Logo of geneticsGeneticsCurrent IssueInformation for AuthorsEditorial BoardSubscribeSubmit a Manuscript
Genetics. 2007 Jul; 176(3): 1469–1482.
PMCID: PMC1931559

Functional Analysis of Maize RAD51 in Meiosis and Double-Strand Break Repair


In Saccharomyces cerevisiae, Rad51p plays a central role in homologous recombination and the repair of double-strand breaks (DSBs). Double mutants of the two Zea mays L. (maize) rad51 homologs are viable and develop well under normal conditions, but are male sterile and have substantially reduced seed set. Light microscopic analyses of male meiosis in these plants reveal reduced homologous pairing, synapsis of nonhomologous chromosomes, reduced bivalents at diakinesis, numerous chromosome breaks at anaphase I, and that >33% of quartets carry cells that either lack an organized nucleolus or have two nucleoli. This indicates that RAD51 is required for efficient chromosome pairing and its absence results in nonhomologous pairing and synapsis. These phenotypes differ from those of an Arabidopsis rad51 mutant that exhibits completely disrupted chromosome pairing and synapsis during meiosis. Unexpectedly, surviving female gametes produced by maize rad51 double mutants are euploid and exhibit near-normal rates of meiotic crossovers. The finding that maize rad51 double mutant embryos are extremely susceptible to radiation-induced DSBs demonstrates a conserved role for RAD51 in the repair of mitotic DSBs in plants, vertebrates, and yeast.

MEMBERS of the RecA-like protein family are required for homologous recombination and DNA repair (Roca and Cox 1990). Rad51p, a recA homolog of yeast, polymerizes in vitro on double-strand DNA to form a helical filament that is nearly identical, at least at low resolution, to the three-dimensional (3-D) structure formed by RecA (Ogawa et al. 1993). Rad51p can also bind single-stranded DNA and mediate DNA strand exchange in vitro (Sung 1994; Benson et al. 1998; New et al. 1998; Shinohara and Ogawa 1998). In vivo Rad51p forms complexes with Rad52p, Rad55p, Rad57p, and Dmc1p (Bishop 1994; Hays et al. 1995). When cultured in sporulation medium, yeast rad51 mutant diploid cells can produce 10–20% spores relative to wild type, implying defects in meiosis. Rarely formed asci contain dyads and triads; very few contain tetrads (Shinohara et al. 1992). On the basis of mutant analysis, it appears that Rad51p and Dmc1p have overlapping functions in meiotic crossovers. Rates of meiotic crossovers decrease substantially as compared to wild type only when both proteins are eliminated (Shinohara et al. 1997; Grishchuk and Kohli 2003).

RAD51 homologs have been cloned from several plant species (Terasawa et al. 1995; Doutriaux et al. 1998; Franklin et al. 1999). Maize has two closely related genes, Zmrad51a and Zmrad51b (Franklin et al. 1999), which encode RecA-related proteins (Figure 1). To be consistent with nomenclature in other organisms, Zmrad51a and Zmrad51b are hereby renamed Zmrad51A1 and Zmrad51A2, respectively. (In this article these two genes will be referred to as rad51A1 and rad51A2 except in those instances where the Zm prefix is required for clarity.) It has been suggested that in addition to its role in mediating DNA transfer, the maize RAD51 protein may be involved in homology search during meiosis (Franklin et al. 1999; Pawlowski et al. 2003).

Figure 1.
A phylogenetic tree of selected RecA-like proteins of maize, Arabidopsis, and yeast. This average distance tree was generated using the clustalx1.83 and viewed with njplot. Statistical support for this phylogenetic tree was assessed by bootstrap analysis ...

A reverse genetic approach was used to functionally characterize the two maize rad51 genes. Analysis of rad51A1 and rad51A2 single and double mutants established that although rad51A1 and rad51A2 have at least partially redundant functions, RAD51 function is required for efficient chromosome pairing, homologous synapsis, proper chromosome segregation, and double-strand break (DSB) repair in meiosis. Although maize plants that lack RAD51 function survive and even flower, dry kernels of rad51 double mutants are highly sensitive to radiation treatment. rad51 double mutants are male sterile and exhibit a greatly reduced production of viable female gametes. Surprisingly, the rate of meiotic crossovers in these surviving female gametes does not differ significantly from that of wild type.


Genetic stocks and crosses:

The female parent of cross 1 is a Mu stock derived from those of Robertson (1978, 1980, 1981) via crosses to the hybrid B77 × B79 and Q66 × Q67. The inbred line B73 is fixed for functional alleles of rad51A1 (Rad51A1-B73) and rad51A2 (Rad51A2-B73); mutant alleles were maintained via crosses to B73. A material transfer agreement governs the distribution of the trait utility system for corn (TUSC) Mu insertion alleles of rad51A1 and rad51A2; inquiries should be directed to R. B. Meeley.

  • Cross 1: Mu Rad51A2/Rad51A2 × rad51A2-98E7/Rad51A2.
  • Cross 2: Mu rad51A2-98E7/Rad51A2 × B73.

Characterization of Mu transposons and insertion sites in rad51 TUSC alleles:

Members of the Mu transposon family share ~200 bp conserved terminal repeats (TIRs). PCR was performed on each TUSC Mu-insertion allele using a series of gene-specific primers in combination with a primer (Mu-TIR) located in the highly conserved Mu TIRs. The resulting PCR products were purified and sequenced. These PCR products contain Mu TIR sequences terminal to the Mu-TIR primer annealing site. Comparing the TIR sequences from each of the Mu-insertion alleles to the left and right TIRs of each of the previously defined classes of Mu transposons established which class of Mu transposon was responsible for each TUSC allele. The resulting PCR products also contained rad51 sequences between the positions of the Mu insertion site and the rad51-specific primer. Comparison of these PCR sequences to the rad51 genomic sequences established the position of the Mu insertion in each TUSC allele.

Screen for deletion derivatives of rad51A2-98E7:

Approximately 1000 kernels were planted in a 96-well format and tissues from up to four seedlings were bulked for DNA extraction (Dietrich et al. 2002) in the initial screen. Individual seedlings from a pool that contained a deletion candidate were transplanted to large pots for further growth. Leaf samples from each plant were recollected and DNA isolated. The PCR screen was repeated to identify the individual plant that carried a putative deletion derivative.

The PCR screen was performed using the Mu-TIR primer in conjunction with the rad51A2-specific primer 13219. In the absence of a structural rearrangement at the rad51A2 locus, this primer pair will produce a 395-bp PCR product. In contrast, PCR amplification of DNA from a plant that carries a deletion of DNA between the annealing sites of primers 13219 and Mu-TIR would be expected to yield a smaller product.

The PCR products derived from deletion candidates were purified using the Qiaquick gel extraction kit (QIAGEN) and sequenced using the Mu-TIR primer. As a control, PCR products from rad51A2-98E7 were purified and sequenced at the same time.

