SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

A heteromeric proliferating cell nuclear antigen-like ring complex 9-1-1 is comprised of Rad9, Hus1 and Rad1. When assembled, 9-1-1 binds to TopBP1 and activates the ATR-Chk1 checkpoint pathway. This binding in vitro depends on the phosphorylation of Ser-341 and Ser-387 in Rad9 and is reduced to 70% and 20% by an alanine substitution for Ser-341 (S341A) and Ser-387 (S387A), respectively, and to background level by their simultaneous substitution (2A). Here, we show the importance of phosphorylation of these two serine residues in vivo. siRNA-mediated knockdown of Rad9 in HeLa cells impaired UV-induced phosphorylation of checkpoint kinase, Chk1, and conferred hypersensitivity to UV irradiation and to methyl methane sulfonate or hydroxyurea treatments. Either siRNA-resistant wild-type Rad9 (Rad9Rr) or Rad9Rr harboring the S341A substitution restored the phosphorylation of Chk1 and damage sensitivity, whereas Rad9Rr harboring S387A or 2A did not. However, high expression of S387A restored Chk1 phosphorylation and partially suppressed the hypersensitivity. Thus, the affinity of Rad9 to TopBP1 correlates with the activation of the cellular DNA damage response and survival after DNA damage in HeLa cells, and phosphorylation of Ser-341 and Ser-387 of Rad9 is critical for full activation of the checkpoint response to DNA damage.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Eukaryotic cells respond to genotoxic stress by activating checkpoint pathways that delay cell cycle progression until DNA damage has been repaired (Abraham 2001; Bakkenist & Kastan 2004; Lambert & Carr 2005). Ataxia-telangiectasia-mutated (ATM)-Rad3-related (ATR) kinase is an upstream regulator of the DNA damage response that exists in a complex with ATR-interacting protein (ATRIP) (Cortez et al. 2001; Byun et al. 2005; Cimprich & Cortez 2008). The ATR–ATRIP complex is recruited to sites of DNA damage via an interaction with replication protein A (RPA), which recognizes and binds to damage-induced single-stranded DNA (ssDNA) (Zou & Elledge 2003; Namiki & Zou 2006). Once activated, ATR–ATRIP phosphorylates various substrates, including Chk1, which in turn activates additional downstream effectors through the phosphorylation of key cell cycle regulatory proteins such as Cdc25 and Wee1 (Abraham 2001; Perry & Kornbluth 2007; Burrows & Elledge 2008).

Activation of ATR–ATRIP requires two factors, TopBP1 and 9-1-1, both of which are recruited to and accumulate at sites of DNA damage and interact with each other (Delacroix et al. 2007; Lee et al. 2007). The 9-1-1 complex is a proliferating cell nuclear antigen (PCNA)-like ring structure comprised of three cell cycle checkpoint factors, Rad9, Hus1 and Rad1 (St. Onge et al. 1999; Volkmer & Karnitz 1999; Griffith et al. 2002; Shiomi et al. 2002). Loading of 9-1-1 onto damaged chromatin is accomplished via a loader complex, Rad17-replication factor-C, and is dependent on RPA, but not ATR–ATRIP (Zou et al. 2003). Rad9−/− mouse embryonic stem (ES) cells exhibit high sensitivity to UV irradiation and hydroxyurea (HU) (Roos-Mattjus et al. 2003). Additionally, in A549 human lung adenocarcinoma cells, siRNA-mediated silencing of endogenous Rad9 results in a decrease in the population of cells in G2-M phase after ionizing radiation compared to nonsilenced cells (Yuki et al. 2008). These results indicate that Rad9 is a key component of the DNA damage checkpoint response in mammalian cells.

