DNA Stimulates Mec1-mediated Phosphorylation of Replication Protein A*

The cellular single-stranded DNA (ssDNA)-binding protein replication protein A (RPA) becomes phosphorylated periodically during the normal cell cycle and also in response to DNA damage. In Saccharomyces cerevisiae, RPA phosphorylation requires the checkpoint protein Mec1, a protein kinase homologous in structure and function to human ATR. We confirm here that immunocomplexes containing a tagged version of Mec1 catalyze phosphorylation of purified RPA, likely reflecting an RPA kinase activity intrinsic to Mec1. A significant stimulation of this activity is observed upon the addition of covalently closed ssDNA derived from the bacteriophage M13. This stimulation is not observed with mutant RPA deficient for DNA binding, indicating that DNA-bound RPA is a preferred substrate. Stimulation is also observed upon the addition of linear ssDNA homopolymers or hydrolyzed M13 ssDNA. In contrast to circular ssDNA, these DNA cofactors stimulate both wild type and mutant RPA phosphorylation. This finding suggests that linear ssDNA can also stimulate Mec1-mediated RPA phosphorylation by activating Mec1 or an associated protein. Although the Mec1-interacting protein Ddc2 is required for RPA phosphorylation in vivo, it is required for neither basal nor ssDNA-stimulated RPA phosphorylation in vitro. Therefore, activation of Mec1-mediated RPA phosphorylation by either circular or linear ssDNA does not operate through Ddc2. Our results provide insight into the mechanisms that function in vivo to specifically induce RPA phosphorylation upon initiation of DNA replication, repair, or recombination.

The cellular single-stranded DNA (ssDNA)-binding protein replication protein A (RPA) becomes phosphorylated periodically during the normal cell cycle and also in response to DNA damage. In Saccharomyces cerevisiae, RPA phosphorylation requires the checkpoint protein Mec1, a protein kinase homologous in structure and function to human ATR. We confirm here that immunocomplexes containing a tagged version of Mec1 catalyze phosphorylation of purified RPA, likely reflecting an RPA kinase activity intrinsic to Mec1. A significant stimulation of this activity is observed upon the addition of covalently closed ssDNA derived from the bacteriophage M13. This stimulation is not observed with mutant RPA deficient for DNA binding, indicating that DNA-bound RPA is a preferred substrate. Stimulation is also observed upon the addition of linear ssDNA homopolymers or hydrolyzed M13 ssDNA. In contrast to circular ssDNA, these DNA cofactors stimulate both wild type and mutant RPA phosphorylation. This finding suggests that linear ssDNA can also stimulate Mec1-mediated RPA phosphorylation by activating Mec1 or an associated protein. Although the Mec1-interacting protein Ddc2 is required for RPA phosphorylation in vivo, it is required for neither basal nor ssDNA-stimulated RPA phosphorylation in vitro. Therefore, activation of Mec1mediated RPA phosphorylation by either circular or linear ssDNA does not operate through Ddc2. Our results provide insight into the mechanisms that function in vivo to specifically induce RPA phosphorylation upon initiation of DNA replication, repair, or recombination.
In the budding yeast Saccharomyces cerevisiae, the protein kinase Mec1 is required both for checkpoint-associated processes that delay cell cycle progression following genotoxic stress (1)(2)(3) and for more direct regulation of various DNA repair processes (4 -6). Mec1 is structurally and functionally related to human ATR (7), which is underexpressed in Seckel syndrome (8), and to ATM, which is mutated in ataxia-telangiectasia (9). Mec1 is also similar to the catalytic subunit of DNA-activated protein kinase (10), which is required for V(D)J recombination and DNA double-strand break repair (for review, see Ref. 11). It is believed that Mec1 and its human homologues are all primary effectors in the DNA damage response, catalyzing phosphorylation of numerous proteins involved in managing genotoxic stress.
