Mutational Analysis of the Drosophila DNA Repair and Recombination Gene mei-9

Corresponding author: Jeff J. Sekelsky, 303 Fordham Hall, University of North Carolina, Chapel Hill, NC 27599., sekelsky{at}unc.edu (E-mail)

Communicating editor: K. G. GOLIC

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS AND DISCUSSION LITERATURE CITED

Drosophila mei-9 is essential for several DNA repair and recombination pathways, including nucleotide excision repair (NER), interstrand crosslink repair, and meiotic recombination. To better understand the role of MEI-9 in these processes, we characterized 10 unique mutant alleles of mei-9. These include a P-element insertion that disrupts repair functions but not the meiotic function; three nonsense mutations, one of which has nearly wild-type levels of protein; three missense mutations, one of which disrupts the meiotic function but not repair functions; two small in-frame deletions; and one frameshift.

THE product of the Drosophila melanogaster mei-9 gene plays an essential role in nucleotide excision repair (NER; BOYD et al. 1976B ), the primary pathway for the removal of damage caused by ultraviolet (UV) light (reviewed in SANCAR 1996 ; DE LAAT et al. 1999 ). MEI-9 is the Drosophila ortholog of the human and yeast NER proteins XPF and Rad1p (SEKELSKY et al. 1995 ). XPF and Rad1p function as heterodimers with Ercc1 and Rad10p, respectively (BARDWELL et al. 1993 ; PARK et al. 1995 ). These heterodimers recognize and nick specific structures in which DNA transits from double-stranded to single-stranded regions (BARDWELL et al. 1994 ; HABRAKEN et al. 1994 ; PARK et al. 1995 ; SIJBERS et al. 1996 ; DE LAAT et al. 1998A ). In NER, the strand containing the damage is nicked 5' to the damage. When coupled to a 3' nick made by another endonuclease with opposite polarity (XPG), an oligonucleotide containing the damage is removed, allowing resynthesis using the intact strand as a template. MEI-9 presumably has this same function in Drosophila, and this function likely requires ERCC1, which has been shown to interact strongly with MEI-9 in a yeast two-hybrid assay (SEKELSKY et al. 2000 ).

The NER defect of mei-9 mutants results in extreme hypersensitivity to UV (BOYD et al. 1976B ). In addition, mei-9 mutants are sensitive to ionizing radiation and crosslinking agents (BAKER et al. 1978 ; MASON et al. 1981 ). These hypersensitivities are believed to reveal additional DNA repair pathways in which the mei-9 gene product is required. Sensitivity to ionizing radiation suggests a role in a double-strand break (DSB) repair pathway, and sensitivity to the bifunctional crosslinking agent nitrogen mustard (HN2) suggests a role in the repair of interstrand DNA crosslinks (ICLs). The functions of MEI-9 in these presumed roles in DSB repair and ICL repair are unknown.

In addition to its multiple DNA repair functions, mei-9 plays a key role in the meiotic recombination pathway. The first mei-9 mutations (mei-9^a and mei-9^b) were recovered by BAKER and CARPENTER 1972 in a screen for X-linked mutations that cause high levels of meiotic nondisjunction of the X chromosome. All nondisjunction in mei-9 mutants occurs at the first meiotic division and can be attributed to a severe decrease in meiotic crossing over (BAKER and CARPENTER 1972 ; CARPENTER and SANDLER 1974 ). Unlike other recombination mutants, the 90–95% decrease in crossing over in mei-9 mutants is uniform throughout the genome. This property led Baker and Carpenter to conclude that MEI-9 acts late in the recombination pathway, in the actual process of exchange.

Since homologs of MEI-9 have DNA structure-specific endonuclease activity, we proposed a model in which MEI-9 cuts Holliday junctions in meiotic recombination intermediates (see below for a detailed description of this model). However, there are no biochemical data to support this model. Indeed, it is not even known whether the meiotic function of MEI-9 requires its nuclease activity or its NER partner ERCC1. The meiotic function does require interaction with a novel protein, MUS312 (YILDIZ et al. 2002 ). MUS312 may facilitate the proposed function of MEI-9 in resolving Holliday junction intermediates.

Thus, although the NER function of MEI-9 is understood, the functions of MEI-9 in other repair pathways and in meiotic recombination are less clear. To gain a better understanding of the role of MEI-9 in these pathways, we characterized a set of mei-9 mutations. We determined the sequences of 14 mei-9 alleles and identified 10 unique mutations. We genetically characterized these mutations for their effects on meiotic recombination and DNA repair pathways.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS AND DISCUSSION LITERATURE CITED

Drosophila stocks and genetics:
Genetic loci not described in the text are described in FLYBASE 2001 . Flies were reared on standard medium at 25°.

For all genetic experiments, males carrying the mei-9 allele to be tested (mei-9^X) were crossed to mei-9^A2/FM7 females, or mei-9^A2 males were crossed to mei-9^X/FM7 females. Female progeny were therefore mei-9^X/mei-9^A2 or mei-9/FM7. The former served as the mutant class and the latter as the control class (mei-9 mutations are completely recessive for the phenotypes tested here; BOYD et al. 1976A ). By assaying phenotypes in mei-9^X/mei-9^A2 females, we hoped to minimize the effects of different strain backgrounds or modifier mutations that may have accumulated in the stocks.

X chromosome nondisjunction and crossing over on chromosome 2 were measured as described previously (SEKELSKY et al. 1995 , SEKELSKY et al. 1999 ). For X nondisjunction, mei-9^X/mei-9^A2 females were crossed to C(1;Y)1, v f B/O males. Progeny of normal disjunction are Bar females (mei-9/C(1:Y)1, v f B) and non-Bar males (mei-9/0). Half of the diplo-X progeny survive as non-Bar females (mei-9^X/mei-9^A2) and half die (mei-9^X/mei-9^A2/C(1;Y)1, v f B). Similarly, half of the nullo-X progeny survive as Bar males (C(1;Y), v f B/O) and half die (0/0). The number of exceptional progeny is calculated as 2 x (non-Bar females + Bar males). Because this assay requires survival to adulthood, progeny resulting from nondisjunction of chromosome 2 or 3 are not counted. Thus, the percentage of nondisjunction is among progeny that live to adulthood.

