A DNA-Damage-Induced Cell Cycle Checkpoint in Arabidopsis

Corresponding author: A. B. Britt, 1 Shields Ave., University of California, Davis, CA 95616., abbritt{at}ucdavis.edu (E-mail)

Communicating editor: R. S. HAWLEY

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Although it is well established that plant seeds treated with high doses of gamma radiation arrest development as seedlings, the cause of this arrest is unknown. The uvh1 mutant of Arabidopsis is defective in a homolog of the human repair endonuclease XPF, and uvh1 mutants are sensitive to both the toxic effects of UV and the cytostatic effects of gamma radiation. Here we find that gamma irradiation of uvh1 plants specifically triggers a G₂-phase cell cycle arrest. Mutants, termed suppressor of gamma (sog), that suppress this radiation-induced arrest and proceed through the cell cycle unimpeded were recovered in the uvh1 background; the resulting irradiated plants are genetically unstable. The sog mutations fall into two complementation groups. They are second-site suppressors of the uvh1 mutant's sensitivity to gamma radiation but do not affect the susceptibility of the plant to UV radiation. In addition to rendering the plants resistant to the growth inhibitory effects of gamma radiation, the sog1 mutation affects the proper development of the pollen tetrad, suggesting that SOG1 might also play a role in the regulation of cell cycle progression during meiosis.

THE ability of an organism to repair DNA damage in a timely fashion is essential for the integrity and maintenance of the genome. To this end, signal transduction pathways are involved in sensing DNA damage, pausing the cell division cycle to provide time for repair, inducing repair, and finally releasing the cell cycle from arrest. This arrest allows the cell's repair machinery time prior to S phase or M phase to mend the damage. Cells proficient in DNA repair, but deficient in their ability to arrest in response to damage, typically exhibit high levels of genomic instability, demonstrating the importance of DNA-damage-dependent cell cycle arrest in maintaining the genome (for review see PAULOVICH et al. 1997 ).

Half a century ago plant biologists demonstrated that when corn kernels were treated with very high levels of ionizing radiation (500 krad), the seedlings would germinate at a normal rate and respond appropriately to their environment. The plants would then produce only a limited number of leaves, which contained few cells, although these cells were much larger than those of unirradiated plants (SCHWARTZ and BAY 1956 ). Further studies on the formation of "gamma plantlets" (EVANS 1965 ) demonstrated that irradiated Vicia faba (fava bean) was delayed by several days in the uptake of tritiated thymidine, indicating that entry into S phase was postponed by radiation treatment. FOARD and HABER 1961 ; and HABER et al. 1961 further characterized irradiated plants, this time working in wheat, and showed that although the plants were arrested there was no evidence of senescence and that protein levels appeared to be normal. These early experiments demonstrated that the treatment of seeds with high levels of gamma radiation could arrest progression through the cell cycle but did not arrest transcription, translation, or basic cellular processes.

In this article, we investigate the mechanistic basis of the gamma-plantlet phenomena in Arabidopsis and seek to determine whether plants in fact do have a checkpoint response to DNA damage. We chose to investigate the responses of Arabidopsis to gamma radiation in the gamma radiation hypersensitive uvh1 mutant background. A screen for UV-sensitive mutants of Arabidopsis (HARLOW et al. 1994 ) identified one mutation, termed uvh1, that rendered the plants sensitive to both UV and gamma radiation (Fig 1). Several groups (FIDANTSEF et al. 2000 ; LIU et al. 2000 ) demonstrated that UVH1 is a homolog of the human repair endonuclease XPF, which is involved in nucleotide excision repair, the excision of intermolecular crosslinks, and the repair of single stranded 3' DNA overhangs that are generated during the joining of double-strand breaks (Fig 2; BARDWELL et al. 1994 ). Like the yeast mutants defective in double-strand-break repair, uvh1 makes an ideal genetic background for the study of DNA-damage-induced cell cycle arrest as very low doses of gamma radiation induce the gamma-plantlet phenomenon, eliminating the varied stress responses observed in wild-type at high doses. Our studies suggest that the gamma-plantlet response in uvh1 is due to a DNA-damage-dependent G₂ cell cycle arrest. Furthermore, the isolation of mutants defective in this response demonstrates that the formation of gamma plantlets is not an intrinsic and direct effect of unrepaired damage but is instead imposed upon the cell by a genetically encoded signal transduction pathway. The genetically unstable phenotype of these mutants indicates that this checkpoint is important for the maintenance of genomic integrity.

View larger version (154K): In this window In a new window Download PPT slide   Figure 1. Effects of UV and gamma radiation on wild-type and uvh1-2 plants: wild-type and uvh1-2 plants with no treatment, seeds irradiated at 10 krad, and plants treated with 300 J/m2 UV-C radiation. Red arrows indicate necrotic leaves in UV-C-treated plants.

View larger version (16K): In this window In a new window Download PPT slide   Figure 2. Roles of the UVH1 protein in nucleotide excision repair (A), nonhomologous end joining (B), and the repair of intermolecular DNA crosslinks (C). (A) Nucleotide excision repair: Following the recognition of base damage by the nucleotide repair machinery, a denaturation bubble is formed. The XPF/ERCC1 heterodimer (UVH1 is the Arabidopsis homolog of XPF), in conjunction with RPA, cleaves the duplex/3' single-strand junction of the bubble while XPG cleaves the 3' end of the bubble. The damaged oligonucleotide is then released and a new strand is synthesized across the gap (FRIEDBERG et al. 1995 ). (B) Nonhomologous end joining: In the repair of a double-strand break the 5' end of the DNA is resected to reveal regions of microhomology. The annealing of the complementary strands then displaces one or more single-strand flaps. Again, the XPF/ERCC1 heterodimer, guided by RPA, specifically cleaves 5' of the double-strand to single-strand DNA transition, releasing the DNA flap and allowing ligation of the ends to proceed (BARDWELL et al. 1994 ). The heterodimer plays a similar role in removing flaps during single-strand annealing, a homology-dependent (but RAD51 independent) process involving recombination between tandem repeats (PRADO and AGUILERA 1995 ). (C) Interstrand crosslinks: Although crosslink repair is still poorly understood, mutants defective in the homologs of XPF or ERCC are extremely sensitive to crosslinking agents.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Stocks:
uvh1-1 was isolated in the Columbia background by HARLOW et al. 1994 and was a gift from David Mount. uvh1-2 was isolated in the Landsberg erecta background in our lab (JIANG et al. 1997 ). Arabidopsis plants (Columbia ecotype) containing the cycB1:GUS reporter construct were a gift of Peter Doerner and Adan Colon-Carmona (COLON-CARMONA et al. 1999 ). The gymnous::GUS reporter line was a gift of Yuval Eshed (ESHED et al. 1999 ). The alb1 line (Columbia ecotype) is listed at the Arabidopsis Biological Resource Center (Columbus, OH) as stock CS26.