We have shown previously that Mec1 is required for phosphorylation of replication protein A (RPA) 1 (12)(13)(14), a heterotrimeric complex that binds strongly to single-stranded DNA (ssDNA) (for review, see Ref. 15). RPA stabilizes ssDNA during DNA replication, repair, and recombination and also interacts directly with other proteins involved in these processes. During normal cell cycle progression, Mec1 directs phosphorylation of the RPA middle subunit (Rfa2) (12). Upon genotoxic stress, Mec1 directs phosphorylation of both Rfa2 (12) and the RPA large subunit (Rfa1) (13), which contains the major ssDNA binding activity of the heterotrimer (15). Mec1 is also required for an RPA phosphorylation reaction during meiosis that is specifically induced upon initiation of recombination (14). In human cells, ATM is required for proper RPA middle subunit phosphorylation in response to ionizing radiation (16). Therefore, a general pathway leading to RPA phosphorylation has been evolutionarily conserved.
Although the exact role of RPA phosphorylation remains unclear, there is accumulating evidence that phosphorylation of the middle subunit regulates DNA metabolism rather than checkpoint-associated processes. For instance, studies in both yeast and human cells have failed to reveal a correlation between RPA middle subunit phosphorylation and the arrest of cell cycle or meiotic progression (12,14,17). In addition, the extensively phosphorylated RPA isolated from DNA-damaged human cells does not support SV40 DNA replication in vitro (18 -20), and phosphorylated RPA isolated from M phase cells exhibits a lower affinity for double-stranded DNA (dsDNA) and DNA polymerase ␣ (21). Furthermore, "phosphomimetic" versions of human RPA exhibit a decreased capacity to destabilize dsDNA in vitro (22) and an apparent decreased capacity to associate with DNA replication centers in vivo (23). Finally, our recent studies have indicated that a phosphomimetic mutation in the yeast RPA middle subunit confers a meiotic recombination phenotype. 2 To investigate the mechanism underlying Mec1-dependent RPA phosphorylation, we examined the reaction in vitro. We demonstrate here that ssDNA can stimulate RPA phosphorylation through two different mechanisms. One of these mechanisms involves activation of RPA, whereas the other appears to involve activation of Mec1. Such DNA stimulation helps to explain the specific induction of RPA phosphorylation that occurs upon initiation of various DNA transactions. * This work was supported by Grant GM61860 from the National Institutes of Health, Grant RPG-00-211-01-CCG from the American Cancer Society, and funds from the Karmanos Cancer Institute. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

EXPERIMENTAL PROCEDURES
Expression Plasmids-A high copy 2-based URA3-containing plasmid (pHA-Mec1) encoding galactose-inducible Mec1 with three hemagglutinin epitope (HA) tags at the N terminus was generously provided by Ted Weinert (University of Arizona). A mutant version of this plasmid (pHA-Mec1 kd ) encoding kinase-dead Mec1 was constructed by the deletion of a 207-base pair BglII fragment from the region encoding the Mec1 kinase domain. Cells were transformed with these plasmids by a lithium acetate procedure (24).
Irradiation-Proliferating cells were placed on ice and exposed to 20 kilorads of ionizing radiation at 552 rads/min with a Shepard Mark I Model 68 137 Cs irradiator. The cells were returned to 30°C for 30 min (see Fig. 6A) or 1 h (see Fig. 1C) prior to analysis.
