Holliday junction-containing DNA structures persist in cells lacking Sgs1 or Top3 following exposure to DNA damage Hocine W. Mankouria,b,1, Thomas M. Ashtona,1, and Ian D. Hicksona,b,2 a Weatherall Institute of Molecular Medicine, University of Oxford, John Radcliffe Hospital, Oxford OX3 9DS, United Kingdom; and bNordea Center for Healthy Aging, Department of Cellular and Molecular Medicine, University of Copenhagen, 2200 Copenhagen N, Denmark
Edited* by Stephen C. Kowalczykowski, University of California, Davis, CA, and approved February 14, 2011 (received for review September 22, 2010)
The Sgs1–Rmi1–Top3 “dissolvasome” is required for the maintenance of genome stability and has been implicated in the processing of various types of DNA structures arising during DNA replication. Previous investigations have revealed that unprocessed (X-shaped) homologous recombination repair (HRR) intermediates persist when S-phase is perturbed by using methyl methanesulfonate (MMS) in Saccharomyces cerevisiae cells with impaired Sgs1 or Top3. However, the precise nature of these persistent DNA structures remains poorly characterized. Here, we report that ectopic expression of either of two heterologous and structurally unrelated Holliday junction (HJ) resolvases, Escherichia coli RusA or human GEN11-527, promotes the removal of these X-structures in vivo. Moreover, other types of DNA replication intermediates, including stalled replication forks and non-HRR-dependent X-structures, are refractory to RusA or GEN11-527, demonstrating specificity of these HJ resolvases for MMS-induced X-structures in vivo. These data suggest that the X-structures persisting in cells with impaired Sgs1 or Top3 contain HJs. Furthermore, we demonstrate that Sgs1 directly promotes X-structure removal, because the persistent structures arising in Sgs1-deficient strains are eliminated when Sgs1 is reactivated in vivo. We propose that HJ resolvases and Sgs1–Top3–Rmi1 comprise two independent processes to deal with HJ-containing DNA intermediates arising during HRR in S-phase. DNA repair
| RecQ | helicase | topoisomerase
T
he RecQ family of DNA helicases is required for the maintenance of genome stability in all organisms. Mutations in at least three (of five) human RecQ helicases is associated with an increased predisposition to the development of cancer and/or premature aging (1). Mutations in BLM cause Bloom’s syndrome (BS), which is associated with increased cancer predisposition, whereas mutations in WRN or RECQ4 cause distinct disorders (Werner’s syndrome and Rothmund–Thomson syndrome, respectively) characterized by premature aging and some developmental abnormalities. In Saccharomyces cerevisiae, there is only one RecQ helicase, called Sgs1. As a consequence, this organism has proved particularly useful for genetic analyses, as well as for characterization of the defects arising in cells lacking RecQ helicases (2). Sgs1 is thought to be the ortholog of BLM, based on a number of observations. First, BLM is the only human RecQ helicase that shares the same structural domain architecture as Sgs1. Second, both BLM and Sgs1 associate with conserved interacting partners that are apparently unique for BLM in human cells. More specifically, BLM exists in a complex with a type IA topoisomerase, hTOPOIIIα, and two oligonucleotide/oligosaccharidebinding (OB)-fold containing proteins, hRMI1 and hRMI2 (3– 7); Sgs1 associates with the yeast orthologs of these proteins, Top3 and Rmi1 (8–14). Together, these proteins likely act cooperatively as a “dissolvasome” that processes multiple kinds of DNA structures arising during DNA replication, repair, recombination, and mitosis (15). One cellular process in which BLM and Sgs1 have both been strongly implicated is homologous recombination repair (HRR). 4944–4949 | PNAS | March 22, 2011 | vol. 108 | no. 12
HRR is a conserved cellular process that allows cells to copy genetic information from a homologous sequence, and is required for the efficient repair of DNA breaks and ssDNA gaps that can arise during S-phase due to DNA damage, discontinuities in DNA replication, or the impediment of replication fork (RF) progression. Evidence that BLM functions in HRR is suggested by the fact that BS cells demonstrate elevated levels of mitotic recombination, sister chromatid exchanges, and genome instability (1). Furthermore, BLM (either alone or in conjunction with hTOPOIIIα, hRMI1, and hRMI2) can resolve different types of HRR intermediates in vitro, such as D-loops and single or double Holliday junctions (HJs; refs. 6, 7, and 16–21). Mutation of SGS1 or the Schizosaccharomyces pombe ortholog of BLM/SGS1, called rqh1+, also causes genome instability, hyperrecombination, and sensitivity to DNA-damaging agents (8, 22–25), and Sgs1, like BLM, is able to unwind HJs in vitro (26). Furthermore, unprocessed HRR intermediates (X-structures) have been directly observed in methyl methanesulfonate (MMS)-treated sgs1, top3, and rmi1 mutants using 2D gel electrophoresis (27–30). Because many of the deleterious phenotypes of sgs1 or rqh1 mutants can be suppressed by the mutation of genes involved in the early steps of HRR (e.g., RAD51 in S. cerevisiae and rhp51+ in S. pombe; refs. 31–35), it is likely that these phenotypes are, at least in part, due to unregulated or incomplete HRR. Although X-shaped HRR structures have been detected by using 2D DNA gel electrophoresis in S. cerevisiae cells with impaired Sgs1, Top3, or Rmi1 (27–30), it is presently unknown whether different types of DNA structures arise in cells deficient in these different proteins. Because abolition of Sgs1 activity (or just its helicase activity) can suppress the poor growth phenotypes of top3 or rmi1 mutants (8, 13, 14, 36), it has been proposed that Sgs1 may create a DNA intermediate that is toxic in cells lacking Top3 or Rmi1. One proposal is that the convergent branch migration of double HJs (dHJs) by Sgs1 creates a hemicatenane structure that can only be resolved by Top3 (in conjunction with Rmi1), in a process known as “dHJ dissolution” (18, 20, 21). However, in vivo evidence for this process is lacking, because the 2D gel methodology cannot definitively distinguish between different types of joint DNA molecules such as HJs and hemicatenanes. The X-structures arising in MMS-treated sgs1 mutants have been proposed to be “pseudo-HJs,” consisting of a region of hemicatenated nascent DNA and concomitant single-stranded regions of parental DNA (27). This conclusion was based on the in vitro characterization of X-structures, including their ability to
Author contributions: H.W.M. and I.D.H. designed research; H.W.M. and T.M.A. performed research; H.W.M. contributed new reagents/analytic tools; H.W.M., T.M.A., and I.D.H. and analyzed data; and H.W.M. wrote the paper. The authors declare no conflict of interest. *This Direct Submission article had a prearranged editor. 1
H.W.M. and T.M.A. contributed equally to this work.
