Inducible expression and cytogenetic effects of the EcoRI restriction endonuclease in Chinese hamster ovary cells.

The cytogenetic endpoints sister chromatid exchange (SCE) and chromosome aberrations are widely used as indicators of DNA damage induced by mutagenic carcinogens. Chromosome aberrations appear to result directly from DNA double-strand breaks, but the lesion(s) giving rise to SCE formation remains unknown. Most compounds that induce SCEs induce a spectrum of lesions in DNA. To investigate the role of double-strand breakage in SCE formation, we constructed a plasmid that gives rise to one specific lesion, a staggered-end ("cohesive") DNA double-strand break. This plasmid, designated pMENs, contains a selectable marker, neo, which is a bacterial gene for neomycin resistance, and the coding sequence for the bacterial restriction endonuclease EcoRI attached to the mouse metallothionein gene promoter. EcoRI recognizes G decreases AATTC sequences in DNA and makes DNA double-strand breaks with four nucleotides overhanging as staggered ends. Cells transfected with pMENS were resistant to the antibiotic G418 and contained an integrated copy of the EcoRI gene, detectable by DNA filter hybridization. The addition of the heavy metal CdSO4 resulted in the intracellular production of EcoRI, as measured by an anti-EcoRI antibody. Cytogenetic analysis after the addition of CdSO4 indicated a dramatic increase in the frequency of chromosome aberrations but very little effect on SCE frequency. Although there was some intercellular heterogeneity, these results confirm that DNA double-strand breaks do result in chromosome aberrations but that these breaks are not sufficient to give rise to SCE formation.

Radon-induced deletions in human cells: role of nonhomologous strand rejoining.

Radon is a ubiquitous inhaled human carcinogen that is thought to be the largest single natural source of human exposure to radiation. We report that a freely replicating episome in human cells exposed to radon gas underwent mutagenic changes, a high proportion of which were large deletions involving many thousands of base pairs. These deletions were not randomly distributed but started and ended in defined regions as if caused by the passage of a single alpha-particle track through a coiled chromatin structure. The sizes appeared to be defined by structural features of chromatin: the minimum size was 2435 base pairs, and the maximum size was 8051 base pairs, close to the upper limit that would leave intact the plasmid sequences required for selection in bacteria. Ends were rejoined by nonhomologous recombination involving up to 6 base pairs of homology. This process may not be confined to the repair of exogenously induced double-strand breaks but may be used for rejoining free DNA ends generated by a variety of cellular processes. The mechanism of alpha-particle deletion mutagenesis may account for the high relative biological effectiveness of radon irradiation for many end points and its consequences for lung carcinogenesis.

Determination of tissue distribution of an intramuscular plasmid vaccine using PCR and in situ DNA hybridization.

The increasing use of nucleic acid-based therapeutics has created a need for new methods of determining tissue distribution and levels. Radiolabel methods may not always be appropriate because nucleic acids are easily degraded. Quantitation using the polymerase chain reaction (PCR) has the advantage that only continuous stretches of DNA will be amplified. In situ hybridization allows detection of specific sequences in histological preparations. We have used quantitative PCR and in situ hybridization techniques to study the pharmacokinetics and distribution of PGagPol (a potential anti-HIV plasmid vaccine) in rabbits. Samples were obtained 4 hr, 24 hr, 7 days, and 28 days after intramuscular injection of 100 micrograms or 400 micrograms of plasmid. A simplified procedure for collecting and processing tissues for PCR that minimizes the risk of contamination was developed. Using PCR, plasmid was found principally in the skin and muscle of the injection site and in blood plasma. At 4 hr after dosing with 400 micrograms, the plasmid was detected at the injection site with mean copy numbers of 10(6) (in muscle) and 4 x 10(4) (in skin) per microgram of tissue. Plasmid copy number declined rapidly in muscle during the first 24 hr and was undetectable at 7 and 28 days after injection. The decline was slower in the skin, and the plasmid was still detectable at 28 days. With in situ hybridization, plasmid was detected in muscle, mainly in the perimysium and to a lesser degree in the endomysium and within the muscle fibers. These data indicate that quantitative PCR and in situ hybridization are sensitive methods for examining tissue distribution of DNA used for gene therapy.

