Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Pyrimidine pool imbalance induced by BLM helicase deficiency contributes to genetic instability in Bloom syndrome

Pyrimidine pool imbalance induced by BLM helicase deficiency contributes to genetic instability... ARTICLE Received 31 Jan 2011 | Accepted 23 may 2011 | Published 28 Jun 2011 DOI: 10.1038/ncomms1363 Pyrimidine pool imbalance induced by BLm helicase deficiency contributes to genetic instability in Bloom syndrome 1,2 1,2, 1,2, 1,2 Pauline Chabosseau , Géraldine Buhagiar-Labarchède *, Rosine onclercq-Delic *, sarah Lambert , 3,4 3,4 1,2 michelle Debatisse , olivier Brison & mounira Amor-Guéret Defects in DnA replication are associated with genetic instability and cancer development, as illustrated in Bloom syndrome. Features of this syndrome include a slowdown in replication speed, defective fork reactivation and high rates of sister chromatid exchange, with a general predisposition to cancer. Bloom syndrome is caused by mutations in the BLM gene encoding a RecQ helicase. Here we report that BLm deficiency is associated with a strong cytidine deaminase defect, leading to pyrimidine pool disequilibrium. In BLm-deficient cells, pyrimidine pool normalization leads to reduction of sister chromatid exchange frequency and is sufficient for full restoration of replication fork velocity but not the fork restart defect, thus identifying the part of the Bloom syndrome phenotype because of pyrimidine pool imbalance. This study provides new insights into the molecular basis of control of replication speed and the genetic instability associated with Bloom syndrome. nucleotide pool disequilibrium could be a general phenomenon in a large spectrum of precancerous and cancer cells. 1 2 Institut Curie, Centre de Recherche, Orsay, France. CNRS UMR 3348, Stress Génotoxiques et Cancer, Centre Universitaire, Bât. 110, 91405 Orsay, France. 3 4 CNRS UMR 3244, Institut Curie, Centre de Recherche, Paris, France. Université Pierre et Marie Curie, 26 rue d’Ulm, 75248 Paris cedex 05, France. *These authors contributed equally to this work. Correspondence and requests for materials should be addressed to M.A.-G. (email: [email protected]). nATuRE C ommunICATIons | 2:368 | DoI: 10.1038/ncomms1363 | www.nature.com/naturecommunications © 2011 Macmillan Publishers Limited. All rights reserved. ARTICLE nATuRE C ommunICATIons | DoI: 10.1038/ncomms1363 loom syndrome (BS) is a rare autosomal recessive disorder CDA gene expression (Fig. 1a), demonstrating the requirement of characterized by marked genetic instability associated with BLM helicase activity for CDA gene expression. We inserted the Bpredisposition to a wide range of cancers commonly ae ff ct - human CDA promoter into a luciferase vector and showed that the ing the general population . BS is caused by mutations in the BLM CDA promoter was significantly less active in the absence of BLM gene, which encodes BLM, a RecQ 3′–5′ DNA helicase, indicating or of its helicase activity (Fig. 1b), demonstrating the requirement an essential role for BLM in maintaining genetic stability and pre- of BLM helicase activity for correct CDA expression and, thus, venting cancer . In vitro, BLM unwinds DNA structures mimick- the direct or indirect involvement of this activity in regulation of ing replication forks and homologous recombination intermedi- the CDA promoter. ates . The specific functions of BLM remain unclear, but it is widely CDA has an essential role in maintaining the cellular pool of thought that BLM is involved in restarting blocked replication nucleotides, which is crucial for genome stability. Indeed, an imbal- 4,5 forks . In the absence of BLM, cells display a high rate of sister ance in the nucleotide pools is highly mutagenic in mammalian 6 16 6,17 chromatid exchanges (SCEs), pathognomonic for BS . BS cells also cells , and BS cells display a mutator and hyper-Rec phenotype . display a slowing of replication fork progression associated with e Th purine pools were not ae ff cted in BS cells, whereas the pyrimi - 4,5 endogenous activation of the ATM-Chk2-γH2AX pathway . This dine pools were significantly unbalanced ( Fig. 1c). Indeed, BS cells pathway is activated in precancerous lesions and is thought to be had about twice as much dCTP and a slightly lower level of dTTP 7,8 a part of an antitumorigenesis barrier . This barrier is based on than GFP-BLM cells (Fig. 1c). Moreover, expression of the GFP- DNA replication stress, leading to activation of the DNA damage BLMI841T non-functional protein in BS cells, which was not able 7,8 checkpoint and, thus, apoptosis or cell cycle arrest . BS cells may to rescue CDA gene expression (Fig 1a), also failed to normalize the therefore be in a precancerous state. Our working hypothesis is dCTP pool size (Supplementary Fig. S1). that BLM deficiency results in a global ‘SOS-like’ cellular response, facilitating progression through the antitumorigenesis barrier CDA expression partially rescues cellular BS phenotype. Stable potentially involved in carcinogenesis in the general population . expression of the CDA gene in BS cells (BS-CDA) was sufficient u Th s, identification of the genes and pathways deregulated in the to restore a dCTP pool similar to that of GFP-BLM cells (Fig. 2a), absence of BLM may facilitate the establishment of a molecular demonstrating the direct responsibility of the CDA defect for signature of the precancerous state. the pyrimidine pool imbalance observed in BLM-deficient cells. Deoxyribonucleoside triphosphate (dNTP) levels