Timeless-Tipin is involved in replication fork stabilization in the presence and absence of genotoxic agents

Depletion of Timeless or Tipin from human cells has been shown to cause accumulation of spontaneous foci containing phosphorylated histone H2AX (γH2AX) (Chou and Elledge, 2006; Urtishak et al., 2009), indicative of DNA damage even in the absence of genotoxic agents. Furthermore, depletion of Timeless causes increased sister chromatid exchange, which is dependent on Brca2 and Rad51 (Urtishak et al., 2009). These results suggest that Timeless-Tipin has a crucial role in preventing DNA damage at replication forks. To examine the effects of Timeless or Tipin depletion on fork stabilization, we used the established siRNA sequences (Tim-#1 and Tim-#2 for Timeless; Tip-#1 for Tipin) that are designed to target Timeless or Tipin (Unsal-Kacmaz et al., 2005; Yoshizawa-Sugata and Masai, 2007) and confirmed that expression of Timeless or Tipin was dramatically reduced (Fig. 2A and supplementary material Fig. S1). As described previously (Chou and Elledge, 2006; Yoshizawa-Sugata and Masai, 2007), we noticed that depletion of Timeless caused a dramatic reduction in the level of Tipin and vice versa (Fig. 2A), indicating that Timeless and Tipin protein levels are mutually dependent. These cells were treated with hydroxyurea (HU), which depletes the nucleotide pool available for DNA replication, resulting in replication fork arrest. Cells were then released into fresh medium to allow cells to recover from replication arrest and processed for chromosome analysis using pulsed-field gel electrophoresis (PFGE) (Fig. 2B). In PFGE, intact human chromosomes are not able to migrate into the gel. However, when chromosomes are fragmented as a result of DNA damage, shorter pieces of chromosomes can enter the gel (Blocher et al., 1989; Joshi and Grant, 2005). Control siRNA-treated cells did not show an increase in fragmented DNA in response to HU (Fig. 2B). After the release from HU arrest, control cells displayed a mild increase in fragmented DNA at the 5 hour time-point, probably because of minor DNA damage when cells restart replication fork progression. However, the amount of fragmented DNA was not significantly increased later over the course of the experiment (Fig. 2B, 5-20 hours in luciferase siRNA). By contrast, when cells were treated with Timeless siRNA (Tim-#1), cells showed further accumulation of fragmented DNA after release into fresh medium (Fig. 2B, 5-20 hours in Tim siRNA). We obtained similar results with HeLa cells transfected with Tim-#2 or Tip-#1 siRNA (data not shown). These results suggest that HU treatment causes replication fork breakage leading to chromosome fragmentation in the absence of Timeless. They also suggest that cells treated with Timeless siRNA fail to efficiently repair DNA damage caused by HU. Thus, Timeless-Tipin is required for stabilization of stalled replication forks induced by HU treatment.

Open in a separate window

Fig. 2.

Timeless is required for replication fork stabilization. (A) HeLa cells were transiently transfected with the indicated siRNAs (Luc, luciferase; Tim, timeless; Tip, tipin). Levels of timeless (Tim), tipin (Tip) and tubulin (Tub), 24 hours after the transfection are shown, as monitored by western blotting (WB). siRNA depletion of timeless also caused a reduction of the tipin level, and vice versa. Representative results of repeat experiments are shown. (B,C) HeLa cells were transfected with timeless siRNA, then treated with 5 mM HU for 5 hours (B) or 10 μM CPT (C) for 3 hours. Cells were washed and harvested at the indicated times. Cells were counted, and 1×10⁶ cells were embedded in an agarose plug for chromosomal DNA preparation. Chromosomal DNA in plugs was separated by PFGE and visualized using ethidium bromide staining. Downregulation of timeless induced strong accumulation of subchromosomal DNA fragments in response to both HU and CPT treatment. Quantification of DNA damage was shown as relative intensity of fragmented DNA by setting the minimum intensity (no drug in control siRNA cells) to 1. Sp and Sc indicate Genomic DNA from S. pombe and S. cerevisiae as size markers, respectively. Representative results of repeat experiments are shown.

