Subnuclear Relocalization of Structure-Specific Endonucleases in ...

Report

Subnuclear Relocalization of Structure-Specific Endonucleases in Response to DNA Damage Graphical Abstract

Authors Irene Saugar, Alberto Jime´nez-Martı´n, Jose´ Antonio Tercero

Correspondence [email protected]

In Brief Saugar et al. find that, under conditions of DNA damage, the structure-specific endonuclease Mus81-Mms4/EME1 relocalizes to subnuclear foci, where it colocalizes with the endonucleases Rad1-Rad10 and Slx1-Slx4. Relocalization takes place to a class of stress-induced foci defined by the Cmr1 protein and correlates with the function of the endonuclease.

Highlights d

The endonuclease Mus81-Mms4 relocalizes to subnuclear foci following DNA damage

d

Mus81-Mms4 colocalizes to foci with the endonucleases Rad1-Rad10 and Slx1-Slx4

d

The endonucleases relocalize to a class of DNA damage response foci defined by Cmr1

d

Mus81-Mms4 relocalization to foci correlates with the function of the endonuclease

Saugar et al., 2017, Cell Reports 20, 1553–1562 August 15, 2017 ª 2017 The Author(s). http://dx.doi.org/10.1016/j.celrep.2017.07.059

Cell Reports

Report Subnuclear Relocalization of Structure-Specific Endonucleases in Response to DNA Damage Irene Saugar,1 Alberto Jime´nez-Martı´n,1 and Jose´ Antonio Tercero1,2,* 1Centro

de Biologı´a Molecular Severo Ochoa (CSIC/UAM), Cantoblanco, 28049 Madrid, Spain Contact *Correspondence: [email protected] http://dx.doi.org/10.1016/j.celrep.2017.07.059 2Lead

SUMMARY

Structure-specific endonucleases contribute to the maintenance of genome integrity by cleaving DNA intermediates that need to be resolved for faithful DNA repair, replication, or recombination. Despite advances in the understanding of their function and regulation, it is less clear how these proteins respond to genotoxic stress. Here, we show that the structurespecific endonuclease Mus81-Mms4/EME1 relocalizes to subnuclear foci following DNA damage and colocalizes with the endonucleases Rad1-Rad10 (XPF-ERCC1) and Slx1-Slx4. Recruitment takes place into a class of stress foci defined by Cmr1/WDR76, a protein involved in preserving genome stability, and depends on the E2-ubiquitin-conjugating enzyme Rad6 and the E3-ubiquitin ligase Bre1. Foci dynamics show that, in the presence of DNA intermediates that need resolution by Mus81-Mms4, Mus81 foci persist until this endonuclease is activated by Mms4 phosphorylation. Our data suggest that subnuclear relocalization is relevant for the function of Mus81-Mms4 and, probably, of the endonucleases that colocalize with it. INTRODUCTION Structure-specific endonucleases play a crucial role in the maintenance of genome stability, as they are necessary to resolve a variety of secondary DNA intermediates that are generated during DNA replication, DNA repair, or DNA recombination (Dehe´ and Gaillard, 2017; Rass, 2013; Schwartz and Heyer, 2011). Their nucleolytic function ensures faithful completion of these processes, thereby contributing importantly to preserving genomic integrity in every cell cycle. The activity of structure-specific endonucleases is highly regulated to guarantee the efficient resolution of their DNA substrates. At the same time, these proteins need to be tightly controlled to avoid the unscheduled cleavage of DNA intermediates, which could lead to genomic instability due to chromosomal rearrangements, high levels of chromatid exchanges, or faulty replication. These controls are achieved by different regulatory mechanisms that include posttranslational modifications and interactions with other factors that stimulate their activity (Cussiol et al., 2017; Dehe´ and Gaillard, 2017; Minocherhomji

and Hickson, 2014; Sarangi and Zhao, 2015; West et al., 2015). For example, in budding yeast and human cells, the activity of the endonuclease Mus81-Mms4/EME1 is strictly regulated by cyclin-dependent kinase (CDK)- and polo-like kinase (PLK)dependent phosphorylation of the non-catalytic subunit Mms4/ EME1 (Gallo-Ferna´ndez et al., 2012; Matos et al., 2011; Saugar et al., 2013; Szakal and Branzei, 2013) and, in fission yeast, by CDK-dependent phosphorylation of Eme1 (Dehe´ et al., 2013). At least in budding yeast, Mms4 phosphorylation also depends on Cdc7-Dbf4 (Princz et al., 2017). This modification restricts the full activity of this endonuclease to the end of S-phase and mitosis. The function of structure-specific endonucleases is important for cell survival under DNA damage conditions, which can be caused by endogenous and environmental agents, as well as by cancer therapeutic treatments. However, despite the role of these proteins in coping with DNA lesions, little is known about how they respond to the presence of DNA damage. Protein relocalization is a hallmark of the cellular response to DNA damage or replicative stress (Lisby et al., 2004; Tkach et al., 2012), but the potential regulation of structure-specific endonucleases by this mechanism remains poorly understood. Here, we show that Mus81-Mms4/EME1 and other endonucleases undergo cellular relocalization upon DNA damage. Mus81-Mms4 is recruited into subnuclear foci following genotoxic stress and colocalizes at these sites with the endonucleases Rad1-Rad10 and Slx1Slx4. This relocalization could be important for the function of these proteins. RESULTS Mus81-Mms4 Relocalizes to Subnuclear Foci in Response to DNA Damage To study the localization of different structure-specific endonucleases, we tagged the proteins with GFP, which yielded Saccharomyces cerevisiae strains that behaved similarly to parental controls (Figure S1A). We started analyzing Mus81Mms4 and found that in untreated cells, the two subunits appeared diffused in the nucleus, whereas they formed subnuclear foci after treatment with the DNA-damaging agent methyl methanesulfonate (MMS) (Figures 1A and 1F). These foci localized to perinuclear regions and usually appeared as one dot per cell (Figure 1A). Mus81 and Mms4 accumulation into foci was not due to an increase in their levels after DNA damage (Figures S1B and S1C), thus reflecting relocalization. Of note, although Mus81 and Mms4 form a complex that is essential for the nuclease activity (Kaliraman et al., 2001), both subunits

Cell Reports 20, 1553–1562, August 15, 2017 ª 2017 The Author(s). 1553 This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

A

B

C

D

E G

F

Figure 1. Mus81-Mms4 Is Recruited to Subnuclear Foci following DNA Damage (A) Cycling GFP-MUS81 (YSG39 strain) and GFP-MMS4 (YSG42 strain) cells were examined by fluorescence microscopy before and after treatment with 0.033% MMS for 60 min. The percentage of cells containing foci is shown in (F). Arrows indicate selected foci. (B) GFP-MUS81 mms4D (YSG41 strain) and GFP-MMS4 mus81D (YSG192 strain) cells were treated as in (A). The percentage of cells containing foci is shown in each case. (C) GFP-MUS81 (YSG413 strain) and GFP-mus81-DD (YSG414 strain) cells were treated as in (A). The percentage of cells containing foci is indicated.

(legend continued on next page)

1554 Cell Reports 20, 1553–1562, August 15, 2017

relocalized to foci independently of each other (Figure 1B). This indicates that neither the formation nor the activity of the Mus81-Mms4 complex affect the relocalization of their subunits. Moreover, Mus81 recruitment into foci does not require the nuclease activity of this catalytic subunit, as a Mus81 nuclease-dead mutant relocalizes normally to foci after DNA damage (Figure 1C). We then studied the formation of Mus81 foci following treatment with other drugs that cause DNA damage or replicative stress. A high percentage of cells showed Mus81 foci after exposure to H2O2, 4NQO, or hydroxyurea, but not to camptothecin or phleomycin (Figure 1D). Previously, it had been shown that Mus81 forms zeocin-induced foci (Sebesta et al., 2017). We also analyzed Mus81 foci formation in different DNA repair, DNA damage tolerance, or genome integrity checkpoint mutants. Cells lacking the RecQ helicase Sgs1 (BLM), which has overlapping functions with Mus81-Mms4 (Kaliraman et al., 2001), or the E3-ubiquitin ligase Rad18, involved in DNA damage tolerance, showed a significant increase in the percentage of cells containing Mus81 foci with respect to the wild-type control (Figure 1E). Likewise, a higher number of cells with Mus81 foci were detected in mutants lacking Mec1 (ATR), but not in those deficient in other checkpoint proteins. There was also an increase in the number of cells with Mus81 foci in mutants lacking nucleotide excision repair or base excision repair, but not homologous recombination (Figure 1E). Collectively, these results show that Mus81Mms4 relocalizes to subnuclear foci in response to DNA damage or replicative stress and also in mutants that may accumulate unrepaired endogenous DNA lesions or unresolved DNA intermediates. We next analyzed Mus81 and Mms4 foci along the cell cycle. Both proteins formed foci after induced DNA damage in G1- and S-phase cells, but not in G2/M cells (Figures 1F, S2A, and S2B). This was not a consequence of the nocodazole treatment used for G2/M synchronization, as cells blocked in G1 and then released into S-phase in medium containing this drug formed Mus81 and Mms4 foci upon DNA damage (Figure S2C). We also examined the chromatin association of this endonuclease throughout the cell cycle. As shown in Figure 1G, Mus81 and Mms4 were mostly bound to chromatin in G1-phase, S-phase, and G2/M-arrested cells, both under unperturbed conditions and in response to DNA damage. This result indicates that Mus81 and Mms4 are chromatin components and therefore that their relocalization to foci under DNA-damaging conditions occurs within the chromatin.