TopBP1 contains nine breast cancer 1 (BRCA1) carboxyl-terminal (BRCT) motifs that mediate protein–protein interactions (Rappas et al. 2011). TopBP1 binds to and activates the ATR–ATRIP complex via an ATR activation domain identified between the seventh and eighth BRCT repeats (Kumagai et al. 2006; Mordes et al. 2008). TopBP1 also binds to Rad9 of the 9-1-1 complex via its N-terminal BRCT motifs (Mäkiniemi et al. 2001; Delacroix et al. 2007; Lee et al. 2007; Rappas et al. 2011), and the phosphorylation of Ser-387 of Rad9 has been shown to be essential for the interaction (St Onge et al. 2003; Delacroix et al. 2007; Lee et al. 2007). We previously reported that phosphorylation of Ser-341 also promotes the interaction between TopBP1 and Rad9 in vitro, although to a lesser extent than Ser-387 (Takeishi et al. 2010). In HeLa cells, over-expression of a variant of Rad9 (2A) in which both Ser-387 and Ser-341 are replaced by alanine resulted in increased UV and methyl methane sulfonate (MMS) sensitivity, presumably through a dominant-negative mechanism. Interestingly, although the contribution of phosphorylated Ser-387 and Ser-341 to TopBP1 binding appears to be quite different in vitro, over-expression of a single alanine substitution mutant of Ser-387 or Ser-341 (S387A or S341A, respectively) did not confer clear sensitivity to UV or MMS treatment compared to wild-type (WT) Rad9. These results indicated that over-expressed S387A as well as S341A could interact functionally with TopBP1, even though Ser-341 appeared to be a lower affinity site in vitro, and the interaction was sufficient for the activation of the DNA damage checkpoint response (Takeishi et al. 2010). However, the approach of over-expression of Rad9 does not clarify whether the interaction of 9-1-1 with TopBP1 through phospho-Ser-341 and phospho-Ser-387 was directly involved in the activation of the DNA damage response.

In the current study, we investigated the role of Rad9 Ser-387 and Ser-341 phosphorylation in the DNA damage response under near-physiological conditions. Cell lines expressing siRNA-resistant (Rad9Rr) WT or the single or double alanine substitution mutants (S341A, S387A or 2A) were generated, and the interaction of Rad9 with TopBP1 was examined by immunoprecipitation from cell extracts. Sensitivity to UV irradiation, MMS and HU and activation of the DNA damage checkpoint, as assessed by the phosphorylation of Chk1, after UV irradiation were investigated. We found that activation of the DNA damage response correlated with the binding of Rad9 to TopBP1, and both serine residues in the C-terminus of Rad9 were necessary for full activation of the DNA damage response.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Rad9 knockdown cells exhibit high sensitivity to UV, MMS and HU treatment

In previous work, we showed that over-expression of exogenous Rad9 2A, a double alanine substitution mutant of Ser-341 and Ser-387, conferred increased sensitivity to UV irradiation and MMS in HeLa cells, whereas expression of the individual substitution mutants (S341A or S387A) did not. These results indicated that both serine residues in the C-terminus of Rad9 are involved in the cellular DNA damage response in vivo. Additionally, casein kinase 2 (CK2)-dependent phosphorylation of Ser-341 and Ser-387 was required for maximal TopBP1 binding activity of 9-1-1 in vitro (Takeishi et al. 2010). However, a dominant-negative approach of over-expression of exogenous Rad9 variants does not elucidate whether increased sensitivity to DNA damage is a direct consequence of the inhibition of Rad9 and TopBP1 binding through phosphorylated Ser-341 and Ser-387. To directly examine the role of Rad9 serine phosphorylation in the binding of 9-1-1 to TopBP1 and activation of the DNA damage response, Rad9 mutants were expressed under more physiological conditions in cells depleted of endogenous Rad9 by RNA interference (RNAi).

Transfection of a Rad9-specific siRNA into HeLa cells resulted in the depletion of approximately 80% of endogenous Rad9 by 72 h post-transfection and a slight slowing of cell growth (Fig. 1A,B). In addition, these cells were clearly more sensitive to UV irradiation, MMS and HU (Fig. 1C). Thus, Rad9 was required for the normal response to DNA-damaging agents, consistent with the work of Roos-Mattjus et al. (2003).

image

Figure 1. siRNA-mediated knockdown of Rad9 in HeLa cells. (A) After Rad9 siRNA transfection, whole-cell extracts were analyzed by immunoblot using an anti-Rad9 antibody (upper). Polδ was analyzed in parallel as a loading control (lower). Samples treated with Rad9 siRNA or mock-transfected are indicated as ‘+’ and ‘−’, respectively. (B) Relative cell number after Rad9 siRNA transfection. Cells were counted every 24 for 96 h as described in 'Experimental procedures'. Cell number at 24 h was set as 1.0. Data represent the means ± SEM of three independent experiments. siRNA- and mock-transfected cells are indicated as ‘KD +’ (dotted line) and ‘KD−’ (solid line), respectively. (C) Cell survival, as assessed by relative cell number, of siRNA- (KD+; broken lines) and mock-transfected (KD−; solid lines) cells after treatment with the indicated doses of UV irradiation, methyl methane sulfonate (MMS), or hydroxyurea (HU). Cell number was measured as described in 'Experimental procedures' with cell number before exposure (dose 0) set as 1. Data represent the means ± SEM of two independent experiments.

