PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of scirepAboutEditorial BoardFor AuthorsScientific Reports
 
Sci Rep. 2017; 7: 11650.
Published online 2017 September 14. doi:  10.1038/s41598-017-11937-z
PMCID: PMC5599684

A cell cycle-independent mode of the Rad9-Dpb11 interaction is induced by DNA damage

Abstract

Budding yeast Rad9, like its orthologs, controls two aspects of the cellular response to DNA double strand breaks (DSBs) – signalling of the DNA damage checkpoint and DNA end resection. Rad9 binds to damaged chromatin via modified nucleosomes independently of the cell cycle phase. Additionally, Rad9 engages in a cell cycle-regulated interaction with Dpb11 and the 9-1-1 clamp, generating a second pathway that recruits Rad9 to DNA damage sites. Binding to Dpb11 depends on specific S/TP phosphorylation sites of Rad9, which are modified by cyclin-dependent kinase (CDK). Here, we show that these sites additionally become phosphorylated upon DNA damage. We define the requirements for DNA damage-induced S/TP phosphorylation of Rad9 and show that it is independent of the cell cycle or CDK activity but requires prior recruitment of Rad9 to damaged chromatin, indicating that it is catalysed by a chromatin-bound kinase. The checkpoint kinases Mec1 and Tel1 are required for Rad9 S/TP phosphorylation, but their influence is likely indirect and involves phosphorylation of Rad9 at S/TQ sites. Notably, DNA damage-induced S/TP phosphorylation triggers Dpb11 binding to Rad9, but the DNA damage-induced Rad9-Dpb11 interaction is dispensable for recruitment to DNA damage sites, indicating that the Rad9-Dpb11 interaction functions beyond Rad9 recruitment.

Introduction

DNA damage (such as double strand breaks (DSBs)) elicits cellular signalling pathways, collectively known as the DNA damage response (reviewed in ref. 1). Among these, checkpoint mechanisms control cell cycle progression as well as transcriptional and post-translational regulation of DNA repair and replication. Furthermore, local signalling events are critical in directing DNA repair pathway choice. Budding yeast Rad9 was the first checkpoint protein to be discovered2. Since then, it has become evident that Rad9, as well as its orthologs such as fission yeast Crb23, 4 and human 53BP1 (reviewed in ref. 5), play a crucial role in the DNA damage response, having at least two functions: signal transduction in the DNA damage checkpoint (reviewed in ref. 1) and control of DNA end resection, a local process that critically determines DSB repair pathway choice (reviewed in ref. 6).

As checkpoint signalling mediator, Rad9 links the signal transduction from the apical kinase Mec1 to the effector kinase Rad53712. As such, it is essential for activation of Rad53 and therefore for the activation of a global checkpoint response upon DNA damage. Moreover, Rad9 is also an inhibitor of DNA end resection1316. Since DNA end resection generates the DNA substrate for recombination-based repair and interferes with ligation-based repair, Rad9 is a critical regulator of DSB repair pathway choice. To fulfil these two functions, Rad9 engages in several protein-protein interactions that occur within damaged chromatin1722.

Rad9 binds to modified histones via two distinct domains. The TUDOR domain of Rad9 interacts with histone H3 in its K79-methylated form19, 22, a widespread modification of chromatin that is introduced by the methyltransferase Dot123, 24. The tandem-BRCT domain of Rad9 interacts with histone H2A in its S129-phosphorylated form (γH2A21, 25), a DNA damage-specific chromatin mark introduced by the apical checkpoint kinases Mec1 and Tel126. As such, Rad9 is a bivalent nucleosome binder, a feature that is conserved among Rad9 orthologs, even though different histone marks are being recognized2731.

Rad9 also binds to the scaffold protein Dpb1117, 18. Dpb11 contains two pairs of BRCT domains, which provide two phospho-protein binding surfaces (reviewed in ref. 32). While Rad9 binds to BRCT1 + 2, Dpb11 also interacts with the 9-1-1 complex via BRCT3 + 417, 33, 34. Physical and genetic interaction data suggest that these interactions generate a second pathway that recruits Rad9 to DNA damage sites: DNA damage-loaded 9-1-1 can tether Dpb11, which in turn can recruit Rad917, 33. Notably, the interaction of Dpb11 with Rad9 depends on Rad9 phosphorylation at S462 and T474 residues17. Both sites match the minimal consensus (S/TP) for phosphorylation by cyclin-dependent kinase (Cdc28, in the following referred to as CDK) and consistently a CDK-dependent interaction between Rad9 and Dpb11 can be observed in G2/M-arrested cells17.

Furthermore, Rad9 binds to the checkpoint effector kinase Rad537, 8, 10, 12. This interaction involves phosphorylation of Rad9 in the S/TQ cluster domain (SCD), which is specifically bound by the FHA domains of Rad53. Rad9 is phosphorylated in the SCD by the apical kinases Mec1 and Tel1 upon association with damaged chromatin7, 12. Current models suggest that Rad53 is transiently recruited to damaged chromatin by this mechanism (reviewed in ref. 1). Here, it becomes activated by Mec1/Tel1 phosphorylation, before it dissociates from the DNA damage site to set off the global DNA damage response.

Promoting Rad53 phosphorylation and activation offers a straightforward mechanism of how Rad9 mediates checkpoint signalling. In contrast, it is less clear by which mechanism Rad9 regulates DNA end resection13, 14, even though an antagonistic relationship between Rad9 and the resection-promoting nucleosome remodeller Fun30 has been demonstrated35, 36.

Rad9 recruitment to damaged chromatin occurs in all cell cycle phases19. However, individual Rad9 recruitment mechanisms are apparently under cell cycle control17, 33. Previous data has therefore led to a model where in G1 only one Rad9 recruitment pathway (via interaction with modified nucleosomes, referred to as the ‘histone pathway’1922, 25) is active, while outside of G1 a second Rad9 recruitment pathway (via Dpb11 and 9-1-1, referred to as the ‘Dpb11 pathway’) is additionally available17, 33. However, the underlying reason for restricting the Rad9-Dpb11 interaction to specific cell cycle phases is not understood.