Growth conditions:
Plants were grown on either Sunshine mix 2 (SunGro, Bellevue, WA) or 1/2x MS (GIBCO, Carlsbad, CA) Phytagel (Sigma, St. Louis) agar, pH 5.2. Plants were grown under cool-white lamps filtered through Mylar at an intensity of 100–150 µmol/m²/sec with a 24-hr day. The temperature was set at 22° and the humidity at 50%.

Mutagenesis and screening:
Two grams of uvh1-2 seeds was treated with 0.225% (w/v) methane sulfonic acid ethyl ester (EMS; Sigma) and placed on a rocking table for 16 hr at room temperature. The EMS-containing solution was decanted and the seeds were washed in 500 ml of water for 4 hr before sowing. Approximately 5000 M₁ plants were harvested in two separate batches. A total of 50,000 M₂ seeds were gamma irradiated (see below) and 12-day-old seedlings were screened for the presence of leaves. The sog1-1 mutation was generated in the first batch of seeds while the sog1-2–sog1-5 mutants were from the second; thus sog1-1 was derived independently from the other mutants.

Radiation treatment:
Prior to radiation, seeds were imbibed in water and placed at 4° for 24 hr. Gamma radiation was carried out in a ¹³⁷Cs reactor at the UC Davis Institute for Toxicology and Environmental Health. UV-irradiated plants were grown for 2 weeks and then treated with 250 kJ/m² of UV-C using a germicidal lamp, placed under orange light for 48 hr, and then returned to normal lighting conditions. UV sensitivity was assayed 72 hr after transfer to the growth chamber.

Histology:
A variety of histological methods were used to analyze tissue in this study.

Paraffin embedding and 4',6-diamidino-2-phenylindole staining: To analyze the mitotic index, seeds were gamma irradiated, grown in 1/2x MS liquid culture, pH 5.2, and 3 days later the plants were fixed in FAA and then dehydrated in an ethanol series followed by paraffin embedding using a Leica (Wetzler, Germany) tissue fixation TP1020 robot (50% ethanol 1 hr, 60% ethanol 1 hr, 75% ethanol 1 hr, 80% ethanol 1 hr, 95% ethanol 0.1% eosin 6 hr, 2x 100% ethanol 1 hr, 2x Histoclear 1 hr, Histoclear saturated with paraffin 4 hr, paraffin 5 hr, and paraffin 6 hr). Embedded tissue was cut to an 8-µm width on a Microm (Walldorf, Germany) HM-340E microtome. Wax sections were placed on microscope slides and then deparaffinized and dehydrated (2 x 10 min Histoclear, 2 x 2 min 100% ethanol, 1 min 95% ethanol, 1 min 90% ethanol, 1 min 80% ethanol, 1 min 60% ethanol, 1 min 30% ethanol, and 2 min water). The tissue was mounted in 1 µg/ml 4',6-diamidino-2-phenylindole (DAPI; Molecular Probes, Eugene, OR), 50% glycerol, 0.1 M Tris, pH 9.2, 1 mg/ml phenylenediamine. The stained material was analyzed on a Zeiss (Jena, Germany) Axiphot microscope.

Staining for GUS: Plants that were to be analyzed for GUS were grown in liquid containing 1/2x MS salts at a pH of 5.2. If embryos were to be harvested the seed coat was first removed. The tissue was cleared and stained according to standard techniques (JEFFERSON et al. 1987 ; RODRIGUES-POUSADA et al. 1993 ). The stained tissue was then placed on slides with 50% glycerol and analyzed on a Zeiss Axiophot microscope.

Tetrad analysis: To visualize pollen tetrads and meiocytes, individual buds were dissected and the anthers removed. The anthers were then pierced with a fine needle, allowing the contents to flow out. The tissue was prepared according to ROSS et al. 1996 with slight modification; the buds were digested at 37° for 30 min in 0.3% pectolyase, 0.3% cytohelicase, and 1% cellulase (Sigma) and then mounted in DAPI as above. The tissue was analyzed on a Zeiss Axiophot microscope and images were captured using a Zeiss MC100 camera with Kodak 160T film (Eastman Kodak, Cherry Hill, NJ).

Cosegregation of tetrad and gamma radiation phenotypes:
To determine whether the abnormal tetrad phenotype cosegregated with the gamma-resistant phenotype we analyzed a population derived from a backcross of uvh1-2 sog1-3 to uvh1-2. F₃ families from this cross were scored for their sensitivity to both UV and gamma radiation. Individuals from 10 gamma-resistant and 10 gamma-sensitive families were then analyzed for the presence of abnormal tetrads as detailed above.

Loss of heterozygosity tests:
All uvh1-2 sog1 mutants were backcrossed to uvh1-2, and the resulting F₃ families were screened for plants that were sensitive to UV but formed leaves in the presence of gamma radiation. Seeds from these families were sown and the resulting plants were crossed to alb1 mutant plants (these plants carry, heterozygously, a mutation that produces an albino phenotype when homozygous). F₃ families were screened for lines that were homozygous for either uvh1-2 or uvh1-2 and sog1 and were heterozygous for the alb1 mutation.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Irradiated uvh1 plants do not progress through mitosis:
While it has long been established that the irradiation of plant seeds induces an arrest of seedling development, it is not known whether this developmental arrest is due to the death of meristematic cells or to an arrest of cell division. To distinguish between these two possibilities, we analyzed DAPI-stained longitudinal sections of irradiated Arabidopsis seedlings for evidence of either cell death or cell cycle arrest. Wild-type and uvh1-2 seeds were either untreated or irradiated at 10 krad and the shoot apical meristem of DAPI-stained sectioned plants was analyzed. In no case did we find any direct evidence for cell death. There were no obviously enucleate cells, no apoptotic bodies (WANG et al. 1996 ), and no evidence of dead crushed cells (data not shown). However, when we compared the shoot apical meristem of arrested seedlings to that of nonarrested seedlings, we found a substantial difference in the ability of cells to progress through mitosis. To quantify the percentage of mitotically active cells in the different treatments, we counted the number of cells in which a mitotic figure was present (Fig 3). We found that both unirradiated wild-type and unirradiated uvh1 plants had approximately the same fraction of cells passing through mitosis, with mitotic indices of 4%. While irradiated wild-type plants were not different from unirradiated plants, none of the 20 gamma-irradiated uvh1 sections had a mitotic figure, indicating that the shoot apical meristems of these plants were arrested at some point outside of M phase.

View larger version (35K): In this window In a new window Download PPT slide   Figure 3. Mitotic index of irradiated and unirradiated wild-type and uvh1-2 seedlings at the shoot apical meristem. Seeds of wild type and uvh1-2 were gamma irradiated at 10 krad. Twelve days following gamma radiation the tissue was fixed and sectioned, and the nuclei were stained with DAPI. Mitotic figures were present in 3.9 ± 2.4% (n = 20) and 3.7 ± 3.0% (n = 19) of unirradiated and irradiated wild-type shoot apical meristem cells, respectively. In uvh1-2 plants, 3.7 ± 2.2% (n = 25) of the unirradiated shoot apical meristem cells contained mitotic figures, while none of the irradiated uvh1-2 cells were in mitosis (n = 20; n = number of plants screened). Each Ler apical meristem contains 100 cells at germination; a minimum of 50 cells were scored from each plant.