Download figure to PowerPoint

Expression of siRNA-resistant Rad9 (Rad9Rr)WT and Rad9 mutants in HeLa cells

Expression plasmids (pcDNA3-based) harboring FLAG epitope-tagged Rad9Rr WT, S341A, S387A or 2A were introduced into HeLa cells. Stable G418-resistant colonies expressing exogenous Rad9 were isolated. The representative cell lines used in this work are shown in Fig. 2A. Cell lines indicated by ‘L’ represent cells in which the expression level of Rad9 was increased approximately twofold, indicating the expression of Rad9Rr at a comparable level to endogenous Rad9 (Fig. 2A, lanes 2, 3, 4 and 6). Two cell lines that expressed S387A and 2A at 10-fold or higher levels compared to endogenous Rad9 (lanes 5 and 7) were also tested and are indicated by ‘H’.

image

Figure 2. Expression levels of Rad9 and Rad9 binding to TopBP1 in HeLa cells expressing exogenous Rad9Rr wild-type (WT) or the indicated Rad9 mutants. (A) Total Rad9 (endogenous and exogenous) was detected by immunoblot analysis using an anti-Rad9 antibody. Polδ was analyzed in parallel as a loading control. ‘L’ and ‘H’ indicate low- and high-level expression cell lines, respectively, as described in the text. (B) Co-immunoprecipitation of TopBP1 with FLAG-Rad9 from L cell lines. Two percent of the input cell lysate (lanes 1–5) and 50% of the bound fraction (lanes 6–10) were analyzed by immunoblot. The upper and lower panels represent TopBP1 and Rad9, respectively. (C) Quantification of relative levels of TopBP1 bound to Rad9. Bound TopBP1 was normalized to the amount of precipitated Rad9 and expressed relative to WT-L, which was set as 1.0. Data represent the means ± SEM of two independent experiments.

Download figure to PowerPoint

None of the Rad9 cell lines exhibited significant growth defects (Fig. S1 in Supporting Information). To determine the ability of exogenous Rad9 WT and Rad9 mutants to engage in protein–protein interactions, cell lysates were subjected to immunoprecipitation using anti-FLAG antibody-coated beads followed by immunoblot analysis to detect co-precipitated Rad1 and Hus1 (Fig. S2 in Supporting Information). Similar levels of Hus1 and Rad1 were co-immunoprecipitated with FLAG-Rad9Rr from all cell lines, which indicated that 9-1-1 complexes containing exogenous Rad9 were formed efficiently. In cells expressing higher levels of Rad9, higher levels of Rad1 and Hus1 were detected in the immunoprecipitates, which indicated that there is a pool of free Rad1 and Hus1 in cells and that under these experimental conditions, the level of 9-1-1 is determined by the level of Rad9 expression. These results were consistent with our previous findings (Takeishi et al. 2010). Thus, the activity of 9-1-1 in the cell lines could be assumed to mirror the levels of exogenous Rad9, even in the H cell lines, if endogenous Rad9 was depleted. To determine whether the Rad9 mutants exhibited similar affinities for TopBP1 in human cell extracts as previously shown in vitro, the presence of TopBP1 in FLAG-Rad9Rr immunoprecipitates from the L cell lines was assessed by immunoblot analysis (Fig. 2B). Less than 1% of cellular TopBP1 was bound to WT Rad9 (WT-L; lanes 2, 7). The amount of co-immunoprecipitated TopBP1 in S387A-L and 2A-L cells was approximately 10% of that seen in WT-L cells (lanes 9, 10), whereas in S341A-L cells, the amount of bound TopBP1 was similar to that of WT-L (lane 8). These results indicated that the affinities of Rad9 WT and the Rad9 mutants for TopBP1 in human cell extracts were similar to those seen in vitro, although quantitative differences were difficult to reproduce by this experiment, perhaps owing to their actual concentrations in human cell extracts different from the in vitro binding condition.