Here, we report new aspects in the regulation of Rad9 in the response to DSBs. We find that the Rad9 S/TP sites, which facilitate Dpb11-binding, are also phosphorylated upon DNA damage independently of the cell cycle phase. DNA damage-dependent phosphorylation of these sites can be detected even in G1 cells or upon inhibition of CDK. Notably, these phosphorylation events depend on prior chromatin-recruitment of Rad9 via the ‘histone pathway’ and on the integrity of the SCD domain of Rad9. Furthermore, the Rad9 phosphorylation facilitates the interaction between Rad9 and Dpb11, similarly to our previous results on the CDK-dependent mode of interaction. These findings suggest that Dpb11 and Rad9 can interact even in G1, where Dpb11 is not involved in recruiting Rad9 to damaged chromatin.

Results

DNA damage induces phosphorylation of Rad9 S/TP sites and binding of Rad9 to Dpb11

Orthologs of Rad9 and Dpb11 were found to interact in different organisms17, 18, 29, 37. In case of budding yeast, our previous work has shown that Rad9 specifically interacts with Dpb11 in cells arrested in M phase, but not in cells arrested in G117. The cell cycle-regulation of the interaction is achieved by CDK-dependent phosphorylation of two S/TP motifs on Rad9 (S462 and T474, referred to as Rad9 S/TP sites hereafter), which are recognized by the BRCT1 + 2 domain of Dpb1117.

We observed that Rad99myc from cell extracts of cells containing MMS-induced DNA damage showed increased interaction with GSTDpb11 in pulldown experiments (Fig. S1A). Strikingly, even when we used cells arrested in G1, we found that DNA damage treatment with the DSB-inducing agent phleomycin resulted in an increased interaction of Rad99myc with GSTDpb11 (Figs 1A and S1B). Phleomycin treatment causes Rad9 to undergo a phospho-shift (Fig. 1A)8, 1012. Notably, we found Dpb11 to associate with this hyperphosphorylated form of Rad9 (Fig. 1A). In contrast, in M phase cell extracts Rad99myc was able to interact with GSTDpb11 even in the absence of DNA damage treatment (Fig. 1A), consistent with our previous result on the CDK regulation of Rad917.

Figure 1
A CDK-independent, DNA damage-dependent mode of Rad9-S462 and -T474 phosphorylation and interaction with Dpb11. (A) DNA damage stimulates the Rad9-Dpb11 interaction in cell extracts. GST pulldown experiment with GSTDpb11-N (contains BRCT1 + 2, ...

The interaction between Rad9 and Dpb11 critically depends on phosphorylation of S462 and T474 on Rad917. We therefore tested, whether phosphorylation of these sites is also induced by DNA damage. To this end, we used our previously generated phosphorylation-specific antibodies directed against Rad9-epitopes containing either phosphorylated S462 or phosphorylated T474, respectively17 (note that anti-Rad9-T474p is highly specific for the phosphorylated form, while anti-Rad9-S462p retains some binding to the unmodified form). When we purified Rad9 via IP from M phase cells, we observed that these Rad9 S/TP sites were phosphorylated in the presence as well as in the absence of DNA damage, consistent with these sites being modified by CDK (Figs 1B and S1C)17. Notably, we observed that the S/TP sites were also phosphorylated specifically in phleomycin-treated G1 cells, but not in the absence of DNA damage (Fig. 1B, note the phleomycin-induced phosphorylation shift). The anti-Rad9-T474p antibody can also detect Rad9 S/TP phosphorylation from cell extracts. Figure 1C shows Rad9-T474 phosphorylation in undamaged M phase cells, as well as damaged G1 and M phase cells, but not in undamaged G1 cells, corroborating the result of the IP experiment. Moreover, cells expressing the rad9-ST462,474AA variant (referred to as rad9-AA hereafter) did not show any reactivity with the Rad9-T474p antibody, confirming specificity (Figs 1C and S1D). We therefore conclude that there are two different modes of Rad9 S/TP phosphorylation: mode 1, which is cell cycle-regulated and depends on CDK17, and mode 2, which is DNA damage-dependent.

In order to verify that the DNA damage-induced phosphorylation of Rad9 in G1 is CDK-independent, we used a cdc28-as1 mutant strain, in which CDK activity was effectively inhibited by addition of 1-NM-PP1, but this did not abrogate Rad9-T474 phosphorylation after DNA damage (Figs 1D and S1E). We furthermore used the same strategy of CDK-inhibition in M phase-arrested cells and found that CDK-dependent phosphorylation of Rad9-T474 in undamaged cells was effectively inhibited in line with previous results (Figs 1E and S1F)17. Notably, phleomycin treatment efficiently stimulated phosphorylation of Rad9-T474 in M phase-arrested cells after CDK inhibition (Fig. 1E). Taken together, these data show that the damage-induced phosphorylation of the Rad9 S/TP sites occurs independently of the cell cycle phase and CDK activity (Fig. 1E).

DNA damage-induced phosphorylation of the Rad9 S/TP sites depends on the apical checkpoint kinases Mec1 and Tel1 and the Rad9 SCD

Upon DNA damage, the apical checkpoint kinases Mec1 and Tel1 target several sites on Rad98, 11, 12. Therefore, we tested whether also the phosphorylation of Rad9 S/TP sites would be dependent on Mec1 and Tel1. Notably, T474 phosphorylation in G1-arrested, phleomycin-treated cells was reduced in mec1Δ and tel1Δ mutant cells and completely abolished in a mec1Δ tel1Δ double mutant (Figs 2A and S2A). Therefore, phosphorylation of Rad9 S/TP sites shows a dependency on the apical checkpoint kinases, which is highly similar to overall damage-induced Rad9 phosphorylation (indicated by the phosphoshift, Fig. 2A). In contrast, the deletion mutants of the checkpoint effector kinases RAD53 or CHK1, alone or in combination, did not affect T474 phosphorylation (Figs 2B and S2B).

Figure 2
Mec1 and Tel1 are required for phosphorylation of Rad9 S/TP sites after DNA damage. (A) Rad9-T474 phosphorylation after DNA damage depends on the apical checkpoint kinases Mec1 and Tel1. G1-arrested cells with indicated genotypes were treated with phleomycin, ...

