Evidence of Meiotic Crossover Control in ... - Semantic Scholar

Copyright 2006 by the Genetics Society of America DOI: 10.1534/genetics.105.047845

Evidence of Meiotic Crossover Control in Saccharomyces cerevisiae Through Mec1-Mediated Phosphorylation of Replication Protein A Amy J. Bartrand, Dagmawi Iyasu, Suzanne M. Marinco and George S. Brush1 Barbara Ann Karmanos Cancer Institute and Department of Pathology, Wayne State University School of Medicine, Detroit, Michigan 48201 Manuscript received July 6, 2005 Accepted for publication August 8, 2005 ABSTRACT Replication protein A (RPA) is the major single-stranded DNA-binding protein in eukaryotes, essential for DNA replication, repair, and recombination. During mitosis and meiosis in budding yeast, RPA becomes phosphorylated in reactions that require the Mec1 protein kinase, a central checkpoint regulator and homolog of human ATR. Through mass spectrometry and site-directed mutagenesis, we have now identified a single serine residue in the middle subunit of the RPA heterotrimer that is targeted for phosphorylation by Mec1 both in vivo and in vitro. Cells containing a phosphomimetic version of RPA generated by mutation of this serine to aspartate exhibit a significant alteration in the pattern of meiotic crossovers for specific genetic intervals. These results suggest a new function of Mec1 that operates through RPA to locally control reciprocal recombination.

H

OMOLOGOUS recombination during meiosis serves to increase genetic variability and, with few exceptions, is required for proper chromosome segregation. For meiotic recombination to occur, DNA must first be ‘‘damaged’’ in the form of programmed double-strand breaks to allow for interaction and genetic exchange between DNA homologs. To complete recombination, the DNA is repaired and returned to its normal stable structure. Because this process involves transient disruption of DNA integrity, strict regulatory controls must operate to prevent deleterious genetic rearrangements. While many of the proteins that carry out meiotic recombination have been identified and characterized, a great deal about the mechanisms that regulate their functions remains to be learned. Our previous studies with the budding yeast Saccharomyces cerevisiae have revealed that the cellular single-stranded DNA (ssDNA)-binding protein replication protein A (RPA), which is required for DNA replication, repair, and recombination (for review, see Wold 1997), becomes phosphorylated upon initiation of meiotic recombination (Brush et al. 2001). The dependence of this phosphorylation event on the generation of DNA double-strand breaks and the essential role of RPA in recombination has led to our hypothesis that RPA phosphorylation is a recombination regulatory mechanism.

1 Corresponding author: Karmanos Cancer Institute, Program in Molecular Biology and Human Genetics, Wayne State University School of Medicine, 3114 Prentis Ctr., 110 E. Warren Ave., Detroit, MI 48201. E-mail: [email protected]

Genetics 172: 27–39 ( January 2006)

RPA phosphorylation in different cell types and under different environmental conditions requires members of the ATM family of protein kinases (Brush et al. 1996; Liu and Weaver 1993). The human ATM protein is mutated in the autosomal recessive disorder ataxiatelangiectasia (A-T), characterized by cerebellar degeneration, ocular telangiectasia, immunodeficiency, infertility, premature aging, and an increased risk of leukemias and lymphomas (Savitsky et al. 1995). ATM is a central checkpoint regulator that protects cells from killing by agents that induce DNA double-strand breaks, such as ionizing radiation (IR). A human homolog of ATM is ATR (ATM and Rad3 related), which unlike ATM is essential for viability and is activated by a variety of genetic insults (for review on ATM and ATR, see Abraham 2001). It has recently been shown that ATR is underexpressed in Seckel syndrome, which is characterized by developmental defects including mental retardation and dwarfism (O’Driscoll et al. 2003). A major function of ATM and ATR is to delay cell-cycle progression upon DNA damage, thereby providing the time necessary for DNA repair to occur (Abraham 2001). However, there is increasing evidence that ATM and ATR are also involved in controlling DNA repair itself (Koundrioukoff et al. 2004; Zhou and Elledge 2000). We have previously found that a number of RPA phosphorylation reactions in S. cerevisiae are directed by Mec1, a protein kinase that is similar in structure and function to ATM and even more similar to ATR (for review, see Durocher and Jackson 2001). Mec1 is essential for viability (Kato and Ogawa 1994) and is required for checkpoints that respond to genotoxic insult during both mitosis and meiosis (Weinert et al. 1994; Paulovich

28

A. J. Bartrand et al.

and Hartwell 1995; Siede et al. 1996; Stuart and Wittenberg 1998). Mec1 also directs a checkpoint response that monitors the presence of recombination intermediates during normal meiotic progression (Lydall et al. 1996). In addition, Mec1 is required for proper homologous recombination during mitosis and meiosis (Kato and Ogawa 1994; Vallen and Cross 1995; Grushcow et al. 1999; Bashkirov et al. 2000) and supports efficient DNA repair by nonhomologous end joining (de la Torre-Ruiz and Lowndes 2000; Downs et al. 2000). Several lines of evidence indicate that Mec1 is critical not only in protecting cells from environmentally inflicted or programmed DNA damage, but also in promoting proper DNA metabolism during normal S phase (Desany et al. 1998; Merrill and Holm 1999; Myung et al. 2001; Cha and Kleckner 2002). In concert with these various functions, Mec1 is required for DNA damageinduced phosphorylation of both the RPA middle subunit (Rfa2) and the large subunit (Rfa1) (Brush et al. 1996; Brush and Kelly 2000), which contains the major ssDNA-binding activity of the heterotrimer (Wold 1997), for recombination-dependent Rfa2 phosphorylation during meiosis (Brush et al. 2001), and for periodic Rfa2 phosphorylation during the normal cell cycle (Din et al. 1990; Brush et al. 1996). Recent work has shown that RPA bound to ssDNA serves as a binding platform for ATR- and Mec1-containing complexes, thereby localizing these protein kinases to sites of DNA damage (Zou and Elledge 2003). In addition, we have recently shown that ssDNA stimulates RPA phosphorylation catalyzed by immunoprecipitates of Mec1 (Bartrand et al. 2004). These results have provided a possible explanation for the induction of RPA phosphorylation in vivo by many different events, all of which involve alterations in DNA structure. On the basis of the functional characterization of ATM family members, RPA phosphorylation could be involved either in checkpoint-mediated cell-cycle delay or directly in DNA metabolism. Studies using human cells that express fragments of ATM have revealed no correlation between RPA phosphorylation and the S-phase DNA damage checkpoint (Morgan and Kastan 1997). In addition, our studies in yeast have indicated that Mec1dependent Rfa2 phosphorylation is independent of the Rad53 protein kinase (Brush et al. 1996), a central checkpoint regulator that lies downstream of Mec1 (Sanchez et al. 1996; Sun et al. 1996). Such evidence would suggest that phosphorylated RPA does not function in arresting cell-cycle progression. However, human RPA is phosphorylated at several residues (Zernik-Kobak et al. 1997), and it is possible that specific RPA phosphoisomers do function in cell-cycle delay. In this regard, it is noteworthy that Mec1-dependent Rfa1 phosphorylation does require Rad53 under certain conditions (Brush and Kelly 2000). In favor of a DNA metabolism-related function, phosphorylated human RPA isolated from DNA-damaged human cells does not

support SV40 DNA replication in vitro (Carty et al. 1994), and phosphorylated human RPA isolated from human M-phase cells exhibits a lower affinity for doublestranded DNA (dsDNA) and DNA polymerase a (Oakley et al. 2003). In addition, phosphomimetic versions of human RPA are less effective than the wild-type version in destabilizing dsDNA in vitro and in associating with DNA replication centers in vivo (Binz et al. 2003; Vassin et al. 2004). Finally, there has been speculation that Mec1dependent RPA phosphorylation is involved in stabilizing replication forks that are stalled due to DNA damage or depleted nucleotide pools, thereby preventing deleterious recombination events (Liberi et al. 2000; Cobb et al. 2003). As described in this study, we have now identified a serine residue in RPA that is phosphorylated in an Mec1dependent manner during both mitosis and meiosis. Through mutation of this residue to alanine and aspartate, we have generated RPA molecules resembling the unphosphorylated and phosphorylated forms, respectively. We have found that the RPA serine-to-aspartate mutation significantly alters the pattern of meiotic crossover events for certain genetic intervals, indicating that Mec1-mediated RPA phosphorylation influences the mechanism of crossing over during meiosis in a regionspecific manner.