Structure-Specific Endonucleases Colocalize to Foci following DNA Damage In metazoans, MUS81-EME1/Mms4 forms a multi-protein complex with other structure-specific endonucleases, such as XPFERCC1 (Rad1-Rad10 in S. cerevisiae) and SLX1-SLX4 (Fekairi et al., 2009; Mun˜oz et al., 2009; Svendsen et al., 2009). In budding yeast, Mus81-Mms4 can associate at least with Slx4 (Gritenaite et al., 2014). We therefore asked whether Rad1Rad10 and Slx1-Slx4 behave similarly to Mus81-Mms4 in the response to DNA damage. The catalytic subunits Rad1 and Slx1 relocalized to subnuclear foci in a high percentage of cells after MMS treatment (Figures 2A and 2B), similar to Mus81. Likewise, Slx4, which is not only a subunit of Slx1-Slx4, but is also a platform for several proteins involved in genome integrity maintenance (Cussiol et al., 2017), formed foci upon exposure to MMS (Figures 2A and 2B), in agreement with previous data (Tkach et al., 2012). As for Mus81-Mms4, the accumulation of Rad1, Slx1, and Slx4 into foci was not due to increased protein levels upon DNA damage (Figures S1D–S1F), thus reflecting relocalization. The observed foci were similar to those of Mus81-Mms4, and, remarkably, the analyzed endonucleases colocalized at these sites (Figures 2C and 2D). These results strongly suggest that Mus81-Mms4, Rad1-Rad10, and Slx1-Slx4 are recruited into the same foci through a common DNA-damage-induced signal. Nevertheless, despite their colocalization, they formed foci independently of each other, even in the absence of the Slx4 scaffold (Figure 2E). Endonuclease Relocalization Is Rad6- and Bre1Dependent and Occurs to a Class of DNA-DamageInduced Foci Defined by Cmr1 To investigate the potential signal(s) for endonuclease recruitment into foci, we analyzed possible genetic dependencies for foci formation. We examined Mus81 foci after DNA damage in the mutants that did not accumulate this protein when cells were untreated (Figure 1E). We detected Mus81 foci in the absence of homologous recombination or checkpoint proteins after MMS treatment. By contrast, no Mus81 foci were found in cells lacking the E2-ubiquitin-conjugating enzyme Rad6 (human HR6A and HR6B) (Figure 3A). Rad6 participates in several processes in conjunction with different E3-ubiquitin ligases, Rad18, Ubr1, and Bre1 (Game and Chernikova, 2009), and we assessed the contribution of these to Mus81 foci formation (Figure 3B). Cells lacking the DNA damage tolerance protein Rad18, and cells lacking Ubr1, which together with Rad6 is necessary for protein degradation by the N-end rule pathway, showed Mus81

(D) Cycling GFP-MUS81 (YSG39 strain) cells were treated for 60 min with 0.4 mM H2O2, 250 mM camptothecin, 50 mg/mL phleomycin, 0.5 mg/mL 4NQO, or 0.2 M hydroxyurea. The percentage of cells showing Mus81 foci is shown in each case. Arrows indicate selected foci. (E) Percentage of cycling cells of different mutants containing spontaneous Mus81 foci. The strains are listed in Table S1. (F) Percentage of cells containing Mus81 (YSG39 strain) or Mms4 (YSG42 strain) foci at different stages of the cell cycle. For G1 analysis, cells were synchronized using a factor and held for an additional 60 min in G1 in medium without or with 0.033% MMS. For S-phase, G1-blocked cells were released in medium either without (30 min) or with 0.033% MMS (60 min). For G2/M, cells were synchronized with nocodazole and held in G2/M for an additional 60 min in medium ± MMS (0.033%). (G) Chromatin-binding assays of Mus81 and Mms4. Whole cell extracts (W) from G1-, S-, or G2/M-phase cells were fractionated into a soluble fraction (S) and a chromatin-enriched fraction (C) and analyzed by immunoblot. Synchronization and MMS treatment were as in (F). Strains: GFP-MUS81 (YSG39); GFP-MMS4 (YSG42). PGK and Orc6 were used as controls for soluble and chromatin fractions, respectively. In all cases, the histograms represent the mean ± SD from three different experiments. See also Figures S1 and S2.

Cell Reports 20, 1553–1562, August 15, 2017 1555

A

C

B

D

E

Figure 2. Rad1, Slx1, and Slx4 Relocalize to DNA-Damage-Induced Foci and Colocalize with Mus81 (A) Cycling GFP-RAD1 (YSG271 strain), GFP-SLX1 (YSG356 strain), and GFP-SLX4 (YSG235 strain) cells were examined by fluorescence microscopy before and after treatment with 0.033% MMS for 60 min. For simplicity, only images of MMS-treated cells are shown. Arrows indicate selected foci. (B) Percentage of cells containing foci of Rad1, Slx1, or Slx4, before and after treatment with 0.033% MMS for 60 min. (C) Colocalization of the endonucleases at DNA-damage-induced foci. Cycling RAD1-mCHERRY GFP-MUS81 (YSG280 strain) (upper panel), SLX1-mCHERRY GFP-MUS81 (YSG363 strain) (middle panel), and SLX4-mCHERRY GFP-MUS81 (YSG274 strain) (bottom panel) cells were examined by fluorescence microscopy after treatment with 0.033% MMS for 60 min. (D) Quantification of the data in (C). (E) Cycling cells were examined by fluorescence microscopy before and after treatment with 0.033% MMS for 60 min. Histograms show the percentage of cells containing foci of every protein in the absence of the others. Strains: GFP-MUS81 rad1D (YSG299), GFP-MUS81 slx1D (YSG353), and GFP-MUS81 slx4D (YSG81) (upper panel, left); GFP-RAD1 mus81D (YSG307), GFP-RAD1 slx1D (YSG376), and GFP-RAD1 slx4D (YSG305) (upper panel, right); GFP-SLX1 mus81D (YSG372), GFP-SLX1 rad1D (YSG373), and GFP-SLX1 slx4D (YSG374) (bottom panel, left); and GFP-SLX4 mus81D (YSG304), GFP-SLX4 rad1D (YSG302), and GFP-SLX4 slx1D (YSG377) (bottom panel, right). In all cases, the histograms represent the mean ± SD from three different experiments. See also Figure S1.

1556 Cell Reports 20, 1553–1562, August 15, 2017

foci after exposure to MMS. However, Mus81 foci formation was highly reduced in the absence of Bre1 (human RNF20 and RNF40) (Figure 3B), which in conjunction with Rad6 monoubiquitylates histone H2B (Hwang et al., 2003; Wood et al., 2003). However, Mus81 foci formation was not affected in cells lacking Dot1 or Set1, which function downstream of Bre1 to methylate histone H3 (Feng et al., 2002; Sun and Allis, 2002). The Rad6/Bre1 dependency for foci formation suggests that histone ubiquitylation might be required for Mus81 relocalization. Notably, the formation of Rad1, Slx1, and Slx4 foci was also Rad6 and Bre1 dependent (Figure 3C), reinforcing the idea of a common signal for their colocalization. Interestingly, although this signal for recruitment into foci is triggered by DNA damage and might involve histone modification, it was independent of H2A phosphorylation, as Mus81 relocalizes to foci in the non-phosphorylatable mutant H2A-S129A and in null mutants of its readers (rtt107D or rad9D) (Figure S3A). To further investigate the nature of endonuclease foci, we analyzed whether they colocalize with any known nuclear structure. We did not find colocalization of Mus81 foci with DNA recombination/repair (represented by Rad52), DNA replication (Rfa1), nucleolus (Nop1), nuclear pore (Nup84), telomeres (Rap1), or spindle pole body (SPC42) proteins (Figures 3D and 3F). Remarkably, however, Mus81 foci colocalized with Cmr1/ WDR76 (Figures 3E and 3F), a protein related to histones that is involved in genome stability maintenance and interacts with DNA repair, chromatin, and replication factors (Gallina et al., 2015; Tkach et al., 2012). Cmr1 has defined a class of DNA damage response foci to where different proteins relocalize, some of which are implicated in mitotic cell cycle, chromosome organization, or spindle organization (Gallina et al., 2015; Tkach et al., 2012). Supporting that these endonucleases are components of this type of foci, their relocalization was dependent on Btn2, a protein that directs Cmr1 to these sites and is likely to be structurally involved in their formation (Gallina et al., 2015) (Figure 3G). Nevertheless, the endonucleases were recruited into these foci independently of Cmr1 (Figure 3H). Importantly, the Rad6/Bre1 dependency for endonuclease relocalization could be a feature for other components of this class of foci, as Cmr1 recruitment into foci is reduced in the absence of these proteins (Figure 3I). Of note, although Cmr1-foci formation is dependent on the histone deacetylase Hos2 (Tkach et al., 2012), Mus81, Rad1, Slx1, and Slx4 formed DNA-damage-induced foci normally in cells lacking this protein (Figure S3B). These results indicate the existence of common and particular signals for protein recruitment into this type of foci. Foci Dynamics Correlate with Endonuclease Function We next investigated the biological significance of endonuclease relocalization. Like Cmr1, but different to other proteins that are recruited into these foci (Gallina et al., 2015), the nucleases did not relocalize to foci for degradation or as a consequence of misfolding, as they did not undergo changes in their levels in response to DNA damage and were functional (Figure S1). Moreover, the relocalization was not affected by the presence of the microtubule polymerization inhibitor nocodazole (Figure S2C), which is thought to be a requisite for a protein to be considered as misfolded (Gallina et al., 2015 and references therein).