It could thus be reasoned that Rad9 S/TP sites are themselves targeted by the apical checkpoint kinases Mec1 and Tel1, similarly to Rad9 S/TQ sites8, 11, 12. However, we did not obtain evidence that purified Mec1 would show activity towards Rad9 S/TP sites in vitro (data not shown). Therefore, we considered the option that the apical checkpoint kinases could promote Rad9 S/TP site phosphorylation indirectly. Possible mechanisms include a priming role of Rad9 S/TQ phosphorylation or Mec1/Tel1 promoting chromatin recruitment of a factor involved in S/TP site phosphorylation, such as the kinase acting on Rad9 or Rad9 itself (via γH2A). Indeed, we found that a Rad9 mutant harbouring six S/T to A exchanges in the S/TQ cluster domain (SCD) (rad9-6AQ)12 abolished phleomycin-induced phosphorylation of Rad9 S/TP sites in G1 (Figs 2C and S2C). In contrast, CDK-dependent phosphorylation of these sites in M phase was unaffected by the rad9-6AQ mutant (Fig. S2D). Previous work has suggested that phosphorylation of the SCD would induce Rad9 dimerization38. However, we excluded dimerization as underlying cause for the SCD-dependency, as the dimerization-defective Rad9-S1129A variant38 showed normal phosphorylation of Rad9-T474 both in G1 after DNA damage and in M phase (Figs 2C and S2D). Overall, we conclude that Mec1/Tel1-dependent phosphorylation of the SCD of Rad9 is required for phosphorylation of the Rad9 S/TP sites upon DNA damage, but additional direct and/or indirect roles of the apical checkpoint kinases are possible.

Chromatin-recruitment of Rad9 is required for phosphorylation of the Rad9 S/TP sites

Previous studies suggest two possible pathways by which Rad9 is recruited to damaged chromatin (‘histone pathway’1922, 25 and ‘Dpb11 pathway’17, 33). In G1 cells, however, the ‘histone pathway’ is apparently uniquely required17, 33. Given our findings, we re-investigated the possibility that the ‘Dpb11 pathway’ may be contributing to Rad9 recruitment and also tested the alternative model that the damage-induced Rad9-Dpb11 interaction in G1 may rely on the ‘histone pathway’.

A critical element of the ‘histone pathway’ is K79-methylation of H3, which is catalysed by the Dot1 methyltransferase23 and recognized by the TUDOR domain of Rad919, 22. We therefore tested Rad9 binding to damaged chromatin by ChIP in G1-arrested cells and used the GAL-HO system to induce a site-specific, non-repairable DSB at the MAT locus39. While Rad9 became enriched in the chromatin region surrounding the DSB in WT cells after DSB induction, Rad9 enrichment was strongly decreased in dot1Δ cells (Figs 3A and S3A). Consistent with a lack of Rad9 recruitment to damaged chromatin, we observed that damage-induced phosphorylation of Rad9 S/TP sites was reduced in G1 cells lacking Dot1 (Figs 3B and S3D).

Figure 3
Dot1 is required for phosphorylation of Rad9 S/TP sites and interaction with Dpb11. (A) Dot1 is required for Rad9 association with a double strand break (DSB). Induction of an non-repairable DSB at MAT locus using galactose-induced HO. ChIP against Rad9 ...

Intriguingly, deletion of DOT1 caused a strong reduction of Rad9-T474 phosphorylation in phleomycin-treated G1 cells (Fig. 3B). To ascertain that this effect originated from a defect in the interaction of Rad9 with nucleosomes (i.e. a deficient ‘histone pathway’), we introduced the corresponding H3 K79-binding-defective mutation in the Rad9 TUDOR domain (rad9-Y798Q 19) and observed a highly similar reduction in Rad9-T474 phosphorylation in this background (Fig. 3B). This effect was again specific for the DNA damage-induced phosphorylation of Rad9 S/TP sites (mode 2), as neither a dot1Δ nor a rad9-Y798Q mutation diminished CDK-dependent phosphorylation of Rad9-T474 in M phase (Fig. S3B,C).

We expected that a lack of Rad9 S/TP phosphorylation would translate into an inability to bind to Dpb11. Indeed, we observed a reduced association of Rad9 in GSTDpb11 pulldowns in the absence of Dot1, when the Rad9-Dpb11 association was induced by phleomycin-treatment of G1-arrested cells (Fig. 3C).

We observed that dot1Δ as well as rad9-Y798Q cells showed minor residual Rad9-T474 phosphorylation in G1 (Fig. 3B), which responded dose-dependently to phleomycin (Fig. S3D). Since M phase cells could compensate a defect in the ‘histone pathway’ by Dpb11-dependent Rad9 recruitment (‘Dpb11 pathway’17, 33), we tested if the ‘Dpb11 pathway’ would be responsible for the residual phosphorylation of Rad9. However, we did not observe any additional defect in Rad9-T474 phosphorylation, when we introduced the Dpb11-binding-deficient ddc1-T602A allele either alone or in combination with dot1Δ (Figs 3D and S3E). Therefore, we conclude that Rad9 S/TP site phosphorylation after DNA damage as well as the interaction of Dpb11 and Rad9 are dependent on the ‘histone pathway’.

Forced Rad9 recruitment to damaged chromatin allows efficient Rad9 S/TP site phosphorylation

The ‘histone pathway’ facilitates Rad9 recruitment to damaged chromatin. We reasoned that the dependency of the damage-induced Rad9 S/TP-phosphorylation on the ‘histone pathway’ could be easily explained, if Rad9 needed to localize to damaged chromatin in order to become phosphorylated. We therefore aimed to create a cellular scenario, which forces Rad9 localization to damaged chromatin independently of the ‘histone pathway’.

We have previously shown that covalent protein-fusions containing the BRCT3 + 4 domain of Dpb11 localized efficiently and cell cycle-independently to damaged chromatin36. In case of Rad9, this fusion protein (Rad9-Dpb11[increment]N, referred to as Rad9-Dpb11 fusion) hyperactivates DNA damage checkpoint signalling17. To ascertain that this fusion acts by forcing Rad9 localization to damaged chromatin, we measured inhibition of DNA end resection by Rad9 as a read-out of Rad9 function13, 14. Therefore, we tested the extent of resection at an HO-induced DSB using ChIP against the ssDNA-binding protein RPA. In the presence of the Rad9-Dpb11 fusion, the spreading of resection was strongly reduced independently of the cell cycle phase and the functionality of the ‘histone pathway’ (Figs 4A and S3A,B). These data therefore suggest a model whereby the Rad9-Dpb11 fusion forces enhanced Rad9 recruitment to damaged chromatin, where it causes hyperactivation of the DNA damage checkpoint, as well as inhibition to DNA end resection, consistent with previous results17, 40.