Cell Extracts-Denatured cell extracts used for Western blot analysis of Rfa1 and Rfa2 were prepared by freezing and heating cells in SDS-PAGE sample buffer (14). Native cell extracts were prepared from spheroplasts. The cells were first grown to stationary phase in SC Gly -Ura and then used to inoculate 1 liter of SC Gal -Ura at a starting A 600 of ϳ0.2. After growth into the exponential phase (A 600 of ϳ1.0), the cells were harvested by centrifugation at 4000 ϫ g for 10 min (this and all other centrifugations were carried out at 4°C) and washed once with cold water. The cells were then collected by centrifugation at 1850 ϫ g for 5 min, resuspended in 0.1 M EDTA and 10 mM dithiothreitol (1 ml/0.4 g of cells), and incubated with rotation for 10 min at 30°C. The cells were collected again by centrifugation at 1850 ϫ g for 5 min and resuspended in YPS (1% yeast extract, 2% peptone, 1 M sorbitol) (1 ml/g of cells). Zymolyase 100T (US Biological) was added (0.5 mg/2.5 g cells), and the suspension was incubated at 30°C for 1.75 h. Resulting spheroplasts were collected by centrifugation for 10 min at 500 ϫ g and subsequently washed twice with YPS. The pellet was then resuspended in lysis buffer (50 mM Hepes, pH 7.4, 100 mM KCl, 0.1 mM EDTA, 0.2% Tween 20, 1 mM dithiothreitol, 25 mM NaF, 100 M Na 3 VO 4 , 1 mM phenylmethylsulfonyl fluoride, 1 g/ml leupeptin, 1 mM benzamidine) and incubated on ice for 30 min. The suspension was then centrifuged at 12,000 ϫ g for 10 min, and the supernatant (cell extract) was distributed into aliquots and stored at Ϫ80°C.
Western Blot Analysis-For detection of Rfa1 and Rfa2 phosphoisomers, denatured cell extract was subjected to electrophoresis through a 12% low cross-linking SDS-polyacrylamide gel (150:1 w/w acrylamide: bisacrylamide). For detection of HA-tagged Mec1 proteins, native cell extract or immunoprecipitate was subjected to electrophoresis through a 5% SDS-polyacrylamide gel with normal cross-linking (37.5:1 w/w acrylamide:bisacrylamide). For detection of tubulin, native cell extract was subjected to electrophoresis through an 8% SDS-polyacrylamide gel with normal cross-linking. The separated proteins were then transferred to a 0.2 nitrocellulose membrane (Schleicher and Schuell) in 25 mM Tris, 192 mM glycine, 20% methanol (for RPA and HA-tagged Mec1 proteins) or 10 mM CAPS-NaOH, pH 11, 10% methanol (for tubulin). Immobilized proteins were immunostained by incubation with rabbit polyclonal antibodies directed against Rfa1 or Rfa2 (14), mouse monoclonal antibody directed against the HA epitope (HA.11, Covance), or rat monoclonal antibody directed against tubulin (YOL1/34, Serotec) followed by incubation with horseradish peroxidase-linked goat antirabbit, anti-mouse, or anti-rat antibody (Pierce), respectively. Protein bands were visualized by autoradiography following incubation with chemiluminescence reagents (Pierce).
Proteins-RPA was purified from W303-1A and DLY285 cells by Affi-Gel Blue (Bio-Rad), ssDNA-cellulose (Sigma), and Mono Q (Amersham Biosciences) chromatography based on methods described previously (33,34). RPA concentration was determined by the Bradford assay (35) using bovine serum albumin as a standard. A purified preparation of recombinant mutant RPA defective for DNA binding (A-B-) and its wild type counterpart (36) were very generously provided by Steven Brill (Rutgers University). Recombinant rat PHAS-I was purchased from Stratagene.
Protein Kinase Assay-For each reaction mixture, immunocomplex was generated by the incubation of 2 mg of native cell extract with 10 -14 g of HA.11 antibody in a final volume of 0.5 ml for 2 h at 4°C. YGB33 and YGB40 were the sources of the cell extract for the experiments shown in Figs. 2-5, and YGB362, YGB363, YGB368, and YGB369 were the sources of the cell extract for the experiment shown in Fig. 6. EZview TM Red protein G affinity gel (Sigma) was then added (20 l of a 50% slurry), and incubation was continued for an additional 2 h at 4°C, this time with mixing. The Protein G beads were collected by centrifugation (1850 ϫ g, 1 min, 4°C) and washed twice with lysis buffer containing 0.5 M NaCl, twice with lysis buffer alone, and three times with kinase buffer (10 mM (37). Reaction mixtures were incubated at 30°C for 40 min and then subjected to 15% SDS-polyacrylamide gel electrophoresis. The gels were dried on 3MM paper (Whatman), and incorporation of radioactive inorganic phosphate into protein was detected by autoradiography and Storm PhosphorImager (Amersham Biosciences) analysis.