One possibility was that the recruitment into foci could be related to the regulation of the nuclease activity. To test this idea, we focused on Mus81-Mms4, whose regulation has been extensively studied. This endonuclease is fully activated by phosphorylation of Mms4, which only occurs after bulk genome replication (Gallo-Ferna´ndez et al., 2012; Matos et al., 2011; Princz et al., 2017; Saugar et al., 2013; Szakal and Branzei, 2013) and is a mode of regulation that is not altered by DNA damage (Saugar et al., 2013). As we detected foci in G1- and S-phases, but not in G2/M, it was possible that Mus81-Mms4 was recruited into foci to help avoid Mms4 phosphorylation until G2/M. To investigate this hypothesis, we analyzed the phosphorylation of this protein during the cell cycle in btn2D cells, in which there is no foci formation, in the absence and in the presence of DNA damage. We found that Mms4 phosphorylation was prevented if cells did not reach G2/M, as it happened when btn2D and BTN2+ control cells were exposed to MMS after release from G1 arrest (Figures S4A and S4B). Therefore, Mus81-Mms4 recruitment into foci following DNA damage is not required to avoid Mms4 phosphorylation until G2/M. Conversely, protein recruitment into foci is not a prerequisite for Mms4 phosphorylation, as rad53Dbtn2D cells, which do not form Mus81 foci, showed premature Mms4 phosphorylation in the presence of MMS with respect to control cells (Figures S4A and S4C), as a consequence of a defective checkpoint (Szakal and Branzei, 2013). We then analyzed whether Mus81-Mms4 relocalization to foci could be related to the function of the endonuclease, rather than to the regulation of the complex by phosphorylation. We reasoned that, if this were the case, Mus81 foci should persist until Mus81-Mms4 can carry out its function under conditions in which some DNA structures need to be resolved by this endonuclease. To test this hypothesis, we evaluated the dynamics of Mus81 foci assembly/disassembly after treatment of cells in S-phase with MMS and subsequent removal of the drug (Figures 4A–4C). In control cells, Mus81 foci began to disappear soon after MMS removal (Figure 4B), indicating that the signal for protein recruitment into foci was no longer present. By contrast, in cells lacking the RecQ helicase Sgs1, which after MMS treatment accumulate DNA recombination intermediates that are resolved later by Mus81-Mms4 (Ashton et al., 2011; Szakal and Branzei, 2013), Mus81 foci persist until the complex is fully activated by Mms4 phosphorylation (Figures 4B and 4C). This suggests that, once activated, Mus81-Mms4 resolves the accumulated intermediates and the signaling for protein relocalization disappears. Consistent with these data, in the absence of Mms4, and therefore of nuclease activation, a high percentage of cells did not disassemble Mus81 foci or did so much later under similar experimental conditions (Figures 4D–4F), very likely because some DNA structures that needed resolution by the endonuclease persisted and there was signaling for Mus81 recruitment. In agreement with these results, Mus81 foci gradually disappeared under the same conditions, even before Mus81-Mms4 activation, in mutants such as sgs1Dctf4D and sgs1Drad51D (Figures S4D and S4E), in which the accumulation of DNA recombination intermediates is greatly reduced (Branzei et al., 2008; Fumasoni et al., 2015; Liberi et al., 2005). Finally, supporting these results, Mus81 foci disappeared earlier in a

Cell Reports 20, 1553–1562, August 15, 2017 1557

A

B

D

C

F

G

H

I

E

Figure 3. Endonuclease Relocalization Occurs to a Class of Foci Defined by Cmr1 and Depends on Rad6 and Bre1 (A) Percentage of cells containing Mus81 foci in different mutants after treatment of cycling cells with 0.033% MMS for 60 min. Strains: GFP-MUS81 (wild-type, WT, YSG39), GFP-MUS81 rad6D (YSG215), GFP-MUS81 rad52D (YSG51), GFP-MUS81 rad53D chk1D (YSG55), GFP-MUS81 rad17D (YSG74), GFP-MUS81 rad9D (YSG54), and GFP-MUS81 mrc1D (YSG188). (B) Percentage of cells containing Mus81 foci in different mutants before and after treatment of cycling cells with 0.033% MMS for 60 min. Strains: GFP-MUS81 rad18D (YSG96), GFP-MUS81 ubr1D (YSG232), GFP-MUS81 bre1D (YJT167), GFP-MUS81 dot1D (YJT168), and GFP-MUS81 set1D (YSG312).

(legend continued on next page)

1558 Cell Reports 20, 1553–1562, August 15, 2017

significant number of cells when Mms4 was prematurely phosphorylated due to ectopic expression of Cdc5 (PLK1) (Figures 4G–4I), most probably because there was anticipated resolution of accumulated DNA intermediates by this endonuclease, which stopped signaling for protein recruitment into foci. Altogether, our results indicate that Mus81 is recruited into foci following DNA damage, and that these foci persist until Mus81-Mms4 is fully activated and carries out its function in the resolution of accumulated DNA intermediates, which in turn eliminates the signaling for recruitment. These results correlate the localization of the Mus81-Mms4 complex at foci with its function, suggesting that subnuclear relocalization in response to DNA damage is important for the role of this endonuclease. DISCUSSION The data presented in this work have shown that the structurespecific endonucleases Mus81-Mms4, Rad1-Rad10, and Slx1Slx4 relocalize to subnuclear foci in response to DNA damage, where all of them colocalize. Using complementary experiments, we have found a direct correlation between the persistence of Mus81 foci and the existence of DNA intermediates that need to be resolved by Mus81-Mms4. Foci dynamics has indicated that, if these intermediates are present: (1) Mus81 foci persist until Mus81-Mms4 is fully activated by Mms4 phosphorylation and can carry out its nucleolytic function; (2) Mus81 foci do not disappear or do so much later when Mus81-Mms4 is not activated; and (3) Mus81 foci disappear earlier when Mus81-Mms4 is prematurely activated. In addition, Mus81 foci do not remain (after eliminating the DNA-damaging agent) when there is no DNA intermediates accumulation, even in the absence of Mus81-Mms4 activation. These findings correlate the relocalization of Mus81-Mms4 with its function, strongly suggesting that shuttling through the DNA-damage-induced foci could be important for the role of this complex and, by extension, of the other endonucleases. Our results have shown that recruitment into foci does not prevent the phosphorylation of Mms4 that leads to Mus81-Mms4 activation, and it is also not necessary for the modification of Mms4. We propose that the accumulation or sequestration of the endonucleases into foci following genotoxic stress could serve as a prepping mechanism that readies these proteins for

further action, probably facilitating their subsequent function if substrates needing nucleolytic resolution are formed. It is also reasonable to think that the colocalization of the endonucleases might favor the coordination among them and, in turn, their correct function when this is necessary. Importantly, endonuclease colocalization is likely a universal mechanism, as in human cells MUS81 colocalizes with ERCC1 after aphidicolin treatment (Naim et al., 2013; Ying et al., 2013), and SLX4 colocalizes with XPF-ERCC1 foci even in the absence of genotoxic stress (Wilson et al., 2013). Interestingly, although the endonucleases colocalize, they are recruited into foci independently of each other, suggesting that their interactions, as well as the formation of complexes and macro-complexes, might occur and perhaps be facilitated after relocalization to these sites. Our results have also shown that, under DNA-damaging conditions, the endonucleases colocalize with Cmr1, which marks a class of DNA damage response foci (Tkach et al., 2012). This protein is involved in the maintenance of genome integrity and has been associated with histones, as well as with DNA repair, DNA replication, and checkpoint proteins (Gallina et al., 2015; Tkach et al., 2012). Recently, it has been proposed that Cmr1 defines a nuclear compartment called INQ that is established in response to genotoxic stress and is composed of a large number of proteins, including some involved in DNA metabolism and cellcycle control (Gallina et al., 2015). The diversity of proteins found at these foci (Gallina et al., 2015; Tkach et al., 2012) indicates that multiple processes can be regulated by protein recruitment into these compartments. Here, we have identified structure-specific endonucleases as components of these foci and therefore new processes than can be controlled by relocalization to these sites. In turn, this finding opens up the possibility that still unexplored interactions exist between the endonucleases and other components of these foci. These data highlight the relevance of this type of foci as subnuclear sites that are important for diverse biological processes, some of which perhaps remain unidentified. Notably, endonuclease recruitment into foci depends on Rad6 and Bre1. Due to the role of these proteins in histone modification (Hwang et al., 2003; Wood et al., 2003), these data suggest that the relocalization of the endonucleases after DNA damage may involve at least histone ubiquitylation. This potential modification might remodel the chromatin to allow the access of some proteins to DNA or to provide a permissive environment for these

