Yoshisada Fujiwara

Cross-link repair in human cells and its possible defect in Fanconi's anemia cells☆

Abstract We investigated the differential repair of DNA lesions induced by bifunctional mitomycin C, monofunctional decarbamoyl mitomycin C and ultraviolet irradiation in normal human, Xeroderma pigmentosum and Fanconis anemia cells using assays for the survival of clone-forming ability, alkaline sucrose sedimentation and hydroxyapatite chromatography of DNA. Four FA cell lines exhibited about 5 to 15 times higher sensitivity to MC killing, despite normal resistance to u.v. and DMC, than did normal human cells. The XP cells, however, were highly sensitive to u.v. and DMC killings due to their deficiency in excision repair, but the cells unexpectedly had an almost normal capacity for surviving MC and repairing the MC interstrand cross-links. In experiments to determine the sedimentation velocity of the DNA in alkaline sucrose gradients, normal and XP cells showed evidence for single-strand cutting following MC treatment. The sedimentation velocity of the DNA covalently cross-linked by MC in an FA strain was 2.5 times faster than that of the untreated control, and remained unaltered during post-incubation due to the lack of half-excision § of cross-links. However, FA cells, but not XP cells, had the normal ability to incise DNA with the DMC monoadducts. Hydroxyapatite chromatography revealed the reversibly bihelical property of MC cross-linked DNA after denaturation. Normal and XP cells lost such reversibility during post-MC incubation as the result of cross-link removal with first-order kinetics (half-life = 2 h). The three FA lines studied exhibited two- to eightfold reduced rates of cross-link removal than normal and XP cells, indicating a difference in the repair deficiency of the FA strain. Thus we have been led to conclude that FA cells may have different levels of deficiency in half-excision repair of interstrand cross-links induced by MC, despite having normal mechanisms for repair of u.v.-induced pyrimidine dimers and DMC monoadducts, and vice versa in XP cells.

Replicative bypass repair of ultraviolet damage to DNA of mammalian cells: Caffeine sensitive and caffeine resistant mechanisms

Replicative bypass repair of UV damage to DNA was studied in a wide variety of human, mouse and hamster cells in culture. Survival curve analysis revealed that in established cell lines (mouse L, Chinese hamster V79, HeLa S3 and SV40-transformed xeroderma pigmentosum (XP)), post-UV caffeine treatment potentiated cell killing by reducing the extrapolation number and mean lethal UV fluence (Do). In the Do reduction as the result of random inactivation by caffeine of sensitive repair there were marked clonal differences among such cell lines, V79 being most sensitive to caffeine potentiation. However, other diploid cell lines (normal human, excision-defective XP and Syrian hamster) exhibited no obvious reduction in Do by caffeine. In parallel, alkaline sucrose sedimentation results showed that the conversion of initially smaller segments of DNA synthesized after irradiation with 10 J/m2 to high-molecular-weight DNA was inhibited by caffeine in transformed XP cells, but not in the diploid human cell lines. Exceptionally, diploid XP variants had a retarded ability of bypass repair which was drastically prevented by caffeine, so that caffeine enhanced the lethal effect of UV. Neutral CsC1 study on the bypass repair mechanism by use of bromodeoxyuridine for DNA synthesis on damaged template suggests that the pyrimidine dimerer acts as a block to replication and subsequently it is circumvented presumably by a new process involving replicative bypassing following strand displacement, rather than by gap-filling de novo. This mechanism worked similarly in normal and XP cells, whether or not caffeine was present, indicating that excision of dimer is not always necessary. However, replicative bypassing become defective in XP variant and transformed XP cells when caffeine was present. It appears, therefore, that the replicative bypass repair process is either caffeine resistant or sensitive, depending on the cell type used, but not necessarily on the excision repair capability.

Inhibition of human excision DNA repair by inorganic arsenic and the co-mutagenic effect in V79 Chinese hamster cells

We investigated the lethal, UV killing-potentiating and repair-inhibiting effects of trivalent arsenic trioxide (As2O3) and pentavalent sodium arsenate (Na2HAsO4) in normal human and xeroderma pigmentosum (XP) fibroblasts. The presence of As2O3 for 24 h after UV irradiation inhibited the thymine dimer excision from the DNA of normal and XP variant cells and thus the subsequent unscheduled DNA synthesis (UDS): excision inhibitions were partial, 30-40%, at a physiological dose of 1 microgram/ml and 100% at a supralethal dose of 5 micrograms/ml. Correspondingly, As2O3 also potentiated the lethal effect of UV on excision-proficient normal and XP variant cells in a concentration-dependent manner, but not on excision-defective XP group A cells. Na2HAsO4 (As5+) was approximately an order of magnitude less effective in preventing all the above repair events than As2O3 (As3+) which is highly affinic to SH-containing proteins. The above results provide the first evidence that arsenic inhibits the excision of pyrimidine dimers. Partially repair-suppressing small doses of As2O3 (0.5 microgram/ml) and Na2HAsO4 (5 micrograms/ml) enhanced co-mutagenically the UV induction of 6-thioguanine-resistant mutations of V79 Chinese hamster cells. Thus, such a repair inhibition may be one of the basic mechanisms for the co-mutagenicity and presumably co-carcinogenicity of arsenic. XP group A and variant strains showed a unique higher sensitivity to As2O3 and Na2HAsO4 killing by a yet unidentified mechanism.