Immunoprecipitates of Mec1
Catalyze RPA Phosphorylation-To analyze Mec1-dependent RPA phosphorylation in vitro, we employed episomal galactose-inducible genes encoding HA-tagged Mec1 (HA-Mec1) and its kinase-dead counterpart (HA-Mec1 kd ). The two versions of Mec1 were expressed at equal levels in the presence of galactose (Fig. 1A). Expression of HA-Mec1 in a mec1 mutant rescued Rfa2 phosphorylation during normal growth and both Rfa2 and Rfa1 phosphorylation in response to DNA damage, whereas HA-Mec1 kd did not (Fig. 1,  B and C). Therefore, the tagged versions of Mec1 behaved in vivo as expected for the wild type protein. In addition, we found that expression of HA-Mec1 kd did not interfere with RPA phosphorylation in wild type cells (Fig. 1, B and C). It is noted that detectable Rfa2 and Rfa1 phosphorylation was observed upon irradiation in the absence of Mec1 (Fig. 1C, left panels). A portion of this residual activity is most likely due to the presence of Tel1, a protein similar in structure and function to Mec1 that is often considered to be the ATM homologue in yeast (38). However, the magnitude of the residual phosphorylation appeared slightly greater in cells harboring pHA-Mec1, suggesting that some low level expression might occur in the presence of glucose.
HA-Mec1 and HA-Mec1 kd were isolated by immunoprecipitation using antibody directed against HA and tested for RPA kinase activity. Although nearly identical levels of the two proteins were routinely obtained in the immune pellets ( Fig.  2A, upper panel), significant Rfa2 phosphorylation was only supported by HA-Mec1 immunoprecipitates ( Fig. 2A, lower  panel). These data strongly suggest that Mec1 is an RPA kinase and further suggest that Mec1 directly catalyzes RPA phosphorylation in vivo. In some experiments, we also observed a low level of apparent Rfa1 phosphorylation under these conditions (see the first two lanes in Fig. 2C). However, we note that the presence of a background band often obscured this activity even in the presence of stimulatory DNA cofactor (see next section). Two recent reports have presented similar Mec1-dependent RPA kinase activities in vitro (39,40).
DNA Activates Mec1-dependent RPA Phosphorylation in Vitro-Phosphorylation of human RPA is catalyzed by purified versions of ATM (41) or the DNA-activated protein kinase catalytic subunit (42), which are both structurally similar to Mec1. Because DNA can stimulate these reactions, we tested the effect of DNA on Mec1-mediated RPA phosphorylation. We found that M13 ssDNA significantly stimulated Rfa2 and Rfa1 phosphorylation in a Mec1-dependent manner (Fig. 2B), and an increase in DNA concentration led to an increase in phosphorylation of both subunits (Fig. 2C). We have shown previously that RPA phosphorylation mediated by the meiosis-specific protein kinase Ime2 is not stimulated by DNA (43). Therefore, DNA does not play a general role in activating phosphorylation of yeast RPA. Finally, we have consistently observed greater phosphorylation of Rfa2 than Rfa1, indicating that Rfa1 either contains fewer phosphorylation sites or is simply a less efficient substrate than Rfa2.