Molecular characterization of derivative alleles:

Any of three possible types of sequence rearrangements can generate derivative alleles from rad51A2-98E7 that yield smaller PCR products when amplifying with a rad51A2-specific primer 13219 in combination with a Mu-TIR primer: an intragenic Mu transposition, an insertion of an additional Mu transposon, or a deletion of rad51A2 sequences adjacent to the original Mu insertion. To differentiate an intragenic transposition from an adjacent deletion, a second PCR reaction was performed using the Mu-TIR primer in combination with a rad51A2-specific primer RB-98E7 whose binding site is on the opposite side of the Mu insertion as compared to the primer 13219 used in the initial screen. If a derivative allele arose via an adjacent deletion, the PCR product from the other side should be the same size between rad51A2-98E7 and the derivative allele. In contrast, if an allele arose via an intragenic transposition, the PCR product from the other side of the Mu insertion would be larger than the PCR product obtained from rad51A2-98E7. To differentiate between the insertion of an additional Mu transposon and an adjacent deletion, a primer rbdel was designed to anneal to the region thought to have been deleted. If a derivative allele arose via an adjacent deletion, this primer pair would not be expected to yield a PCR product. In contrast, if a derivative allele arose via the insertion of a second Mu transposon, amplification with Mu-TIR and rbdel would be expected to yield a rad51A2-specific PCR product.

Deletion derivative of rad51A1-54F11:

While genotyping a family segregating for rad51A1-54F11 using primer pair Mu-TIR and 11981, one plant (00-2608-1) with a smaller-size PCR product was identified. PCR-based experiments similar to those conducted on rad51A2-98E7 derivatives established that this allele (designated rad51A1-54F11d1) is a deletion derivative of rad51A1-54F11 (Figures 2 and and33).

Figure 2.
Structures of the rad51A1 and rad51A2 genes. Exons are represented by boxes. The splice sites of rad51A1 and rad51A2 are conserved except that exon II of rad51A2 is split into two exons (II and III) in rad51A1. Mu insertions in TUSC alleles are indicated ...
Figure 3.
Structures of rad51A1-54F11 and rad51A2-98E7 deletion derivatives. Shaded boxes represent exons; Mu insertions are designated by triangles. The 5′ end of each deletion is located at a Mu insertion site. Deleted regions are designated by dashed ...


The primer pair Mu-TIR and 11981 amplify Mu insertion alleles rad51A1-54F11 and rad51A1-54F11d1. Because the wild-type Rad51A1 allele in our stock is from inbred line B73, one indel polymorphism (IDP) marker (raidp31) can anneal to Rad51A1-B73 but not to rad51A1-54F11 and rad51A1-54F11d1, and PCR amplification with the primer pair rad51a-244 and raidp31 will be positive for Rad51A1-B73/rad51A1-54F11 and Rad51A1-B73/rad51A1-54F11d1 plants and negative for rad51A1-54F11/rad51A1-54F11 and rad51A1-54F11d1/rad51A1-54F11d1 plants (Figure 2). In addition, the primer radel is located on the region deleted in rad51A1-54F11d1 relative to rad51A1-54F11; PCR amplification with the primer pair rad51a-244 and radel will be positive for Rad51A1-B73/rad51A1-54F11d1 plants and negative for rad51A1-54F11d1/rad51A1-54F11d1 plants.

The primer pair Mu-TIR and RB-98E7 can amplify the Mu insertion allele rad51A2-98E7d4. The primer rbdel is located on the region deleted in rad51A2-98E7d4 relative to rad51A2-98E7. Hence, PCR amplification with the primer pair RB-98E7 and rbdel will be positive for Rad51A2/rad51A2-98E7d4 plants and negative for rad51A2-98E7d4/rad51A2-98E7d4 plants.

Irradiation of dry kernels:

Irradiation was conducted at the Linear Accelerator Facility (LAF) at the Iowa State University Meat Laboratory. Kernels arising from the self-pollination of plants with the genotype (Rad51A1/rad51A1; rad51A2/rad51A2) were irradiated using a Circe IIIR electron beam irradiator (Linac Technologies, SA, Orsay, France) with an energy level of 10 MeV and a dose rate of 29 kGy/min. Dry kernels were placed on a conveyor cart in a single layer. Alanine dosimeters were placed on the top and bottom surfaces of these samples. Samples were exposed to the electron beam using a single-sided pass. Actual absorbed doses were measured by placing the alanine dosimeters into the e-scan electron paramagnetic resonance instrument (Bruker BioSpin). The actual dose kernels absorbed in this treatment was 0.37 kGy delivered by an electron beam that had a dose rate of 28.8 kGy/min. Treated kernels were immediately germinated in 96-well flats. DNA was isolated from germinated seedlings (Dietrich et al. 2002) and their genotypes were determined via PCR.

Preparation and observation of microsporocytes:

Microsporocyte samples were fixed in a freshly prepared 3:1 mixture of 95% ethanol and acetic acid (v:v) and then stored at −20°. Anthers were smeared and stained with acetocarmine following the procedure of Dempsey (1994).

Fixation and preparation of meiocytes for fluorescence in situ hybridization and immunostaining:

Anthers were prepared following the procedure of Golubovskaya et al. (2002). Briefly, anthers from developing tassels were staged with the acetocarmine squash technique. Anthers from the same floret and from those in close proximity, and thus close in developmental stage, were fixed at room temperature in 4 ml of 4% formaldehyde in buffer A [15 mm PIPES, NaOH, pH 6.8, 80 mm KCl, 20 mm NaCl, 0.5 mm EGTA, 2 mm EDTA, 0.15 mm spermine tetra HCL, 0.05 mm spermidine, 1 mm dithiothreitol, 0.32 m sorbitol (Bass et al. 1997)] for 45 min in a gently shaking 10-ml petri dish. They were then washed three times, 30 min each in fresh buffer A, and stored at 4° in the buffer. Fixed anthers were cut open at the tip to release the meiocytes into 100–200 μl of buffer A. Ten microliters of meiocytes suspended in the buffer A were then transferred by micropipette onto a glass cover slip (22 × 22 mm) followed by the immediate addition of 5 μl of activated acrylamide stock. Acrylamide was activated by the addition of 5 μl of 20% ammonium persulfate and 5 μl of 20% sodium sulfite to 100 μl of a 15% (29:1 acrylamide:bis acrylamide) gel stock in 1× buffer A. Slides were quickly stirred and a second cover slip was placed on top for 45 min, then removed with a razor blade, leaving a thin pad of acrylamide with embedded meiocytes on the slide.


A 27-bp oligonucleotide, 5′-CCTAAAGTAGTGGATTGGGCATGTTCG-3′, labeled with either Cy5 or FITC, was obtained from Genset (Paris) and was used to detect the CentC sequence that resides near maize centromeres (Ananiev et al. 1998). Oligonucleotides complementary to the telomere repeat (5′-{CCCTAAA}4-3′) and labeled with either Cy5 or FITC (Genset) were used to detect maize telomeres (Bass et al. 1997). A 5S rDNA probe was made by PCR. Approximately 1–10 ng of a plasmid containing 5S rDNA sequence from maize (Zimmer et al. 1988) was added to a standard PCR reaction mix: 2 μl 10× buffer with 15 mm MgCl2 from Perkin-Elmer, 2 μl of forward and reverse primers, 2 μl 1 mm dATP, dGTP, dCTP, a mixture of dTTP and dUTP-FITC or dUTP-Cy5, 2 units AmpliTaq (Perkin-Elmer), and water to 20 μl. For each labeling reaction, a 20 μl unlabeled control reaction was performed, and an aliquot of equal molar volume was run side by side with the labeling reaction in a 4% gel. Incorporation of fluorescent label could sometimes be seen on the transilluminator, but empirical determination of probe effectiveness by fluorescence in situ hybridization (FISH) was required for each batch of probe.

