal UVA doses. 4-thiothymidine induced UVA dose enhancements of around 100-fold in DNA repair-proficient cells. Nucleotide excision repair-defective xeroderma pigmentosum cells were sensitized up to 1000-fold, implicating bulky DNA photoproducts in the lethal effect. The synergistic action of thiothymidine plus UVA required thymidine kinase, indicating a selective toxicity toward rapidly proliferating cells. Cooperative UVA cytotoxicity is a general property of DNA thiobases, and 6-thioguanine and 4-thiodeoxyuridine were also UVA sensitizers. Thiobase/UVA treatment may offer a novel therapeutic approach for the clinical management of nonmalignant conditions like psoriasis or for superficial tumors that are accessible to phototherapy.     

UV damage to plant life in a photobiologically dynamic environment: the case of marine phytoplankton

The effect of UV-B radiation on growth of marine phytoplankton was investigated in relation to DNA damage induced by a range of biologically effective doses (BEDs). Emiliania huxleyi (Prymnesiophyceae) was chosen as a model organism of the ocean's phytoplankton because of its importance in global biogeochemical cycling of carbon and sulphur, elements that influence the world's climate as components of the trace gases carbon dioxide (CO₂) and dimethylsulfide (DMS). A marine diatom, Cyclotella, was studied for its capacity to repair the DNA damage, quantified as thymine dimers by the application of a monoclonal antibody against these photoproducts. DNA repair was shown to be complete after just a few hours of exposure to visible light; the repair rate increased with PAR intensity. E. huxleyi appeared to be most sensitive to UV-B radiation: growth was already affected above a dose of 100 J m^-2 d^-1 (biologically effective radiation, weighted with Setlow's DNA action spectrum), probably through effects on the cell cycle related to damage to nuclear DNA: mean specific growth rates were inversely correlated with thymine dimer contents in cells. Near the ocean's surface UV-B radiation conditions that induce the changes observed by us in cultures can be expected during the growing season of phytoplankton, not only in the tropics but also at higher latitudes. Nevertheles, blooms of species such as E. huxleyi are often excessive in the field. It is suggested that exposure duration of cells near the surface of the ocean can be shorter than our artificial 3 h in the laboratory due to vertical mixing, a phenomenon that is typical for the ocean's upper 50–100 m. When mixing reaches depths greater than the layer where most UV-B is attenuated, negative effects on cells through UV-A-induced inhibition of photosynthesis may prevail over DNA damage, the action spectrum of which has been shown to be limited to the UV-B part of the spectrum. Moreover, the radiation wavelengths that induce DNA damage repair (UV-A and visible) are attenuated vertically much less than UV-B. The photobiological situation in the upper ocean is much more complicated than on land, and effects of UV radiation on plankton biota can only be modelled realistically here when both the spectrally differential attenuation in the UV and visual part of the spectrum and the rate of vertical mixing are taken into account. Action spectra of both damage and repair of DNA and of photosynthesis inhibition of representative microalgal species are the second conditio sine qua non if we want to predict the effect of stratospheric ozone depletion on marine phytoplankton performance.     

Studies on the reactivation of bacteria photodamaged by psoralen

Summary In photoreaction with the pyrimidine bases (thymine, cytosine, uracil) as well as with nucleic acids (DNA, RNA) a C₄-cycloaddition of furocoumarins to the 5.6-double bond of pyrimidine bases takes place. The simple photoadduct furocoumarin-pyrimidine base can be split by reirradiation at wavelengths shorter than 334 nm. Reactivation of bacterial cells photodamaged by psoralen (365 nm) was tried experimentally. However, reirradiation at shorter wavelengths and with visible light of the psoralen-inactivated bacterial cells was without any effect. The inability of the shorter wavelengths to repair this photodamage was probably due to a filter effect of DNA for such wavelengths, as shown by experiments on a DNA-psoralen combination. On the other hand the observed ability of psoralen to form inter-strand cross-linkages in the photoreaction with DNA may be significant for explaining the absence of photoreactivation when the inactivated bacterial cells are irradiated with visible light.     