Figure 4
A Rad9-Dpb11 fusion forces Rad9 recruitment to DSBs and T474 phosphorylation independently of the ‘histone pathway’. (A) The Rad9-Dpb11 fusion blocks resection, also in the absence of Dot1. RPA-ChIP at the indicated positions from an HO-induced ...

Next, we used the Rad9-Dpb11 fusion to test its effects on Rad9 S/TP site phosphorylation. We found that after DNA damage induction Rad9-T474 phosphorylation was enhanced in the context of the Rad9-Dpb11 fusion and even present to low levels without induction of exogenous damage (Figs 4B and S4C,D). Importantly, in the context of the fusion Rad9-T474 phosphorylation was largely independent of Dot1 (Fig. 4B), while it still showed dependency on the apical kinases Mec1 and Tel1 (Figs 4C and S4E). Overall, these data suggest that the function of the ‘histone pathway’ in damage-induced Rad9 S/TP phosphorylation lies entirely in the recruitment of Rad9 to damaged chromatin.

Rad9 S/TP phosphorylation in G1 is dispensable for DNA end resection and the DNA damage checkpoint

Outside of G1, CDK-phosphorylation of Rad9 S/TP sites provides a pathway of Rad9 recruitment to damaged chromatin17. However, in case of the damage-induced Rad9 phosphorylation mode, our data rather suggest a function downstream of recruitment (Figs 3 and and4).4). So far, Rad9 is known to have two functions – (A) inhibition of DNA end resection and (B) activation of the DNA damage checkpoint. Therefore, we tested if the rad9-AA variant would show a G1-specific defect in any of these functions.

To measure DNA end resection, we again used the GAL-HO system and ChIP against RPA. Consistent with previous studies13, 14, we observed enhanced spreading of the RPA-ChIP signal away from the site of the DSB in rad9Δ and dot1Δ strains, indicating enhanced DNA end resection in the absence of chromatin-bound Rad9 (Figs 5A and S5A). However, we did not observe any significant change in DNA end resection in G1-arrested rad9-AA cells, even in the absence of Yku70, suggesting that the Rad9-Dpb11 interaction on its own is not required for regulation of DNA end resection in G1 (Figs 5A,B and S5B).

Figure 5
Lack of damage-induced Rad9 S/TP phosphorylation does not directly affect checkpoint signalling or DNA end resection. (A,B) The rad9-AA mutant – in contrast to the rad9Δ mutant – does not induce hyper-resection in G1-arrested cells. ...

For checkpoint activation, we have previously shown that the rad9-AA mutant on its own does not induce any defects in the phosphorylation of the Rad53 effector kinase in G1 cells17 (see also Figs 5C and S5C). We therefore considered the possibility that a defect in damage-induced Rad9 S/TP phosphorylation may be compensated by other factors. Specifically, we tested compensation by the 9-1-1 complex, since both Rad9 and 9-1-1 could in principle serve to recruit Dpb11 to sites of DNA damage. Therefore, we combined the rad9-AA mutant with the ddc1-T602A mutant, which abolishes the 9-1-1-Dpb11 interaction. However, while the ddc1-T602A mutation strongly reduced Dpb11 association with a site-specific DSB in G1-arrested cells, the rad9-AA mutant did not induce a measurable defect (Fig. S5D). Consistently, checkpoint activation was still largely functional in the rad9-AA mutant, even in the ddc1-T602A background (Fig. 5C).

Overall, the functional relevance of the damage-induced mode of Rad9 S/TP phosphorylation therefore remains unclear. Given the high degree of redundancy in the checkpoint signalling network, it is highly likely that a defect in the rad9-AA mutant is compensated, perhaps by phosphorylation of an additional factor or by other phosphorylation sites in Rad9.

Discussion

S/TP site phosphorylation has been shown to be an important cellular mechanism that facilitates cell cycle controls (see ref. 41 for a review on control of the DNA damage response by S/TP phosphorylation). Our study provides experimental evidence for DNA damage-dependent, but cell cycle-independent phosphorylation of the budding yeast checkpoint protein Rad9 at S/TP sites. These sites have previously been shown to be phosphorylated by CDK and to facilitate interaction with Dpb1117, 42. Notably, we found that also the DNA damage-induced, CDK-independent phosphorylation of Rad9 leads to an interaction with Dpb11. When testing the attributes of DNA damage-induced phosphorylation, we found that it requires the histone methyltransferase Dot1, indicating a dependency on the ‘histone pathway’, which is known to target Rad9 to damaged chromatin1922, 25. Moreover, the covalent Rad9-Dpb11 fusion, which is known to tether Rad9 to damaged chromatin17, 40, bypasses this dependency on the ‘histone pathway’.

We found that damage-induced Rad9 S/TP phosphorylation is abolished, when Rad9 cannot be recruited to damaged chromatin. In this regard, damage-induced Rad9 S/TP phosphorylation is highly similar to Rad9 S/TQ phosphorylation1923, which can be measured as an overall Rad9 phosphorylation shift. Conversely, we observed that forced localization of Rad9 to chromatin, reinstates S/TP phosphorylation, suggesting that Rad9 has to be recruited to damaged chromatin in order to become phosphorylated for both damage-induced S/TP phosphorylation and S/TQ phosphorylation.