are also of CDA expression in BS cells also decreased the frequency of SCE major importance for maintaining genomic integrity. Indeed, by 25% (P = 0.0015, Student’s t-test), reflected in a great decrease nucleotide pool imbalance, particularly for the pyrimidine pools, in the number of cells with high levels of SCE (Fig. 2b), suggest- is known to result in genetic instability and oncogenic transforma- ing a direct relationship between the high rate of SCE observed in 10,11 tion . Pyrimidine biosynthesis occurs through the de novo path- BS cells and nucleotide pool disequilibrium. Addition of 500 µM way, in which nucleotides are synthesized from small metabolites, or dC to the culture medium of GFP-BLM cells led to a 1.75 times the salvage pathway, which recycles nucleosides and nucleotides . increase in dCTP pool size (Fig. 2c) and ~42% increase in SCE In mammalian cells, dec fi iencies of many of the enzymes involved in frequency (P = 0.0327, Student’s t-test; Fig. 2d), confirming that the two pyrimidine biosynthesis pathways, such as thymidine kinase, nucleotide pool disequilibrium promotes an increase in SCE fre- dTMP synthase, dCTP kinase, CTP synthase and dCMP deami- quency. CDA expression in BS cells also increased colony-forming nase, have been associated with the induction of mutations, DNA efficiency significantly ( Fig. 2e). Moreover, CDA expression in BS breaks, sensitization to DNA-damaging agents, increase in the rate cells was sufficient to increase replication fork speed, by up to 10,11 of homologous recombination and chromosomal abnormalities . 24%, restoring fork velocity to levels similar to those in GFP-BLM We report here that BLM deficiency leads to cytidine deaminase cells (Fig. 2f ). These results suggest that the slowdown of replica - (CDA) downregulation, resulting in pyrimidine pool disequilib- tion speed in BS cells could be the consequence of the pyrimidine rium, accounting for the slowing of replication fork progression and pool imbalance. contributing to the increase in SCE frequency associated with the BS phenotype. CDA downregulation reproduces part of the BS phenotype. e Th short interfering RNA (siRNA)-mediated downregulation of Results CDA in GFP-BLM cells increased the frequency of SCE by 46% BLM deficiency is associated with a strong CDA defect . For the (P = 0.0134, Student’s t-test), reflecting a significant increase in the identification of genes specifically deregulated in the absence of number of cells with a high SCE frequency (Fig. 3a). By contrast, BLM, we performed a microarray analysis comparing the transient transfection of BS cells (with no detectable CDA expres- transcriptome of wild-type HeLa cells with that of HeLa cells in sion) with siRNAs specific for CDA had no ee ff ct on SCE level which BLM was downregulated , and the transcriptome of the (Supplementary Fig. S2). CDA downregulation in HeLa cells and GM8505B BS cell line (BS) with that of its BLM-complemented in BLM-downregulated HeLa cells (in which CDA downregulation counterpart (green uo fl rescent protein (GFP)-BLM). We identified was not complete as seen in Fig. 1a) also significantly increased 77 candidate genes for which regulation was potentially modified the number of cells with a high SCE frequency (Fig. 3b). Moreo- in the same way in both HeLa cells depleted of BLM and in BS ver, CDA downregulation significantly decreased replication fork cells (Supplementary Table S1). One of these genes, encoding CDA velocity in GFP-BLM cells, but not in BS cells (Fig. 3c). us Th , the (EC 3.5.4.5; locus 1p36.2-p35), particularly attracted our attention. level of CDA gene expression inu fl ences both SCE frequency and CDA is an enzyme of the pyrimidine salvage pathway catalysing the speed of the replication fork, CDA downregulation being asso- the hydrolytic deamination of cytidine (C) and deoxycytidine (dC) ciated with an increase in the level of SCE and a decrease in the to uridine (U) and deoxyuridine (dU), respectively . We confirmed replication fork velocity. that CDA mRNA and protein levels were strongly reduced in BS cells and in BLM-downregulated HeLa cells, and in HEK 293 human Addition of dU partially rescues BS phenotype. Finally, we found embryonic kidney cells depleted of BLM (Fig. 1a). e Th expression of that the addition of 25 µM dU (the final CDA product) or thymidine a functional BLM protein (GFP-BLM) in BS cells was sufficient to (T) to the culture medium led to ~40% decrease in SCE frequency in restore CDA expression (Fig. 1a). By contrast, expression of a BLM BS cells (Fig. 4a). As the addition of dU at this concentration had no protein with an inactive helicase domain but a functional DNA- effect on residual SCE frequency in GFP-BLM cells or on the qual - binding domain (GFP-BLMI841T) in BS cells failed to rescue ity of the differential labeling of sister chromatids in either cell line, nATuRE C ommunICATIons | 2:368 | DoI: 10.1038/ncomms1363 | www.nature.com/naturecommunications © 2011 Macmillan Publishers Limited. All rights reserved. n ATu RE Commun ICATIons | Do I: 10.1038/ncomms1363 ARTICLE P<0.0001 P<0.0001 P=0.0002 1.2 1.2 1.0 kDa 0.8 0.8 0.60 0.56 170 - BLM 0.6 16 - CDA 0.4 0.4 42 - β-Actin 0.2 0.008 0.0166 0 0 BS GFP-BLM GFP-BLMI841T GFP-BLM BS GFP-BLMI841T P=0.0069 P<0.0001 P=0.0299 1.2 7.45 1 9 60 1.0 46.85 kDa 38.75 0.8 6 40 170 - BLM 3.84 0.6 0.456 16 - CDA 0.4 3 20 0.2 β-Actin 42 - 0 0 0 HeLaV-siCtrl HeLash-siBLM BS GFP-BLM BS GFP-BLM P<0.0001 1.2 9 9 1.0 4.67 4.65 kDa 0.8 6 6 0.6 BLM 170 - 0.4 16 - 3 CDA 1.07 0.109 0.97 0.2 