MATERIALS AND METHODS Yeast strains: The yeast strains employed in this study are listed in Table 1. Rfa2 phosphorylation site mutants were generated in progenitor haploids by two-step replacement (Scherer and Davis 1979) using 5-fluoroorotic acid (Toronto Research Chemicals) for counterselection. All strains used for meiotic experiments are isogenic or congenic to SK-1 (Kane and Roth 1974). RPA phosphorylation site determination: For analysis of radiation-induced RPA phosphorylation, W303-1A cells were grown in YPD at 30, chilled on ice during exposure to 20-krad ionizing radiation using a Shepard Mark I model 68 137Cs irradiator, and then returned to 30 for 1 hr prior to harvesting. For analysis of meiotic RPA phosphorylation, DSY1089 cells were allowed to progress synchronously into meiosis as previously described (Brush et al. 2001) and were harvested at 6 hr. RPA was then purified from these cell populations (see Bartrand et al. 2004) and samples were electrophoresed through 12% SDS-polyacrylamide gels. After gel staining with Coomassie blue, gel slices containing Rfa2 were excised and dried in vacuo without heating. Further steps were conducted by the Michigan State University Mass Spectrometry Facility. Gel slices were minced, treated with trypsin, and analyzed by matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry using an Applied Biosystems mass spectrometer with a-cyano-4-hydroxycinnamic acid as the matrix. Secondary MALDI-TOF mass spectrometry analyses were performed after treatment of samples with alkaline phosphatase. Peaks whose mass decreased by 80 Da upon phosphatase treatment were identified and analyzed for sequence on the basis of expected masses of Rfa2 tryptic fragments, which were determined using software supplied at http://www. prospector.ucsf.edu. For the RPA samples derived from either

RPA Phosphorylation and Recombination

29

TABLE 1 Yeast strains Straina W303-1Ab YGB18 YGB23 DSY1089c YGB214 YGB217 DSY1057c YGB298 DSY1031c YGB300 YGB210 YGB302 YGB212 YGB401d YGB400d DSY1030c YGB299 YGB211 YGB301 YGB213 YGB303 YGB213 YGB299 YGB298 YGB441 YGB302 YGB442 HTY1212e HTY1213e YGB386 YGB372 YGB387 YGB378

Genotype MATa MATa MATa his4B his4X

ade2-1 can1-100 his3-11,15 leu2-3,112 trp1-1 ura3-1 RFA2 ade2-1 can1-100 his3-11,15 leu2-3,112 trp1-1 ura3-1 rfa2-S122A ade2-1 can1-100 his3-11,15 leu2-3,112 trp1-1 ura3-1 rfa2-S122D leu2 ::hisG MAT a arg4-BglII ho ::LYS2 trp1::hisG lys2 ura3 RFA2 leu2 ::hisG MAT a arg4-NspI ho ::LYS2 trp1::hisG lys2 ura3 RFA2 his4B leu2 ::hisG MAT a arg4-BglII ho ::LYS2 trp1::hisG lys2 ura3 rfa2-S122A his4X leu2 ::hisG MAT a arg4-NspI ho ::LYS2 trp1::hisG lys2 ura3 rfa2-S122A his4B leu2 ::hisG MAT a arg4-BglII ho ::LYS2 trp1::hisG lys2 ura3 rfa2-S122D his4X leu2 ::hisG MAT a arg4-NspI ho ::LYS2 trp1::hisG lys2 ura3 rfa2-S122D his4X ::LEU2 ::URA3 leu2 ::hisG MAT a arg4-NspI ho ::LYS2 TRP1 lys2 mec1-1 smlx ura3 RFA2 his4B ::LEU2 leu2 ::hisG MAT a arg4-BglII ho ::LYS2 TRP1 lys2 mec1-1 smlx ura3 RFA2 HIS4 leu2 ::hisG MAT a ARG4 ho ::LYS2 TRP1 lys2 ura3 RFA2 his4X leu2 ::hisG MAT a arg4-NspI ho ::LYS2 trp1::hisG lys2 ura3 RFA2 HIS4 leu2 ::hisG MAT a ARG4 ho ::LYS2 TRP1 lys2 ura3 rfa2-S122A his4X leu2 ::hisG MAT a arg4-NspI ho ::LYS2 trp1::hisG lys2 ura3 rfa2-S122A HIS4 leu2 ::hisG MAT a ARG4 ho ::LYS2 TRP1 lys2 ura3 rfa2-S122D his4X leu2 ::hisG MAT a arg4-NspI ho ::LYS2 trp1::hisG lys2 ura3 rfa2-S122D HIS4 leu2 ::hisG MAT a ARG4 ho ::LYS2 TRP1 lys2 ura3 RFA2 ::URA3 his4X leu2 ::hisG MAT a arg4-NspI ho ::LYS2 trp1::hisG lys2 ura3 RFA2 ::URA3 his4B leu2 ::hisG MAT a arg4-BglII ho ::LYS2 trp1::hisG lys2 ura3 RFA2 HIS4 leu2 ::hisG MAT a ARG4 ho ::LYS2 TRP1 lys2 ura3 RFA2 :: :: :: his4B leu2 hisG MAT a arg4-BglII ho LYS2 trp1 hisG lys2 ura3 rfa2-S122A HIS4 leu2 ::hisG MAT a ARG4 ho ::LYS2 TRP1 lys2 ura3 rfa2-S122A his4B leu2 ::hisG MAT a arg4-BglII ho ::LYS2 trp1::hisG lys2 ura3 rfa2-S122D HIS4 leu2 ::hisG MAT a ARG4 ho ::LYS2 TRP1 lys2 ura3 rfa2-S122D :: :: :: his4B leu2 hisG MAT a arg4-BglII ho LYS2 trp1 hisG lys2 ura3 rfa2-S122D HIS4 leu2 ::hisG MAT a ARG4 ho ::LYS2 TRP1 lys2 ura3 RFA2 HIS4 leu2 ::hisG MAT a ARG4 ho ::LYS2 TRP1 lys2 ura3 RFA2 his4X LEU2 MAT a arg4-NspI ho ::LYS2 trp1::hisG lys2 ura3 RFA2 :: HIS4 leu2 hisG MAT a ARG4 ho ::LYS2 TRP1 lys2 ura3 rfa2-S122D his4X LEU2 MAT a arg4-NspI ho ::LYS2 trp1::hisG lys2 ura3 rfa2-S122D can1 URA3 HOM3 HIS1 TRP2 leu2 ::hisG MAT a ho ::LYS2 lys2 RFA2 CAN1 ura3 hom3-10 his1 trp2 leu2 ::hisG MAT a ho ::LYS2 lys2 RFA2 can1 URA3 HOM3 HIS1 TRP2 leu2 ::hisG MAT a ho ::LYS2 lys2 rfa2-S122A CAN1 ura3 hom3-10 his1 trp2 leu2 ::hisG MAT a ho ::LYS2 lys2 rfa2-S122A can1 URA3 HOM3 HIS1 TRP2 leu2 ::hisG MAT a ho ::LYS2 lys2 rfa2-S122D CAN1 ura3 hom3-10 his1 trp2 leu2 ::hisG MAT a ho ::LYS2 lys2 rfa2-S122D

a

All strains were generated in this study except those indicated by a footnote. Genotypes of diploids and of haploids involved in crosses include horizontal lines to represent individual chromosomes. Certain haploid strains were used in two crosses and have been listed twice to indicate the nature of these crosses. b Thomas and Rothstein (1989). c Lydall et al. (1996), Stuart and Wittenberg (1998). d YGB401 and YGB400 were derived from YGB302 and YGB212, respectively. e Tsubouchi and Ogawa (2000). mitotic or meiotic cells, the following peptide was predicted: GYGS122QVAQQFEIGGYVK. To identify the phosphorylated residue, electrospray mass spectrometry analyses were carried out on the digests using a capillary liquid chromatography system (Waters, Milford, MA) coupled to an LCQ-Deca ion trap mass spectrometer (Thermo Finnigan) equipped with a

nanospray ionization source. Peptide fragmentation in this procedure led to identification of serine 122 as the phosphorylated residue in both RPA preparations. RPA phosphorylation assays: Phosphorylation of Rfa1 and Rfa2 in vivo was detected by Western blot analysis as previously described (Brush et al. 2001). Where indicated, cells were

