Replication stress induces specific enrichment of RECQ1 at common fragile sites FRA3B and FRA16D

Silencing of RECQ1 by siRNA leads to reduced cell proliferation [37, 40, 41]. A key step in the regulation of cell proliferation is initiation of replication. DNA replication initiates from a defined initiation site, from which replication progresses in both directions. One of the best characterized metazoan origins is located at the 3’ end of the human lamin B2 gene (Figure 1A) [42]. RECQ1 binds to the lamin B2 origin in unperturbed cells and knockdown of RECQ1 results in reduced origin firing and defective replication elongation [34]. To test whether RECQ1 plays a role in mammalian fork progression after initiation, we investigated the effects of fork stalling on binding of RECQ1 to lamin B2 origin in HeLa cells by ChIP (Figure 1B, C). ChIP experiments were done using cells that were either untreated or treated with hydroxyurea which specifically inhibits class I ribonucleotide reductase depleting dNTP pools [43, 44] or aphidicolin which is an inhibitor of DNA polymerase α, δ and ϵ [45, 46] to induce replication forks arrest. Thus, replication stress induced by hydroxyurea (2 mM, 24 h) and aphidicolin (0.5 μM, 24 h) treatment is mechanistically different. Cross-linked chromatin was immunoprecipitated with a control IgG or a specific antibody against RECQ1 that has been successfully used to identify protein and chromatin interactions of RECQ1 (Figure 1B) [34, 47]. Immunoprecipitation of ORC2, a known lamin B2 origin binding protein and a part of the pre-replication complex [42], served as positive control in these experiments. Following cross-link reversal, the immunoprecipitated chromatin was used to determine the lamin B2 origin-containing DNA as well as an adjacent region, B13 that does not contain origin by quantitative real time PCR (qPCR). ORC2 interacted with lamin B2 origin, and approximately 4-, 13- and 18-fold enrichment of lamin B2 origin-specific DNA was found in ORC2 immunoprecipitate as compared to IgG in untreated, hydroxyurea, or aphidicolin treated HeLa cells, respectively (Figure 1B). Lamin B2 origin-specific DNA was enriched nearly 6-fold in RECQ1-immunoprecipitate as compared to IgG in untreated HeLa cells (Figure 1B). Treatment with replication inhibitors induced a further enrichment of RECQ1 at lamin B2 origin; approximately 11- and 30-fold enrichment of lamin B2 origin specific DNA was obtained in RECQ1 immunoprecipitate as compared to IgG in cells treated with hydroxyurea or aphidicolin, respectively (Figure 1B). In comparison, minimal binding to the control B13 containing DNA was observed for ORC2 or RECQ1-immunoprecipitates (Figure 1B, C). Thus, RECQ1 binds to the lamin B2 origin of replication in unperturbed HeLa cells and replication stress significantly enhances origin binding of RECQ1.

RECQ1 protein is expressed throughout the cell cycle and is not regulated in cell cycle stage specific manner [48]; although a small increase is reported in T98G cells synchronized in S-phase using serum starvation or mimosine treatment [34]. We monitored cell cycle progression in HeLa cells utilized in ChIP assay by FACS analysis (Figure 1D) and also determined total cellular RECQ1 protein level (Figure 1E). Total RECQ1 protein level was not modulated appreciably in HeLa cells following treatment with hydroxyurea or aphidicolin (Figure 1E). We next examined cellular RECQ1 in chromatin containing fractions following treatment with agents that introduce replication fork blocking lesions. HeLa cells either untreated or treated with hydroxyurea (2 mM, 24 h) or aphidicolin (0.5 μM, 24 h) were subjected to detergent extraction to isolate insoluble nuclear pellet containing proteins that were tightly bound to chromatin and/or nuclear matrix. RECQ1 protein in untreated cells predominantly fractionated with soluble proteins and only a minor fraction associated with chromatin in the insoluble nuclear pellet (Figure 1F). Interruption of DNA synthesis by hydroxyurea or aphidicolin resulted in increased RECQ1 in the insoluble fraction that also contained histones (Figure 1F). Apparently, increased chromatin association of RECQ1 was also seen following treatment with the DNA inter-strand crosslinking agent mitomycin C (0.5 μg/ml, 24 h) that induces double strand breaks on encounter with progressing replication fork (Figure 1F). When replication fork progression was inhibited by incubating the cells with aphidicolin prior to mitomycin C treatment, chromatin-bound RECQ1 signal was reduced as compared to mitomycin C alone but was greater than the untreated cells (Figure 1F). Thus, stalled and collapsed replication forks induce re-localization of endogenous RECQ1 to the chromatin.