FISH and indirect immunofluorescence:

FISH was performed following the procedure of Golubovskaya et al. (2002). Briefly, newly polymerized acrylamide pads were washed with 1× PBS, followed by four washes with a prehybridization buffer (50% deionized formamide, and 2× SSC). Cover slips were placed on a slide, then 50 μl of probe in prehybridization buffer was added, then sealed under a second cover slip using rubber cement, and incubated at 37° for 30–45 min. Slides were denaturated on a PCR block at 96° for 6 min followed by overnight incubation at 30°. Slides were then washed for 30 min sequentially with 1× PBS and 1× SSC (three times), 1× PBS and 0.1% Tween-20 (four times), 1× PBS (three times), 1× TBS (one time). Slides were then stained with 10 μg/ml DAPI (4′,6-diamidino-2-phenylindole, dihydrochloride) in 1× TBS for 30 min at room temperature. Excess DAPI was removed by washing with 1× TBS (three times) for a total of 30 min. Slides were then mounted in 1,4 diazabicyclo-[2,2,2] octane (DABCO) and sealed with clear fingernail polish and stored at 20°. The procedure for staining RAD51 foci using the anti-HsRAD51 rDNA antibody was as described previously (Franklin et al. 1999).

Three dimensional deconvolution light microscopy and image generation:

Images were acquired on a Delta Vision (Applied Precision) imaging station: an Olympus IX70 inverted microscope with 100×, 1.35 NA oil-immersion lens and a photometric (Roper Scientific) CCD. All images were taken with a Z step size of 0.2 μm, saved as 3-D stacks, and subjected to constrain iterative deconvolution. Three-dimensional data analysis and two-dimensional image creation were performed using the DeltaVision/Soft WoRx software package (Applied Precision) on a Silicon Graphics work station. Two-dimensional images were converted to TIFF and opened in Photoshop on a Macintosh computer. Photoshop was used to manipulate false colors and to convert colors from RGB to CMYK for printing.


Structures of the rad51A1 and rad51A2 genes:

The maize genome contains two homologs of the yeast rad51 gene: Zmrad51A1 and Zmrad51A2, which map to chromosomes 7 and 3, respectively (Franklin et al. 1999). These two genes are most closely related to AtRAD51 and only distantly related to AtRAD51 paralogs (AtRAD51B, AtRAD51C, AtRAD51D, AtXRCC2, and AtXRCC3; Bleuyard et al. 2005; Figure 1). Full-length cDNA sequences of each gene have been reported previously (GenBank accession nos. AF079428 and AF079429). The structures of the Zmrad51A1 and Zmrad51A2 genes were constructed on the basis of alignments of these apparently full-length cDNA sequences with assembled genome survey sequences (GSSs) supplemented by additional sequencing of a Zmrad51A1 genomic clone (Figure 2).

Analysis of rad51 copy number in the maize genome:

The coding regions of rad51A1 and rad51A2 are 84 and 90% identical at the nucleotide and amino acid levels, respectively. DNA gel blot analysis was conducted to determine the number of rad51 genes in the maize genome. When the final wash was performed at 65°, the rad51A1 (supplemental Figure 1, lane 1, at http://www.genetics.org/supplemental/) and rad51A2 (supplemental Figure 1, lane 3) cDNA probes are gene specific. Under less stringent wash conditions (55°) the rad51A1 cDNA probe detects fragments derived from both the rad51A1 and rad51A2 genes (supplemental Figure 1, lane 2), but no other strongly hybridizing fragments. Consistent with DNA gel blot results, the MAGI maize genomic (http://magi.plantgenomics.iastate.edu/) and EST databases do not contain another gene that is highly similar to rad51A1 or rad51A2 (using a cut-off value of 75% nucleotide identity). In summary, these results indicate that there are only two rad51 genes in the maize genome.

Isolation of Mu-insertion alleles of rad51A1 and rad51A2:

Three Mu-insertion alleles of rad51A1 (rad51A1-42A8, rad51A1-54F11, and rad51A1-family6) and two of rad51A2 (rad51A2-63H3 and rad51A2-98E7) were recovered via the reverse genetic screen TUSC (Bensen et al. 1995).

Of the five TUSC alleles, only one, rad51A1-54F11, contained a Mu transposon insertion in a coding region (Figure 2). The other alleles had insertions in introns or 5′ of the apparent transcription start sites. Because such alleles might confer residual gene function and might also be subject to “Mu suppression” (Barkan and Martienssen 1991; Martienssen and Baron 1994; Cui et al. 2003;), they were not considered ideal reagents for the study of RAD51 functions.

Deletion derivatives of rad51A2-98E7:

Although germinal excision events are recovered only rarely from Mu-insertion alleles (Brown et al. 1989; Levy et al. 1989; Schnable et al. 1989), ~1% of gametes carry deletions adjacent to a given Mu insertion (Taylor and Walbot 1985; Levy and Walbot 1991; Das and Martienssen 1995). A PCR-based screen modified from that described by Das and Martienssen (1995) was used to identify rare derivatives of rad51 TUSC alleles that contained deletions adjacent to the Mu insertions.

The rad51A2-98E7 allele was selected to screen for deletions because its Mu insertion is located near an intron–exon junction. Initially, rad51A2-98E7 was crossed into a Mu active line (cross 1, materials and methods). Progeny of this cross that carried rad51A2-98E7 were crossed by the inbred line B73 (cross 2); ~1000 of the resulting progeny were germinated and analyzed via PCR using the Mu-TIR primer in conjunction with the gene-specific primer 13219. Seedlings that yielded smaller PCR products than those obtained from rad51A2-98E7 were grown to maturity and backcrossed to B73. The progeny from these crosses were genotyped via PCR; four heritable derivative alleles were recovered (Figures 2 and and33).

Experiments conducted as described in materials and methods indicated that all four of the heritable derivative alleles arose via adjacent deletions (data not shown). Sequence analysis of the PCR products associated with these alleles established that the sizes of the deletions range from 69 to 179 bp (Figure 3). In the case of deletion derivative rad51A2-98E7d4, all of exon VII has been deleted. Because only half of the progeny from cross 2 would be expected to carry rad51A2-98E7, the rate at which adjacent deletions were recovered from rad51A2-98E7 was ~0.8% (4/~500).

A deletion derivative of rad51A1-54F11:

The rad51A1-54F11d1 deletion derivative was isolated from a family segregating for rad51A1-54F11 (materials and methods). Sequence analysis of a PCR product amplified from rad51A1-54F11d1 established that the 363 bp (~1447–1809 in GenBank accession no. AY359681) that was deleted includes part of exon II, all of intron II and exon III, and part of intron III (Figure 3).

RT–PCR analysis of rad51 double mutants:

RT–PCR experiments were conducted to test whether the rad51A1 and rad51A2 deletion derivative alleles are null mutants. Consistent with observations of Franklin et al. (1999), rad51A1 and rad51A2 transcripts of the expected sizes were detected during meiotic prophase in maize tassels that carry Rad51A1 and Rad51A2, respectively (Figure 4). In contrast, rad51A1 transcripts were not detected in plants homozygous for rad51A1-54F11d1. Hence, rad51A1-54F11d1 is an apparently null mutant allele.

Figure 4.
RT–PCR analysis of rad51A1 and rad51A2. RNA was isolated from tassels undergoing meiotic prophase from plants with the following genotypes: homozygous for Rad51A1 and Rad51A2 (lanes 1 and 2), homozygous for rad51A1-54F11d1 and Rad51A2 (lanes 3 ...