We also treated control and Timeless siRNA-transfected HeLa cells with camptothecin (CPT), a compound that traps topoisomerase I on DNA and effectively induces chromosome breakage at replication forks during S phase (Pommier, 2006). In control cells, CPT treatment induced a mild increase in subchromosomal DNA as visualized by PFGE (Fig. 2C, no CPT and CPT in luciferase siRNA), indicating that CPT treatment caused fork breakage. However, the amount of subchromosomal DNA did not increase significantly after the removal of CPT from the culture medium (Fig. 2C, 5-20 hours in luciferase siRNA). When cells treated with Timeless siRNA (Tim-#1) were treated with CPT and released into fresh medium, fork breakage was further enhanced as evident from a robust increase in the amount of subchromosomal DNA after the removal of CPT (Fig. 2C, 5-20 hours in Tim siRNA). We obtained similar results with Tim-#2 or Tip-#1 siRNA-treated cells (data not shown), suggesting that broken forks were not efficiently repaired in the absence of Timeless-Tipin. Therefore, these results suggest that Timeless-Tipin is involved in prevention of replication fork breakage and/or subsequent recovery of broken replication forks in human cells.

It should also be noted that Timeless siRNA-treated cells showed a reproducible increase in the amount of subchromosomal DNA in the absence of HU or CPT (Fig. 2B,C; no HU/CPT). This result suggests that Timeless-Tipin is also involved in preventing replication fork collapse in the absence of exogenous DNA damaging agents. Consistently, chromosome spread experiments demonstrated a high incidence of chromosome fragmentation in a metaphase of Timeless or Tipin siRNA cells (Fig. 3A). As shown in Fig. 3B, 34.0±6.4% of Timeless RNAi cells and 29.8±5.2% of Tipin RNAi cells displayed chromosome fragmentation, whereas 9.9±1.9% of control (luciferase) cells showed fragmented chromosomes. These data suggest that Timeless-Tipin is also required for stabilization of replication forks during unperturbed DNA replication.

Open in a separate window

Fig. 3.

The timeless-tipin complex is involved in sister chromatid cohesion. (A) HeLa cells were transfected with luciferase (Luc), timeless (Tim) or tipin (Tip) siRNA, grown for an additional 30 hours, treated with colcemid for 90 minutes, incubated in hypotonic buffer, and fixed. Fixed cells were dropped onto a glass slide from a height of 0.75 m and stained with DAPI. Representative images of cells with fragmented chromosomes and cells with cohesion defects are shown. (B) The frequency of different phenotypes shown in A was determined and expressed as percentage of total metaphase cells. At least 100 metaphase cells were counted for each experiment. Average percentages of three independent experiments are shown.

Timeless is required for proper establishment of sister chromatid cohesion

Studies in C. elegans have indicated that the Timeless homolog Tim-1 is associated with the cohesin complex and is involved in chromosome cohesion (Chan et al., 2003). Moreover, in budding and fission yeast, Timeless-Tipin homologs have been reported to be required for proper establishment of sister chromatid cohesion (Ansbach et al., 2008; Mayer et al., 2004; Warren et al., 2004; Xu et al., 2007). However, the mechanism by which Timeless-Tipin homologs regulate this process is unknown. To determine whether human Timeless-Tipin is involved in chromosome cohesion, we used a chromosome-spread method to visualize metaphase chromosomes of HeLa cells treated with Timeless or Tipin siRNA. Cells were incubated with 0.1 μg/ml colcemid for 1.5 hours to increase the population of metaphase cells. Most control (luciferase) cells showed normal chromatid pairing (Fig. 3A), although a small number of cells displayed cohesion defects in the control experiment as reported previously (Parish et al., 2006; Watrin et al., 2006). When Timeless or Tipin was depleted, cells appeared to display loose pairing of sister chromatids, which is indicative of cohesion defects (Fig. 3A). As shown in Fig. 3A (arrows, panels of cohesion defect), pairing sister chromatids were significantly further apart at the centromeric region. Quantification of these results revealed that 8.8±2.7% of control siRNA-treated cells showed cohesion defects, whereas 27.3±11.0% of Timeless siRNA cells and 20.8±5.8% of Tipin siRNA cells displayed cohesion defects (Fig. 3B); suggesting that Timeless-Tipin is involved in proper establishment of sister chromatid cohesion.