30


treated with 0.1 m hydroxyurea (HU) for 2 hr. Mec1-dependent phosphorylation of RPA in vitro was performed by an immunoprecipitation kinase assay as described previously (Bartrand et al. 2004). For use as substrate, RPA was purified from either W303-1A (RFA2) or YGB23 (rfa2-S122D) cells. RPA concentration in purified preparations was initially estimated by the Bradford method (Bradford 1976) using bovine serum albumin as a standard. RPA was included at estimated concentrations of 24 mg/ml for wild type and 31, 47, and 77.5 mg/ml for S122D. The titration was included so that equal concentrations of fulllength wild-type and mutant RPA, as assessed by silver staining, could be compared. Where indicated, poly(dT) (Sigma, St. Louis) was added to a final concentration of 4.8 mg/ml. Sensitivity to killing by ionizing radiation: Stationary-phase diploid cells were placed on ice and exposed to increasing doses of ionizing radiation with a Shepard Mark I model 68 137 Cs irradiator. The cells were then diluted in water and plated in triplicate on rich media. Colonies were counted after 3 days of incubation at 30. Sporulation and tetrad analysis: Synchronous sporulation and analysis of DNA content by flow cytometry using SYBR Green I (Molecular Probes, Eugene, OR) were conducted essentially as described previously (Brush et al. 2001). Sporulation efficiency was determined by scoring the number of asci formed after 32 hr. For tetrad analysis, a minimal-growth mating/sporulation protocol was employed (Reenan and Kolodner 1992). Haploid cells were patched from frozen stock onto YPD (1% yeast extract/2% peptone/2% glucose/ 2% agar) and allowed to grow overnight at 30. The haploids were mixed together on YPD and incubated at 30 for 6 hr. The mating mixtures were then patched onto sporulation plates [1% potassium acetate and 2% agar containing both uracil (50 mm) and leucine (250 mm) or only leucine for uracil prototrophs] and incubated for 3 days at 30. Asci were dissected on YPD and isolated spores were subsequently germinated at 30. After 3 days of growth, the dissected spores were analyzed for auxotrophies and mating types by standard procedures. Only tetrads with four viable colonies were included for calculating recombination frequencies. Tetrads exhibiting gene conversions for two or more markers or 8:0 segregation for a single marker were considered false and therefore discounted. Genetic distances between markers were determined from tetrads exhibiting 2:2 segregation for both markers according to the following equation (Perkins 1949): cM ¼ 100 3 (TT 1 6NPD)/(2(PD 1 TT 1 NPD)). Genetic distances of HIS4 and MAT from centromere III were estimated using the TRP1 marker according to the following equation (see Sherman and Wakem 1991): cM9 ¼ 100 3 (TT)/(2(PD 1 TT 1 NPD)). Only tetrads segregating 2:2 for both TRP1 and the marker of interest were employed. For analysis of chromatid usage (Table 4) and coincident crossovers (Figure 5), tetrads exhibiting a gene conversion at HIS4, LEU2, or MAT were not included. For the interference data shown in Table 2 (and Table S2 at http://www.genetics.org/supplemental/), the expected NPD frequency ( f NPDe) was determined from TT frequency ( f TT) by the following equation (Papazian 1952): f NPDe ¼ 1/2(1 f TT (1 (3/2)f TT)2/3). For the data shown in Table 5, f TT was greater than two-thirds in most cases for the CAN1–URA3 and URA3–HOM3 intervals, necessitating the use of a different equation for these intervals (Papazian 1952; Nakagawa and Ogawa 1999): f NPDe ¼ ( f TT2/8)(1 1 (2/3) f TT). Standard errors of genetic distances were generated at the Stahl Lab Online Tools website (http://www.groik. com/stahl/). Chi-square tests were employed for other statistical analyses. Mitotic recombination: Homozygosis at the MAT locus was monitored to estimate the rate of mitotic recombination during mitosis. A fluctuation analysis was conducted in which

Figure 1.—Mec1 targets Rfa2 S122 in vivo. (A) MALDITOF mass spectrometry analysis of trypsin-digested Rfa2. The arrow indicates the fragment (containing S122) whose mass is altered by phosphatase treatment. (B) RPA phosphorylation in vivo. Wild-type and Rfa2 S122 mutant cells were compared for Rfa1 and Rfa2 phosphorylation during normal cell-cycle progression and after HU exposure. The positions of phosphorylated Rfa1 (P-Rfa1) and phosphorylated Rfa2 (P-Rfa2) are shown. The asterisk indicates an immunoreactive species presumed to be an Rfa1 degradation product. nine wild-type and eight rfa2-S122D colonies (2–3 mm diameter) from YPD plates were each inoculated into 1 ml YPD and incubated overnight to generate saturated cultures. A 10-ml aliquot of each culture was diluted and plated onto YPD to determine the number of colony-forming units. The remainder of each culture was concentrated and applied to a YPD plate seeded with a mating-type tester strain. The plated cells were incubated overnight and replica plated to a minimal medium that allowed growth of triploids. Recombination rates were calculated as described (Luria and Delbruck 1943).

RESULTS

Mec1 catalyzes phosphorylation of Rfa2 serine 122: To map sites in RPA that become phosphorylated in Mec1-dependent reactions, we subjected RPA purified from IR-treated yeast cells to mass spectrometry analysis (see materials and methods). MALDI-TOF mass spectrometry revealed a tryptic fragment with an m/z value of 1910.89 Da that disappeared after phosphatase treatment (Figure 1A). This mass corresponds to an Rfa2 peptide containing amino acids 119–135, GYGS QVAQQFEIGGYVK, with a single phosphate group


31 Figure 2.—Mec1 targets Rfa2 S122 in vitro. RPA containing a serine-to-aspartate mutation at residue 122 (S122D) was tested as a substrate of Mec1 protein kinase activity in vitro using immunoprecipitate (IP) of hemagglutinin (HA)-tagged Mec1 or its kinase-dead counterpart (Mec1kd) (Bartrand et al. 2004). A titration of the S122D mutant was included so that approximately equal concentrations of full-length wild-type (wt) and S122D RPA, as assessed by silver staining, could be compared. Where indicated, ssDNA [poly(dT)] was added. (Top) An autoradiograph of the silver-stained gel shown below. The positions of the three RPA subunits (Rfa1, Rfa2, and Rfa3) are indicated. Rfa1* is a presumed degradation product of Rfa1. The locations of the antibodies contributed by the IP reactions are indicated as IgG. Note that the silverstained Rfa1* is obscured by heavy-chain IgG.

