From DNA damage to chromosome aberrations: joining the break.

Despite many years of experimental studies on radiation-induced chromosomal aberrations, and the recent progress in elucidating the molecular mechanisms of the DNA damage response, the link between DNA double-strand break repair and its expression as microscopically visible chromosomal rearrangements remains, in many ways, obscure. Some long standing controversies have partially been resolved to the satisfaction of most investigators, including the linearity of the dose-response for DNA double-strand break induction, the necessity of pairwise interaction of radiogenic damaged sites in the formation of exchange aberrations, and the importance of proximity between lesions in misrejoining. However, the contribution of different molecular DNA repair mechanisms (e.g., alternative end-joining pathways) and their impact on the kinetics of aberration formation is still unclear, as is the definition of "complex" radiogenic damaged sites - in either the chemical or spatial sense - which ostensibly lead to chromosome rearrangements. These topics have been recently debated by molecular biologists and cytogeneticists, whose opinions are summarized in this paper.

Levels of gamma-H2AX Foci after low-dose-rate irradiation reveal a DNA DSB rejoining defect in cells from human ATM heterozygotes in two at families and in another apparently normal individual.

We have investigated the use of the gamma-H2AX assay, reflecting the presence of DNA double-strand breaks, as a possible means for identifying individuals who are mildly hypersensitive to ionizing radiation, such as some ATM heterozygotes. We compared levels of gamma-H2AX foci after irradiation in cells from six apparently normal individuals as well as from individuals from two separate AT families including the proband, mother, father and three unaffected siblings in each family. After a 1-Gy single acute (high-dose-rate) gamma-ray dose delivered to noncycling contact-inhibited monolayers of cells, clear differences were seen between samples from normal individuals (ATM(+/+)) and probands (ATM(-/-)) at nearly all sampling times after irradiation, but no clear distinctions were seen for cells from normal compared to obligate heterozygotes (ATM(+/-)). In contrast, after 24 h of continuous irradiation at a dose rate of 10 cGy/h, appreciable differences in numbers of foci per cell were observed for cells from individuals for all the known ATM genotypes compared with controls. Four unaffected siblings had mean numbers of foci per cell similar to that for the obligate heterozygotes, whereas the other two had mean values similar to that for normal controls. We determined independently that those siblings with mean numbers of foci per cell in the range of ATM heterozygotes carried the mutant allele, while both siblings with a normal number of foci per cell after irradiation had normal alleles. A more limited set of experiments using lymphoblastoid cell strains in the low-dose-rate assay also revealed distinct differences for normal compared to ATM heterozygotes from the same families and opens the possibility of using peripheral blood lymphocytes as a more suitable material for an assay to detect mild hypersensitivities to radiation among individuals.

gamma-H2AX foci after low-dose-rate irradiation reveal atm haploinsufficiency in mice.

We have investigated the use of the gamma-H2AX assay, reflecting the presence of DNA double-strand breaks (DSBs), as a possible means for identifying individuals who may be intermediate with respect to the extremes of hyper-radiosensitivity phenotypes. In this case, cells were studied from mice that were normal (Atm+/+), heterozygous (Atm+/-), or homozygous recessive (Atm-/-) for a truncating mutation in the Atm gene. After single acute (high-dose-rate) exposures, differences in mean numbers of gamma-H2AX foci per cell between samples from Atm+/+ and Atm-/- mice were clear at nearly all sampling times, but at no sampling time was there a clear distinction for cells from Atm+/+ and Atm+/- mice. In contrast, under conditions of low-dose-rate irradiation at 10 cGy/h, appreciable differences in the levels of gamma-H2AX foci per cell were observed in synchronized G1 cells derived from Atm+/- mice relative to cells from Atm+/+ mice. The levels were intermediate between those for cells from Atm+/+ and Atm-/- mice. After 24 h exposure at this dose rate, measurements in cells from four different mice for each genotype yielded mean frequencies of foci per cell of 1.77 +/- 0.13 (SEM) for Atm+/+ cells, 4.75 +/- 0.20 for the Atm+/- cells, and 11.10 +/- 0.33 for the Atm-/-cells. The distributions of foci per G1 cell were not significantly different from Poisson. To the extent that variations in sensitivity with respect to gamma-H2AX focus formation reflect variations in radiosensitivity for biological effects of concern, such as carcinogenesis, and that similar differences are seen for other genetic DNA DSB processing defects in general, this assay may provide a relatively straightforward means for distinguishing individuals who may be mildly hypersensitive to radiation such as we observed for Atm heterozygous mice.