Meristematic cells of uvh1-2 enter G₂ but do not proceed to M:
To pinpoint in which phase of the cell cycle the gamma plantlets were arrested, we crossed a late G₂-/early M-phase specific reporter construct in the Columbia background into the uvh1-2 line. This construct consists of the Arabidopsis cycB1 promoter (formerly designated cyc1aAt; see FERREIRA et al. 1994 ) driving a protein fusion of the cycB1 mitotic destruction box and the ß-glucuronidase gene. ß-Glucuronidase reacts with X-GlcU to create a blue semi-cell-autonomous precipitate. The cycB1 promoter drives expression of ß-glucuronidase in late G₂/early M while the mitotic destruction box targets the protein for degradation as the cell begins to exit M phase (COLON-CARMONA et al. 1999 ).

As the plant embryo becomes fully mature, the cells cease division and prepare for long-term quiescence. Across the plant kingdom there is variation as to whether these cells completely arrest in G₁ or arrest in both G₁ and G₂ (DELTOUR 1985 ). On the basis of previous work we expected that the bulk of the Arabidopsis embryo's cells would be arrested in G₁ (LAUFS et al. 1998 ). To obtain a qualitative estimate of the frequencies of cells in G₂, we looked for expression of the cycB1::GUS construct in the mature embryo. We removed the seed coat of mature cycB1::GUS seeds, imbibed the embryos in X-GlcU, and could find no GUS staining cells in the Arabidopsis embryo. To ensure that the product of the GUS gene was stable in the seeds, we also looked at the same reporter driven by the embryo-specific Gymnos promoter (ESHED et al. 1999 ) and found ample staining (data not shown). This observation is consistent with previously published observations (LAUFS et al. 1998 ) suggesting that the Arabidopsis embryo is a naturally "synchronized" G₁ population of cells.

We next analyzed the effects of radiation on cell cycle progression in uvh1-2 cycB1::GUS vs. UVH1 cycB1::GUS plants. Seeds, having been soaked in water for 24 hr, were either untreated or gamma irradiated at 10 krad. Three days after irradiation the plants were analyzed for expression of the reporter. We found no evidence for a substantial G₁ arrest; at 24 hr after irradiation (48 hr after imbibition) no cells had entered G₂, while at 36 hr after irradiation the cells of both wild-type and uvh1 root tips had begun to progress into G₂. However, at later time points the apical and root tip meristem cells of irradiated uvh1-2 plants accumulated in G₂ while irradiated wild-type or unirradiated wild-type and uvh1-2 meristem cells continued through the cell cycle (Fig 4). These data, taken together with the mitotic index investigations, suggest that irradiation of uvh1-2 seeds triggers a G₂ arrest. In contrast, undamaged or repair-proficient cells continue to cycle normally.

View larger version (93K): In this window In a new window Download PPT slide   Figure 4. Shoot apical meristem of irradiated uvh1-2 roots arrest in G2 phase. Seeds were imbibed in water for 24 hr and then gamma irradiated at 10 krad. Seedlings were harvested either 36 hr or 5 days after gamma irradiation and then stained for the presence of the GUS protein. (A) Both wild-type and uvh1-2 plants showed an accumulation of GUS and, therefore, an entry into G2 36 hr after imbibition. While wild type never had more than a few cells staining for GUS at any one time, a large proportion of the uvh1-2 root apical cells stained for the presence of GUS. (B) A x20 magnification of 5-day-old uvh1-2 roots, either unirradiated or treated with 10 krad of gamma radiation. Two representative roots are displayed at each dose.

sog mutants in the uvh1-2 background lack the gamma-plantlet response:
Radiation-induced G₂ arrest of uvh1-2 seedlings could be due to one of two causes. DNA damage induced by gamma radiation might directly impede progression through the cell cycle by physically blocking the cellular machinery. For example, damaged bases might block the advancement of RNA polymerase and therefore prevent the expression of genes required to positively regulate cell cycle progression. Alternatively, an extrinsic signaling network may sense even very small amounts of DNA-damage and enforce an arrest of cell division, providing time for the cell to repair this damage before the chromosomes segregate into separate daughter cells and the opportunity for repair is lost. Isolation of mutants that are defective in DNA-damage-induced arrest provides evidence that this arrest is due to an extrinsic signaling mechanism. A signal transduction network that senses DNA damage and arrests cell division is well documented in the yeast and mammalian literature (for example, see WEINERT and HARTWELL 1988 ; ZHOU and ELLEDGE 2000 ).

To differentiate between the two possible causes of arrest we screened for mutants in the uvh1-2 background that did not arrest cell division following gamma irradiation. Since uvh1-2 is defective in DNA repair, we could then screen for suppressors of arrest using a very low dose of gamma radiation without generating excessive damage throughout the cell. Because Arabidopsis is diploid, we reasoned that some loss of chromosome arms resulting from segregation of unrepaired chromatids would be tolerated. Interpolating from the yeast and mammalian data on the rate of induction of double-strand breaks (DSBs) by gamma radiation, a 10-krad dose should induce 70 DSBs per haploid Arabidopsis genome (WARD 1990 ), and it is unclear what fraction of double-strand breaks (or other gamma-induced lesions) are repaired via a UVH1-dependent pathway. Approximately 5000 uvh1-2 seeds were EMS mutagenized. As we assumed that most of the mutations would be recessive, we allowed the M₁ plants to self-fertilize and then screened the M₂ generation. Of 50,000 M₂ plants, 52 were able to form leaves following -radiation. The M₃ progeny of these 52 plants were retested for their ability to form leaves following gamma radiation. Six families were found to reproducibly develop true leaves after radiation. These mutants were termed sog (suppressor of gamma radiation; Fig 5).

View larger version (157K): In this window In a new window Download PPT slide   Figure 5. sog1 uvh1 mutants are gamma resistant but UV hypersensitive. uvh1-2 and uvh1-2 sog1 plants from untreated seeds, seeds gamma irradiated at 10 krad, or seedlings treated with 300 J/m2 UV-C radiation. Red arrows indicate necrotic leaves in UV-C-treated plants.