2
To whom correspondence should be addressed. E-mail:
[email protected].
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. 1073/pnas.1014240108/-/DCSupplemental.
www.pnas.org/cgi/doi/10.1073/pnas.1014240108
Results Heterologous HJ Resolvases Diminish the MMS-Induced X-Structures in sgs1 Mutants. To determine the specific nature of the MMS-
induced X-structures arising in haploid sgs1 mutants, we investigated whether any proteins with well-characterized in vitro substrate specificities could diminish the level of the X-structures when ectopically expressed in vivo. Interestingly, expression of two heterologous HJ resolvases, E. coli RusA and a fragment of human GEN1, GEN11-527, have been demonstrated to suppress the MMS- and UV-sensitivity of S. pombe rqh1 mutants (39–41), suggesting that unprocessed HJs could be directly responsible for various rqh1 phenotypes. We therefore examined whether expression of these HJ resolvases could reduce the level of the X structures in MMS-treated sgs1 mutants using 2D gel electrophoresis. To permit inducible expression of heterologous proteins and their efficient targeting to the nucleus, NLS–GFP (control), NLS–RusA–GFP, and NLS–GEN11-527-GFP proteins were subcloned into pYES2, allowing galactose-inducible protein expression. These proteins will henceforth be denoted as GFP, RusA, or GEN11-527, respectively. Proteins were induced (by the addition of 2% galactose) during α-factor arrest and throughout the subsequent analyses. Samples of wild-type (WT)–GFP, sgs1– GFP, sgs1–RusA, and sgs1–GEN11-527 cells were taken at specific intervals following the release from G1 arrest to observe DNA replication intermediates on 2D gels originating from the earlyfiring replication origin, ARS305. Parallel analyses, using the same G1-arrested/galactose-induced culture, were performed to directly compare the cellular responses to MMS and to the DNA replication inhibitor, hydroxyurea (HU; Fig. 1A). To permit efficient origin firing, cells were released from G1 arrest into Mankouri et al.
medium containing MMS or HU at 30 °C for 1 h. Cultures were then incubated at 37 °C for a further 6 h to promote robust HJ resolvase activity. Genomic DNA was extracted by using the hexadecyltrimethylammonium bromide (CTAB) method to restrain branch migration of joint (X-shaped) molecules (42). After 7-h exposure to 0.033% MMS, all DNA replication intermediates had disappeared from ARS305 in WT cells expressing GFP, consistent with the completion of DNA replication at this region of the genome by this stage. In sgs1 mutants expressing GFP, bubbles and Y arcs had also disappeared, but, consistent with previous findings (27, 28, 30), persistent X-structures at ARS305 were observed (Fig. 1B). Previous analyses have identified these structures to be Rad51-dependent HRR intermediates (27). Interestingly, these MMS-induced X-structures were not detectable in sgs1 cells expressing RusA and GEN11-527 (Fig. 1B). All strains examined exhibited similar mid-S-phase FACS profiles (Fig. S1), suggesting that the observed differences on 2D gels were not attributable to any differences in cell cycle progression or MMS-induced checkpoint arrest. We conclude that expression of RusA and GEN11-527 leads to a marked reduction in the level of unprocessed HRR intermediates in MMStreated sgs1 cells. Stalled RFs and Non-HRR-Dependent X-Structures Are Refractory to HJ Resolvases. Biochemical analysis of RusA and GEN11-527 has
demonstrated robust, and substrate-specific, HJ resolvase activity for these enzymes in vitro (40, 43–47). However, unlike RusA (40), GEN11-527 also exhibits some ability to cleave RF-like structures in vitro (44, 45). To assess whether the RusA and GEN11-527 HJ resolvases demonstrate specificity for MMSinduced X-structures in vivo, we examined whether they could cleave other types of replication intermediates detectable using 2D gels. To test this, we directly analyzed the fate of DNA replication intermediates arising at ARS305 in response to HU. In the presence of 0.2 M HU, we observed robust ARS305 origin firing (as revealed by the presence of bubbles, Y arcs, and origin-associated X-structures) in WT cells and sgs1 mutants expressing GFP (Fig. 1C). Previous analyses have demonstrated that the origin-associated X-structures are normal DNA replication intermediates that are not dependent on DNA damage or HRR proteins for their formation (42). Furthermore, the formation and disappearance of these structures is unaffected in sgs1 mutants (27). ARS305 origin-firing was also unaffected in sgs1 mutants expressing RusA or GEN11-527 (Fig. 1C). Therefore, ectopic expression of these HJ resolvases does not noticeably affect any early DNA replication intermediates arising at ARS305. Cultures were then incubated at 37 °C for a further 6 h in the presence of 0.2 M HU, as