Role of homologous recombination in the alpha-particle-induced bystander effect for sister chromatid exchanges and chromosomal aberrations.

The bystander effect for sister chromatid exchanges (SCEs) and chromosomal aberrations was examined in hamster cell lines deficient in either DNA-PKcs (V3 cells, deficient in nonhomologous end joining, NHEJ) or RAD51C (irs3 cells, deficient in homologous recombination, HR). Cells synchronized in G0/G1 phase were irradiated with very low fluences of alpha particles such that < 1% of the nuclei were traversed by an alpha particle. Wild-type cells showed a prominent bystander response for SCE induction; an even greater effect was observed in V3 cells. On the other hand, no significant induction of SCE was observed in the irs3 RAD51C-deficient bystander cells irradiated at various stages in the cell cycle. Whereas a marked bystander effect for chromosomal aberrations occurred in V3 cells, the induction of chromosomal aberrations in irs3 bystander cells was minimal and similar to that of wild-type cells. Based on these findings, we hypothesize that HR is essential for the induction of SCE in bystander cells; however, HR is unable to repair the DNA damage induced in NHEJ-deficient bystander cells that leads to either SCE or chromosomal aberrations.

Effects of radiation on the environment: a need to question old paradigms and enhance collaboration among radiation biologists and radiation ecologists.

A historical perspective is given of the current paradigm that does not explicitly protect nonhuman biota from radiation but instead relies on the concept that if dose limits are set to protect humans, then the environment is automatically protected as well. We summarize recent international questioning of this paradigm and briefly present three frameworks for protecting biota that are being considered by the U.S. Department of Energy, the Canadian Nuclear Safety Commission, and the International Commission on Radiological Protection. We point out a controversial component in each of the three frameworks and suggest topics that need additional research. We emphasize that to properly address radiation protection of the environment, we need to understand how effects are integrated across different levels of biological organization. We caution that the proposed use of molecular end points to estimate ecological risks from radioactive contamination is applicable only if we understand the extent of the impact that molecular damage has on individual organisms and populations of exposed biota. To accomplish the latter, enhanced collaborations are required among the traditionally separate disciplines of radiation biology and radiation ecology.

Environmental biodosimetry: a biologically relevant tool for ecological risk assessment and biomonitoring.

Biodosimetry, the estimation of received doses by determining the frequency of radiation-induced chromosome aberrations, is widely applied in humans acutely exposed as a result of accidents or for clinical purposes, but biodosimetric techniques have not been utilized in organisms chronically exposed to radionuclides in contaminated environments. The application of biodosimetry to environmental exposure scenarios could greatly improve the accuracy, and reduce the uncertainties, of ecological risk assessments and biomonitoring studies, because no assumptions are required regarding external exposure rates and the movement of organisms into and out of contaminated areas. Furthermore, unlike residue analyses of environmental media, environmental biodosimetry provides a genetically relevant biomarker of cumulative lifetime exposure. Symmetrical chromosome translocations can impact reproductive success, and could therefore prove to be ecologically relevant as well. We describe our experience in studying aberrations in the yellow-bellied slider turtle as an example of environmental biodosimetry.

Conservation of chromosome 1 in turtles over 66 million years.

Fluorescence in situ hybridization of a whole chromosome 1-specific probe from the yellow-bellied slider turtle (Trachemys scripta) to cells from four other species of turtle ranging from a desert tortoise to a loggerhead sea turtle resulted in specific and exclusive hybridization to chromosome 1 in all five species. Previous observations of conservation in the giemsa banding pattern and chromosome morphology and number among turtles are thus extended to the DNA sequence level, revealing a cytogenetic stability of chromosome 1 in these turtles during the past 66-144 million years. This contrasts with the situation for various hominoid species where, in many instances, extensive chromosomal rearrangements have been reported in one third of that time period. Our probe, which was prepared by microdissecting whole chromosomes from embryonic T. scripta fibroblasts and amplifying using DOP-PCR, is the first report of a whole-chromosome FISH probe for any reptile.

