Initiation of DNA damage responses through XPG‐related nucleases

Lesion‐specific enzymes repair different forms of DNA damage, yet all lesions elicit the same checkpoint response. The common intermediate required to mount a checkpoint response is thought to be single‐stranded DNA (ssDNA), coated by replication protein A (RPA) and containing a primer‐template junction. To identify factors important for initiating the checkpoint response, we screened for genes that, when overexpressed, could amplify a checkpoint signal to a weak allele of chk1 in fission yeast. We identified Ast1, a novel member of the XPG‐related family of endo/exonucleases. Ast1 promotes checkpoint activation caused by the absence of the other XPG‐related nucleases, Exo1 and Rad2, the homologue of Fen1. Each nuclease is recruited to DSBs, and promotes the formation of ssDNA for checkpoint activation and recombinational repair. For Rad2 and Exo1, this is independent of their S‐phase role in Okazaki fragment processing. This XPG‐related pathway is distinct from MRN‐dependent responses, and each enzyme is critical for damage resistance in MRN mutants. Thus, multiple nucleases collaborate to initiate DNA damage responses, highlighting the importance of these responses to cellular fitness.     

Base excision repair of oxidative DNA damage activated by XPG protein

Oxidized pyrimidines in DNA are removed by a distinct base excision repair pathway initiated by the DNA glycosylase--AP lyase hNth1 in human cells. We have reconstituted this single-residue replacement pathway with recombinant proteins, including the AP endonuclease HAP1/APE, DNA polymerase beta, and DNA ligase III-XRCC1 heterodimer. With these proteins, the nucleotide excision repair enzyme XPG serves as a cofactor for the efficient function of hNth1. XPG protein promotes binding of hNth1 to damaged DNA. The stimulation of hNth1 activity is retained in XPG catalytic site mutants inactive in nucleotide excision repair. The data support the model that development of Cockayne syndrome in XP-G patients is related to inefficient excision of endogenous oxidative DNA damage.     

Complementation of the DNA repair-deficient swi10 mutant of fission yeast by the human ERCC1 gene.

In human cells DNA damage caused by UV light is mainly repaired by the nucleotide excision repair pathway. This mechanism involves dual incisions on both sides of the damage catalyzed by two nucleases. In mammalian cells XPG cleaves 3' of the DNA lesion while the ERCC1-XPF complex makes the 5' incision. The amino acid sequence of the human excision repair protein ERCC1 is homologous with the fission yeast Swi10 protein. In order to test whether these proteins are functional homologues, we overexpressed the human gene in a Schizosaccharomyces pombe swi10 mutant. A swi10 mutation has a pleiotropic effect: it reduces the frequency of mating type switching (a mitotic transposition event from a silent cassette into the expression site) and causes increased UV sensitivity. We found that the full-length ERCC1 gene only complements the transposition defect of the fission yeast mutant, while a C-terminal truncated ERCC1 protein also restores the DNA repair capacity of the yeast cells. Using the two-hybrid system of Saccharomyces cerevisiae we show that only the truncated human ERCC1 protein is able to interact with the S . pombe Rad16 protein, which is the fission yeast homologue of human XPF. This is the first example yet known that a human gene can correct a yeast mutation in nucleotide excision repair.     

Rad51 protein controls Rad52-mediated DNA annealing

In Saccharomyces cerevisiae, Rad52 protein plays an essential role in the repair of DNA double-stranded breaks (DSBs). Rad52 and its orthologs possess the unique capacity to anneal single-stranded DNA (ssDNA) complexed with its cognate ssDNA-binding protein, RPA. This annealing activity is used in multiple mechanisms of DSB repair: single-stranded annealing, synthesis-dependent strand annealing, and cross-over formation. Here we report that the S. cerevisiae DNA strand exchange protein, Rad51, prevents Rad52-mediated annealing of complementary ssDNA. Efficient inhibition is ATP-dependent and involves a specific interaction between Rad51 and Rad52. Free Rad51 can limit DNA annealing by Rad52, but the Rad51 nucleoprotein filament is even more effective. We also discovered that the budding yeast Rad52 paralog, Rad59 protein, partially restores Rad52-dependent DNA annealing in the presence of Rad51, suggesting that Rad52 and Rad59 function coordinately to enhance recombinational DNA repair either by directing the processed DSBs to repair by DNA strand annealing or by promoting second end capture to form a double Holliday junction. This regulation of Rad52-mediated annealing suggests a control function for Rad51 in deciding the recombination path taken for a processed DNA break; the ssDNA can be directed to either Rad51-mediated DNA strand invasion or to Rad52-mediated DNA annealing. This channeling determines the nature of the subsequent repair process and is consistent with the observed competition between these pathways in vivo.     