Our data therefore suggest that Rad9 S/TP sites are targeted by a chromatin-localized kinase. The apical checkpoint kinases Mec1 and Tel1 would fulfil this requirement, as they are specifically active at damaged chromatin43. Consistently, we found that damage-induced Rad9 S/TP phosphorylation is abolished in a mec1Δ tel1Δ double mutant. However, this influence could also be indirect, since Mec1 and Tel1 are necessary for efficient phosphorylation of the Rad9 SCD, which itself is required for damage-induced phosphorylation of Rad9 S/TP sites. Moreover, we could not find any in vitro evidence to support that Mec1 or Tel1 would directly target S/TP motifs. Currently, the best candidates for this novel mode of Rad9 S/TP phosphorylation, are the transcriptional kinases of the CDK family – Kin28, Srb10, Bur1 and Ctk1 – given their similarity to Cdc28 and their chromatin-localization. In the future, it will therefore be interesting to test the connection between transcriptional CDKs and the DNA damage checkpoint.

Several studies have collectively suggested a model of cell cycle-regulated Rad9 recruitment and activation in budding yeast17, 33, 42 and fission yeast4. These models suggest that the ‘histone pathway’ is exclusively required for Rad9 recruitment to damaged chromatin in G1, while in M phase both ‘histone pathway’ and ‘Dpb11 pathway’ are active. While our study suggests that Rad9 and Dpb11 can interact in G1 as well, this view of Rad9 recruitment pathways is not affected, since Rad9 recruitment via the ‘histone pathway’ is upstream of and required for damage-induced Rad9 S/TP site phosphorylation and the Rad9-Dpb11 interaction in G1. Dpb11, therefore, does apparently not function as Rad9 recruiter in G1. As such, it is currently unresolved what function the damage-induced phosphorylation of Rad9 S/TP sites and subsequent binding to Dpb11 could have. We have not found any phenotypes in the G1 checkpoint or in the control of DNA end resection in G1, when we used the rad9-AA mutant. So far, we have investigated possible redundancies in Dpb11 recruitment (using the ddc1-T602A allele, Fig. 5) and Rad9 recruitment (using the dot1Δ allele, ref. 17), but also this did not reveal a defect. Therefore, the damage-induced Rad9 phosphorylation at S/TP sites may either act redundantly with a currently unknown factor or mediate an entirely new function.

Eukaryotic orthologs of Rad9 have been shown to be recruited to damaged chromatin by related mechanisms4, 2731, 4447. Specifically, both fission yeast Crb2 and human 53BP1 were found to interact with the respective Dpb11 orthologs29, 37. Notably, in human cells 53BP1 and TOPBP1 were found to interact specifically in G137. This interaction, therefore, does seemingly not require CDK-phosphorylation, but would rather be consistent with a DNA damage-induced mode of interaction as described here. The phosphorylation sites on 53BP1 that mediate TOPBP1-binding are currently unknown and it remains to be established whether the DNA damage-induced mode of the Rad9-Dpb11 interaction is evolutionary conserved.

Given the abundance of target proteins that are modified at S/TP sites by CDK48, S/TP site phosphorylation is often interpreted as phosphorylation by CDK. Our results caution, however, that this may be an oversimplified view. It will be interesting to see if CDK-independent S/TP site phosphorylation is a general phenomenon that can be observed on other proteins as well. Phosphoproteomic experiments in human cells treated with etoposide or γ-irradiation have rather suggested an opposite trend, as S/TP phosphorylation was generally decreased49. However, this decrease is caused by the inhibition of Cdk1 and Cdk2 after DNA damage in human cells. In order to test whether a substantial number of S/TP phosphorylation substrates become modified specifically after DNA damage, a system would be required, where CDK is not generally downregulated after DNA damage. While budding yeast fulfils this requirement, previous phosphoproteomic studies of the DNA damage response in budding yeast have primarily focussed on damage-induced S/TQ phosphorylation and checkpoint kinase dependencies50, 51. A systematic investigation of DNA damage-induced S/TP phosphorylation, as well as the involved kinases, therefore appears worthwhile.

Methods

Materials

All yeast strains used in this study were derived from W303 MATa and were constructed using standard methods52. Cells were grown in YP glucose or YP raffinose media at 30 °C. All strains used in this study are listed in Supplementary Table 1, all antibodies in Supplementary Table 2.

Measurement of Rad9 and Rad53 phosphorylation

Cells were grown in YP glucose media at 30 °C or 24 °C. Cell cycle synchronization was performed using α-factor (5 μg/ml or 0.25 μg/ml for bar1Δ mutants) or nocodazole (5 μg/ml) for 2–3 hours. To inhibit CDK, a strain containing the cdc28-as1 allele53 was treated with 1 µM 1-NM-PP1. To induce DNA damage, phleomycin (Invivogen) was added to the medium to a final concentration of 50 µg/ml - or concentrations as indicated. Denaturing cell extracts were prepared by alkaline lysis followed by trichloroacetic acid (TCA) precipitation and precipitated proteins were collected by centrifugation and resuspended in SDS-PAGE sample buffer containing 8 M urea for subsequent SDS-PAGE analysis.

To detect Rad9 phosphorylation on S462 and T474, previously described phospho-specific antibodies were used17. Rad53 phospho-shifts were resolved on 10% acrylamide gels.

Rad9 Immunoprecipitations

For Rad93FLAG IPs cell extracts were prepared from 200 OD yeast cells treated as above for cell cycle arrest and DNA damage. Cells were harvested, washed in ice-cold sorbitol buffer (1 M sorbitol, 25 mM Hepes pH 7.6), and resuspended in a 1:1 ratio with lysis buffer supplemented with protease and phosphatase inhibitors (100 mM Hepes, 200 mM KOAc, 0.1% NP-40, 10% glycerol, 2 mM β-mecaptoethanol, 100 nM okadaic acid, 10 mM NaF, 20 mM β-glycerophosphate, 400 μM PMSF, 4 μM aprotinin, 4 mM benzamidin, 400 μM leupeptin, 300 μM pepstatin A), snap-frozen to liquid nitrogen and lysed using a Spex Sample Prep cryo mill. The extracts were cleared by centrifugation and incubated with anti‐FLAG agarose resin (Sigma) for 1 hour (4 °C, rotation). After five washes with lysis buffer, Rad93FLAG was eluted twice with 0.5 mg/ml 3xFLAG peptide (Sigma). The elutions were pooled and proteins were precipitated with TCA prior to analysis on 4–12% NuPAGE gels (Invitrogen) and standard western blotting.

