Removal of O6-methylguanine from DNA of normal and xeroderma pigmentosum-derived lymphoblastoid lines

The origin of O6-methylguanine-DNA methyltransferase in Chinese hamster ovary cells transfected with human DNA

Transfection of Chinese hamster ovary (CHO) cells with human DNA has been shown in several laboratories to produce clones which stably express the DNA-repair protein, O6-methylguanine-DNA methyltransferase (MGMT), that is lacking in the parent cell lines (Mex- phenotype). We have investigated the genetic origin of the MGMT in a number of such MGMT-positive (Mex+) clones by using human MGMT cDNA and anti-human MGMT antibodies as probes. None of the five independently isolated Mex+ lines has human MGMT gene sequences. Immunoblot analysis confirmed the absence of the human protein in the extracts of these cells. The MGMT mRNA in the lines that express low levels of MGMT (0.6-1.4 x 10(4) molecules/cell) is of the same size (1.1 kb) as that present in hamster liver. One cell line, GC-1, with a much higher level of MGMT (4 x 10(4) molecules/cell) has two MGMT mRNAs, a major species of 1.3 kb and a minor species of 1.8 kb. It has also two MGMT polypeptides (32 and 28 kDa), both of which are larger than the 25 kDa MGMT present in hamster liver and other Mex+ transfectants. These results indicate that the MGMT in all Mex+ CHO cell clones is encoded by the endogenous gene. While spontaneous activation of the MGMT gene cannot be ruled out in the Mex+ cell clones, the intervention of human DNA sequences may be responsible for activation of the endogenous gene in the GC-1 line.

Inducibility of the DNA repair gene encoding O6-methylguanine-DNA methyltransferase in mammalian cells by DNA-damaging treatments

The inducibility of the mammalian O6-methylguanine-DNA methyltransferase (MGMT) gene encoding the MGMT protein (EC 2.1.1.63) responsible for removal of the procarcinogenic and promutagenic lesion O6-alkylguanine from DNA was examined by an analysis of transcription of the MGMT gene following exposure of repair-competent (Mex+) and repair-deficient (Mex-) cells to N-methyl-N'-nitro-N-nitrosoguanidine (MNNG). While human and rodent Mex- cells (CHO-9, V79, HeLa MR) showed no detectable MGMT mRNA despite the presence of the gene in their genome, the amount of it in several Mex+ lines (NIH 3T3, HeLa S3, HepG2) paralleled their MGMT activity. However, none of these cell lines showed an increase in the MGMT mRNA level after treatment with various concentrations of MNNG. In contrast, MNNG-treated rat hepatoma cells, H4IIE and FTO-2B, both Mex+, had three- to fivefold more MGMT mRNA than the corresponding untreated controls as measured 12 to 72 h after alkylation. N-Methyl-N-nitrosourea, methyl methanesulfonate, N-hydroxyethyl-N-chloroethylnitrosourea, UV light, and X rays caused a similar accumulation of MGMT mRNA in rat hepatoma cells. Studies with inhibitors of RNA and protein synthesis indicate that the induced increase in the amount of MGMT mRNA was due to enhanced transcription of the gene. Furthermore, they revealed the turnover of the MGMT mRNA to be relatively low (half-life, greater than 7 h). Mutagen-induced increase of transcription of MGMT mRNA in H4IIE cells was accompanied by elevation of MGMT repair activity and resulted in reduction of mutation frequency after a challenge dose of MNNG. Although induction of MGMT mRNA transcription has been observed in two rodent hepatoma cell lines so far, this appears to be the first demonstration of inducibility of a mammalian gene encoding a clearly define DNA repair function. The transcription activation of the MGMT gene protects cells from the mutagenic effects of methylating agents.

Repair of O6-ethylguanine in DNA protects rat 208F cells from tumorigenic conversion by N-ethyl-N-nitrosourea

O6-Ethylguanine (O6-EtGua) is one of about a dozen different alkylation products formed in the DNA of cells exposed to the alkylating N-nitroso carcinogen N-ethyl-N-nitrosourea (EtNU). We have evaluated selectively the relative capacity of cells for the specific enzymatic repair of O6-EtGua as a determinant for the probability of malignant conversion. Eleven O6-EtGua-repair-proficient (R+) variant subclones were isolated from the O6-EtGua-repair-deficient (R-) clonal rat fibroblast line 208F by selection for resistance to 1,3-bis-(2-chloroethyl)-1-nitrosourea (frequency, approximately equal to 10(-5). Contrary to the 208F wild-type cells, all variants expressed O6-methylguanine-DNA methyltransferase activity, while both kinds of cells were deficient for repair of the DNA ethylation products O2- and O4-ethylthymine. After exposure to EtNU (less than or equal to 500 micrograms/ml; 20 min), cells were analyzed for the formation of piled-up foci in monolayer culture and of anchorage-independent colonies in semisolid agar medium. Depending on the EtNU concentration, the frequencies of piled-up foci and agar colonies, respectively, in the R+ variants were as low as 1/28th and 1/56th of those in the R- wild type. Contrasting with the cells from R+ variant-derived agar colonies, cells from 208F (R-) agar colonies gave rise to highly malignant tumors when implanted subcutaneously into syngeneic rats. No significant differences in the frequencies of piled-up foci were found between wild-type and variant cells after exposure to the major reactive metabolite of benzo[a]pyrene, (+)-7 beta, 8 alpha-dihydroxy-9,10 alpha-epoxy-7,8,9,10 alpha-tetrahydrobenzo[a] pyrene, for which stable binding to guanine O6 in cellular DNA has not been observed. The relative capacity of cells for repair of O6-alkylguanine is, therefore, a critical determinant for their risk of malignant conversion by N-nitroso carcinogens.