The sensitivity of developing larvae to DNA-damaging agents was assessed as in BOYD et al. 1976A . For HN2, adults were crossed in plastic vials and removed after 2 days of egg laying. After 1 additional day, 250 µl of 0.008% of the nitrogen mustard mechloramine (Sigma, St. Louis) in water was added to the medium. Percentage survival is expressed as (mutant/expected) x 100, where mutant is the number of mei-9^X/mei-9^A2 in treated vials, and expected is the number of mei-9^X/FM7 (control) in the treated vial or plate times the ratio of mutant to control in untreated vials. Each experiment consisted of summed counts from 10 vials. For UV assays, embryos were collected on grape agar plates overnight and then allowed to develop for 4 days. The resulting third instar larvae were washed and spread in a monolayer on chilled petri plates and irradiated in a Stratalinker (Stratagene, La Jolla, CA). Percentage survival was calculated as described above. Means and standard deviations were determined from at least three independent experiments.

Sequencing of mutants:
The entire mei-9 protein-coding region was sequenced from all 14 mutant lines. Individual flies homozygous for each mutation were homogenized and PCR was performed using gene-specific primers. PCR products were isolated on an agarose gel, purified, and sequenced directly. Mutations were confirmed by sequencing the opposite strand from an independent amplification.

Western blot analysis:
Ovaries from mei-9 mutants were dissected on ice and then ground and boiled in SDS sample buffer. Samples were loaded onto a polyacrylamide gel at the equivalent of one pair of ovaries per lane. After separation by electrophoresis, proteins were transferred to polyvinyl difluoride (PVDF) membrane. MEI-9 was detected with rabbit polyclonal anti-MEI-9 serum, using the ECL detection kit (Amersham, Arlington Heights, IL).

Yeast two-hybrid assay:
Yeast two-hybrid assays were done as in YILDIZ et al. 2002 .

	RESULTS AND DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS AND DISCUSSION LITERATURE CITED

Architecture of the MEI-9 protein:
Drosophila MEI-9 shares extensive amino acid sequence similarity with human XPF and yeast Rad1p (SEKELSKY et al. 1995 ; SIJBERS et al. 1996 ). The highest degree of sequence similarity is found in the amino-terminal third (residues 3–326 of MEI-9) and the carboxy-terminal third (residues 534–875 of MEI-9). The central region shows limited similarity in primary sequence, but is rich in glutamic acid (23 of 208 residues), glutamine (20 residues), and proline (18 residues).

Several motifs and domains are recognizable in the primary sequence (Fig 1A and Fig 2). The amino-terminal conserved region has several motifs found in RNA helicase superfamily 2 members. Of the seven conserved motifs found in members of this helicase superfamily, MEI-9, XPF, and Rad1p share the first four (I, Ia, II, and III) and possibly the fifth (ARAVIND et al. 1999 ; SGOUROS et al. 1999 ). However, key residues necessary for helicase activity are absent, so it is unlikely that these proteins have this activity.

View larger version (30K): In this window In a new window Download PPT slide   Figure 1. Conserved motifs in MEI-9, XPF, and Rad1p. (A) Schematic of the primary structure of MEI-9, compared to XPF and Rad1p. Shaded boxes represent the positions of the helicase-like motifs (I, Ia, II, and III), the nuclease domain, and the carboxy-terminal helix-hairpin-helix motifs (HhH). Positions of mei-9 alleles are indicated. The bar below each schematic delimits the amino-terminal and carboxy-terminal conserved regions. Numbers below XPF and Rad1p are percentage of amino acid identity and similarity in these regions and in the central, less-conserved region. (B) Interactions between MEI-9 and MUS312 and between MEI-9 and ERCC1 in a yeast two-hybrid screen. The regions encoding the indicated residues of MEI-9 were expressed as fusion proteins with Gal4 DNA-binding domain, and full-length MUS312 and ERCC1 were expressed as fusion proteins with Gal4 activation domain. Growth on +HIS indicates the presence of both plasmids; growth on –HIS indicates an interaction between the two fusion proteins.

View larger version (144K): In this window In a new window Download PPT slide   Figure 2. Alignment of MEI-9, XPF, and Rad1p and positions of MEI-9 mutations. An alignment of MEI-9 with human XPF and S. cerevisiae Rad1p (residues 90–1100) is shown. Residues that are identical or similar in all three species are indicated by white lettering on a black background; residues that are identical or similar only between MEI-9 and XPF are indicated by white lettering on a gray background. Motifs and domains that are shown include the helicase-like motifs (labeled I, Ia, II, and III), the nuclease domain, and the two helix-hairpin-helix (HhH) motifs (open boxes represent helices, connecting line represents hairpin). Mutations that we identified are shown above the MEI-9 sequence. Nonsense mutations are labeled with a hexagon; residues that are deleted in mei-9D2 and mei-9L1 are marked with the symbol . We also found several polymorphisms, which are indicated by parentheses above the MEI-9 sequence.

Perhaps the most significant conserved sequence in MEI-9 is the nuclease domain, which is in the carboxy-terminal region of high conservation. This domain is related to the nuclease domain of eukaryotic Mus81 and several archaeal proteins and is characterized by the conserved signature VERKX₃D (ARAVIND et al. 1999 ; SGOUROS et al. 1999 ). The crystal structure for the related nuclease domain of the Pyrococcus furiosus protein Hef was recently solved and was found to have an active site similar in structure to that of type II restriction endonucleases and of the P. furiosus Holliday junction resolvase Hjc (NISHINO et al. 2003 ).

Carboxy-terminal to the nuclease domain are two helix-hairpin-helix motifs, which are thought to be involved in binding DNA substrates and in dimerization with ERCC1/Rad10p. XPF interacts with Ercc1 via residues 726–905 (DE LAAT et al. 1998B ), and Rad1p interacts with Rad10p via residues 809–997 (BARDWELL et al. 1993 ). Consistent with this, we found that a fragment of MEI-9 encompassing residues 600–926 interacts with Drosophila ERCC1 in a yeast two-hybrid assay, but a fragment encompassing residues 1–626 does not (Fig 1B). In contrast, MUS312 interacts with MEI-9 residues 1–626, but not with residues 600–926 (Fig 1B). The interaction domains for ERCC1 and MUS312 therefore map to different regions on the MEI-9 polypeptide.

Mutations in mei-9 alleles:
We collected 16 existing mei-9 alleles, which were isolated in several different screens over a period of almost 30 years (Table 1). Two of these, mei-9^D1 and mei-9^RT4, showed neither hypersensitivity to DNA-damaging agents nor increased levels of chromosome nondisjunction, so they were not studied further. We sequenced mei-9 from the other 14 stocks to identify the molecular lesions responsible for the mutant phenotypes (Table 1 and Fig 2). The mei-9^D2 and mei-9^D3 alleles were found to be identical and are referred to as mei-9^D2. Similarly, mei-9^A1, mei-9^A2, and mei-9^A3 were identical and are referred to as mei-9^A2. Finally, mei-9^RT2 and mei-9^RT3 were identical and are referred to as mei-9^RT3. This analysis therefore identified 10 unique mei-9 mutations.