GST-Dpb11 pulldowns

The Dpb11-Rad9 interaction was tested as described17. GST, GST-Dpb11 FL or a GST-Dpb11 fragment containing BRCT1 + 2 were immobilized on glutathione sepharose 4B (GE Healthcare) and incubated with 600 ml ammonium sulphate-precipitated (57%) cell extracts (in 200 mM KOAc, 100 mM Hepes pH 7.6, 10% glycerol, 0.02% NP-40, 2 mM β-mercaptoethanol, 20 mM β-glycerophosphate, 10 mM NaF, 100 mM okadaic acid, protease inhibitors) corresponding to 50 OD yeast cells. The pulldown was incubated 1 hour (4 °C, rotation), washed and eluted by boiling in SDS-PAGE sample buffer.

Chromatin Immunoprecipitation (ChIP) to a DSB and qPCR analysis

For chromatin immunoprecipitation of Rad9, RPA and Dpb11, cells were grown in YP raffinose to an OD of 0.5 and cell cycle arrest was induced with α-factor or nocodazole. A single double strand break at the MAT locus was introduced by inducing the HO endonuclease from the galactose promoter by addition of galactose to the cultures to a 2% final concentration. 100 ODs of cells were crosslinked with formaldehyde (final 1%) for 16 minutes at timepoints as indicated and the reaction was quenched with glycine. Cells were harvested by centrifugation, washed in ice-cold PBS and snap-frozen. Cell pellets were resuspended in 800 μl lysis buffer (50 mM HEPES KOH pH 7.5, 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, 0.1% Na-deoxycholate, 0.1% SDS) and lysed with zirconia beads using a bead beating device. The chromatin was sonified to shear the DNA to a size of 200–500 bp. The obtained extracts were cleared by centrifugation, 1% was taken as input sample and 40% were incubated with either anti-FLAG-M2 magnetic beads (Sigma) for 2 hours (Rad93FLAG ChIPs) or with anti-RPA antibody (AS07-214, Agrisera) followed by 30 min with Dynabeads ProteinA (Invitrogen, for RPA ChIPs). The beads were washed 3x in lysis buffer, 2x in lysis buffer with 500 mM NaCl, 2x in wash buffer (10 mM Tris-Cl pH 8.0, 0.25 M LiCl, 1 mM EDTA, 0.5% NP-40, 0.5% Na-deoxycholate) and 2x in TE pH 8.0. DNA-protein complexes were eluted in 1% SDS, proteins were removed via proteinase K digestion (3 h, 42 °C) and crosslinks were reversed (8 h or overnight, 65 °C). The DNA was subsequently purified using phenol-chloroform extraction and ethanol precipitation and quantified by quantitative PCR (Roche LightCycler480 System, KAPA SYBR FAST 2x qPCR Master Mix, KAPA Biosystems) at indicated positions with respect to the DNA double strand break. As a control, 2-3 control regions on other chromosomes were quantified.

Electronic supplementary material

Acknowledgements

We thank J. Diffley, M.P. Longhese, J. Petrini for antibodies, plasmid constructs and strains, members of the Jentsch and Pfander labs for stimulating discussion and critical reading of the manuscript. This work was supported by the German Research Council (DFG; project grant PF794/1-1, to BP) and the Max Planck Society (to BP).

Author Contributions

Author Contributions

B.P. initiated the study. G.D.C. and B.P. conducted all experiments. S.C.S.B. and K.U.R. provided preliminary data and materials. All authors analysed the data. B.P. wrote the paper, all authors commented on the manuscript

Notes

Competing Interests

The authors declare that they have no competing interests.

Footnotes

Electronic supplementary material

Supplementary information accompanies this paper at 10.1038/s41598-017-11937-z.