attached. Further analysis by electrospray mass spectrometry indicated phosphorylation of Rfa2 serine 122 (S122) (data not shown). These techniques were then applied to RPA purified from meiotic cells, which also indicated phosphorylation of this residue (data not shown). ATM-like protein kinases such as Mec1 often catalyze phosphorylation of a serine or threonine directly upstream of a glutamine (Kim et al. 1999), and S122 is one of two such residues in Rfa2. Mutants in the chromosomal RFA2 S122 codon were generated and examined for RPA phosphorylation during normal mitotic growth and in response to genotoxic stress (Figure 1B). As might be expected from the mass spectrometry data, Rfa2 containing a serine-to-alanine mutation (S122A) resembled unphosphorylated Rfa2 as assessed by electrophoretic migration, and Rfa2 containing a serine-to-aspartate mutation (S122D) resembled phosphorylated Rfa2. In both cases, only a single species of Rfa2 was detected during exponential growth or upon exposure of cells to the DNA replication inhibitor HU. In mec1 mutant cells, HU exposure induces a low level of Tel1-dependent Rfa2 phosphorylation (Brush et al. 1996). Because this residual Rfa2 phosphorylation was abolished by the S122A mutation, we conclude that both Mec1 and Tel1 direct S122 phosphorylation. In contrast to Rfa2, the S122 mutations had no apparent affect on HU-induced Rfa1 phosphorylation (Figure 1B). To further assess Rfa2 S122 as a Mec1 target, we purified a mutated form of RPA containing the S122D substitution in Rfa2 and determined its effectiveness as a substrate for Mec1-dependent protein kinase activity in vitro (Figure 2). We found that Rfa2 phosphorylation was greatly reduced by the S122D mutation in both the absence and the presence of ssDNA. We note that our mutant RPA preparation did have a slightly greater level of a 50-kDa truncated form of Rfa1, an effect that is not

likely due to the mutation itself. Characterization of human RPA large subunit has revealed a similar truncated form that is capable of interacting with the other RPA subunits and with ssDNA (Wold 1997). However, to provide confidence that the decrease in phosphorylation of the mutant RPA was not due to a slight difference in the levels of full-length Rfa1, we performed a titration with mutant RPA. Even when mutant RPA was included at 2.5 times the initial concentration, we detected little if any basal Rfa2 phosphorylation and only 25% of the wild-type level in the presence of ssDNA (Figure 2). These results strongly support the conclusion that Rfa2 S122 is a major Mec1 target. We also found significant Mec1-dependent phosphorylation of Rfa1 and/or its truncated form when either wild-type or mutant RPA was employed (Figure 2), consistent with our studies in vivo indicating that the Rfa2 phosphorylation site mutation does not affect Mec1-dependent Rfa1 phosphorylation. Rfa2 S122 phosphorylation does not have an obvious mitotic function: Despite the strong evidence that Mec1 targets Rfa2 S122 during vegetative growth, we have yet to detect any mitotic phenotypes conferred by mutation of this single residue. For example, both mutants progressed through the cell cycle normally and neither was hypersensitive to killing by HU, ultraviolet light, methyl methanesulfonate, or IR (Figure 3 and data not shown). This lack of phenotype is consistent with a previous study in which this same residue was mutated (Mallory et al. 2003). The insensitivity of the S122A and S122D mutants to various DNA-damaging agents would suggest that DNA repair mechanisms are generally intact in these cells. However, it is important to note that Mec1 could direct phosphorylation of at least one other Rfa2 site, as evidenced by the low level of Rfa2 S122D phosphorylation observed in vitro (Figure 2). Furthermore,

32


Figure 3.—Rfa2 S122 mutants are not hypersensitive to killing by IR. Viability of stationary-phase diploid cells was analyzed after exposure to increasing IR doses: wild type (¤), rfa2-S122A ( ), rfa2-S122D (n), and mec1-1 ( ). Data from three independent experiments were averaged.

amino acid substitution of the Rfa2 phosphorylation site did not affect HU-induced Rfa1 phosphorylation (Figure 1B). Therefore, additional mutagenesis of RPA could be required to unveil DNA metabolism-related mitotic phenotypes. Mutation of Rfa2 S122 alters meiotic recombination on chromosome III: The effects of the two Rfa2 S122 mutations were further investigated by allowing cells to progress into and through meiosis. We first examined the electrophoretic mobility of RPA in the two mutants (Figure 4A). In considering these data, note that Mec1independent, Ime2-dependent RPA phosphorylation occurs early in meiosis (Clifford et al. 2004). As a result, Mec1 activity leads to the generation of hyperphosphorylated Rfa2 upon initiation of recombination, and wild-type cells contain three discernible Rfa2 isoforms during meiotic progression. In contrast, both mutant versions of RPA exhibited only two Rfa2 isoforms. As in mitotic cells, the S122D mutation decreased the electrophoretic mobility of Rfa2, leading to a species that appeared to be constitutively phosphorylated throughout meiotic progression. We have repeatedly observed that the level of transiently generated hyperphosphorylated Rfa2 in wild-type cells is low relative to total Rfa2 (Brush et al. 2001). Although accurate analysis of relative levels depends on the synchrony of the population, we can conclude from our data that the three cell types exhibit significantly different Rfa2 profiles. The Rfa2 phosphorylation site mutants were compared for kinetics of meiotic progression and for efficiency of haploid product formation. Analysis of DNA

Figure 4.—Rfa2 S122 mutants exhibit altered electrophoretic profiles during meiosis. Wild-type and mutant cells were synchronously sporulated and analyzed for (A) Rfa2 status by Western blotting and (B) DNA content by flow cytometry. P indicates the normal position of the hyperphosphorylated Rfa2 that is generated upon initiation of meiotic recombination. Chromosomal DNA content is 2C prior to and 4C after premeiotic DNA synthesis.

content in cells progressing synchronously through meiosis revealed no significant effect of either S122 mutation on premeiotic DNA replication (Figure 4B). Although close examination of the flow cytometry data might suggest subtle differences in the behavior of the cell populations, these differences were not reproducible and most likely result from minor differences in synchrony. However, we have consistently observed a difference in the flow cytometry profile of mec1 cells, which exhibited a broad 2C peak that may consist of two subpopulations (Figure 4B). Nonetheless, mec1 cells appeared to complete premeiotic DNA synthesis at a rate similar to the other strains. The absence of an effect on premeiotic DNA synthesis in the Rfa2 phosphorylation site mutant strains may have been anticipated given that meiotic Mec1-dependent RPA phosphorylation occurs after DNA synthesis has already occurred (Brush et al.


33

TABLE 2 Reciprocal recombination in the HIS4-MAT interval Interferencec

Tetrads Crossa A

B

Rfa2 status Wild type S122A S122D Wild typed Wild type S122A S122D Wild type/S122D

Distance (cM)b

Total

PD

TT

NPD

NPDe

NPD/NPDe

P-value

6 6 6 6 6 6 6 6

384 359 384 323 216 230 231 279

141 136 136 123 65 82 79 86

228 210 221 186 141 139 135 181

15 13 27 14 10 9 17 12

34.1 30.2 30.5 25.7 29.3 21.8 19.4 36.5

0.44 0.43 0.89 0.54 0.34 0.41 0.88 0.33

0.001 0.002 0.53 0.02 0.0004 0.006 0.58 ,0.0001

41.4 40.1 49.9 41.8 46.5 42.0 51.3 45.3

2.9 2.9 3.7 3.3 4.1 3.8 4.9 3.5

PD, parental ditype; TT, tetratype; NPD, nonparental ditype. a Cross A, MATa HIS4 3 MATa his4X; cross B, MATa his4B 3 MATa HIS4. b Genetic distance as determined by Perkins (1949) (see materials and methods). c The expected number of NPD tetrads (NPDe) was calculated from the TT frequency (see materials and methods). P is the probability, based on a x2 test, that the difference between NPD and NPDe is due to chance. d Derived from rfa2-S122D strains.