Two mei-9 alleles were recovered in a hybrid dysgenesis screen in which nonautonomous P elements were mobilized (YAMAMOTO et al. 1990 ). The mei-9^RT1 allele has a 672-bp P element inserted into the 5' untranslated region, 119 bp upstream of the translation start site (SEKELSKY et al. 1995 ). The mei-9^RT3 allele does not have a P element, but contains a single-base-pair insertion just past the nuclease domain, resulting in a frameshift. The predicted product contains residues 1–742 followed by 50 novel residues.

The other eight mutations were induced by the alkylating agent ethyl methanesulfonate (EMS). Of these, six have the most common alteration produced by EMS, a G to A transition (PASTINK et al. 1991 ). In three cases (mei-9^A2, mei-9^b, and mei-9¹¹) the result is a nonsense mutation, and in the other three (mei-9¹², mei-9^a, and mei-9^D4) the result is a missense mutation. The nonsense mutations are predicted to truncate the polypeptide after residue 103 (mei-9^A2), 125 (mei-9^b), or 498 (mei-9¹¹). Both mei-9^A2 and mei-9¹¹ behave as null mutations in genetic assays (see below), but mei-9^b has been considered to be hypomorphic (BAKER and CARPENTER 1972 ).

Two of the missense mutations are within the active site of the nuclease domain. In mei-9^a, aspartic acid 658 is changed to asparagine. This residue is conserved in all proteins with the related nuclease domain. When the corresponding residue of XPF is changed to alanine, nuclease activity is abolished (ENZLIN and SCHARER 2002 ). The same change made to Hef severely decreases, but does not completely abolish nuclease activity (NISHINO et al. 2003 ). In mei-9^D4, arginine 697 is replaced by glutamine. In Hef, this arginine is involved in stabilizing the active site through electrostatic interactions with glutamic acid residues 661 and 731. Although R697 is also absolutely conserved, substitution with alanine decreases but does not abolish the in vitro nuclease activities of XPF and Hef (ENZLIN and SCHARER 2002 ; NISHINO et al. 2003 ). The third missense mutation, mei-9¹², changes glycine 306 to glutamic acid, resulting in a meiosis-specific mutant phenotype (YILDIZ et al. 2002 ).

The remaining two EMS-induced alleles of mei-9 are small in-frame deletions. There is a 6-bp deletion in mei-9^D2, resulting in the loss of two conserved residues at the beginning of helicase-like motif III. In mei-9^L1, a 39-bp deletion results in the loss of 13 residues between helicase-like motif III and the poorly conserved central region of the polypeptide.

Expression of MEI-9 mutant proteins:
To investigate the effects of mei-9 mutations on expression of MEI-9 protein, we probed blots of ovary extracts with a polyclonal rabbit serum raised against residues 316–665 (Fig 3). The predicted size of wild-type MEI-9 is 106 kD, but the anti-MEI-9 serum recognizes a protein that runs at a relative molecular mass of 125 kD. Several polypeptides of lower molecular weight are also detected, but since they do not change in mei-9 mutants, we conclude that they are unrelated to MEI-9. The two nuclease domain missense mutations (mei-9^a and mei-9^D4) express MEI-9 at approximately wild-type levels.

View larger version (89K): In this window In a new window Download PPT slide   Figure 3. Expression of MEI-9 in mutant ovaries. Ovary extracts from wild-type (WT) and various mei-9 mutant alleles were separated on a polyacrylamide gel, transferred to PVDF membrane, and detected with polyclonal anti-MEI-9 serum. MEI-9 is indicated by an arrowhead. The antiserum also detects unknown bands of lower molecular weight, unrelated to MEI-9; one of these is included as a loading control. The open arrowhead in A indicates a band that may correspond to truncated MEI-911. In this blot, mei-9RT1 appears to have a band that runs just under wild-type MEI-9. In other blots, however, this band appears to be similar in intensity in all lanes (B).

We do not detect any MEI-9 protein in ovaries of females homozygous for the nonsense mutation mei-9^A2. It is unknown whether the predicted 103-residue amino-terminal fragment is produced and is stable. However, ARAJ and SMITH 1996 were unable to detect mei-9 transcripts in this mutant, suggesting that it is removed by nonsense-mediated mRNA decay. We therefore consider mei-9^A2 to be a null allele molecularly. Consistent with this, mei-9^A2 behaves as a null mutation in genetic assays (see below). We did not detect full-length MEI-9 in ovaries of mei-9¹¹ females, but we did detect a faint band at a relative molecular mass of 65 kD. The predicted size of the MEI-9¹¹ truncation is 54 kD, so it is likely that this band corresponds to the truncated protein. Surprisingly, we detect a protein of approximately normal size and abundance in ovaries of females homozygous for mei-9^b, which has a nonsense mutation at codon 126. Thus, there is some type of suppression of this nonsense codon, such as translational read-through or mRNA editing. Suppression of nonsense mutations has been observed previously in Drosophila (WASHBURN and O'TOUSA 1992 ; SAMSON et al. 1995 ).

We also failed to detect MEI-9 protein in mei-9^RT1, mei-9^RT3, or mei-9^D2 ovaries. For mei-9^RT1, the P-element insertion into the 5' untranslated region may affect transcription, splicing, or translation of the mei-9 mRNA, resulting in decreased protein production. However, since this allele is not null (see below) MEI-9 must be expressed at a level below our detection. The predicted MEI-9^RT3 protein is 15 kD smaller than wild-type MEI-9. We do not detect any of this mutant protein on our blots, even after prolonged exposure times. Some researchers have proposed that the stability of the mammalian ortholog XPF requires the presence of its binding partner Ercc1 (VAN VUUREN et al. 1993 ; GAILLARD and WOOD 2001 ). It is possible that MEI-9^RT3 is unstable because it lacks the ERCC1-binding region and therefore cannot heterodimerize.