described in Fig. 1B for MMS-treated cells. We observed that the levels of the different DNA replication intermediates detectable after 1 h in 0.2 M HU were substantially reduced in all strains examined by 7 h in the NcoI–NcoI 5-kb ARS305 fragment under analysis. The main species of DNA replication intermediates present in the NcoI–NcoI fragment now consisted of large Ys and origin-associated X-structures (Fig. 1C). Enhancing the exposure of these data revealed that low levels of intact bubble structures were also detectable in these extracts (Fig. S2). Furthermore, because these DNA replication intermediates were unaffected by the expression of HJ resolvases, we conclude that RusA and GEN11-527 do not cleave RFs or originassociated (Rad51-independent) X-structures in vivo. To test whether the reduction in replication intermediates in the NcoI–NcoI fragment at the 7-h time point was due to RFs migrating outside of the 5-kb fragment under analysis, or whether HU-induced RF collapse had occurred, we processed the DNA samples using different restriction enzymes that permit analysis of replication intermediates in the region just distal to the NcoI–NcoI fragment (see Fig. 1D for a diagrammatic representation). Analysis of a BamHI–SpeI ARS305 fragment revealed that RFs and origin-associated X-structures were detectable after 7 h of 0.2 M HU treatment, suggesting that RF PNAS | March 22, 2011 | vol. 108 | no. 12 | 4945
GENETICS
branch migrate (even in the presence of Mg2+, which generally inhibits HJ migration), their apparent resistance to the RuvC HJ resolvase, and their partial sensitivity to ssDNA nucleases (27). MMS also causes increased interhomolog X-structures in sgs1 diploid cells (37), although the precise nature of these interchromosomal joint DNA molecules remains to be determined. Interestingly, a recent study demonstrated that interchromosomal dHJs arise during double-strand break (DSB) repair in mitotic diploid cells (38). Furthermore, the turnover of these structures was altered in sgs1 mutants, consistent with a possible role for Sgs1 in the prevention and/or direct processing of HJs in vivo (38). Inter-sister chromatid X-structures were also identified as intermediates of DSB repair in this study, although whether these contained HJs or not could not be determined. Therefore, despite several independent studies demonstrating the existence of unprocessed X-shaped DNA structures in sgs1 mutants under a number of different experimental conditions, it remains unknown whether these comprise a common type of DNA structure or not. Indeed, the precise nature and abundance of the X-structures could vary depending on the context (e.g., in DSB repair vs. the repair of replication-associated lesions). Consistent with this proposal, RecQ helicases can process a variety of different DNA structures in addition to HJs in vitro, and the BLM–hTOPOIIIα– hRMI1–hRMI2 and Sgs1–Top3–Rmi1 complexes probably act as DNA structure-specific “dissolvasomes” in vivo (15). We sought to characterize the composition of MMS-induced, replication-associated X-structures by promoting their resolution in vivo using heterologous enzymes with well-characterized in vitro substrate specificities. Here, we report that ectopic expression of either of two heterologous HJ resolvases, Escherichia coli RusA or human GEN11-527, enhances the in vivo removal of the MMS-induced X-structures in cells impaired for Sgs1 or Top3. We propose that the X-structures persisting in both of these mutants contain one or more HJs. Furthermore, because X-structure processing in Sgs1-deficient strains can also be promoted by the reactivation of Sgs1 in vivo, we propose that the Sgs1 complex directly processes HJ-containing DNA structures arising during HRR, facilitating their elimination.
Fig. 1. Expression of HJ resolvases reduces the levels of MMS-induced X-structures in sgs1 mutants. (A) Outline of the experimental protocol. Cultures were released from G1 arrest at 30 °C and incubated at 37 °C after 1 h to promote robust HJ resolvase activity. (B) Analysis of DNA structures in WT–GFP, sgs1–GFP, sgs1–RusA, and sgs1–GEN11-527 strains after exposure to MMS. DNA replication intermediates were analyzed in a 5-kb NcoI–NcoI ARS305 fragment by 2D gel electrophoresis after 7-h exposure to 0.033% MMS. The key on the right denotes DNA structures that can be identified by the 2D gel technique. The arrow denotes X-shaped structures present in sgs1 mutants. (C) Analysis of DNA structures after exposure to HU. DNA replication intermediates were analyzed as in B, except after 1 h or 7 h of exposure to 0.2 M HU. (D) A schematic diagram of ARS305 highlighting the relevant restriction sites. The DNA extracts from cells exposed to 0.2 M HU for 7 h (C) were digested with BamHI and SpeI and then analyzed by 2D gel electrophoresis.