To further understand the role of Timeless-Tipin in sister chromatid cohesion, we investigated whether Timeless-Tipin interacts with components of the cohesin complex. The Timeless-FLAG or Tipin-FLAG fusion proteins were overexpressed in HEK293 cells and immunoprecipitated by anti-FLAG antibodies. As shown in Fig. 4A, endogenous Smc1, Smc3 and SA1, which are subunits of the cohesin complex, copurified with Timeless-FLAG or Tipin-FLAG, whereas actin failed to co-precipitate with Timeless or Tipin. To investigate the possibility that Timeless and Tipin interact with cohesin subunits through DNA, we treated cell extracts with DNaseI before immunoprecipitation. In this condition, Smc1 and SA1 still readily co-precipitated with Timeless (Fig. 4B). However, there was no detectable interaction between Smc3 and Timeless (Fig. 4B). Interestingly, Tipin failed to interact with cohesin subunits in this condition (Fig. 4B). These results suggest that Timeless is able to associate with Smc1 and SA1 independently of DNA. They also suggest that Timeless-Smc3 and Tipin-cohesin interactions are dependent on DNA.

Open in a separate window

Fig. 4.

Timeless is involved in stable maintenance of cohesin subunits on chromatin. (A,B) HEK293 cells were transfected with pcDNA3-3FLAG (Vec), pcDNA3-timeless-3FLAG (Tim), or pcDNA-tipin-3FLAG (Tip) and grown for 48 hours. Cell extracts were prepared from these cells in the absence (A) or presence (B) of DNaseI. FLAG-tagged proteins were immunoprecipitated from cell extracts with anti-FLAG antibody. Immunoprecipitates were probed with the indicated antibodies to determine interacting proteins. Representative results of repeat experiments are shown. WCE, whole cell extract; anti-FLAG IP, immunoprecipitated fraction; OE, overexpressed protein. (C) HEK293 cells were transfected with the indicated siRNA, synchronized at the G1-S transition, and released into the fresh medium to allow cells to progress through the cell cycle. Cells were collected at the indicated times and fractionated into Triton-X-100-soluble and - insoluble fractions for protein analyses. Levels of the indicated proteins were determined by western blotting. Representative results of repeat experiments are shown. A, asynchronous. (D) DNA content of cells used in fractionation experiments in C was determined by flow cytometry.

These results suggest that Timeless-Tipin facilitates efficient sister chromatid cohesion by specifically associating with the cohesin complex. Therefore, we investigated whether depletion of Timeless leads to the destabilization of cohesin subunits on chromatin. HEK293 cells treated with control siRNA or Timeless siRNA were synchronized at the G1-S boundary with L-mimosine, released into the cell cycle, and collected at 0, 1, 2, 3, 6 and 9 hours after release (Fig. 4C,D). Cells were fractionated into a Triton-X-100-soluble fraction containing cytosol and nucleoplasm, and a Triton-X-100-insoluble fraction enriched with chromatin- and nuclear matrix-bound proteins as described by Yoshizawa-Sugata and Masai (Yoshizawa-Sugata and Masai, 2007). Actin and CENP-B were exclusively fractionated into Triton-soluble and Triton-insoluble fractions, respectively, indicating that fractionation was successful (Fig. 4C). As shown in Fig. 4C, Smc1, Smc3 and SA1 were recovered in the Triton-insoluble fraction in control siRNA cells. Interestingly, the level of Smc1 in Triton X-100-insoluble fraction was greatly reduced in G1-S and S phase of Timeless-depleted cells compared with control cells (Fig. 4C). There was also a significant reduction in the levels of Smc3 and SA1 during S phase but not G1-S, when Timeless was downregulated. These results suggest that Timeless is involved in the stable association of the cohesin subunits with chromatin during S phase, which in turn regulates proper establishment of sister chromatid cohesion.