Sensitivity of S387A-L and 2A-L cells to UV, MMS and HU compared to Rad9 knockdown cells

Rad9 levels were assessed in Rad9Rr-expressing cell lines after siRNA transfection (Fig. 3A). In parental HeLa, WT-L, 2A-L, S341A-L and S387A-L cells, expression level of Rad9 reduced by approximately 80%, 42%, 57%, 25% and 36%, respectively, by treatment with a Rad9-specific siRNA. With the exception of parental HeLa cells, the level of Rad9 remaining was comparable to that of endogenous Rad9. After siRNA treatment, cells were exposed to UV irradiation, MMS or HU and then cell survival was assessed. In S387A-L and 2A-L cells, there was a clear decrease in survival after knockdown of endogenous Rad9, similar to siRNA-treated parental HeLa cells (Fig. 3B). In contrast, S341A-L and WT-L were unaffected by any of the treatments, with the exception of a partial sensitivity to UV in WT-L cells. Given that WT-L cells were resistant to MMS and HU and exhibited only partial sensitivity to UV treatment (Fig. 3B), we concluded that WT-L cells were largely resistant to DNA-damaging agents.

image

Figure 3. Knockdown of endogenous Rad9 in wild-type (WT)-L, S341A-L, S387A-L and 2A-L cells and cell sensitivity DNA damage. (A) Levels of Rad9 in low-level Rad9Rr-expressing cells with or without Rad9 siRNA treatment. After transfection with Rad9 siRNA for 72 h, whole-cell extracts were analyzed by immunoblot using an anti-Rad9 antibody (upper left panels). Polδ was analyzed in parallel as a loading control (lower). Samples treated with Rad9 siRNA or mock-transfected are indicated ‘+’ and ‘−’, respectively. Rad9 expression was normalized to Polδ expression and is expressed relative to mock-transfected HeLa cells (bar graph; right). Black bars, mock-transfected; white bars, cells transfected with Rad9 siRNA. (B) Cell survival of siRNA-transfected HeLa (KD+; squares), WT-L (crosses), S341A (triangles), S-387-L (crosses with solid line) and 2A-L (circles) cells after treatment with the indicated doses of UV irradiation (left), methyl methane sulfonate (MMS) (middle) or hydroxyurea (HU) (right). Cell number was measured as described for Fig. 1 and is expressed relative to unexposed cells (0 dose), which was set as 1. Mock-transfected HeLa cells (KD−; diamonds) were analyzed as a control. Data represent the means ± SEM of two independent experiments.

Download figure to PowerPoint

Unlike previous results derived from experiments in which Rad9 was over-expressed (Takeishi et al. 2010), the current results suggested that there are functional differences between Ser-341 and Ser-387. When expressed at levels comparable to endogenous Rad9 in a Rad9-depleted background, S341A was as functional as Rad9 WT, whereas S387A was nearly inactive, similar to 2A. These results indicated that Ser-387 plays a major role in the DNA damage response, which is consistent with other reports (Delacroix et al. 2007; Lee et al. 2007), whereas Ser-341 is not absolutely required. These results were consistent with the levels of TopBP1 bound to the various Rad9 mutants in HeLa cells (Fig. 2b), but were in disagreement with their apparent affinities in vitro, where Ser-341 and Ser-387 contributed approximately 20% and 70% to TopBP1-binding capacity, respectively (Takeishi et al. 2010).

Partial suppression of UV, MMS and HU sensitivity in S387A-H cells

Given that Ser-341 contributed approximately 20% to the interaction of 9-1-1 with TopBP1 in vitro, we evaluated the role of Ser-341 in the DNA damage response further, using Rad9-depleted HeLa cells expressing high levels of S387A and 2A. In the presence of endogenous Rad9, over-expression of 2A, but not S387A, resulted in significantly increased sensitivity to UV and MMS treatment (Takeishi et al. 2010). In both H cell lines, there were no obvious quantitative differences in total cellular Rad9 after the depletion of endogenous Rad9, as the Rad9Rr mutants were expressed at approximately 10-fold higher levels than endogenous Rad9 (Fig. 4A). 2A-H cells exhibited a high sensitivity to UV, MMS and HU treatment (Fig. 4B), which indicated that substitution of both serine residues eliminated the response to DNA damage, regardless of the expression level of 2A. However, S387A-H cells exhibited an intermediate sensitivity to DNA damage (Fig. 4B), which indicated a partial suppression of the defect observed in S387A-L cells. These results suggested that S387A, which contains only the Ser-341 phosphorylation site, is functional and can contribute to the DNA damage response when expressed at high levels.

