The relationship between synaptonemal complex formation (synapsis) and double-strand break formation (recombination initiation) differs between organisms. Although double-strand break creation is required for normal synapsis in Saccharomyces cerevisiae and the mouse, it is not necessary for synapsis in Drosophila and Caenorhabditis elegans. To investigate the timing of and requirements for double-strand break formation during Drosophila meiosis, we used an antibody that recognizes a histone modification at double-strand break sites,phosphorylation of HIS2AV (γ-HIS2AV). Our results support the hypothesis that double-strand break formation occurs after synapsis. Interestingly, we detected a low (10-25% of wildtype) number of γ-HIS2AV foci in c(3)G mutants, which fail to assemble synaptonemal complex,suggesting that there may be both synaptonemal complex-dependent and synaptonemal complex-independent mechanisms for generating double-strand breaks. Furthermore, mutations in Drosophila Rad54 (okr) and Rad51 (spnB) homologs cause delayed and prolonged γ-HIS2AV staining, suggesting that double-strand break repair is delayed but not eliminated in these mutants. There may also be an interaction between the recruitment of repair proteins and phosphorylation.

Introduction

Meiotic recombination in Drosophila melanogaster requires mei-W68 (McKim and Hayashi-Hagihara, 1998), a Spo11 homolog. On the basis of physical studies and homology to the archaea topoisomerase II subunit Top6A, Spo11 is presumed to be the protein responsible for double-strand break (DSB) formation(and therefore meiotic recombination initiation) in Saccharomyces cerevisiae and Schizosaccharomyces pombe(Cervantes et al., 2000; Keeney et al., 1997). Furthermore, Spo11 homologs have been identified in a wide variety of organisms, leading to the hypothesis that the creation of DSBs is a conserved mechanism for initiating meiotic recombination(Keeney, 2001). Similarly,repair of meiotic DSBs requires members of a conserved group of genes such as homologs of Rad51 and Rad54 (McKim et al.,2002; Villeneuve and Hillers,2001).

Another hallmark of meiotic prophase is the synaptonemal complex (SC), a specialized protein-chromosome structure that physically connects aligned homologous chromosomes. Completion of synapsis between homologous chromosomes is marked by the presence of SC along their entire lengths. SC can form in the absence of DSBs in Drosophila(Liu et al., 2002; McKim et al., 1998) and Caenorhabditis elegans (Dernburg et al., 1998), but not in budding yeast(Roeder, 1997) or mouse(Baudat et al., 2000; Romanienko and Camerini-Otero,2000). The relative timing of DSBs and SC formation in the mouse was determined using an antibody that recognizes the phosphorylated form of a histone H2A variant, H2AX (γ-H2AX). On induction of DSBs in mammalian mitotic and meiotic cells, H2AX is rapidly phosphorylated(Rogakou et al., 1999).γ-H2AX staining was detected before the appearance of SC proteins,suggesting that DSBs appear before SC formation during meiotic prophase in the mouse (Mahadevaiah et al.,2001), consistent with time-course studies of DSB formation in S. cerevisiae (Padmore et al.,1991).

Drosophila has a single H2A variant (HIS2AV) that, like H2AX, is phosphorylated at a conserved SQ motif within an extended C-terminal tail following chromosome breakage in mitotic cells(Madigan et al., 2002). Phosphorylation of Drosophila HIS2AV (γ-HIS2AV) in mitotic cells is rapid, reaching its maximum within five minutes of exposure to agents that induce DSBs but not single-strand nicks(Madigan et al., 2002). Here we provide evidence that HIS2AV is phosphorylated in response to meiotic DSB formation. Using γ-HIS2AV staining as a marker for DSB formation, we found evidence that DSB formation occurs after synapsis and is partially dependent on the SC protein C(3)G. In addition, γ-HIS2AV staining suggests that DSB repair is delayed in okr (Rad54 homolog) and spnB (Rad51/Dmc1 homolog) mutants, but it does occur eventually.

Materials and Methods

Genetic techniques

Drosophila mei-W681 and mei-W684572are strong alleles that eliminate meiotic recombination(McKim and Hayashi-Hagihara,1998) (R.B. and K.M., unpublished). His2Av810is a 31 bp deletion that removes the second exon and causes recessive lethality (van Daal and Elgin,1992). The P{His2AvΔCT} transgene is a 4 kb fragment containing a His2Av gene with a stop codon inserted at amino acid Q127 that truncates the protein 14 amino acids early and eliminates the phosphorylation site (Clarkson et al.,1999). X-chromosome nondisjunction was assayed by crossing females to C(1;Y)1, v f B; C(4)RM, ci ey/0 males. The nondisjunction frequency was calculated as 2(exceptional progeny)/[2(exceptional progeny)+(regular progeny)]. Crossing over on the second chromosome was measured by crossing His2AvΔCT/+; al dp b pr cn/+;His2Av810 females to al dp b pr cn males.

Cytology

For immunolocalization experiments, females were aged for 16-40 hours at room temperature, dissected and fixed using either the `Buffer A' protocol(Belmont et al., 1989) or the PBS-based protocol described by Page and Hawley(Page and Hawley, 2001). To detect DSBs, the Anti-phospho-H2AX (Ser139) rabbit polyclonal antibody(Upstate Biotechnology) was used at 1:100 and a Cy3-labeled goat anti-rabbit secondary antibody (Amersham) was used at 1:250. The guinea pig C(3)G antibody(Page and Hawley, 2001) was used at 1:500 and a FITC-labeled goat anti-guinea pig secondary antibody(Vector) was used at 1:250. A combination of two Orb antibodies (4H8 and 6H4)(Lantz et al., 1994) was used at 1:150 and a FITC-labeled goat anti-mouse secondary antibody (Vector) was used at 1:250 or a Cy5-labeled goat anti-mouse (Amersham) was used at 1:100. Chromosomes were stained with 4',6'-diamidino-2-phenylindole hydrochloride(DAPI) (0.2 μM) for 10 minutes or Hoechst (0.1 μl/ml of a 10 mg/ml solution) for 5 minutes. Images were collected using a Leica TCS SP2 confocal microscope or a Zeiss Axioplan II imaging microscope equipped with a Cooke Corp. Sensicam CCD camera using a 63× or 100× objective. Sections were collected at 0.2 μm or 0.3 μm intervals.

The counts of γ-HIS2AV foci were made from a subset of germaria where the foci were clear. Foci were manually counted by examining the full series of optical sections containing an oocyte nucleus. This method might underestimate foci because two foci that were in the same X-Y position but in consecutive Z-sections may have been counted as a single focus. In addition,the foci were not always the same intensity, which may reflect their asynchronous development. It was not difficult to distinguish the nuclear foci from background because the latter was cytoplasmic. For example, the cytoplasmic background was variable between germaria but uniform throughout the germarium. By contrast, nuclear foci in wildtype were only present in region 2a and early 2b. Nuclei at other stages were generally free of staining.