2001). We further found that mutation of S122 affected neither the efficiency of ascus formation (for example, wt ¼ 59%, S122A ¼ 63%, and S122D ¼ 65% in one experiment) nor the viability of haploid spore products (for example, 99% for each strain as estimated from tetrad dissection in Table 2, cross A). Importantly, mec1 cells are reported to only modestly decrease spore formation and viability (Lydall et al. 1996). In fact, we observed 54% sporulation efficiency for mec1 cells in the aforementioned experiment, similar to the levels observed in wild-type and RPA phosphorylation site mutant cells. Therefore, specific inactivation of the meiotic Mec1-RPA pathway through mutation of Rfa2 might not be expected to provoke general replication or sporulation defects. The dependence of meiotic Mec1-dependent RPA phosphorylation on initiation of recombination (Brush et al. 2001) suggests that phosphorylation of Rfa2 S122 functions in the meiotic recombination process. To explore this possibility, we employed tetrad analysis to evaluate both nonreciprocal and reciprocal recombination events. We did not observe effects on gene conversion that were attributable to RPA phosphorylation site mutation (see Table S1 at http://www.genetics.org/supplemental/). However, we did observe changes in crossover pattern. We initially examined the interval between the HIS4 and MAT loci, which are positioned 40 cM apart on opposite sides of the chromosome III centromere. In the first cross, we observed that the S122A mutation had no effect on genetic distance but that the S122D mutation led to an increase (Table 2, cross A). The degree of this increase varied with different crosses, but showed a consistent trend (see below). Further inspection of our data revealed a pronounced increase in the proportion of nonparental ditype (NPD) tetrads with rfa2-S122D cells. Replacement of rfa2-S122D with RFA2 verified that this increase was due

to the S122D mutation and not to a secondary mutation introduced during strain construction (Table 2, cross A). We also analyzed mitotic recombination in this region by examining MAT homozygosis in wild-type and rfa2-S122D diploids. We found that the recombination rates in both wild-type and mutant cells were 108 events/generation. It is recognized that these rates are likely to be underestimates due to the nature of the assay (see materials and methods). Nonetheless, they are many orders of magnitude below what is observed during meiosis and are essentially identical in the two strains. These data in combination with the minimal mitotic growth dictated by our sporulation protocol indicate that the increased NPD frequency observed in rfa2-S122D cells is a meiotic phenomenon. The frequency of NPD tetrads can provide an estimate of the double-crossover frequency between markers. For most genetic intervals in wild-type cells, a crossover interference mechanism operates to inhibit neighboring crossover events, leading to fewer NPDs observed than would be expected if they were randomly distributed (for recent discussion, see Bishop and Zickler 2004). Assuming the absence of chromatid interference, which would serve to increase NPD frequency but to our knowledge has not been detected in S. cerevisiae (Mortimer et al. 1989), crossover interference can be measured by comparing the observed and expected NPD frequencies. Expected NPD frequency is routinely determined from the tetratype (TT) frequency using a calculation based on the Poisson distribution (Papazian 1952) (see materials and methods). While the wild-type and rfa2S122A cells exhibited a significant degree of interference in the HIS4–MAT interval by this measure, interference was nearly abolished in the rfa2-S122D cells (Table 2, cross A). A second set of crosses measuring the same interval with different haploid strains led to nearly

34

A. J. Bartrand et al. TABLE 3 Genetic distance from centromere III Genetic distance cM9b Cross A

B

a

cMc

HIS4-MAT ðcMÞ

Rfa2 status

HIS4-CENIII

CENIII-MAT

Sum

HIS4-MAT

Sum ðcM9Þ

Wild type S122A S122D Wild typed Wild type S122A S122D Wild type/S122D

24.5 21.5 24.4 25.5 27.3 27.2 26.3 26.5

18.8 16.7 16.6 14.4 18.0 16.3 17.5 17.4

43.3 38.2 41.0 39.9 45.3 43.5 43.8 43.9

41.4 40.1 49.9 41.8 46.5 42.0 51.3 45.3

0.96 1.05 1.22 1.05 1.03 0.97 1.17 1.03

a

Crosses correspond to those in Table 2. Genetic distances between the indicated markers and centromere III (see materials and methods). c Genetic distance data from Table 2. d Derived from rfa2-S122D strains. b

identical results (Table 2, cross B). This apparent decrease in crossover interference for the HIS4–MAT region is equal to, if not greater than, the decrease that results from deletion of MSH4 (Novak et al. 2001). We also found that the S122D mutation was recessive to wild type (see Table 2, cross B). Due to heterozygosity at the TRP1 locus, which is tightly linked to the chromosome IV centromere, we were able to monitor second division segregation of HIS4 and MAT. This analysis provides an estimate of the genetic distances between these markers and the chromosome III centromere. In contrast to direct mapping of the HIS4–MAT interval, the genetic distance between these two loci determined through centromere III linkage did not reveal an obvious difference between the wild-type and mutant cells (Table 3). Therefore, the S122D mutation does not appear to simply increase the crossover frequency between HIS4 and MAT. Subinterval analysis of chromosome III reveals an unusual recombination phenotype: To address the possibility that the S122D mutation increased the specific frequency of four-chromatid double crossovers (NPDs) in the HIS4–MAT interval by enforcing chromatid interference rather than diminishing crossover interference, a third cross including heterozygosity at the LEU2 locus lying between HIS4 and MAT was conducted (see Figure 5 and Table S2 at http://www.genetics.org/ supplemental/). As with the previous crosses, we observed an apparent decrease in crossover interference for the entire HIS4–MAT interval in rfa2-S122D cells, this time accompanied by only a slight increase in genetic distance. (Note that the introduction of heterozygosity at LEU2 appeared to decrease the HIS4–MAT distance in both wild-type and mutant cells.) Once again, genetic distance as measured by centromere III linkage was not increased by the S122D mutation (Figure 5). Tetrads with TT patterns in both of the two subintervals (HIS4–

LEU2 and LEU2–MAT) were further classified as having undergone two-, three-, or four-chromatid double crossovers, which in the absence of chromatid interference should occur with a 1:2:1 distribution. Our data do not exclude such a distribution for either wild-type or rfa2-S122D cells (Table 4), suggesting that Mec1-mediated Rfa2 phosphorylation does not influence chromatid usage for coincident crossovers occurring in the two neighboring intervals. The introduction of a third marker on chromosome III also allowed for measurement of crossover interference in the HIS4–MAT interval by a second method. In this case, the product of the frequencies of tetrads displaying recombination events (TT or NPD) in the single subintervals predicts the frequency of tetrads that would be expected to display simultaneous recombination events in both intervals ( fe) if crossover interference were absent (see Chua and Roeder 1997). Crossover interference leads to an observed frequency of coincident events (fo) that is less than the predicted frequency ( fo/fe , 1). To our surprise, we detected similar levels of crossover interference in wild-type and rfa2-S122D cells by this method (Figure 5). The apparent discrepancy between the two measurements of crossover interference (NPD ratio vs. coincident crossovers) can be explained by the specific elevation of subinterval NPD frequencies in the rfa2-S122D mutant (Figure 5). Mutation of Rfa2 S122 affects meiotic recombination in a specific region of chromosome V: To investigate the generality of the recombination phenotype conferred by the S122D mutation, we examined several genetic intervals on chromosome V (Table 5). For three of four intervals, mutation of the RPA phosphorylation site to either alanine or aspartate did not alter recombination frequency or apparent crossover interference. However, the genetic distance of the HOM3–HIS1 interval, which was only 0.4 cM in wild type and 0.5 cM in


35

genetic distances is statistically significant (see Stahl Lab Online Tools at http://www.groik.com/stahl/). Therefore, the frequency of crossovers in this specific region is elevated when RPA resembles the constitutively phosphorylated form.

DISCUSSION

Figure 5.—Recombination is locally altered by phosphomimetic Rfa2. Genetic distance and interference of the HIS4MAT region were compared in wild-type and rfa2-S122D cells by tetrad analysis. The NPD distributions in the subintervals are shown (NPD frequencies in parentheses), as are the NPD ratios for the subintervals and entire region. NPDe is the number of NPD tetrads expected in the absence of crossover interference. The frequencies of TT plus NPD tetrads in the HIS4-LEU2 ( fa) and in the LEU2-MAT ( fb) subintervals were multiplied to determine the frequency ( fe) of coincident recombination events in the two subintervals expected in the absence of crossover interference. This value was compared with the observed frequency ( fo) of coincident recombination events. Complete tetrad data related to this figure are provided as supplementary data (at http://www.genetics.org/ supplemental/).

the rfa2-S122A mutant, increased by approximately threefold to 1.3 cM in the rfa2-S122D mutant. On the basis of the standard error of these two measurements, the difference between the wild-type and rfa2-S122D TABLE 4 Chromatid usage for double crossovers in the HIS4-LEU2-MAT interval Chromatids involved in double crossovera Rfa2 status

2

3

4

Pb

Wild type S122D

13 13

28 20

10 13

0.66 0.68

a

Tetrads with TT patterns in both the HIS4-LEU2 and LEU2-MAT intervals were analyzed. b P is the probability, based on a x2 test, that the 2:3:4chromatid double-crossover distribution corresponds to a 1:2:1 ratio.