Silent repair accounts for cell cycle specificity in the signaling of oxidative DNA lesions

Reactive oxygen species are the most important source of DNA lesions in aerobic organisms, but little is known about the activation of the DNA checkpoints in response to oxidative stress. We show that treatment of yeast cells with sublethal concentrations of hydrogen peroxide induces a Mec1-dependent phosphorylation of Rad53 and a Rad53-dependent cell cycle delay specifically during S phase. The lack of Rad53 phosphorylation after hydrogen peroxide treatment in the G1 and G2 phases is due to the silent repair of oxidative DNA lesions produced at these stages by the base excision repair (BER) pathway. Only the disruption of the BER pathway and the accumulation and/or treatment of DNA intermediates by alternative repair pathways reveal the existence of primary DNA lesions induced at all phases of the cell cycle by hydrogen peroxide. Our data illustrate both the concept of silent repair of DNA damage and the high sensitivity of S-phase cells to hydrogen peroxide.     

Interaction with Rad51 is indispensable for recombination mediator function of Rad52

In the yeast Saccharomyces cerevisiae, the RAD52 gene is indispensable for homologous recombination and DNA repair. Rad52 protein binds DNA, anneals complementary ssDNA strands, and self-associates to form multimeric complexes. Moreover, Rad52 physically interacts with the Rad51 recombinase and serves as a mediator in the Rad51-catalyzed DNA strand exchange reaction. Here, we examine the functional significance of the Rad51/Rad52 interaction. Through a series of deletions, we have identified residues 409-420 of Rad52 as being indispensable and likely sufficient for its interaction with Rad51. We have constructed a four-amino acid deletion mutation within this region of Rad52 to ablate its interaction with Rad51. We show that the rad52delta409-412 mutant protein is defective in the mediator function in vitro even though none of the other Rad52 activities, namely, DNA binding, ssDNA annealing, and protein oligomerization, are affected. We also show that the sensitivity of the rad52delta409-412 mutant to ionizing radiation can be complemented by overexpression of Rad51. These results thus demonstrate the significance of the Rad51-Rad52 interaction in homologous recombination.     

Coordinated Binding of Single-Stranded and Double-Stranded DNA by UvsX Recombinase

Homologous recombination is important for the error-free repair of DNA double-strand breaks and for replication fork restart. Recombinases of the RecA/Rad51 family perform the central catalytic role in this process. UvsX recombinase is the RecA/Rad51 ortholog of bacteriophage T4. UvsX and other recombinases form presynaptic filaments on ssDNA that are activated to search for homology in dsDNA and to perform DNA strand exchange. To effectively initiate recombination, UvsX must find and bind to ssDNA within an excess of dsDNA. Here we examine the binding of UvsX to ssDNA and dsDNA in the presence and absence of nucleotide cofactor, ATP. We also examine how the binding of one DNA substrate is affected by simultaneous binding of the other to determine how UvsX might selectively assemble on ssDNA. We show that the two DNA binding sites of UvsX are regulated by the nucleotide cofactor ATP and are coordinated with each other such that in the presence of ssDNA, dsDNA binding is significantly reduced and correlated with its homology to the ssDNA bound to the enzyme. UvsX has high affinity for dsDNA in the absence of ssDNA, which may allow for sequestration of the enzyme in an inactive form prior to ssDNA generation.     

Characterization of the Interaction between Rfa1 and Rad24 in Saccharomyces cerevisiae