Results

A phosphorylated H2A variant is a molecular marker of Drosophila meiotic DSBs

To monitor the early steps of DSB repair in Drosophila female meiosis, we used an antibody to human γ-H2AX to detect the phosphorylation of HIS2AV (γ-HIS2AV). The Drosophila germarium is favorable material for this analysis because the oocytes are contained within a developmentally ordered array of cysts, allowing successive stages of meiotic prophase to be observed simultaneously(Fig. 1). In wild-type female meiosis, the γ-H2AX antibody detected nuclear foci in early pachytene cells (region 2a and occasionally region 2b of the germarium) (Figs 1, 2). We cannot directly measure the rate of HIS2AV phosphorylation in Drosophila meiotic cells but in somatic cells, it reaches its maximum within 5 minutes of exposure to agents that induce DSBs and is almost completely removed within 3 hours(Madigan et al., 2002). Given that the life-span of γ-HIS2AV foci is likely to be short relative to the developmental age difference between each cyst in the germarium (∼12 hours) (King, 1970), the foci could appear at DSBs that are induced early in pachytene (region 2a) and disappear later in pachytene (regions 2b and 3) because of repair.

Fig. 1.

γ-HIS2AV staining in wild-type female meiosis. (A) Projection of an image stack containing a single germarium stained for ORB (green),γ-HIS2AV (red) and DNA (blue). Before region 2a, four incomplete premeiotic divisions produce a cyst of 16 cells interconnected by ring canals. Because developing cysts are propelled to the posterior end of the germarium(in direction of arrow), they are usually arrayed in order of their relative developmental ages (Carpenter,1975a). The ORB protein (Lantz et al., 1994) was used as a cytoplasmic marker to identify the 16-cell cysts, which are numbered in order of developmental stage. When cysts with two pro-oocytes are initially formed, they are in region 2a (cysts 1-4)and have a round shape. This region contains early zygotene and pachytene oocytes. Several cells within each cyst show evidence of entering meiosis, and two cells, the pro-oocytes, have four ring-canals and reach pachytene with complete assembly of synaptonemal complex (SC)(Carpenter, 1975a). One of these cells will become the oocyte. When the cysts flatten out and become encircled by follicle cells, they are in region 2b (cyst 5 and 6) and 3 (cyst 7). Region 3 oocytes are in mid-late pachytene. ORB becomes enriched in the oocyte in regions 2b and 3 (arrows, cysts 5-7), but initially (more anterior)it appears uniformly in all 16 cells of a cyst (region 2a, cysts 1-4). In projections such as this, it is difficult to determine which nuclei contain foci. It is clear from this view, however, that abundant γ-HIS2AV foci are not observed until cysts 3. C(3)G staining is usually present in the first ORB staining cysts (1 and 2) (Page and Hawley, 2001) (data not shown), suggesting that γ-HIS2AV staining appears after SC formation (see Fig. 2). Single optical sections show γ-HIS2AV foci in nuclei from cyst 3 (B) and cyst 5 (C) but not later stage nuclei from cysts 6 and 7 (D). Arrowheads mark the oocytes. Bar, 10 μm.

Fig. 1.

γ-HIS2AV staining in wild-type female meiosis. (A) Projection of an image stack containing a single germarium stained for ORB (green),γ-HIS2AV (red) and DNA (blue). Before region 2a, four incomplete premeiotic divisions produce a cyst of 16 cells interconnected by ring canals. Because developing cysts are propelled to the posterior end of the germarium(in direction of arrow), they are usually arrayed in order of their relative developmental ages (Carpenter,1975a). The ORB protein (Lantz et al., 1994) was used as a cytoplasmic marker to identify the 16-cell cysts, which are numbered in order of developmental stage. When cysts with two pro-oocytes are initially formed, they are in region 2a (cysts 1-4)and have a round shape. This region contains early zygotene and pachytene oocytes. Several cells within each cyst show evidence of entering meiosis, and two cells, the pro-oocytes, have four ring-canals and reach pachytene with complete assembly of synaptonemal complex (SC)(Carpenter, 1975a). One of these cells will become the oocyte. When the cysts flatten out and become encircled by follicle cells, they are in region 2b (cyst 5 and 6) and 3 (cyst 7). Region 3 oocytes are in mid-late pachytene. ORB becomes enriched in the oocyte in regions 2b and 3 (arrows, cysts 5-7), but initially (more anterior)it appears uniformly in all 16 cells of a cyst (region 2a, cysts 1-4). In projections such as this, it is difficult to determine which nuclei contain foci. It is clear from this view, however, that abundant γ-HIS2AV foci are not observed until cysts 3. C(3)G staining is usually present in the first ORB staining cysts (1 and 2) (Page and Hawley, 2001) (data not shown), suggesting that γ-HIS2AV staining appears after SC formation (see Fig. 2). Single optical sections show γ-HIS2AV foci in nuclei from cyst 3 (B) and cyst 5 (C) but not later stage nuclei from cysts 6 and 7 (D). Arrowheads mark the oocytes. Bar, 10 μm.

Fig. 2.

γ-HIS2AV staining in wild-type female meiosis appears after the initiation of SC formation. All images are single sections, withγ-HIS2AV (red), C(3)G (green) and DNA (blue) staining. (A) An early region 2a pro-oocyte in zygotene with no nuclear γ-HIS2AV staining. (B)A pachytene nucleus with five γ-HIS2AV foci associated with the C(3)G-staining. A color merge and the γ-HIS2AV channel are shown. (C) A pro-nurse cell with γ-HIS2AV foci and fragments of C(3)G staining at the same developmental stage as the cell in B. The three γ-HIS2AV foci are all associated with segments of C(3)G staining. Next to this cell and in the same cyst is a pro-oocyte in pachytene. Bars, 1 μm.

Fig. 2.

γ-HIS2AV staining in wild-type female meiosis appears after the initiation of SC formation. All images are single sections, withγ-HIS2AV (red), C(3)G (green) and DNA (blue) staining. (A) An early region 2a pro-oocyte in zygotene with no nuclear γ-HIS2AV staining. (B)A pachytene nucleus with five γ-HIS2AV foci associated with the C(3)G-staining. A color merge and the γ-HIS2AV channel are shown. (C) A pro-nurse cell with γ-HIS2AV foci and fragments of C(3)G staining at the same developmental stage as the cell in B. The three γ-HIS2AV foci are all associated with segments of C(3)G staining. Next to this cell and in the same cyst is a pro-oocyte in pachytene. Bars, 1 μm.

Consistent with the hypothesis that the foci disappear because of DSB repair, a dramatic change in the γ-HIS2AV staining pattern was observed in spnB (a meiosis-specific Rad51 homolog) and okr (a Rad54 homolog) mutants, which are proposed to be defective in DSB repair(Ghabrial et al., 1998). Although the wild-type γ-HIS2AV foci were limited to early pachytene(region 2a and occasionally 2b, Fig. 1B,C) and were always absent from late pachytene oocytes (region 3, Fig. 1D, Table 2), the okrWS and spnBBU mutant germaria always exhibited foci in late pachytene (region 3 and early vitellarium)(Fig. 3A-D, Fig. 4). In addition, the number of γ-HIS2AV foci in region 3 oocytes of the DSB repair-defective mutants was consistent and usually higher than in wildtype, often in excess of 20 γ-HIS2AV foci (average 21.1, Table 1). The foci persisted in the okr and spnBmutants until stage 4 of the vitellarium, at which point they disappeared,just before the dissolution of the SC (Fig. 5). The persistent and numerous foci in these mutants may result from the failure to efficiently repair DSBs. Indeed, the γ-HIS2AV staining in okrWS and spnBBU mutants gradually became brighter until region 3, suggesting that the γ-HIS2AV was accumulating (Fig. 3C,D).