To determine whether the sog mutations were dominant or recessive we backcrossed the sog uvh1-2 mutants to uvh1-2, selfed the resulting F₁, and scored the segregating F₂ generation for resistance to gamma radiation. We found that the F₂ progeny segregated in a 3:1 ratio [3 gamma sensitive (arrested) to 1 resistant], indicating that the sog alleles are recessive (Table 1). We then determined how many genes the six radiation-resistant families represented by performing a complementation test. Individuals from each family were crossed in a pairwise manner, the F₁ was selfed, and the F₂ plants were scored for radiation resistance. Crosses between mutants affecting the same gene should produce 100% gamma-sensitive progeny, while mutants in different genes, when crossed, will produce an F₂ generation that segregates for wild-type vs. mutant phenotypes [in an expected ratio of 9 sensitive to 7 resistant (sog/sog) plants, if unlinked]. At 10 krad, occasional "pseudosensitive" individuals (plantlets that produced no true leaves at the time of scoring) are observed in even homozygous uvh1 sog mutants, and chi-square tests were performed to distinguish between the presence of these rare "escapers" and the higher frequency of segregation expected from noncomplementation. This test revealed that the mutants fell into a single complementation group (Table 2). The sog1 mutants were generated from two independently mutagenized populations (and thus might represent only two alleles). sog1-6 has severe fertility defects and therefore has proved recalcitrant to genetic analysis; for this reason this article focuses on the sog1-1 through 1-5 alleles.

It is possible that the sog1 mutation did not represent a second-site suppressor of the gamma-sensitive phenotype of uvh1, but instead was merely a reversion of the uvh1 mutation. To test this possibility we utilized the dual role of UVH1 in DNA repair; as well as being hypersensitive to gamma radiation uvh1 is also hypersensitive to the effects of UV light. The UVH1 gene encodes a homolog of the human repair endonuclease XPF and the ensuing sensitivity to UV light derives from the persistence of toxic pyrimidine dimers (FIDANTSEF et al. 2000 ; LIU et al. 2000 ). These UV-induced lesions can impair the progression of DNA and RNA polymerase and therefore lead to cell death. UV toxicity is observed as the browning and necrosis of exposed tissue. To determine whether the UV-sensitive phenotype of uvh1 mutation was still present, the uvh1-2 sog1 mutants were grown for 2 weeks on soil and then irradiated with UV-C light. The plants were subsequently shielded from photoreactivating blue light for 3 days and then placed in the growth chamber. All of the mutants exhibited the typical necrotic browning of the leaves that is the hallmark of UV hypersensitivity (Fig 5). Therefore, the sog mutation does not represent a complete reversion of uvh1.

The possibility remained that the sog1 mutation represented a partial reversion of the gamma-radiation-induced damage repair capacity of uvh1-2. To test this possibility we analyzed the segregation ratio of uvh1-2 sog1 plants crossed to wild type. If the sog1 phenotype were due to reversion of the gamma sensitivity conferred by the uvh1-2 allele, one would expect all F₂ progeny to be gamma resistant. Rather, four independently generated F₂ families segregated 13:3 for resistance to gamma radiation (² = 0.35 2.4), confirming the genetic separability of the uvh1-2 and sog1 mutations.

sog1 mutants produce leaves at wild-type rates:
To understand the extent of sog1 suppression of the gamma- plantlet phenomena we compared the rate of leaf production in irradiated wild-type, uvh1, and uvh1 sog1 plants. It is possible that irradiated plants with only a partial checkpoint defect would have a growth rate intermediate between wild-type and uvh1. On the other hand, if the sog1 mutation was pleiotropic and was involved in regulating the development of new organs, it is possible that the sog1 mutants would show growth rates in excess of wild type. To investigate these two possibilities we irradiated wild-type, uvh1-2, and uvh1-2 sog1 seeds and measured the rate of true leaf development over 2.5 weeks. On average, the first pair of leaves from irradiated wild-type plants was visible to the naked eye 11 days after irradiation while on average uvh1-2 plants displayed their first pair of leaves 20 days after irradiation. The five different sog1 alleles showed a range of developmental rates with the first pair of leaves emerging from 10 to 14 days. The growth curves of the sog1 mutants demonstrated that the mutant plants did not grow faster than irradiated wild-type plants; rather, the sog1 uvh1 seedlings grew at the same or at a slightly slower rate than did wild-type plants (Fig 6). Although the rate of leaf production in the uvh1-2 sog1 lines was essentially unchanged from wild-type, we noticed that at the higher dose of 20 krad the leaves were misshapen, with a ragged appearance, perhaps indicating the death of certain cell files. This phenotype worsened in uvh1 sog1 plants as we further increased the dosage of radiation, whereas in wild-type plants the leaves became progressively smaller with increasing doses of radiation but did not appear to be "ragged" or missing files of cells.

View larger version (39K): In this window In a new window Download PPT slide   Figure 6. Rates of leaf production in wild-type, uvh1-2, and uvh1-2 sog1 plants. Seeds from wild-type, uvh1-2, and uvh1-2 sog1 were irradiated at 20 krad and sown on soil, and the rate of leaf production was followed. Diamonds denote wild-type, squares are uvh1-2, and circles are the uvh1-2 sog1 lines. (A) uvh1-2 sog1-1, (B) uvh1-2 sog1-2, (C) uvh1-2 sog1-3, (D) uvh1-2 sog1-4, and (E) uvh1-2 sog1-5.

The fact that leaves developed in all of the irradiated sog1 uvh1-2 double mutants suggests that the G₂ arrest that was seen in the single mutant has been abolished. To further test this hypothesis we crossed the cycB1::GUS reporter construct into lines uvh1-2 sog1-1 and uvh1-2 sog1-2. We irradiated seeds homozygous for uvh1-2, sog1-1, and cycB1::GUS, grew the plants for 3 days on agar plates, and then stained for GUS expression (Fig 7). In contrast to the irradiated uvh1-2 cycB1::GUS, which showed extensive GUS staining, these plants showed an accumulation of GUS similar to that of irradiated wild-type plants. As expected, approximately half a dozen cells expressed the construct, but there was no large-scale accumulation of GUS-expressing cells. These findings confirmed our hypothesis that the sog1 mutation prevents a DNA-damage-induced G₂ arrest.

View larger version (87K): In this window In a new window Download PPT slide   Figure 7. uvh1-2 sog1 plants no longer arrest in G2. uvh1-2 cycB1::GUS plants were crossed to uvh1-2 sog1-1 and uvh1-2 sog1-2 plants to generate uvh1-2 sog1 cycB1::GUS plants. Seeds from uvh1-2 sog1-1 cycB1::GUS plants were imbibed in water for 24 hr and then were either left untreated or gamma irradiated at 10 krad. Five days after gamma irradiation seedlings were stained for the presence of GUS. Two representative uvh1-2 sog1-1 cycB1::GUS roots are displayed at each dose. There was no apparent difference in the staining pattern between irradiated and unirradiated plants. The uvh1-2 sog1-2 plants did not stain differently from the uvh1-2 sog1-1 plants.

uvh1-2 sog1 plants show increased levels of genomic instability:
Although our data suggest that the irradiated sog1 mutants develop leaves because of the lack of a G₂ arrest, it is also possible that the cells were no longer arresting because there was a second-site mutation that partially rescued the repair deficiency of the uvh1 mutation. In this case the mutation might specifically increase the ability of the plant to repair ionizing radiation-induced damage while not promoting the excision repair of UV-induced dimers. To differentiate between an upregulation of repair and the loss of a DNA-damage-dependent cell cycle checkpoint, we analyzed the impact of the sog1 mutation on genomic integrity by both indirect and direct methods. We scored plants heterozygous for an albino mutation for sectors representing loss of the wild-type Albino allele. We also used fluorescence microscopy to look at DAPI-stained pollen meiocytes for evidence of chromosome loss and/or improper meioses.