We have demonstrated that Rfa2 S122 becomes phosphorylated through Mec1-mediated catalysis during both mitosis and meiosis. Mutation of S122 to aspartate generates an Rfa2 isomer that mimics a constitutively phosphorylated form. This condition of RPA is considerably different from that observed in wild-type cells, particularly those undergoing meiotic recombination in which only a fraction of the total RPA becomes phosphorylated in the Mec1-mediated reaction (Brush et al. 2001). Cells harboring the S122D mutation exhibit a significant alteration in the pattern of meiotic crossovers for specific genetic regions. It has been observed previously that mutation of MEC1, or members of the same epistasis group such as RAD17, alters the choice of recombination templates during meiosis (Grushcow et al. 1999; Thompson and Stahl 1999). We suggest from our data here that Mec1 also functions to control the pattern of certain meiotic crossover events through its effect on RPA. On the basis of analysis of NPD frequency, the S122D mutation appears to greatly decrease crossover interference in the HIS4–MAT interval or, more specifically, in both of two subintervals that make up the HIS4–MAT interval. Interference was first described early in the 20th century as a mechanism in Drosophila that inhibits generation of neighboring crossovers (Muller 1916). As yet, the exact function of interference is unclear. However, it is often suggested that the nonrandom distribution of crossovers afforded by crossover interference helps to ensure that even the smallest chromosomes will sustain at least one crossover, a phenomenon commonly referred to as ‘‘obligatory chiasma’’ (for review, see Bishop and Zickler 2004). This reasoning is based on the notion that a limited number of crossover events can be established per meiosis. Through gene deletion analysis, several proteins associated with crossover interference control have been identified, including the synaptonemal complex protein Zip1 (Sym and Roeder 1994), the DNA helicase Mer3 (Nakagawa and Ogawa 1999), the telomere-associated protein Ndj1 (Tam1) (Chua and Roeder 1997), the MutS homologs Msh4 (Novak et al. 2001) and Msh5 (Argueso et al. 2004) that function in a single complex (Pochart et al. 1997), and the recombinase Dmc1 as well as its accessory protein Tid1 (Shinohara et al. 2003). In addition to a decrease in crossover interference for most genetic intervals tested, a lack of any one of these proteins leads to some degree of spore death and, in the cases of Zip1, Mer3,

36

A. J. Bartrand et al. TABLE 5 Reciprocal recombination on chromosome V Tetrads

Interval

Rfa2 status

CAN1–URA3

Wild type S122A S122D Wild type S122A S122D Wild type S122A S122D Wild type S122A S122D

URA3–HOM3

HOM3–HIS1

HIS1–TRP2

Interference a

Distance (cM)

Total

PD

TT

NPD

NPDe

NPD/NPDe

P

6 6 6 6 6 6 6 6 6 6 6 6

651 641 657 652 640 655 638 635 645 630 633 638

176 196 182 191 171 199 633 629 628 529 511 526

461 428 460 435 440 435 5 6 17 100 118 109

14 17 15 26 29 21 0 0 0 1 4 3

60.1 51.6 59.1 52.4 55.1 52.1

0.23 0.33 0.25 0.50 0.53 0.40 — — — — — —

,0.0001 ,0.0001 ,0.0001 0.0003 0.0004 ,0.0001

41.9 41.3 41.9 45.3 48.0 42.8 0.4 0.5 1.3 8.4 11.2 10.0

1.7 1.9 1.8 2.2 2.3 2.0 0.2 0.2 0.3 0.9 1.2 1.1

See Table 2 for definitions. a An alternative Papazian equation was used to determine NPDe for the CAN1-URA3 and URA3-HOM3 intervals (see materials and methods) because the TT frequencies for these intervals were . 2/3 in most cases. Due to the small NPD values, interference was not determined for the HOM3-HIS1 and HIS1-TRP2 intervals.

Msh4, or Msh5, a coincident reduction in recombination frequency. Our observation that the S122D mutation elevated the NPD frequency in the HIS4–MAT interval initially suggested that RPA phosphorylation state might also contribute to regulation of crossover interference. However, further analysis did not provide such evidence. Specifically, measurement of crossover interference in the same HIS4–MAT interval through enumeration of coincident recombination events in the neighboring HIS4–LEU2 and LEU2–MAT subintervals suggests that crossover interference still operates between the two subintervals. Furthermore, the HIS4–MAT genetic distance measured through centromere linkage is similar in wild-type and rfa2-S122D cells, but a decrease in crossover interference would increase this genetic distance if chromatid choice were random. Interestingly, bisection of the HIS4–MAT region revealed an increased NPD frequency for the individual subintervals. These data might suggest that the S122D mutation specifically affects recombination events occurring in close proximity. Recombination patterns on chromosome III have been extensively studied, and it has been established that two clusters of recombination hotspots, as assessed by DNA double-strand break generation, flank a ‘‘cold’’ region that includes the centromere (Baudat and Nicolas 1997; Gerton et al. 2000). Therefore, this region could naturally be subject to adjacent recombination initiation events. With this unusual characteristic of chromosome III in mind, as well as our previous studies demonstrating that Mec1-dependent RPA phosphorylation occurs after generation of DNA doublestrand breaks (Brush et al. 2001), we can envision that the S122D mutation modifies the mechanism by which

closely juxtaposed recombination events proceed after initiation. Such a distance effect would also be compatible with our model of natural Mec1-dependent RPA phosphorylation, which presumably occurs at the DNA double-strand break (Bartrand et al. 2004) and would most likely be limited to affecting nearby events. Even if two crossovers occur in close proximity, a simple reduction in crossover interference in the absence of chromatid interference would increase the genetic distance from a marker to its centromere when calculated through second division segregation analysis. An alternative explanation for the increase in NPD frequency that accompanies the S122D mutation could be that specific chromatids are employed in adjoining crossovers when RPA resembles the phosphorylated form. A reasonable assumption is that chromatid choice of such closely situated double crossovers is normally random, leading to a 1:2:1 distribution of two-, three-, and four-chromatid events. The suggestion has been made, however, that these double crossovers would optimally involve three chromatids because the resulting structure would provide greater stability to the synapsed chromosomal complex than two- or four-chromatid double crossovers (Maguire 1980). Nonetheless, if the normal distribution were altered by a chromatid interference mechanism such that adjoining double crossovers preferentially involved four chromatids, thereby resulting in NPD tetrads, an apparent increase in genetic distance as measured by the Perkins equation (Perkins 1949) would result. However, the calculated distances between the two markers and centromere III would not be increased in the second division segregation equation. In fact, a decrease would be expected, the magnitude of which would depend on the prevalence of