Induction of chromosome damage by restriction enzymes during mitosis.

Once electroporated into the nucleus of eukaryotic cells, restriction enzymes will bind at specific DNA sequences and cleave DNA to make double-strand breaks. These induced breaks can lead to chromosome aberrations and consequently offer one approach to determining the mechanism(s) of aberration formation. Because the higher-order structure of DNA in eukaryotic cells might influence the ability of restriction enzymes to locate their recognition sequence, bind, and cleave DNA, we have investigated whether enzymes will cut DNA during metaphase when the chromosomes are most condensed. Chinese hamster ovary cells synchronized in mitosis and treated with either AluI or Sau3AI showed few chromosome aberrations when held in mitosis for 1, 2, or 3 h after enzyme treatment. However, some disruption of chromosome morphology was seen, especially after exposure to Sau3AI. When cells were allowed to complete one cell cycle after enzyme treatment in the preceding mitosis, there was extensive chromosome damage, with the most abundant type of lesion being the interstitial deletion. It appears that restriction enzymes will cleave the highly condensed DNA in mitotic cells but that decondensation, DNA replication, and recondensation are required before the aberrations are manifested.

Analysis of restriction enzyme-induced DNA double-strand breaks in Chinese hamster ovary cells by pulsed-field gel electrophoresis: implications for chromosome damage.

Restriction enzymes can be electroporated into mammalian cells, and the induced DNA double-strand breaks can lead to aberrations in metaphase chromosomes. Chinese hamster ovary cells were electroporated with PstI, which generates 3' cohesive-end breaks, PvuII, which generates blunt-end breaks, or XbaI, which generates 5' cohesive-end breaks. Although all three restriction enzymes induced similar numbers of aberrant metaphase cells, PvuII was dramatically more effective at inducing both exchange-type and deletion-type chromosome aberrations. Our cytogenetic studies also indicated that enzymes are active within cells for only a short time. We used pulsed-field gel electrophoresis to investigate (i) how long it takes for enzymes to cleave DNA after electroporation into cells, (ii) how long enzymes are active in the cells, and (iii) how the DNA double-strand breaks induced are related to the aberrations observed in metaphase chromosomes. At the same concentrations used in the cytogenetic studies, all enzymes were active within 10 min of electroporation. PstI and PvuII showed a distinct peak in break formation at 20 min, whereas XbaI showed a gradual increase in break frequency over time. Another increase in the number of breaks observed with all three enzymes at 2 and 3 h after electroporation was probably due to nonspecific DNA degradation in a subpopulation of enzyme-damaged cells that lysed after enzyme exposure. Break frequency and chromosome aberration frequency were inversely related: The blunt-end cutter PvuII gave rise to the most aberrations but the fewest breaks, suggesting that it is the type of break rather than the break frequency that is important for chromosome aberration formation.

Modulation of restriction enzyme-induced damage by chemicals that interfere with cellular responses to DNA damage: a cytogenetic and pulsed-field gel analysis.

The electroporation of restriction enzymes into mammalian cells results in DNA double-strand breaks that can lead to chromosome aberrations. Four chemicals known to interfere with cellular responses to DNA damage were investigated for their effects on chromosome aberrations induced by AluI and Sau3AI; in addition, the number of DNA double-strand breaks at various times after enzyme treatment was determined by pulsed-field gel electrophoresis (PFGE). The poly(ADP-ribose) polymerase inhibitor 3-aminobenzamide (3AB) dramatically increased the yield of exchanges and deletions and caused a small but transitory increase in the yield of double-strand breaks induced by the enzymes. 1-beta-D-Arabinofuranosylcytosine, which can inhibit DNA repair either by direct action on DNA polymerases alpha and delta or by incorporation into DNA, potentiated aberration induction but to a lesser extent than 3AB and did not affect the amount of DNA double-strand breakage. Aphidicolin, which inhibits polymerases alpha and delta, had no effect on AluI-induced aberrations but did increase the aberration yield induced by Sau3AI. The postreplication repair inhibitor caffeine had no effect on aberration yields induced by either enzyme. Neither aphidicolin nor caffeine modulated the amount of DNA double-strand breakage as measured by PFGE. These data implicate poly(ADP-ribosyl)ation and polymerases alpha and delta as important components of the cellular processes required for the normal repair of DNA double-strand breaks with blunt or cohesive ends. Comparison of these data with the effect of inhibitors on the frequency of X-ray-induced aberrations leads us to the conclusion that X-ray-induced aberrations can result from the misjoining or nonrejoining of double-strand breaks, particularly breaks with cohesive ends, but that this process accounts for only a portion of the induced aberrations.