Sensitivity of fibroblasts derived from ataxia-telangiectasia patients to calicheamicin gamma 1I

We report the hypersensitivity of SV40-transformed fibroblasts derived from ataxia telangiectasia (AT) patients to calicheamicin gamma 1I. In common with other free-radical generating agents such as bleomycin and ionizing radiation, treatment with calicheamicin gamma 1I reveals AT derived lines to be 6-fold more sensitive to this drug when compared to controls. Furthermore, in common with ionizing radiation, AT cells did not show dose-dependent inhibition of DNA synthesis after treatment with calicheamicin gamma 1I.

Regulation of O6-methylguanine-DNA methyltransferase expression in the Burkitt's lymphoma cell line Raji

We have investigated the expression of the DNA-repair enzyme O6-methylguanine-DNA methyltransferase in the Burkitt's lymphoma cell line Raji. An existing mutant Raji cell line which lacks thymidine kinase activity had previously been shown to be Mex- and to no longer express O6-methylguanine-DNA methyltransferase. We report here that in addition to the methyltransferase and thymidine kinase, a third enzyme with an unrelated function, galactokinase, is also not expressed in Raji cells. The control of thymidine kinase expression is post-transcriptional and it is possible that galactokinase and methyltransferase can share a common post-transcriptional regulation with thymidine kinase.

The control of O6-methylguanine-DNA methyltransferase (MGMT) activity in mammalian cells: a pre-molecular view

Human peripheral blood lymphocytes (PBLs) can have a range of O6-methylguanine-DNA methyltransferase (MGMT) activities. PBLs from some individuals may have almost no MGMT activity. Such individuals have most often been subject to malignancy or to immunodeficiency disease. Long-term lymphoblastoid lines (LCLs) prepared from PBLs of normal subjects by Epstein-Barr virus (EBV) transformation have MGMT activities which are in general somewhat higher than the PBLs from which they derive. Such cultures are therefore generally MGMT-positive. Only in rare cases, and generally from patients with low MGMT activity, are freshly obtained lines with very low activity obtained. There is however a 4-fold range of MGMT activity over which multiple lines derived from the same PBL sample can be found. Long-term cultivation can lead to LCLs with low activity as well as to lines of high activity. On rare occasions an MGMT-positive line may, within a few divisions, give a negative line. Some (but not all) MGMT-negative (or very low) lines have been known to gain (some) activity. Chinese hamster ovary (CHO) cell lines are in general very low in MGMT activity. Lines of higher activity can be selected by treatment with mutagenic crosslinking alkylating agents. Chinese hamster lines with high MGMT activity can be obtained by transfection with human DNA from MGMT-positive cells. Lines with significant activity can also be obtained by transfection of CHO cells with human DNA from MGMT-negative (or very low) cells. Resistance to MNNG treatment can be acquired without the acquisition of significant MGMT activity. Crosses of lines of high and low MGMT activity give equivocal results. Hybrids of low x low activity have no activity. Crosses of positive x positive strains give varied results. It has not been possible to identify MGMT-positive hybrids as including one particular chromosome by this type of experiment. There is no evidence for a general adaptive effect on MGMT synthesis greater than the variation within the cell cycle.

Mismatch binding proteins and tolerance to alkylating agents in human cells

The Mex- (Mer-) phenotype of human cells is characterised by a sensitivity to agents such as N-methyl-N'-nitro-N-nitrosoguanidine (MNNG) and N-methyl-N-nitrosourea (MNU). The hypersensitivity of Mex- cells is a consequence of their failure to express the DNA-repair enzyme m6-Gua-DNA methyltransferase. Resistance to MNNG and MNU may be acquired by Mex- cells either by reexpression of a methyltransferase function or by an ill-defined process of tolerance in which the cytotoxic potential of m6-Gua is circumvented without the altered base being removed from DNA. It has been suggested that tolerance might involve an altered mismatch correcting function. We have investigated proteins which recognise and bind specifically to DNA fragments containing single-base mismatches. Cell-free extracts of a Burkitt's lymphoma cell line (Raji) contain two such mismatch binding activities. Neither protein appears to have a high affinity for m6-Gua-containing base pairs. The data indicate that m6-Gua-containing base pairs might be poor substrates for mismatch repair processes in human cells.

Complementation of N-hydroxyethyl-N-chloroethylnitrosourea hypersensitivity of Mex- cells by microinjection of mRNA from Mex+ cells

Chinese hamster ovary (CHO-9) and HeLa MR cells lack detectable amounts of O6-alkylguanine DNA alkyltransferase (phenotypically Mex-) and are hypersensitive to the toxic effect of N-hydroxyethyl-N-chloroethylnitrosourea (HeCNU), as compared to Mex+ derivatives. Microinjection of size-fractionated polyA+ mRNA extracted from HeLa S3 (Mex+) into CHO-9 and HeLa MR cells, or from ataxia telangiectasia (Mex+) into HeLa MR cells, gave rise to an increase in survival following treatment with toxic doses of HeCNU. Transient complementation of the Mex- phenotype was achieved with an RNA population 0.8-1 kb in size.