Replication stress particularly affects genomic loci where progression of replication forks is slow or problematic [2]. To test a putative role of RECQ1 in promoting fork recovery or repair, we examined whether RECQ1 is recruited to CFS since aphidicolin treatment introduces stalled replication forks at fragile sites [5]. In order to determine whether RECQ1 is recruited to stalled replication forks induced at the FHIT region in the aphidicolin-sensitive fragile site FRA3B, HeLa cells were either untreated or treated with 0.5 μM aphidicolin for 24 h. The cross-linked chromatin prepared from each condition was then processed for ChIP by using either control IgG or a specific RECQ1 antibody. Immunoprecipitation using a specific antibody against phosphorylated H2AX (γH2AX) which is known to bind across the FHIT region of FRA3B served as positive control [18, 49]. To determine whether RECQ1 occupies FRA3B locus, primers specific for two separate regions in FRA3B fragile locus including the distal aphidicolin induced breakpoints cluster (FDR) located within intron 4 of the FHIT gene were used (Figure 2A) [49, 50]. As shown in Figure 2B, RECQ1 or γH2AX did not bind the FRA3B locus in untreated cells but were recruited to this fragile site when cells were exposed to aphidicolin. Treatment with aphidicolin induced enrichment of RECQ1 and γH2AX at FRA3B; approximately 4.5- and 6.5-fold enrichment of FDR-specific DNA was obtained in RECQ1 and γH2AX immunoprecipitate as compared to IgG, respectively (Figure 2B, C). RECQ1 and γH2AX immunoprecipitate also contained a relatively modest, but reproducible, enrichment of FCR-specific DNA over IgG, respectively (Figure 2B, C). Thus, aphidicolin treatment induced an increase in RECQ1 occupancy at the FRA3B fragile site in HeLa cells. ChIP experiments using hyroxyurea treated cells (2 mM, 24 h) revealed nearly 2.7- and 3.6-fold enrichment of FDR and FCR-specific DNA in RECQ1 immunoprecipitate as compared to IgG whereas γH2AX immunoprecipitate displayed 3- and 2-fold enrichment of FDR and FCR-specific DNA, respectively (Figure 2D). The relatively reduced binding observed here is likely due to the fact that hydroxyurea and other inhibitors are less specific than aphidicolin in inducing lesions primarily at fragile sites [5].

To ascertain preferential binding of RECQ1 to CFS, we further analyzed RECQ1 immunoprecipitates by qPCR for the enrichment of DNA corresponding to FRA16D, the second most active and aphidicolin-sensitive fragile site in the human genome (Figure 3A) [51]. As control, we also examined two non-CFS DNA sequences located within the GAPDH and β-actin, respectively (Figure 3B, D) [18]. ChIP from untreated HeLa cells did not show enrichment of FRA16D-specific DNA in RECQ1 immunoprecipitate relative to IgG suggesting that RECQ1 does not occupy FRA16D fragile locus in unstressed replicating cells (Figure 3B). Treatment of HeLa cells with aphidicolin induced recruitment of RECQ1 at the FRA16D locus; approximately 14-fold enrichment of FRA16D-specific DNA was obtained in RECQ1 immunoprecipitate as compared to IgG (Figure 3B). In contrast, only 1.5- and 1.8-fold enrichment of control sequence spanning GAPDH was obtained in RECQ1 immunoprecipitate relative to IgG (Figure 3B, C). As compared to IgG, ChIP of RECQ1, γH2AX, or ORC2 did not show enrichment of control sequence spanning β-actin in untreated or aphidicolin treated cells indicating the specificity of respective antibodies used in these experiments (Figure 3D, E).