The predicted MEI-9^D2 protein lacks two conserved residues (V143 and K144) from the motif similar to helicase motif II. The function of this motif in MEI-9 and orthologs is unknown, so it is difficult to hypothesize why we fail to detect this mutant protein. In contrast, the mei-9^L1 deletion predicts a protein lacking 13 residues (199–212), 7 of which are conserved between MEI-9, XPF, and Rad1. MEI-9^L1 protein is readily detectable, though at a reduced level. Therefore, the 39-bp deletion [or the 13-amino-acid (aa) deletion] either decreases transcription or translation efficiency or decreases stability of the protein.

Meiotic defects in mei-9 mutants:
Proper segregation of homologous chromosomes is dependent on the generation of reciprocal crossovers by the meiotic recombination pathway (reviewed in HAWLEY 1988 ). Failure to generate crossovers results in nondisjunction, which will give rise to aneuploidy. BAKER and CARPENTER 1972 reported high levels of meiotic chromosome nondisjunction for the first two alleles of mei-9, mei-9^a, and mei-9^b. We performed the same X chromosome nondisjunction assay on the 10 alleles we sequenced (Table 2). To control for genetic background effects, we examined each allele as a heterozygote with the null allele mei-9^A2. Most alleles caused a high level of nondisjunction—between a 70- and 110-fold increase relative to wild-type females.

The mei-9^RT1 allele stands out in that it causes only a sixfold increase. As described above, we did not detect MEI-9 protein in mei-9^RT1 ovaries. Because meiotic cells comprise only a small fraction of the ovary, it is possible that MEI-9 levels are not decreased as substantially in meiotic cells as in the rest of the ovary. Alternatively, very low levels of MEI-9 may be sufficient to carry out the meiotic functions of the gene. A single crossover per pair of homologous chromosomes is sufficient to ensure disjunction, and only six to eight crossovers occur in a normal Drosophila meiosis (CARPENTER and BAKER 1982 ). In contrast, the mutagen hypersensitivity assays are likely to induce extensive amounts of DNA damage and therefore may require more MEI-9 protein for efficient repair.

We further characterized the meiotic phenotype of mei-9^RT1 females by directly measuring levels of crossing over along the left arm of chromosome 2. We observed a relatively small decrease in crossing over in this region. We conclude that sufficient MEI-9 protein is present in meiotic cells to provide nearly complete function.

Although the meiotic phenotype of mei-9^b females was thought to be weaker than that of mei-9^a females (BAKER and CARPENTER 1972 ), in our assay there was no significant difference (P = 0.73 by Fisher's exact test). We assayed each allele in trans to mei-9^A2, so the difference could be due to genetic background effects or to the stronger phenotype of a hypomorphic allele in trans to a null allele relative to the phenotype of a homozygous hypomorphic allele. The amount of protein in mei-9^b ovaries is far greater than that in mei-9^RT1 ovaries, yet mei-9^b females have a fairly strong meiotic defect, whereas mei-9^RT1 females have a very weak defect. Read-though of the nonsense mutation in mei-9^b may replace the glutamine residue, which is conserved in XPF, with a residue that reduces or eliminates protein function, and this may explain the mei-9^b phenotype.

Three other alleles, mei-9^D4 (R697Q, in nuclease domain), mei-9¹² (G306E, prevents interaction with MUS312), and mei-9^RT3 (frameshift at 742), result in a somewhat lower level of nondisjunction than the null allele mei-9^A2. Although the differences are statistically significant (P < 0.0001 in each case), it is unclear what the biological significance is, if any. It is noteworthy that mus312 null alleles result in a level of nondisjunction similar to that of mei-9¹², the meiosis-specific allele that results in a protein incapable of interaction with MUS312 (YILDIZ et al. 2002 ). Like mei-9¹² mutants (see below), the level of crossing over in mus312 mutants is similar to or lower than that of the null mutation mei-9^A2. Similar differences in nondisjunction frequencies have been reported for other Drosophila meiotic recombination mutants (HALL 1972 ; MCKIM et al. 1996 ).

We further investigated the meiotic phenotypes of these mutants by directly measuring crossing over between net and pr, a 54-map-unit interval spanning the entire euchromatic portion of the left arm of chromosome 2 (Table 3). All three of these mutations resulted in a severe decrease in crossing over across the entire chromosome arm. The magnitude of the decrease is similar to that of mei-9^A2 females, except in the case of mei-9^D4, which has slightly higher levels of crossing over. The decreased level of nondisjunction in mei-9^D4 may therefore be due to residual function of the missense protein, but it is unclear why mei-9^RT3 and mei-9¹² produce 20–25% X nondisjunction, whereas others (including null mutations) produce 30–35% nondisjunction.

The only known requirement for mei-9 in female meiosis is in the meiotic recombination pathway, and it is believed that nondisjunction in these mutants is due entirely to a defect in generating reciprocal crossovers (CARPENTER and SANDLER 1974 ). It is possible that there is a second function for mei-9 in meiotic chromosome segregation that is independent of the crossover function. In Schizosaccharomyces pombe, for example, certain mutations in rec12, which encodes the homolog of Spo11, have effects on the segregation of achiasmate chromosomes and on meiosis II disjunction (SHARIF et al. 2002 ). The higher nondisjunction rates of some mei-9 alleles could be due to effects on these processes. However, neither we nor others have observed meiosis II nondisjunction in mei-9 mutants (BAKER and CARPENTER 1972 ; CARPENTER and SANDLER 1974 ; Ö. YILDIZ and J. J. SEKELSKY, unpublished data), and specific effects on achiasmate chromosome segregation have not been seen for mei-9^a (BAKER and CARPENTER 1972 ).

Sensitivity of mei-9 mutants to DNA-damaging agents:
We determined the relative sensitivity of the different mei-9 mutants to two DNA-damaging agents. UV introduces pyrimidine dimers and 6-4 photo products, which are repaired by the NER pathway. HN2 is a bifunctional alkylating agent that can introduce interstrand crosslinks, which are substrates for the less well understood ICL repair pathways. We applied these agents at doses at which 80–90% of mei-9 null mutants die. Most mei-9 mutations cause extreme hypersensitivity to UV, consistent with the known function of MEI-9 in NER (Fig 4A). Only mei-9¹² mutants were not hypersensitive to UV. This allele was previously characterized as a separation-of-function mutation that disrupts the interaction between MEI-9 and its meiotic recombination partner MUS312, a protein that is not involved in NER (YILDIZ et al. 2002 ).

View larger version (42K): In this window In a new window Download PPT slide   Figure 4. Sensitivity of mei-9 mutants to ultraviolet light (A) and nitrogen mustard (B). Percentage survival, relative to wild-type controls, after exposure of larvae to 750 erg/mm2 of ultraviolet light or addition of 0.008% HN2 to food is shown for each allele tested. Bars indicate standard deviations. Allele names and molecular defects are indicated, with alleles ordered according to the position of the molecular lesion responsible, from 5' to 3'.