Table 2.

Quantitation of γ-HIS2AV foci in wild-type cysts

Germarium*
RegionA-1A-2B-2C-1C-3
2a 0/0 5/5 0/0 0/0 0/0 
 16/17 7/9 4/1 1/2 0/0 
 10/9 9/8 6/3 8/12 4/0 
  4/4 24/26 3/0 9/7 
    15/16 9/10 
2b 2/4 0/0 6/6 4/9 4/2 
 0/0 0/0 2/1 0/0 3/4 
 0/0     
0/0 0/0 0/0 0/0 0/0 
Germarium*
RegionA-1A-2B-2C-1C-3
2a 0/0 5/5 0/0 0/0 0/0 
 16/17 7/9 4/1 1/2 0/0 
 10/9 9/8 6/3 8/12 4/0 
  4/4 24/26 3/0 9/7 
    15/16 9/10 
2b 2/4 0/0 6/6 4/9 4/2 
 0/0 0/0 2/1 0/0 3/4 
 0/0     
0/0 0/0 0/0 0/0 0/0 

Average number of C(3)G staining cysts/germaria with >1 foci=3.8

Each vertical column represents a staged series of cysts from a single germarium. The cysts are listed from the earliest stage with C(3)G staining to the last stage in the germarium (region 3). The two numbers separated by a slash are the numbers of γ-HIS2AV foci in the two pro-oocytes in each cyst, identified on the basis of C(3)G staining. Cyst 1 is the first where C(3)G staining could be observed

*

Each germarium is labelled according to the experiment number (first letter) and a unique identifier given to each germarium imaged (second number). Thus, these germaria come from three different experiments

Fig. 3.

Increased γ-HIS2AV in late pachytene oocytes of DSB repair defective mutants. All oocytes were stained for γ-HIS2AV (red), ORB (green) and DNA (blue) and, unless otherwise noted, were in region 3 (late pachytene). (A)γ-HIS2AV foci are rare in region 2a nuclei (a lower magnification image to show more nuclei) and (B) abundant in a region 3 oocyte from a spnBBU female. (C) Region 2b and (D) region 3 oocytes from the same okrWS mutant germarium, showing that the foci become brighter in more posterior oocytes, which are in a later stage of pachytene. In both spnB and okr mutants, large numbers of foci were present in every region 3 oocyte examined (n>20 for each genotype). By contrast, nuclear γ-HIS2AV foci were absent in region 3 oocytes from (E) His2AvΔCT; spnBBUHis2Av810 (F) and okrWSmei-W681 females. (G) Nuclear γ-HIS2AV foci were present in okrWS; c(3)G68 double mutants, suggesting that a few DSBs are made in the absence of SC. (H) In a region 3 oocyte from an okrZ0682 mutant female, γ-HIS2AV foci are present, but less abundant than in an okrWS mutant. All images are maximum projections of the sections needed to view the entire nucleus. Bar, 5 μm; images B-H are approximately the same scale.

Fig. 3.

Increased γ-HIS2AV in late pachytene oocytes of DSB repair defective mutants. All oocytes were stained for γ-HIS2AV (red), ORB (green) and DNA (blue) and, unless otherwise noted, were in region 3 (late pachytene). (A)γ-HIS2AV foci are rare in region 2a nuclei (a lower magnification image to show more nuclei) and (B) abundant in a region 3 oocyte from a spnBBU female. (C) Region 2b and (D) region 3 oocytes from the same okrWS mutant germarium, showing that the foci become brighter in more posterior oocytes, which are in a later stage of pachytene. In both spnB and okr mutants, large numbers of foci were present in every region 3 oocyte examined (n>20 for each genotype). By contrast, nuclear γ-HIS2AV foci were absent in region 3 oocytes from (E) His2AvΔCT; spnBBUHis2Av810 (F) and okrWSmei-W681 females. (G) Nuclear γ-HIS2AV foci were present in okrWS; c(3)G68 double mutants, suggesting that a few DSBs are made in the absence of SC. (H) In a region 3 oocyte from an okrZ0682 mutant female, γ-HIS2AV foci are present, but less abundant than in an okrWS mutant. All images are maximum projections of the sections needed to view the entire nucleus. Bar, 5 μm; images B-H are approximately the same scale.

Fig. 4.

Defects in DSB repair cause a delay in γ-HIS2AV staining. Wildtype and DSB repair mutant germaria were stained for γ-HIS2AV (red), DNA(blue) and C(3)G (green). Also shown are the single channels forγ-HIS2AV and C(3)G staining. (A) A region 3 oocyte from a wild-type female that stains for C(3)G but not for γ-HIS2AV foci. (B) A region 3 cyst of an okrWS mutant germarium showing γ-HIS2AV foci in the oocyte (with extensive C(3)G staining, arrow) and nurse cells(arrowheads). (C) Pro-oocytes in region 2a and 2b cysts of the a spnBBU mutant germarium. In wildtype, these stages would normally have the maximum number of γ-H2AX foci for the germarium. Instead, these pro-oocytes have just begun to form foci associated with the SC(arrows). (D) A Region 3 cyst of the spnBBU mutant shown in (c) that has a large number of foci in the oocyte (arrow) and nurse cells (arrowheads). In this germarium, the previous cyst (region 2b, not shown) was the first one to strongly stain for γ-HIS2AV and the sixth in the germarium that stained with C(3)G (see C). As in Fig. 3, all images are maximum projections. Bar, 5 μm.

Fig. 4.

Defects in DSB repair cause a delay in γ-HIS2AV staining. Wildtype and DSB repair mutant germaria were stained for γ-HIS2AV (red), DNA(blue) and C(3)G (green). Also shown are the single channels forγ-HIS2AV and C(3)G staining. (A) A region 3 oocyte from a wild-type female that stains for C(3)G but not for γ-HIS2AV foci. (B) A region 3 cyst of an okrWS mutant germarium showing γ-HIS2AV foci in the oocyte (with extensive C(3)G staining, arrow) and nurse cells(arrowheads). (C) Pro-oocytes in region 2a and 2b cysts of the a spnBBU mutant germarium. In wildtype, these stages would normally have the maximum number of γ-H2AX foci for the germarium. Instead, these pro-oocytes have just begun to form foci associated with the SC(arrows). (D) A Region 3 cyst of the spnBBU mutant shown in (c) that has a large number of foci in the oocyte (arrow) and nurse cells (arrowheads). In this germarium, the previous cyst (region 2b, not shown) was the first one to strongly stain for γ-HIS2AV and the sixth in the germarium that stained with C(3)G (see C). As in Fig. 3, all images are maximum projections. Bar, 5 μm.