Caffeine-sensitive repair of ultraviolet light-damaged DNA of mouse L cells.

Abstract Mouse L5 cells, although as resistant to UV as excision-capable HeLa S3 cells, do not excise thymine dimers yielded at a rate of 0.05% thymine per 100 ergs/mm 2 . The daughter DNA newly replicated for 2 hours in 200 ergs/mm 2 -irradiated L5 cells is discontinuous with gaps spaced at about 5 × 10 6 daltons and seemingly complementary to photoproduct on the parental strand. The gapped DNA is subsequently sealed by the postreplication repair. Caffeine (2mM) inhibits such a DNA repair in irradiated L5 cells, but not in HeLa S3 cells. This result is correlated with caffeine-effected depression of recovery in the colony-forming ability of L5 cells after irradiation.

Defective repair of mitomycin C crosslinks in Fanconi's anemia and loss in confluent normal human and xeroderma pigmentosum cells.

Crosslink repair of mitomycin C-induced interstrand crosslinks was studied in exponentially growing and confluent normal human, transformed W138CT-1, Fanconis anemia (FA) and xeroderma pigmentosum (XP) group-A fibroblasts by the assay methods of alkaline sucrose centrifugation, hydroxyapatite column chromatography and S1-nuclease digestion. These three methods demonstrated unequivocally that crosslinking occurred at a rate of 0.13 crosslinks/10(8) Da per microgram per ml mitomycin C (less than or equal to 10 micrograms/ml) and the first half-excision of crosslinks followed the rapid first-order kinetics of 2-3 h half-life in exponentially-growing normal, WI38CT-1 and XP group-A cells. However, the first half-excision was completely defective in three out of the four FA strains tested and severely retarded in an FA strain. These results strongly support our previous observations in different strains of normal human, FA and XP group-A cells. An important new addition is that confluent, otherwise proficient, normal and XP cells almost completely lost the ability of the first, rapid half-excision of mitomycin C crosslinks in their DNA. This probably suggests that the enzyme or regulatory factor responsible for the half-excision, which differs from that for nucleotide excision repair, present constitutively in confluent cells, may be induced or activated only in the cycling cells. However, its relation to a defective FA factor is not clear at present.

Mitochondrial and intracellular free-calcium regulation of radiation-induced apoptosis in human leukemic cells.

PURPOSE To investigate the mechanisms and pathways of X-ray apoptosis in Molt-4 cells, focusing on mitochondrial and cytosolic Ca2+ ([Ca2+]i) regulation. MATERIALS AND METHODS X-irradiated Molt-4 cells and cell extract (CE) were used to analyse: (1) induced apoptosis (Giemsa stain), (2) p53, Bcl-2 and Bax expressions (immunoblot), (3) mitochondrial potential deltapsi(m) and (4) [Ca2+]i (flow cytometry), (5) caspase-3 activity, and (6) roles of [Ca2+]- and caspase-3-mediated pathways by inhibiting either or both pathways for induced apoptosis. RESULTS Molt-4 cells were sensitive to apoptosis since 5 Gy induced 57 and 94% apoptosis at 6 and 24 h. After 5Gy, p53 was accumulated that upregulated Bax but which repressed Bcl-2 with time, resulting in a 7-fold increase in Bax/Bxl-2 at 6 h. Predominant Bax reduced deltapsi(m), and low-deltapsi(m) cells increased 45 min earlier than apoptosis after 5 Gy. Caspase-3 was activated in apoptotic CE. The caspase-3 inhibitor Ac-DEVD-CHO inhibited apoptosis and DNA-ladder formation by approximately 50%, suggesting a approximately 50% role of caspase-3-activated DNase (CAD). [Ca2+]i was increased after 5 Gy. [Ca2+]i-chelating BAPTA-AM (5 microM) and/or DNase gamma-inhibiting Zn2+ (0.5 mM) inhibited approximately 50% of induced apoptosis and DNA-laddering, indicating a 50% participation of Ca2+/Mg2+-dependent DNase gamma. CONCLUSIONS The p53-Bax-mitochondria-caspase-3-CAD pathway and the [Ca+2]i-mediated DNase gamma pathway were involved in the regulation of X-ray apoptosis in sensitive Molt-4 cells.