Timeless and ChlR1 cooperate to control sister chromatid cohesion

We have previously shown that overproduction of Chl1, a DNA helicase known to have a role in sister chromatid cohesion in yeast and humans (Parish et al., 2006; Petronczki et al., 2004; Skibbens, 2004), can suppress DNA damage sensitivity of the swi1 (a Timeless homolog) deletion mutants in fission yeast (Ansbach et al., 2008). However, how these proteins are involved in cohesion establishment was unknown. Therefore, to determine the mechanisms by which Timeless-Tipin depletion causes defective sister chromatid cohesion, we investigated the interaction between Timeless and ChlR1, a human homolog of Chl1 (Parish et al., 2006). Timeless-FLAG was overexpressed in HEK293 cells and immunoprecipitated. As shown in Fig. 5A, endogenous ChlR1 protein from HEK293 cells was consistently co-precipitated with Timeless-FLAG. We also overexpressed ChlR1-FLAG in HEK293 cells and found that endogenous Timeless co-purified with ChlR1-FLAG (Fig. 5A). The interaction between Timeless and ChlR1 was maintained in DNase-I-treated extracts (data not shown). Thus, Timeless interacts with ChlR1 in human cells. To further examine Timeless and ChlR1 interaction, we fractionated HEK293 cells into Triton-soluble and -insoluble fractions, as described in Fig. 4C. This experiment revealed that ChlR1 was associated with chromatin in control cells (Fig. 4C). However, RNAi-dependent downregulation of Timeless led to the disappearance of ChlR1 (Fig. 4C), suggesting that Timeless is required for stabilization of ChlR1 or maintenance of ChlR1 on chromatin to prevent ChlR1 degradation.

Open in a separate window

Fig. 5.

ChlR1 and timeless-tipin cooperate in establishment of sister chromatid cohesion. (A) ChlR1 associates with timeless. HEK293 cells were transfected with pcDNA3-3FLAG (Vec), pcDNA3-timeless-3FLAG (Tim) or pcDNA3-hChlR1-3FLAG (Chl) and grown for 48 hours. Immunoprecipitation (IP) was performed with the anti-FLAG antibody. Associated proteins were examined by western blotting using the indicated antibodies. Representative results of repeat experiments are shown. WCE, whole cell extract; IP, immunoprecipitated fraction; WB, protein detected by western blotting; OE, overexpressed protein. Representative results of repeat experiments are shown. (B) HeLa cells were transiently transfected with the indicated siRNAs. Cells were collected 48 or 72 hours after transfection, and levels of ChlR1 were monitored by western blotting, using antibodies against ChlR1. Luc, Luciferase. (C) Chromosome spread analysis was performed as described in Fig. 3. Cells treated with tipin siRNA and ChlR1-2 siRNA showed significant cohesion defects. (D) ChlR1 overexpression partially suppressed cohesion defects of timeless or tipin-depleted cells. HeLa cells stably overexpressing ChlR1 or vector alone (Vec) were transfected with the indicated siRNA. Cohesion was evaluated by chromosome spread analysis. At least 50 metaphase cells were counted for each experiment. Error bars correspond to the s.d. obtained from three independent experiments. OE, overexpression.

When HeLa cells were transfected with ChlR1 siRNAs, cells showed sister chromatid cohesion defects consistent with the previous reports (Fig. 5B,C) (Farina et al., 2008; Parish et al., 2006) and comparable with the defects observed in Timeless- or Tipin-depleted cells (Fig. 3B, Fig. 5C). To investigate how Timeless-Tipin and ChlR1 participate in the establishment of sister chromatid cohesion, we generated a HeLa cell line stably overexpressing ChlR1-FLAG. We depleted Timeless or Tipin from these cells and performed chromosome-spread analyses. Overexpression of ChlR1 significantly reduced the cohesion defects caused by depletion of Timeless or Tipin by siRNA (Fig. 5D). Cohesion defects in Tipin-depleted cells were reduced from 36.01% to 18.14% when ChlR1 was overexpressed (P=0.0068 by paired Student's t-test). A similar reduction in cohesion defects was also observed in Timeless-depleted cells in the presence of overproduced ChlR1 (from 29.67% to 24.15%; P=0.0895) (Fig. 5D). These data suggest that Timeless-Tipin and ChlR1 cooperate to regulate proper sister chromatid cohesion to preserve genomic integrity.