Natural ssDNA molecules from bacterial viruses such as M13 have regions of secondary structure (44) and therefore contain both ssDNA and dsDNA. To test the effectiveness of pure ssDNA, we employed long homopolymers (poly(dA) or poly(dT) populations containing molecules with hundreds to thousands of nucleotides) and found efficient stimulation of Rfa2 phosphorylation (Fig. 3A). Inclusion of an RNA homopolymer did not activate Rfa2 phosphorylation (Fig. 3A), confirming the DNA specificity of the observed stimulation. As shown in Fig. 3B, we could detect a similar profile of DNA-stimulated Rfa1 phosphorylation despite the presence of the background band. It is interesting to note that both ssDNA homopolymers were more effective in stimulating RPA phosphorylation than M13 ssDNA. To rule out the possibility that differences in DNA concentrations were responsible for these disparate effects, we increased the M13 ssDNA concentration and found no further stimulation of Rfa2 phosphorylation activity (data not shown). Therefore, the conditions used were maximal for activation of Rfa2 phosphorylation, suggesting that the differences in degree of stimulation resulted from differences in the ssDNA structures.
Because RPA is also capable of binding short ssDNA molecules, we were interested in determining whether oligonucleotides could stimulate Mec1-mediated RPA phosphorylation. We employed a series of oligo(dT) molecules and found that a polymer as short as 7 nucleotides modestly stimulated Mec1dependent Rfa2 phosphorylation, whereas a 3-nucleotide homopolymer was incapable of stimulating the reaction (Fig. 3A). We observed slightly increased stimulation with 50-and 70nucleotide polymers, but the degree of stimulation was much less than that observed with the longer poly(dA) and poly(dT) molecules (Fig. 3A). We further investigated the specificity of DNA stimulation by adding duplex DNA molecules. We first employed M13 RFI, a replicative form that is simply a double-stranded version of M13 ssDNA (see Ref. 44), and found that it efficiently stimulated the reaction (Fig. 3A). However, subsequent experiments showed a variability in the effectiveness of M13 RF I that was strictly dependent on the preparation (Fig. 3C). Our analysis suggested that the M13 RF I capable of stimulating RPA phosphorylation contained a significant quantity of residual M13 ssDNA (data not shown). Titration with duplex plasmid DNA provided a clear indication that covalently closed dsDNA was incapable of stimulating Mec1-mediated RPA phosphorylation (Fig. 3C).
M13 ssDNA Stimulation Is Substrate-specific-To determine whether M13 ssDNA might have a direct effect on Mec1-dependent activity, we examined Mec1-mediated phosphorylation of another protein substrate. We employed purified recombinant rat PHAS-I, a regulator of translation that serves as an effective substrate of Mec1 in vitro (45). Although our immunoprecipitated Mec1 supported PHAS-I phosphorylation, M13 ssDNA did not stimulate the reaction as seen with RPA (Fig. 4). Experiments employing a higher PHAS-I concentration gave identical results (data not shown). Therefore, M13 ssDNA does not appear to generally affect Mec1-dependent protein kinase activity. Unfortunately, attempts to define the substrate specificity with linear ssDNA failed, as an unacceptable level of background Mec1-independent phosphorylation was observed in reaction mixtures containing HA-Mec1 kd immunoprecipitates, poly(dT), and protein substrates PHAS-I or casein (data not shown). Although some background activity was also observed when RPA was tested as the substrate (for example, see Fig. 5A), a much higher signal-to-noise ratio was observed under these conditions.

Analysis of DNA-binding Defective RPA Reveals Two ssDNA
Stimulatory Mechanisms-To further investigate the role of RPA in the M13 ssDNA stimulatory effect, we employed recombinant RPA containing mutations in the A and B DNA-binding domains of Rfa1. This mutant form of RPA is incapable of binding short oligonucleotides and is extremely impaired for binding to longer ssDNA molecules (36). We found that M13 ssDNA could stimulate Mec1-dependent phosphorylation of the wild type recombinant RPA control but not the mutant form (Fig. 5). These results indicate that RPA-DNA binding is required for the stimulation observed in the presence of covalently closed ssDNA.