Maintaining the integrity of the genome requires the high fidelity duplication of the genome and the ability of the cell to recognize and repair DNA lesions. The heterotrimeric single stranded DNA (ssDNA) binding complex Replication Protein A (RPA) is central to multiple DNA processes, which are coordinated by RPA through its ssDNA binding function and through multiple protein-protein interactions. Many RPA interacting proteins have been reported through large genetic and physical screens; however, the number of interactions that have been further characterized is limited. To gain a better understanding of how RPA functions in DNA replication, repair, and cell cycle regulation and to identify other potential functions of RPA, a yeast two hybrid screen was performed using the yeast 70 kDa subunit, Replication Factor A1 (Rfa1), as a bait protein. Analysis of 136 interaction candidates resulted in the identification of 37 potential interacting partners, including the cell cycle regulatory protein and DNA damage clamp loader Rad24. The Rfa1-Rad24 interaction is not dependent on ssDNA binding. However, this interaction appears affected by DNA damage. The regions of both Rfa1 and Rad24 important for this interaction were identified, and the region of Rad24 identified is distinct from the region reported to be important for its interaction with Rfc2 5. This suggests that Rad24-Rfc2-5 (Rad24-RFC) recruitment to DNA damage substrates by RPA occurs, at least partially, through an interaction between the N terminus of Rfa1 and the C terminus of Rad24. The predicted structure and location of the Rad24 C-terminus is consistent with a model in which RPA interacts with a damage substrate, loads Rad24-RFC at the 5’ junction, and then releases the Rad24-RFC complex to allow for proper loading and function of the DNA damage clamp.     

Rad51 gain-of-function mutants that exhibit high affinity DNA binding cause DNA damage sensitivity in the absence of Srs2

We previously identified several rad51 gain-of-function alleles that partially suppress the requirement for RAD55 and RAD57 in DNA repair. To gain further insight into the mechanism of action of these alleles, we compared the activities of Rad51-V328A, Rad51-P339S and Rad51-I345T with wild-type Rad51, for DNA binding, filament stability, strand exchange and interaction with the antirecombinase helicase, Srs2. These alleles were chosen because they show the highest activity in suppression of ionizing radiation sensitivity of the rad57 mutant, and Val 328 and Ile 345 are conserved in the human Rad51 protein. All three mutant proteins exhibited higher affinity for single-stranded DNA (ssDNA) and showed more robust strand exchange activity with oligonucleotide substrates than wild-type Rad51, with the Rad51-I345T and Rad51-V328A proteins displaying higher activity than Rad51-P339S. However, the Srs2 antirecombinase was able to disrupt Rad51–ssDNA complexes formed with all the mutant proteins. In vivo, the rad51-I345T mutant strain exhibited high resistance to methyl methane sulfonate that was dependent on functional SRS2. These results suggest the Srs2 translocase is able to disrupt Rad51–ssDNA complexes at stalled replication forks, but in the absence of Srs2 the enhanced DNA binding of the Rad51-I345T protein is detrimental to cell survival.     

Checkpoint-dependent phosphorylation of Exo1 modulates the DNA damage response

Exo1 is a nuclease involved in mismatch repair, DSB repair, stalled replication fork processing and in the DNA damage response triggered by dysfunctional telomeres. In budding yeast and mice, Exo1 creates single-stranded DNA (ssDNA) at uncapped telomeres. This ssDNA accumulation activates the checkpoint response resulting in cell cycle arrest. Here, we demonstrate that Exo1 is phosphorylated when telomeres are uncapped in cdc13-1 and yku70Delta yeast cells, and in response to the induction of DNA damage. After telomere uncapping, Exo1 phosphorylation depends on components of the checkpoint machinery such as Rad24, Rad17, Rad9, Rad53 and Mec1, but is largely independent of Chk1, Tel1 and Dun1. Serines S372, S567, S587 and S692 of Exo1 were identified as targets for phosphorylation. Furthermore, mutation of these Exo1 residues altered the DNA damage response to uncapped telomeres and camptothecin treatment, in a manner that suggests Exo1 phosphorylation inhibits its activity. We propose that Rad53-dependent Exo1 phosphorylation is involved in a negative feedback loop to limit ssDNA accumulation and DNA damage checkpoint activation.     

Crystal structure of the catalytic core of Rad2: insights into the mechanism of substrate binding

Rad2/XPG belongs to the flap nuclease family and is responsible for a key step of the eukaryotic nucleotide excision DNA repair (NER) pathway. To elucidate the mechanism of DNA binding by Rad2/XPG, we solved crystal structures of the catalytic core of Rad2 in complex with a substrate. Rad2 utilizes three structural modules for recognition of the double-stranded portion of DNA substrate, particularly a Rad2-specific α-helix for binding the cleaved strand. The protein does not specifically recognize the single-stranded portion of the nucleic acid. Our data suggest that in contrast to related enzymes (FEN1 and EXO1), the Rad2 active site may be more accessible, which would create an exit route for substrates without a free 5′ end.     