Plasmids

Full-length cDNAs against human Timeless, Tipin and ChlR1/DDX11 were amplified by PCR and ligated to 3× FLAG in the pcDNA3 vector, resulting in pcDNA3-Timeless-3FLAG, pcDNA3-Tipin-3FLAG, and pcDNA3-ChlR1-3FLAG, respectively.

Antibodies

Antibodies to Timeless and Tipin were generated by immunizing rabbits with the purified glutathione S-transferase (GST)-fused Timeless C-terminal region (914-1208 amino acids, Timeless-CT) and GST-fused Tipin, respectively. Sera were affinity purified over GST-Timeless-CT or GST-Tipin crosslinked to glutathione-Sepharose using dimethylpimelimidate, as described (Harlow and Lane, 1988). Antibodies to proliferating cell nuclear antigen (PCNA) (PC10), centromere protein B (CENP-B) (2D-7), Cdc45 (M-300) and actin (C-11) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA); SA-1 (ab4455) from Abcam (Cambridge, MA); Smc1 (PC737) from Calbiochem (San Diego, CA); Smc3 (BL313) from Bethyl Laboratory (Montgomery, TX); ChlR1/DDX11 from Novus Biologicals (Littleton, CO); FLAG (M2) from Sigma (St Louis, MO); and BubR1 from BD Biosciences (San Jose, CA).

siRNA

Transfection of small interfering RNA (siRNA) duplexes was performed by using Oligofectamine (Invitrogen, Carlsbad, CA) as recommended by the supplier. siRNA oligonucleotides were purchased from Invitrogen. The sense strands of siRNA oligonucleotides for Timeless (Tim-#1, Tim-#2), Tipin (Tip-#1) and control (luciferase) were reported previously (Unsal-Kacmaz et al., 2007; Yoshizawa-Sugata and Masai, 2007). The sequences of ChlR1 siRNA are as follows: ChlR1-#1, 5′-UCC UGC AUG GCU GAG AGC CAG GCU U-3′; ChlR1-#2, 5′-CCA ACU GGC ACU GGG AAG UCC UUA A-3′; ChlR1-#3, 5′-CCU GUG UCU GUC UUC UUC CUG CGA A-3′.

Chromatin immunoprecipitation

Chromatin immunoprecipitation (ChIP) assays were performed as described elsewhere (Nelson et al., 2006) with modification. Briefly, cells were fixed in culture medium with 1.42% formaldehyde for 15 minutes. The cells were then quenched with 125 mM glycine for 5 minutes, and washed twice in PBS and collected by centrifugation at 2000 g at 4°C. Cells were resuspended in ChIP lysis buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 5 mM EDTA, 0.5% NP-40, 1.0% Triton-X-100, 0.2 mM p-APMSF (4-amidinobenzylsulfonyl fluoride hydrochloride), and Roche Complete EDTA-free protease inhibitor cocktail), incubated on ice for 10 minutes, and sonicated using a Misonix Sonicator 3000 (Misonix, Farmingdale, NY) until chromatin DNA was sheared into 500-700 bp fragments. Cell lysates were clarified by maximum-speed centrifugation in an Eppendorf 5415D centrifuge at 4°C. Immunoprecipitations were performed in the cell extracts using either affinity-purified anti-Timeless, Cdc45 antibodies or Rabbit IgG in combination with Protein-A-Sepharose. Immunoprecipitated DNA and input DNA were recovered using Chelex-100 resin (Bio-Rad) as described (Nelson et al., 2006). Recovered DNA was analyzed by triplicate SYBR Green-based real-time PCR (Bio-Rad) using primers that are designed to amplify the MYC origin, MYC 11-kb proximal, HBB origin, HBB 31 kb proximal, MCM4 origin, and ATCG1 gene regions. Primer sequences were reported previously (Schaarschmidt et al., 2002; Sibani et al., 2005b; Tan et al., 2006). Raw percentage precipitated DNA values (percentage raw precipitation) were calculated based on ΔCT between input and immunoprecipitated samples and corrected for PCR efficiency. To obtain percentage precipitation values, raw percentage precipitated DNA values of IgG control samples were subtracted from percentage raw precipitation values of Timeless or Cdc45 ChIP samples.