(C) Percentage of cells containing Mus81-, Rad1-, Slx1-, or Slx4-foci in rad6D or bre1D mutants after treatment of cycling cells with 0.033% MMS. Strains: GFPMUS81 (WT, YSG39), GFP-MUS81 rad6D (YSG215), GFP-MUS81 bre1D (YJT167), GFP-RAD1 (WT, YSG271), GFP-RAD1 rad6D (YSG287), GFP-RAD1 bre1D (YSG315), GFP-SLX1 (WT, YSG356), GFP-SLX1 rad6D (YSG368), GFP-SLX1 bre1D (YSG370), GFP-SLX4 (WT, YSG235), GFP-SLX4 rad6D (YSG246), and GFPSLX4 bre1D (YSG264). (D) Analysis of Mus81 foci nuclear localization. Exponentially growing cultures were examined by fluorescence microscopy after treatment with 0.033% MMS for 60 min. Strains: RAD52-GFP MUS81-mCHERRY (YSG283), RFA1-GFP MUS81-mCHERRY (YSG309), NOP1-mCHERRY GFP-MUS81 (YSG276), NUP84mCHERRY GFP-MUS81 (YSG278), RAP1-mCHERRY GFP-MUS81 (YSG326), and SPC42-EQFP GFP-MUS81 (YSG308). Quantification is shown in (F). (E) Colocalization of Mus81 and Cmr1 foci. Cycling CMR1-mCHERRY GFP-MUS81 (YSG324 strain) cells were examined by fluorescence microscopy after treatment with 0.033% MMS for 60 min. Quantification is shown in (F). (F) Quantification of the experiments in (D) and (E). (G and H) Percentage of cells containing foci of each nuclease in cells lacking Btn2 (G) or Cmr1 (H), before and after treatment of exponentially growing cultures with 0.033% MMS for 60 min. Strains: GFP-MUS81 btn2D (YSG310), GFP-RAD1 btn2D (YSG322), and GFP-SLX1 btn2D (YSG369); GFP-SLX4 btn2D (YSG320); GFP-MUS81 cmr1D (YSG338), GFP-RAD1 cmr1D (YSG341), GFP-SLX1 cmr1D (YSG375), and GFP-SLX4 cmr1D (YSG340). (I) Percentage of cells containing Cmr1 foci in rad6D and bre1D mutants after treatment of exponentially growing cultures with 0.033% MMS for 60 min. Strains: CMR1-GFP (WT, YSG342), CMR1-GFP rad6D (YSG344), and CMR1-GFP bre1D (YSG345). In all cases, the histograms represent the mean ± SD from three different experiments. See also Figure S3.

Cell Reports 20, 1553–1562, August 15, 2017 1559

A

B

C

D

E

G

H

F

I

Figure 4. Foci Dynamics Correlate with Mus81-Mms4 Activity (A–C) Cycling GFP-MUS81 HA-MMS4 SGS1+ (YSG45 strain) and GFP-MUS81 HA-MMS4 sgs1D (YSG133 strain) cells were synchronized in G1 and released into S-phase in medium containing 0.033% MMS for 60 min. MMS was then removed to allow cell-cycle progression. (A) Cell-cycle progression was monitored by flow cytometry. (B) Percentage of cells containing Mus81 foci at every time point. (C) Immunoblot analysis of Mms4 during the course of the experiment. (D–F) GFP-MUS81 HA-MMS4 sgs1D (YSG133 strain) and GFP-MUS81 pGAL-HA-MMS4 sgs1D (YSG99 strain) cells were synchronized in G1 in medium with 0.5% galactose. The cells were held in G1 for 60 min in medium with 2% glucose (to switch off the GAL promoter to allow Mms4 elimination) and then released in medium containing 0.033% MMS. The MMS was removed after 60 min to allow cell-cycle progression. (D) Cell-cycle progression was monitored by flow cytometry. (E) Percentage of cells containing Mus81 foci at every time point. (F) Immunoblot analysis of Mms4 during the course of the experiment. (G–I) GFP-MUS81 HA-MMS4 sgs1D CDC5+ (YSG133 strain) and GFP-MUS81 HA-MMS4 sgs1D pGAL-CDC5 (YSG95 strain) cells were grown in medium with 0.1% glucose (GAL promoter OFF), synchronized in G1, and released in the presence of 0.2 M hydroxyurea (HU) for 60 min. The cultures were transferred to medium with 2% galactose (GAL promoter ON) for 60 min to induce Cdc5 expression. After this arrest in early S-phase, HU was removed and the cells were treated with 0.033% MMS for 60 min, after which, the drug was removed to allow cell-cycle progression.

(legend continued on next page)

1560 Cell Reports 20, 1553–1562, August 15, 2017

proteins to accumulate when cells need to cope with DNA damage. Alternatively, or additionally, histone modification might allow the recruitment of some protein(s) that could be necessary for nuclease relocalization to foci. Efficient regulation of structure-specific endonucleases is required to ensure resolution of certain DNA intermediates, as well as to prevent the potential genomic instability derived from their incorrect or untimely function. Subnuclear relocalization upon DNA damage should contribute to the important role of these proteins in the maintenance of genome integrity. EXPERIMENTAL PROCEDURES Yeast Strains The yeast strains are listed in Table S1. Yeasts were routinely grown in YP medium with 2% glucose. Cell-Cycle Synchronization, Flow Cytometry, and ChromatinBinding Assays Cells were synchronized in G1 with a factor (5–10 mg/mL) and in G2/M with nocodazole (2 mg/mL). Samples for flow cytometry were processed as described (Saugar et al., 2013) and analyzed using a FACSCalibur flow cytometer (BD Biosciences). Chromatin isolation was as described (Donovan et al., 1997). Fluorescence Microscopy For fluorescence microscopy analysis, cells were usually grown at 30 C in minimal medium supplemented with adenine (130 mg/mL) and yeast synthetic drop out. The experiments with rad6D and bre1D strains were carried out at 25 C, due to the described temperature sensitivity of these mutants. Live cells were examined using a Zeiss Axiovert 200 fluorescence microscope with a 100 3 oil immersion objective (1.45 NA) and MetaMorph imaging software. For each field of cells captured, five fluorescent images were obtained at 1 mm intervals along the z axis using a Hamamatsu charge-coupled device (CCD) camera. The images were processed and colored using ImageJ software and prepared for publication using Adobe Photoshop. The percentage of cells containing foci was calculated in each case after analyzing at least three independent experiments. Approximately 300 cells were counted in each experiment. Protein Analysis Protein extracts for immunoblot were prepared from trichloroacetic acidtreated cells, as described (Saugar et al., 2013). Haemagglutinin (HA)- and GFP-tagged proteins were detected with 12CA5 (Centro de Biologı´a Molecular Severo Ochoa [CBMSO]) and anti-GFP (Roche) antibodies, respectively. The anti-Orc6 antibody (SB49) was a gift from Cancer Research UK. The antiPGK antibody was from Invitrogen. The secondary antibody was horseradish peroxidase (HRP)-coupled anti-mouse (Vector Labs). Statistical Analysis Statistical analysis was performed using a two-tailed Student’s t test. A p value of < 0.05 was considered statistically significant.