image

Figure 4. Knockdown of endogenous Rad9 in S387A-H and 2A-H cells and cell sensitivity to DNA damage. (A) Levels of Rad9 in high-level Rad9Rr-expressing cells with or without Rad9 siRNA treatment. After transfection with (+) or without (−) Rad9 siRNA for 72 h, whole-cell extracts were analyzed by immunoblot using an anti-Rad9 antibody (upper). Polδ was analyzed in parallel as a loading control (lower). Rad9 expression was normalized to Polδ expression and is expressed relative to mock-transfected HeLa cells, which was set as 1.0 (bar graph; right). Black bars, mock-transfected cells; white bars, Rad9 siRNA-transfected cells. (B) Cell survival of siRNA-transfected HeLa (KD+; crosses), S387A-H (circles) and 2A-H (boxes) cells after treatment with the indicated doses of UV irradiation (left), methyl methane sulfonate (MMS) (middle) or hydroxyurea (HU) (right). Cell survival was measured as described for Fig. 1 and is expressed relative to unexposed cells (dose 0), which was set as 1. Mock-transfected HeLa cells (KD−; diamonds) were analyzed as a control. Data represent the means ± SEM of two independent experiments.

Download figure to PowerPoint

S387A, but not 2A, can overcome defects in Chk1 activation in response to DNA damage

Binding of TopBP1 to 9-1-1 results in the activation of ATR and phosphorylation of Chk1 (Delacroix et al. 2007; Lee et al. 2007). The differences in cell survival among the various Rad9-expressing cell lines could reflect differences in Chk1 phosphorylation levels, depending on the relative affinities of 9-1-1 for TopBP1. Cell lines were exposed to UV irradiation, and then Chk1 phosphorylation in whole-cell extracts was analyzed by immunoblotting (Fig. 5). Phosphorylation of Chk1 in response to UV irradiation was reduced to approximately 60% in Rad9-depleted HeLa cells. A similar reduction was observed in S387A-L cells, but not in S387A-H cells. Thus, when expressed at high levels, S387A could compensate for the defect in Chk1 phosphorylation in response to UV in Rad9-depleted cells, although it was unable to do so at low levels. These results were consistent with the sensitivity of the cell lines to DNA damage (Figs 3b,4b). In 2A-L cells, Chk1 phosphorylation was reduced by approximately 50% under Rad9-depleted conditions, which indicated that exogenous 2A was unable to rescue the checkpoint defect (Fig. 5). These results were consistent with cell survival results showing that low-level expression of 2A was unable to overcome the sensitivity to DNA damage caused by the loss of TopBP1 binding activity. In 2A-H cells, Chk1 phosphorylation was reduced to approximately 60%, even without siRNA treatment, with an additional 10% reduction under Rad9-depleted conditions. This was also consistent with the cell survival results showing that over-expression of 2A impairs the DNA damage response in a dominant-negative manner. Thus, the ability of Rad9 to bind to TopBP1 regulates the cellular DNA damage response primarily through the activation of the ATR-Chk1 signaling axis.