integrity (and ongoing RF progression) remained largely intact in the presence of HU (Fig. 1D). Flow cytometric data also confirmed that DNA replication proceeded, albeit very slowly, in all strains in the presence of 0.2 M HU (Fig. S1). Furthermore, the replication intermediates detectable at this time were again unaffected by expression of HJ resolvases in sgs1 mutants (Fig. 1D). These data indicate that the HJ resolvases analyzed here do not cleave, or prevent the maturation of, other types of DNA replication intermediates detectable at ARS305. This result suggests that RusA and GEN11-527 retain their biochemically characterized substrate specificities (40, 43–47) when expressed in S. cerevisiae and therefore implies that the MMS-induced X-structures contain one or more HJs, or that they arise from a HJ-containing precursor. Furthermore, we conclude that the RAD51-independent origin-associated X-structures do not contain HJs, in agreement with previous findings (42). 4946 | www.pnas.org/cgi/doi/10.1073/pnas.1014240108
HJ Resolvases Do Not Prevent X-Structure Accumulation in sgs1 Mutants. Next, we analyzed the mechanism of X-structure re-
moval by HJ resolvases in sgs1 mutants. We analyzed DNA structures at ARS305 after 3 h of 0.033% MMS treatment because the ARS305 fragment under analysis is generally fully replicated in the presence of MMS by 2–3 h, which coincides with the peak levels of Rad51-dependent X-structures in sgs1 mutants (27, 28, 30). After 3-h exposure to 0.033% MMS, we observed that X-structures were detectable in sgs1 mutants expressing GFP, RusA, or GEN11-527 (Fig. 2). However, quantification of relative X-structure intensity revealed a small reduction in the levels of X-structures in sgs1 cells expressing RusA or GEN11-527 (28% and 35%, respectively). We propose that the expression of HJ resolvases is not able to strongly counteract the rapid surge in X-structure formation that occurs in the first 2–3 h of 0.033% MMS treatment in sgs1 mutants (27, 28, 30). After Mankouri et al.
an additional 5-h incubation at 37 °C, we observed that Xstructures were still evident in sgs1 mutants expressing GFP, but they were no longer detectable in sgs1 mutants expressing RusA or GEN11-527 (Fig. 2). We therefore propose that the majority (>65%) of sgs1 X-structures are removed by HJ resolvases only after they form. We acknowledge, however, that the 28–35% decrease in X-structure accumulation in sgs1 strains expressing HJ resolvases could also be due to the additional cleavage of an X-structure precursor by these enzymes. Nuclease Activity of HJ Resolvases Is Required to Diminish XStructures in sgs1 Mutants. To test whether X-structure removal
using HJ resolvases in sgs1 cells was dependent on there being either an extended S-phase arrest or a prolonged exposure to MMS (Figs. 1B and 2), we analyzed strains for their recovery from 0.033% MMS treatment (Fig. 3A). Protein expression was induced during α-factor arrest (Fig. 3B) and was maintained throughout the subsequent analyses. After 3 h of 0.033% MMS exposure, sgs1 cells were harvested, washed, and released into drug-free medium at 37 °C. Also, to investigate whether the nuclease activity of RusA and GEN11-527 is required for sgs1 X-structure removal, we compared the effects of expressing biochemically validated, nuclease-defective alleles of RusA (RusAD70N) and GEN11-527 (GEN11-527 D157A; refs. 44, 48, and 49; Fig. 3B). All strains examined were proficient in recovery from MMS-induced S-phase arrest and successfully completed DNA replication by 5 h after MMS removal, as revealed by FACS analysis (Fig. S3). We note, however, that GEN11-527 expression did cause a very small, but reproducible, delay in sgs1 S-phase progression in the presence of 0.033% MMS. We conclude that expression of HJ resolvases does not cause, or exacerbate, any significant DNA replication defects in sgs1 mutants, consistent with our observations that HJ resolvases do not cleave RFs, or hinder their progression, in vivo (Fig. 1 C and D). At 5 h after the removal of MMS, we observed that unprocessed X-structures were still evident at ARS305 in sgs1–GFP strains (Fig. 3C). Interestingly, we again observed that sgs1 X-structures were not detectable in cells expressing RusA or GEN11-527 at 5 h after the removal of MMS (Fig. 3C). Importantly, this effect was dependent on the nuclease activity of both proteins, because the nuclease-defective alleles were indistinguishable from the GFP control in demonstrating prominent X-structures (Fig. 3C). HJ Resolvases also Diminish Persistent X-Structures Arising in Cells Lacking Functional Top3. To determine whether the reduction in
X-structures caused by RusA and GEN11-527 in vivo is specific for X-structures arising in sgs1 mutants, we assessed whether these HJ resolvases could also diminish MMS-induced X-structures arising in cells impaired for Top3. Although MMS-induced X-structures have been detected in top3 deletion mutants (27), it is not feasible to grow large cultures of plasmid-transformed top3 mutants because of their Mankouri et al.