42 - β-Actin 0 0 0 HEK293V-siCtrl HEK293sh-siBLM BS GFP-BLM BS GFP-BLM Figure 1 | BLM dec fi iency is associated with CDA dec fi iency and pyrimidine pool imbalance. (a) CDA mRn A and protein levels determined by real- time quantitative PCR and western blotting, respectively, in Bs , GFP-BLm and GFP-BLm I841T cells (upper panels), in HeLash-siBLm and HeLaV-siCtrl cells (middle panels) and in human embryonic kidney HEK293sh-siBLm and HEK293V-siCtrl cells (lower panels). Error bars represent the s.e.m. of three independent experiments. The signic fi ance of differences was assessed by s tudent’s t-tests. (b) Analysis of CDA promoter activity based on the luciferase assay in Bs , GFP-BLm and GFP-BLm I841T cells. Firey fl luciferase activity was normalized with respect to that of the Renilla luciferase expressed from a control vector introduced by co-transfection. Error bars represent the s.e.m. of seven independent experiments. The signic fi ance of differences was assessed by s tudent’s t-tests. (c) Deoxycytidine triphosphate (dCTP) and deoxythymidine (dTTP) levels in Bs and GFP-BLm cells. Error bars represent the s.e.m. of seven independent experiments. The signic fi ance of differences was assessed by s tudent’s t-tests (upper panels). Deoxyadenosine triphosphate (dATP) and deoxyguanosine triphosphate (dGTP) levels in Bs -GFP and GFP-BLm cells. Error bars represent the s.e.m. of three independent experiments (lower panels). and did not modify the rate of 5-bromodeoxyuridine incorporation substrates for the recombinant BLM protein are G-quadruplex 18,19 into DNA (Supplementary Fig. S3), we concluded that the addition structures . However, we identified no potential intramolecular of dU or T mimicked the activity of CDA, thus accounting for the G-quadruplex-forming sequences or particular sequences likely observed decrease in SCE frequency in BS cells (Fig. 4a). Moreover, to form secondary structures in the CDA promoter (using Quad dU addition led to an 18.2% increase in fork velocity in BS cells Parser, website http://www.quadruplex.com), suggesting that CDA but had no effect on fork velocity in GFP-BLM cells ( Fig. 4b). e Th expression is not repressed by secondary structures formed in number of asymmetric replication bubbles in BS cells, reflecting the the absence of BLM. Furthermore, in our microarray study, none number of arrested forks , was not ae ff cted by dU addition ( Fig. 4c). of the genes with deregulated expression in the absence of BLM es Th e results show that the addition of dU efficiently decreased the encoded a known transcriptional factor (Supplementary Table S1). frequency of SCE and restored fork velocity in BS cells (although Another possibility is that BLM directly targets certain promoters, not to the same extent as CDA expression), demonstrating that the as reported for the WRN helicase . contribution of the CDA defect to the BS phenotype results from e Th key n fi dings of this work are that the defect in replication fork pyrimidine pool disequilibrium. es Th e results also make it pos - speed and, a fraction of the increase in SCE frequency in BS cells sible to distinguish between the replication fork velocity and fork are the consequences of a cascade reaction in which the defect in restart as two independent events, both ae ff cted by BLM helicase BLM helicase activity leads to a CDA expression defect that in turn deficiency, through nucleotide pool disequilibrium in the case of provokes a pyrimidine pool disequilibrium. Importantly, replication the replication fork velocity. fork progression defect can be rescued and the rate of SCEs reduced by 25–40% by acting on the nucleotide pool, without restoring BLM Discussion activity. Indeed, addition of dU, the n fi al CDA product, was found to This study demonstrates for the first time that BLM deficiency be suc ffi ient to restore the replication fork velocity and to decrease leads to CDA downregulation, resulting in pyrimidine pool dis- signic fi antly the level of SCEs in BLM-dec fi ient cells. e Th se data equilibrium and accounting, in part, for the BS phenotype. How- highlight the direct mechanistic connection between the replication ever, it remains unclear how BLM controls CDA expression. We defect, part of the increase in SCE and pyrimidine pool disequili- demonstrate here that the helicase activity of BLM regulates CDA brium in BS cells. Decrease in the fork velocity because of the nucleo- promoter function either directly or indirectly. The preferred tide pool disequilibrium may lead to replication stress and, thus, to n ATu RE Commun ICATIons | 2:368 | Do I: 10.1038/ncomms1363 | www.nature.com/naturecommunications © 2011 Macmillan Publishers Limited. All rights reserved. BS GFP-BLM GFP-BLMI841T GFP-BLM HeLaV-siCtrl HeLash-siBLM HEK293V-siCtrl HEK293sh-siBLM CDA mRNA relative quantity CDA mRNA relative quantity CDA mRNA relative quantity dATP pool dCTP pool 6 6 (pmol per 10 cells) (pmol per 10 cells) Normalized luciferase activity dGTP pool dTTP pool 6 6 (pmol per 10 cells) (pmol per 10 cells) ARTICLE n ATu RE Commun ICATIons | Do I: 10.1038/ncomms1363 50% 47.5% 45.2% a b 40% 70% 64.7% kDa 32.5% 60% 16 - CDA 30% 50% 44.8% 42 - β-Actin kDa 40% 20.0% 19.0% 19.0% 8 6.80 170- 20% 27.6% BLM 30% 20.6% 14.3% 20.7% 16- 6 CDA 20% 14.7% 10% 3.58 42- 6.9% β-Actin 4 10% 2.5% 0% 0% 0% BS-V BS-CDA BS-V BS-CDA GFP-BLM siCtrl GFP-BLM siCDA c d 84.4% 60% 80% 51.4% 50% 60.6% 40.5% 4.79 40% 60% 29.7% 30% 24.3% 2.74 18.9% kDa 40% 16.2% 18.9% 33.3% 20% 170- BLM 10% 16- CDA 15.6% 20% 0% 0 0% 42- β-Actin 