Chromosome translocations in T. scripta: the dose-rate effect and in vivo lymphocyte radiation response.

Using a whole-chromosome FISH painting probe we previously developed for chromosome 1 of the yellow-bellied slider turtle (Trachemys scripta), we investigated the dose-rate effect for radiation-induced symmetrical translocations in T. scripta fibroblasts and lymphocytes. The dose rate below which no reduction in effect per unit dose is observed with further dose protraction was approximately 23 cGy h(-1). We estimated the whole-genome spontaneous background level of complete, apparently simple symmetrical translocations in T. scripta lymphocytes to be approximately 1.20 x 10(-3)/cell projected from aberrations occurring in chromosome 1. Similar spontaneous background levels reported for humans are some 6- to 25-fold higher, ranging from about 6 x 10(-3) to 3.4 x 10(-2) per cell. This relatively low background level for turtles would be a significant advantage for resolution of effects at low doses and dose rates. We also chronically irradiated turtles over a range of doses from 0-8 Gy delivered at approximately 5.5 cGy h(-1) and constructed a lymphocyte dose-response curve for complete, apparently simple symmetrical translocations suitable for use with animals chronically exposed to radiation in contaminated environments. The best-fitting calibration curve (not constrained through the zero dose estimate) was of the form Y(as) = c + aD + bD(2), where Y(as) was the number of apparently simple symmetrical translocations per cell, D was the dose (Gy), a = (0.0058 +/- 0.0009), b = (-0.00033 +/- 0.00011), and c = (0.0015 +/- 0.0013). With additional whole-chromosome probes to improve sensitivity, environmental biodosimetry using stable chromosome translocations could provide a practical and genetically relevant measurement end point for ecological risk assessments and biomonitoring programs.

A deficiency in DNA repair and DNA-PKcs expression in the radiosensitive BALB/c mouse.

We have studied the efficiency of DNA double strand break (DSB) rejoining in primary cells from mouse strains that show large differences in in vivo radiosensitivity and tumor susceptibility. Cells from radiosensitive, cancer-prone BALB/c mice showed inefficient end joining of gamma ray-induced DSBs as compared with cells from all of the other commonly used strains and F1 hybrids of C57BL/6 and BALB/c mice. The BALB/c repair phenotype was accompanied by a significantly reduced expression level of DNA-PKcs protein as well as a lowered DNA-PK activity level as compared with the other strains. In conjunction with published reports, these data suggest that natural genetic variation in nonhomologous end joining processes may have a significant impact on the in vivo radiation response of mice.

Chromosome translocations in turtles: a biomarker in a sentinel animal for ecological dosimetry.

Nonhuman organisms are being exposed to ionizing radiations at radionuclide-contaminated sites around the world. Direct methods are seldom available for measuring biologically relevant doses received by these organisms. Here we extend biological dosimetry techniques, which are much better developed for humans and a few other mammalian species, to a nonmammalian species. Turtles were chosen because a long-lived animal would best serve the need for low-level, chronic exposure conditions. We chose the yellow-bellied slider turtle (Trachemys scripta), which is known to have a maximum life span of at least 22 years. As reported elsewhere, we first isolated an embryonic fibroblast cell line and constructed whole-chromosome-specific DNA libraries for chromosome 1 by microdissection and PCR. A FISH painting probe was prepared and used to establish a dose-response curve for ionizing radiation-induced chromosome interchange aberrations in turtle fibroblasts. This was compared to the dose response for human fibroblasts treated under similar conditions in our laboratory. With respect to induction of chromosome interchange aberrations, human fibroblasts were approximately 1.7 times more sensitive than the T. scripta fibroblasts. To the extent that symmetrical interchanges are persistent over long periods, this approach could eventually provide a measure of the integrated lifetime dose these organisms receive from radionuclides in their environment and give a measure of the extent of relevant genetic damage over that time.