Sensitivity to crosslinking agents such as HN2 indicates that mei-9 also has a role in the repair of ICLs (MASON et al. 1981 ). Except for the meiosis-specific allele mei-9¹², all mei-9 alleles showed strong hypersensitivity to HN2 (Fig 4A). The missense mutation mei-9^D4 exhibited an intermediate phenotype by this assay. This result is not unexpected given that the human homolog of MEI-9 (XPF) still has residual nuclease activity when an analogous mutation is introduced (ENZLIN and SCHARER 2002 ; NISHINO et al. 2003 ).

Summary of mei-9 mutations:
mei-9^A2 has a nonsense mutation predicted to truncate the protein after 103 residues. We cannot detect any full-length MEI-9 protein in mei-9^A2 mutant ovaries. This allele confers a severe phenotype in every assay that we conducted, suggesting to us that it is completely devoid of MEI-9 activity.

The mei-9¹¹ allele also has a nonsense mutation. This allele appears to produce a truncated protein comprising the amino-terminal half of the MEI-9 protein. This allele behaves as a genetic null, with no apparent dominant effects (Ö.YILDIZ and J. J. SEKELSKY, unpublished data), suggesting that the small amount of truncated protein does not interfere with wild-type MEI-9 function.

The third nonsense mutation, mei-9^b, is a strong allele, despite the presence of nearly wild-type levels of MEI-9 protein. Although the protein appears to be nonfunctional or poorly functional, cases such as this should be taken as a caution against concluding that a nonsense mutation is necessarily a null mutation.

Two of the 10 mutations are formally separation-of-function alleles. One of these, mei-9¹², is a missense mutation that results in a severe meiotic defect, but no detectable effect on DNA repair. The basis of this phenotype is believed to be an inability to interact with the meiosis-specific partner MUS312 (YILDIZ et al. 2002 ). The other is mei-9^RT1, which has a P-element insertion in the 5' untranslated region of the gene. This mutation results in severe hypersensitivity to DNA-damaging agents, but only a mild meiotic defect. We were unable to detect MEI-9 in whole-ovary extracts of mei-9^RT1 females, providing an explanation for the severe mutagen hypersensitivity. Given the low number of crossovers that must be generated to ensure segregation of meiotic chromosomes, a level of MEI-9 below detection on our blots may be sufficient to fulfill the meiotic function. Alternatively, the insertion may disrupt expression in meiotic cells to a lesser extent than in other cells of the ovary.

A 6-bp deletion in mei-9^D2 results in the loss of two conserved residues, valine and lysine, from helicase-like motif III. Similarly, MEI-9^L1 lacks 13 residues from the amino-terminal conserved region. MEI-9 protein was undetectable in mei-9^D2 females and reduced in mei-9^L1 females. These mutant proteins may fold incorrectly, leading to decreased stability in vivo. Both alleles result in severe defects in all assays we conducted.

A single-base-pair insertion in mei-9^RT3 results in a frameshift at amino acid 742. This would make a MEI-9 protein lacking the ERCC1-binding region and also lacking the helix-hairpin-helix motifs essential for DNA binding. We do not yet know whether ERCC1 is required for meiotic recombination, but the observed meiotic phenotype in mei-9^RT3 mutants, as well as the sensitivities to UV and HN2, can be explained by an inability to bind DNA or by instability of the truncated protein.

Both mei-9^a and mei-9^D4 have missense mutations in the nuclease domain. A conserved aspartic acid residue at 658 is changed to asparagine in mei-9^a, and a conserved arginine at residue 697 is changed to a glutamine in mei-9^D4. By analogy to similar changes made to XPF, MEI-9^a is predicted to lack nuclease activity, and MEI-9^D4 is predicted to have decreased nuclease activity. Both alleles increase sensitivity to UV and to HN2, although mei-9^D4 may be hypomorphic, and both mutations decrease meiotic crossing over and increase X chromosome nondisjunction. Although we have not characterized these mutant proteins biochemically, the most plausible explanation is that these mutant proteins, which are present at levels comparable to MEI-9 in wild-type females, partially or completely lack nuclease activity. We conclude that the nuclease activity of MEI-9 is essential for generating meiotic crossovers.

The function of MEI-9 in meiotic recombination:
According to current models for meiotic recombination, crossovers are derived from an intermediate that has two Holliday junctions and adjacent heteroduplex DNA (reviewed in STAHL 1994 , STAHL 1996 ). Resolution of this intermediate is accomplished through symmetric nicks made at each Holliday junction (Fig 5). In the canonical model of SZOSTAK et al. 1983 , resolution involves symmetrical nicking at each junction by a Holliday junction resolvase. To produce a crossover, different strands must be nicked at each junction (i.e., each of the four strands is cut once). If the same two strands are nicked at both junctions, noncrossover products result.

View larger version (35K): In this window In a new window Download PPT slide   Figure 5. Model for meiotic recombination in wild-type (A) and mei-9 mutant (B) Drosophila females. The intermediate predicted by the canonical double-strand-break repair model is shown at each top. In A, the arrowheads represent nicks made by MEI-9. These nicks are then used to repair mismatches in heteroduplex DNA. The product of the nicks and repair diagrammed is a crossover without associated gene conversion (five of nine crossover half-tetrads analyzed by CURTIS et al. 1989 had no detectable gene conversion). In B, the arrowheads represent the direction of branch migration of the Holliday junctions. The strands are eventually decatenated by a type I topoisomerase, resulting in one pristine chromatid and one that has heteroduplex DNA remaining.

The two strands in regions of heteroduplex DNA are derived from different homologous chromosomes, so there may be mismatches or other heterologies present. Most such mismatches and heterologies are repaired efficiently during the recombination process. Repair can restore the original 2:2 ratio of alleles at a locus, or it can result in gene conversion, in which the ratio is changed to 3:1.

Although crossover recombinants are severely decreased in mei-9 mutants, CARPENTER 1982 found that noncrossover recombination events either were not reduced or were increased in mei-9 mutants. However, she found that heteroduplex DNA often went unrepaired in mei-9 mutants, leading to postmeiotic segregation (PMS), which manifests as recombinant progeny mosaic for two maternal alleles. The chromosomes used in her experiments were highly polymorphic, with 28 sequence heterologies within the 5.6-kb locus examined (0.5% polymorphism). Unlike the case in fungi (BORTS and HABER 1987 ), this high level of heterology does not affect meiotic recombination frequencies in Drosophila (HILLIKER et al. 1991 ). CURTIS and BENDER 1991 used these heterologies to show that gene conversion tracts in mei-9 mutants are of similar lengths to those of wild-type females.