6.1% GFP-BLM GFP-BLM+dC 0% GFP-BLM GFP-BLM+dC e f HeLaV-siCtrl/Ctrl HeLaV-siCtrl/CDA P = 0.0366 1.4 1.26 50% 45.8% 1.2 1 BS-V 0.885 41.2% 1.0 37.5% * 40% 0.8 1.100 BS-CDA 0.6 0.4 30% 26.5% GFP-BLM 1.066 0.2 17.6% 1.2 kDa 20% 0 0.2 0.4 0.6 0.8 1.0 BS-V BS-CDA 14.7% 14.6% –1 Replication fork velocity (kb min ) CDA 16- 10% 42- β-Actin 2.1% Figure 2 | CDA expression in BLM-deficient cells partially complements 0% BS phenotype. (a) dCTP pool level in Bs -V and Bs -CDA cells and CDA expression assessed by immunoblotting. Error bars represent the range of two independent experiments. (b) Percentage of metaphases with n s CE HeLash-siBLM/Ctrl HeLash-siBLM/CDA per chromosome from Bs -V and Bs -CDA cells. In all, 40–50 metaphases from three independent experiments were analysed for each condition. GFP-BLM (c) dCTP pool level in GFP-BLm cells untreated or treated with dC. Error 1.061 siCtrl bars represent the range of two independent experiments. (d) Percentage GFP-BLM 0.884 of metaphases with n s CE per chromosome from GFP-BLm cells untreated siCDA or treated with dC. In all, 40–50 metaphases from three independent 0.782 BS siCtrl experiments were analysed for each condition. (e) Bs -V or Bs -CDA cells were plated in triplicate at three dilutions (200, 400 and 600 cells). 0.770 BS siCDA Error bars represent the s.e.m. of three independent experiments. The significance of differences was assessed by s tudent’s t-tests. 0 0.2 0.4 0.6 0.8 1.0 1.2 –1 (f) Replication fork velocity in Bs -V, Bs -CDA and GFP-BLm cells. Error Replication fork velocity (kb min ) bars represent the range of two independent experiments. m ann–Whitney Figure 3 | siRNA-mediated CDA depletion partially reproduces BS statistical tests have been performed on the total number of fork analysed phenotype. (a) Percentage of metaphases with n s CE per chromosome from the two experiments (Bs -V, n = 880; Bs -CDA, n = 568; and GFP-BLm : from GFP-BLm cells transfected with the indicated siRn As. In all, 40–50 n = 800). Asterisk indicates a P-value < 0.00001. metaphases from three independent experiments were analysed for each condition. CDA levels were assessed by immunoblotting. (b) Percentage of an increase in replication-associated DNA breaks, accounting in metaphases with n s CEs per chromosome from HeLaV-siCtrl and HeLash- part for SCE formation. e Th se data indicate that the high frequency siBLm cells transiently transfected with the indicated siRn As. CDA protein of SCEs in BS cells has at least two die ff rent origins: 60–75% of levels were assessed by immunoblotting. About 40–50 metaphases SCEs may ree fl ct the restarting of blocked replication forks through from two independent experiments were analysed for each condition. RAD51-mediated homologous recombination , as proposed in (c) Replication fork velocity in GFP-BLm and Bs cells transfected with 9,21,22 several models . e Th remaining 25–40% of SCEs may ree fl ct the the indicated siRn As. Error bars represent the range of two independent repair of DNA breaks because of pyrimidine pool disequilibrium. i Th s experiments. m ann–Whitney statistical tests have been performed on second type of SCEs may also be dependent on RAD51-dependent the total number of forks analysed from the two experiments (GFP-BLm homologous recombination . i Th s study also demonstrates that siCtrl, n = 312; GFP-BLm siCDA, n = 452; Bs siCtrl: n = 459; and Bs siCDA, CDA downregulation in BLM-expressing cells leads to a signic fi ant n = 467). Asterisks indicate a P-value < 0.00001. increase in SCE frequency and replication stress, potentially confer- ring a predisposition to cancer. A previous French study reported low levels of CDA activity, rendering gemcitabine highly toxic . that 7% of adult patients with cancer treated with gemcitabine (5′- We therefore suggest that CDA dec fi iency may be a marker of azacytidine, 2′,2′-diu fl orodeoxycytidine), a CDA substrate, had very predisposition to cancer in the general population. n ATu RE Commun ICATIons | 2:368 | Do I: 10.1038/ncomms1363 | www.nature.com/naturecommunications © 2011 Macmillan Publishers Limited. All rights reserved. BS-V BS-CDA n < 1 1< n < 1.2 1.2 < n < 1.4 n > 1.4 n < 1 1 < n < 1.2 1.2 < n < 1.4 n > 1.4 GFP-BLM siCtrl GFP-BLM siCDA n < 0.3 0.3 < n < 0.4 0.4 < n < 0.5 n > 0.5 n < 0.3 0.3 < n < 0.4 0.4 < n < 0.5 n > 0.5 HeLaV-siCtrl/Ctrl 0 < n < 0.2 HeLaV-siCtrl/CDA 0.2 < n < 0.3 0.3 < n < 0.4 n > 0.4 0 < n < 0.2 0.2 < n < 0.3 0.3 < n < 0.4 n > 0.4 HeLash-siBLM/Ctrl HeLash-siBLM/CDA 0 < n < 0.2 0.2 < n < 0.3 0.3 < n < 0.4 0 < n < 0.2 0.2 < n < 0.3 0.3 < n < 0.4 0 < n < 0.2 0.2 < n < 0.3 0.3 < n < 0.4 n > 0.4 0 < n < 0.2 0.2 < n < 0.3 0.3 < n < 0.4 n > 0.4 dCTP pool dCTP pool Clonogenic survival 6 6 (pmol per 10 cells) (pmol per 10 cells) Percentage of metaphases Percentage of metaphases with n SCE per chromosome with n SCE per chromosome Percentage of metaphases Percentage of metaphases Percentage of metaphases with n SCE per chromosome with n SCE per chromosome with n SCE per chromosome n ATu RE Commun ICATIons | Do I: 10.1038/ncomms1363 ARTICLE P = 0.003 analysis of microarray expression experiments was performed as previously de- P = 0.0052 scribed using a P-value cutoff of 0.05. 1.6 1.415 Raw and normalized transcriptomic data are available at the Institut Curie GFP-BLM 1.066 microarray dataset repository (http://microarrays.curie.fr/. Direct link: http:// 1.2 0.879 GFP-BLM+dU 1.069 0.822 microarrays.curie.fr/publications/UMR3348/pyrimidine-bloom-syndrome/). 