UV damage-specific DNA-binding protein in xeroderma pigmentosum complementation group E

The gel mobility shift assay method revealed a specifically ultraviolet (UV) damage recognizing, DNA-binding protein in nuclear extracts of normal human cells. The resulted DNA/protein complexes caused the two retarded mobility shifts. Four xeroderma pigmentosum complementation group E (XPE) fibroblast strains derived from unrelated Japanese families were not deficient in such a DNA damage recognition/binding protein because of the normal complex formation and gel mobility shifts, although we confirmed the reported lack of the protein in the European XPE (XP2RO and XP3RO) cells. Thus, the absence of this binding protein is not always commonly observed in all the XPE strains, and the partially repair-deficient and intermediately UV-hypersensitive phenotype of XPE cells are much similar whether or not they lack the protein.

Repair of mitomycin C damage to DNA in mammalian cells and its impairment in Fanconi's anemia cells

Abstract Repair of DNA cross-links by mitomycin C (MMC) was studied in mammalian cells. Skin cells from a patient with Fanconis anemia (FA9 cells) were about 6 times as sensitive to MMC killing as HeLa S3 cells with normal excision repair ability, while excision-reduced mouse L and human xeroderma pigmentosum (XP2OS) cells were more resistant to it than HeLa S3 cells. Alkaline sucrose sedimentation of DNA revealed that perhaps half-excision of cross-links and its repair occurred efficiently until 4 h of post-MMC time in L-cells and, though more slowly, in HeLa S3 cells. Thus, the excision repair pathway is the first step of the cross-link repair in mammalian cells, but it seems different from the uvrA -dependent pathway in E. coli , since XP2OS cells survived MMC almost normally. Contrarily, FA9 DNA sedimented much faster at 4 h of post-MMC time, suggesting a possible impairment in FA cells ability to unhook cross-links.

Characteristics of DNA synthesis following ultraviolet light irradiation in mouse L cells: Postreplication repair☆

Abstract Mouse L cells partially synchronized as to DNA synthesis were tested for ability to bypass UV-induced DNA photoproducts during postirradiation DNA synthesis. The amount of 3 H-thymidine incorporated into the DNA of irradiated cells indicates that dimers remaining unexcised in the DNA, block DNA synthesis, though not permanently. Alkaline sucrose sedimentation has revealed that daughter DNA newly synthesized for 2 h in irradiated cells is discontinuous due to a gap presumably opposite each dimer in parental DNA. Such smaller daughter strands exhibit a dose-dependent reduction in the rate of sedimentation and differ from normally replicating short segments. Taking into account the replication rate reduced by UV irradiation, these gaps in daughter DNA seem to be almost closed by the time the synthesis of the involved DNA units is completed, as the sedimentation rate of daughter DNA approaches that of parental DNA during the subsequent chase incubation of irradiated cells. These results provide evidence for the bypass of photoproducts remaining unexcised in DNA, by which mouse L cells may recover.

Strand-scission of HeLa cell deoxyribonucleic acid by bleomycin in vitro and in vivo

Abstract The ability of bleomycin (BLM) to cause strand breaks of DNA in vitro has been confirmed by means of alkaline sucrose sedimentation. BLM-induced breakage of extracted HeLa S3 cell DNA occurs extensively after dialysis following reaction with 50 μg/ml of BLM. 2-Mercaptoethanol (0·003 M) is not obligatory for BLM-induced degradation of DNA in vitro , but enhances it. The presence of 0·025 M EDTA in the reaction mixture completely prevents single-strand breaks of DNA by BLM. In good agreement with the in vitro results, the DNA from the HeLa cell lysate prepared in a lysing medium containing 0·015 M of EDTA is fragmented a little after a 30-min treatment with 25 μg/ml of BLM, while nonspecific and extensive degradation of DNA occurs when the cells are lysed in the absence of EDTA. A 30-min treatment with BLM also provokes a small amount of unscheduled incorporation of 3 H-thymidine into non-S phase cells, indicating that a small number of single-strand breaks induced are repair-patched. Moreover, 25 μg/ml of BLM exerts a somewhat inhibitory effect on the joining of short segments of replicating DNA after a 30-min 3 H-thymidine pulse, but the joining ability is soon resumed. These data suggest that BLM may either hardly enter HeLa S3 cells or may be readily inactivated.