Expression of the rad51A2-98E7d4 was analyzed similarly. Although ethidium bromide staining of the RT–PCR products did not reveal any rad51A2 transcripts in plants homozygous for rad51A2-98E7d4, RT–PCR products of two transcripts could be detected via hybridization with a rad51A2 cDNA probe (data not shown). Sequence analysis demonstrated that one of these transcripts is missing all of exon VII as expected on the basis of the structure of rad51A2-98E7d4 (Figures 2 and and3).3). The deletion of exon VII causes a frameshift in this transcript. The second transcript is also missing exon VII but includes 142 bp from the TIR of the MuDR transposon. In this transcript, a stop codon has been introduced before exon VIII. Hence, in both transcripts ~ 80 amino acids of the normal RAD51A2 protein, including a large portion of the ATP binding site (Shinohara et al. 1992; Nara et al. 2000), are lost due to deletions, frameshift, or a premature stop codon. Hence, it is very unlikely that rad51A2-98E7d4 encodes a functional protein.

The rad51 double mutant does not accumulate RAD51 foci:

To further test whether rad51A1-54F11d1 and rad51A2-98E7d4 are null alleles, meiocytes from wild type and rad51 double mutants were subjected to immunolocalization using RAD51 polyclonal antibodies (Franklin et al. 1999). In wild type, RAD51 appears on chromosomes as distinct foci and reaches a maximum of 500–600 foci in mid-zygotene (Franklin et al. 1999; Figure 5A). In contrast, true RAD51 foci were not observed in rad51 double mutant cells (Figure 5B). This dramatic difference, together with RT–PCR results (Figure 4), demonstrates that the rad51A1-54F11d1 and rad51A2-98E7d4 alleles do not encode functional RAD51.

Figure 5.
RAD51 foci in zygotene cells and phenotypes of rad51 double mutant. RAD51 immunostaining is shown in green and DAPI stained chromosomes are in red. (A) Wild-type nucleus showing hundreds of distinct RAD51 foci. (B) A rad51 double mutant nucleus showing ...

RAD51 is essential for the repair of radiation-induced DSBs during vegetative development:

To address the role of RAD51 in DNA repair during early vegetative development, ~200 dry kernels that were segregating for rad51 double mutants were irradiated with a dose of 0.37 kGy prior to germination. While control siblings survived to develop into apparently healthy seedlings, all the rad51 double mutants (N = 31) germinated 3–4 days later than control siblings, were able to produce only one leaf, and died within 2 weeks of germination (Table 1). This result indicates that RAD51 function is essential for the repair of radiation-induced DSBs in maize.

RAD51 is required for the repair of radiation-induced DSBs

Effects of rad51 double mutants on gamete production:

Plants homozygous for rad51A1-54F11, rad51A1-54F11d1, or rad51A2-98E7d4 (i.e., single mutants) develop apparently normally, are fully male fertile, and have normal seed sets. To determine whether the rad51 double mutant exhibits a phenotype, the progeny resulting from the self-pollinations of plants heterozygous for rad51A1-54F11 or rad51A1-54F11d1 and homozygous for rad51A2-98E7d4 were genotyped and scored for male and female fertility. Among the progeny of these crosses, all double mutant plants were male sterile (i.e., that had anthers that failed to exert and that contained no morphologically normal pollen grains) and exhibited dramatic reductions in female seed sets (Table 2 and Figure 5, C and D). When crossed as females (i.e., serving as pollen recipients) with wild-type pollen, 14 rad51 double mutants produced 800 kernels, while six control wild-type siblings produced 1555 kernels. Hence, in this experiment, seed set of rad51 double mutants was ~22% ((800/14)/(1555/6) × 100% = 22%) of wild type. In all other respects double mutant plants developed apparently normally.

Correlation between male sterility and the rad51 double mutant

Reduced number of bivalents and chiasmata in rad51 double mutants:

Meiosis was analyzed using acetocarmine squashes in microsporocytes of plants that were wild type, homozygous for either of the two single mutations (i.e., rad51A1/rad51A1; Rad51A2/Rad51A2 or Rad51A1/Rad51A1; rad51A2/rad51A2) or the rad51 double mutant. The meiotic sequence appeared to be normal in both of the rad51 single mutant genotypes.

At diakinesis the average number of bivalents per wild type and rad51 double mutant cell was 10 (10.0 ± 0.1) and 4.1 ± 2.1, respectively (Table 3, Figure 6, A–D). To determine the number of chiasmata, the numbers of ring-shaped bivalents, which have two chiasmata, and rod-shaped bivalents, which have one chiasma, were counted (Figure 6). In rad51 double mutants, most of the bivalents are rod shaped, indicating a decrease in the number of chiasmata per bivalent. In rad51 double mutants, the average number of chiasmata per bivalent is 1.0 ± 0.3, while in wild type it is 1.9 ± 0.1 (Table 3). Although the average number of chiasmata per double mutant cell is only 4.5 ± 2.5 (as compared to 20 in normal cells), the existence of chiasmata suggests that some degree of recombination and synapsis occurs in the absence of RAD51.

Figure 6.
Bivalents and multivalents in rad51 double mutants at diakinesis. Male meiocytes from wild type (A) or from rad51A1-54F11d1/rad51A1-54F11d1; rad51A2-98E7d4/rad51A2-98E7d4 (B–H) were stained with acetocarmine and observed under light microscopy. ...
Average number of bivalents and chiasmata per diakinesis cell in rad51 double mutants and wild type control

Chiasmata between nonhomologous chromosomes in rad51 double mutants:

A unique and significant feature of the rad51 double mutants was discovered while counting chiasmata in metaphase I cells. In addition to univalents and bivalents, multivalents (trivalents and quadrivalents) and heteromorphic bivalents were observed in the rad51 double mutants (Figure 6, E–H). Heteromorphic bivalents have been studied in wheat and interspecies hybrids and multivalents can be easily seen in translocation heterozygotes in maize, and both have characteristic morphology. A heteromorphic bivalent appears as two chromosomes of different sizes held together at metaphase I by chiasmata (Figure 6H). Although it is possible that this is the result of a homologously paired bivalent that has suffered breakage, we consider this unlikely because the breakage would have to involve both sister chromatids. Even more convincing evidence of chiasmata between nonhomologous chromosomes comes from the multivalents, which appear as three or four chromosomes forming chain-shaped (Figure 6, E and G) or ring-shaped (Figure 6F) quadrivalents. The frequency of multivalent and heteromorphic bivalents was ~6%: among 82 metaphase I cells analyzed, only one heteromorphic bivalent and four multivalents were found (Figure 6). Thus, the nonhomologous synapsis observed in pachytene sometimes allows for chiasmata between nonhomologous chromosomes. Chiasmata between nonhomologous chromosomes do not occur in any other pairing-defective maize meiotic mutants that have been analyzed, even those with elevated levels of nonhomologous synapsis, such as phs1 (I. N. Golubovskaya, data not shown, and Pawlowski et al. 2004). This indicates that in the absence of RAD51, nonhomologous chromosomes can synapse. To our knowledge, multivalents and heteromorphic bivalents have been observed only in the ph1 mutant of common wheat, where homeologous chromosomes can synapse (in addition to homologous chromosomes), recombine, and form chiasmata (Riley and Chapman 1964; Riley et al. 1966) and Arabidopsis zyp1 mutants (Higgins et al. 2005).