The sog mutation increases the frequency of sectoring in irradiated plants: To assay genomic instability we crossed plants containing the albino1 mutation (alb1) to uvh1-2 sog1 plants. The alb1 mutation is recessive and homozygous albino plants are white and die at the seedling stage. The ALB1 gene lies close to the tip of chromosome 1, at a map position of 12 cM (VAN-DER-VEEN 1973 ). In a heterozygote, loss of the wild-type ALB1 allele in a cell lineage results in a white or light green sector (Fig 8A). We hypothesized that if wild-type SOG1 is required for maintaining genomic stability, plants that were homozygous for uvh1-2 and sog1-5 and heterozygous for alb1 should exhibit a rate of sectoring higher than that of alb/ALB or uvh1-2/uvh1-2 alb1/ALB1 plants. This loss of heterozygosity might be due to breakage of the ALB1 chromosome between the locus and centromere, with subsequent loss of the acentric fragment, or it might be due to mitotic crossing over or the formation of smaller deletions that include the wild-type locus. The instabilities observed at ALB1 would presumably be typical of instabilities occurring throughout the genome. In fact, it is possible that some of the sectors scored might be due to loss of other loci required for chlorophyl production, although sectors were extremely rare in irradiated plants that were not already heterozygous for alb1.

View larger version (46K): In this window In a new window Download PPT slide   Figure 8. uvh1-2 sog1-5 plants show increased genomic instability. Wild-type, uvh1, and uvh1-2 sog 1-5 seeds carrying a heterozygous albino mutation were gamma irradiated, grown, and screened for somatic sectoring due to the loss of the wild-type Albina (Alb1) allele. (A) Loss of the Alb1 allele generates a white sector. (B) A total of 0.92 ± 0.2% of plants wild-type for SOG1 and UVH1 but heterozygous for albino were sectored, 1.3 ± 0.9% of the alb/Alb uvh1-2/uvh1-2 were sectored, and 6.4 ± 1.5% of the alb/Alb uvh1-2/uvh1-2 sog1-5/sog1-5 plants contained sectors.

After crossing uvh1-2 sog1-5 homozygous plants to ALB1/alb1 plants, we screened for F₃ families that were either uvh1-2/uvh1-2 SOG1/SOG1 ALB1/alb1 or uvh1-2/uvh1-2 sog1-5/sog1-5 ALB1/alb1. These two families and the original alb1/Alb1 family allowed us to analyze the relative rates of genomic instability in wild-type, uvh1-2, and uvh1-2 sog1-5 backgrounds. F₃ families of each genotype were either untreated or irradiated at 10 krad and leaves were analyzed for sectors over the vegetative life of the plants. In no instance did we find sectors in any of the unirradiated plants (n > 1000).

In plants derived from seeds treated with 10 krad of gamma radiation we found two different types of sectors, white and pale green, presumably representing sectors that span all the inner layers of the leaf or those that occurred in only one of the layers (the plant epidermis is colorless). Plants that carried the albino mutation alone or both albino and uvh1 had a low frequency of sectoring (Fig 8B). In contrast, uvh1-2 sog1-5 plants carrying the albino mutation had a rate of sectoring approximately fivefold greater than that of uvh1 (Fig 8B). This high level of chromosomal instability in the alb1 heterozygous, sog1 uvh1-2 mutant plants is consistent with the notion that the cells of sog1 plants are progressing through the cell cycle in spite of the persistence of damage.

The sog1 mutation causes meiotic defects: In the course of investigating the sog1 mutation it became apparent that several of the uvh1-2 sog1 mutant lines were significantly less fertile than the uvh1-2 progenitor, even in the absence of gamma radiation. These observations led us to believe that there may be problems in proper development of the pollen meiocytes. To better understand the nature of the fertility defect, we analyzed DAPI-stained sog1-3 uvh1-2 pollen precursor cells as well as the resulting pollen tetrads. We first analyzed cells in the process of division. After a thorough screening of 50 meiotic anaphase cells (a small subset of the total number of meiotic cells) we found only one dicentric bridge in the uvh1-2 sog1-3 pollen and none in uvh1-2 or wild-type pollen. Because these bridges are both rare and transient, we could not make any conclusions as to the frequency of such lesions. We then analyzed the products of pollen meiosis in the least fertile line, uvh1-2 sog1-3, and found that this line displayed a high frequency of abnormal meioses (Fig 9). Anthers were gathered from individual buds of unirradiated wild-type, uvh1-1, and uvh1-1 sog1-3 plants. The anthers were dissected and stained with DAPI. One hundred percent of wild-type tetrads contained four nuclei (n = 153). In contrast, 91.5% (n = 206) of uvh1-2 tetrads contained four nuclei and only 65% (n = 170) of uvh1-1 sog1-3 tetrads contained four nuclei. To confirm that this meiotic defect is due to the sog1 mutation, we analyzed the cosegregation of the abnormal meiosis with the gamma-resistant phenotype. Ten of 10 plants that were gamma resistant also bore abnormal tetrads. In contrast, 0 of 10 gamma-sensitive plants had abnormal tetrads, demonstrating that the two phenotypes are tightly linked. These findings indicate that either the sog1 mutation may interfere with the proper progression of meiosis or the accumulation of defects throughout somatic development can lead to a high frequency of improper meioses.

View larger version (77K): In this window In a new window Download PPT slide   Figure 9. uvh1-2 sog1-3 plants are defective in meiosis. uvh1-2 and uvh1-2 sog1-3 pollen meiocytes (tetrads) were dissected from immature anthers. The tetrads were stained with DAPI to determine the number of intact nuclei. Additionally, the cells were visualized under bright field to determine both the shape and the number of the spores. In the photograph of uvh1-2, two nuclei (in one focal plane) were false colored dark blue while the other two (in another focal plane) were colored light blue. The uvh1-2 tetrads conformed both to the typical number of nuclei and to cell shape. The uvh1-2 sog1-3 had a variable number of nuclei and the cells were misshapen. The lightly fluorescing material in the uvh1-2 sog1-3 tetrad may be either organellar DNA or chromatid fragments.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Our analysis of the effects of gamma radiation on DNA repair-defective plants demonstrates that the uvh1-2 gamma-plantlet response is caused by a DNA damage-dependent G₂-phase arrest. Our ability to identify mutants that eliminate the checkpoint response shows that the arrest is not a direct effect of DNA damage. Rather, the gamma-plantlet response is triggered by a signal-transduction mechanism that monitors the cell for damage and then arrests the cell cycle. Checkpoint mutants, defective in this response, show high levels of genomic instability and meiotic defects.