Collectively, these experiments demonstrate preferential and specific binding of RECQ1 to the two well characterized aphidicolin-sensitive fragile sites FRA3B and FRA16D in response to replication stress.

We previously reported that siRNA mediated depletion of RECQ1 renders HeLa cells more sensitive to camptothecin, an anti-tumor drug that inhibits the topoisomerase-induced DNA breakage-reunion reaction resulting in DNA double strand breaks at stalled replication forks [37]. Given the enrichment of RECQ1 at aphidicolin-sensitive fragile sites, we postulated that RECQ1 may be important in the cellular resistance to aphidicolin. Control or RECQ1 siRNA-transfected cells were exposed to increasing concentrations of aphidicolin and their survival was measured 72 h later by MTS assay. In multiple experiments, RECQ1-depleted HeLa cells displayed increased and dose-dependent sensitivity to aphidicolin when compared to control siRNA-transfected cells (Figure 4A). Increased aphidicolin sensitivity of RECQ1-depleted cells was also observed when proliferation was assayed by cell counting (Figure 4B). This is consistent with the reported sensitivity of RECQ1-depleted cells to HU, and other replication blocking agents [35].

Increased sensitivity to genotoxic agents may be caused not only by defects in DNA repair but also by disruption of cell cycle checkpoint responses to DNA damage [52]. ATR-dependent Chk1 phosphorylation is induced to stabilize replication complexes on stalled forks in cells treated with aphidicolin or hydroxyurea and the RPA coated single-stranded DNA is considered to be the predominant signal for activation of the ATR-Chk1 signal transduction pathway [53]. We examined by Western blotting Chk1 phosphorylation at Ser317 that is central to the normal DNA damage response to replication stress [54]. As compared to control siRNA transfected cells, RECQ1-depleted HeLa cells exhibit constitutively activated Chk1 as shown by phosphorylation at Ser317 in the absence of exogenously induced stress (Figure 4C). Appearance of RPA32 phosphorylation in undamaged RECQ1-depleted cells suggests that the checkpoint response is triggered by DNA structures generated during replication such as collapsed replication forks, long stretches of single-stranded DNA, or unresolved complex intermediates in RECQ1-deficiency. Pharmacological induction of replication stress revealed aberrant checkpoint activation in RECQ1-depleted cells (Figure 4C). Hydroxyurea treatment (2 mM, 24 h) resulted in the phosphorylation of Chk1 at Ser317 and RPA32 in control siRNA transfected HeLa cells (Figure 4C). In contrast, hydroxyurea treatment of RECQ1-depleted cells displayed consistently attenuated phosphorylation of Chk1 at Ser317 and reduced RPA32 phosphorylation (Figure 4C). Reduced Chk1 phosphorylation was also observed in response to aphidicolin treatment (0.5 μM, 24 h) in RECQ1-depleted cells as compared to control HeLa cells (Figure 4C). RPA32 phosphorylation was not evident in control HeLa cells upon arresting replication with aphidicolin [55] and a minimal increase in the signal for RPA phosphorylation was seen in RECQ1-depleted cells as compared to their untreated condition (Figure 4C). We observed a modest increase in signal for γH2AX in untreated and aphidicolin treated RECQ1-depleted cells as compared to control siRNA transfected cells (Figure 4C).