The meiotic recombination defect in mei-9 mutants therefore has two components: a decrease in crossovers without a corresponding decrease in noncrossovers and failure to repair heteroduplex DNA, without altering the frequency or length of heteroduplex DNA. Although it is possible that these two defects represent different functions for MEI-9, we prefer a model in which both phenotypes are consequences of a single defect. In a mei-9 mutant, crossovers are reduced even in the absence of sequence polymorphisms, when there are no heterologies in heteroduplex DNA (BAKER and CARPENTER 1972 ; RUTHERFORD and CARPENTER 1988 ). The crossover defect is therefore not due to failure to repair heteroduplex DNA.

Given the DNA structure-specific endonuclease activities of mammalian and yeast homologs of MEI-9, we propose that MEI-9 functions as a Holliday junction resolvase during meiotic recombination (SEKELSKY et al. 1995 , SEKELSKY et al. 1998 ). According to this model, resolution of a recombination intermediate begins when a complex containing MEI-9 cuts one or both Holliday junctions. These nicks are used both to resolve the junctions and to allow repair of heteroduplex DNA.

How can noncrossover recombinants be produced if MEI-9 does not make cuts at the Holliday junctions? We propose that the double-Holliday-junction intermediate is "resolved" instead through branch migration and decatenation, as in THALER et al. 1987 (Fig 5B). This process can generate only noncrossover products, which explains the lack of crossovers in mei-9 mutants. In addition, because no nicks are present, there is no repair of heteroduplex DNA, which explains the PMS phenotype of mei-9 mutants. Although there is no direct evidence that meiotic recombination in Drosophila proceeds through a double-Holliday-junction intermediate, we believe that ours is the most parsimonious model to explain the meiotic phenotypes of mei-9 mutants.

CARPENTER 1982 detected PMS in only 60% of the noncrossover recombinants recovered from mei-9 mutants. She may have failed to detect all cases of PMS, given the techniques available. Alternatively, some noncrossovers may have arisen through a pathway that does not require Holliday junction cleavage. Recent studies of meiotic recombination in Saccharomyces cerevisiae suggest that, in contrast to the canonical DSB repair model, most double-Holliday-junction intermediates are resolved to give crossovers, whereas most noncrossovers arise through an alternative branch of the pathway termed synthesis-dependent strand annealing (SDSA, ALLERS and LICHTEN 2001 ; HUNTER and KLECKNER 2001 ). In Drosophila, SDSA accounts for most mitotic DSB repair, which is usually not associated with crossing over (KURKULOS et al. 1994 ; NASSIF et al. 1994 ; ADAMS et al. 2003 ). Although there are no data concerning the use of SDSA in meiotic recombination in Drosophila, noncrossover events generated by an SDSA pathway in mei-9 mutants would not be expected to have PMS. It is therefore possible that some noncrossovers in Drosophila meiotic recombination occur through SDSA.

The few crossovers that remain in mei-9 mutants are of unknown origin, although they are dependent on the Spo11 homolog MEI-W68 (MCKIM and HAYASHI-HAGIHARA 1998 ). It is possible that a secondary Holliday junction resolvase acts with lower efficiency in mei-9 mutants. Alternatively, these crossovers may arise through a different pathway. OSMAN et al. 2003 recently proposed a recombination model in which crossovers are generated through an intermediate that does not contain fully ligated Holliday junctions. This model was proposed to reconcile the in vitro activities of Mus81-Eme1 with the meiotic phenotype of mus81 and eme1 mutants. We do not favor this model for Drosophila, because it does not easily explain the defect in repair of heteroduplex DNA seen in mei-9 mutants. However, this model could explain the MEI-9-independent crossovers. It will be interesting to see whether these crossovers have PMS and whether they require Drosophila MUS81-MMS4.

	FOOTNOTES

¹ Present address: Department of Radiation and Cellular Oncology, University of Chicago, Chicago, IL 60637.
² Present address: McLendon Clinical Labs, University of North Carolina, Chapel Hill, NC 27599.
³ Present address: Washington University School of Medicine, St. Louis, MO 63130.

	ACKNOWLEDGMENTS

We thank Scott Hawley and Kim McKim for providing mei-9 alleles, Jan LaRocque and Sarah Radford for assistance with experiments, and members of the Sekelsky laboratory for helpful comments. H.K. was supported by a National Research Service Award postdoctoral fellowship. B.K. was supported by a Summer Undergraduate Research Fellowship from the Smallwood Foundation. This work was supported by a grant from the National Institute of General Medical Science to J.J.S. (GM61252).

Manuscript received November 24, 2003; Accepted for publication January 22, 2004.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS AND DISCUSSION LITERATURE CITED

ADAMS, M. D., M. MCVEY, and J. J. SEKELSKY, 2003  Drosophila BLM in double-strand break repair by synthesis-dependent strand annealing. Science 299:265-267.[Abstract/Free Full Text]

ALLERS, T. and M. LICHTEN, 2001  Differential timing and control of noncrossover and crossover recombination during meiosis. Cell 106:47-57.[CrossRef][Medline]

ARAJ, H. and P. D. SMITH, 1996  Positional cloning of the Drosophila melanogaster mei-9 gene, the putative homolog of the Saccharomyces cerevisiae RAD1 gene. Mutat. Res. 364:209-215.[Medline]

ARAVIND, L., D. R. WALKER, and E. V. KOONIN, 1999  Conserved domains in DNA repair proteins and evolution of repair systems. Nucleic Acids Res. 27:1223-1242.[Abstract/Free Full Text]

BAKER, B. S. and A. T. C. CARPENTER, 1972  Genetic analysis of sex chromosomal meiotic mutants in Drosophila melanogaster.. Genetics 71:255-286.[Abstract/Free Full Text]

BAKER, B. S., A. T. C. CARPENTER, and P. RIPOLL, 1978  The utilization during mitotic cell division of loci controlling meiotic recombination in Drosophila melanogaster.. Genetics 90:531-578.[Abstract/Free Full Text]

BARDWELL, A. J., L. BARDWELL, D. K. JOHNSON, and E. C. FRIEDBERG, 1993  Yeast DNA recombination and repair proteins Rad1 and Rad10 constitute a complex in vivo mediated by localized hydrophobic domains. Mol. Microbiol. 8:1177-1188.[Medline]