Genetic toxicology testing of the antimalarial drugs chloroquine and a new analog, AQ-13.

AQ-13 ([N1-(7-chloro-quinolin-4yl)-3-(N3,N3-diethylamino)propylamine] dihydrochloride trihydrate) is an aminoquinoline antimalarial drug that is effective against chloroquine-resistant strains of Plasmodium falciparum. It is structurally similar to the widely used chloroquine diphosphate (CQ). We evaluated these drugs in the three assays currently recommended by the International Conference on Harmonization (ICH): bacterial mutagenesis in Salmonella typhimurium and Escherichia coli, mammalian cell mutagenesis in L5178Y mouse lymphoma cells, and micronucleus induction in rat bone marrow. A small but statistically significant increase in revertant colonies was produced by CQ with Salmonella tester strain TA98 without metabolic activation (MA) and by AQ-13 with strain TA1537 both with and without MA. In L5178Y cells, testing of CQ and AQ-13 up to cytotoxic concentrations with and without MA produced no increase in mutant colonies and no increase in the numbers of small colonies. Slight decreases in the ratio of polychromatic erythrocytes (PCE) to red blood cells (RBC) were observed in male and female rats treated with CQ and in females only treated with AQ-13; however, none of these changes was statistically significant. No increases in the frequency of micronucleated PCE were observed at any dose level of CQ or AQ-13. Although both CQ and AQ-13 showed weak bacterial mutagenicity, this mutagenic effect was not confirmed in either the mouse lymphoma mutagenesis assay or the micronucleus assay. These results indicate that CQ and AQ-13 should pose minimal risk of genotoxic damage in human populations being administered these drugs.

The induction of chromosome aberrations by restriction endonucleases that produce blunt-end or cohesive-end double-strand breaks.

Restriction endonucleases have been used to study the involvement of specific types of DNA damages in the production of chromosome aberrations. In this study restriction endonucleases were introduced into viable CHO cells using osmolytic shock of pinocytic vesicles. We compared two cohesive-end cutters, Msp I (CCGG-2-base overlap) and Sau3A I (GATC-4-base overlap) with two blunt-end cutters, Alu I (AGCT) and Rsa I (GTAC). All 4 enzymes were effective at inducing aberrations. The 4-base overlap cohesive-end cutter Sau3A I was approximately as effective as the blunt-end cutter Alu I. We present evidence that cutting frequency rather than cut end-structure is important in determining efficiency of aberration induction. There is over-dispersion of the distribution of dicentrics and rings among cells, and the data could be fitted to a Neyman Type A distribution, a modified Poisson, that indicates that there is a probability distribution both for the entry of the enzyme into a cell nucleus and for the induction of aberrations once the enzyme has entered a cell nucleus. In addition, we used Alu I to determine the sensitivity of cells to aberration induction in the different stages of the cell cycle. Alu I induced aberrations in all stages of the cycle, chromatid-type in S/G2 and chromosome-type in G1. In agreement with data of others, there were variations in sensitivity with cycle stage, and changes in the proportions of the different aberration classes for chromatid-type aberrations.

pHAZE: a shuttle vector system for the detection and analysis of ionizing radiation-induced mutations.

We have designed and constructed a shuttle vector system, pHAZE, that is maintained as an episome in normal human fibroblasts, has a low background mutation frequency, and is capable of detecting a spectrum of mutations, including deletions up to 8.3 kb. The efficacy of this system was demonstrated by using it to analyze mutations produced by X-rays, which induced both point mutations and large deletions in pHAZE.