AUTHOR CONTRIBUTIONS I.S. and A.J.-M. performed the experiments. I.S. and J.A.T. designed the experiments and analyzed the data. J.A.T. supervised the project and wrote the manuscript. ACKNOWLEDGMENTS We thank Crisanto Gutie´rrez for critical reading of the manuscript, Karim Labib for the SPC42-EQFP and RFA1-GFP strains, Dana Branzei for the H2A-S129A mutant, and Miguel G. Blanco for the mus81-dd mutant. This work was supported by the Spanish Ministry of Economy, Industry, and Competitiveness (MINECO), grants BFU2013-43766-P and BFU2016-77663-P AEI/FEDER UE. A.J.-M. was the recipient of a predoctoral fellowship (FPI) from MINECO. The CBMSO receives an institutional grant from Fundacio´n Ramo´n Areces. Received: December 28, 2016 Revised: May 9, 2017 Accepted: July 20, 2017 Published: August 15, 2017 REFERENCES Ashton, T.M., Mankouri, H.W., Heidenblut, A., McHugh, P.J., and Hickson, I.D. (2011). Pathways for Holliday junction processing during homologous recombination in Saccharomyces cerevisiae. Mol. Cell. Biol. 31, 1921–1933. Branzei, D., Vanoli, F., and Foiani, M. (2008). SUMOylation regulates Rad18mediated template switch. Nature 456, 915–920. Cussiol, J.R., Dibitetto, D., Pellicioli, A., and Smolka, M.B. (2017). Slx4 scaffolding in homologous recombination and checkpoint control: lessons from yeast. Chromosoma 126, 45–58. Dehe´, P.M., and Gaillard, P.H. (2017). Control of structure-specific endonucleases to maintain genome stability. Nat. Rev. Mol. Cell Biol. 18, 315–330. Dehe´, P.M., Coulon, S., Scaglione, S., Shanahan, P., Takedachi, A., Wohlschlegel, J.A., Yates, J.R., 3rd, Llorente, B., Russell, P., and Gaillard, P.H. (2013). Regulation of Mus81-Eme1 Holliday junction resolvase in response to DNA damage. Nat. Struct. Mol. Biol. 20, 598–603. Donovan, S., Harwood, J., Drury, L.S., and Diffley, J.F.X. (1997). Cdc6pdependent loading of Mcm proteins onto pre-replicative chromatin in budding yeast. Proc. Natl. Acad. Sci. USA 94, 5611–5616. Fekairi, S., Scaglione, S., Chahwan, C., Taylor, E.R., Tissier, A., Coulon, S., Dong, M.Q., Ruse, C., Yates, J.R., 3rd, Russell, P., et al. (2009). Human SLX4 is a Holliday junction resolvase subunit that binds multiple DNA repair/ recombination endonucleases. Cell 138, 78–89. Feng, Q., Wang, H., Ng, H.H., Erdjument-Bromage, H., Tempst, P., Struhl, K., and Zhang, Y. (2002). Methylation of H3-lysine 79 is mediated by a new family of HMTases without a SET domain. Curr. Biol. 12, 1052–1058. Fumasoni, M., Zwicky, K., Vanoli, F., Lopes, M., and Branzei, D. (2015). Errorfree DNA damage tolerance and sister chromatid proximity during DNA replication rely on the Pola/Primase/Ctf4 Complex. Mol. Cell 57, 812–823.

SUPPLEMENTAL INFORMATION

Gallina, I., Colding, C., Henriksen, P., Beli, P., Nakamura, K., Offman, J., Mathiasen, D.P., Silva, S., Hoffmann, E., Groth, A., et al. (2015). Cmr1/ WDR76 defines a nuclear genotoxic stress body linking genome integrity and protein quality control. Nat. Commun. 6, 6533.

Supplemental Information includes four figures and one table and can be found with this article online at http://dx.doi.org/10.1016/j.celrep.2017.07.059.

Gallo-Ferna´ndez, M., Saugar, I., Ortiz-Baza´n, M.A., Va´zquez, M.V., and Tercero, J.A. (2012). Cell cycle-dependent regulation of the nuclease activity of Mus81-Eme1/Mms4. Nucleic Acids Res. 40, 8325–8335.

(G) Cell-cycle progression was monitored by flow cytometry. (H) Percentage of cells containing Mus81 foci at every time point. (I) Immunoblot analysis of Mms4 during the course of the experiment. In all cases, the histograms represent the mean ± SD from three different experiments. Data were analyzed using Student’s t test. **p < 0.01 and ***p < 0.001. See also Figure S4.

Cell Reports 20, 1553–1562, August 15, 2017 1561

Game, J.C., and Chernikova, S.B. (2009). The role of RAD6 in recombinational repair, checkpoints and meiosis via histone modification. DNA Repair (Amst.) 8, 470–482. Gritenaite, D., Princz, L.N., Szakal, B., Bantele, S.C., Wendeler, L., Schilbach, S., Habermann, B.H., Matos, J., Lisby, M., Branzei, D., and Pfander, B. (2014). A cell cycle-regulated Slx4-Dpb11 complex promotes the resolution of DNA repair intermediates linked to stalled replication. Genes Dev. 28, 1604–1619. Hwang, W.W., Venkatasubrahmanyam, S., Ianculescu, A.G., Tong, A., Boone, C., and Madhani, H.D. (2003). A conserved RING finger protein required for histone H2B monoubiquitination and cell size control. Mol. Cell 11, 261–266. Kaliraman, V., Mullen, J.R., Fricke, W.M., Bastin-Shanower, S.A., and Brill, S.J. (2001). Functional overlap between Sgs1-Top3 and the Mms4-Mus81 endonuclease. Genes Dev. 15, 2730–2740.

Sarangi, P., and Zhao, X. (2015). SUMO-mediated regulation of DNA damage repair and responses. Trends Biochem. Sci. 40, 233–242. Saugar, I., Va´zquez, M.V., Gallo-Ferna´ndez, M., Ortiz-Baza´n, M.A., Segurado, M., Calzada, A., and Tercero, J.A. (2013). Temporal regulation of the Mus81Mms4 endonuclease ensures cell survival under conditions of DNA damage. Nucleic Acids Res. 41, 8943–8958. Schwartz, E.K., and Heyer, W.D. (2011). Processing of joint molecule intermediates by structure-selective endonucleases during homologous recombination in eukaryotes. Chromosoma 120, 109–127. Sebesta, M., Urulangodi, M., Stefanovie, B., Szakal, B., Pacesa, M., Lisby, M., Branzei, D., and Krejci, L. (2017). Esc2 promotes Mus81 complex-activity via its SUMO-like and DNA binding domains. Nucleic Acids Res. 45, 215–230. Sun, Z.W., and Allis, C.D. (2002). Ubiquitination of histone H2B regulates H3 methylation and gene silencing in yeast. Nature 418, 104–108.

Liberi, G., Maffioletti, G., Lucca, C., Chiolo, I., Baryshnikova, A., Cotta-Ramusino, C., Lopes, M., Pellicioli, A., Haber, J.E., and Foiani, M. (2005). Rad51dependent DNA structures accumulate at damaged replication forks in sgs1 mutants defective in the yeast ortholog of BLM RecQ helicase. Genes Dev. 19, 339–350.

Svendsen, J.M., Smogorzewska, A., Sowa, M.E., O’Connell, B.C., Gygi, S.P., Elledge, S.J., and Harper, J.W. (2009). Mammalian BTBD12/SLX4 assembles a Holliday junction resolvase and is required for DNA repair. Cell 138, 63–77.

Lisby, M., Barlow, J.H., Burgess, R.C., and Rothstein, R. (2004). Choreography of the DNA damage response: spatiotemporal relationships among checkpoint and repair proteins. Cell 118, 699–713.

Szakal, B., and Branzei, D. (2013). Premature Cdk1/Cdc5/Mus81 pathway activation induces aberrant replication and deleterious crossover. EMBO J. 32, 1155–1167.

Matos, J., Blanco, M.G., Maslen, S., Skehel, J.M., and West, S.C. (2011). Regulatory control of the resolution of DNA recombination intermediates during meiosis and mitosis. Cell 147, 158–172. Minocherhomji, S., and Hickson, I.D. (2014). Structure-specific endonucleases: guardians of fragile site stability. Trends Cell Biol. 24, 321–327. Mun˜oz, I.M., Hain, K., De´clais, A.C., Gardiner, M., Toh, G.W., Sanchez-Pulido, L., Heuckmann, J.M., Toth, R., Macartney, T., Eppink, B., et al. (2009). Coordination of structure-specific nucleases by human SLX4/BTBD12 is required for DNA repair. Mol. Cell 35, 116–127. Naim, V., Wilhelm, T., Debatisse, M., and Rosselli, F. (2013). ERCC1 and MUS81-EME1 promote sister chromatid separation by processing late replication intermediates at common fragile sites during mitosis. Nat. Cell Biol. 15, 1008–1015. Princz, L.N., Wild, P., Bittmann, J., Aguado, F.J., Blanco, M.G., Matos, J., and Pfander, B. (2017). Dbf4-dependent kinase and the Rtt107 scaffold promote Mus81-Mms4 resolvase activation during mitosis. EMBO J. 36, 664–678. Rass, U. (2013). Resolving branched DNA intermediates with structure-specific nucleases during replication in eukaryotes. Chromosoma 122, 499–515.