Table 1.

Summary of γ-HIS2AV staining in wildtype and meiotic mutants

Average number of foci in:
GenotypeRegion 2Region 3n (cells)
+/+* 12.3 18 
okrWS 21.1 10 
spnBBU 24.3 
His2AvΔCT spnBBU  
mei-W684572  
okrWS mei-W68  
c(3)G68 2.2 14 
okrWS c(3)G68 5.4 14 
Average number of foci in:
GenotypeRegion 2Region 3n (cells)
+/+* 12.3 18 
okrWS 21.1 10 
spnBBU 24.3 
His2AvΔCT spnBBU  
mei-W684572  
okrWS mei-W68  
c(3)G68 2.2 14 
okrWS c(3)G68 5.4 14 

W, weak staining, not counted. N, no staining

*

Data from Table 2. Average of the pro-oocytes with the largest numbers of foci; two pro-oocytes from two cysts in each germarium were counted, except only one was counted from B-2. As discussed in the text, for wildtype these are probably underestimates of the DSB total

Fig. 5.

γ-HIS2AV foci persist into late stages of meiotic prophase in okr and spnB mutants. Following region 3 in the germarium,each cyst enters the vitellarium, which is divided into several developmental stages (numbered 2-14). The meiotic nucleus condenses into a tight mass, the karyosome, meiosis arrests and the SC dissolves from the nucleus by stage 6. The remaining cells become polyploid nurse cells. The oocyte (arrows) is marked by the ORB protein (green), DNA is stained blue and the γ-HIS2AV foci are red. (A) A merge of five confocal sections from an okrWS vitellarium showing foci in a stage 3 oocyte and their absence at stage 5. (B) A merge of three confocal sections from a spnBBU vitellarium. The γ-HIS2AV foci are present in a stage 4 cyst but absent in the stage 6 cyst (see inset). Bar, 10 μm.

Fig. 5.

γ-HIS2AV foci persist into late stages of meiotic prophase in okr and spnB mutants. Following region 3 in the germarium,each cyst enters the vitellarium, which is divided into several developmental stages (numbered 2-14). The meiotic nucleus condenses into a tight mass, the karyosome, meiosis arrests and the SC dissolves from the nucleus by stage 6. The remaining cells become polyploid nurse cells. The oocyte (arrows) is marked by the ORB protein (green), DNA is stained blue and the γ-HIS2AV foci are red. (A) A merge of five confocal sections from an okrWS vitellarium showing foci in a stage 3 oocyte and their absence at stage 5. (B) A merge of three confocal sections from a spnBBU vitellarium. The γ-HIS2AV foci are present in a stage 4 cyst but absent in the stage 6 cyst (see inset). Bar, 10 μm.

To determine whether the foci detected by the γ-H2AX antibody were dependent on phosphorylation of HIS2AV, we stained germaria from females carrying a mutant form of the protein that lacks the phosphorylation site. Homozygotes for the null allele His2Av810 are lethal(van Daal and Elgin, 1992). However, when this homozygote is coupled with a transgene expressing His2AvΔCT, which lacks the last 14 amino acids inclusive of the phosphorylation site, the flies are viable and fertile(Clarkson et al., 1999). When the His2AvΔCT; His2Av810 females were stained with the γ-H2AX antibody, the nuclear foci were absent (data not shown). Because the spnBBU mutation causes an increase ofγ-HIS2AV foci in late pachytene oocytes, we used it as a tool to confirm the absence of foci in the His2AvΔCT mutant. Indeed, His2AvΔCT; spnBBUHis2Av810 females exhibited no nuclear foci of γ-HIS2AV staining (Fig. 3E). The dependence of staining on the C-terminal tail of the His2Av gene is consistent with the conclusion that the γ-H2AX antibody detects the phosphorylated form of HIS2AV.

To determine whether the foci of γ-HIS2AV were dependent on DSB formation, we examined staining patterns in mei-W68 mutant oocytes,where meiotic recombination does not occur(McKim et al., 1998). We did not observe nuclear γ-HIS2AV staining in either a mei-W68single mutant (compare with wildtype where foci were normally observed early in pachytene), or okrWS mei-W681 and mei-W684572; spnBBU double mutants (compare with okrWS or spnBBU single mutants where a large number of γ-HIS2AV foci are normally observed in late pachytene oocytes) (Fig. 3D,F). The dependence of staining on mei-W68 shows that the γ-H2AX antibody is detecting a response to DSBs in wild-type pachytene. In conjunction with western blot analysis showing that HIS2AV is phosphorylated when DSBs are induced (Madigan et al.,2002), our results suggest that Drosophila HIS2AV is phosphorylated in response to meiotic DSBs.

Relationship of the SC to γ-HIS2AV formation

To correlate DSB repair with the prophase events of synapsis and SC formation, we used an antibody to C(3)G, which is proposed to be a structural component of the transverse elements (Page and Hawley, 2001). Formation of the SC occurs early in region 2a(Carpenter, 1979), and C(3)G staining is visible in pro-oocytes of region 2a cysts at the same time or earlier than ORB staining (Page and Hawley, 2001). γ-HIS2AV foci were only observed in pro-oocytes with extensive C(3)G staining, indicative of pachytene. In some germaria, earlier stage (more anterior) pro-oocytes had patches of C(3)G staining, indicative of zygotene where the chromosomes have not fully synapsed. These pro-oocytes lacked γ-HIS2AV foci(Fig. 2, Table 2). The appearance ofγ-HIS2AV foci only after the appearance of C(3)G staining suggests that DSBs are not phosphorylated until after SC formation. As in the experiments using ORB staining, the γ-HIS2AV foci were consistently absent from region 3 oocytes (Fig. 4A, n=13, and Table 2),suggesting the completion of an important step in DSB repair by this stage.

Carpenter previously described the distribution of early recombination nodules (RN) in Drosophila, which are predicted to be protein complexes situated at the sites of DSB repair(Anderson et al., 1997; Bishop, 1994; Carpenter, 1979; Plug et al., 1996; Tarsounas et al., 1999; Terasawa et al., 1995). Theγ-HIS2AV foci and early RN patterns are similar, both appearing at the same stages of cyst development. Both the γ-HIS2AV foci and early RNs usually appear in three to six consecutive cysts, first appearing in small numbers, then increasing in the following cyst or two, and then decreasing in number again before disappearing altogether(Table 2). For example, in germarium B-2, the cyst with the most foci in the pro-oocytes (the fourth) had two cysts before it and two cysts after it with fewer foci. A similar pattern was also observed for early RNs(Carpenter, 1979). Considering that the cysts are usually arranged in temporal order, this pattern probably reflects that, as was suggested for early RNs, the appearance and disappearance of γ-HIS2AV foci is asynchronous. Asynchrony in appearance of γ-HIS2AV foci means that the average number observed per cell in wildtype is probably less than the total number of DSB sites. Exceptions to the temporal ordering of cysts do occur, as noted by Carpenter(Carpenter, 1979), such as the fourth cyst in germarium C-1, which is probably out of order.