Partial inhibition of DNA synthesis by aphidicolin results in incomplete replication in late replicating chromosomal areas such as fragile sites leading to visible breaks in metaphase chromosomes [56]. We therefore examined whether RECQ1-depletion increases the basal and aphidicolin-induced metaphase chromosome breaks. Depletion of RECQ1 led to a significant increase in the number of chromosome breaks per cell even in the absence of aphidicolin when compared to the control siRNA (Figure 4D). This is consistent with spontaneous chromosome breaks and gaps observed in RECQ1 knockout mouse embryonic fibroblasts [36] and a recent report using stable knockdown of RECQ1 in human cells [35].

Given the demonstrated ability of RECQ1 to act upon replication fork like structures in vitro [33, 35], these results indicate that depletion of RECQ1 as such may contribute to replication fork collapse and also diminish recovery from replication arrest.

Human HeLa (ATCC) were grown in Dulbecco's modified Eagle's medium (DMEM) (Invitrogen) supplemented with 10% fetal bovine serum (Hyclone Laboratories), 100 U/ml penicillin and 100 μg/ml streptomycin (Invitrogen). Cells were grown in a humidified 5% CO₂ incubator at 37°C. To induce replication stress, exponentially growing cells were treated with hydroxyurea (Sigma), aphidicolin (Calbiochem), mitomycin C (Sigma) as indicated. Depletion of RECQ1 was achieved by transfecting HeLa cells with a scrambled control siRNA or a SMARTpool of four distinct siRNA species targeting different regions of RECQ1 mRNA (siGenome SMARTpool, Dharmacon) at a final concentration of 10 nM using Lipofectamine 2000 transfection reagent as per the manufacturer’s instructions (Invitrogen). Knockdown specificity and efficacy of RECQ1 smartpool siRNA was evaluated by real-time PCR and Western blotting.

HeLa cells were cultured overnight at a density of 1 × 10⁷ per 15 cm diameter dish and subjected to either no treatment or treatment with 0.5 μM aphidicolin for 24 h. Chromatin and proteins were cross-linked by incubating cells in 1% formaldehyde for 15 min at room temperature and the reaction was stopped by 10 min incubation with 125 mM glycine. Cells were collected and washed sequentially with solution I (10 mM HEPES [pH 7.5], 10 mM EDTA, 0.5 mM EGTA, 0.75% Triton X-100) and solution II (10 mM HEPES [pH 7.5], 200 mM NaCl, 1 mM EDTA, 0.5 mM EGTA). The cell pellets were resuspended in 2ml lysis buffer (25 mM Tris-HCl [pH 7.5], 150 mM NaCl, 0.1% SDS, 1% Triton X-100, 0.5% deoxycholate freshly supplemented with 1X protease inhibitor cocktail (Roche)) and sonicated on ice by 10 s pulses at 30% of maximal power on a Misonix 2000 sonicator (Misonix). This sonication method consistently yielded chromatin fragments corresponding to an average DNA length of 400-1000 bp as checked on a 1.5% agarose gel. After centrifugation at 20,000X g for 15 min to remove any debris, the supernatant was pre-cleared with protein-G-sepharose/salmon sperm DNA beads (Millipore) at 4°C for 1 h. For each immunoprecipitation, 600 μl (equivalent of 3 × 10⁶ cells) of the pre-cleared chromatin was incubated overnight at 4°C with 3 μg of antibodies specific for either RECQ1 (Bethyl Lab, A300-450A), γH2AX (Millipore, 05-636, clone JBW301), or ORC2 (Enzo Life Science, ADI-KAM-cc235); antibodies were confirmed for their immunoprecipitation specificity using Western blot. A reaction containing an equivalent amount of rabbit IgG was included as the background control. 10% of the pre-cleared chromatin was set aside as input control. Antibody-chromatin complexes were pulled down by adding 50 μl of protein-G-sepharose/salmon sperm DNA beads and incubated for 2 h at 4°C. The beads were washed for 10 min each with the following solutions: lysis buffer (as mentioned above), high-salt wash buffer (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris-HCl [pH 8.1], 500 mM NaCl), LiCl wash buffer (250 mM LiCl, 1% NP-40, 1% deoxycholate, 1 mM EDTA, 10 mM Tris-HCl [pH 8.0]), and TE buffer (10 mM Tris-HCl [pH 8.0], 1 mM EDTA). Finally, DNA was eluted with elution buffer (1% SDS, 100 mM NaHCO₃). Eluates were incubated at 65°C for overnight with the addition of 5 M NaCl to a final concentration of 200 mm to reverse the formaldehyde cross-linking and digested at 55°C for 3 h with proteinase K at a final concentration of 50 μg per ml. Following phenol/chloroform extraction and ethanol precipitation, sheared DNA fragments served as template in qPCR analysis. qPCR were performed using Taq Universal SYBR Green Supermix (Bio-Rad) with technical triplicates, and threshold cycle numbers (C_t) were determined with an iQ5 thermal cycler (Bio-Rad). Fold enrichment of the targeted genomic sequences were calculated over IgG as: fold enrichment = 2^-(Ct_IP^{− Ct}_IgG⁾, where Ct_IP and Ct_IgG are mean threshold cycles of PCR done in triplicates on DNA samples immunoprecipitated with specific antibody and control IgG, respectively. All qPCR reactions were also checked by melt curve analyses and agarose gel electrophoresis to confirm the presence of a single specific product. The sequences of the qPCR primers are listed in Table 1.