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

BORTS, R. H. and J. E. HABER, 1987  Meiotic recombination in yeast: alteration by multiple heterozygosities. Science 237:1459-1465.[Abstract/Free Full Text]

BOYD, J. B., M. D. GOLINO, T. D. NGUYEN, and M. M. GREEN, 1976a  Isolation and characterization of X-linked mutants of Drosophila melanogaster which are sensitive to mutagens. Genetics 84:485-506.[Abstract/Free Full Text]

BOYD, J. B., M. D. GOLINO, and R. B. SETLOW, 1976b  The mei-9^a mutant of Drosophila melanogaster increases mutagen sensitivity and decreases excision repair. Genetics 84:527-544.[Abstract/Free Full Text]

CARPENTER, A. T. C., 1982  Mismatch repair, gene conversion, and crossing-over in two recombination-defective mutants of Drosophila melanogaster. Proc. Natl. Acad. Sci. USA 79:5961-5965.[Abstract/Free Full Text]

CARPENTER, A. T. C. and B. S. BAKER, 1982  On the control of the distribution of meiotic exchange in Drosophila melanogaster.. Genetics 101:81-89.[Abstract/Free Full Text]

CARPENTER, A. T. C. and L. SANDLER, 1974  On recombination-defective meiotic mutants in Drosophila melanogaster.. Genetics 76:453-475.[Abstract/Free Full Text]

CURTIS, D. and W. BENDER, 1991  Gene conversion in Drosophila and the effects of meiotic mutants mei-9 and mei-218.. Genetics 127:739-746.[Abstract]

CURTIS, D., S. CLARK, A. CHOVNICK, and W. BENDER, 1989  Molecular analysis of recombination events in Drosophila.. Genetics 122:653-661.[Abstract/Free Full Text]

DE LAAT, W. L., E. APPELDOORN, N. G. J. JASPERS, and J. H. J. HOEIJMAKERS, 1998a  DNA structural elements required for ERCC1-XPF endonuclease activity. J. Biol. Chem. 273:7835-7842.[Abstract/Free Full Text]

DE LAAT, W. L., A. M. SIJBERS, H. ODIJK, N. G. JASPERS, and J. H. HOEIJMAKERS, 1998b  Mapping of interaction domains between human repair proteins ERCC1 and XPF. Nucleic Acids Res. 26:4146-4152.[Abstract/Free Full Text]

DE LAAT, W. L., N. G. JASPERS, and J. H. HOEIJMAKERS, 1999  Molecular mechanism of nucleotide excision repair. Genes Dev. 13:768-785.[Free Full Text]

ENZLIN, J. H. and O. D. SCHARER, 2002  The active site of the DNA repair endonuclease XPF-ERCC1 forms a highly conserved nuclease motif. EMBO J. 21:2045-2053.[CrossRef][Medline]

FLYBASE, 2001 FlyBase—A Database of the Drosophila Genome (http://flybase.bio.indiana.edu).

GAILLARD, P. H. and R. D. WOOD, 2001  Activity of individual ERCC1 and XPF subunits in DNA nucleotide excision repair. Nucleic Acids Res. 29:872-879.[Abstract/Free Full Text]

GRAF, U., B. VOGEL, S. BIBER, and G. WURGLER, 1979  Genetic control of mutagen sensitivity in Drosophila melanogaster: a new allele at the mei-9 locus on the X-chromosome. Mutat. Res. 59:129-133.[CrossRef][Medline]

HABRAKEN, Y., P. SUNG, L. PRAKASH, and S. PRAKASH, 1994  Holliday junction cleavage by yeast Rad1 protein. Nature 371:531-534.[CrossRef][Medline]

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, pp. 497–527 in Genetic Recombination, edited by R. KUCHERLAPATI and G. SMITH. American Society of Microbiology, Washington, DC.

HILLIKER, A. J., S. H. CLARK, and A. CHOVNICK, 1991  The effect of DNA sequence polymorphisms on intragenic recombination in the rosy locus of Drosophila melanogaster.. Genetics 129:779-781.[Abstract]

HUNTER, N. and N. KLECKNER, 2001  The single-end invasion: an asymmetric intermediate at the double-strand break to double-holliday junction transition of meiotic recombination. Cell 106:59-70.[CrossRef][Medline]

KURKULOS, M., J. M. WEINBERG, D. ROY, and S. M. MOUNT, 1994  P element-mediated in vivo deletion analysis of white-apricot: deletions between direct repeats are strongly favored. Genetics 136:1001-1011.[Abstract]

LINDSLEY, D. L., and G. G. ZIMM, 1992 The Genome of Drosophila melanogaster. Academic Press, San Diego.

MASON, J. M., M. M. GREEN, K. E. S. SHAW, and J. B. BOYD, 1981  Genetic analysis of X-linked mutagen-sensitive mutants of Drosophila melanogaster.. Mutat. Res. 81:329-343.[CrossRef][Medline]

MCKIM, K. S. and A. HAYASHI-HAGIHARA, 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.[Abstract/Free Full Text]

MCKIM, K. S., J. B. DAHMUS, and R. S. HAWLEY, 1996  Cloning of the Drosophila melanogaster meiotic recombination gene mei-218: a genetic and molecular analysis of interval 15E. Genetics 144:215-228.[Abstract]

NASSIF, N., J. PENNEY, S. PAL, W. R. ENGELS, and G. B. GLOOR, 1994  Efficient copying of nonhomologous sequences from ectopic sites via P-element-induced gap repair. Mol. Cell. Biol. 14:1613-1625.[Abstract/Free Full Text]

NISHINO, T., K. KOMORI, Y. ISHINO, and K. MORIKAWA, 2003  X-Ray and biochemical anatomy of an Archaeal XPF/Rad1/Mus81 family nuclease. Similarity between its endonuclease domain and restriction enzymes. Structure 11:445-457.[Medline]

OSMAN, F., J. DIXON, C. L. DOE, and M. C. WHITBY, 2003  Generating crossovers by resolution of nicked Holliday junctions: a role for Mus81-Eme1 in meiosis. Mol. Cell 12:761-774.[CrossRef][Medline]

PARK, C.-H., T. BESSHO, T. MATSUNAGA, and A. SANCAR, 1995  Purification and characterization of the XPF-ERCC1 complex of human DNA repair excision nuclease. J. Biol. Chem. 230:22657-22660.