Complete SC formation (pachytene) was not necessary for γ-HIS2AV foci to be observed. Although the most numerous foci were observed in the two pro-oocytes with thread-like C(3)G staining(Fig. 2B), foci were also observed in nurse cells, which have only short segments of C(3)G staining(Fig. 2C). For example, cyst 3 in the germarium shown in Fig. 1A had one cell with 16 foci (some shown in the single section of Fig. 1B) and at least five nuclei with four or less foci. Cyst 4 had nuclei with 17 and 19 foci but also five nuclei with five or less foci. Thus, the γ-HIS2AV foci are not uniformly distributed among the nuclei in each cell of a 16-cell cyst, but instead the largest numbers were found in the pro-oocytes. That the largest number of γ-HIS2AV foci are in the pro-oocytes parallels Carpenter's(Carpenter, 1979) observation that there is a gradient of SC formation; the pro-oocytes form complete SC and the adjacent nurse cells form progressively less SC. Similarly, in spnB and okr mutants, it was clear that several cells in a 16-cell cyst stained with γ-HIS2AV (e.g. Fig. 4B,D). These data indicate that DSBs are induced in the nurse cells that do not complete SC formation, in agreement with previous studies showing early RNs(Carpenter, 1975b) and MEI-P22 localization, a protein required for DSBs(Liu et al., 2002), in nurse cell nuclei.

DSB formation occurs in the absence of C(3)G

To test directly the dependence of DSB formation on the SC, we examinedγ-HIS2AV staining in c(3)G68 mutants, which lack SC. Surprisingly, a small number of γ-HIS2AV foci were consistently observed in early prophase (region 2a) of c(3)G68 mutant germaria(data not shown). No more than three foci were counted in a nucleus (average 2.2, Table 1) but, due to the absence SC staining, we could not determine whether these cells were pro-oocytes. To enhance the detection of γ-HIS2AV foci, we constructed okrWS; c(3)G68 double mutant females with the expectation that any DSBs would persist and be visible by antibody staining in late prophase (region 3) oocytes. This experiment confirmed the presence ofγ-HIS2AV foci in c(3)G68 mutant oocytes; however, in comparison to the okr single mutant (average 21.1, see above), there were clearly fewer foci in region 2b or 3 oocytes (average 5.4, Table 1; compare Fig. 3D and 3G). On the basis of these observations, DSBs appear to be induced in a c(3)G mutant,but possibly less frequently than in wildtype (see Discussion).

Phosphorylation of HIS2AV is delayed in DSB repair mutants okr and spnB

The onset of SC formation consistently appears at the same time (early in region 2a of the germarium) in both wildtype and the DSB-repair defective mutants (data not shown). Furthermore, the SC disappears at the same developmental time point as wildtype, vitellarium stage 5-6, suggesting there are no delays in the cell cycle of DSB repair-defective mutants. However,there was a slight delay in the onset of γ-HIS2AV staining in the DSB repair mutants relative to wildtype. In wildtype, γ-HIS2AV foci were observed in the first or second cyst with C(3)G staining(Table 2). In all 13 okrWS and eight spnBBU mutant germaria examined, γ-HIS2AV foci were not observed until the third or fourth C(3)G staining cysts (Figs 3, 4).

Comparing the defects of SC-deficient and DSB repair-defective mutants

The few γ-HIS2AV foci in a c(3)G mutant were observed in the same developmental stages as in wildtype, present in regions 2a and 2b but absent in region 3. The c(3)G pattern of fewer foci appearing and disappearing in an otherwise normal time course is a different phenomenon than that observed with DSB-repair defective mutants. In okr and spnB mutants, we suggest that all DSB sites are phosphorylated but in a time course different from wildtype. The absence of SC does not appear to delay the phosphorylation of HIS2AV as in the okr and spnBmutants, and the smaller number of foci in c(3)G mutants probably reflects a reduction in DSB formation. This conclusion also makes it likely that, in wildtype, the appearance of foci after SC formation reflects when DSBs are induced.

The failure to repair DSBs does not induce sterility in a weak allele of okr

Mutations in DSB repair genes like okr and spnB confer sterility as a consequence of a checkpoint-like mechanism that is triggered by the presence of unrepaired DSBs (Ghabrial and Schupbach, 1999). Surprisingly, c(3)G mutants suppress okr mutants (Ghabrial and Schupbach, 1999), despite our evidence for DSBs in the okrWS; c(3)G68 double mutant. One explanation for this result is that the low number of unrepaired DSBs (implied from the low number of observed foci) is not sufficient to activate the checkpoint mechanism. We tested this hypothesis by examining γ-HIS2AV staining in a hypomorphic allele of okr, okrZ0682, which causes a reduction in meiotic crossing over (to 30% of wildtype) but not the sterility associated with amorphic okr alleles (D.S. and K.M., unpublished). In all okrZ0682 late pachytene (e.g. region 3) and vitellarium oocytes that were examined (n=15 ovarioles),γ-HIS2AV foci were present (Fig. 3H). The presence of staining indicated that there were unrepaired DSBs in late pachytene oocytes but their presence was not sufficient to cause sterility. There were, however, obviously fewer γ-HIS2AV foci in region 3 okrZ0682 oocytes than in okrWS or spnBBU mutants (compare Fig. 3D and 3H). Because okrZ0682 is a hypomorph, the reduced number of foci relative to a null allele may indicate that a significant number of breaks are repaired before region 3. We suggest that okrZ0682 females are fertile and c(3)G mutants suppress strong okr mutants for the same reason. In both cases there is a low number of unrepaired DSBs and these are not sufficient to trigger the cell-cycle response that leads to sterility.

Deleting the HIS2AV phosphorylation site has negligible effects on meiotic recombination

We performed genetic tests on the His2AvΔCT;His2Av810 females to investigate the consequences of failing to phosphorylate HIS2AV following meiotic DSB formation. The frequency of X-chromosome nondisjunction was not significantly elevated relative to wildtype (0.2%, n=954). Elevated levels of X-chromosome nondisjunction correlate with a decrease in meiotic crossing over(Hawley, 1988); therefore,these results suggest that there are no defects in repairing meiotic DSBs as crossovers. This was confirmed by direct analysis of crossing over: the frequency of 2nd chromosome crossing over in His2AvΔCT;His2Av810 females within the aldp(12.4%) and dpb (24.3%, n=420) intervals was similar to controls (al – dp=13.5%; dp –b=25.5%). The results of these two experiments indicate that the lack ofγ-HIS2AV is not a serious detriment to meiotic DSB repair.