Immunological alteration of the Bloom's syndrome uracil DNA glycosylase in Epstein-Barr virus-transformed human lymphoblastoid cells

The immunological reactivity of the uracil DNA glycosylase was investigated in three Epstein-Barr virus-transformed human lymphoblastoid cell lines. Two were derived from normal human lymphocytes while the third was derived from a Bloom's syndrome patient. A panel of 3 anti-human placental uracil DNA glycosylase monoclonal antibodies (37.04.12, 40.10.09 and 42.08.07) was used. Immunological reactivity was determined in a double-blind enzyme-linked immunosorbent assay (ELISA); by inhibition of enzyme activity; and by immunoblot analysis. In the ELISA, the glycosylase from each lymphoblastoid cell line was recognized by glycosylase antibodies 37.04.12 and 42.08.07. In contrast, antibody 40.10.09 failed to recognize the glycosylase from the Bloom's syndrome cell line. Further analysis demonstrated that the 40.10.09 antibody was unable to inhibit catalysis by the Bloom's syndrome lymphoblast glycosylase. In contrast, the 40.10.09 antibody inhibited the activity of the two normal human lymphoblast enzymes. Denaturation of the Bloom's syndrome lymphoblast glycosylase rendered that protein immunoreactive with the 40.10.09 antibody. These results demonstrated that: (1) the immunological alteration in the Bloom's syndrome uracil DNA glycosylase was detected in hematopoietic cells; and (2) viral transformation did not affect the immunoreactivity of the enzyme from either normal human or Bloom's syndrome cells.

Active site amino acid sequence of the bovine O6-methylguanine-DNA methyltransferase

An O6-methylguanine-DNA methyltransferase has been partially purified from calf thymus by conventional biochemical techniques. The enzyme was specifically radioactively labelled at the cysteine residue of the active site and further purified by attachment to a solid support. Following digestion with trypsin, a radioactive peptide containing the active site region of the protein was purified by size fractionation, ion exchange chromatography and reverse phase HPLC. The technique yielded an essentially homogeneous oligopeptide which was subjected to amino acid sequencing. The sequence adjacent to the acceptor cysteine residue of the bovine protein exhibits striking homology to the C-terminal methyl acceptor site of the E. coli Ada protein and the proposed acceptor sites of the E. coli Ogt and the B. subtilis Dat1 proteins.

Isolation and structural characterization of a cDNA clone encoding the human DNA repair protein for O6-alkylguanine

O6-Methylguanine-DNA methyltransferase (MGMT; DNA-O6-methylguanine:protein-L-cysteine S-methyltransferase, EC 2.1.1.63), a unique DNA repair protein present in most organisms, removes the carcinogenic and mutagenic adduct O6-alkylguanine from DNA by stoichiometrically accepting the alkyl group on a cysteine residue in a suicide reaction. The mammalian protein is highly regulated in both somatic and germ-line cells. In addition, the toxicity of certain alkylating drugs in tumor and normal cells is inversely related to the levels of this protein. The cDNA of the human gene, henceforth named MGMT, has been cloned in an expression vector on the basis of its rescue of a methyltransferase-deficient (ada-) Escherichia coli host. A 22-kDa active methyltransferase encoded entirely by the cDNA contains an amino acid sequence of 61 residues that bears 60-65% similarity with segments of E. coli methyltransferase (products of the ada and ogt genes), which encompass the alkyl-acceptor residues. The human cDNA has no sequence similarity with the ada and ogt genes, due in part to differences in codon usage, and shows no detectable homology with E. coli genomic DNA. However, it hybridizes with distinct restriction fragments of human, mouse, and rat DNAs. The lack of methyltransferase observed in many human cell lines is due to the absence of the MGMT gene or to lack of synthesis and/or stability of its 0.95-kilobase poly(A)+ RNA transcript.

Streptozotocin interactions with pancreatic beta cells and the induction of insulin-dependent diabetes

The MSZ diabetic male mouse represents one of the most useful tools available to researchers interested in analyzing the consequences of insulin dependent diabetes in male mice. In contrast to the high mortality induced by single high doses of SZ, protracted administration of smaller SZ dosages yields a more stable diabetic condition. Moreover, in insulitis prone strains such as BKs, the model allows "synchronization" of beta cell destruction such that the inflammatory events occur on a predictable timescale. The MSZ-diabetic mouse represents a diabetic condition in which the primary etiopathologic effect is produced by an environmental toxin, and not by a genetically programmed loss of tolerance to beta cell specific antigens. In this regard, etiopathogenesis in the MSZ model is quite distinct from that underlying autoimmune type I diabetes in humans, NOD mice, and BB rats, and it is probably not appropriate to refer to pathogenesis in the MSZ model as one of "autoimmune insulitis" as has sometimes been done. The fact that insulitis in the MSZ model may not be "autoimmune," but may actually be a normal response to either tissue damage or to beta cells that have been structurally modified by a chemical, makes the model of special interest. Clearly, there is no single cause of insulin dependent diabetes, with disease induction representing a genetic susceptibility interacting with environmental triggers, such as toxins in the diet (including nitrosamines and fungal metabolites) as well as pathogenic viruses. The MSZ model will continue to be actively investigated because of insights it will afford regarding the genetic bases for susceptibility and resistance to diabetogenic environmental toxins. The model will be of further value by contributing to knowledge of the complicated interactions between pancreatic islet cells, other endocrine cells, and leukocytes in maintenance of glucose homeostasis.