The recombination-deficient mutant RPA (rfa1-t11) is displaced slowly from single-stranded DNA by Rad51 protein

Replication protein-A (RPA) is involved in many processes of DNA metabolism, including DNA replication, repair, and recombination. Cells carrying a mutation in the largest subunit of RPA (rfa1-t11: K45E) have defects in meiotic recombination, mating-type switching, and survival after DNA damage caused by UV and methyl methanesulfonate, as well as increased genome instability; however, this mutant has no significant defect in DNA replication. We purified the RPA heterotrimer containing the rfa1-t11 substitution (RPA(rfa1-t11)). This mutant RPA binds single-stranded DNA (ssDNA) with the same site size, and the RPA(rfa1-t11).ssDNA complex shows a similar sensitivity to disruption by salt as the wild-type RPA.ssDNA complex. RPA(rfa1-t11) stimulates DNA strand exchange, provided that the Rad51 protein.ssDNA nucleoprotein complex is assembled prior to introduction of the mutant RPA. However, RPA(rfa1-t11) is displaced from ssDNA by Rad51 protein more slowly than wild-type RPA and, as a consequence, Rad51 protein-mediated DNA strand exchange is inhibited when the ssDNA is in a complex with RPA(rfa1-t11). Rad52 protein can stimulate displacement of RPA(rfa1-t11) from ssDNA by Rad51 protein, but the rate of displacement remains slow compared with wild-type RPA. These in vitro results suggest that, in vivo, RPA is bound to ssDNA prior to Rad51 protein and that RPA displacement by Rad51 protein is a critical step in homologous recombination, which is impaired in the rfa1-t11 mutation.     

Roles of Rad23 protein in yeast nucleotide excision repair

Nucleotide excision repair (NER) removes many different types of DNA lesions. Most NER proteins are indispensable for repair. In contrast, the yeast Rad23 represents a class of accessory NER proteins, without which NER activity is reduced but not eliminated. In mammals, the complex of HR23B (Rad23 homolog) and XPC (yeast Rad4 homolog) has been suggested to function in the damage recognition step of NER. However, the precise function of Rad23 or HR23B in NER remains unknown. Recently, it was suggested that the primary function of RAD23 protein in NER is its stabilization of XPC protein. Here, we tested the significance of Rad23-mediated Rad4 stabilization in NER, and analyzed the repair and biochemical activities of purified yeast Rad23 protein. Cellular Rad4 was indeed stabilized by Rad23 in the absence of DNA damage. Persistent overexpression of Rad4 in rad23 mutant cells, however, largely failed to complement the ultraviolet sensitivity of the mutant. Consistently, deficient NER in rad23 mutant cell extracts could not be complemented by purified Rad4 protein in vitro. In contrast, partial complementation was observed with purified Rad23 protein. Specific complementation to the level of wild-type repair was achieved by adding purified Rad23 together with small amounts of Rad4 protein to rad23 mutant cell extracts. Purified Rad23 protein was unable to bind to DNA, but stimulated the binding activity of purified Rad4 protein to N-acetyl-2-aminofluorene-damaged DNA. These results support two roles of Rad23 protein in NER: (i) its direct participation in the repair biochemistry, possibly due to its stimulatory activity on Rad4-mediated damage binding/recognition; and (ii) its stabilization of cellular Rad4 protein.     

The human Rad51 K133A mutant is functional for DNA double-strand break repair in human cells

The human Rad51 protein requires ATP for the catalysis of DNA strand exchange, as do all Rad51 and RecA-like recombinases. However, understanding the specific mechanistic requirements for ATP binding and hydrolysis has been complicated by the fact that ATP appears to have distinctly different effects on the functional properties of human Rad51 versus yeast Rad51 and bacterial RecA. Here we use RNAi methods to test the function of two ATP binding site mutants, K133R and K133A, in human cells. Unexpectedly, we find that the K133A mutant is functional for repair of DNA double-strand breaks when endogenous Rad51 is depleted. We also find that the K133A protein maintains wild-type-like DNA binding activity and interactions with Brca2 and Xrcc3, properties that undoubtedly promote its DNA repair capability in the cell-based assay used here. Although a Lys to Ala substitution in the Walker A motif is commonly assumed to prevent ATP binding, we show that the K133A protein binds ATP, but with an affinity approximately 100-fold lower than that of wild-type Rad51. Our data suggest that ATP binding and release without hydrolysis by the K133A protein act as a mechanistic surrogate in a catalytic process that applies to all RecA-like recombinases. ATP binding promotes assembly and stabilization of a catalytically active nucleoprotein filament, while ATP hydrolysis promotes filament disassembly and release from DNA.     