Discussion

The foci detected by the γ-H2AX antibody in Drosophilaoocytes are dependent on MEI-W68, the Spo11 homolog presumed to generate meiotic DSBs (Keeney et al.,1997; McKim and Hayashi-Hagihara, 1998), and HIS2AV, the histone H2A variant that is phosphorylated in mitotic cells in response to DSB formation. Thus, it is likely that Drosophila HIS2AV is phosphorylated at DSB sites created during meiosis. Although phosphorylation of an H2A variant in response to DSBs occurs in many organisms and cell types, its contribution to DSB repair may be specialized. Madigan et al. (Madigan et al., 2002) found that His2AvΔCT mutants are not sensitive to X-rays and we found that Drosophila females lacking γ-HIS2AV have normal meiotic recombination and fertility,suggesting no defects in meiotic DSB repair. Similarly, H2AX mutant female mice are fertile, consistent with no gross defects in meiotic DSB repair (Celeste et al., 2002). It has been shown, however, that phosphorylation is required for some forms of repair: mouse H2AX mutant cells or animals are sensitive to ionizing radiation, have elevated levels of genomic instability and are immune deficient (Bassing et al.,2002; Celeste et al.,2002). In both S. cerevisiae(Downs et al., 2000) and Drosophila (D.S. and K.M., unpublished), eliminating H2A variant phosphorylation causes sensitivity to the mutagen MMS and reduces the incidence of damage-induced apoptosis(Madigan et al., 2002). In summary, γ-H2AX or γ-HIS2AV probably accumulate at all DSBs but have a relatively specialized function, such as contributing to nonhomologous end joining (Downs et al.,2000) or a cell-cycle checkpoint(Fernandez-Capetillo et al.,2002).

The rapid onset of the phosphorylation makes γ-HIS2AV a useful cytological marker for DSB formation(Madigan et al., 2002). However, there are two important caveats to interpreting the cytological data. First, the rate at which HIS2AV is phosphorylated during Drosophilameiosis is not known. We also expect mutations that directly or indirectly affect the activity of relevant kinases such as the Drosophila ATM homolog (Burma et al., 2001)would prevent γ-HIS2AV from appearing, even though DSBs are present. Second, in meiotic cells we do not know that the removal of phosphorylation always corresponds to repair of the break. However, we have no reason to suspect that the relationship between phosphorylation and the presence of a DSB is different than mitotic cells. Furthermore, phosphorylation and dephosphorylation of H2AX appears to be rapid in mouse meiotic cells(Mahadevaiah et al., 2001). Thus, we favor the simplest interpretation of the cytological data; the appearance of γ-HIS2AV closely reflects the presence or absence of DSBs.

In wild-type Drosophila females, the number of foci observed in the pro-oocytes is probably less than the total number of DSBs because of asynchrony in either DSB formation or phosphorylation. In the DSB repair-defective mutants, however, the total number of foci could be close to the number of DSBs if most or all of the breaks are not repaired. In fact, the average number of foci in a okr null mutant (21.1) is close to the number of initiation events predicted from genetic data. Chovnick and colleagues estimated that only 20% of gene conversion events at ryare associated with a crossover (Hilliker et al., 1988). Taking into account that the rosy locus is in a relatively crossover-depressed region of the genome(McKim et al., 2002), we predict from the genetic data that there are three or four initiation events per chromosome arm or 15-20 per nucleus.

Does DSB formation occur after SC formation?

Genetic studies in Drosophila showed that meiotic recombination is not required for SC formation (McKim et al., 1998), but did not rule out the possibility that the two events are independent. γ-HIS2AV foci are absent at zygotene (short segments of C(3)G in the oocyte) but are present at early pachytene (once C(3)G accumulates into filaments), supporting the model that, in wildtype, DSB formation does not occur until after SC formation. This is the same relationship between the SC and early RNs in Drosophila(Carpenter, 1979). We have not, however, ruled out the possibility that DSBs are initiated earlier but phosphorylation occurs after SC formation. This is an unlikely scenario for two reasons. First, SC formation, or C(3)G accumulation in particular, appears to stimulate DSB formation (see below). Second, while the absence of SC reduces the number of foci, it does not prevent or delay HIS2AV phosphorylation. A direct comparison to our approach of using a phosphorylated histone to monitor DSB formation is a similar study of male mouse meiosis(Mahadevaiah et al., 2001). In contrast to our results, but like S. cerevisiae, mouse meiotic DSB formation occurs before and is required for SC formation(Baudat et al., 2000; Roeder, 1997; Romanienko and Camerini-Otero,2000).

The observation of γ-HIS2AV foci in c(3)G mutant females is the first evidence for SC-independent DSB formation in Drosophila. This result was unexpected because gene conversion(Carlson, 1972) and crossing over (Hall, 1972) have been reported to be virtually eliminated in c(3)G mutants. However, the data from the gene conversion study were also consistent with a reduction in DSB frequency, which is consistent with our observation that the number ofγ-HIS2AV foci is reduced in c(3)G mutant females (10-25% of wildtype). Therefore, the presence of SC probably has a significant stimulatory role in DSB formation. The absence of crossing over, despite our evidence that DSBs form in c(3)G mutants, can also be explained;genetic evidence suggests that c(3)G is required for repairing DSBs as crossovers (Roberts, 1969)(R.B. and K.M., unpublished). Thus, even if DSBs are present in c(3)Gmutants (either endogenous or from X-irradiation), they cannot be repaired as crossovers. Surprisingly, the dependence of meiotic DSB formation and crossing over on a central element component is not a conserved feature in all organisms. In mutants of the S. cerevisiae c(3)G homolog zip1, DSB formation occurs at a normal frequency and the production of crossovers is only mildly reduced (Sym et al., 1993; Xu et al.,1997). The effects of c(3)G mutants on DSB formation are more reminiscent of mutations in the S. cerevisiae axial element components hop1 and red1, where the frequency of DSB formation is reduced to 10-12% and 25-47%, respectively, of wild-type levels(Schwacha and Kleckner, 1997; Woltering et al., 2000; Xu et al., 1997).

DSB repair is delayed, but not eliminated, in the absence of spnB and okr

The simplest hypothesis to explain the disappearance of γ-HIS2AV foci in wild-type oocytes before mid-pachytene is that a crucial step in DSB repair has been completed. The presence of γ-HIS2AV foci at a later stage of prophase in okr and spnB mutants, in which DSB repair is defective, supports the conclusion that the disappearance of the foci corresponds to a step in DSB repair. Similar to the conclusions drawn from the physical studies of DSB sites in S. cerevisiae rad51 and dmc1 mutants (Bishop et al.,1992; Shinohara et al.,1992), DSBs probably persist longer in okr and spnB mutants. The γ-HIS2AV foci in spnB and okr mutants disappear at a consistent developmental time point,vitellarium stage 4, just before dissolution of the SC. One explanation for the late disappearance of foci is that an alternative mechanism to repair the DSBs in spnB and okr mutants becomes available late in meiotic prophase. One reason we favor this explanation is that the high fertility of okrZ0682 mutants, which also haveγ-HIS2AV foci that persist until stage 4 oocytes, can only be explained if all of the DSBs are repaired. Furthermore, a similar observation was made in S. cerevisiae rad51 mutants, where the repair of DSBs was delayed but did eventually occur (Shinohara et al., 1992). Nonetheless, we have not proven that the disappearance of foci in spnB and okr mutants corresponds precisely to the repair of DSBs. Thus, another, more complex, explanation for these observations is that the HIS2AV phosphorylation is removed even though the DSB has not been repaired.