N-methyl-N-nitrosourea-induced mutations in a shuttle plasmid replicated in human cells

The supF gene of the recombinant shuttle plasmid pZ190 (modified pZ189) was used as a target to study the nature of mutations induced by N-methyl-N-nitrosourea (MNU) in human cells. Treatment of the intact plasmid with MNU followed by its replication in human lymphoblastoid cells led to extensive inactivation and no detectable mutations of the plasmid. However, exposure of the supF DNA fragment alone, followed by its ligation into the vector, caused a ten-fold increase in mutant frequency when replicated in O6-methylguanine-DNA methyltransferase-deficient cells (from 0.54 x 10(-3) to 5.8 x 10(-3)) and an 80-fold increase when replicated in cells containing normal levels of the enzyme (from 0.047 x 10(-3) to 3.8 x 10(-3)). About 45% of the mutant plasmid molecules recovered from human cells contained deletions and insertions. Sixty to 70% of the mutant molecules of wild-type size contained a single-base substitution. Most of these changes were of the G.C----A.T type, consistent with the hypothesis that O6-methylguanine is the primary mutagenic adduct induced by MNU. However, the distribution of mutation sites was highly nonrandom; more than half of all mutations were localized at the G.C position 123, and the rest were distributed in about a dozen sites. The high yield of mutations induced in the supF DNA in a host cell whose capacity for the removal of O6-methylguanine far exceeded the amount present in the supF suggests that the repair of damages in extrachromosomal DNA may be inefficient. This is supported by the observation that the yield of mutations in supF transfected into lymphoblastoid cells devoid of repair activity for O6-methylguanine was comparable to that observed with repair-proficient host cells. The present data, together with results of mutations induced in pZ189 by other agents, strongly suggest that one major determinant of mutational hot spots is the structure of the target DNA itself.

Affinity purification and characterization of human O6-alkylguanine-DNA alkyltransferase complexed with BCNU-treated, synthetic oligonucleotide

Tumor cells resistant to chloroethylnitrosourea (CENU) therapy contain high levels of O6-alkylguanine DNA-alkyltransferase (GATase), a DNA repair enzyme that aborts DNA interstrand cross-linking by removing CENU-induced O6-alkylguanine adducts. Because the transferase binds covalently to CENU-treated oligonucleotides, we reacted partially purified GATase from cultured human lymphoblasts with a BCNU-treated, 35S-5'-end-labeled, synthetic oligonucleotide designed to have a polyadenylated 3' terminus. Immunoprobing Western blots of this reaction mixture with GATase-specific monoclonal antibody indicated that 25-30% of the transferase became complexed. We purified this complex by affinity chromatography with oligo(dT) cellulose, recovering homogenous material that appeared as a discrete 35-kDa Coomassie blue or silver-stained band after SDS-polyacrylamide gel electrophoresis. Autoradiography and Western immunoblotting confirmed that this band contained both the radiolabeled oligonucleotide and the GATase protein.

A reduced rate of bulky DNA adduct removal is coincident with differentiation of human neuroblastoma cells induced by nerve growth factor

Human SY5Y neuroblastoma cells which were differentiated in culture by treatment with 7S murine nerve growth factor for 5 weeks and selection with aphidicolin (L. Jensen, Dev. Biol. 120:56-64, 1987) demonstrated a considerably slower rate of removal of DNA adducts of benzo[a]pyrene, benzo[a]pyrenediolepoxide, and N7-methylguanine than did undifferentiated mitotic cells. A dramatic decline in unscheduled DNA synthesis induced by UV radiation was similarly observed. DNA polymerase beta and uracil DNA glycosylase were unchanged after differentiation, DNA polymerase alpha and DNA methylase decreased roughly threefold, and total apurinic-apyrimidinic endonuclease activity increased roughly threefold after treatment.

Absence of DNA damage-mediated induction of human methyltransferase specific for precarcinogenic O6-methylguanine

The ability of cultured normal human fetal liver and kidney epithelial cells to repair the premutagenic and precarcinogenic O6-methylguanine (O6-MeGua) DNA adduct was determined by directly monitoring its loss in cellular DNA and quantitating the number of O6-MeGua-DNA-methyltransferase (O6-MT) molecules per cell. Following treatment of the epithelial cells with the direct acting carcinogen N-methyl-N-nitrosourea (MNU), the loss of the O6-MeGua adduct was biphasic, exhibiting a half-life of 2.0 and 1.5 h in the liver and kidney cells, respectively. The activity of O6-MT in the liver and kidney epithelial cells in culture was 0.19 pmol/mg protein or 18,500 molecules/cell. The activity of O6-MT was maintained throughout the life of the cultures, i.e., 20 subpassages or 50 cumulative population doublings for the liver and kidney. In order to ascertain whether human fetal epithelial cells exhibit an induction of O6-MT, the cell cultures were treated with single and multiple conditioning doses of N-methyl-N-nitro-N-nitroso-guanidine (MNNG) or gamma-irradiated and assayed for the amount of O6-MT. A 1 h exposure of cells to 2, 4, and 8 microM MNNG resulted in an 80-100% decrease of the initial O6-MT activity which was restored to the constitutive levels within 48 and 72 h post-treatment. Rat hepatoma cells, used as a positive control, increased their levels of O6-MT to 2.8-fold the constitutive levels following treatment with MNNG. Treatment of the human liver and kidney epithelial cells with chronic low doses of MNNG exhibited O6-MT levels identical to untreated cells. The O6-MT activity in epithelial cells remained unaffected upon pre-irradiation with 1.2 or 2.5 Gy of gamma-irradiation.

A reduced rate of bulky DNA adduct removal is coincident with differentiation of human neuroblastoma cells induced by nerve growth factor

Human SY5Y neuroblastoma cells which were differentiated in culture by treatment with 7S murine nerve growth factor for 5 weeks and selection with aphidicolin (L. Jensen, Dev. Biol. 120:56-64, 1987) demonstrated a considerably slower rate of removal of DNA adducts of benzo[a]pyrene, benzo[a]pyrenediolepoxide, and N7-methylguanine than did undifferentiated mitotic cells. A dramatic decline in unscheduled DNA synthesis induced by UV radiation was similarly observed. DNA polymerase beta and uracil DNA glycosylase were unchanged after differentiation, DNA polymerase alpha and DNA methylase decreased roughly threefold, and total apurinic-apyrimidinic endonuclease activity increased roughly threefold after treatment.