0.8 * BS 0.839 0.350 0.349 0.4 0.284 * Plasmid construction and site-directed mutagenesis. Site-directed mutagenesis 0.992 BS + dU was carried out in the EGFP-C1 vector containing the full-length BLM cDNA: 0 0.2 0.4 0.6 0.8 1.0 1.2 Ile-841 was mutated with the QuikChange XL site-directed mutagenesis kit (Strata- –1 Replication fork velocity (kb min ) gene) according to the manufacturer’s instructions. Symmetric replication bubble BS GFP-BLM Primer 1 (5′-CCCAGGGTACAGAAGGACACCCTGACTCAGCTGAAG-3′) IdU incorporation and primer 2 (5′-CTTCAGCTGAGTCAGGGTGTCCTTCTGTACCCTGGG-3′) 40% CldU incorporation 31.7% 30.9% were used. 30% 17.7% 18.5% 20% Reverse transcription and real-time quantitative PCR. Total RNA was extracted Asymmetric replication bubble with RNeasy Mini kit (Qiagen) including a DNAse digestion step. RNA quality was 10% IdU incorporation assessed with the Experion system (BioRad), and cDNAs were synthesized with CldU incorporation 0% 250 ng of random hexamers (Invitrogen), 2 µg of RNA and Superscript II reverse transcriptase (Invitrogen). Quantitative PCR experiments were performed accord- ing to the MIQE Guideline . Amplification mixtures contained the cDNA template (1/100 dilution), SYBR Green Supermix 1× (BioRad) and 300 nM forward and reverse primers. Amplification was performed with the CFX96 detection system Figure 4 | The BS phenotype is partially complemented by dU or T. (BioRad). The relative quantities of the CDA and BLM cDNAs were normalized (a) sCE frequencies in Bs and GFP-BLm cell lines untreated or treated with du against four reference genes (RPL32, HPRT1, HMBS and SDHA) chosen on the basis or T (about 1,800 chromosomes analysed for each condition). Error bars of their low M-value . The primer sequences for BLM, RPL32, HPRT1, HMBS and 28–30 represent the s.e.m. of three independent experiments. The significance of SDHA have been described in previous publications . CDA primer sequences: differences was assessed by s tudent’s t-tests. (b) Replication fork velocity primer 1 (5′-CCCTACAGTCACTTTCCTG-3′) and primer 2 (5′-CGGGTAGCAGG CATTTTCTA-3′). in GFP-BLm and Bs cell lines untreated or treated with du . Error bars represent the range of three independent experiments. m ann–Whitney Western blot analysis and antibodies. Cells were lysed in 350 mM NaCl, 50 mM statistical tests have been performed on the total number of forks analysed Tris–HCl, pH7.5, 1% NP-40 and protease inhibitors (Roche), sonicated and heated from the three experiments (GFP-BLm , n = 530; GFP-BLm + du , n = 642; Bs , at 70 °C for 10 min. Samples equivalent to 25 µg of protein were subjected to electrophoresis in NuPAGE Novex 4–12% Bis-Tris pre-cast gels (Invitrogen). The n = 858; and Bs + du , n = 440). Asterisks indicate a P-value < 0.00001. procedures used for gel electrophoresis and immunoblotting have been described (c) Percentage of asymmetric replication bubbles in Bs and GFP-BLm elsewhere . Primary and secondary antibodies were used at the following concen- cell lines untreated or treated with du , and examples of symmetric and trations: rabbit anti-BLM antibody (1:1,000; ab476 from Abcam); rabbit anti-CDA asymmetric bubbles (70–130 replication bubbles analysed for each antibody (1:500; ab56053 from Abcam); rabbit anti-β-actin antibody (1:10,000; condition). Cldu , chlorodeoxyuridine; Idu , iododeoxyuridine. Error bars Sigma); and horseradish peroxidase-conjugated goat anti-rabbit IgG (1:5,000; Santa Cruz Biotechnology). represent the range of two independent experiments. dNTP assays. Assays were performed as previously described . Briefly, dNTPs were extracted from fresh cell pellets (10–40×10 cells) in 0.6 M trichloroace- Methods Cell culture and treatments. Cell lines were cultured in DMEM supplemented tic acid and 15 mM MgCl . Extracts were neutralized and used immediately or stored at − 80 °C. For each dNTP assay, an oligonucleotide template primer was with 10% FCS. BS and GFP-BLM cells were obtained by transfecting BS GM08505B designed , and DNA polymerization reactions were carried out with Sequenase cells with the EGFP-C1 vector alone (Clontech) or with the same vector containing the full-length BLM cDNA , respectively, using JetPEI reagent (Ozyme). Ae ft r 48 h, (USB Corporation) and a radiolabeled nucleotide (Perkin Elmer) on samples or on − 1 dNTP standard solutions (0–4 pM). Samples were spotted on to DE81 membranes selection with 800–1,600 µg ml G418 (PAA) was applied. Individual colonies were − 1 (Whatman), which were washed with 5% (w/v) Na HPO (Sigma) and counted in a isolated and cultured in medium containing 500 µg ml G418. 2 4 liquid scintillation counter. GFP-BLMI841T cells were obtained by transfecting BS GM08505B cells as dTTP and dCTP pools were determined in GFP-BLM cells treated with 2 mM described above, with the EGFP-C1 vector containing the full-length BLM cDNA hydroxyurea (Sigma) for 2 h as a positive control for the dNTP assay (Supplemen- mutated at codon 841. tary Fig. S4). HeLaV and HeLashBLM cells, HEK293V and HEK293shBLM cells were obtained by transfecting cells with an empty pSM2 vector or with the same vector encoding a short hairpin RNA sequence directed against BLM (Open Biosystems, SCE assays. Cells were transferred to slides and cultured in the presence of 10 µM 5-bromodeoxyuridine (BrdU) (Sigma) during two cell divisions. Aer 40–48 ft h, clone V2HS-89234), respectively, using JetPEI reagent. Aer 48 ft h, selection with − 1 − 1 1–5 µg ml puromycin (Invivogen) was done. Individual colonies were isolated colchicine (Sigma) was added (0.1 µg ml ) and the cells were incubated for 1 h. − 1 and