In contrast to M13 ssDNA, poly(dT) and poly(dA) stimulated Mec1-mediated phosphorylation of either wild type or DNAbinding defective mutant RPA (Fig. 5). One interpretation of these data is that linear ssDNA molecules can stimulate Mec1mediated RPA phosphorylation through a second mechanism. To further explore this possibility, we compared the activation by intact M13 ssDNA and M13 ssDNA digested with HhaI restriction endonuclease. This digestion generates numerous linear ssDNA molecules theoretically ranging in size from 8 to 966 nucleotides. We observed an enhanced stimulation of wild type RPA phosphorylation with the digested M13 ssDNA (Fig.  5B). Furthermore, this preparation was capable of stimulating phosphorylation of DNA-binding defective RPA (Fig. 5B). Therefore, a change in M13 ssDNA structure significantly alters its stimulatory properties.
DNA-stimulated RPA Phosphorylation Does Not Require Ddc2-Mec1 stably interacts with Ddc2 (28 -30), an orthologue of the ATR-interacting protein ATRIP (46). Ddc2 and ATRIP are thought to function similarly, directing their cognate kinases to sites of DNA damage (47,48). To directly test whether Ddc2 is required for Mec1-dependent RPA kinase activity, we generated ddc2 deletion mutants and tested for RPA phosphorylation both in vivo and in vitro. We found that Ddc2 was required for normal levels of RPA phosphorylation during proliferation and in response to DNA damage (Fig. 6A). However, immunoprecipitates of HA-Mec1 isolated from pHA-Mec1transformed mec1 ddc2 cells contained a fully functional Mec1dependent RPA kinase (Fig. 6, B and C). Therefore, Ddc2 is not required for basal Mec1-mediated RPA phosphorylation. Importantly, activation of Mec1-dependent RPA phosphorylation by either M13 ssDNA or poly(dT) was also independent of Ddc2 (Fig. 6, B and C). Therefore, the reported DNA binding activity of Ddc2 is not responsible for either form of ssDNA-stimulated Mec1-mediated RPA phosphorylation. DISCUSSION Mechanisms that regulate the timing and extent of protein phosphorylation can involve the protein kinase that catalyzes the reaction or the protein substrate that becomes phosphorylated. We have shown here that naked DNA stimulates Mec1mediated RPA phosphorylation in vitro. Based on our analyses of different DNA cofactors and different protein substrates, we propose that DNA imparts its control on RPA phosphorylation in two ways. One mechanism depends on RPA interaction with DNA and is therefore substrate-driven. The other does not appear to rely on RPA-DNA interaction and could involve interaction of DNA with Mec1 itself or an associated protein.
The stimulation of RPA phosphorylation observed upon the addition of covalently closed ssDNA occurs upon the binding of RPA to ssDNA. This activation could result from a conformational change that unveils RPA phosphoacceptors targeted by Mec1 catalysis. In this model, RPA becomes a more effective Mec1 substrate when bound to ssDNA, leading to RPA phosphorylation specifically when ssDNA is generated. Such a stimulatory mechanism has been reported previously for DNAactivated protein kinase-catalyzed phosphorylation of human RPA (49). By necessity, this mode of stimulation would not regulate the initial interaction between RPA and DNA, but instead would regulate a subsequent step such as dissociation from DNA, interaction with other DNA structures, or interaction with other proteins involved in DNA metabolism.
In the case of linear ssDNA, activation of RPA phosphorylation is not entirely dependent on RPA-ssDNA interaction, as stimulation is observed with a DNA-binding defective RPA mutant. It is important to note that this RPA mutant is not completely incapable of binding ssDNA (36). Therefore, the possibility that DNA-binding defective RPA binds ssDNA under the conditions that we have employed must be considered. However, such an explanation would require specific interaction of the mutant RPA with long ssDNA homopolymers and linearized M13 ssDNA but not with circular M13 ssDNA. Given that this specificity is unlikely, a strong possibility is that stimulation by linear ssDNA involves activation of enzyme, either Mec1 itself or a Mec1-interacting protein that is required for RPA phosphorylation. Unlike the substrate-driven mechanism, such an enzyme-driven mechanism would not necessarily restrict the targeted RPA to the DNA-bound state and therefore would not limit the role of this reaction to a post-ssDNA-binding RPA function. Nonetheless, it seems likely that an ssDNA-bound form of Mec1 would first encounter RPA that is bound to the same ssDNA molecule. It will be interesting to determine whether the two ssDNA stimulatory mechanisms lead to phosphorylation of a common set of residues or generate different RPA phosphoisomers with different functions.