The gamma plantlet response is the result of a G₂-specific cell cycle arrest:
We found that when we gamma-irradiated repair-defective uvh1-2 seeds the plants germinated normally, but the meristematic cells arrested in the G₂ phase of the cell cycle. Our findings are similar to those of earlier investigators (FOARD and HABER 1961 , FOARD and HABER 1970 ; EVANS 1965 ), who observed that unrepaired DNA damage triggers an arrest of plant development. While the early work indicated that gamma radiation interfered with cell division, it remained unclear as to where and why the plants arrested. Our data suggest that the gamma-plantlet response is induced in our repair-defective line by lesions that require expression of UVH1 for their repair or processing. Unfortunately, the broad range of lesions induced by ionizing radiation, combined with the broad substrate specificity of the XPF/ERCC1 repair endonuclease, makes it difficult to positively identify the nature of the arrest-inducing lesion(s); these might be double-strand breaks, intermolecular crosslinks, or even modified bases that require this endonuclease for their repair.

Our investigations of the sog1 mutation demonstrate that the G₂ arrest is imposed on the cell by a DNA-damage-sensing signal transduction mechanism. Although studies on the effects of radiation on plants indicate that the cell cycle is arrested in response to radiation, it was previously unclear whether plants, like humans and yeast, harbor a classic checkpoint response to DNA damage. In fact, some data suggest that plants may lack some of the DNA-damage-dependent checkpoint responses that exist in other organisms (COUTEAU et al. 1999 ). The lack of any noticeable meiotic arrest in Arabidopsis meiotic recombinase-deficient dmc1 mutants highlights the difference between the response of plants and animals to DNA damage, as discussed below.

DMC1 encodes a meiosis-specific homolog of the RAD51 recombinase and is required for strand exchange during recombination (MASSON and WEST 2001 ). Yeast and mouse dmc1 mutants are unable to repair these meiotic DSBs through homologous recombination and arrest in a checkpoint-dependent manner without completing meiosis (LYDALL et al. 1996 ; PITTMAN et al. 1998 ; YOSHIDA et al. 1998 ). In Arabidopsis, the DMC1 gene is also required for meiotic recombination (COUTEAU et al. 1999 ). However, rather than undergoing arrest, Arabidopsis dmc1 cells proceed through anaphase without establishing chiasma and the chromosomes migrate randomly to the poles (COUTEAU et al. 1999 ). Further evidence of a lack of a meiotic checkpoint comes from the large collection of maize meiotic mutants (for examples, see LIU et al. 1993 ; CHAN and CANDE 1998 ). The different mutants are defective in various stages of meiosis but none exhibit a checkpoint response (Z. CANDE, personal communication).

In both animals and yeast, unprocessed meiotic double-strand breaks are sensed by checkpoint proteins and division is arrested (CHIN et al. 1999 ). Why does Arabidopsis seem to arrest in response to DNA damage generated by gamma radiation but not to breaks generated by other means? One possibility is that plants have a robust DSB-dependent mitotic checkpoint, as seen in our irradiated uvh1 plants, but lack a meiotic DSB checkpoint. While this would explain the fact that irradiated uvh1 plants arrest while dmc1 plants do not arrest, it is also possible that Arabidopsis entirely lacks a DSB-dependent G₂ arrest and that the G₂-phase checkpoint observed here is a response to lesions other than double-strand breaks.

The damage induced by ionizing radiation is complex. In addition to strand breaks, ionizing radiation induces oxidative damage at sugars and bases, often in the form of multiply damaged sites (MDSs; WARD 1998 ). MDSs that include closely spaced lesions on opposing strands are similar to interstrand crosslinks in that they are recalcitrant to excision repair due to the lack of an undamaged template strand, can block replication of both strands of DNA, and may be processed into double-strand breaks during S phase. Crosslinks and lesions like them block both transcription and DNA replication and can trigger a checkpoint response (FRIEDBERG et al. 1995 ). The repair of a crosslink is a multistep process involving both the creation of a double-strand break 5' of the crosslink and the resolution of the crosslinked strands (DE SILVA et al. 2000 ). The processing of interstrand crosslinks in mammals requires both XPF (of which UVH1 is a homolog) and ERCC1 (of which UVR7 is a homolog). Both uvh1 and uvr7 mutants exhibit a G₂ checkpoint response to gamma radiation (JIANG et al. 1997 ; HEFNER et al. 2003 ). Similarly, Saccharomyces cerevisiae rad1 (the yeast homolog of UVH1) mutants arrest in G₂/M in response to the crosslinking agent cisplatin (GROSSMANN et al. 1999 ).

The argument for a DNA-damage-dependent but not DSB-dependent checkpoint is buttressed by a collection of gamma-radiation-hypersensitive mutants recently identified in our laboratory (HEFNER et al. 2003 ). Only those lines that were sensitive to both a crosslinking agent (mitomycin C) and gamma radiation arrested in G₂. In contrast, the ionizing radiation-sensitive 1 (irs1) mutant (defective in the DSB-specific ligase LIG4; E. HEFNER, J. FRIESNER and A. B. BRITT, unpublished data) is sensitive to gamma radiation but not to MMC, and does not exhibit a G₂ arrest response to ionizing radiation. Thus two separate and independent experiments, the dmc1 reverse genetic study and our analysis of gamma-sensitive lines, suggest that Arabidopsis may have a less stringent response than other organisms to double-strand breaks.

Genomic instability of sog mutants:
In addition to an abrogation of the DNA-damage-dependent checkpoint response, Arabidopsis uvh1-2 sog1 double mutants also displayed a number of hallmarks of chromosomal instability. Irradiated uvh1-2 sog1 plants showed a dramatic increase in loss of the ALB1 locus, although in the absence of radiation none of the plants were sectored. The uvh1-2 sog1 plants also showed gross defects in meiotic segregation in the absence of radiation. While the loss of the albino1 gene is likely due to a failure to correctly repair damage induced in irradiated mitotic cells, an explanation of the meiotic defect is less obvious. It is possible that either the aberrant tetrads arose as an accumulation of genetic abnormalities during mitotic development that prevents the proper progression of meiosis or the sog1 mutation causes a defect in the regulation of meiosis per se. Although we cannot rule out the possibility that accumulated chromosomal aberrations would lead to improper meioses, we did look for evidence of meiotic bridges in the uvh1-2 sog1 mutant and found only 1 among 50 anaphase figures, while we found none in uvh1 or wild-type plants. We also failed to observe any sectoring in the unirradiated ALB1/alb1 sog1 uvh1 mutants, suggesting that the instability-inducing lesions are rarely generated spontaneously. In contrast, the consistent meiotic defect in the distribution of chromosomes combined with the aberration in "tetrad" shape and cell number suggest a cytokinetic defect resembling the Arabidopsis tardy asynchronous meiosis (tam; MAGNARD et al. 2001 ) or ask1 (YANG et al. 1999 ) mutants. The tam mutants behave similarly to the uvh1 sog1 mutants in that they commonly form from one to four spores per tetrad, while occasional "polyads" have more than four spores. Additionally, tam mutants are no longer meiotically synchronized, each spore proceeding through meiosis independently of its neighbor, suggesting a defect in cell cycle regulation. ask1 mutant pollen commonly has from one to six spores per polyad and a varying amount of DNA per cell, and the polyads resemble the abnormal shapes observed in uvh1-2 sog1. On the basis of its homology to the yeast SKP1 gene, ASK1 is thought to be involved in controlling the transition of the meiotic spindle from metaphase to anaphase (YANG et al. 1999 ). The similarity of the uvh1-2 sog1 phenotype to these pollen mutants and the proposed role of the ask1 and tam mutants in controlling the transition between the stages of meiosis suggest that sog1 might also be a regulator of progression through meiosis.