Immunoprecipitation

HEK293 cells were transfected with pcDNA3-3FLAG, pcDNA3-Timeless-3FLAG, pcDNA3-Tipin-3FLAG, or pcDNA3-ChlR1-3FLAG with TransIT LT-1 Transfection Reagent (Mirus Bio, Madison, WI) according to the manufacturer's protocol and harvested 48 hours after transfection. Cells were washed twice in ice-cold PBS and lysed in immunoprecipitation (IP) lysis buffer (50 mM Tris-HCl, pH 8.0, 150 mM NaCl, 10% glycerol, 0.1% NP-40, 50 mM NaF, 1 mM Na₃VO₄, 5 mM EDTA, 5 mM N-methylmaleimide, 1 μM okadaic acid, 0.2 mM p-APMSF, and Roche Complete EDTA-free protease inhibitor cocktail) by Branson Digital Sonifier (Danbury, CT) for eight cycles of 2 seconds at output 9%, with a 5 second interval on ice between each cycle. Protein extracts were clarified by maximum-speed centrifugation in an Eppendorf 5415D centrifuge for 10 minutes at 4°C. When needed, protein extracts were incubated with 5 U of DNase I at room temperature for 1 hour, followed by 3 hours at 4°C. Protein extracts were then mixed with anti-FLAG M2 agarose (Sigma), Protein-A/G-Sepharose (Sigma), or Protein-G dynabeads (Invitrogen) and incubated for 2 hours at 4°C. The beads were collected and washed three times in IP lysis buffer. Proteins associated with the beads were analyzed by western blotting.

Cell fractionation

Cell fractionation was performed as described elsewhere (Yoshizawa-Sugata et al., 2005). Cells were synchronized in S phase by L-mimosine as described above, rinsed twice and released into growth medium. Harvested cell were incubated in cell fractionation (CF) buffer (20 mM PIPES, pH 6.8, 400 mM NaCl, 1 mM MgCl₂, 1 mM EGTA, 300 mM sucrose, 1 mM dithiothreitol, 1 mM Na₃VO₄, and Roche Complete EDTA-free protease inhibitor cocktail) containing 0.1% Triton X-100 for 20 minutes on ice. Triton-soluble and -insoluble fractions were separated by centrifugation in an Eppendorf 5415D centrifuge at 800 g for 4 minutes at 4°C. The insoluble fraction was washed twice in CF buffer and resuspended in the same buffer. The soluble fraction was clarified by maximum speed centrifugation. Fractions were analyzed by western blotting.

Pulsed-field gel electrophoresis

Preparation and pulsed-field gel electrophoresis (PFGE) of chromosomal DNA were performed as described (Blocher et al., 1989; Joshi and Grant, 2005), using CHEF-DR II system (Bio-Rad) at the following settings. Block 1: field strength, 1.9 V/cm; initial and final switch times, 30 and 120 seconds, respectively; running temperature, 14°C; pump speed, 70; running time, 30 hours. Block 2: field strength, 1.9 V/cm; initial and final switch times, 120 seconds and 42 minutes, respectively; running temperature, 14°C; pump speed, 70; running time, 51 hours. Gels were stained with 0.5 μg/ml ethidium bromide in water for 30 minutes and destained in water for 1 to 2 hours. DNA amount in each lane was quantified using EZ Quant-Gel software (EZQuant LTD, Tel-Aviv, Israel).

In situ cell fractionation, immunofluorescence and chromosome spreads

In situ fractionation of HeLa cells and immunofluorescence were performed as described (Mirzoeva and Petrini, 2001). Chromosome spreads were conducted as described elsewhere (Henegariu et al., 2001).

This work is supported by funds from Pennsylvania Department of Health (to E.N.), W. W. Smith Charitable Trust (to E.N.), National Institutes of Health (GM077604, to E.N.) and Drexel University College of Medicine. We thank Joseph Takahashi for generously providing the human Timeless cDNA; Peter Adams, Jane Clifford, Dale Haines, Christian Sell, Mauricio Reginato and Laura Roseaulin and anonymous reviewers for helpful comments; Mark Lechner, Taotao Rao, Ian Klansek and Michelle Villasmil for technical assistance. Deposited in PMC for release after 12 months.