Analysis of the mutagenic potential of ENU and MMS in germ cells of male C57BL/6 lacI transgenic mice.

Mutant frequencies in male germ cells were determined in mice 3 days after exposure to saline, methylmethane sulfonate (MMS), or ethylnitrosourea (ENU). DNA was isolated from seminiferous tubules by a modified version of the drop dialysis method. A 5-fold increase in mutant frequency was observed in mice treated with ENU. No statistically significant increase was observed in mice treated with MMS.

Mechanisms involved in rejoining DNA double-strand breaks induced by ionizing radiation and restriction enzymes.

DNA double-strand breaks are considered to be the most deleterious lesion induced by ionizing radiation. However, the mechanism of rejoining of these lesions has not been extensively studied at the molecular level. We have used a shuttle vector, pHAZE, to analyze the mechanism of rejoining of DNA double-strand breaks in human cells. The advantage of this vector system is that, unlike many previously described shuttle vectors, it has a large target gene for the detection of deletions and it is maintained as a freely replicating episome with chromatin conformation in the nucleus of human cells. In this study we compare data obtained on the spectrum of mutations induced in pHAZE by ionizing radiation (alpha-particles) and restriction enzymes (PvuII, ClaI, and PvuI). Unlike ionizing radiation, restriction enzymes induce double-strand breaks in DNA with known end structures at defined locations and therefore provide a model system for analyzing cellular responses to DNA double-strand breaks. Exposure of human cells containing the vector to alpha-particle irradiation produced both point mutations and large deletions in pHAZE. When the junction regions of the deletions were sequenced it was found that 65% were rejoined with up to 6 bp of homology at the junction region. Analysis of restriction-enzyme-induced mutations suggests that double-strand break ends are modified to facilitate rejoining and that the type of modification is characteristic for different end structures. Double-strand breaks with cohesive ends appear to have fewer modifications introduced at the break points before rejoining than breaks with blunt ends. When considered in relation to the data obtained with ionizing radiation this suggests that the presence of cohesive sequences either at, or in proximity to, the ends enhances rejoining of DNA double-strand breaks.

Restriction endonucleases do not induce sister-chromatid exchanges in Chinese hamster ovary cells.

Bacterial restriction enzymes offer the unique opportunity to determine the biological and cytogenetic consequences of DNA double-strand breakage. To examine the role of various types of breaks in sister-chromatid exchange (SCE) formation, we used restriction enzymes with different recognition sequences and different cutting frequencies to generate DNA double-strand breaks in Chinese hamster ovary (CHO) cells. The restriction enzymes were introduced by electroporation into exponentially growing cells during the second replication cycle in bromodeoxyuridine, and SCEs were analyzed at mitosis. Contrary to results reported by others, we found no increase in SCE frequency in cells exposed to restriction enzymes despite the presence of numerous cells with chromatid aberrations. These data suggest that DNA double-strand breaks do not lead to SCE formation.

Cell electroporation is a highly efficient method for introducing restriction endonucleases into cells.

Restriction endonucleases that make either blunt- or cohesive-end DNA double-strand breaks can induce chromosome aberrations. We have used cell electroporation with great success to permeabilize Chinese hamster ovary cells for the introduction of restriction enzymes. The introduction of restriction enzymes by this method resulted in extremely high frequencies (greater than 90%) of aberrant metaphase cells and also a dramatic decrease in cell survival, as measured by subsequent colony formation. Cell electroporation by itself caused no increase in aberrant chromosomes and had only a slight effect on cell survival.

Radiation-induced point mutations, deletions and micronuclei in lacI transgenic mice.