Conclusion:
Our investigation into the radiation-induced arrest response of uvh1 plants and the isolation of mutants defective in this response demonstrate that plants, like animals, harbor a classic checkpoint response to DNA damage. Mutants deficient in this checkpoint response exhibited both gamma-induced somatic instability and consistent meiotic defects, indicating that this checkpoint plays an important role in the maintenance of genomic integrity. These findings, coupled with the fact that the Arabidopsis genome contains many of the same checkpoint-related genes that are known in humans, demonstrate that plants can react as animals do to DNA damage. We suspect, however, that this G₂ arrest is not a response to double-strand breaks, but instead to another gamma-induced lesion.

	FOOTNOTES

¹ Present address: Department of Biology, Washington University, St. Louis, MO 63130.

	ACKNOWLEDGMENTS

We thank Eli Hefner, Joanna Friesner, Kevin Culligan, and James Hatteroth for helpful discussions. The gift of the cycB1::GUS construct by Peter Doerner and Adan Colon-Carmona was extremely helpful. This work was done in partial fulfillment of a doctorate in philosophy in the Plant Biology Graduate Group at the University of California, Davis. This work was funded by National Science Foundation grant MCB-9983142.

Manuscript received November 5, 2002; Accepted for publication January 16, 2003.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

BARDWELL, A. J., L. BARDWELL, A. E. TOMKINSON, and E. C. FRIEDBERG, 1994  Specific cleavage of model recombination and repair intermediates by the yeast Rad1-Rad10 DNA endonuclease. Science 265:2082-2085.[Abstract/Free Full Text]

CHAN, A. and W. Z. CANDE, 1998  Maize meiotic spindles assemble around chromatin and do not require paired chromosomes. J. Cell Sci. 111(Pt. 23):3507-3515.[Abstract]

CHIN, L., S. E. ARTANDI, Q. SHEN, A. TAM, and S. L. LEE et al., 1999  p53 deficiency rescues the adverse effects of telomere loss and cooperates with telomere dysfunction to accelerate carcinogenesis. Cell 97:527-538.[Medline]

COLON-CARMONA, A., R. YOU, T. HAIMOVITCH-GAL, and P. DOERNER, 1999  Technical advance: spatio-temporal analysis of mitotic activity with a labile cyclin-GUS fusion protein. Plant J. 20:503-508.[Medline]

COUTEAU, F., F. BELZILE, C. HORLOW, O. GRANDJEAN, and D. VEZON et al., 1999  Random chromosome segregation without meiotic arrest in both male and female meiocytes of a dmc1 mutant of Arabidopsis. Plant Cell 11:1623-1634.[Abstract/Free Full Text]

DELTOUR, R., 1985  Nuclear activation during early germination of the higher plant embryo. J. Cell Sci. 75:43-83.[Medline]

DE SILVA, I. U., P. J. MCHUGH, P. H. CLINGEN, and J. A. HARTLEY, 2000  Defining the roles of nucleotide excision repair and recombination in the repair of DNA interstrand cross-links in mammalian cells. Mol. Cell. Biol. 20:7980-7990.[Abstract/Free Full Text]

ESHED, Y., S. F. BAUM, and J. L. BOWMAN, 1999  Distinct mechanisms promote polarity establishment in carpels of Arabidopsis. Cell 99:199-209.[Medline]

EVANS, H. J., 1965  Effects of radiation on meristematic cells. Radiat. Bot. 5:171-182.

FERREIRA, P. C., A. S. HEMERLY, J. D. ENGLER, M. VAN MONTAGU, and G. ENGLER et al., 1994  Developmental expression of the Arabidopsis cyclin gene cyc1At. Plant Cell 6:1763-1774.[Abstract/Free Full Text]

FIDANTSEF, A. L., D. L. MITCHELL, and A. B. BRITT, 2000  The Arabidopsis UVH1 gene is a homolog of the yeast repair endonuclease RAD1. Plant Physiol. 124:579-586.[Abstract/Free Full Text]

FOARD, D. E. and A. H. HABER, 1961  Anatomic studies of gamma-irradiated wheat growing without cell division. Am. J. Bot. 48:438-446.

FOARD, D. E. and A. H. HABER, 1970  Physiologically normal senescence in seedlings grown without cell division after massive gamma-irradiation of seeds. Radiat. Res. 42:372-380.[Medline]

FRIEDBERG, E. C., G. C. WALKER and W. SIEDE, 1995 DNA Repair and Mutagenesis. ASM Press, Washington, DC.

GROSSMANN, K. F., J. C. BROWN, and R. E. MOSES, 1999  Cisplatin DNA cross-links do not inhibit S-phase and cause only a G2/M arrest in Saccharomyces cerevisiae. Mutat. Res. 434:29-39.[Medline]

HABER, A. H., W. L. CARRIER, and D. E. FOARD, 1961  Metabolic studies of gamma-irradiated wheat growing without cell division. Am. J. Bot. 48:351-429.

HARLOW, G. R., M. E. JENKINS, T. S. PITTALWALA, and D. W. MOUNT, 1994  Isolation of uvh1, an Arabidopsis mutant hypersensitive to ultraviolet light and ionizing radiation. Plant Cell 6:227-235.[Abstract]

HEFNER, E., S. B. PREUSS, and A. B. BRITT, 2003  Arabidopsis mutants sensitive to gamma radiation include the homologue of the human repair gene ERCC1. J. Exp. Bot. 54:669-680.[Abstract/Free Full Text]

JEFFERSON, R. A., T. A. KAVANAGH, and M. W. BEVAN, 1987  GUS fusions: beta-glucuronidase as a sensitive and versatile gene fusion marker in higher plants. EMBO J. 6:3901-3907.[Medline]

JIANG, C. Z., C. N. YEN, K. CRONIN, D. MITCHELL, and A. B. BRITT, 1997  UV- and gamma-radiation sensitive mutants of Arabidopsis thaliana. Genetics 147:1401-1409.[Abstract]