Chromosome breaks from male meiocytes in rad51 double mutants:

The first indication of chromosome breakage in the rad51 double mutants was seen at diakinesis and metaphase I. Numerous chromosome fragments are evident at anaphase I, indicating a high degree of chromosome breakage in the rad51 double mutants (Figure 7). Of 35 anaphase I cells specifically analyzed for breakage (and in hundreds of other cells looked at for other purposes), all showed chromosome breakage. Different sizes of chromosome fragments range from a whole chromosome arm to tiny, single, and double fragments. This is very unusual in maize, where chromosome breaks in other meiotic mutants are “healed” by late pachytene (Pawlowski et al. 2004). In addition to chromosome breaks, abnormal stretches of chromosomes are seen at anaphase I (Figure 7, B–F). These appear broken randomly throughout the spindle. These are distinguished from classical chromosome bridges, which are also observed in rad51 double mutants at anaphase II accompanied by fragments (Figure 7J). Classical chromosome bridges (as in Figure 7J), resulting from dicentric chromosomes being pulled to two poles at anaphase I, span the distance from one daughter nucleus to the other, are very stable, and are usually broken only by the phragmoplast (as can be seen in Figure 7I).

Figure 7.
Chromosome breakage and bridges occur in rad51 double mutants. These are acetocarmine chromosome squashes of rad51 double mutant meiocytes in pachytene through telophase II. (A) Pachytene looks completely normal at this level of examination. (B) Metaphase ...

RAD51 is required for normal meiotic disjunction:

The meiotic abnormalities observed in the rad51 double mutant would be expected to affect meiotic chromosome segregation. Quartets of microspores were examined in eight rad51 double mutant plants. In normal maize plants, one nucleolus is present in each of the four microspores of each quartet because each microspore contains only one copy of chromosome 6, which carries the only nucleolar organizing region (NOR) (McClintock 1934). Many of the quartets contained one or two microspores that lacked an organized nucleolus (Figure 8A) but contained numerous micronucleoli (smaller bodies that resembled nucleoli, Figure 8, B–F). Because two nucleoli can fuse together, the types of aberrant quartets illustrated in Figure 8, B–D, could arise via nondisjunction or loss of chromosome 6 in one or more cells. Microspores that contained two nucleoli (Figure 8, E and F) have two copies of chromosome 6. The observed frequencies of the various types of quartets in four of the double mutant plants are provided in Table 4. Because cells with two nucleoli possess two copies of chromosome 6, the abnormal quartet types observed in the rad51 double mutant in this study could be produced only if chromosome 6 underwent nondisjunction at one of the meiotic divisions. For example, quartet type VIII (Table 4) could be produced only via nondisjunction of chromosome 6 during the first meiotic division. Similarly, quartet types III and IV (Table 4) could be produced only via nondisjunction of chromosome 6 during the second meiotic division. Hence, rad51 function is required for the normal disjunction of chromosome 6 (and presumably the other chromosomes) during both meiotic divisions.

Figure 8.
Types of quartets observed from rad51 double mutants. Normal (A) and abnormal (B–F) quartets. The arrowhead in A indicates an organized nucleolus. The arrowheads in B and C indicate microspores missing an organized nucleolus. The arrowheads in ...
Frequencies of quartet types in rad51 single and double mutants

Meiotic homologous pairing is reduced in the rad51 double mutants:

Even though chromosome pairing at pachytene in rad51 double mutants looks indistinguishable from wild-type plants at the light microscope level (Figure 7A), we wanted to determine if the abnormalities observed from metaphase I to quartet stage were a consequence of an earlier meiotic malfunction. The degree of homologous pairing in rad51 double mutants was assessed using FISH with a 5S rDNA probe that hybridizes to a single locus in maize (Golubovskaya et al. 2002). Since maize is diploid, two FISH foci are seen if homologous pairing has not occurred and only one if pairing is complete. In wild type at pachytene, 100% of cells show one single, bright 5S rDNA focus, indicating that homologous pairing has occurred. In 41 rad51 double mutant pachytene meiocytes, 18 (44%) had a single focus as in wild type (Figure 9A) and 23 (56%) had two unpaired foci, indicating a lack of homologous pairing. Of these 23 cell with unpaired 5S rDNA foci, 14 (34% of total) had two foci on chromosomes that were closely appressed (i.e., apparently synapsed) to a nonhomologous chromosome (Figure 9B), and 9 (22%) had two very close, but unpaired foci on short stretches of unpaired chromosomes (Figure 9C). These data indicate that while homologous pairing is possible in the rad51 double mutant, it is not nearly as efficient as in wild type. The reduced frequency of correct homologous pairing (44% of normal) correlates well with the reduced degree of bivalents (40%) in rad51 double mutants. In addition, the apparent synapsis between nonhomologous chromosomes (Figure 9B) also correlates with the presence of multivalents and heteromorphic bivalents (Figure 6, E–H).

Figure 9.
FISH analysis of 5S rDNA loci in pachytene rad51 double mutants meiocytes Chromosomes are shown in red (DAPI) and 5S rDNA loci in green. (A) Paired 5S rDNA foci. (B) Two unpaired 5S rDNA foci. Each 5S rDNA locus pairs with and appears to be synapsed with ...

Nonhomologous synapsis is evident in the rad51 double mutants:

To assess the integrity of synapsis in the rad51 double mutants, double immunolocalizations were performed with antibodies against AFD1 and ASY/HOP1. AFD1 is the maize REC8 homolog and marks both axial (before synapsis) and lateral (after synapsis) elements of the synaptonemal complex. When meiocytes are fixed in 4% formaldehyde and buffer A, AFD1/REC8 marks lateral elements more strongly. ASY1/HOP1 is an antibody toward the Arabidopsis HOP1 homolog, and in maize meiocytes fixed in 4% formaldehyde it marks only the unsynapsed axial elements (Golubovskaya et al. 2006). Thus, together, these antibodies can show regions where chromosomes are synapsed and unsynapsed. In wild type at pachytene, only AFD1/REC8 stains chromosomes (Golubovskaya et al. 2006). To examine whether synapsis is defective in rad51 double mutants, pachytene meiocytes were selected by their appearance of the chromosomes in the DAPI channel. In deconvolution microscopy, this can easily be determined because synapsed chromosomes are exactly twice the width of chromosomes at leptotene, when chromosomes are completely unpaired. Looking at the whole nuclei, rad51 double mutants look normal at zygotene and pachytene (Figure 10, A–C). However, upon close inspection of individual optical Z sections in the AFD channel, nonhomologous synapsis could be easily detected. This was evident mostly by synaptic partner switches (Figure 10, D–G). These Y-shaped structures represent two chromosomes, which synapse for some distance and then are interrupted at some point (arrows) by a partner switch. Thus, three chromosomes are involved in these structures and synapsis is nonhomologous for some distance. One to four partner switches were detected per nucleus, but there may be more because it is only possible to detect partner switches in the X–Y plane. Of 14 carefully examined cells, there was one event containing two sets of nonhomologously synapsed chromosomes forming an incomplete cross-shaped quadrivalent (Figure 10F).

Figure 10.
Nonhomologous synpasis in the rad51 double mutant. (A–C) Immunolocalization of ASY1/HOP1 (green) showing unsynapsed axial elements and AFD1 (red) showing the lateral elements of synapsed chromosome regions. DAPI stained chromosomes are shown in ...

Meiotic crossovers in rad51 double mutants:

Our finding that frequency of chiasmata at diakinesis was low in male gametes of rad51 double mutants (Table 3) is consistent with the fact that reduced chromosome pairing was observed at pachytene (Figure 9) and only ~62% quartets were normal due to aberrant segregation of chromosome 6 (Table 4).

The rates of meiotic crossovers between two genetic markers (a1 and et1) were assayed in female double mutants and rad51A2 single mutant controls. In this 10-cM interval (Goodman et al. 1980; da Costa e Silva et al. 2004) no significant differences in the rates of crossovers were detected between rad51 double mutants and controls (Table 5). The rates of crossovers were also measured at another genetic interval (i.e., between IDP1440 and IDP1983) using an independent population. Similarly, no significant differences in the rates were detected between rad51 double mutants and wild-type controls (Table 5). These results indicate that RAD51 is not essential for meiotic crossovers in surviving female meiocytes.