Phosphorylation of H2AX in the mouse is one of the earliest responses to DSB formation in the mouse, and has been observed before accumulation of the DSB repair proteins Rad50, Rad51 and Dmc1(Mahadevaiah et al., 2001; Paull et al., 2000). However,the relationship between the phosphorylation of H2A variants and the recruitment of DSB repair proteins may be complex. H2AX is required for Brca1,Nbs1 and 53BP1 (Bassing et al.,2002; Celeste et al.,2002; Fernandez-Capetillo et al., 2002) but not Rad51 foci formation(Celeste et al., 2002). Our observation that the onset of γ-HIS2AV foci is delayed in okrand spnB mutants suggests that phosphorylation of DrosophilaHIS2AV is influenced by some DNA repair proteins. We have not, however, ruled out the possibility that okr and spnB mutants have a delay in the DSB formation, although such an activity for these proteins has not previously been described.

The abnormal development of the oocyte in spnB and okrmutants is proposed to be caused by a checkpoint response to unrepaired DSBs;there is a failure to translate gurken with subsequent defects in dorsal-ventral (D-V) polarity of the oocyte(Ghabrial and Schupbach,1999). Our γ-HIS2AV experiments revealed this DSB repair defect in spnB and okr mutants. However, the γ-HIS2AV foci disappear in spnB and okr mutants before the D-V phenotype is visible. If indeed the disappearance of foci corresponds to DSB repair, cell cycle events in the germarium or early in the vitellarium, such as a checkpoint response, may not have developmental effects until later in oogenesis. Interestingly, Abdu et al. (Abdu et al., 2002) suggested that the cell cycle and developmental effects reflect a divergence in the checkpoint pathway. It is possible that one pathway regulates the DNA repair functions, allowing repair by a relatively early stage (vitellarium stage 4), whereas the second pathway has effects on oocyte patterning later in oocyte development.

Acknowledgements

We are grateful to Li Nguyen for technical assistance and Cordelia Rauskolb and Ruth Steward for providing insightful discussions and comments on the manuscript. We also thank Scott Page and Scott Hawley for providing the C(3)G antibody, Trudi Schupach for spnB and okr alleles and Robert Glaser for the His2Av mutant stocks. Additional stocks used in this study were received from the Bloomington Stock Center. The ORB antibodies were obtained from the Developmental Studies Hybridoma Bank at the University of Iowa. An NIH Biotechnology Training Grant and Charles and Johanna Busch fellowship to E.A.M. and a grant from the National Science Foundation(MCB-0077705) to K.S.M. supported this work.