image

Figure 5. Hyperphosphorylation of Chk1 in S387A-L, S387A-H, 2A-L and 2A-H cells after UV irradiation. After transfection with (+) or without (−) Rad9 siRNA for 72 h, cells were exposed to UV irradiation (80 J/m2) and then allowed to incubate for 2 h. Cell lysates were analyzed by immunoblot using an anti-phospho-Chk1 (Ser-345) antibody (phosphorylated Chk1; upper) and an anti-Chk1 antibody (total Chk1; lower). (b) Relative levels of phosphorylated Chk1 in cells with (+; white bars) or without (−; gray bars) Rad9 siRNA treatment. Levels of phosphorylated Chk1 were normalized to total Chk1 and expressed relative to mock-transfected HeLa cells, which was set as 1.0. Data represent the means ± SEM of two independent experiments.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Rad9 binding to TopBP1 is mediated by two target serine phosphorylation sites in the Rad9 C-terminus. Single amino acid substitution mutants of each site, S341A and S387A, exhibit approximately 70% and 20% binding capacity for TopBP1, whereas the double mutant, 2A, exhibits background levels of binding (Takeishi et al. 2010). Previously, using a dominant-negative approach of over-expression of these three Rad9 mutants, both Ser-341 and Ser-387 appeared to be involved in the DNA damage response in HeLa cells. To investigate the function of these residues under more physiological conditions, we generated cell lines that stably expressed FLAG-tagged Rad9Rr WT and Rad9 mutants. In these cells, treatment with a Rad9-specific siRNA would deplete endogenous Rad9 while leaving exogenous Rad9Rr unaffected. Under conditions in which exogenous Rad9 was expressed at levels comparable to endogenous Rad9, cells expressing S387A and 2A were as sensitive to DNA damage as Rad9-depleted HeLa cells, whereas cells expressing S341A were resistant to DNA damage, similar to HeLa cells that were not depleted of Rad9. Interestingly, when expressed at high levels, S387A conferred partial resistance to DNA damage, whereas 2A did not, even at high levels of expression. Thus, S387A, which contains only the Ser-341 target phosphorylation site, is functional and can interact with TopBP1 and suppress the defect in DNA damage response when expressed in a high enough levels, whereas 2A appears to be completely nonfunctional. Given the current results, we propose that both phospho-Ser-387 and phospho-Ser-341 are necessary for the full DNA damage response in HeLa cells, but that the former plays a more important role than the latter, consistent with the TopBP1 binding affinities of 2A, S341A and S387A in vitro. Unlike S387A, S341A was almost fully active in terms of the DNA damage response under our assay conditions. By comparison, Ser-341, which contributed approximately 20% to the binding of 9-1-1 to TopBP1 in vitro, appeared to be unnecessary. Because these experiments were carried out at near-lethal doses of DNA-damaging agents, one possibility is that 9-1-1 and TopBP1 would accumulate extensively at heavily damaged sites and that the 70% binding capacity is sufficient to activate the cellular DNA damage response. From an apparent role of phospho-Ser-341 for the binding, we suggest that the interaction between 9-1-1 and TopBP1 through both phospho-Ser-341 and phospho-Ser-387 would be needed to stabilize the functional complex at DNA-damaged sites of sublethal doses, where their accumulation is limited.

TopBP1 binds to Rad9 via two N-terminal BRCT motifs, BRCT1 and BRCT2 (Delacroix et al. 2007; Lee et al. 2007). Rappas et al. (2011) showed that BRCT1 is the primary binding site for Ser-387 and suggested that BRCT2 might function as a second binding site for something other than Ser-387, although phospho-Thr-355, phospho-Ser-375 and phospho-Ser-380 in the Rad9 C-terminus were subsequently ruled out as putative binding targets. Thus, phospho-Ser-341 appears to be the most plausible target for BRCT2, and our results suggest that engagement of both sites confers maximum affinity binding of 9-1-1 and TopBP1. This model remains to be clarified by structural analysis of the binding complex.

As expected, phosphorylation of Chk1 was decreased by Rad9 mutations that impaired the interaction between 9-1-1 and TopBP1. The inability of S387A at low levels or 2A (at low or high levels) to complement the defect in DNA damage response, as well as the partial complementation by high-level expression of S387A, correlated with the levels of Chk1 phosphorylation in cells after siRNA treatment. These results indicate that Rad9 directly regulates the activity of ATR/ATRIP. Interestingly, the level of phosphorylated Chk1 in S387A-H cells after Rad9 knockdown was approximately 110% of that in nontreated HeLa cells, although the abilities in response to UV, MMS and HU treatment were partially abrogated. This result indicated that whereas S387A fully restored activation of Chk1, survival after DNA damage was not completely restored and required active endogenous Rad9. The binding of 9-1-1 to TopBP1 may be required for the activation of other factors involved in the DNA damage response, which act either downstream of or in parallel to Chk1. Additional studies aimed at identifying novel binding partners of Rad9, 9-1-1 or TopBP1 will shed light on this interesting possibility.