Effect of RAD51 on rates of meiotic crossovers across two intervals

Further support for this conclusion comes from cytological analyses of progeny of rad51 double mutant females pollinated with wild-type pollen. Root tips of 20 such progeny were examined cytologically; each contained the diploid number (N = 20) of chromosomes. Meiocytes from similarly produced progeny of rad51 double mutant females were also examined; all had 10 pairs of chromosomes.


The two rad51 genes are functionally redundant:

Consistent with the hypothesis that maize is a segmental allotetraploid (Gaut et al. 2000), the maize genome contains two rad51 homologs: rad51A1 and rad51A2 (Franklin et al. 1999; supplemental Figure 1 at http://www.genetics.org/supplemental/). Apparently null or severely disrupted alleles have been isolated from both genes. On the basis of the observations that there is no evidence for more than two rad51 genes in maize (supplemental Figure 1), that the rad51 double mutants do not accumulate normal rad51A1 or rad51A2 transcripts (Figure 4), and that RAD51 foci are not detectable at zygotene (Figure 5), we conclude that the rad51 double mutant does not exhibit residual RAD51 activity.

Plants homozygous for single mutants of either gene develop normally, are fully male fertile, and do not display reduced seed set. Cytological observations do not reveal any differences between these single mutants and the inbred line B73, which carries functional alleles of both genes. In contrast, plants that are homozygous for mutant alleles at both loci are completely male sterile (Figure 5C) and exhibit dramatically reduced seed set (Figure 5D) and abnormal male meiosis (Figures 699).). Hence, rad51A1 and rad51A2 appear to be functionally redundant in maize.

Roles of RAD51 in DSBs repair:

This study demonstrates both significant differences and similarities in the roles of RAD51 in mitotic cells of vertebrates and plants. Consistent with the finding of Li et al. (2004), we demonstrate that under normal conditions, the RAD51-dependent homologous recombination repair pathway plays at best only a minor role in mitotic plant cells. In contrast, disruptions of rad51 in vertebrates are lethal (Tsuzuki et al. 1996; Sonoda et al. 1998) due to the accumulation of unrepaired DSBs (Sonoda et al. 1998).

It has not previously been shown that RAD51 is involved in the repair of mitotic DSBs in plants (Bleuyard et al. 2006). The finding that maize rad51 double mutant embryos are extremely susceptible to radiation-induced DSBs (Table 1) demonstrates a conserved role for RAD51 in the repair of mitotic DSBs in plants, vertebrates, and yeast.

The maize rad51 double mutant is unique among some 20 other meiotic maize mutants in that it shows chromosome fragmentation. Most clearly established in the phs1 mutant, meiotic chromosome breaks that are not repaired in meiotic mutants are somehow “healed” in late prophase in maize (Pawlowski et al. 2004). On the basis of this observation, it has been postulated that a backup DNA repair mechanism exists late in maize meiosis (Pawlowski et al. 2004). This is different from Arabidopsis, where many meiotic mutants show chromosome fragmentation (Bleuyard and White 2004; Li et al. 2004, 2005; Hamant et al. 2006). The maize rad51 double mutant points to a unique and specific role of RAD51 in maize, viz., it may be required for this backup repair mechanism.

RAD51 is critical but not essential for meiotic chromosome pairing, synapsis, and crossovers:

The rad51 double mutant exhibits a reduced amount of homologous pairing (44% of wild type), a reduced number of bivalents (4 vs. 10), and nearly completely synapsed chromosomes (Figure 10C). This is consistent with observations in yeast, in which rad51 mutants undergo normal chromosome pairing and nearly perfect synapsis (Rockmill et al. 1995). Our data differ from the previous report that in Arabidopsis rad51 plays an essential role in chromosome pairing and synapsis (Li et al. 2004).

The residual female fertility associated with the maize rad51 double mutant allowed us to demonstrate that in maize, as is the case for yeast (Shinohara et al. 1992, 1997), rates of meiotic crossovers in the absence of RAD51 are not statistically distinguishable from controls (Table 5). Because these data were collected from the ~22% of surviving female gametes, we cannot, however, exclude the possibility that there may be selection that favors gametes with higher rates of meiotic crossovers. Even so, the significant number of chiasmata observed in male meiocytes (Table 3) provides additional evidence that meiotic crossovers occur in the rad51 double mutant. Due to the complete sterility or lethality associated with rad51 mutants, it has not been possible to measure rates of meiotic crossovers in Arabidopsis (Li et al. 2004) and vertebrates (Tsuzuki et al. 1996; Sonoda et al. 1998).

Does maize dmc1 partially complement rad51 in meiosis?

This study demonstrates that in maize, as is true in yeast (Shinohara et al. 1992, 1997; Rockmill et al. 1995), rad51 genes affect meiotic chromosome pairing, synapsis, and crossover, but are not essential. We do not imply that the RAD51 protein is not involved in these processes. Rather, we hypothesize that in maize (unlike Arabidopsis) these roles of RAD51 can be at least partially complemented by other RecA homologs. If true, this maize homolog would be predicted to exhibit functional differences relative to the corresponding Arabidopsis homolog (Couteau et al. 1999; Bleuyard et al. 2005). Although there are multiple rad51 paralogs in plants (Figure 1; Bleuyard et al. 2005), our working hypothesis, based on colocalization data from multiple species (Bishop 1994; Anderson et al. 1997; Tarsounas et al. 1999), in vitro functional analyses from humans and yeast (Sung 1994; Sehorn et al. 2004), and genetic data from yeast (Shinohara et al. 1997), is that one or both of the two maize dmc1 genes (Franklin et al. 1999) can complement the meiotic chromosome pairing, synapsis, and crossover functions normally provided by RAD51, as has been proposed for yeast (Shinohara et al. 1997; Tsubouchi and Roeder 2003).

Absence of RAD51 leads to nonhomologous pairing and synapsis in meiosis:

The maize rad51 double mutant exhibits a phenotype that has not been observed in rad51 mutants of other species. Specifically, we observed pairing, synapsis, and chiasmata between nonhomologous chromosomes. It has previously been demonstrated that nonhomologous synapsis occurs in mutants such as phs1 (Pawlowski et al. 2004); the chiasmata observed between nonhomologous chromosomes in the rad51 double mutant suggests that crossovers can occur between these nonhomologous chromosomes as occurs in the ph mutant of wheat (Riley and Chapman 1964; Riley et al 1966). In yeast, Rad51p has been shown to function in mediating strand exchange (Sung 1994; Benson et al. 1998; New et al. 1998; Shinohara and Ogawa 1998). It is possible that other RecA-like maize proteins, in the absence of RAD51, cannot efficiently mediate strand invasion and this inefficiency leads to nonhomologous pairing and synapsis in meiosis. On the other hand, the low percentage of nonhomologous pairing (Figure 9), nearly normal synapsis (Figure 10, A–C), and close to normal rates of meiotic crossovers from female surviving gametes (Table 5) suggest that other maize RecA homologs do an imperfect but adequate job to mediate strand invasion in rad51 double mutants. These findings lead us to propose that in maize, RAD51 functions to increase, directly or indirectly, the precision of homologous pairing during meiosis.