An obvious distinction between linear and circular ssDNA is the presence of ends. An intriguing possibility is that ssDNA ends activate Mec1 to catalyze phosphorylation of RPA and perhaps other proteins as well. With this possibility in mind, it is important to note the degree of stimulation observed in the presence of short ssDNA molecules, which also provide ssDNA ends. In the case of oligo(dT) 7 , putative RPA binding sites were present in ϳ10-fold molar excess over RPA; therefore, the majority of the ssDNA ends in this reaction were not hidden by bound RPA. Nonetheless, only modest stimulation was observed. Therefore, ssDNA ends alone cannot account for Mec1 activation, and it is possible that a minimum DNA length is required. The low level activation of RPA phosphorylation observed with oligonucleotides probably indicates a simple transition from unbound to DNA-bound RPA, leading to a stimulation that is mechanistically similar to that observed with circular M13 ssDNA. As the length of ssDNA is increased, RPA phosphorylation is augmented because of the increased influence of both stimulatory mechanisms.
Induction of RPA phosphorylation in vivo occurs during the S phase of the normal cell cycle and in response to various forms of DNA damage (12,13,16,18,50,51). Both ssDNA and ssDNA ends (either in the context of duplex DNA or freely exposed) are generated during these processes and would conceivably be available to activate RPA phosphorylation. A recent study suggests that Mec1 and its human homologue, ATR, are recruited FIG. 6. Ddc2 is required for RPA phosphorylation in vivo but not in vitro. A, Western blot analysis of RPA from exponentially proliferating (left) or irradiated (right) cells. P-Rfa1, phosphorylated Rfa1; P-Rfa2, phosphorylated Rfa2. B and C, RPA kinase assays in the presence of M13 ssDNA (ss) or poly(dT) (dTn) (both at 4.8 g/ml) employing HA-Mec1 and HA-Mec1 kd immunocomplexes prepared from wild type or mec1 ddc2 cells harboring pHA-Mec1 or pHA-Mec1 kd .
to sites of DNA damage through direct association of their respective partners, Ddc2 and ATRIP, with ssDNA-bound RPA (48). We have shown here that Ddc2 is required for proper RPA phosphorylation during the normal cell cycle and in response to DNA damage but not for RPA phosphorylation in vitro. Such a dichotomy has been observed previously with Rad53 phosphorylation (30), which is catalyzed in a Mec1-and Ddc2-dependent manner upon genotoxic stress (52,53). These observations indicate that Ddc2 is not required for the protein kinase activity associated with Mec1 but support the proposed auxiliary role in which Ddc2 directs Mec1 to sites of DNA damage (47,48). However, it should be noted that Mec1 is required for proper formation of Ddc2 foci following DNA damage (54), suggesting that Mec1 is not simply a passenger in the DNA localization process. Furthermore, there is considerable evidence that ATR can directly interact with both undamaged and damaged DNA (48,55,56), and it is possible that Mec1 shares this capacity. We propose that once positioned correctly, perhaps initially through direct interaction between Ddc2 and RPA, Mec1 encounters and catalyzes phosphorylation of ssDNA-bound RPA. This reaction would lead to specific and localized RPA phosphorylation. Subsequent encounters with ssDNA, possibly exposed ssDNA ends, could then stimulate Mec1 to catalyze further RPA phosphorylation. It has now been shown that DNA also activates ATR-mediated RPA phosphorylation in vitro, although the nature of the DNA required for this activation has not been addressed (56,57). Given the apparent evolutionary conservation of this pathway, our future efforts to precisely characterize the Mec1-mediated reaction should provide considerable insight into RPA phosphorylation in both yeast and humans.