three-chromatid events (TT tetrads) in wild-type cells that become supplanted by four-chromatid events (NPD tetrads) in rfa2-S122D cells. It is emphasized that such chromatid selection would be confined to proximal crossovers, as we have not observed chromatid interference with more distal events (Table 4). In addition to an increase in NPD frequency for the HIS4–MAT interval, we have observed an increase in the genetic distance of the HOM3–HIS1 interval as a result of the S122D mutation. For this small region, no NPD tetrads were generated in any of the strains. Therefore, the recombination effect that we have detected could be mechanistically distinct from that of the HIS4–MAT interval, perhaps explained by a simple increase in crossover frequency for this particular region. Given that normal meiotic Mec1-dependent RPA phosphorylation depends upon initiation of recombination (Brush et al. 2001), we would favor a mechanism that occurs after the generation of a DNA double-strand break, such as the channeling of a noncrossover recombination event into a crossover. RPA has an early function in meiotic recombination as it binds to the 39 ssDNA extensions at the DNA doublestrand breaks in preparation for subsequent binding by the recombinases Rad51 and Dmc1, which perform strand invasion (for review, see Sung et al. 2003). RPA also functions later in strand invasion by binding to and stabilizing the ssDNA generated upon strand displacement of the target DNA (Eggler et al. 2002). Thus, RPA is well positioned to regulate functions associated with generation of recombination intermediates and commitment to crossover recombination, such as chromatid choice. This regulation could involve interaction of phosphorylated RPA with other recombination proteins or with the DNA itself. In this regard, it is interesting that S122 is located directly between two aromatic residues that contribute to certain RPA–ssDNA interactions in vitro (Bastin-Shanower and Brill 2001). The specificity that we observe in which many genetic regions are completely unaffected by the S122D mutation has not been observed previously and could explain the lack of phenotype resulting from the S122A mutation, which might lead only to recombination effects in other regions not tested in this study. Because rfa2S122D is recessive to wild type, we suggest that a certain threshold of phosphorylated RPA is required to increase the frequency of specific crossover events in regions. We further suggest that the phosphorylated RPA generated in wild-type meiotic cells is not of sufficient local concentration to affect the regions affected by the S122D mutation. However, it is likely that recombination in other genetic intervals is influenced by normal RPA phosphorylation. In this regard, it will be important to determine whether RPA phosphorylation is induced at all DNA double-strand breaks generated during meiosis or at only a subset. While we have found that neither lack of RPA phosphorylation nor its apparent

37

constitutive generation adversely affects meiotic progression or completion, it is possible that RPA phosphorylation is particularly important for resolving deleterious recombination issues that are rare under our controlled conditions but more prevalent under the nonideal conditions encountered in nature. Further studies on Mec1-dependent RPA phosphorylation should provide added insight into the role of crossover control in promoting proper meiosis and help to identify related functions that operate during mitotic growth. We thank David Stuart, Douglas Bishop, and Hideo Tsubouchi for providing yeast strains and Susanne Hoffmann-Benning and Rhonda D. Husain-Ponnampalam from the Michigan State Mass Spectrometry facility for expert mass spectrometry analysis. We also thank Dawn Clifford, Craig Giroux, and John Lopes for helpful discussions. This work was supported by grant GM061860 from the National Institutes of Health.

LITERATURE CITED Abraham, R. T., 2001 Cell cycle checkpoint signaling through the ATM and ATR kinases. Genes Dev. 15: 2177–2196. Argueso, J. L., J. Wanat, Z. Gemici and E. Alani, 2004 Competing crossover pathways act during meiosis in Saccharomyces cerevisiae. Genetics 168: 1805–1816. Bartrand, A. J., D. Iyasu and G. S. Brush, 2004 DNA stimulates Mec1-mediated phosphorylation of replication protein A. J. Biol. Chem. 279: 26762–26767. Bashkirov, V. I., J. S. King, E. V. Bashkirova, J. Schmuckli-Maurer and W. D. Heyer, 2000 DNA repair protein Rad55 is a terminal substrate of the DNA damage checkpoints. Mol. Cell. Biol. 20: 4393–4404. Bastin-Shanower, S. A., and S. J. Brill, 2001 Functional analysis of the four DNA binding domains of replication protein A. The role of RPA2 in ssDNA binding. J. Biol. Chem. 276: 36446–36453. Baudat, F., and A. Nicolas, 1997 Clustering of meiotic doublestrand breaks on yeast chromosome III. Proc. Natl. Acad. Sci. USA 94: 5213–5218. Binz, S. K., Y. Lao, D. F. Lowry and M. S. Wold, 2003 The phosphorylation domain of the 32-kDa subunit of replication protein A (RPA) modulates RPA-DNA interactions. Evidence for an intersubunit interaction. J. Biol. Chem. 278: 35584–35591. Bishop, D. K., and D. Zickler, 2004 Early decision: meiotic crossover interference prior to stable strand exchange and synapsis. Cell 117: 9–15. Bradford, M. M., 1976 A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72: 248–254. Brush, G. S., and T. J. Kelly, 2000 Phosphorylation of the replication protein A large subunit in the Saccharomyces cerevisiae checkpoint response. Nucleic Acids Res. 28: 3725–3732. Brush, G. S., D. M. Morrow, P. Hieter and T. J. Kelly, 1996 The ATM homologue MEC1 is required for phosphorylation of replication protein A in yeast. Proc. Natl. Acad. Sci. USA 93: 15075– 15080. Brush, G. S., D. M. Clifford, S. M. Marinco and A. J. Bartrand, 2001 Replication protein A is sequentially phosphorylated during meiosis. Nucleic Acids Res. 29: 4808–4817. Carty, M. P., M. Zernik-Kobak, S. McGrath and K. Dixon, 1994 UV light-induced DNA synthesis arrest in HeLa cells is associated with changes in phosphorylation of human singlestranded DNA-binding protein. EMBO J. 13: 2114–2123. Cha, R. S., and N. Kleckner, 2002 ATR homolog Mec1 promotes fork progression, thus averting breaks in replication slow zones. Science 297: 602–606. Chua, P. R., and G. S. Roeder, 1997 Tam1, a telomere-associated meiotic protein, functions in chromosome synapsis and crossover interference. Genes Dev. 11: 1786–1800. Clifford, D. M., S. M. Marinco and G. S. Brush, 2004 The meiosisspecific protein kinase Ime2 directs phosphorylation of replication protein A. J. Biol. Chem. 279: 6163–6170.

38


Cobb, J. A., L. Bjergbaek, K. Shimada, C. Frei and S. M. Gasser, 2003 DNA polymerase stabilization at stalled replication forks requires Mec1 and the RecQ helicase Sgs1. EMBO J. 22: 4325–4336. de la Torre-Ruiz, M. A., and N. F. Lowndes, 2000 The Saccharomyces cerevisiae DNA damage checkpoint is required for efficient repair of double-strand breaks by non-homologous end joining. FEBS Lett. 467: 311–315. Desany, B. A., A. A. Alcasabas, J. B. Bachant and S. J. Elledge, 1998 Recovery from DNA replicational stress is the essential function of the S-phase checkpoint pathway. Genes Dev. 12: 2956–2970. Din, S., S. J. Brill, M. P. Fairman and B. Stillman, 1990 Cell-cycleregulated phosphorylation of DNA replication factor A from human and yeast cells. Genes Dev. 4: 968–977. Downs, J. A., N. F. Lowndes and S. P. Jackson, 2000 A role for Saccharomyces cerevisiae histone H2A in DNA repair. Nature 408: 1001–1004. Durocher, D., and S. P. Jackson, 2001 DNA-PK, ATM and ATR as sensors of DNA damage: Variations on a theme? Curr. Opin. Cell Biol. 13: 225–231. Eggler, A. L., R. B. Inman and M. M. Cox, 2002 The Rad51-dependent pairing of long DNA substrates is stabilized by replication protein A. J. Biol. Chem. 277: 39280–39288. Gerton, J. L., J. DeRisi, R. Shroff, M. Lichten, P. O. Brown et al., 2000 Global mapping of meiotic recombination hotspots and coldspots in the yeast Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 97: 11383–11390. Grushcow, J. M., T. M. Holzen, K. J. Park, T. Weinert, M. Lichten et al., 1999 Saccharomyces cerevisiae checkpoint genes MEC1, RAD17 and RAD24 are required for normal meiotic recombination partner choice. Genetics 153: 607–620. Kane, S. M., and R. Roth, 1974 Carbohydrate metabolism during ascospore development in yeast. J. Bacteriol. 118: 8–14. Kato, R., and H. Ogawa, 1994 An essential gene, ESR1, is required for mitotic cell growth, DNA repair and meiotic recombination in Saccharomyces cerevisiae. Nucleic Acids Res. 22: 3104–3112. Kim, S. T., D. S. Lim, C. E. Canman and M. B. Kastan, 1999 Substrate specificities and identification of putative substrates of ATM kinase family members. J. Biol. Chem. 274: 37538–37543. Koundrioukoff, S., S. Polo and G. Almouzni, 2004 Interplay between chromatin and cell cycle checkpoints in the context of ATR/ATM-dependent checkpoints. DNA Repair 3: 969–978. Liberi, G., I. Chiolo, A. Pellicioli, M. Lopes, P. Plevani et al., 2000 Srs2 DNA helicase is involved in checkpoint response and its regulation requires a functional Mec1-dependent pathway and Cdk1 activity. EMBO J. 19: 5027–5038. Liu, V. F., and D. T. Weaver, 1993 The ionizing radiation-induced replication protein A phosphorylation response differs between ataxia telangiectasia and normal human cells. Mol. Cell. Biol. 13: 7222–7231. Luria, S. E., and M. Delbruck, 1943 Mutations of bacteria from virus sensitivity to virus resistance. Genetics 28: 491–511. Lydall, D., Y. Nikolsky, D. K. Bishop and T. Weinert, 1996 A meiotic recombination checkpoint controlled by mitotic checkpoint genes. Nature 383: 840–843. Maguire, M. P., 1980 Adaptive advantage for chiasma interference: a novel suggestion. Heredity 45: 127–131. Mallory, J. C., V. I. Bashkirov, K. M. Trujillo, J. A. Solinger, M. Dominska et al., 2003 Amino acid changes in Xrs2p, Dun1p and Rfa2p that remove the preferred targets of the ATM family of protein kinases do not affect DNA repair or telomere length in Saccharomyces cerevisiae. DNA Repair 2: 1041–1064. Merrill, B. J., and C. Holm, 1999 A requirement for recombinational repair in Saccharomyces cerevisiae is caused by DNA replication defects of mec1 mutants. Genetics 153: 595–605. Morgan, S. E., and M. B. Kastan, 1997 Dissociation of radiationinduced phosphorylation of replication protein A from the S-phase checkpoint. Cancer Res. 57: 3386–3389. Mortimer, R. K., D. Schild, C. R. Contopoulou and J. A. Kans, 1989 Genetic map of Saccharomyces cerevisiae, Edition 10. Yeast 5: 321–403. Muller, H. J., 1916 The mechanism of crossing-over. Am. Nat. 50: 193–221, 284–305, 350–366, 421–434. Myung, K., A. Datta and R. D. Kolodner, 2001 Suppression of spontaneous chromosomal rearrangements by S phase checkpoint functions in Saccharomyces cerevisiae. Cell 104: 397–408.