cultured in medium containing 0.5 µg ml puromycin. Cells were then incubated in hypotonic solution (1:5 (vol/vol) FCS-distilled water) and fixed with a 3:1 (vol/vol) mixture of methanol-acetic acid. Cells were then BS-V and BS-CDA cell lines were obtained by transfecting BS cells with an 25 − 1 stained by incubation with 10 µg ml Hoechst 33258 (Sigma) in distilled water for empty pCI-puro vector , or with the same vector containing the full-length CDA cDNA (NM001785), using JetPEI reagent. Aer 48 ft h, selection was carried out with 20 min, rinsed with 2×SSC (Euromedex), exposed to ultraviolet light at 365 nm at a − 1 − 1 distance of 10 cm for 105 min, rinsed in water, stained by incubation with 2% Giemsa 0.2 µg ml puromycin (Invivogen) and 500 µg ml G418 (Invitrogen). Individual − 1 solution (VWR) for 16 min, rinsed in water, dried and mounted. Chromosomes colonies were isolated and maintained in culture with 0.1 µg ml puromycin and − 1 500 µg ml G418. were observed with a Leica DMRB microscope at ×100 magnification. Metaphases 5 5 5 For siRNA transfection assays, 3×10 to 4×10 HeLa or HEK cells or 8×10 to were captured with a SONY DXC 930 P camera and SCEs were analysed. 9×10 BS or GFP-BLM cells were used to seed the wells of a six-well plate. Cells were transfected with an siRNA specic fi for BLM or CDA (ON-TARGETplus SMART - DNA molecular combing. Asynchronous populations of cells were labeled by pool, Dharmacon) or negative control siRNAs (ON-TARGETplus siCONTROL Non incubation for 40 min with 100 µM IdU (Sigma-Aldrich), washed with medium Targeting Pool, Dharmacon; 100 nM n fi al concentration) for 48 h for BLM, or twice at 37 °C and then labeled by incubation for 40 min with 100 µM CldU (Sigma- Aldrich). The DNA solution was prepared as previously described , and DNA was successively, for a total of 120 h for CDA using DharmaFect (Dharmacon). Deoxyribonucleosides (Sigma) were added to the cell culture medium at a final combed on silane-treated coverslips with a combing apparatus (Genomic Vision). concentration of 500 µM (dC) and 25 µM (dU, T) for 96 h (2×48 h). Coverslips with combed DNA were baked overnight at 60 °C and incubated in 0.5 M NaOH–1 M NaCl solution for 15 min, with gentle shaking, for DNA Transcriptomic analysis with Affymetrix microarrays. Analysis was performed denaturation. Coverslips were washed several times in PBS and dried by successive on BS, GFP-BLM, HeLaV-siCtrl and HeLash-siBLM cell lines. For each cell line, incubations, for 5 min each, in 70, 90 and 100% ethanol. They were then incubated three independent total RNA samples were extracted using the RNeasy system with the primary antibodies. All antibodies were diluted in BlockAid (Invitrogen). (Qiagen). Fragmented biotinylated antisense RNA was generated and hybridized to Coverslips were incubated at room temperature, first with 1/5 mouse anti-BrdU Affymetrix Human Genome U133 Plus 2.0 Arrays. Aer h ft ybridization, arrays were (BD Biosciences) + 1/25 rat anti-CldU (ABserotec) for 1 h, then with 1/25 goat anti- scanned following guidelines from Affymetrix (http://www.affymetrix.com). These mouse 488 + 1/25 Alexa Fluor goat anti-rat 555 antibodies (Invitrogen) for 40 min, arrays contain ~54,000 probesets, representing ~47,000 transcripts. Statistical followed by 1/50 mouse anti-ssDNA antibody (Millipore) for 40 min, then 1/25 n ATu RE Commun ICATIons | 2:368 | Do I: 10.1038/ncomms1363 | www.nature.com/naturecommunications © 2011 Macmillan Publishers Limited. All rights reserved. GFP-BLM GFP-BLM+dU BS BS+dU Untreated dU 25µM T 25µM Untreated dU 25µM T 25µM Percentage of No. of SCE per asymmetric bubbles chromosome ARTICLE nATuRE C ommunICATIons | DoI: 10.1038/ncomms1363 rabbit anti-mouse 350 antibody (Invitrogen) for 30 min and finally with 1/25 goat 19. Sun, H., Karow, J. K., Hickson, I. D. & Maizels, N. The Bloom’s syndrome anti-rabbit 350 antibody (Alexa Fluor, Invitrogen) for 30 min. Between incuba- helicase unwinds G4 DNA. J. Biol. Chem. 273, 27587–27592 (1998). tions, coverslips were washed three times, for 5 min each, in PBS. Coverslips were 20. Lachaud, A. A. et al. Werner’s syndrome helicase participates in transcription mounted in Prolong Gold antifade reagent (Invitrogen). Images were acquired with of phenobarbital-inducible CYP2B genes in rat and mouse liver. Biochem. a Leica DM RXA microscope equipped with a motorized XY stage, using a ×40 Pharmacol. 79, 463–470 (2010). PlanApo N.A. 1.25 objective and a CoolSNAP HQ interline CCD camera (Photo- 21. Bachrati, C. Z. & Hickson, I. D. RecQ helicases: guardian angels of the DNA metrics). For each slide, a mosaic of 10×10 partly overlapping images was collected replication fork. Chromosoma 117, 219–233 (2008). with a Metamorph software (Molecular Devices) routine developed in-house. 22. Wu, L. Role of the BLM helicase in replication fork management. DNA Repair Image collections were assembled into a mosaic with the ‘Stitching 2D/3D’ plugin (Amst) 6, 936–944 (2007). (available from http://fly.mpi-cbg.de/~preibisch/software.html) for ImageJ software 23. Ciccolini, J. et al. Cytidine deaminase residual activity in serum is a (Rasband, W.S., ImageJ, US National Institutes of Health, Bethesda, Maryland, predictive marker of early severe toxicities in adults aer g ft emcitabine-based USA, http://rsb.info.nih.gov/ij/, 1997–2009.) chemotherapies. J. Clin. Oncol. 28, 160–165 (2010). 24. Eladad, S. et al. Intra-nuclear trafficking of the BLM helicase to DNA damage- CDA promoter cloning and luciferase reporter assays. e Th CDA promoter induced foci is regulated by SUMO modification. Hum. Mol. Genet. 