Fig. 3. HJ resolvases remove X-structures arising in sgs1 and TOP3Y356F mutants. (A) Outline of the experimental protocol. (B) Protein extracts were prepared from G1-arrested sgs1–GFP, sgs1–RusA, sgs1–RusAD70N, sgs1– GEN11-527, and sgs1–GEN11-527 D157A strains induced with 2% galactose, and equivalent levels of protein were resolved by SDS/PAGE. Levels of the indicated protein were determined by Western blotting. (C) DNA replication intermediates around ARS305 were analyzed in the indicated sgs1 strains by 2D gel electrophoresis after 5-h recovery from MMS treatment. X-shaped DNA structures are denoted by arrows. (D) HJ resolvases process X-structures arising in cells impaired for Top3. WT–GFP, WT–RusA, and WT–GEN11-527 strains were transformed with pWJ1347 (GAL1–TOP3Y356F). Coexpression of proteins was induced during G1 arrest and throughout the subsequent incubation. DNA replication intermediates around ARS305 were analyzed in the indicated strains by 2D gel electrophoresis after 5 h of recovery from MMS treatment.
poor growth rate and the rapid appearance of suppressor mutations (8, 13, 14, 36). To permit the acute impairment of Top3 function in vivo, we therefore used a previously validated, dominant-negative allele of TOP3, TOP3Y356F, that, when overexpressed, causes the accumulation of Rad51-dependent X-structures after 2-h exposure to 0.033% MMS (29, 50). Consistent with our previous findings (29), overexpression of TOP3Y356F caused persistent MMS-induced Xstructures that were still detectable at 5 h after recovery from 0.033% MMS (Fig. 3D), similar to what was observed in sgs1 mutants (Fig. 3C). Interestingly, we observed that expression of RusA and GEN11-527 also diminished the level of TOP3Y356F Xstructures (Fig. 3D). We conclude that heterologous HJ resolvases can remove X-structures arising in cells with impaired Sgs1 or Top3, suggesting that a common type of HJ-containing DNA structure likely persists in both of these mutants. Reactivation of Sgs1 Removes X-Structures in sgs1-36 Mutants.
Having demonstrated that unprocessed HJ-containing structures persist in sgs1 mutants, we examined whether the normal function of Sgs1 is to prevent these structures from accumulating, or whether Sgs1 primarily acts to directly eliminate them in vivo. We note, however, that these roles are not necessarily mutually exclusive. To test these possibilities, we used a validated temperature-sensitive mutant of sgs1, sgs1-36, that permits the temporary, and reversible, inactivation of Sgs1 (51). Strains were grown at 25 °C, arrested in G1 at the restrictive temperature (35 °C), and then released into drug-free medium containing 0.033% MMS. At the restrictive temperature (35 °C), sgs1-36 PNAS | March 22, 2011 | vol. 108 | no. 12 | 4947
GENETICS
Fig. 2. HJ resolvases do not prevent the accumulation of MMS-induced Xstructures in sgs1 mutants. sgs1–GFP, sgs1–RusA, and sgs1–GEN11-527 strains were released from G1 arrest into medium containing 0.033% MMS at 30 °C. Protein expression was induced during G1 arrest and throughout the subsequent incubation. After 3 h, cells were incubated at 37 °C (to promote robust HJ resolvase activity) for a further 5 h. DNA replication intermediates were analyzed around ARS305 by 2D gel electrophoresis at the indicated times, as per Fig. 1B. X-shaped DNA structures are denoted by arrows.
Fig. 4. Reactivation of Sgs1 removes X molecules in SGS1-deficient cells. (A) Conditional inactivation of SGS1 causes unprocessed X-structures after exposure to MMS. SGS1+, sgs1Δ, and sgs1-36 strains were released from G1 arrest at 35 °C into medium containing 0.033% MMS at the indicated temperatures. DNA replication was monitored after 2 h by 2D gel electrophoresis and flow cytometry. X-shaped DNA structures are denoted by arrows. (B) After 2 h of MMS-treatment at 35 °C, sgs1Δ and sgs1-36 cells were harvested, washed, and resuspended in drug-free medium at 35 °C or 25 °C, as indicated. The phenotype of the sgs1-36 strain (denoted as sgs1 or SGS1+) at each step of the protocol is indicated in the schematic diagram. (C) DNA replication intermediates around ARS305 were analyzed in the indicated strains by 2D gel electrophoresis.
mutants resemble sgs1Δ mutants and accumulate MMS-induced X-structures at ARS305 (Fig. 4A). As expected, MMS-induced Xstructures were not detectable in SGS1+ cells or in sgs1-36 mutants returned to the permissive temperature (25 °C) immediately after the release from G1 arrest (Fig. 4A). After 2 h of treatment with 0.033% MMS at 35 °C, we followed the fate of MMS-induced X-structures in an sgs1Δ strain and in the sgs1-36 mutant at both the permissive and the restrictive temperature (Fig. 4B). In the sgs1Δ mutant, X-structures were detectable at 4 h after MMS removal at both 25 °C and 35 °C, demonstrating that sgs1 X-structure persistence is independent of temperature under these conditions (Fig. 4C). Similarly, sgs136 strains held at 35 °C after the MMS treatment resembled sgs1 deletion strains and exhibited unprocessed X-structures. Interestingly, sgs1-36 strains returned to the permissive temperature immediately after the removal of MMS did not exhibit unprocessed X-structures after 4 h in drug-free medium (Fig. 4C). Therefore, reactivation of Sgs1 in the sgs1-36 strain background promotes the processing of MMS-induced X-structures, in a manner similar to that observed for sgs1 mutants expressing HJ resolvases (Figs. 1–3). Together, these data are consistent with a role for Sgs1 in the direct removal of HJ-containing structures arising during MMS-induced HRR. Discussion In this study, we have demonstrated that expression of two evolutionarily divergent HJ resolvases, E. coli RusA and human GEN11-527, can diminish the level of MMS-induced X-structures 4948 | www.pnas.org/cgi/doi/10.1073/pnas.1014240108