The phosphorylation of Chk1 in 2A-L and 2A-H cells in the absence of Rad9 siRNA treatment was decreased to 95% and 59% compared to HeLa (KD−) cells even in the presence of functional endogenous Rad9. 2A formed 9-1-1 complexes but exhibited negligible binding to TopBP1. Thus, it acted in a dominant-negative manner regardless of its expression level and inhibited Chk1 phosphorylation. A similar dominant-negative effect on the DNA damage response by over-expression of 2A has been reported previously (Takeishi et al. 2010). In contrast, the phosphorylation of Chk1 in S387A-L and S387A-H cells was increased to 117% and 138%, respectively. These results suggested that despite the low affinity of TopBP1 for phospho-Ser-341, the excess exogenous S387A would function additively to that of endogenous Rad9. These results again support the idea that Ser-341 plays a significant role in the DNA damage response through the activation of the ATR-Chk1 pathway.

In summary, we have shown that Rad9 Ser-341 and Ser-387 are both required for full activation of the DNA damage response in HeLa cells and that the affinity of Rad9 for TopBP1 strongly correlates with the activation of the DNA damage response. This work clarifies the importance of the interaction between 9-1-1 and TopBP1 as a major determinant of checkpoint activity in human cells and suggests that the ability to properly regulate this interaction is critical for responding to and tolerating DNA damage. Thus, there are likely to be mechanisms in place to fine-tune checkpoint activity based on the levels of phospho-Ser-dependent interactions. Of note, phosphorylation of Ser-341 and Ser-387 appears to be constitutive (Takeishi et al. 2010). This implies dephosphorylation of particular functional Rad9 or other modifications of the neighboring regions for the mechanism. Further analysis of the 9-1-1–TopBP1 interaction will provide insight into how cells respond so precisely to DNA damage. The current results also suggest a potential therapeutic target for sensitizing cancer cells to therapy. The checkpoint response in most cancer cells is impaired, enabling increased growth rates, a well-known property of cancer cells that is exploited by the various chemotherapeutic agents used in cancer therapy. If the Rad9–TopBP1 interaction is fully functional, challenging these cells with a molecule that can weaken this interaction might induce the cells to die at a lower dose of chemotherapeutic agent. The system developed herein could serve as a useful tool for the development of such novel anticancer therapies.

Experimental procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Preparation of plasmids for the expression of Rad9

To construct the cDNA encoding Rad9Rr, the five siRNA target nucleotides in pcDNA-FLAG-Rad9 were mutated using the Quick-Change Site-Directed Mutagenesis kit (Stratagene, LaJolla, CA, USA) and the following primers: Rad9siR2#FW, 5′-CAGAGTCAGCAAACTTGAACCTATCGATTCATTTTGATGCTCCAGG-3′; and Rad9siR2#RV, 5′-CCTGGAGCATCAAAATGAATCGATAGGTTCAAGTTTGCTGACTCTG-3′. Underlined nucleotides indicated the targeted nucleotides. Substitutions were confirmed by sequencing, and the resultant plasmid, pcDNAFLR9WTdup2, was used for the construction of Rad9 mutants. S341A, S387A and 2A were generated by replacing the ClaI-NotI fragment with the corresponding fragment from expression plasmids encoding S341A, S387A and 2A (Takeishi et al. 2010) to generate pcDNAFLR9341Adup2, pcDNAFLR9387Adup2 and pcDNAFLR92Adup2, respectively. Rad9 proteins expressed from these plasmids were referred to as Rad9Rr-S341A, Rad9Rr-S387A and Rad9Rr-2A, respectively.

Cell culture and isolation of stable transformants

Cells were cultured in Dulbecco's modified Eagle's medium (DMEM; Sigma-Aldrich, St. Louis, MO, USA) supplemented with 10% fetal bovine serum (Invitrogen, Carlsbad, CA, USA), 100 μg/mL of streptomycin and 100 unit/mL of penicillin (Sigma-Aldrich) at 37 °C in 5% CO2. To obtain stable cell lines expressing exogenous Rad9, HeLa cells were transfected with the appropriate expression plasmid using FuGENE 6 (Roche, Mannheim, Germany) and cultured in the presence of 400 μg/mL G418 (Nacalai Tesque, Kyoto, Japan) to select stable transformants. All cell lines were subsequently maintained in medium containing 100 μg/mL G418.