1562 Cell Reports 20, 1553–1562, August 15, 2017

Tkach, J.M., Yimit, A., Lee, A.Y., Riffle, M., Costanzo, M., Jaschob, D., Hendry, J.A., Ou, J., Moffat, J., Boone, C., et al. (2012). Dissecting DNA damage response pathways by analysing protein localization and abundance changes during DNA replication stress. Nat. Cell Biol. 14, 966–976. West, S.C., Blanco, M.G., Chan, Y.W., Matos, J., Sarbajna, S., and Wyatt, H.D. (2015). Resolution of recombination intermediates: mechanisms and regulation. Cold Spring Harb. Symp. Quant. Biol. 80, 103–109. Wilson, J.S., Tejera, A.M., Castor, D., Toth, R., Blasco, M.A., and Rouse, J. (2013). Localization-dependent and -independent roles of SLX4 in regulating telomeres. Cell Rep. 4, 853–860. Wood, A., Krogan, N.J., Dover, J., Schneider, J., Heidt, J., Boateng, M.A., Dean, K., Golshani, A., Zhang, Y., Greenblatt, J.F., et al. (2003). Bre1, an E3 ubiquitin ligase required for recruitment and substrate selection of Rad6 at a promoter. Mol. Cell 11, 267–274. Ying, S., Minocherhomji, S., Chan, K.L., Palmai-Pallag, T., Chu, W.K., Wass, T., Mankouri, H.W., Liu, Y., and Hickson, I.D. (2013). MUS81 promotes common fragile site expression. Nat. Cell Biol. 15, 1001–1007.

Cell Reports, Volume 20

Supplemental Information

Subnuclear Relocalization of Structure-Specific Endonucleases in Response to DNA Damage Irene Saugar, Alberto Jiménez-Martín, and José Antonio Tercero

Figure S1. Saugar et al. A no MMS

0.005% MMS

0.075% MMS

D GFP-RAD1

W303 GFP-MUS81 GFP-MMS4 GFP-RAD1 GFP-SLX1 GFP-SLX4 mus81Δ

mms4Δ

rad1Δ

slx1Δ

slx4Δ

120 105 90 75 60 45 30 15 G1 (αF) Log. 1C 2C

B

1C 2C

GFP-Rad1 Loading/transfer control (Ponceau S)

G1 release + MMS (min) (αF) 15 30 45 60 75 90 105120

GFP-Rad1 Loading/transfer control (Ponceau S)

E GFP-MUS81

G1 release + MMS (min) (αF) 15 30 45 60 75 90 105120

GFP-Mus81 Loading/transfer control (Ponceau S)

1C 2C

1C 2C

G1 release no MMS (min) (αF) 15 30 45 60 75 90 105 120

Release -/+ MMS (min)

1C 2C

GFP-Mms4-P GFP-Mms4

1C 2C

Loading/transfer control (Ponceau S)

G1 release + MMS (min) (αF) 15 30 45 60 75 90 105 120

GFP-Slx1 Loading/transfer control (Ponceau S)

G1 release no MMS (min) (αF) 15 30 45 60 75 90 105 120

+ MMS 120 105 90 75 60 45 30 15 G1 (αF) Log.

Loading/transfer control (Ponceau S)

G1 release + MMS (min) (αF) 15 30 45 60 75 90 105 120

GFP-Mms4 Loading/transfer control (Ponceau S)

GFP-Slx1

GFP-SLX4

no MMS

Release -/+ MMS (min)

F

120 105 90 75 60 45 30 15 G1 (αF) Log. 1C 2C

120 105 90 75 60 45 30 15 G1 (αF) Log.

Loading/transfer control (Ponceau S)

+ MMS

G1 release no MMS (min) (αF) 15 30 45 60 75 90 105 120

+ MMS

GFP-Mus81

GFP-MMS4

no MMS

no MMS

Release -/+ MMS (min)

+ MMS 120 105 90 75 60 45 30 15 G1 (αF) Log.

1C 2C

GFP-SLX1

G1 release no MMS (min) (αF) 15 30 45 60 75 90 105 120 Release -/+ MMS (min)

no MMS

C

G1 release no MMS (min) (αF) 15 30 45 60 75 90 105120

+ MMS Release -/+ MMS (min)

no MMS

1C 2C

1C 2C

GFP-Slx4 Loading/transfer control (Ponceau S)

G1 release + MMS (min) (αF) 15 30 45 60 75 90 105 120

GFP-Slx4 Loading/transfer control (Ponceau S)

Figure S1 (related to Figures 1 and 2). (A) The strains used in this work to study the different proteins by fluorescence microscopy behave like the parental control in terms of growth and DNA damage sensitivity. The figure shows a drop assay to compare the sensitivity to MMS of the constructed GFP-tagged strains with that of the wild-type parental control and null mutants of the genes encoding the proteins of interest. For the assay, serial dilutions (10-fold) of normalized cycling cell cultures were spotted onto YPD plates containing the indicated amounts of MMS and incubated at 30ºC for 48–72 h. Strains: W303-1a (parental wild-type control), GFP-MUS81 (YSG39), GFP-MMS4 (YSG42), GFP-RAD1 (YSG271), GFP-SLX1 (YSG356), GFP-SLX4 (YSG235), mus81Δ (YMV39), mms4Δ (YMV48), rad1Δ (YSG360), slx1Δ (YSG352) and slx4Δ (YSG80). (B-F) The protein levels of GFP-Mus81, GFP-Mms4, GFP-Rad1, GFP-Slx1 or GFP-Slx4 do not change significantly after treatment with MMS. Exponentially growing cultures of GFP-MUS81 (YSG39 strain) (B), GFP-MMS4 (YSG42 strain) (C), GFP-RAD1 (YSG271 strain) (D), GFP-SLX1 (YSG356 strain) (E) and GFP-SLX4 (YSG235 strain) (F) cells were synchronized in G1 phase using α factor and then released in fresh medium in the absence or in the presence of 0.033% MMS. The cells were collected at the indicated time points and the DNA content was determined by flow cytometry to follow cell cycle progression (left panels). Protein levels were analyzed in each case by immunoblot (right panels). A Ponceau S stained membrane was used as a loading and transfer control. Note that, as previously shown (Gallo-Fernandez et al., 2012; Matos et al., 2011; Saugar et al., 2013; Szakal and Branzei, 2013), Mms4 was phosphorylated around G2/M (C).

Figure S2. Saugar et al. A

B GFP-MMS4

GFP-MUS81

S phase no MMS + MMS

S phase no MMS + MMS

GFP

GFP

G2/M (noc.)

G2/M (noc.) +MMS

G1(αF) + MMS

G1(αF)

G2/M (noc.)

G2/M (noc.) +MMS

GFP

GFP

bright field

G1(αF) + MMS

bright field

G1(αF)

release

bright field

G1(αF)

bright field

G1(αF)

release

C G1 (αF) à S phase (+ nocod) (-/+ MMS) no MMS +MMS

% cells with foci

100 80 60 40 20 0

-

+

GFP-MUS81

-

+

(MMS)

GFP-MMS4

Figure S2 (related to Figure 1). Mus81 and Mms4 accumulate and form DNA damage-induced foci in G1 and S phases, but not in G2/M phase. The formation of Mus81- (YSG39 strain) (A) and Mms4- (YSG42 strain) (B) foci was examined at different stages of the cell cycle by fluorescence microscopy. In each case, to study foci formation in G1 (lower panel, left), cells were synchronized with α factor and then held for an additional 60 min in G1 in medium containing 0.033% MMS. To analyze foci formation in S phase (upper panel), cells were synchronized in G1 using α factor and then released in fresh medium without or with 0.033% MMS (for 30 and 60 min, respectively). To study foci formation in G2/M (lower panel, right), cells were synchronized with nocodazole and then held in G2/M for an additional 60 min in the presence of 0.033% MMS. Arrows indicate selected foci. (C) Nocodazole treatment does not affect Mus81 or Mms4 foci formation. GFP-MUS81 (YSG39 strain) and GFP-MMS4 (YSG42 strain) cells were synchronized in G1 with α factor and then released into S phase in fresh medium containing nocodazole for 60 min, either in the absence or in the presence of 0.033% MMS. Cells were examined by fluorescence microscopy. The percentage of cells showing Mus81 or Mms4 foci is shown in each case. The histograms represent the mean ± SD from three different experiments.

Figure S3. Saugar et al.

A

GFP-MUS81

no MMS + MMS

80 60 40 20 0

B

hos2Δ

100

% cells with foci

% cells with Mus81 foci

100

- + wt

- +

- +

- + (MMS)

H2A- rtt107Δ

rtt107Δ

rad9Δ

S129A

no MMS + MMS

80 60 40 20 0

- +

- +

GFP- GFPMUS81

RAD1

- +

- + (MMS)

GFPSLX1

GFPSLX4

Figure S3 (related to Figure 3). (A) Endonuclease foci formation is independent of H2A phosphorylation. Percentage of cells containing Mus81 foci in different mutants before and after treatment of exponentially growing cultures with 0.033% MMS for 60 min. Strains: GFP-MUS81, ‘wt’ (YSG39), GFP-MUS81 H2AS129A (YSG399), GFP-MUS81 rtt107Δ (YSG245) and GFP-MUS81 rtt107Δ rad9Δ (YSG398). (B) Endonuclease foci formation does not depend on the histone deacetylase Hos2. The figure shows the percentage of hos2Δ cells containing foci of every protein before and after treatment of exponentially growing cultures with 0.033% MMS for 60 min. Strains: GFP-MUS81 hos2Δ (YAJ351), GFP-RAD1 hos2Δ (YAJ352), GFP-SLX1 hos2Δ (YAJ353) and GFP-SLX4 hos2Δ (YAJ354). In all cases, the histograms represent the mean ± SD from three different experiments.