PASTINK, A., E. HEEMSKERK, M. J. M. NIVARD, C. J. VAN VLIET, and E. VOGEL, 1991  Mutational specificity of ethyl methanesulfonate in excision-repair-proficient and -deficient strains of Drosophila melanogaster.. Mol. Gen. Genet. 229:213-218.[CrossRef][Medline]

RUTHERFORD, S. L. and A. T. C. CARPENTER, 1988  The effect of sequence homozygosity on the frequency of X-chromosomal exchange in Drosophila melanogaster females. Genetics 120:725-732.[Abstract/Free Full Text]

SAMSON, M. L., M. J. LISBIN, and K. WHITE, 1995  Two distinct temperature-sensitive alleles at the elav locus of Drosophila are suppressed nonsense mutations of the same tryptophan codon. Genetics 141:1101-1111.[Abstract]

SANCAR, A. S., 1996  DNA excision repair. Annu. Rev. Biochem. 65:43-81.[CrossRef][Medline]

SEKELSKY, J. J., K. S. MCKIM, G. M. CHIN, and R. S. HAWLEY, 1995  The Drosophila meiotic recombination gene mei-9 encodes a homologue of the yeast excision repair protein Rad1. Genetics 141:619-627.[Abstract]

SEKELSKY, J. J., K. C. BURTIS, and R. S. HAWLEY, 1998  Damage control: the pleiotropy of DNA repair genes in Drosophila melanogaster.. Genetics 148:1587-1598.[Abstract/Free Full Text]

SEKELSKY, J. J., K. S. MCKIM, L. MESSINA, R. L. FRENCH, and W. D. HURLEY et al., 1999  Identification of novel Drosophila meiotic genes recovered in a P element screen. Genetics 152:529-542.[Abstract/Free Full Text]

SEKELSKY, J. J., K. J. HOLLIS, A. I. EIMERL, K. C. BURTIS, and R. S. HAWLEY, 2000  Nucleotide excision repair endonuclease genes in Drosophila melanogaster.. Mutat. Res. 459:219-228.[Medline]

SGOUROS, J., P.-H. L. GAILLARD, and R. D. WOOD, 1999  A relationship between a DNA-repair/recombination nuclease family and archeal helicases. Trends Biochem. Sci. 24:95-97.[CrossRef][Medline]

SHARIF, W. D., G. G. GLICK, M. K. DAVIDSON, and W. P. WAHLS, 2002  Distinct functions of S. pombe Rec12 (Spo11) protein and Rec12-dependent crossover recombination (chiasmata) in meiosis I; and a requirement for Rec12 in meiosis II. Cell Chromosome 1:1-14.[CrossRef][Medline]

SIJBERS, A. M., W. L. DE LAAT, R. R. ARIZE, M. BIGGERSTAFF, and Y.-F. WEI et al., 1996  Xeroderma pigmentosum group F caused by a defect in a structure-specific DNA repair endonuclease. Cell 86:811-822.[CrossRef][Medline]

SMITH, P. D., 1976  Mutagen sensitivity of Drosophila melanogaster. III. X-linked loci governing sensitivity to methyl methanesulfonate. Mol. Gen. Genet. 149:73-85.[CrossRef][Medline]

STAHL, F., 1996  Meiotic recombination in yeast: coronation of the double-strand break repair model. Cell 87:965-968.[CrossRef][Medline]

STAHL, F. W., 1994  The Holliday junction on its thirtieth anniversary. Genetics 138:241-246.[Medline]

SZOSTAK, J. W., T. L. ORR-WEAVER, R. J. ROTHSTEIN, and F. W. STAHL, 1983  The double-strand-break repair model for recombination. Cell 33:25-35.[CrossRef][Medline]

THALER, D. S., M. M. STAHL, and F. W. STAHL, 1987  Tests of the double-strand-break repair model for Red-mediated recombination of phage l and plasmid ldv. Genetics 116:501-511.[Abstract/Free Full Text]

VAN VUUREN, A. J., E. APPELDOORN, H. ODIJK, A. YASUI, and N. G. J. JASPERS et al., 1993  Evidence for a repair enzyme complex involving ERCC1 and complementing activities of ERCC4, ERCC11 and xeroderma pigmentosum group F. EMBO J. 12:3693-3701.[Medline]

WASHBURN, T. and J. E. O'TOUSA, 1992  Nonsense suppression of the major rhodopsin gene of Drosophila. Genetics 130:585-595.[Abstract]

YAMAMOTO, A. H., R. K. BRODBERG, S. S. BANGA, J. B. BOYD, and J. M. MASON, 1990  Recovery and characterization of hybrid dysgenesis-induced mei-9 and mei-41 alleles of Drosophila melanogaster.. Mutat. Res. 29:17-28.

YILDIZ, O., S. MAJUMDER, B. C. KRAMER, and J. J. SEKELSKY, 2002  Drosophila MUS312 interacts with the nucleotide excision repair endonuclease MEI-9 to generate meiotic crossovers. Mol. Cell 10:1503-1509.[CrossRef][Medline]

This article has been cited by other articles:

				E. R. Taylor and C. H. McGowan Cleavage mechanism of human Mus81-Eme1 acting on Holliday-junction structures PNAS, March 11, 2008; 105(10): 3757 - 3762. [Abstract] [Full Text] [PDF]

				M. McVey, S. L. Andersen, Y. Broze, and J. Sekelsky Multiple Functions of Drosophila BLM Helicase in Maintenance of Genome Stability Genetics, August 1, 2007; 176(4): 1979 - 1992. [Abstract] [Full Text] [PDF]

				K. Trowbridge, K. McKim, S. J. Brill, and J. Sekelsky Synthetic Lethality of Drosophila in the Absence of the MUS81 Endonuclease and the DmBlm Helicase Is Associated With Elevated Apoptosis Genetics, August 1, 2007; 176(4): 1993 - 2001. [Abstract] [Full Text] [PDF]

				S. J. Radford, S. McMahan, H. L. Blanton, and J. Sekelsky Heteroduplex DNA in Meiotic Recombination in Drosophila mei-9 Mutants Genetics, May 1, 2007; 176(1): 63 - 72. [Abstract] [Full Text] [PDF]

				A. J. Wijeratne, C. Chen, W. Zhang, L. Timofejeva, and H. Ma The Arabidopsis thaliana PARTING DANCERS Gene Encoding a Novel Protein Is Required for Normal Meiotic Homologous Recombination Mol. Biol. Cell, March 1, 2006; 17(3): 1331 - 1343. [Abstract] [Full Text] [PDF]