Ionizing radiation induces gene mutations (point mutations, deletions and insertions) as well as chromosome damage in mammalian cells. Although these effects have been studied extensively in cells in culture, until recently it has not been possible to analyze the mutagenic potential of ionizing radiation in vivo, especially at the molecular level. The development of transgenic mutagenesis systems has now made it possible to study the effects of ionizing radiation at both the molecular and chromosomal levels in the same animal. In this report we present preliminary data on the response of Big Blue lacI transgenic mice to ionizing radiation as measured by lacI mutations and micronuclei. C57Bl/6 transgenic mice were irradiated with 137Cs gamma-rays at doses ranging from 0.1 to 14 Gy, and expression times ranging from 2 to 14 days. Dose-related increases in the mutant frequency were observed after irradiations with longer expression times. Mutant plaques were analyzed by restriction enzyme digestion to detect large structural changes in the target sequence. Of 34 gamma-ray-induced mutations analyzed, 4 were large-scale rearrangements. 3 of these rearrangements were deletions within the lacI gene characterized by the presence of short regions of homology at the breakpoint junctions. The fourth rearrangement was a deletion that extended from within the alpha lacZ gene into downstream sequences and that had 43 bp of homology at the junction. These data indicate that the Big Blue lacI transgenic mouse system in sensitive to the types of mutations induced by ionizing radiation. To determine whether the presence of the transgene affects micronucleus induction we compared the response of nontransgenic to hemizygous transgenic B6C3F1 mice and the response of nontransgenic to hemizygous and homozygous transgenic C57Bl/6 mice. The presence or absence of the lacI transgene had no effect on spontaneous micronucleus frequencies for either strain. However, radiation-induced micronucleus frequencies were significantly higher in hemizygous lacI B6C3F1 mice than in nontransgenic litter mates; the converse was true in C57Bl/6 mice. These data suggest that the lacI transgene does not cause chromosome instability as measured by spontaneous micronucleus levels. However, the response of these transgenic mice to a variety of clastogenic agents needs to be investigated before they are integrated into standard in vivo assays for chromosome damage.

Increased chromosomal radiosensitivity of a Chinese hamster ovary cell line that inducibly expresses the Eco RI restriction endonuclease.

We have transfected a Chinese hamster ovary cell line (CHO 6) with a plasmid that inducibly expresses the Eco RI restriction endonuclease gene in the presence of cadmium sulfate (CdSO4). Expression of Eco RI results in DNA double-strand breaks, which can lead to chromosome aberrations. The new line, designated CHO 10, also has a low level of constitutive expression of Eco RI in the absence of CdSO4 without any cytogenetic effect. This suggested that these cells may be efficient at repairing low levels of DNA double-strand breaks. To test this, both cell lines were exposed to ionizing radiation, and aberration yields were analyzed with or without induction of Eco RI. CHO 10 cells showed increased radiosensitivity after G1 irradiation, but after G2 exposure, only doses greater than or equal to 0.4 Gy caused more damage in CHO 10 cells. We conclude that CHO 10 cells can tolerate constitutive expression of Eco RI, but that when the cells are subjected to additional stress, in this case ionizing radiation, they become very sensitive to DNA double-strand breaks.

Transgenic animal models for measuring mutations in vivo.

Transgenic animal models for measuring mutations provide a powerful tool for rapidly assessing tissue-specific mutations following in vivo treatment. These models are based on the insertion into the rodent genome of the Escherichia coli lacI (lac repressor) or lacZ (beta-galactosidase) genes that serve as targets for mutations. Following in vivo treatment of animals, genomic DNA is isolated from various tissues and the target gene is packaged into lambda-phage heads; the lambda-phage are used to infect E. coli in order to produce plaques. Mutations in the target gene are then detected using colorimetric or selective procedures. In this review methods are discussed for producing these transgenic models, the target genes used, gene rescue techniques, sequencing of isolated mutants, and parameters that affect dosing regimens and design of studies. We also present a summary of data published to date with these systems and present our conclusions and proposed directions for future research.

Transgenic animal models for detection of in vivo mutations.

Transgenic rodent models for measuring mutations provide a tool for assessing tissue-specific mutations following in vivo treatment. These systems are based on the insertion into the rodent genome Escherichia coli lacI (lac repressor) or lacZ (beta-galactosidase) genes that serve as targets for mutations. Following in vivo treatment of animals, genomic DNA is isolated from tissues of interest, and the target gene is screened for mutations using either lambda-phage packaging or isolation of the target gene with magnetic affinity capture. In this paper we review the various experimental methods used in the conduct of transgenic mutation assays and discuss critical factors that affect the interpretations of results of these assays.