Cells were collected by trypsinization and fixed with 70% ethanol for 30 min. Cells were subsequently washed in PBS, treated with RNase in PBS, and resuspended in PBS with 4 μg/ml propidium iodide. Cells were analyzed on a FACSCalibur™ flow cytometer from BD Science using ModFit LT software.

Chromatin enriched fractions were prepared as previously described [47] from cells that were either untreated or treated with indicated concentration of hydroxyurea or aphidicolin. To obtain total extracts, cells were lysed in RIPA buffer (50 mM Tris-HCl [pH 8.0], 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS) containing cocktail of protease and phosphatase inhibitors (Roche). Equal amounts of total protein for each sample was run on 8-16% SDS-PAGE and transferred to PVDF membrane for immunoblotting with antibodies for phospho-Chk1 Ser 317 (Cell Signaling, 2344), γH2AX (Cell Signaling, 2577), RPA32 (Bethyl Lab, A300-244A), RECQ1 (Bethyl Lab, A300-450A), GAPDH (Cell Signaling, 2118), and Histone H3 (Cell Signaling, 4499); all antibodies were used at 1:1000 dilution.

Cells (24 h after siRNA transfection) were seeded in quadruplicate at a density of 3000 cells/well in 96-well plates and allowed to adhere for 16 h and subsequently exposed to increasing concentration of aphidicolin (0-5 μM) and allowed to grow at 37°C for 3 days in 5% CO₂. Cell proliferation was determined by CellTiter 96® AQueous Non-Radioactive Cell Proliferation Assay (MTS) (Promega). Additionally, cells were transfected with the indicated siRNA, split 1:4 onto a multiwell plate, and grown in the presence or absence of aphidicolin for further 72 h. The cells in one well were counted every 24 h for the duration of the experiment.

Cells, 48 h after transfection, were grown in the presence or absence of aphidicolin (0.5 μM) for 24 h and treated with 0.5 μg/ml colchicine for 4 h at 37°C before collection. To prepare metaphase spreads, cells were resuspended in hypotonic solution (0.06 M KCl) for 15 min at room temperature, and then fixed with 3:1 (vol/vol) methanol-glacial acetic acid. Fixed cell suspension was dropped onto precleaned microscope slides and air-dried overnight. Metaphase chromosomes were visualized by Giemsa staining. Images were documented with a Nikon microscope and at least 10 metaphase chromosome spreads per treatment were scored in a blinded fashion.