Culture methods for turtle lymphocytes.

Optimization of culture techniques for turtle and other reptilian lymphocytes is essential for facilitating cytogenetic and immunologic research for these animals. We examined a variety of conditions and parameters relevant to turtle lymphocyte culture including: different mitogenic agents, alone and in combination; lymphocyte separation protocols; culture volume; time required to stimulate lymphocytes to mitosis; importance of humidity and gas exchange in culture incubation; suitability of different culture media; effects of varying serum concentrations; ability of interleukin-2 (IL-2) to stimulate lymphocyte growth and prevent apoptosis; and feasibility of inducing premature chromosome condensation. The best conditions of those we studied for obtaining mitotic cells were (1) the combined use of phytohemagglutinin-M form (2%) and lipopolysaccharides (0.55 microg/ml), (2) the use of 5% autologous turtle serum (as opposed to fetal bovine serum), and (3) collection of mitotic cells around 96 hours after mitogenic stimulation. Human, recombinant IL-2 did not increase the fraction of lymphocytes in mitosis over the range of concentrations tested and calyculin A was ineffective at inducing premature chromosome condensation in turtle lymphocytes over the range of concentrations tested. This test regime provides a guideline for determination of appropriate lymphocyte culture conditions in turtles and other reptiles.

Nonrandom degradation of DNA in human leukemic cells during radiation-induced apoptosis.

In many cells, the process of apoptosis is accompanied by endonuclease-mediated double-strand cleavage of DNA between nucleosomes, resulting in the production of discrete fragments of 200 bp or multiples thereof. To address the question of whether this endonuclease attack occurs randomly or nonrandomly along chromosomes, we first constructed chromosome fluorescence in situ hybridization probes from the 200- and 400-bp fragments from gamma-irradiated apoptotic human T cells along with similar-sized probes from randomly sheared DNA of nonirradiated cells. These probes were compared for their binding along normal human metaphase chromosomes after fluorescence in situ hybridization with and without the presence of unlabeled total human blocking DNA. The addition of blocking DNA to the apoptotic probes revealed a nonrandom pattern of hybridization that was not observed for the nonirradiated control probes. The most obvious areas of selective binding occurred around the centromeric and other heterochromatic regions along the chromosome arms, such as the long (q arm) of the Y chromosome. The converse of this experiment was also carried out. DNA probes from heterochromatic and euchromatic regions of the human Y chromosome were hybridized onto slot blots of apoptotic ladder-sized and randomly sheared nonirradiated human T-lymphocyte DNA. The slot blot results showed that for an equal mass of ladder-sized apoptotic DNA and randomly sheared nonirradiated control DNA, the apoptotic DNA sample contains a relatively larger proportion of Y heterochromatin DNA sequences (approximately 2.5-fold). Together, these results indicate that apoptosis-mediated endonuclease attack does not occur randomly in the genome but occurs preferentially in heterochromatin.

Vitamin E (d-alpha-tocopheryl succinate) decreases mitotic accumulation in gamma-irradiated human tumor, but not in normal, cells.

Previous studies have shown that treatment of tumor cells in vitro with d-alpha-tocopheryl succinate (alpha-TS), a most effective form of vitamin E, alone or in combination with X-irradiation, reduced the growth of these cells more than that produced by individual agents. However, it is unknown whether alpha-TS, alone or in combination with gamma-irradiation, would produce similar effects on normal cells. To study this, we have compared the effects of alpha-TS on three human tumor cell lines, HeLa (cervical carcinoma), OVGI (ovarian carcinoma), and A549 (lung carcinoma), with the effects on three human normal fibroblast lines, GM2149, AG1522, and HF19. Results showed that alpha-TS treatment of HeLa cells for 20 hours caused inhibition of growth in a dose-dependent manner, but normal human fibroblasts treated similarly with alpha-TS did not show such an effect. alpha-TS treatment for 20 hours also decreased mitotic accumulation in all three tumor cell lines but did not produce such an effect in any of the normal fibroblasts. As expected, gamma-irradiation with 1 Gy decreased mitotic accumulation in human tumor cells and normal fibroblasts; however, alpha-TS treatment for 24 hours before, during, and after irradiation for the entire experimental period further decreased mitotic accumulation in human tumor cells but not in normal cells. These data suggest that effects of alpha-TS, alone or in combination with gamma-irradiation, are selective for tumor cells. Therefore, existing fear that antioxidants such as vitamin E may protect cancer cells from free radical damage during radiation therapy is not justified.