Instability of Mex- phenotype in human lymphoblastoid cell lines

Three lymphoblastoid cell lines (LCLs) had extremely low activities of O6-alkylguanine-DNA alkyltransferase (O6-AGT), and were classified as Mex-. They were highly sensitive to cell killing by 1-(4-amino-2-methyl-5-pyrimidinyl)-methyl-3-(2-chloroethyl)-3-nitrosoure a hydrochloride (ACNU), whereas NMO2, a Mex+ LCL with a high O6-AGT activity, was resistant to the agent. Small fractions of these Mex- LCLs survived the treatment with 10 micrograms/ml of ACNU for 24 h, and the surviving cells were found to be resistant to subsequent treatments with the agent. In addition, they contained O6-AGT activities comparable to that of NMO2 and were therefore regarded as Mex+. These results suggest that the Mex- phenotype in LCLs is unstable.

Unique DNA repair properties of a xeroderma pigmentosum revertant

A group A xeroderma pigmentosum revertant with normal sensitivity was created by chemical mutagenesis. It repaired (6-4) photoproducts normally but not pyrimidine dimers and had near normal levels of repair replication, sister chromatid exchange, and mutagenesis from UV light. The rate of UV-induced mutation in a shuttle vector, however, was as high as the rate in the parental xeroderma pigmentosum cell line.

Structural alterations of DNA ligase I in Bloom syndrome

Cell lines derived from seven patients with Bloom syndrome all contain a DNA ligase I with unusual properties. Six lines were shown to have a reduced level of this enzyme activity and the residual enzyme was anomalously heat-labile. The seventh line contained a dimeric rather than monomeric form of ligase I. Several cell lines representative of other inherited human syndromes have apparently normal DNA ligases. The data indicate that Bloom syndrome is due to a defect in the structure of DNA ligase I caused by a "leaky" point mutation occurring at one of at least two alternative sites.

Comparative analysis of O6-methylguanine methyltransferase activity and cellular sensitivity to alkylating agents in cell strains derived from a variety of animal species

Using 26 cultured cell lines derived from 17 different animal species, we have measured both the activity of O6-methylguanine (O6-MeG) methyltransferase (MT) in cell extracts and the sensitivity of the strains to the lethal effects of the alkylating agents, N-methyl-N'-nitro-N-nitrosoguanidine (MNNG) and 1-(4-amino-2-methyl-5-pyrimidinyl)methyl-3-(2-chloroethyl)-3-nitrosourea (ACNU). The MT activity was assayed by measuring the amount of 3H radioactivity transferred from methyl-[3H]-labeled O6-MeG in DNA to acceptor protein molecules in the extracts. In all the 21 mammalian cell strains, lethal sensitivity to ACNU as measured by colony-forming ability correlated well with cellular MT activity, indicating that the major lethal ACNU damage is reparable by the MT. On the other hand, MNNG sensitivity did not necessarily correlate with the MT activity.

Unique DNA repair properties of a xeroderma pigmentosum revertant

A group A xeroderma pigmentosum revertant with normal sensitivity was created by chemical mutagenesis. It repaired (6-4) photoproducts normally but not pyrimidine dimers and had near normal levels of repair replication, sister chromatid exchange, and mutagenesis from UV light. The rate of UV-induced mutation in a shuttle vector, however, was as high as the rate in the parental xeroderma pigmentosum cell line.

The 1987 Walter Hubert lecture. Regulation and deficiencies in DNA repair

A number of rare human inherited syndromes are associated with apparent defects in DNA repair and a greatly increased frequency of cancer. Cell lines derived from such individuals phenotypically resemble certain bacterial mutant strains that have increased sensitivity to radiation or chemical agents and well characterised repair defects. This analogy provides leads for unravelling the molecular alterations in such cancer-prone human cells. The inducibility of DNA repair enzymes is also reviewed. Exposure of bacteria to alkylating agents, or oxygen radicals, causes the overproduction of several novel and interesting repair activities, and the induced bacteria provide an abundant source of these proteins for purification and biological characterisation. Enzymes with the same defined specificities are often present in human cells, presumably serving the same functions as in microorganisms, but these activities are only constitutively expressed at low levels.

Transfer of human genes conferring resistance to methylating mutagens, but not to UV irradiation and cross-linking agents, into Chinese hamster ovary cells

Chinese hamster ovary cells were transfected by human DNA ligated to the bacterial gpt (xanthine-guanine-phosphoribosyltransferase) gene which was used either in its native form or after partial inactivation with methylnitrosourea. The gpt+ transfectants were screened for resistance to high doses of N-methyl-N'-nitro-N-nitrosoguanidine. Using this approach, we showed that Chinese hamster ovary cells can acquire N-methyl-N'-nitro-N-nitrosoguanidine resistance upon transfection with DNA from diploid human fibroblasts, that this resistance is transferable by secondary transfection and is specific for methylating mutagens, and that it is not caused by increased removal of O6-methylguanine, 3-methyladenine, and 7-methylguanine from DNA.

Difference in O6-methylguanine methyltransferase activity among transformed NIH3T3 cell clones

We examined the sensitivity to the lethal effects of methylating agents and the O6-methylguanine methyltransferase (MTR) activities of in vitro transformed NIH3T3 cell clones. The sensitivities to the lethal effects of MNNG were not different among all 49 transformed cell clones examined and do not correlate with the MTR activities. All 8 spontaneously transformed cell clones showed the same sensitivities to ACNU as the parental cell line. 2 of 20 transformants induced by UV or MNNG showed higher sensitivities to the ACNU although the MTR activity was normal. One cell clone transformed by UV was sensitive to ACNU and showed about half MTR activity. 5 of 19 cell clones transformed by oncogenes (Ha-ras or SV40 ori-) were sensitive to the lethal effects of ACNU and showed the low MTR activities, but were not as much sensitive as a Ha-MuSV transformed cell clone, Ha821.