Human flap endonuclease structures, DNA double-base flipping, and a unified understanding of the FEN1 superfamily

Flap endonuclease (FEN1), essential for DNA replication and repair, removes RNA and DNA 5' flaps. FEN1 5' nuclease superfamily members acting in nucleotide excision repair (XPG), mismatch repair (EXO1), and homologous recombination (GEN1) paradoxically incise structurally distinct bubbles, ends, or Holliday junctions, respectively. Here, structural and functional analyses of human FEN1:DNA complexes show structure-specific, sequence-independent recognition for nicked dsDNA bent 100° with unpaired 3' and 5' flaps. Above the active site, a helical cap over a gateway formed by two helices enforces ssDNA threading and specificity for free 5' ends. Crystallographic analyses of product and substrate complexes reveal that dsDNA binding and bending, the ssDNA gateway, and double-base unpairing flanking the scissile phosphate control precise flap incision by the two-metal-ion active site. Superfamily conserved motifs bind and open dsDNA; direct the target region into the helical gateway, permitting only nonbase-paired oligonucleotides active site access; and support a unified understanding of superfamily substrate specificity.     

The DNA repair protein rad23 is a negative regulator of multi-ubiquitin chain assembly

Rad23 is a nucleotide-excision repair protein with a previously unknown biochemical function. We determined that yeast and human Rad23 inhibited multi-ubiquitin (Ub) chain formation and the degradation of proteolytic substrates. Significantly, Rad23 could be co-precipitated with a substrate that contained a short multi-Ub chain. The UV sensitivity of rad23Delta was reduced in mutants lacking the E2 enzyme Ubc4, or the multi-Ub chain-promoting factor Ufd2. These studies suggest that the stability of proteolytic substrates is governed by the competing action of multi-Ub chain-promoting and chain-inhibiting factors. The stabilization of DNA repair and stress factors could represent an important biological function of Rad23.     

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.     

Tailed duplex DNA is the preferred substrate for Rad51 protein-mediated homologous pairing

The repair of potentially lethal DNA double-stranded breaks (DSBs) by homologous recombination requires processing of the broken DNA into a resected DNA duplex with a protruding 3'-single-stranded DNA (ssDNA) tail. Accordingly, the canonical models for DSB repair require invasion of an intact homologous DNA template by the 3'-end of the ssDNA, a characteristic that the bacterial pairing protein RecA possesses. Unexpectedly, we find that for the eukaryotic homolog, Rad51 protein, the 5'-end of ssDNA is more invasive than the 3'-end. This pairing bias is unaffected by Rad52, Rad54 or Rad55-57 proteins. However, further investigation reveals that, in contrast to RecA protein, the preferred DNA substrate for Rad51 protein is not ssDNA but rather dsDNA with ssDNA tails. This important distinction permits the Rad51 proteins to promote DNA strand invasion using either 3'- or 5'-ends with similar efficiency.     

Yeast Rad52 and Rad51 recombination proteins define a second pathway of DNA damage assessment in response to a single double-strand break

Saccharomyces cells with a single unrepaired double-strand break adapt after checkpoint-mediated G(2)/M arrest. We have found that both Rad51 and Rad52 recombination proteins play key roles in adaptation. Cells lacking Rad51p fail to adapt, but deleting RAD52 suppresses rad51Delta. rad52Delta also suppresses adaptation defects of srs2Delta mutants but not those of yku70Delta or tid1Delta mutants. Neither rad54Delta nor rad55Delta affects adaptation. A Rad51 mutant that fails to interact with Rad52p is adaptation defective; conversely, a C-terminal truncation mutant of Rad52p, impaired in interaction with Rad51p, is also adaptation defective. In contrast, rad51-K191A, a mutation that abolishes recombination and results in a protein that does not bind to single-stranded DNA (ssDNA), supports adaptation, as do Rad51 mutants impaired in interaction with Rad54p or Rad55p. An rfa1-t11 mutation in the ssDNA binding complex RPA partially restores adaptation in rad51Delta mutants and fully restores adaptation in yku70Delta and tid1Delta mutants. Surprisingly, although neither rfa1-t11 nor rad52Delta mutants are adaptation defective, the rad52Delta rfa1-t11 double mutant fails to adapt and exhibits the persistent hyperphosphorylation of the DNA damage checkpoint protein Rad53 after HO induction. We suggest that monitoring of the extent of DNA damage depends on independent binding of RPA and Rad52p to ssDNA, with Rad52p's activity modulated by Rad51p whereas RPA's action depends on Tid1p.