Figure S4. Saugar et al. A

B

GFP-MUS81 HA-MMS4 BTN2+ RAD53+ (control cells)

no MMS

release no MMS (min) G1 15 30 45 60 75 90 105 120

+ MMS

GFP-MUS81 HA-MMS4 btn2Δ

no MMS

120’ HA-Mms4-P HA-Mms4 105’ Loading/transfer 90’ control 75’ (Ponceau S) 60’ Release + MMS (min) 45’ G1 15 30 45 60 75 90 105120 30’ 15’ HA-Mms4 G1(αF) Loading/transfer log control (Ponceau S)

1C 2C

release -/+ MMS

HA-Mms4-P 120’ HA-Mms4 105’ Loading/transfer 90’ control 75’ (Ponceau S) 60’ Release + MMS (min) 45’ G1 15 30 45 60 75 90 105120 30’ 15’ HA-Mms4 G1(αF) Loading/transfer log

release -/+ MMS

1C 2C

1C 2C

C

release no MMS (min) G1 15 30 45 60 75 90 105 120

+ MMS

control (Ponceau S)

1C 2C

GFP-MUS81 HA-MMS4 btn2Δ rad53Δ sml1Δ no MMS

release no MMS (min) G1 15 30 45 60 75 90 105 120

+ MMS release -/+ MMS

120’ 105’ 90’ 75’ 60’ 45’ 30’ 15’ G1(αF) log 1C 2C

1C 2C

HA-Mms4-P HA-Mms4 Loading/transfer control (Ponceau S)

Release + MMS (min) G1 15 30 45 60 75 90 105120 HA-Mms4-P HA-Mms4 Loading/transfer control (Ponceau S)

D GFP-MUS81 GFP-MUS81 HA-MMS4 HA-MMS4 sgs1Δctf4Δ

sgs1Δ

90’ 45’

+MMS 60’ G1(αF)

% cells with Mus81 foci

120’

80

sgs1Δ

G1 MMS

release no MMS (min) (αF) 60’ 15 30 45 60 75 90 105 120 135 150

HA-Mms4-P HA-Mms4 Loading/transfer control (Ponceau S)

60 40

sgs1Δctf4Δ

G1 MMS

20

HA-Mms4-P HA-Mms4

0

log 1C 2C

HA-MMS4 sgs1Δ

HA-MMS4 sgs1Δ ctf4Δ



100 release no MMS

150’

GFP-MUS81 HA-MMS4

1C 2C

release no MMS (min) (αF) 60’ 15 30 45 60 75 90 105 120 135 150

+ MMS 60’

45

90 120 150 release no MMS (min)

Loading/transfer control (Ponceau S)

E GFP-MUS81 GFP-MUS81 HA-MMS4 HA-MMS4 sgs1Δrad51Δ

sgs1Δ

90’ 45’

+MMS 60’ G1(αF) log 1C 2C

1C 2C

% cells with Mus81 foci

120’

release no MMS

150’

GFP-MUS81 HA-MMS4 HA-MMS4 sgs1Δ

HA-MMS4 sgs1Δ rad51Δ



100 80

sgs1Δ

HA-Mms4-P HA-Mms4

release no MMS (min) G1 MMS (αF) 60’ 15 30 45 60 75 90 105 120 135 150

60

Loading/transfer control (Ponceau S)

40

sgs1Δrad51Δ

20 0

release no MMS (min) G1 MMS (αF) 60’ 15 30 45 60 75 90 105 120 135 150

HA-Mms4 + MMS 60’

45

90 120 150 release no MMS (min)

Loading/transfer control (Ponceau S)

Figure S4 (related to Figure 4). (A–C) Recruitment of Mus81-Mms4 into foci is not required to prevent Mms4 phosphorylation and is not necessary for the modification of this subunit. Exponentially growing cultures of control GFP-MUS81 HAMMS4 BTN2+ RAD53+ (YSG45 strain) (A), GFP-MUS81 HA-MMS4 btn2Δ (YSG334 strain) (B) and GFPMUS81 HA-MMS4 sml1Δ rad53Δ btn2Δ (YSG348 strain) (C) cells were synchronized in G1 phase using α factor and then released in fresh medium in the absence or in the presence of 0.033% MMS. The cells were collected at the indicated time points and the DNA content was determined by flow cytometry to follow cell cycle progression (left panels). Protein levels were analyzed in each case by immunoblot (right panels). (D-E) Mus81 foci disappear (after removal of the DNA-damaging agent) when there is no accumulation of DNA recombination intermediates, even in the absence of Mus81-Mms4 activation. Cycling cultures of GFPMUS81 HA-MMS4 sgs1Δ (YSG133 strain) and GFP-MUS81 HA-MMS4 sgs1Δctf4Δ (YSG426 strain) cells (D), as well as cycling cultures of GFP-MUS81 HA-MMS4 sgs1Δ (YSG133 strain) and GFP-MUS81 HAMMS4 sgs1Δrad51Δ (YSG417 strain) cells (E), were synchronized in G1 with α factor and released into Sphase in medium containing 0.033% MMS for 60 min. The MMS was then removed to allow cell cycle progression, which was monitored by flow cytometry (left panels). Central panels indicate the percentage of cells containing Mus81 foci at every time point. Right panels show the immunoblot analysis of Mms4 during the course of the experiment.

Table S1. Saugar et al.

Strain

Relevant genotype

Source

YSG39

MATa PADH1-yeGFP-MUS81::natNT2

This work

YSG41

MATa PADH1-yeGFP-MUS81::natNT2 mms4Δ::URA3

This work

YSG42

MATa PADH1-yeGFP-MMS4::natNT2

This work

YSG45

MATa PADH1-yeGFP-MUS81::natNT2 PADH1-3HA-MMS4::natNT2

This work

YSG48

MATa PADH1-yeGFP-MUS81::natNT2 sml1Δ::URA3 rad53Δ::LEU2

This work

YSG49

MATa PADH1-yeGFP-MUS81::natNT2 rad5Δ::URA3

This work

YSG50

MATa PADH1-yeGFP-MUS81::natNT2 sml1Δ::URA3 mec1Δ::kanMX

This work

YSG51

MATa PADH1-yeGFP-MUS81::natNT2 rad52Δ::URA3

This work

YSG53

MATa PADH1-yeGFP-MUS81::natNT2 chk1Δ::HIS3

This work

YSG54

MATa PADH1-yeGFP-MUS81::natNT2 rad9Δ::HIS3

This work

YSG55

MATa PADH1-yeGFP-MUS81::natNT2 sml1Δ::URA3 rad53Δ::LEU2 chk1Δ::HIS3

This work

YSG61

MATa PADH1-yeGFP-MUS81::natNT2 rad54Δ::URA3

This work

YSG67

MATa PADH1-yeGFP-MUS81::natNT2 sgs1Δ::kanMX

This work

YSG74

MATa PADH1-yeGFP-MUS81::natNT2 rad17Δ::URA3

This work

YSG80

MATa slx4Δ::TRP1

This work

YSG81

MATa PADH1-yeGFP-MUS81::natNT2 slx4Δ::TRP1

This work

YSG95

MATa PADH1-yeGFP-MUS81::natNT2 PADH1-3HA-MMS4::natNT2 sgs1Δ::kanMX ura3-1::PGAL1-10-CDC5::URA3