siRNA-mediated knockdown of Rad9

HeLa cells (2.5 × 105) were transfected with 20 pmol annealed RAD9-specific siRNA (target sequence, UCAGCAAACUUGAAUCUUAGCAUUC) (Integrated DNA Technologies, Coralville, IA, USA) using Lipofectamine RNAiMAX and Opti-MEM (Invitrogen), according to the manufacturer's instructions. Transfections were carried out in a volume of 2 mL of DMEM in 35-mm dishes. To assess the expression levels of Rad9, cells were lysed in 1 x Laemmli buffer (50 mm Tris–HCl, pH6.8, 100 mm DTT, 2% SDS, 0.05% bromophenol blue and 10% glycerol) 72 h post-transfection and then analyzed by immunoblot using an anti-Rad9 antibody (sc-8324; Santa Cruz Biotech, Santa Cruz, CA, USA).

For cell survival assays, cells transfected with siRNA for 48 h were collected from 35-mm dishes (2.0 × 103), dispensed into a 96-well plate and then allowed to adhere for 24 h. Cells were exposed to UV irradiation at the indicated doses and then incubated at 37 °C for 72 h, or treated with MMS or HU at the indicated concentrations for 24 h and then incubated in fresh medium for 48 h. Viable cells were determined using a Cell Counting Kit8 (Dojindo, Kumamoto, Japan), according to the manufacturer's instructions. Optical density at 450 nm (OD450) was measured using a Multiskan FC (Thermo, Waltham MA, USA) after a chromogenic reaction for 1 and 2 h at 37 °C. Cell growth rate was measured by counting the number of cells every 24 h using the same procedure but without treatment with DNA-damaging agents.

Immunoprecipitation of FLAG-Rad9

Cells (1.6 × 106) were inoculated in a 10-cm dish and allowed to incubate at 37 °C for 48 h. Cells were trypsinized, collected by centrifugation at 640 g for 5 min, washed with 3 mL of PBS (Takara, Shiga, Japan) two times and then lysed with 97.5 μL of buffer H (25 mm HEPES, pH 7.8, 1 mm EDTA, 10% glycerol, 150 mm NaCl) containing 0.1 mm phenylmethylsulfonyl fluoride, 2 μg/mL of leupeptin, 16.7 mm β-glycerophosphate (Sigma-Aldrich), 1.7 mm Na3VO4 (Sigma-Aldrich), 10.4 mm NaF (Sigma-Aldrich) and 0.5% NP40. After incubation for 30 min on ice, the lysate was subjected to ultracentrifugation at 84,000 g for 15 min. A 95-μL aliquot of the cell lysate was incubated with 8 μL of anti-FLAG beads (Sigma-Aldrich) for 1.5 h at 4 °C. The beads were washed three times with 100 μL of buffer H and then bound proteins were eluted in 10 μL of 1 x Laemmli buffer. Proteins were separated by 12.5% SDS-PAGE, then transferred to a PVDF membrane (GE Healthcare, Buckinghamshire, UK). The following primary antibodies were used, as indicated: anti-Rad9 (sc-8324), anti-Polδ (sc-8739) and anti-Chk1 (sc-8408) (all from Santa Cruz Biotech) and anti-phospho-Ser345 Chk1 (#2341; Cell Signaling Technology, Danvers, MA, USA).

Analysis of Chk1 phosphorylation

After siRNA transfection for 48 h, cells (2.5 × 105 cells/well) in a 24-well plate were allowed to adhere for 24 h and then exposed to UV irradiation at 80 J/m2. Cells were allowed to incubate at 37 °C for 2 h, after which they were washed with PBS, lysed in 50 μL of a 1 : 1 mixture of 2 x buffer H and 2 x Laemmli buffer and then analyzed by immunoblot using the indicated antibodies.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

This study was supported by a grant-in-aid for Scientific Research (KAKENHI) and by a grant from the Uehara Memorial Foundation for E. O. and by a grant-in-aid for Research Fellow of the Japan Society for the Promotion of Science for Y. T.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information
FilenameFormatSizeDescription
gtc1630-sup-0001-figs.pdfWord document134KFigure S1 Co-immunoprecipitation of Hus1 and Rad1 with FLAG-Rad9 from HeLa, WT-L, S341A-L, S387A-L, S387-H, 2A-L and 2A-H cells. Figure S2 Growth of WT-L (crosses with broken line), S341A-L (triangles), S387A-L (crosses with solid line) and 2A-L (circles) (a), S387A-H (b) and 2A-H (c) cells.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.