Molecular and biochemical characterisation of DNA-dependent protein kinase-defective rodent mutant irs-20.

The catalytic subunit of the DNA-dependent protein kinase (DNA-PKcs) is a member of a sub-family of phosphatidylinositol (PI) 3-kinases termed PIK-related kinases. A distinguishing feature of this sub-family is the presence of a conserved C-terminal region downstream of a PI 3-kinase domain. Mutants defective in DNA-PKcs are sensitive to ionising radiation and are unable to carry out V(D)J recombination. Irs-20 is a DNA-PKcs-defective cell line with milder gamma-ray sensitivity than two previously characterised mutants, V-3 and mouse scid cells. Here we show that the DNA-PKcs protein from irs-20 cells can bind to DNA but is unable to function as a protein kinase. To verify the defect in irs-20 cells and provide insight into the function and expression of DNA-PKcs in double-strand break repair and V(D)J recombination we introduced YACs encoding human and mouse DNA-PKcs into defective mutants and achieved complementation of the defective phenotypes. Furthermore, in irs-20 we identified a mutation in DNA-PKcs that causes substitution of a lysine for a glutamic acid in the fourth residue from the C-terminus. This represents a strong candidate for the inactivating mutation and provides supportive evidence that the extreme C-terminal motif is important for protein kinase activity.

Regional gene mapping using mixed radiation hybrids and reverse chromosome painting.

We describe a new approach for low-resolution physical mapping using pooled DNA probe from mixed (non-clonal) populations of human-CHO cell hybrids and reverse chromosome painting. This mapping method is based on a process in which the human chromosome fragments bearing a complementing gene were selectively retained in a large non-clonal population of CHO-human hybrid cells during a series of 12- to 15-Gy gamma irradiations each followed by continuous growth selection. The location of the gene could then be identified by reverse chromosome painting on normal human metaphase spreads using biotinylated DNA from this population of "enriched" hybrid cells. We tested the validity of this method by correctly mapping the complementing human HPRT gene, whose location is well established. We then demonstrated the method's usefulness by mapping the chromosome location of a human gene which complemented the defect responsible for the hypersensitivity to ionizing radiation in CHO irs-20 cells. This method represents an efficient alternative to conventional concordance analysis in somatic cell hybrids where detailed chromosome analysis of numerous hybrid clones is necessary. Using this approach, it is possible to localize a gene for which there is no prior sequence or linkage information to a subchromosomal region, thus facilitating association with known mapping landmarks (e.g. RFLP, YAC or STS contigs) for higher-resolution mapping.

An ionizing radiation-sensitive CHO mutant cell line: irs-20. IV. Genetic complementation, V(D)J recombination and the scid phenotype.

The genetic defect responsible for hypersensitivity of Chinese hamster ovary (CHO) irs-20 cells to ionizing radiation was found to be recessive in nature and could be complemented to produce wild-type radiosensitivity in irs-20/human hybrids. The radiosensitivities of six hybrid clones were determined based on their colony-forming ability under continuous irradiation at 6 cGy/h. A parallel cytogenetic analysis revealed a concordance between the presence or absence of human chromosome 8 and the resistant or sensitive phenotype. Confirming evidence was obtained using human chromosome 8-specific PCR primers. Positive amplification was obtained in hybrids with wild-type radiosensitivity, while no amplification was obtained in sensitive hybrids. Complementation analysis between radiosensitive CHO irs-20 and murine scid cell lines was carried out to determine whether the defects leading to their ionizing radiation hypersensitivity could be corrected by genetic complementation in the hybrids. Complementation did not occur. A transient V(D)J recombination assay after the introduction of the RAG1 and RAG2 genes indicated that the V(D)J recombination ability of the CHO irs-20 cells was about 10% of that for the CHO wild-type cells for signal join formation with an 80% joining fidelity and only 3% of the parental level for coding join formation. These data show that murine scid and irs-20 mutant hamster cells fall into the same complementation group and show similar defects in V(D)J recombination.