This work

YSG96

MATa PADH1-yeGFP-MUS81::natNT2 rad18Δ::kanMX

This work

YSG99

MATa PADH1-yeGFP-MUS81::natNT2 PGAL1-10-3HA-MMS4::kanMX sgs1Δ::kanMX

This work

YSG112

MATa PADH1-yeGFP-MUS81::natNT2 tel1Δ::HIS3

This work

YSG114

MATa PADH1-yeGFP-MUS81::natNT2 tel1Δ::HIS3 sml1Δ::URA3 mec1Δ::kanMX

This work

YSG123

MATa PADH1-yeGFP-MUS81::natNT2 rad14Δ::kanMX

This work

YSG124

MATa PADH1-yeGFP-MUS81::natNT2 mag1Δ::kanMX

This work

YSG133

MATa PADH1-yeGFP-MUS81::natNT2 PADH1-3HA-MMS4::natNT2 sgs1Δ::kanMX

This work

YSG188

MATa PADH1-yeGFP-MUS81::natNT2 mrc1Δ::HIS3

This work

YSG192

MATa PADH1-yeGFP-MMS4::natNT2 mus81Δ::kanMX

This work

YSG215

MATa PADH1-yeGFP-MUS81::natNT2 rad6Δ::URA3

This work

YSG216

MATa PADH1-yeGFP-MUS81::natNT2 rad26Δ::URA3

This work

YSG217

MATa PADH1-yeGFP-MUS81::natNT2 rad26Δ::URA3 rad14Δ::kanMX mag1Δ::hphNT1

This work

YSG218

MATa PADH1-yeGFP-MUS81::natNT2 rad30Δ::URA3 rev3Δ::kanMX rev1Δ::LEU2

This work

YSG232

MATa PADH1-yeGFP-MUS81::natNT2 ubr1Δ::HIS3

This work

YSG235

MATa PADH1-yeGFP-SLX4::natNT2

This work

YSG245

MATa PADH1-yeGFP-MUS81::natNT2 rtt107Δ::hphNT1

This work

YSG246

MATa PADH1-yeGFP-SLX4:natNT2 rad6Δ::URA3

This work

YSG264

MATa PADH1-yeGFP-SLX4::natNT2 bre1Δ::hphNT1

This work

YSG271

MATa PADH1-yeGFP-RAD1::natNT2

This work

YSG274

MATa PADH1-yeGFP-MUS81::natNT2 kanMX::PADH1-SLX4mCherry::hphNT1

This work

YSG276

MATa PADH1-yeGFP-MUS81::natNT2 NOP1-mCherry::hphNT1

This work

YSG278

MATa PADH1-yeGFP-MUS81::natNT2 NUP84-mCherry::hphNT1

This work

YSG280

MATa PADH1-yeGFP-MUS81::natNT2 kanMX::PADH1-RAD1mCherry::hphNT1

This work

YSG283

MATa kanMX::PADH1-MUS81-mCherry::hphNT1 RAD52-yeGFP::HisMX

This work

YSG287

MATa PADH1-yeGFP-RAD1::natNT2 rad6Δ::URA3

This work

YSG299

MATa PADH1-yeGFP-MUS81::natNT2 rad1Δ::URA3

This work

YSG302

MATa PADH1-yeGFP-SLX4::natNT2 rad1Δ::URA3

This work

YSG304

MATa PADH1-yeGFP-SLX4::natNT2 mus81Δ::kanMX

This work

YSG305

MATa PADH1-yeGFP-RAD1::natNT2 slx4Δ::HISMX

This work

YSG307

MATa PADH1-yeGFP-RAD1::natNT2 mus81Δ::kanMX

This work

YSG308

MATa PADH1-yeGFP-MUS81::natNT2 SPC42-eqFP::HIS3

This work

YSG309

MATa kanMX::PADH1-MUS81-mCherry::hphNT1 RFA1-GFP::TRP1

This work

YSG310

MATa PADH1-yeGFP-MUS81::natNT2 btn2Δ::HISMX

This work

YSG312

MATa PADH1-yeGFP-MUS81::natNT2 set1Δ::HISMX

This work

YSG315

MATa PADH1-yeGFP-RAD1::natNT2 bre1Δ::hphNT1

This work

YSG320

MATa PADH1-yeGFP-SLX4::natNT2 btn2Δ::HISMX

This work

YSG322

MATa PADH1-yeGFP-RAD1::natNT2 btn2Δ::HISMX

This work

YSG324

MATa PADH1-yeGFP-MUS81::natNT2 CMR1-mCherry::hphNT1

This work

YSG326

MATa PADH1-yeGFP-MUS81::natNT2 RAP1-mCherry::hphNT1

This work

YSG334

MATa PADH1-yeGFP-MUS81::natNT2 PADH1-3HA-MMS4::natNT2 btn2Δ::HISMX

This work

YSG338

MATa PADH1-yeGFP-MUS81::natNT2 cmr1Δ::HISMX

This work

YSG340

MATa PADH1-yeGFP-SLX4::natNT2 cmr1Δ::HISMX

This work

YSG341

MATa PADH1-yeGFP-RAD1::natNT2 cmr1Δ::HISMX

This work

YSG342

MATa CMR1-yeGFP::HISMX

This work

YSG344

MATa CMR1-yeGFP::HISMX rad6Δ::URA3

This work

YSG345

MATa CMR1-yeGFP::HISMX bre1Δ::hphNT1

This work

YSG348

MATa PADH1-yeGFP-MUS81::natNT2 PADH1-3HA-MMS4::natNT2 rad53Δ::LEU2 sml1Δ::URA3 btn2Δ::HISMX

This work

YSG352

MATa slx1Δ::HISMX

This work

YSG353

MATa PADH1-yeGFP-MUS81::natNT2 slx1Δ::HISMX

This work

YSG356

MATa PADH1-yeGFP-SLX1::natNT2

This work

YSG360

MATa rad1Δ::URA3

This work

YSG363

MATa PADH1-yeGFP-MUS81::natNT2 kanMX::PADH1-SLX1mCherry::hphNT1

This work

YSG368

MATa PADH1-yeGFP-SLX1::natNT2 rad6Δ::URA3

This work

YSG369

MATa PADH1-yeGFP-SLX1::natNT2 btn2Δ::HISMX

This work

YSG370

MATa PADH1-yeGFP-SLX1::natNT2 bre1Δ::hphNT1

This work

YSG372

MATa PADH1-yeGFP-SLX1::natNT2 mus81Δ::kanMX

This work

YSG373

MATa PADH1-yeGFP-SLX1::natNT2 rad1Δ::URA3

This work

YSG374

MATa PADH1-yeGFP-SLX1::natNT2 slx4Δ::HIS3

This work

YSG375

MATa PADH1-yeGFP-SLX1::natNT2 cmr1Δ::HISMX

This work

YSG376

MATa PADH1-yeGFP-RAD1::natNT2 slx1Δ::HISMX

This work

YSG377

MATa PADH1-yeGFP-SLX4::natNT2 slx1Δ::HISMX

This work

YSG398

MATa PADH1-yeGFP-MUS81::natNT2 rad9Δ::HIS3 rtt107Δ::hphNT1

This work

YSG399

MATa PADH1-yeGFP-MUS81::natNT2 hta1-htb1::LEU2 hta2htb2::TRP1(pJD151 [hta1-S129A-HTB1])

This wotk

YSG400

MATa PADH1-yeGFP-MUS81::natNT2 rad30Δ::URA3 rev3Δ::kanMX rev1Δ::LEU2 rad5Δ::HIS3

This work

YSG413

MATa mus81Δ::HIS3::YIp128::kanMX::PADH1-MUS81yeGFP::hphNT1::LEU2

This work

YSG414

MATa mus81Δ::HIS3::YIp128::kanMX::PADH1-mus81(D414A, D415A)yeGFP::hphNT1::LEU2

This work

YSG417

MATa PADH1-yeGFP-MUS81::natNT2 PADH1-3HA-MMS4::natNT2 rad51Δ::hphNT1 sgs1Δ::kanMX

This work

YSG426

MATa PADH1-yeGFP-MUS81::natNT2 PADH1-3HA-MMS4::natNT2 ctf4Δ::URA3 sgs1Δ::kanMX

This work

YAJ351

MATa PADH1-yeGFP-MUS81::natNT2 hos2Δ::HISMX

This work

YAJ352

MATa PADH1-yeGFP-RAD1::natNT2 hos2Δ::HISMX

This work

YAJ353

MATa PADH1-yeGFP-SLX1::natNT2 hos2Δ::HISMX

This work

YAJ354

MATa PADH1-yeGFP-SLX4::natNT2 hos2Δ::HISMX

This work

YJT166

MATa PADH1-yeGFP-MUS81::natNT2 rad14Δ::kanMX mag1Δ::hphNT1

This work

YJT167

MATa PADH1-yeGFP-MUS81::natNT2 bre1Δ::hphNT1

This work

YJT168

MATa PADH1-yeGFP-MUS81::natNT2 dot1Δ::hphNT1

This work

YMV39

MATa mus81Δ::kanMX

YMV48

MATa mms4Δ::URA3

Saugar et al. (2013) Saugar et al. (2013)

Table S1 (related to all main figures and all supplemental figures). S. cerevisiae strains used in this study. All strains were based on W303 (ade2-1 ura3-1 his3-11,15 trp1-1 leu2-3,112 can1-100), except YSG413 and YSG414, which derived from BY4741 (his3Δ1 leu2Δ0 met15Δ0 ura3Δ0 lys2∆0). The strains were constructed by transformation with PCR-products or by crosses and subsequent tetrad dissection, using standard techniques. All strains were verified by PCR, drug sensitivity assays and immunoblot when appropriate. For epitope-tagging and gene deletions, the pML (Longtine et al., 1998), pYM (Janke et al., 2004) plasmid series were usually used as templates for PCR.

SUPPLEMENTAL REFERENCES Janke, C., Magiera, M.M., Rathfelder, N., Taxis, C., Reber, S., Maekawa, H., Moreno-Borchart, A., Doenges, G., Schwob, E., Schiebel, E., et al. (2004). A versatile toolbox for PCR-based tagging of yeast genes: new fluorescent proteins, more markers and promoter substitution cassettes. Yeast 21, 947-962. Longtine, M.S., McKenzie, A., 3rd, Demarini, D.J., Shah, N.G., Wach, A., Brachat, A., Philippsen, P., and Pringle, J.R. (1998). Additional modules for versatile and economical PCR-based gene deletion and modification in Saccharomyces cerevisiae. Yeast 14, 953-961.