Combination of the mutation process with the sensitization and repair processes leading to increased frequencies of mutations in algal populations

The possibility of using a combination of the mutation process with the induction of the repair processes has been studied to increase the mutation frequencies in algal populations after UV-treatment. From this study it follows that the repair process induced by visible light is much more effective than the dark repair processes in the chlorococcal algae used. In these algae, visible light perhaps does not induce only those repair processes which affect their DNA, but probably also some recovery ones which affect their damaged structures and physiological functions. A suitable combination of the sensitization of algal cells by a DNA-base analogue before UV-treatment and the induction of the light repair and recovery processes resulted in a rather high increase of viable mutations in chlorococcal algae. These findings may be useful in the breeding of chlorococcal algae, which have no possibility of hybridization (except somatic).     

The active form of Escherichia coli DNA photolyase contains a fully reduced flavin and not a flavin radical, both in vivo and in vitro

Escherichia coli DNA photolyase is a flavoprotein that when purified is blue in color and contains a stable neutral radical FAD (E-FADH). In the presence of a suitable electron donor (i.e., thiols, tyrosine, or NADH) the radical FAD adsorbs visible light and undergoes photoreduction to the fully reduced FAD (E-FADH2). The in vitro quantum yield of dimer repair for E-FADH is 0.07 while that of E-FADH2 approaches the in vivo value of 1. Electron paramagnetic resonance studies on whole cells indicate that the in vivo form of photolyase is E-FADH2 with enzyme containing radical FAD generated predominantly during the ammonium sulfate precipitation step of the purification. Activity measurements of E-FADH using long-wavelength photoreactivating light indicate that enzyme containing FAD in the radical form is not active in dimer repair. Dimer repair observed with E-FADH at shorter wavelengths is probably photoreduction of E-FADH followed by dimer repair by E-FADH2.     

Inactivation of Escherichia coli, F′ Episomes at Transfer, and Bacteriophage Lambda by Psoralen Plus 360-nm Light: Significance of Deoxyribonucleic Acid Cross-Links

We have investigated some biological consequences of light-induced psoralen-deoxyribonucleic acid (DNA) adducts and find that for several Escherichia coli functions (killing of strain AB2480 recA13 uvrA6, inactivation of phage lambda plaque-forming ability in wild type and uvrA6 hosts, loss of ability to transmit intact Flac(+) episomes), a light exposure sufficient for production of a single cross-link per DNA molecule correlates well with the biological consequence. Although one cross-link per genome is apparently lethal to recA13 uvr(-) strains, mutants carrying the recA13 or uvrA6 markers survive light exposures producing 6.7 and 16 cross-links per genome, respectively, and wild-type cells recover from 65 psoralen cross-links. Evidently, the excision and recombinational repair systems complement one another in reconstructing an intact genome from cellular DNA containing psoralen photoproducts. The above bacterial and phage strains, in which DNA repair processes are minimized, are also extremely sensitive to pyrimidine dimer-forming 254-nm UV light (without psoralen), and were expected to respond similarly to formation of psoralen-pyrimidine base monoadducts in their DNA. Since the biological inactivation by psoralen correlates well with cross-link formation, we suggest that the sensitizing action of this drug primarily derives from its ability to form DNA cross-links.     

Repair of DNA damage in shuttle vectors, virus, and chromosomal DNAs may depend on their biological imprinting--a 'Pygmalion' effect

We have created a cell line that can repair damage in chromosomal DNA and in herpes virus, while not repairing the same damage in shuttle vectors (pZ189 and pRSVcat). This cell line, a xeroderma pigmentosum (XP) revertant, repairs the minor (6-4)-photoproducts, but not cyclobutane dimers, in chromosomal DNA. The phenotype of this revertant after irradiation with ultraviolet (UV) light is the same as that of normal cells for survival, repair replication, recovery of rates of DNA and RNA synthesis, and sister-chromatid exchange formation, which indicates that a failure to mend cyclobutane dimers may be irrelevant to the fate of irradiated human cells. The two shuttle vectors were grown in Escherichia coli and assayed during transient passage in human cells, whereas the herpes virus was grown and assayed exclusively in mammalian cells. The ability of the XP revertant to distinguish between the shuttle vector and herpes virus DNA molecules according to their 'cultural background', i.e., bacterial or mammalian, may indicate that one component of the repair of UV damage involves gene products that recognize DNA markers that are uniquely mammalian, such as DNA methylation patterns. This component of excision repair may be involved in the original defect and the reversion of XP group A cells.