Chromosome aberration induction in Chinese hamster ovary cells by restriction enzymes with different methylation sensitivity.

The isoschizomer pair MspI and HpaII were used to investigate whether the putative specificity of restriction endonucleases would be maintained when they were introduced into mammalian cells. Although both enzymes recognize the sequence CCGG, HpaII will cut only if the internal cytosine is unmethylated, whereas MspI will cut regardless of the methylation status. Cleavage results in a cohesive-end DNA double-strand break, which can lead to the formation of chromosome aberrations. Since mammalian DNA is heavily methylated, one would expect MspI to be much more effective than HpaII at inducing chromosome aberrations in Chinese hamster ovary cells. In fact, during G1, MspI induced a greater than 90-fold higher number of aberrations than did HpaII. Cell cycle studies indicated that during early S there was a 30-fold increase in HpaII-induced aberrations. This increase may be due to increased accessibility of replicating hypomethylated DNA. Cells that were treated with the demethylating agent 5-aza-2'-deoxycytidine (AzdC) displayed only a moderate increase in HpaII-induced aberrations during G1. This observation, together with the results of restriction enzyme analysis of genomic DNA, indicated that demethylation was incomplete. The effects of AzdC on the induction of aberrations by MspI suggested that AzdC increases chromatin accessibility. Our results were consistent with the expected specificity of MspI and HpaII. Thus, it appears that restriction endonucleases can play a useful role in determining the biological consequences of DNA double-strand breaks.

Spectrum of mutations produced by specific types of restriction enzyme-induced double-strand breaks.

Rejoining of DNA double-strand breaks (DSB) plays a central role in the various processes leading to DNA rearrangements. We have analyzed DNA alterations induced by restriction enzymes that produce DSB with specific types of ends. Restriction enzymes were electroporated into a human lymphoblastoid cell line that stably maintains pHAZE, an EBV-based vector containing the lacZ gene. After allowing time for DSB repair, pHAZE DNA was rescued and screened in Escherichia coli. Mapping and sequence analysis of mutant copies of pHAZE indicated that restriction enzymes induced all classes of alterations except base substitutions (base deletions and insertions, large-scale deletions, inversions, and insertions). The spectra of alterations were distinctive for each enzyme and appear to be the consequence of specific end-modification processes.

p53 deficiency alters the yield and spectrum of radiation-induced lacZ mutants in the brain of transgenic mice.

Exposure to heavy particle radiation in the galacto-cosmic environment poses a significant risk in space exploration and the evaluation of radiation-induced genetic damage in tissues, especially in the central nervous system, is an important consideration in long-term manned space missions. We used a plasmid-based transgenic mouse model system, with the pUR288 lacZ transgene integrated in the genome of every cell of C57Bl/6(lacZ) mice, to evaluate the genetic damage induced by iron particle radiation. In order to examine the importance of genetic background on the radiation sensitivity of individuals, we cross-bred p53 wild-type lacZ transgenic mice with p53 nullizygous mice, producing lacZ transgenic mice that were either hemizygous or nullizygous for the p53 tumor suppressor gene. Animals were exposed to an acute dose of 1 Gy of iron particles and the lacZ mutation frequency (MF) in the brain was measured at time intervals from 1 to 16 weeks post-irradiation. Our results suggest that iron particles induced an increase in lacZ MF (2.4-fold increase in p53+/+ mice, 1.3-fold increase in p53+/- mice and 2.1-fold increase in p53-/- mice) and that this induction is both temporally regulated and p53 genotype dependent. Characterization of mutants based on their restriction patterns showed that the majority of the mutants arising spontaneously are derived from point mutations or small deletions in all three genotypes. Radiation induced alterations in the spectrum of deletion mutants and reorganization of the genome, as evidenced by the selection of mutants containing mouse genomic DNA. These observations are unique in that mutations in brain tissue after particle radiation exposure have never before been reported owing to technical limitations in most other mutation assays.