that normally persist in S. cerevisiae sgs1 and TOP3Y356F strains. Given that both RusA and GEN11-527 exhibit robust HJ processing activity in vitro (40, 43–47) and do not appear to nonspecifically cleave other types of (Rad51-independent) DNA replication intermediates in vivo, we propose that the Rad51dependent X-structures we detect in sgs1 and TOP3Y356F strains probably contain HJs between sister chromatids. However, we note that an ideal specificity control would be a Rad51-dependent X-structure that definitively does not contain HJs. However, to our knowledge, a mutant exhibiting such a structure has not been identified. Previous studies have suggested that the sgs1 MMS-induced X-structures are “pseudo-Holliday junctions,” consisting of a region of hemicatenated nascent DNA and concomitant parental ssDNA (27). However, this type of DNA structure would not be efficiently cleaved by RusA or GEN11-527, which both demonstrate a preference for specific nucleotide sequences being present at the four-way junction center for efficient cutting (44, 46). Furthermore, because the X-structures arising in TOP3Y356F cells are also cleaved by RusA and GEN11527 , we suggest that the poor growth of top3 strains is unlikely to be caused by unresolved hemicatenane structures created by Sgs1 during HRR. We note, however, that we cannot rule out the possibility that different types of X-structures could exhibit some degree of interconversion and that HJ resolvase cleavage occurs once X-structures temporarily contain one or more canonical HJs. Our data are in agreement with a recent study revealing the first demonstration of HJs arising as intermediates of DSB repair in mitotic cells and a defect in the prevention and/or processing of these HJs in sgs1 mutants (38). Although X-structure intermediates of DSB repair were detectable in both haploid and diploid WT cells, confirmation that these contained HJs was only possible in diploid cells (due to strand-specific probing of “Mom” and “Dad” recombinant homologous chromosomes). Our data independently verify that unprocessed HJ-containing structures arise in haploid sgs1 cells and suggest that the processing of HJs is altered in sgs1 mutants during both DSB repair and the repair of replication-induced lesions. Furthermore, we also propose that Sgs1 functions in the direct removal of HJ-containing intermediates, because the reactivation of Sgs1 in sgs1-36 strains promotes the processing of otherwise persistent X-structures. Because unprocessed HJ-containing DNA structures also persist in TOP3Y356F cells, our data are consistent with Sgs1 and Top3 acting together, rather than sequentially, to directly process interchromatid HJs arising during HRR of MMS-induced DNA lesions. One possibility is that X-structures comprise dHJs that can be removed by Sgs1–Top3–Rmi1-mediated dHJ dissolution (18, 20, 21, 52). Top3 is likely to be required for relieving the inevitable buildup of torsional stress associated with convergent branch migration of HJs, because Sgs1 or (Drosophila) BLM alone are incapable of extensive convergent branch migration of dHJs in the absence of Top3 or TopoIIIα, respectively (52, 53). Another possibility is that X-structures comprise single HJs and that Sgs1 and Top3 act cooperatively to process these. For example, partially extended D-loops would contain a fourway junction that could theoretically either be cleaved by HJ resolvases or disrupted by the helicase activity of Sgs1. Future studies could be aimed at testing whether the controlled expression of a HJ resolvase can ameliorate certain BS phenotypes, as has been observed for human WS cells and S. pombe rqh1 mutants expressing heterologous HJ resolvases (39–41, 54). Materials and Methods S. cerevisiae Strains and Plasmids. The genotypes of strains used in this study can be found in Table S1. Details of the plasmids used can be found in SI Materials and Methods. Growth Conditions, Cell Synchronization, and Flow Cytometry Analysis. Strains were grown at 30 °C, unless indicated otherwise. Strains were synchronized in G1 with 5–20 μg/mL α-factor mating pheromone as described (29). For protein induction, galactose was added during G1 arrest and maintained at 2% throughout the subsequent analyses. Release of cells from α-factor ar-
Mankouri et al.
Western Blot Analysis. Protein extraction, SDS/PAGE, and Western blot analysis were performed as described (29). The anti-GFP antibody (Roche Diagnostics) was used at a dilution of 1:1,000.
2D Gel Electrophoresis. The CTAB method of DNA extraction and 2D gel procedures were performed as described (42). Unless stated otherwise, DNA extracts were digested with NciI and NcoI before running the firstdimension gels. Quantification of X molecules on 2D gels was performed as described (30).
ACKNOWLEDGMENTS. We thank Drs. S. Brill, L. Cox, M. A. Resnick, R. Rothstein, M. Seki, S. West, and M. Whitby for strains and plasmids; and Drs. P. McHugh and W. Niedzwiedz and various members of the I.D.H. laboratory for helpful discussions. This work was supported by Cancer Research UK and the Nordea Foundation.