We thank Pramod Mahajan (Pioneer Hi-Bred International) for stimulating discussions, Suling Zhao (Pioneer-Hi Bred International) for technical assistance with identifying TUSC alleles for rad51 genes, Marianne Smith of the Schnable laboratory for technical assistance with light microcopy, and graduate student Sanzhen Liu for help with modifying figures. This research was supported by grants to P.S.S. from Pioneer Hi-Bred International and from the National Research Initiative of the USDA Cooperative State Research, Education and Extension Service, grant nos. 2001-35301-10685 and 2005-35301-15715. Additional support was provided by Hatch Act and State of Iowa funds.


Sequence data from this article have been deposited with the EMBL/GenBank Data Libraries under accession no. AY359681.


  • Ananiev, E. V., R. L. Phillips and H. W. Rines, 1998. Chromosome-specific molecular organization of maize (Zea mays L.) centromeric regions. Proc. Natl. Acad. Sci. USA 95: 13073–13078. [PMC free article] [PubMed]
  • Anderson, L. K., H. H. Offenberg, W. M. Verkuijlen and C. Heyting, 1997. RecA-like proteins are components of early meiotic nodules in lily. Proc. Natl. Acad. Sci. USA 94: 6868–6873. [PMC free article] [PubMed]
  • Barkan, A., and R. A. Martienssen, 1991. Inactivation of maize transposon Mu suppresses a mutant phenotype by activating an outward-reading promoter near the end of Mu1. Proc. Natl. Acad. Sci. USA 88: 3502–3506. [PMC free article] [PubMed]
  • Bass, H. W., W. F. Marshall, J. W. Sedat, D. A. Agard and W. Z. Cande, 1997. Telomeres cluster de novo before the initiation of synapsis: a three-dimensional spatial analysis of telomere positions before and during meiotic prophase. J. Cell Biol. 137: 5–18. [PMC free article] [PubMed]
  • Bensen, R. J., G. S. Johal, V. C. Crane, J. T. Tossberg, P. S. Schnable et al., 1995. Cloning and characterization of the maize An1 gene. Plant Cell 7: 75–84. [PMC free article] [PubMed]
  • Benson, F. E., P. Baumann and S. C. West, 1998. Synergistic actions of Rad51 and Rad52 in recombination and DNA repair. Nature 391: 401–404. [PubMed]
  • Bishop, D. K., 1994. RecA homologs Dmc1 and Rad51 interact to form multiple nuclear complexes prior to meiotic chromosome synapsis. Cell 79: 1081–1092. [PubMed]
  • Bleuyard, J. Y., and C. I. White, 2004. The Arabidopsis homologue of Xrcc3 plays an essential role in meiosis. EMBO J. 23: 439–449. [PMC free article] [PubMed]
  • Bleuyard, J. Y., M. E. Gallego, F. Savigny and C. I. White, 2005. Differing requirements for the Arabidopsis Rad51 paralogs in meiosis and DNA repair. Plant J. 41: 533–545. [PubMed]
  • Bleuyard, J. Y., M. E. Gallego and C. I. White, 2006. Recent advances in understanding of the DNA double-strand break repair machinery of plants. DNA Repair 5: 1–12. [PubMed]
  • Brown, W. E., D. S. Robertson and J. L. Bennetzen, 1989. Molecular analysis of multiple Muatotor-derived alleles of the Bronze locus of maize. Genetics 122: 439–445. [PMC free article] [PubMed]
  • Couteau, F., F. Belzile, C. Horlow, O. Grandjean, 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. [PMC free article] [PubMed]
  • Cui, X., A. P. Hsia, F. Liu, D. A. Ashlock, R. P. Wise et al., 2003. Alternative transcription initiation sites and polyadenylation sites are recruited during Mu suppression at the rf2a locus of maize. Genetics 163: 685–698. [PMC free article] [PubMed]
  • da Costa e Silva, O., R. Lorbiecke, P. Garg, L. Muller, M. Wassmann et al., 2004. The Etched1 gene of Zea mays (L.) encodes a zinc ribbon protein that belongs to the transcriptionally active chromosome (TAC) of plastids and is similar to the transcription factor TFIIS. Plant J. 38: 923–939. [PubMed]
  • Das, L., and R. Martienssen, 1995. Site-selected transposon mutagenesis at the hcf106 locus in maize. Plant Cell 7: 287–294. [PMC free article] [PubMed]
  • Dempsey, E., 1994. Traditional analysis of maize pachytene chromosomes, pp. 432–441 in The Maize Handbook, edited by M. Freeling and V. Walbot. Springer-Verlag, New York.
  • Dietrich, C. R., F. Cui, M. L. Packila, J. Li, D. A. Ashlock et al., 2002. Maize Mu transposons are targeted to the 5′ untranslated region of the gl8 gene and sequences flanking Mu target-site duplications exhibit nonrandom nucleotide composition throughout the genome. Genetics 160: 697–716. [PMC free article] [PubMed]
  • Doutriaux, M. P., F. Couteau, C. Bergounioux and C. White, 1998. Isolation and characterisation of the RAD51 and DMC1 homologs from Arabidopsis thaliana. Mol. Gen. Genet. 257: 283–291. [PubMed]
  • Franklin, A. E., J. McElver, I. Sunjevaric, R. Rothstein, B. Bowen et al., 1999. Three-dimensional microscopy of the Rad51 recombination protein during meiotic prophase. Plant Cell 11: 809–824. [PMC free article] [PubMed]
  • Gaut, B. S., M. Le Thierry d'Ennequin, A. S. Peek and M. C. Sawkins, 2000. Maize as a model for the evolution of plant nuclear genomes. Proc. Natl. Acad. Sci. USA 97: 7008–7015. [PMC free article] [PubMed]
  • Golubovskaya, I. N., L. C. Harper, W. P. Pawlowski, D. Schichnes and W. Z. Cande, 2002. The pam1 gene is required for meiotic bouquet formation and efficient homologous synapsis in maize (Zea mays L.). Genetics 162: 1979–1993. [PMC free article] [PubMed]
  • Golubovskaya, I. N., O. Hamant, L. Timofejeva, C. R. Wang, D. Braun et al., 2006. Alleles of afd1 dissect REC8 functions during meiotic prophase I. J. Cell Sci. 119: 3306–3315. [PubMed]
  • Goodman, M. M., C. W. Stuber, K. Newton and H. H. Weissinger, 1980. Linkage relationships of 19 enzymes loci in maize. Genetics 96: 697–710. [PMC free article] [PubMed]
  • Grishchuk, A. L., and J. Kohli, 2003. Five RecA-like proteins of Schizosaccharomyces pombe are involved in meiotic recombination. Genetics 165: 1031–1043. [PMC free article] [PubMed]
  • Hamant, O., H. Ma and W. Z. Cande, 2006. Genetics of meiotic prophase I in plants. Annu. Rev. Plant Biol. 57: 267–302 [PubMed]
  • Hays, S. L., A. A. Firmenich and P. Berg, 1995. Complex formation in yeast double-strand break repair: participation of Rad51, Rad52, Rad55, and Rad57 proteins. Proc. Natl. Acad. Sci. USA 92: 6925–6929. [PMC free article] [PubMed]
  • Higgins, J. D., E. Sanchez-Moran, S. J. Armstrong, G. H. Jones and F. C. Franklin, 2005. The Arabidopsis synaptonemal complex protein ZYP1 is required for chromosome synapsis and normal fidelity of crossing over. Genes Dev. 19: 2488–2500. [PMC free article] [PubMed]