4NQO- or MNNG-resistant variants established from a human cell line, RSb, with high sensitivity to both agents

From a human cell line, RSb, with high sensitivity to the killing effects of 4-nitroquinoline 1-oxide (4NQO), N-methyl-N'-nitro-N-nitrosoguanidine (MNNG) and 254-nm ultraviolet light, a 4NQO-resistant variant, Qr-10, and an MNNG-resistant one, Gr-10, were established using ethyl methanesulfonate as the mutagen. Cell proliferation studies and colony-formation assays revealed that Qr-10 and Gr-10 cells actively proliferated under conditions where RSb cell proliferation was greatly inhibited by 4NQO and MNNG, respectively. Total cellular DNA synthesis, as estimated by [Me-3H]thymidine uptake into acid-insoluble cell materials, was depressed in 4NQO-treated Qr-10 and MNNG-treated Gr-10 cells as it was in chemical-treated RSb cells, but recovered more markedly from such inhibition in the variants. 4NQO- and MNNG-induced DNA-repair replication synthesis was enhanced to a greater extent in Qr-10 and Gr-10 cells, respectively, than in RSb cells. The Qr-10 and Gr-10 cells showed the same respective susceptibility to the effects of MNNG and 4NQO, on cell growth and DNA synthesis and DNA-repair synthesis as did the parent cells. But, Qr-10 cells had more resistance to UV-killing and higher levels of UV-induced DNA-repair synthesis than did RSb cells, while UV-susceptibility of Gr-10 cells was the same as that of the latter.

Transfer of human genes conferring resistance to methylating mutagens, but not to UV irradiation and cross-linking agents, into Chinese hamster ovary cells

Chinese hamster ovary cells were transfected by human DNA ligated to the bacterial gpt (xanthine-guanine-phosphoribosyltransferase) gene which was used either in its native form or after partial inactivation with methylnitrosourea. The gpt+ transfectants were screened for resistance to high doses of N-methyl-N'-nitro-N-nitrosoguanidine. Using this approach, we showed that Chinese hamster ovary cells can acquire N-methyl-N'-nitro-N-nitrosoguanidine resistance upon transfection with DNA from diploid human fibroblasts, that this resistance is transferable by secondary transfection and is specific for methylating mutagens, and that it is not caused by increased removal of O6-methylguanine, 3-methyladenine, and 7-methylguanine from DNA.

DNA-mediated transfer and expression of a human DNA repair gene that demethylates O6-methylguanine

Human liver DNA was transfected into CHO cells (mex-) along with pSV2gpt and colonies were selected first for resistance to mycophenolic acid and then to chloroethylnitrosourea. Transformants were obtained that contained approximately 10,000 molecules of O6-alkylguanine alkyltransferase (mex+) per cell. Their genome contained at least three copies of the human Alu sequence.

Reduction of the toxicity and mutagenicity of alkylating agents in mammalian cells harboring the Escherichia coli alkyltransferase gene

The toxic, mutagenic, and carcinogenic effects of alkylating agents have been attributed to their ability to damage DNA. Reaction at the O6 position of guanine results in miscoding during DNA replication, has been shown to be mutagenic in both bacteriophage and bacteria, and may be responsible for malignant transformation. In common with many other prokaryotes and eukaryotes the Escherichia coli B strain contains a protein that repairs O6-alkylation damage in DNA by transferring the alkyl group to one of its own cysteine residues. We have recently cloned the E. coli O6-alkylguanine alkyltransferase gene and shown it to encode a 37-kDa protein containing an additional activity that removes alkyl groups from alkylphosphotriesters in DNA. To examine the biological effects of this gene in mammalian cells, we have now inserted the coding sequence into a retrovirus-based selectable expression vector and transfected it into Chinese hamster V79 cells that lack endogenous alkyltransferase activity. A clone expressing high levels of the bacterial protein was selected and shown to produce a 37-kDa alkyltransferase protein and to rapidly repair O6-methylguanine produced in the host genome following exposure to N-methyl-N-nitrosourea. In comparison with a control population, this clone is considerably more resistant to the toxic and mutagenic effects of alkylating agents that react extensively with oxygen atoms in DNA. The usefulness of these clones in examining the role of DNA alkylation and other biological effects of alkylating agents is discussed.

Suppression of human DNA alkylation-repair defects by Escherichia coli DNA-repair genes

The ada-alkB operon protects Escherichia coli against the effects of many alkylating agents. We have subcloned it into the pSV2 mammalian expression vector to yield pSV2ada-alkB, and this plasmid has been introduced into Mer- HeLa S3 cells, which are extremely sensitive to killing and induction of sister chromatid exchange by alkylating agents. One transformant (the S3-9 cell line) has several integrated copies of pSV2ada-alkB and was found to express a very high level of the ada gene product, the 39-kDa O6-methylguanine-DNA methyltransferase. S3-9 cells were found to have become resistant to killing and induction of sister chromatid exchange by two alkylating agents, N-methyl-N'-nitro-N-nitrosoguanidine and N,N'-bis(2-chloroethyl)-N-nitro-sourea. This shows that bacterial DNA alkylation-repair genes are able to suppress the alkylation-repair defects in human Mer- cells.