Nakagawa, T., and H. Ogawa, 1999 The Saccharomyces cerevisiae MER3 gene, encoding a novel helicase-like protein, is required for crossover control in meiosis. EMBO J. 18: 5714–5723. Novak, J. E., P. B. Ross-Macdonald and G. S. Roeder, 2001 The budding yeast Msh4 protein functions in chromosome synapsis and the regulation of crossover distribution. Genetics 158: 1013–1025. Oakley, G. G., S. M. Patrick, J. Yao, M. P. Carty, J. J. Turchi et al., 2003 RPA phosphorylation in mitosis alters DNA binding and protein-protein interactions. Biochemistry 42: 3255–3264. O’Driscoll, M., V. L. Ruiz-Perez, C. G. Woods, P. A. Jeggo and J. A. Goodship, 2003 A splicing mutation affecting expression of ataxia-telangiectasia and Rad3-related protein (ATR) results in Seckel syndrome. Nat. Genet. 33: 497–501. Papazian, H. P., 1952 The analysis of tetrad data. Genetics 37: 175–188. Paulovich, A. G., and L. H. Hartwell, 1995 A checkpoint regulates the rate of progression through S phase in S. cerevisiae in response to DNA damage. Cell 82: 841–847. Perkins, D. D., 1949 Biochemical mutants in the smut fungus Ustilago maydis. Genetics 34: 607–626. Pochart, P., D. Woltering and N. M. Hollingsworth, 1997 Conserved properties between functionally distinct MutS homologs in yeast. J. Biol. Chem. 272: 30345–30349. Reenan, R. A. G., and R. D. Kolodner, 1992 Characterization of insertion mutations in the Saccharomyces cerevisiae MSH1 and MSH2 genes: evidence for separate mitochondrial and nuclear functions. Genetics 132: 975–985. Sanchez, Y., B. A. Desany, W. J. Jones, Q. Liu, B. Wang et al., 1996 Regulation of RAD53 by the ATM-like kinases MEC1 and TEL1 in yeast cell cycle checkpoint pathways. Science 271: 357–360. Savitsky, K., A. Bar-Shira, S. Gilad, G. Rotman, Y. Ziv et al., 1995 A single ataxia telangiectasia gene with a product similar to PI-3 kinase. Science 268: 1749–1753. Scherer, S., and R. W. Davis, 1979 Replacement of chromosome segments with altered DNA sequences constructed in vitro. Proc. Natl. Acad. Sci. USA 76: 4951–4955. Sherman, F., and P. Wakem, 1991 Mapping yeast genes. Methods Enzymol. 194: 38–57. Shinohara, M., K. Sakai, A. Shinohara and D. K. Bishop, 2003 Crossover interference in Saccharomyces cerevisiae requires a TID1/RDH54- and DMC1-dependent pathway. Genetics 163: 1273–1286. Siede, W., J. B. Allen, S. J. Elledge and E. C. Friedberg, 1996 The Saccharomyces cerevisiae MEC1 gene, which encodes a homolog of the human ATM gene product, is required for G1 arrest following radiation treatment. J. Bacteriol. 178: 5841–5843. Stuart, D., and C. Wittenberg, 1998 CLB5 and CLB6 are required for premeiotic DNA replication and activation of the meiotic S/ M checkpoint. Genes Dev. 12: 2698–2710. Sun, Z., D. S. Fay, F. Marini, M. Foiani and D. F. Stern, 1996 Spk1/Rad53 is regulated by Mec1-dependent protein phosphorylation in DNA replication and damage checkpoint pathways. Genes Dev. 10: 395–406. Sung, P., L. Krejci, S. Van Komen and M. G. Sehorn, 2003 Rad51 recombinase and recombination mediators. J. Biol. Chem. 278: 42729–42732. Sym, M., and G. S. Roeder, 1994 Crossover interference is abolished in the absence of a synaptonemal complex protein. Cell 79: 283– 292. Thomas, B. J., and R. Rothstein, 1989 Elevated recombination rates in transcriptionally active DNA. Cell 56: 619–630. Thompson, D. A., and F. W. Stahl, 1999 Genetic control of recombination partner preference in yeast meiosis: isolation and characterization of mutants elevated for meiotic unequal sisterchromatid recombination. Genetics 153: 621–641. Tsubouchi, H., and H. Ogawa, 2000 Exo1 roles for repair of DNA double-strand breaks and meiotic crossing over in Saccharomyces cerevisiae. Mol. Biol. Cell 11: 2221–2233. Vallen, E. A., and F. Cross, 1995 Mutations in RAD27 define a potential link between G1 cyclins and DNA replication. Mol. Cell. Biol. 15: 4291–4302. Vassin, V. M., M. S. Wold and J. A. Borowiec, 2004 Replication protein A (RPA) phosphorylation prevents RPA association with replication centers. Mol. Cell. Biol. 24: 1930–1943.

RPA Phosphorylation and Recombination Weinert, T. A., G. L. Kiser and L. H. Hartwell, 1994 Mitotic checkpoint genes in budding yeast and the dependence of mitosis on DNA replication and repair. Genes Dev. 8: 652–665. Wold, M. S., 1997 Replication protein A: a heterotrimeric, singlestranded DNA-binding protein required for eukaryotic DNA metabolism. Annu. Rev. Biochem. 66: 61–92. Zernik-Kobak, M., K. Vasunia, M. Connelly, C. W. Anderson and K. Dixon, 1997 Sites of UV-induced phosphorylation of the p34

39

subunit of replication protein A from HeLa cells. J. Biol. Chem. 272: 23896–23904. Zhou, B. B. S., and S. J. Elledge, 2000 The DNA damage response: putting checkpoints in perspective. Nature 408: 433–439. Zou, L., and S. J. Elledge, 2003 Sensing DNA damage through ATRIP recognition of RPA-ssDNA complexes. Science 300: 1542–1548. Communicating editor: L. S. Symington