14, was inserted into a pGL4.14 vector (Promega), upstream from the r fi ey fl luciferase 1351–1365 (2005). gene. Cells were transfected by incubation for 72 h with 2.5 µg of the CDA 25. Girard, P. M., Kysela, B., Harer, C. J., Doherty, A. J. & Jeggo, P. A. Analysis of promoter-containing plasmid and 0.25 µg of pGL4.74 vector (Promega, internal DNA ligase IV mutations found in LIG4 syndrome patients: the impact of two control, Renilla luciferase) in the presence of JetPEI reagent. Firey fl luciferase and linked polymorphisms. Hum. Mol. Genet. 13, 2369–2376 (2004). Renilla luciferase activities were assessed with the Dual-Luciferase Reporter Assay 26. Irizarry, R. A. et al. Exploration, normalization, and summaries of high density System protocol (Promega). oligonucleotide array probe level data. Biostatistics 4, 249–264 (2003). 27. Bustin, S. A. et al. The MIQE guidelines: minimum information for publication Clonogenic survival assays. Cells were plated in a drug-free medium at three of quantitative real-time PCR experiments. Clin. Chem. 55, 611–622 (2009). different densities, in triplicate, for the counting of 30–300 clones, depending 28. Vandesompele, J. et al. Accurate normalization of real-time quantitative RT- on expected survival. Aer 14–21 d ft ays of incubation, colonies were fixed and PCR data by geometric averaging of multiple internal control genes. Genome − 1 stained with methylene blue (5 g l in 50% water and 50% methanol) and scored. Biol. 3, RESEARCH0034 (2002). Only experiments giving a linear correlation between the different dilutions were 29. Bischof, O. et al. Regulation and localization of the Bloom syndrome protein in considered. Colony-forming efficiency was estimated by dividing the number of response to DNA damage. J. Cell. Biol. 153, 367–380 (2001). colony-forming units by the number of cells plated. 30. Viegas, M. H., Gehring, N. H., Breit, S., Hentze, M. W. & Kulozik, A. E. The abundance of RNPS1, a protein component of the exon junction complex, Statistical analysis. The significance of differences was assessed with Student’s can determine the variability in efficiency of the Nonsense Mediated Decay t-test or Mann–Whitney test. P < 0.05 was considered statistically significant. pathway. Nucleic Acids Res. 35, 4542–4551 (2007). 31. Sherman, P. A. & Fyfe, J. A. Enzymatic assay for deoxyribonucleoside References triphosphates using synthetic oligonucleotides as template primers. Anal. 1. German, J. Bloom’s syndrome. XX. The first 100 cancers. Cancer Genet. Biochem. 180, 222–226 (1989). Cytogenet. 93, 100–106 (1997). 32. Preibisch, S., Saalfeld, S. & Tomancak, P. Globally optimal stitching of tiled 3D 2. Ellis, N. A. et al. The Bloom’s syndrome gene product is homologous to RecQ microscopic image acquisitions. Bioinformatics 25, 1463–1465 (2009). helicases. Cell 83, 655–666 (1995). 33. Fitzgerald, S. M. et al. Identification of functional single nucleotide 3. Bennett, R. J. & Keck, J. L. Structure and function of RecQ DNA helicases. polymorphism haplotypes in the cytidine deaminase promoter. Hum. Genet. Crit. Rev. Biochem. Mol. Biol. 39, 79–97 (2004). 119, 276–283 (2006). 4. Davies, S. L., North, P. S. & Hickson, I. D. Role for BLM in replication-fork restart and suppression of origin firing aer r ft eplicative stress. Nat. Struct. Mol. Acknowledgements Biol. 14, 677–679 (2007). We thank N. Ellis for providing us with the GFP-BLM construct, F. Cordelières for help 5. Rao, V. A. et al. Endogenous gamma-H2AX-ATM-Chk2 checkpoint activation in automating DNA combing analysis, B. Albaud and D. Gentien from the Affymetrix in Bloom’s syndrome helicase deficient cells is related to DNA replication Platform for microarray experiments, P. Hupé and N. Servant from the Bioinformatic arrested forks. Mol. Cancer Res. 5, 713–724 (2007). Platforms for helping us to analyse transcriptomic data. We also thank P-M. Girard and 6. Chaganti, R. S., Schonberg, S. & German, J. A manifold increase in sister D. Graindorge for advice about molecular combing and the members of UMR 3348 chromatid exchanges in Bloom’s syndrome lymphocytes. Proc. Natl Acad. Sci. for helpful discussions. This work was supported by grants from the Institut Curie, the USA 71, 4508–4512 (1974). CNRS, the Ligue contre le Cancer (Comité de l’Essonne) and by a fellowship awarded 7. Bartkova, J. et al. DNA damage response as a candidate anti-cancer barrier in to P.C. by the CNRS (BDI), the Ligue contre le Cancer (Comité de l’Essonne) and early human tumorigenesis. Nature 434, 864–870 (2005). Association pour la Recherche sur le Cancer. 8. Gorgoulis, V. G. et al. Activation of the DNA damage checkpoint and genomic instability in human precancerous lesions. Nature 434, 907–913 (2005). 9. Amor-Gueret, M. Bloom syndrome, genomic instability and cancer: the SOS- Author contributions like hypothesis. Cancer Lett. 236, 1–12 (2006). P.C. performed the transcriptomic analysis and all experiments, unless otherwise 10. Meuth, M. The molecular basis of mutations induced by deoxyribonucleoside indicated, participated in the design of experiments and data analysis and generated the triphosphate pool imbalances in mammalian cells. Exp. Cell. Res. 181, 305–316 (1989). figures. G.B.L. performed the work presented in Figures 2b,d and 4a. R.O.D. performed 11. Kunz, B. A. et al. International Commission for Protection Against the work presented in Figure 1b and in Supplementary Figure S1, CDA cloning and Environmental Mutagens and Carcinogens. Deoxyribonucleoside triphosphate established GFP-BLM, GFP-BLMI841T, BS-CDA and BS-V cell lines, and participated levels: a critical factor in the maintenance of genetic stability. Mutat. Res. 318, in the work in Figure 2c. S.L. and M.D. contributed to data analysis, O.B. contributed to 1–64 (1994). the work presented in Figure 1c and in Supplementary Figure S1. M.A.