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

1. Finn K, Lowndes NF, Grenon M. Eukaryotic DNA damage checkpoint activation in response to double-strand breaks. Cell. Mol. Life Sci. 2011;69:1447–1473. doi: 10.1007/s00018-011-0875-3. [PubMed] [Cross Ref]
2. Weinert TA, Hartwell LH. The RAD9 gene controls the cell cycle response to DNA damage in Saccharomyces cerevisiae. Science. 1988;241:317–322. doi: 10.1126/science.3291120. [PubMed] [Cross Ref]
3. Saka Y, Esashi F, Matsusaka T, Mochida S, Yanagida M. Damage and replication checkpoint control in fission yeast is ensured by interactions of Crb2, a protein with BRCT motif, with Cut5 and Chk1. Genes Dev. 1997;11:3387–3400. doi: 10.1101/gad.11.24.3387. [PubMed] [Cross Ref]
4. Du L-L, Nakamura TM, Russell P. Histone modification-dependent and -independent pathways for recruitment of checkpoint protein Crb2 to double-strand breaks. Genes Dev. 2006;20:1583–1596. doi: 10.1101/gad.1422606. [PubMed] [Cross Ref]
5. Panier S, Boulton SJ. Double-strand break repair: 53BP1 comes into focus. Nat. Rev. Mol. Cell Biol. 2014;15:7–18. doi: 10.1038/nrm3719. [PubMed] [Cross Ref]
6. Symington LS. End Resection at Double-Strand Breaks: Mechanism and Regulation. Cold Spring Harbor Perspectives in Biology. 2014;6:a016436–a016436. doi: 10.1101/cshperspect.a016436. [PMC free article] [PubMed] [Cross Ref]
7. Sweeney FD, et al. Saccharomyces cerevisiae Rad9 acts as a Mec1 adaptor to allow Rad53 activation. Current Biology. 2005;15:1364–1375. doi: 10.1016/j.cub.2005.06.063. [PubMed] [Cross Ref]
8. Vialard JE, Gilbert CS, Green CM, Lowndes NF. The budding yeast Rad9 checkpoint protein is subjected to Mec1/Tel1-dependent hyperphosphorylation and interacts with Rad53 after DNA damage. EMBO J. 1998;17:5679–5688. doi: 10.1093/emboj/17.19.5679. [PubMed] [Cross Ref]
9. Durocher D, Henckel J, Fersht AR, Jackson SP. The FHA domain is a modular phosphopeptide recognition motif. Mol. Cell. 1999;4:387–394. doi: 10.1016/S1097-2765(00)80340-8. [PubMed] [Cross Ref]
10. Sun Z, Hsiao J, Fay DS, Stern DF. Rad53 FHA domain associated with phosphorylated Rad9 in the DNA damage checkpoint. Science. 1998;281:272–274. doi: 10.1126/science.281.5374.272. [PubMed] [Cross Ref]
11. Emili A. MEC1-dependent phosphorylation of Rad9p in response to DNA damage. Mol. Cell. 1998;2:183–189. doi: 10.1016/S1097-2765(00)80128-8. [PubMed] [Cross Ref]
12. Schwartz MF, et al. Rad9 phosphorylation sites couple Rad53 to the Saccharomyces cerevisiae DNA damage checkpoint. Mol. Cell. 2002;9:1055–1065. doi: 10.1016/S1097-2765(02)00532-4. [PubMed] [Cross Ref]
13. Lazzaro F, et al. Histone methyltransferase Dot1 and Rad9 inhibit single-stranded DNA accumulation at DSBs and uncapped telomeres. EMBO J. 2008;27:1502–1512. [PubMed]
14. Trovesi C, Falcettoni M, Lucchini G, Clerici M, Longhese MP. Distinct Cdk1 requirements during single-strand annealing, noncrossover, and crossover recombination. PLoS Genet. 2011;7:e1002263. doi: 10.1371/journal.pgen.1002263. [PMC free article] [PubMed] [Cross Ref]
15. Ferrari M, et al. Functional Interplay between the 53BP1-Ortholog Rad9 and the Mre11 Complex Regulates Resection, End-Tethering and Repair of a Double-Strand Break. PLoS Genet. 2015;11:e1004928. doi: 10.1371/journal.pgen.1004928. [PMC free article] [PubMed] [Cross Ref]
16. Lydall D, Weinert T. Yeast checkpoint genes in DNA damage processing: implications for repair and arrest. Science. 1995;270:1488–1491. doi: 10.1126/science.270.5241.1488. [PubMed] [Cross Ref]
17. Pfander B, Diffley JFX. Dpb11 coordinates Mec1 kinase activation with cell cycle-regulated Rad9 recruitment. EMBO J. 2011;30:4897–4907. doi: 10.1038/emboj.2011.345. [PMC free article] [PubMed] [Cross Ref]
18. Granata M, et al. Dynamics of Rad9 Chromatin Binding and Checkpoint Function Are Mediated by Its Dimerization and Are Cell Cycle–Regulated by CDK1 Activity. PLoS Genet. 2010;6:e1001047. doi: 10.1371/journal.pgen.1001047. [PMC free article] [PubMed] [Cross Ref]
19. Wysocki R, et al. Role of Dot1-dependent histone H3 methylation in G1 and S phase DNA damage checkpoint functions of Rad9. Mol. Cell. Biol. 2005;25:8430–8443. doi: 10.1128/MCB.25.19.8430-8443.2005. [PMC free article] [PubMed] [Cross Ref]
20. Javaheri A, et al. Yeast G1 DNA damage checkpoint regulation by H2A phosphorylation is independent of chromatin remodeling. Proceedings of the National Academy of Sciences. 2006;103:13771–13776. doi: 10.1073/pnas.0511192103. [PubMed] [Cross Ref]
21. Hammet A, Magill C, Heierhorst J, Jackson SP. Rad9 BRCT domain interaction with phosphorylated H2AX regulates the G1 checkpoint in budding yeast. EMBO Rep. 2007;8:851–857. doi: 10.1038/sj.embor.7401036. [PubMed] [Cross Ref]
22. Grenon M, et al. Docking onto chromatin via the Saccharomyces cerevisiae Rad9 Tudor domain. Yeast. 2007;24:105–119. doi: 10.1002/yea.1441. [PubMed] [Cross Ref]
23. Giannattasio M, Lazzaro F, Plevani P, Muzi-Falconi M. The DNA damage checkpoint response requires histone H2B ubiquitination by Rad6-Bre1 and H3 methylation by Dot1. J. Biol. Chem. 2005;280:9879–9886. doi: 10.1074/jbc.M414453200. [PubMed] [Cross Ref]
24. van Leeuwen F, Gafken PR, Gottschling DE. Dot1p modulates silencing in yeast by methylation of the nucleosome core. Cell. 2002;109:745–756. doi: 10.1016/S0092-8674(02)00759-6. [PubMed] [Cross Ref]
25. Toh GW-L, et al. Histone H2A phosphorylation and H3 methylation are required for a novel Rad9 DSB repair function following checkpoint activation. DNA Repair (Amst) 2006;5:693–703. doi: 10.1016/j.dnarep.2006.03.005. [PubMed] [Cross Ref]
26. Downs JA, Lowndes NF, Jackson SP. A role for Saccharomyces cerevisiae histone H2A in DNA repair. Nature. 2000;408:1001–1004. doi: 10.1038/35050000. [PubMed] [Cross Ref]