1. Chu WK, Hickson ID (2009) RecQ helicases: Multifunctional genome caretakers. Nat Rev Cancer 9:644–654. 2. Ashton TM, Hickson ID (2010) Yeast as a model system to study RecQ helicase function. DNA Repair (Amst) 9:303–314. 3. Johnson FB, et al. (2000) Association of the Bloom syndrome protein with topoisomerase IIIalpha in somatic and meiotic cells. Cancer Res 60:1162–1167. 4. Wu L, et al. (2000) The Bloom’s syndrome gene product interacts with topoisomerase III. J Biol Chem 275:9636–9644. 5. Yin J, et al. (2005) BLAP75, an essential component of Bloom’s syndrome protein complexes that maintain genome integrity. EMBO J 24:1465–1476. 6. Xu D, et al. (2008) RMI, a new OB-fold complex essential for Bloom syndrome protein to maintain genome stability. Genes Dev 22:2843–2855. 7. Singh TR, et al. (2008) BLAP18/RMI2, a novel OB-fold-containing protein, is an essential component of the Bloom helicase-double Holliday junction dissolvasome. Genes Dev 22:2856–2868. 8. Gangloff S, McDonald JP, Bendixen C, Arthur L, Rothstein R (1994) The yeast type I topoisomerase Top3 interacts with Sgs1, a DNA helicase homolog: a potential eukaryotic reverse gyrase. Mol Cell Biol 14:8391–8398. 9. Fricke WM, Kaliraman V, Brill SJ (2001) Mapping the DNA topoisomerase III binding domain of the Sgs1 DNA helicase. J Biol Chem 276:8848–8855. 10. Bennett RJ, Noirot-Gros MF, Wang JC (2000) Interaction between yeast sgs1 helicase and DNA topoisomerase III. J Biol Chem 275:26898–26905. 11. Ui A, et al. (2005) The ability of Sgs1 to interact with DNA topoisomerase III is essential for damage-induced recombination. DNA Repair (Amst) 4:191–201. 12. Onodera R, et al. (2002) Functional and physical interaction between Sgs1 and Top3 and Sgs1-independent function of Top3 in DNA recombination repair. Genes Genet Syst 77:11–21. 13. Mullen JR, Nallaseth FS, Lan YQ, Slagle CE, Brill SJ (2005) Yeast Rmi1/Nce4 controls genome stability as a subunit of the Sgs1-Top3 complex. Mol Cell Biol 25:4476–4487. 14. Chang M, et al. (2005) RMI1/NCE4, a suppressor of genome instability, encodes a member of the RecQ helicase/Topo III complex. EMBO J 24:2024–2033. 15. Mankouri HW, Hickson ID (2007) The RecQ helicase-topoisomerase III-Rmi1 complex: A DNA structure-specific ‘dissolvasome’? Trends Biochem Sci 32:538–546. 16. Karow JK, Constantinou A, Li JL, West SC, Hickson ID (2000) The Bloom’s syndrome gene product promotes branch migration of holliday junctions. Proc Natl Acad Sci USA 97:6504–6508. 17. van Brabant AJ, et al. (2000) Binding and melting of D-loops by the Bloom syndrome helicase. Biochemistry 39:14617–14625. 18. Wu L, Hickson ID (2003) The Bloom’s syndrome helicase suppresses crossing over during homologous recombination. Nature 426:870–874. 19. Bachrati CZ, Borts RH, Hickson ID (2006) Mobile D-loops are a preferred substrate for the Bloom’s syndrome helicase. Nucleic Acids Res 34:2269–2279. 20. Raynard S, Bussen W, Sung P (2006) A double Holliday junction dissolvasome comprising BLM, topoisomerase IIIalpha, and BLAP75. J Biol Chem 281:13861–13864. 21. Wu L, et al. (2006) BLAP75/RMI1 promotes the BLM-dependent dissolution of homologous recombination intermediates. Proc Natl Acad Sci USA 103:4068–4073. 22. Watt PM, Hickson ID, Borts RH, Louis EJ (1996) SGS1, a homologue of the Bloom’s and Werner’s syndrome genes, is required for maintenance of genome stability in Saccharomyces cerevisiae. Genetics 144:935–945. 23. Murray JM, Lindsay HD, Munday CA, Carr AM (1997) Role of Schizosaccharomyces pombe RecQ homolog, recombination, and checkpoint genes in UV damage tolerance. Mol Cell Biol 17:6868–6875. 24. Stewart E, Chapman CR, Al-Khodairy F, Carr AM, Enoch T (1997) rqh1+, a fission yeast gene related to the Bloom’s and Werner’s syndrome genes, is required for reversible S-phase arrest. EMBO J 16:2682–2692. 25. Yamagata K, et al. (1998) Bloom’s and Werner’s syndrome genes suppress hyperrecombination in yeast sgs1 mutant: implication for genomic instability in human diseases. Proc Natl Acad Sci USA 95:8733–8738. 26. Cejka P, Kowalczykowski SC (2010) The full-length Saccharomyces cerevisiae Sgs1 protein is a vigorous DNA helicase that preferentially unwinds holliday junctions. J Biol Chem 285:8290–8301. 27. Liberi G, et al. (2005) Rad51-dependent DNA structures accumulate at damaged replication forks in sgs1 mutants defective in the yeast ortholog of BLM RecQ helicase. Genes Dev 19:339–350. 28. Branzei D, et al. (2006) Ubc9- and mms21-mediated sumoylation counteracts recombinogenic events at damaged replication forks. Cell 127:509–522. 29. Mankouri HW, Hickson ID (2006) Top3 processes recombination intermediates and modulates checkpoint activity after DNA damage. Mol Biol Cell 17:4473–4483.