LAUFS, P., O. GRANDJEAN, C. JONAK, K. KIEU, and J. TRAAS, 1998  Cellular parameters of the shoot apical meristem in Arabidopsis. Plant Cell 10:1375-1390.[Abstract/Free Full Text]

LIU, Q., I. GOLUBOVSKAYA, and W. Z. CANDE, 1993  Abnormal cytoskeletal and chromosome distribution in po, ms4 and ms6; mutant alleles of polymitotic that disrupt the cell cycle progression from meiosis to mitosis in maize. J. Cell Sci. 106(Pt. 4):1169-1178.[Abstract]

LIU, Z., G. S. HOSSAIN, M. A. ISLAS-OSUNA, D. L. MITCHELL, and D. W. MOUNT, 2000  Repair of UV damage in plants by nucleotide excision repair: Arabidopsis UVH1 DNA repair gene is a homolog of Saccharomyces cerevisiae Rad1. Plant J. 21:519-528.[Medline]

LYDALL, D., Y. NIKOLSKY, D. K. BISHOP, and T. WEINERT, 1996  A meiotic recombination checkpoint controlled by mitotic checkpoint genes. Nature 383:840-843.[Medline]

MAGNARD, J. L., M. YANG, Y. C. CHEN, M. LEARY, and S. MCCORMICK, 2001  The Arabidopsis gene tardy asynchronous meiosis is required for the normal pace and synchrony of cell division during male meiosis. Plant Physiol. 127:1157-1166.[Abstract/Free Full Text]

MASSON, J. Y. and S. C. WEST, 2001  The Rad51 and Dmc1 recombinases: a non-identical twin relationship. Trends Biochem. Sci. 26:131-136.[Medline]

PAULOVICH, A. G., D. P. TOCZYSKI, and L. H. HARTWELL, 1997  When checkpoints fail. Cell 88:315-321.[Medline]

PITTMAN, D. L., J. COBB, K. J. SCHIMENTI, L. A. WILSON, and D. M. COOPER et al., 1998  Meiotic prophase arrest with failure of chromosome synapsis in mice deficient for Dmc1, a germline-specific RecA homolog. Mol. Cell 1:697-705.[Medline]

PRADO, F. and A. AGUILERA, 1995  Role of reciprocal exchange, one-ended invasion crossover and single-strand annealing on inverted and direct repeat recombination in yeast: different requirements for the RAD1, RAD10 and RAD52 genes. Genetics 139:109-123.[Abstract]

RODRIGUES-POUSADA, R. A., R. DE RYCKE, A. DEDONDER, W. VAN CAENEGHEM, and G. ENGLER et al., 1993  The Arabidopsis 1-aminocyclopropane-1-carboxylate synthase gene 1 is expressed during early development. Plant Cell 5:897-911.[Abstract/Free Full Text]

ROSS, K. J., P. FRANSZ, and G. H. JONES, 1996  A light microscopic atlas of meiosis in Arabidopsis thaliana. Chromosome Res. 4:507-516.[Medline]

SCHWARTZ, D. and C. BAY, 1956  Further studies on the reversal in the seedling height dose curve at very high levels of ionizing radiation. Am. Nat. 90:323-327.

VAN-DER-VEEN, J. L., 1973  Double reduction in tetraploid Arabidopsis thaliana, studied by means of a chlorophyll mutant with a distinct simplex phenotype. Arabidopsis Inf. Serv. 10:11-12.

WANG, H., J. LI, R. M BOSTOCK, and D. G. GILCHRIST, 1996  Apoptosis: a functional paradigm for programmed plant cell death induced by a host-selective phytotoxin and invoked during development. Plant Cell 8:375-391.[Abstract]

WARD, J. F., 1990  The yield of DNA double-strand breaks produced intracellularly by ionizing radiation: a review. Int. J. Radiat. Biol. 57:1141-1150.[Medline]

WARD, J. F., 1998 Nature of lesions formed by ionizing radiation, pp. 65–84 in DNA Damage and Repair, Vol. 2, edited by J. A. NICKOLOFF and M. F. HOEKSTRA. Humana Press, Totowa, NJ.

WEINERT, T. A. and L. H. HARTWELL, 1988  The RAD9 gene controls the cell cycle response to DNA damage in Saccharomyces cerevisiae. Science 241:317-322.[Abstract/Free Full Text]

YANG, M., Y. HU, M. LODHI, W. R. MCCOMBIE, and H. MA, 1999  The Arabidopsis SKP1-LIKE1 gene is essential for male meiosis and may control homologue separation. Proc. Natl. Acad. Sci. USA 96:11416-11421.[Abstract/Free Full Text]

YOSHIDA, K., G. KONDOH, Y. MATSUDA, T. HABU, and Y. NISHIMUNE et al., 1998  The mouse RecA-like gene Dmc1 is required for homologous chromosome synapsis during meiosis. Mol. Cell 1:707-718.[Medline]

ZHOU, B. B. and S. J. ELLEDGE, 2000  The DNA damage response: putting checkpoints in perspective. Nature 408:433-439.[Medline]

This article has been cited by other articles:

				C-A Whittle and M. Johnston Moving forward in determining the causes of mutations: the features of plants that make them suitable for assessing the impact of environmental factors and cell age J. Exp. Bot., June 1, 2006; 57(9): 1847 - 1855. [Abstract] [Full Text] [PDF]

				H. Shaked, N. Avivi-Ragolsky, and A. A. Levy Involvement of the Arabidopsis SWI2/SNF2 Chromatin Remodeling Gene Family in DNA Damage Response and Recombination Genetics, June 1, 2006; 173(2): 985 - 994. [Abstract] [Full Text] [PDF]

				K. Abe, K. Osakabe, S. Nakayama, M. Endo, A. Tagiri, S. Todoriki, H. Ichikawa, and S. Toki Arabidopsis RAD51C Gene Is Important for Homologous Recombination in Meiosis and Mitosis Plant Physiology, October 1, 2005; 139(2): 896 - 908. [Abstract] [Full Text] [PDF]

				K. Zhang, L. Diederich, and P. C.L. John The Cytokinin Requirement for Cell Division in Cultured Nicotiana plumbaginifolia Cells Can Be Satisfied by Yeast Cdc25 Protein Tyrosine Phosphatase. Implications for Mechanisms of Cytokinin Response and Plant Development Plant Physiology, January 1, 2005; 137(1): 308 - 316. [Abstract] [Full Text] [PDF]

				K. Culligan, A. Tissier, and A. Britt ATR Regulates a G2-Phase Cell-Cycle Checkpoint in Arabidopsis thaliana PLANT CELL, May 1, 2004; 16(5): 1091 - 1104. [Abstract] [Full Text] [PDF]

				T. V. Reddy, J. Kaur, B. Agashe, V. Sundaresan, and I. Siddiqi The DUET gene is necessary for chromosome organization and progression during male meiosis in Arabidopsis and encodes a PHD finger protein Development, December 15, 2003; 130(24): 5975 - 5987. [Abstract] [Full Text] [PDF]