Adaptation-like response to the chemical induction of sister chromatid exchanges in human lymphocytes

Experiments have been performed to determine whether human lymphocytes in primary cultures can show an "adaptive" response to the induction of cellular lesions (manifested as a production of sister chromatid exchanges, SCEs) as previously found in bacteria and established human and mammalian cell lines. Human lymphocytes were pretreated with various subtoxic concentrations (5-50 ng/ml) of N-methyl-N'-nitro-N-nitrosoguanidine (MNNG) once every 6h for 72 h, and subsequently challenged by a high dose (4 micrograms/ml) of MNNG. The lymphocytes in MNNG-challenged cultures had the lowest frequency of SCEs when pretreated with 10 ng/ml MNNG. Further cross-resistance study revealed that repeated pretreatments of lymphocytes with 10 ng/ml MNNG for 72 h can render the cells resistant to the induction of SCEs by the following challenge with a high dose of MNNG, but not of mitomycin C or ethyl nitrosourea. The data also suggest variations in the degree of the adaptation-like response among individuals.

Adenovirus replication as an in vitro probe for drug sensitivity in human tumors

The feasibility of using adenovirus 5 as an in vitro probe for chemosensitivity in short-term cultures of human tumors was evaluated using human melanoma cell lines and primary cultures of melanoma biopsies. A convenient immunoperoxidase method was developed for quantitating viral replication 2 days after infection. Two different approaches were explored: the host cell reactivation assay (HCR) using drug-treated virus; and the viral capacity assay using drug-treated cells. The HCR assay detected sensitivity to 5-(3-methyl-1-triazeno)imidazole-4-carboxamide (MTIC) in Mer- (methyl excision repair deficient) cell lines as decreased ability of the cells to replicate MTIC-treated virus. This test should be applicable to DNA-damaging agents and repair-deficient tumors. Adenovirus replicated readily in nonproliferating primary cultures of melanoma biopsies; application of the HCR assays to this material identified one Mer- sample of 11 tested. Herpes viruses were not suitable for use in HCR because herpes simplex virus type 1 failed to distinguish Mer- from Mer+ melanoma cells; and nonproductive infection of MTIC-sensitive lymphoid cells with Epstein-Barr virus yielded an MTIC-resistant cell line. The second assay (viral capacity) involved determination of the inhibition of replication of untreated virus in treated cells. This approach correctly predicted sensitivity to hydroxyurea and deoxyadenosine in melanoma cell lines when compared with clonogenic survival assay. Viral capacity was also inhibited by cytosine arabinoside, fluorouracil, vincristine, adriamycin, 6-mercaptopurine and ionising radiation, and may therefore be useful for detecting sensitivity to a wide range of antitumor agents.

DNA repair and replication in xeroderma pigmentosum and related disorders

Xeroderma pigmentosum (XP), ataxia telangiectasia (AT), and Cockayne syndrome (CS) are human diseases that exhibit increased sensitivity to environmental carcinogens [e.g., ultraviolet (UV) light, ionizing radiations, chemicals] because of genetic defects in the patient's capacity to repair and replicate damaged DNA accurately. The major defect in XP is a failure to repair UV damage to DNA; in AT, the failure is in repair or replication of double-strand breaks in DNA; in CS, the failure is in recovery of DNA replication after UV irradiation. Cancer is a major clinical feature of XP and AT, but not of CS. Each disease is complex, with multiple groups defined by complementation in cell-cell hybridization. Overlap is reported between some XP and CS groups. UV-sensitive hamster cell mutants are also known: most of these complement XP groups, and a human gene on chromosome 19 can correct the defects in hamster mutants, but not XP. XP group C is distinct from the other groups in exhibiting a strongly clustered mode of repair, as if only certain regions of the genome can be mended. This mode mainly occurs in confluent group C cells under conditions that permit much greater survival than in exponential growth, and therefore represents a more efficient mode of repair. These diseases all represent important examples of perturbation in the way carcinogen damage in DNA is metabolized, and further research aimed at identifying the kinds of molecular changes involved in the malignancy will be important.

Enzymology of DNA replication and repair in the brain

A number of enzymes thought to be involved in DNA replication have been identified in the brain. These include single-stranded DNA-binding proteins, topoisomerases I and II, DNA polymerase alpha, a protein that binds Ap4A and might be classified as a DNA polymerase alpha accessory protein, RNase H, DNA polymerase beta, DNA ligase, an endo- and an exonuclease of unknown function, DNA methyl transferase and poly(ADPR) synthase. In contrast, little is known about the enzymology of DNA repair in brain. The few enzymes identified comprise uracil-DNA glycosylase, DNA polymerase beta, DNA polymerase alpha (which in neurons is present only at immature stages), DNA ligase, poly(ADPR) synthase, and O6-alkylguanine-DNA alkyltransferase. In addition, an exonuclease acting on depurinated single-stranded DNA (tentatively listed here as 3'----5' exonuclease), an endonuclease of unknown function as well as ill-defined acid and alkaline deoxyribonucleases also occur in brain.