Application of 5-bromo-2'deoxyuridine as a label for in situ hybridization in chromosome microdissection and painting, and 3' OH DNA end labeling for apoptosis.

We have utilized 5-bromo-2'deoxyuridine (BrdU) substituted DNA as a probe for a number of applications including, principally, for chromosome painting by fluorescence in situ hybridization (FISH) but also for DNA end-labeling to detect apoptotic cell death and for filter hybridization. Br-dUTP was used as a substitute for biotin or digoxigenin-dUTP in probe labeling techniques, such as random priming, nick translation, end-labeling or PCR. An especially useful application is that it may be incorporated into probe DNA while cells or plasmids in bacteria are growing in the presence of BrdU. This can be particularly advantageous when large quantities of probe are needed, since the cost per mole of digoxigenin-dUTP or biotin-dUTP is nearly 1000 times that of Br-dUTP. Also, if probe is prepared by growth in BrdU, the difference in cost to prepare equal quantities of labeled DNA is more than 10,000 times greater for biotin-dUTP.

The presence of DNA breaks and the formation of chromatid aberrations after incorporation of 125IdUrd may be necessary but are not sufficient to block cell cycle progression in G2 phase.

Cell progression into mitosis and chromatid aberration frequencies were compared in two Chinese hamster ovary (CHO) cell lines after incorporation of 125IdUrd. Asynchronous, exponentially growing populations of CHO K1 and the DNA repair-deficient, radiation-sensitive CHO irs-20 cells were compared after a 10-min exposure to 14.8 kBq/ml 125IdUrd. Essentially no differences were seen for either end point between the cells of the two cell lines. As the cells in S phase at the time of labeling entered the mitotic cell selection window, the number of mitotic cells of each cell line declined to approximately 60% of the respective unlabeled control. Chromosome analysis of the mitotically selected cells indicated an 125I decay-dependent increase in the number of chromatid aberrations in cells of both cell lines. The appearance of aberrations together with the known rates of production and rejoining of DNA double-strand breaks show that cells are able to progress through G2 phase and into mitosis in the presence of such breaks. The data suggest that DNA damage may be necessary, but is not sufficient to cause a radiation-induced blockade of cell progression through G2 phase.

Comparison of gamma-ray-induced chromosome ring and inversion frequencies.

A method was used to detect chromosome inversions as apparent or false sister chromatid exchanges (SCEs) in the first mitosis after gamma irradiation of human G0 cells. Dose-response relationships for small inversions have not been measured and reported previously, but it has been assumed that these are induced with a frequency equal to that of their easily measured asymmetrical counterpart, the interstitial deletion. Our experiments confirm this expectation. The results also demonstrate, as others have suggested, that in protocols where SCEs have been reported in the first postirradiation mitosis after incorporation of BrdU in the previous cell cycle, the X- or gamma-ray treatment of G0- or G1-phase cells produces virtually no true SCEs.

Cell cycle arrest by Colcemid differs in human normal and tumor cells.

In the continuous presence of Colcemid, the mitotic index in cultures of nine human tumor cell lines began to increase immediately upon addition of the drug. For 12 human normal (nontumorigenic) cell lines, the mitotic index did not begin to increase for some 2 to 3 h after the addition of Colcemid. The effect was independent of whether the cells were of fibroblast or epithelial origin and occurred over a 1000-fold range of Colcemid concentrations. No such differential effect was seen with single concentrations of either Taxol or nocodazole, but a similar delayed effect was seen for two concentrations of vinblastine. These observations suggest a fundamental difference between human normal and human tumor cells involving a cell cycle checkpoint in G2, about 1 to 2 h before mitosis.