References

Abdu, U., Brodsky, M. and Schupbach, T. (
2002
). Activation of a meiotic checkpoint during Drosophila oogenesis regulates the translation of Gurken through Chk2/Mnk.
Curr. Biol.
12
,
1645
-1651.
Anderson, L. K., Offenberg, H. H., Verkuiljen, W. M. H. C. and Heyting, C. (
1997
). RecA-like proteins are components of early meiotic nodules in lily.
Proc. Natl. Acad. Sci. USA
94
,
6868
-6873.
Bassing, C. H., Chua, K. F., Sekiguchi, J., Suh, H., Whitlow, S. R., Fleming, J. C., Monroe, B. C., Ciccone, D. N., Yan, C., Vlasakova, K. et al. (
2002
). Increased ionizing radiation sensitivity and genomic instability in the absence of histone H2AX.
Proc. Natl. Acad. Sci. USA
99
,
8173
-8178.
Baudat, F., Manova, K., Yuen, J. P., Jasin, M. and Keeney,S. (
2000
). Chromosome synapsis defects and sexually dimorphic meiotic progression in mice lacking Spo11.
Mol. Cell
6
,
989
-998.
Belmont, A. S., Braunfeld, M. B., Sedat, J. W. and Agard, D. A. (
1989
). Large-scale chromatin structural domains within mitotic and interphase chromosomes in vivo and in vitro.
Chromosoma
98
,
129
-143.
Bishop, D. K. (
1994
). RecA homologs Dmc1 and Rad51 interact to form multiple nuclear complexes prior to meiotic chromosome synapsis.
Cell
79
,
1081
-1092.
Bishop, D. K., Park, D., Xu, L. and Kleckner, N.(
1992
). DMC1: A meiosis-specific yeast homolog of E. coli recA required for recombination, synaptonemal complex formation, and cell cycle progression.
Cell
69
,
439
-456.
Burma, S., Chen, B. P., Murphy, M., Kurimasa, A. and Chen, D. J. (
2001
). ATM phosphorylates histone H2AX in response to DNA double-strand breaks.
J. Biol. Chem.
276
,
42462
-42467.
Carlson, P. S. (
1972
). The effects of inversions and the c(3)G mutation on intragenic recombination in Drosophila.
Genet. Res. Camb.
19
,
129
-132.
Carpenter, A. T. C. (
1975a
). Electron microscopy of meiosis in Drosophila melanogaster females. I. Structure, arrangement, and temporal change of the synaptonemal complex in wild-type.
Chromosoma
51
,
157
-182.
Carpenter, A. T. C. (
1975b
). Electron microscopy of meiosis in Drosophila melanogaster females. II. The recombination nodule – a recombination-associated structure at pachytene?
Proc. Natl. Acad. Sci. USA
72
,
3186
-3189.
Carpenter, A. T. C. (
1979
). Synaptonemal complex and recombination nodules in wild-type Drosophila melanogaster females.
Genetics
92
,
511
-541.
Celeste, A., Petersen, S., Romanienko, P. J.,Fernandez-Capetillo, O., Chen, H. T., Sedelnikova, O. A., Reina-San-Martin,B., Coppola, V., Meffre, E., Difilippantonio, M. J. et al.(
2002
). Genomic instability in mice lacking histone H2AX.
Science
296
,
922
-927.
Cervantes, M. D., Farah, J. A. and Smith, G. R.(
2000
). Meiotic DNA breaks associated with recombination in S. pombe.
Mol. Cell
5
,
883
-888.
Clarkson, M. J., Wells, J. R., Gibson, F., Saint, R. and Tremethick, D. J. (
1999
). Regions of variant histone His2AvD required for Drosophila development.
Nature
399
,
694
-697.
Dernburg, A. F., McDonald, K., Moulder, G., Barstead, R.,Dresser, M. and Villeneuve, A. M. (
1998
). Meiotic recombination in C. elegans initiates by a conserved mechanism and is dispensable for homologous chromosome synapsis.
Cell
94
,
387
-398.
Downs, J. A., Lowndes, N. F. and Jackson, S. P.(
2000
). A role for Saccharomyces cerevisiae histone H2A in DNA repair.
Nature
408
,
1001
-1004.
Fernandez-Capetillo, O., Chen, H. T., Celeste, A., Ward, I.,Romanienko, P. J., Morales, J. C., Naka, K., Xia, Z., Camerini-Otero, R. D.,Motoyama, N. et al. (
2002
). DNA damage-induced G(2)-M checkpoint activation by histone H2AX and 53BP1.
Nat. Cell Biol.
4
,
993
-997.
Ghabrial, A. and Schupbach, T. (
1999
). Activation of a meiotic checkpoint regulates translation of Gurken during Drosophila oogenesis.
Nat. Cell Biol.
1
,
354
-357.
Ghabrial, A., Ray, R. P. and Schupbach, T.(
1998
). okra and spindle-B encode components of the RAD52 DNA repair pathway and affect meiosis and patterning in Drosophila oogenesis.
Genes Dev.
12
,
2711
-2723.
Hall, J. C. (
1972
). Chromosome segregation influenced by two alleles of the meiotic mutant c(3)G in Drosophila melanogaster.
Genetics
71
,
367
-400.
Hawley, R. S. (
1988
). Exchange and chromosomal segregation in eucaryotes. In
Genetic Recombination
(eds R. Kucherlapati and G. Smith), pp.
497
-527. Washington, DC: American Society of Microbiology.
Hilliker, A. J., Clark, S. H. and Chovnick, A.(
1988
). Genetic analysis of intragenic recombination in Drosophila. In
The Recombination of Genetic Material
(ed. K. B. Low), pp.
73
-90. New York: Academic Press.
Keeney, S. (
2001
). Mechanism and control of meiotic recombination initiation.
Curr. Top. Dev. Biol.
52
,
1
-53.
Keeney, S., Giroux, C. N. and Kleckner, N.(
1997
). Meiosis-specific DNA double-strand breaks are catalyzed by Spo11, a member of a widely conserved protein family.
Cell
88
,
375
-384.
King, R. C. (
1970
).
Ovarian Development in
Drosophila Melanogaster. New York: Academic Press.
Lantz, V., Chang, J. S., Horabin, J. I., Bopp, D. and Schedl,P. (
1994
). The Drosophila ORB RNA-binding protein is required for the formation of the egg chamber and establishment of polarity.
Genes Dev.
8
,
598
-613.
Liu, H., Jang, J. K., Kato, N. and McKim, K. S.(
2002
). mei-P22 encodes a chromosome-associated protein required for the initiation of meiotic recombination in Drosophila melanogaster.
Genetics
162
,
245
-258.
Madigan, J. P., Chotkowski, H. L. and Glaser, R. L.(
2002
). DNA double-strand break-induced phosphorylation of Drosophila histone variant H2Av helps prevent radiation-induced apoptosis.
Nucleic Acids Res.
30
,
3698
-3705.
Mahadevaiah, S. K., Turner, J. M., Baudat, F., Rogakou, E. P.,de Boer, P., Blanco-Rodriguez, J., Jasin, M., Keeney, S., Bonner, W. M. and Burgoyne, P. S. (
2001
). Recombinational DNA double-strand breaks in mice precede synapsis.
Nat. Genet.
27
,
271
-276.
McKim, K. S., Green-Marroquin, B. L., Sekelsky, J. J., Chin, G.,Steinberg, C., Khodosh, R. and Hawley, R. S. (
1998
). Meiotic synapsis in the absence of recombination.
Science
279
,
876
-878.
McKim, K. S. and Hayashi-Hagihara, A. (
1998
). mei-W68 in Drosophila melanogaster encodes a Spo11 homolog:evidence that the mechanism for initiating meiotic recombination is conserved.
Genes Dev.
12
,
2932
-2942.
McKim, K. S., Jang, J. K. and Manheim, E. A.(
2002
). Meiotic recombination and chromosome segregation in Drosophila females.
Annu. Rev. Genet.
36
,
205
-232.
Padmore, R., Cao, L. and Kleckner, N. (
1991
). Temporal comparison of recombination and synaptonemal complex formation during meiosis in S. cerevisiae.
Cell
66
,
1239
-1256.
Page, S. L. and Hawley, R. S. (
2001
). c(3)G encodes a Drosophila synaptonemal complex protein.
Genes Dev.
15
,
3130
-3143.
Paull, T. T., Rogakou, E. P., Yamazaki, V., Kirchgessner, C. U.,Gellert, M. and Bonner, W. M. (
2000
). A critical role for histone H2AX in recruitment of repair factors to nuclear foci after DNA damage.
Curr. Biol.
10
,
886
-895.
Plug, A. W., Xu, J., Reddy, G., Golub, E. I. and Ashley, T.(
1996
). Presynaptic association of Rad51 protein with selected sites in meiotic chromatin.
Proc. Natl. Acad. Sci. USA
93
,
5920
-5924.
Roberts, P. A. (
1969
). Some components of X ray-induced crossing over in females of Drosophila melanogaster.
Genetics
63
,
387
-404.
Roeder, G. S. (
1997
). Meiotic chromosomes: it takes two to tango.
Genes Dev.
11
,
2600
-2621.
Rogakou, E. P., Boon, C., Redon, C. and Bonner, W. M.(
1999
). Megabase chromatin domains involved in DNA double-strand breaks in vivo.
J. Cell Biol.
146
,
905
-916.
Romanienko, P. J. and Camerini-Otero, R. D.(
2000
). The mouse spo11 gene is required for meiotic chromosome synapsis.
Mol. Cell
6
,
975
-987.
Schwacha, A. and Kleckner, N. (
1997
). Interhomolog bias during meiotic recombination: meiotic functions promote a highly differentiated interhomolog-only pathway.
Cell
90
,
1123
-1135.
Shinohara, A., Ogawa, H. and Ogawa, T. (
1992
). Rad51 protein involved in repair and recombination in S. cerevisiae is a RecA-like protein.
Cell
69
,
457
-470.
Sym, M., Engebrecht, J. and Roeder, G. S.(
1993
). ZIP1 is a synaptonemal complex protein required for meiotic chromosome synapsis.
Cell
72
,
365
-378.
Tarsounas, M., Morita, T., Pearlman, R. E. and Moens, P. B.(
1999
). RAD51 and DMC1 form mixed complexes associated with mouse meiotic chromosome cores and synaptonemal complexes.
J. Cell Biol.
147
,
207
-220.
Terasawa, M., Shinohara, A., Hotta, Y., Ogawa, H. and Ogawa,T. (
1995
). Localization of RecA-like recombination proteins on chromosomes of the lily at various meiotic stages.
Genes Dev.
9
,
925
-934.
van Daal, A. and Elgin, S. C. (
1992
). A histone variant, H2AvD, is essential in Drosophila melanogaster.
Mol. Biol. Cell
3
,
593
-602.
Villeneuve, A. M. and Hillers, K. J. (
2001
). Whence meiosis.
Cell
106
,
647
-650.
Woltering, D., Baumgartner, B., Bagchi, S., Larkin, B., Loidl,J., de los Santos, T. and Hollingsworth, N. M. (
2000
). Meiotic segregation, synapsis, and recombination checkpoint functions require physical interaction between the chromosomal proteins Red1p and Hop1p.
Mol. Cell. Biol.
20
,
6646
-6658.
Xu, L., Weiner, B. M. and Kleckner, N. (
1997
). Meiotic cells monitor the status of the interhomolog recombination complex.
Genes Dev.
11
,
106
-118.