27. Fradet-Turcotte A, et al. 53BP1 is a reader of the DNA-damage-induced H2A Lys 15 ubiquitin mark. Nature. 2013;499:50–54. doi: 10.1038/nature12318. [PMC free article] [PubMed] [Cross Ref]
28. Wilson MD, et al. The structural basis of modified nucleosome recognition by 53BP1. Nature. 2016;536:100–103. doi: 10.1038/nature18951. [PubMed] [Cross Ref]
29. Du LL. Histone modification-dependent and -independent pathways for recruitment of checkpoint protein Crb2 to double-strand breaks. Genes Dev. 2006;20:1583–1596. doi: 10.1101/gad.1422606. [PubMed] [Cross Ref]
30. Sanders SL, et al. Methylation of histone H4 lysine 20 controls recruitment of Crb2 to sites of DNA damage. Cell. 2004;119:603–614. doi: 10.1016/j.cell.2004.11.009. [PubMed] [Cross Ref]
31. Botuyan MV, et al. Structural Basis for the Methylation State-Specific Recognition of Histone H4-K20 by 53BP1 and Crb2 in DNA Repair. Cell. 2006;127:1361–1373. doi: 10.1016/j.cell.2006.10.043. [PMC free article] [PubMed] [Cross Ref]
32. Leung CCY, Glover JNM. BRCT domains. Cell Cycle. 2014;10:2461–2470. doi: 10.4161/cc.10.15.16312. [PMC free article] [PubMed] [Cross Ref]
33. Puddu F, et al. Phosphorylation of the Budding Yeast 9-1-1 Complex Is Required for Dpb11 Function in the Full Activation of the UV-Induced DNA Damage Checkpoint. Mol. Cell. Biol. 2008;28:4782–4793. doi: 10.1128/MCB.00330-08. [PMC free article] [PubMed] [Cross Ref]
34. Gritenaite D, et al. A cell cycle-regulated Slx4-Dpb11 complex promotes the resolution of DNA repair intermediates linked to stalled replication. Genes Dev. 2014;28:1604–1619. doi: 10.1101/gad.240515.114. [PubMed] [Cross Ref]
35. Chen X, et al. The Fun30 nucleosome remodeller promotes resection of DNA double-strand break ends. Nature. 2012;489:576–580. doi: 10.1038/nature11355. [PMC free article] [PubMed] [Cross Ref]
36. Bantele SC, Ferreira P, Gritenaite D, Boos D, Pfander B. Targeting of the Fun30 nucleosome remodeller by the Dpb11 scaffold facilitates cell cycle-regulated DNA end resection. eLife. 2017;6:e21687. doi: 10.7554/eLife.21687. [PMC free article] [PubMed] [Cross Ref]
37. Cescutti R, Negrini S, Kohzaki M, Halazonetis TD. TopBP1 functions with 53BP1 in the G1 DNA damage checkpoint. EMBO J. 2010;29:3723–3732. doi: 10.1038/emboj.2010.238. [PubMed] [Cross Ref]
38. Usui T, Foster SS, Petrini JHJ. Maintenance of the DNA-damage checkpoint requires DNA-damage-induced mediator protein oligomerization. Mol. Cell. 2009;33:147–159. doi: 10.1016/j.molcel.2008.12.022. [PMC free article] [PubMed] [Cross Ref]
39. Lee SE, et al. Saccharomyces Ku70, mre11/rad50 and RPA proteins regulate adaptation to G2/M arrest after DNA damage. Cell. 1998;94:399–409. doi: 10.1016/S0092-8674(00)81482-8. [PubMed] [Cross Ref]
40. Liu Y, et al. TOPBP1 Dpb11plays a conserved role in homologous recombination DNA repair through the coordinated recruitment of 53BP1 Rad9. J. Cell Biol. 2017;33:jcb.201607031–17. [PMC free article] [PubMed]
41. Ferretti LP, Lafranchi L, Sartori AA. Controlling DNA-end resection: a new task for CDKs. Frontiers in Genetics. 2013;4:1–7. doi: 10.3389/fgene.2013.00099. [PMC free article] [PubMed] [Cross Ref]
42. Wang G, Tong X, Weng S, Zhou H. Multiple phosphorylation of Rad9 by CDK is required for DNA damage checkpoint activation. Cell Cycle. 2012;11:3792–3800. doi: 10.4161/cc.21987. [PMC free article] [PubMed] [Cross Ref]
43. Gobbini E, Cesena D, Galbiati A, Lockhart A, Longhese MP. Interplays between ATM/Tel1 and ATR/Mec1 in sensing and signaling DNA double-strand breaks. DNA Repair (Amst.) 2013;12:791–799. doi: 10.1016/j.dnarep.2013.07.009. [PubMed] [Cross Ref]
44. Wakeman TP, Wang Q, Feng J, Wang X-F. Bat3 facilitates H3K79 dimethylation by DOT1L and promotes DNA damage-induced 53BP1 foci at G1/G2 cell-cycle phases. EMBO J. 2012;31:2169–2181. doi: 10.1038/emboj.2012.50. [PubMed] [Cross Ref]
45. Huyen Y, et al. Methylated lysine 79 of histone H3 targets 53BP1 to DNA double-strand breaks. Nature. 2004;432:406–411. doi: 10.1038/nature03114. [PubMed] [Cross Ref]
46. Greeson NT, Sengupta R, Arida AR, Jenuwein T, Sanders SL. Di-methyl H4 lysine 20 targets the checkpoint protein Crb2 to sites of DNA damage. J. Biol. Chem. 2008;283:33168–33174. doi: 10.1074/jbc.M806857200. [PMC free article] [PubMed] [Cross Ref]
47. Pellegrino S, Michelena J, Teloni F, Imhof R, Altmeyer M. Replication-Coupled Dilution of H4K20me2 Guides 53BP1 to Pre-replicative Chromatin. CellReports. 2017;19:1819–1831. [PubMed]
48. Holt LJ, et al. Global analysis of Cdk1 substrate phosphorylation sites provides insights into evolution. Science. 2009;325:1682–1686. doi: 10.1126/science.1172867. [PMC free article] [PubMed] [Cross Ref]
49. Beli P, et al. Proteomic investigations reveal a role for RNA processing factor THRAP3 in the DNA damage response. Mol. Cell. 2012;46:212–225. doi: 10.1016/j.molcel.2012.01.026. [PMC free article] [PubMed] [Cross Ref]
50. Smolka MB, Albuquerque CP, Chen S-H, Zhou H. Proteome-wide identification of in vivo targets of DNA damage checkpoint kinases. Proceedings of the National Academy of Sciences. 2007;104:10364–10369. doi: 10.1073/pnas.0701622104. [PubMed] [Cross Ref]
51. de Oliveira, F. M. B. et al. Phosphoproteomics Reveals Distinct Modes of Mec1/ATR Signaling during DNA Replication. Mol. Cell 1–10, 10.1016/j.molcel.2015.01.043 (2015). [PMC free article] [PubMed]
52. Janke C, et al. A versatile toolbox for PCR-based tagging of yeast genes: new fluorescent proteins, more markers and promoter substitution cassettes. Yeast. 2004;21:947–962. doi: 10.1002/yea.1142. [PubMed] [Cross Ref]
53. Bishop AC, et al. A chemical switch for inhibitor-sensitive alleles of any protein kinase. Nature. 2000;407:395–401. doi: 10.1038/35030148. [PubMed] [Cross Ref]

Articles from Scientific Reports are provided here courtesy of Nature Publishing Group