O6 alkylguanine-DNA alkyltransferase activity in human myeloid cells

The association between alkylating agent exposure and acute nonlymphocytic leukemia in humans indicates that myeloid cells may be particularly susceptible to mutagenic damage. Alkylating agent mutagenesis is frequently mediated through formation and persistence of a particular DNA base adduct, O6alkylguanine, which preferentially mispairs with thymine rather than cytosine, leading to point mutations. O6alkylguanine is repaired by O6alkylguanine-DNA alkyltransferase (alkyltransferase), a protein that removes the adduct, leaving an intact guanine base in DNA. We measured alkyltransferase activity in myeloid precursors and compared it with levels in other cells and tissues. In peripheral blood granulocytes, monocytes, T lymphocytes, and B lymphocytes, there was an eightfold range of activity between individuals but only a twofold range in the mean activity between cell types. Normal donors maintained stable levels of alkyltransferase activity over time. In bone marrow T lymphocytes and myeloid precursors, there was an eightfold range of alkyltransferase activity between donors. Alkyltransferase activity in the two cell types was closely correlated in individual donors, r = 0.69, P less than 0.005, but was significantly higher in the T lymphocytes than the myeloid precursors, P less than 0.05. Liver contained the highest levels of alkyltransferase of all tissues tested. By comparison, small intestine contained 34%, colon 14%, T lymphocytes 11%, brain 11%, and myeloid precursors 6.6% of the activity found in liver. Thus, human myeloid precursors have low levels of O6alkylguanine-DNA alkyltransferase compared with other tissues. Low levels of this DNA repair protein may increase the susceptibility of myeloid precursors to malignant transformation after exposure to certain alkylating agents.

DNA repair: relationship to drug and radiation resistance, metastasis and growth factors

DNA repair mechanisms are important for the recovery of both normal and malignant tissues from radiation and chemotherapy. Drug 'resistance' may merely reflect the similarity of cancer to normal tissues. Investigating the normal repair mechanisms by cloning human DNA repair genes will permit a much better comparison. Therapeutic inhibition of DNA repair may be possible with poly-ADP-ribose polymerase inhibitors. A differential effect may be obtained since less-differentiated cells have a higher poly-ADP-ribose polymerase activity. Clinical application of repair inhibitors can be achieved by using antimetabolites such as high-dose hydroxyurea which produces levels of 1-3 mmol litre -1/24 hours. The whole cell and tissue response to DNA damage is more complex than removal of adducts and joining strand breaks. DNA damage can result in an increase in growth-factor receptors, the release of soluble mediators that affect undamaged cells and stimulation of plasminogen activator. These changes may enhance growth and recovery as well as bypass or repair the damage. The generation of heterogeneity in a tumour population may be mediated by DNA rearrangements. Genetic instability is much higher in metastatic clones and a comparison of DNA strand-break repair in a metastatic and a non-metastatic line showed more rapid repair in the former. Aberrant use of DNA repair stimulated by growth factors may mediate tumour progression and heterogeneity as well as drug resistance.

DNA-mediated transfer and expression of a human DNA repair gene that demethylates O6-methylguanine

Human liver DNA was transfected into CHO cells (mex-) along with pSV2gpt and colonies were selected first for resistance to mycophenolic acid and then to chloroethylnitrosourea. Transformants were obtained that contained approximately 10,000 molecules of O6-alkylguanine alkyltransferase (mex+) per cell. Their genome contained at least three copies of the human Alu sequence.

Exposure of HeLa cells to 0(6)-alkylguanines increases sensitivity to the cytotoxic effects of alkylating agents

Exposure of HeLa S3 cells to 0.4mM 0(6)-methylguanine or 0(6)-n-butylguanine for 24 h led to a substantial decrease in the activity of 0(6)-alkylguanine-DNA alkyltransferase. Such pretreatment caused a marked increase in the sensitivity of the cells to the cytotoxic effects of the cross-linking alkylating agent 1-(2-chloroethyl)-1-nitroso-3-cyclohexylurea and a smaller increase in the sensitivity to N-methyl-N'-nitro-N-nitrosoguanidine. These results indicate that the repair of DNA by the alkyltransferase plays an important role in the protection of cells from the cytotoxic effects of certain alkylating agents particularly those such as 1-(2-chloroethyl)-1-nitroso-3-cyclohexylurea which ultimately lead to the formation of lethal interstrand cross-links.

Possible depletion of a DNA repair enzyme in human lymphoma cells by subversive repair

Mex+ human lymphoma cell lines contain O6-methylguanine-DNA methyltransferase, a DNA repair enzyme that undergoes suicide inactivation on interaction with its substrate. The cells are therefore competent to remove the alkylation lesion O6-methylguanine from their DNA. However, several repair-deficient lymphoma cell lines (Mex-) are also known. It is shown here that Mex+ cells can be converted temporarily to a Mex- phenotype by growth in nontoxic concentrations of free O6-methylguanine. The depletion of methyltransferase activity is not a result of O6-methylguanine incorporation into DNA and subsequent demethylation by the enzyme. It is proposed that O6-methylguanine is mistakenly incorporated into tRNA molecules by means of a post-transcriptional ribosyl transfer reaction. The demethylation of such bases in tRNA has been demonstrated by using bacterial and human DNA repair enzymes. The existence of such a subversive repair of a methylated base in tRNA raises the possibility of competition between DNA and RNA for cellular DNA repair enzymes. Furthermore, it is proposed that the known aberrant methylation of tRNA in certain transformed cells, together with subversive tRNA repair, could account for the Mex- phenotype.

Quantitation of DNA repair capacities of human tumor cells by estimation of transfer of DNA adducts to repair proteins

The repair of O6-methylguanine produced in DNA by alkylating agents is accomplished by a unique lesion reversal mechanism which recognizes the methyl group and transfers it to itself in a suicide reaction. Much of what we know about the importance of O6-methylguanine-DNA methyltransferase repair in human cells comes from the study of Mer- tumor cell strains which are deficient in transferase activity. The human transferase has a preference for repair of methyl groups, but will also act on other substrates. Assays for transferase activity detect either the loss of O6-methylguanine from DNA or the appearance of methylated protein. A new assay detects the recovery of a restriction site in a synthetic polymer following demethylation. Inhibition of transferase activity can be produced in cells by several methods.