  • Levy, A. A., and V. Walbot, 1991. Molecular analysis of the loss of somatic instability in the bz2::mu1 allele of maize. Mol. Gen. Genet. 229: 147–151. [PubMed]
  • Levy, A. A., A. B. Britt, K. R. Luehrsen, V. L. Chandler, C. Warren et al., 1989. Developmental and genetic aspects of Mutator excision in maize. Dev. Genet. 10: 520–531. [PubMed]
  • Li, W., C. Chen, U. Markmann-Mulisch, L. Timofejeva, E. Schmelzer et al., 2004. The Arabidopsis AtRAD51 gene is dispensable for vegetative development but required for meiosis. Proc. Natl. Acad. Sci. USA 101: 10596–10601. [PMC free article] [PubMed]
  • Li, W., X. Yang, Z. Lin, L. Timofejeva, R. Xiao et al., 2005. The AtRAD51C gene is required for normal meiotic chromosome synapsis and double-stranded break repair in Arabidopsis. Plant Physiol. 138: 965–976. [PMC free article] [PubMed]
  • Martienssen, R., and A. Baron, 1994. Coordinate suppression of mutations caused by Robertson's mutator transposons in maize. Genetics 136: 1157–1170. [PMC free article] [PubMed]
  • McClintock, B., 1934. The relation of a particular chromosomal element to the development of the nucleoli in Zea mays. Z. Zellforsch. Mikroskop. Anat. 21: 294–328.
  • Nara, T., T. Yamamoto and K. Sakaguchi, 2000. Characterization of interaction of C- and N-terminal domains in LIM15/DMC1 and RAD51 from a basidiomycetes, Coprinus cinereus. Biochem. Biophys. Res. Commun. 275: 97–102. [PubMed]
  • New, J. H., T. Sugiyama, E. Zaitseva and S. C. Kowalczykowski, 1998. Rad52 protein stimulates DNA strand exchange by Rad51 and replication protein A. Nature 391: 407–410. [PubMed]
  • Ogawa, T., X. Yu, A. Shinohara and E. H. Egelman, 1993. Similarity of the yeast RAD51 filament to the bacterial RecA filament. Science 259: 1896–1899. [PubMed]
  • Pawlowski, W. P., I. N. Golubovskaya and W. Z. Cande, 2003. Altered nuclear distribution of recombination protein RAD51 in maize mutants suggests the involvement of RAD51 in meiotic homology recognition. Plant Cell 15: 1807–1816. [PMC free article] [PubMed]
  • Pawlowski, W. P., I. N. Golubovskaya, L. Timofejeva, R. B. Meeley, W. F. Sheridan et al., 2004. Coordination of meiotic recombination, pairing, and synapsis by PHS1. Science 303: 89–92. [PubMed]
  • Riley, R., and V. Chapman, 1964. Cytological determination of the homoeology of chromosomes of Triticum aestivum. Nature 203: 156–158.
  • Riley, R., V. Chapman, R. M. Young and A. M. Belfield, 1966. Control of meiotic chromosome pairing by the chromosomes of homoeologous group 5 of Triticum aestivum. Nature 212: 1475–1477.
  • Robertson, D. S., 1978. Characterization of a mutator system in maize. Mutat. Res. 51: 21–28.
  • Robertson, D. S., 1980. The timing of Mu activity in maize. Genetics 94: 969–978. [PMC free article] [PubMed]
  • Robertson, D. S., 1981. Mutator activity in maize: timing of its activation in ontogeny. Science 213: 1515–1517. [PubMed]
  • Roca, A. I., and M. M. Cox, 1990. The RecA protein: structure and function. Crit. Rev. Biochem. Mol. Biol. 25: 415–456. [PubMed]
  • Rockmill, B., M. Sym, H. Scherthan and G. S. Roeder, 1995. Roles for two RecA homologs in promoting meiotic chromosome synapsis. Genes Dev. 9: 2684–2695. [PubMed]
  • Schnable, P. S., P. A. Peterson and H. Saedler, 1989. The bz-rcy allele of the Cy transposable element system of Zea mays contains a Mu-like element insertion. Mol. Gen. Genet. 217: 459–463. [PubMed]
  • Sehorn, M. G., S. Sigurdsson, W. Bussen, V. M. Unger and P. Sung, 2004. Human meiotic recombinase Dmc1 promotes ATP-dependent homologous DNA strand exchange. Nature 429: 433–437. [PubMed]
  • Shinohara, A., and T. Ogawa, 1998. Stimulation by Rad52 of yeast Rad51-mediated recombination. Nature 391: 404–407. [PubMed]
  • Shinohara, A., H. Ogawa and T. Ogawa, 1992. Rad51 protein involved in repair and recombination in S. cerevisiae is a RecA-like protein. Cell 69: 457–470. [PubMed]
  • Shinohara, A., S. Gasior, T. Ogawa, N. Kleckner and D. K. Bishop, 1997. Saccharomyces cerevisiae recA homologues RAD51 and DMC1 have both distinct and overlapping roles in meiotic recombination. Genes Cells 2: 615–629. [PubMed]
  • Sonoda, E., M. S. Sasaki, J. M. Buerstedde, O. Bezzubova, A. Shinohara et al., 1998. Rad51-deficient vertebrate cells accumulate chromosomal breaks prior to cell death. EMBO J. 17: 598–608. [PMC free article] [PubMed]
  • Sung, P., 1994. Catalysis of ATP-dependent homologous DNA pairing and strand exchange by yeast RAD51 protein. Science 265: 1241–1243. [PubMed]
  • Tarsounas, M., T. Morita, R. E. Pearlman and P. B. Moens, 1999. RAD51 and DMC1 form mixed complexes associated with mouse meiotic chromosome cores and synaptonemal complexes. J. Cell Biol. 147: 207–220. [PMC free article] [PubMed]
  • Taylor, L. P., and V. Walbot, 1985. A deletion adjacent to the maize transposable element Mu-1 accompanies loss of Adh1 expression. EMBO J. 4: 869–876. [PMC free article] [PubMed]
  • Terasawa, M., A. Shinohara, Y. Hotta, H. Ogawa and T. Ogawa, 1995. Localization of RecA-like recombination proteins on chromosomes of the lily at various meiotic stages. Genes Dev. 9: 925–934. [PubMed]
  • Tsubouchi, H., and G. S. Roeder, 2003. The importance of genetic recombination for fidelity of chromosome pairing in meiosis. Dev. Cell 5: 915–925. [PubMed]
  • Tsuzuki, T., Y. Fujii, K. Sakumi, Y. Tominaga, K. Nakao et al., 1996. Targeted disruption of the Rad51 gene leads to lethality in embryonic mice. Proc. Natl. Acad. Sci. USA 93: 6236–6240. [PMC free article] [PubMed]
  • Zimmer, E. A., E. R. Jupe and V. Walbot, 1988. Ribosomal gene structure, variation and inheritance in maize and its ancestors. Genetics 120: 1125–1136. [PMC free article] [PubMed]

Articles from Genetics are provided here courtesy of Genetics Society of America
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • EST
    Published EST sequences
  • Gene
    Gene links
  • Gene (nucleotide)
    Gene (nucleotide)
    Records in Gene identified from shared sequence links
  • GEO Profiles
    GEO Profiles
    Related GEO records
  • GSS
    Published GSS sequences
  • MedGen
    Related information in MedGen
  • Nucleotide
    Published Nucleotide sequences
  • Protein
    Published protein sequences
  • PubMed
    PubMed citations for these articles
  • Taxonomy
    Related taxonomy entry
  • Taxonomy Tree
    Taxonomy Tree

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...