30. Mankouri HW, Ngo HP, Hickson ID (2007) Shu proteins promote the formation of homologous recombination intermediates that are processed by Sgs1-Rmi1-Top3. Mol Biol Cell 18:4062–4073. 31. Gangloff S, Soustelle C, Fabre F (2000) Homologous recombination is responsible for cell death in the absence of the Sgs1 and Srs2 helicases. Nat Genet 25:192–194. 32. Fabre F, Chan A, Heyer WD, Gangloff S (2002) Alternate pathways involving Sgs1/ Top3, Mus81/ Mms4, and Srs2 prevent formation of toxic recombination intermediates from single-stranded gaps created by DNA replication. Proc Natl Acad Sci USA 99:16887–16892. 33. Laursen LV, Ampatzidou E, Andersen AH, Murray JM (2003) Role for the fission yeast RecQ helicase in DNA repair in G2. Mol Cell Biol 23:3692–3705. 34. Hope JC, Maftahi M, Freyer GA (2005) A postsynaptic role for Rhp55/57 that is responsible for cell death in Deltarqh1 mutants following replication arrest in Schizosaccharomyces pombe. Genetics 170:519–531. 35. Miyabe I, Morishita T, Hishida T, Yonei S, Shinagawa H (2006) Rhp51-dependent recombination intermediates that do not generate checkpoint signal are accumulated in Schizosaccharomyces pombe rad60 and smc5/6 mutants after release from replication arrest. Mol Cell Biol 26:343–353. 36. Chakraverty RK, et al. (2001) Topoisomerase III acts upstream of Rad53p in the Sphase DNA damage checkpoint. Mol Cell Biol 21:7150–7162. 37. Carotenuto W, Liberi G (2010) Mitotic inter-homologue junctions accumulate at damaged DNA replication forks in recQ mutants. DNA Repair (Amst) 9:661–669. 38. Bzymek M, Thayer NH, Oh SD, Kleckner N, Hunter N (2010) Double Holliday junctions are intermediates of DNA break repair. Nature 464:937–941. 39. Doe CL, Dixon J, Osman F, Whitby MC (2000) Partial suppression of the fission yeast rqh1(-) phenotype by expression of a bacterial Holliday junction resolvase. EMBO J 19: 2751–2762. 40. Doe CL, Ahn JS, Dixon J, Whitby MC (2002) Mus81-Eme1 and Rqh1 involvement in processing stalled and collapsed replication forks. J Biol Chem 277:32753–32759. 41. Lorenz A, West SC, Whitby MC (2010) The human Holliday junction resolvase GEN1 rescues the meiotic phenotype of a Schizosaccharomyces pombe mus81 mutant. Nucleic Acids Res 38:1866–1873. 42. Lopes M, Cotta-Ramusino C, Liberi G, Foiani M (2003) Branch migrating sister chromatid junctions form at replication origins through Rad51/Rad52-independent mechanisms. Mol Cell 12:1499–1510. 43. Sharples GJ, Chan SN, Mahdi AA, Whitby MC, Lloyd RG (1994) Processing of intermediates in recombination and DNA repair: identification of a new endonuclease that specifically cleaves Holliday junctions. EMBO J 13:6133–6142. 44. Ip SC, et al. (2008) Identification of Holliday junction resolvases from humans and yeast. Nature 456:357–361. 45. Rass U, et al. (2010) Mechanism of Holliday junction resolution by the human GEN1 protein. Genes Dev 24:1559–1569. 46. Chan SN, Harris L, Bolt EL, Whitby MC, Lloyd RG (1997) Sequence specificity and biochemical characterization of the RusA Holliday junction resolvase of Escherichia coli. J Biol Chem 272:14873–14882. 47. Chan SN, Vincent SD, Lloyd RG (1998) Recognition and manipulation of branched DNA by the RusA Holliday junction resolvase of Escherichia coli. Nucleic Acids Res 26: 1560–1566. 48. Giraud-Panis MJ, Lilley DM (1998) Structural recognition and distortion by the DNA junction-resolving enzyme RusA. J Mol Biol 278:117–133. 49. Bolt EL, Sharples GJ, Lloyd RG (1999) Identification of three aspartic acid residues essential for catalysis by the RusA holliday junction resolvase. J Mol Biol 286:403–415. 50. Wagner M, Price G, Rothstein R (2006) The absence of Top3 reveals an interaction between the Sgs1 and Pif1 DNA helicases in Saccharomyces cerevisiae. Genetics 174: 555–573. 51. Kaliraman V, Brill SJ (2002) Role of SGS1 and SLX4 in maintaining rDNA structure in Saccharomyces cerevisiae. Curr Genet 41:389–400. 52. Cejka P, Plank JL, Bachrati CZ, Hickson ID, Kowalczykowski SC (2010) Rmi1 stimulates decatenation of double Holliday junctions during dissolution by Sgs1-Top3. Nat Struct Mol Biol 17:1377–1382. 53. Plank JL, Wu J, Hsieh TS (2006) Topoisomerase IIIalpha and Bloom’s helicase can resolve a mobile double Holliday junction substrate through convergent branch migration. Proc Natl Acad Sci USA 103:11118–11123. 54. Rodriguez-Lopez AM, Whitby MC, Borer CM, Bachler MA, Cox LS (2007) Correction of proliferation and drug sensitivity defects in the progeroid Werner’s Syndrome by Holliday junction resolution. Rejuvenation Res 10:27–40.
Mankouri et al.
PNAS | March 22, 2011 | vol. 108 | no. 12 | 4949
GENETICS
rest, or after MMS treatment, was achieved by centrifugation, washing, and resuspension in drug-free medium. Cell cycle progression was monitored by using flow cytometry as described (29).