-G. designed the 12. Evans, D. R. & Guy, H. I. Mammalian pyrimidine biosynthesis: fresh insights experiments, analysed the data and wrote the manuscript. into an ancient pathway. J. Biol. Chem. 279, 33035–33038 (2004). 13. Lahkim Bennani-Belhaj, K et al. The Bloom syndrome protein limits the lethality associated with RAD51 deficiency. Mol. Cancer Res. 8, 385–394 (2010). Additional information 14. Nygaard, P. On the role of cytidine deaminase in cellular metabolism. Adv. Exp. Supplementary Information accompanies this paper on http://www.nature.com/ Med. Biol. 195 (Pt B), 415–420 (1986). naturecommunications 15. Guo, R. B. et al. Structural and functional analyses of disease-causing missense Competing financial interests: M.A.-G is the sole inventor of patent application mutations in Bloom syndrome protein. Nucleic Acids Res. 35, 6297–6310 (2007). ‘Low levels of cytidine deaminase as a marker for predisposition to develop cancer’, 16. Kunz, B. A. Mutagenesis and deoxyribonucleotide pool imbalance. Mutat. Res. PCT/EP2011/050784. The remaining authors declare no competing financial interests. 200, 133–147 (1988). 17. Warren, S. T., Schultz, R. A., Chang, C. C., Wade, M. H. & Trosko, J. E. Elevated Reprints and permission information is available online at http://npg.nature.com/ spontaneous mutation rate in Bloom syndrome fibroblasts. Proc. Natl Acad. Sci. reprintsandpermissions/ USA 78, 3133–3137 (1981). 18. Mohaghegh, P., Karow, J. K., Brosh, R. M. Jr., Bohr, V. A. & Hickson, I. D. The How to cite this article: Chabosseau, P. et al. Pyrimidine pool imbalance induced by Bloom’s and Werner’s syndrome proteins are DNA structure-specific helicases. BLM helicase deficiency contributes to genetic instability in Bloom syndrome. Nucleic Acids Res. 29, 2843–2849 (2001). Nat. Commun. 2:368 doi: 10.1038/ncomms1363 (2011). nATuRE C ommunICATIons | 2:368 | DoI: 10.1038/ncomms1363 | www.nature.com/naturecommunications © 2011 Macmillan Publishers Limited. All rights reserved. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Nature Communications Springer Journals

Pyrimidine pool imbalance induced by BLM helicase deficiency contributes to genetic instability in Bloom syndrome

Download PDF
6 pages

Loading next page...
 
/lp/springer-journals/pyrimidine-pool-imbalance-induced-by-blm-helicase-deficiency-wse5HdIFN0

References (39)

Publisher
Springer Journals
Copyright
Copyright © 2011 by Nature Publishing Group, a division of Macmillan Publishers Limited. All Rights Reserved.
Subject
Science, Humanities and Social Sciences, multidisciplinary; Science, Humanities and Social Sciences, multidisciplinary; Science, multidisciplinary
eISSN
2041-1723
DOI
10.1038/ncomms1363
Publisher site
See Article on Publisher Site

Abstract

ARTICLE Received 31 Jan 2011 | Accepted 23 may 2011 | Published 28 Jun 2011 DOI: 10.1038/ncomms1363 Pyrimidine pool imbalance induced by BLm helicase deficiency contributes to genetic instability in Bloom syndrome 1,2 1,2, 1,2, 1,2 Pauline Chabosseau , Géraldine Buhagiar-Labarchède *, Rosine onclercq-Delic *, sarah Lambert , 3,4 3,4 1,2 michelle Debatisse , olivier Brison & mounira Amor-Guéret Defects in DnA replication are associated with genetic instability and cancer development, as illustrated in Bloom syndrome. Features of this syndrome include a slowdown in replication speed, defective fork reactivation and high rates of sister chromatid exchange, with a general predisposition to cancer. Bloom syndrome is caused by mutations in the BLM gene encoding a RecQ helicase. Here we report that BLm deficiency is associated with a strong cytidine deaminase defect, leading to pyrimidine pool disequilibrium. In BLm-deficient cells, pyrimidine pool normalization leads to reduction of sister chromatid exchange frequency and is sufficient for full restoration of replication fork velocity but not the fork restart defect, thus identifying the part of the Bloom syndrome phenotype because of pyrimidine pool imbalance. This study provides new insights into the molecular basis of control of replication speed and the genetic instability associated with Bloom syndrome. nucleotide pool disequilibrium could be a general phenomenon in a large spectrum of precancerous and cancer cells. 1 2 Institut Curie, Centre de Recherche, Orsay, France. CNRS UMR 3348, Stress Génotoxiques et Cancer, Centre Universitaire, Bât. 110, 91405 Orsay, France. 3 4 CNRS UMR 3244, Institut Curie, Centre de Recherche, Paris, France. Université Pierre et Marie Curie, 26 rue d’Ulm, 75248 Paris cedex 05, France. *These authors contributed equally to this work. Correspondence and requests for materials should be addressed to M.A.-G. (email: [email protected]). nATuRE C ommunICATIons | 2:368 | DoI: 10.1038/ncomms1363 | www.nature.com/naturecommunications © 2011 Macmillan Publishers Limited. All rights reserved. ARTICLE nATuRE C ommunICATIons | DoI: 10.1038/ncomms1363 loom syndrome (BS) is a rare autosomal recessive disorder CDA gene expression (Fig. 1a), demonstrating the requirement of characterized by marked genetic instability associated with BLM helicase activity for CDA gene expression. We inserted the Bpredisposition to a wide range of cancers commonly ae ff ct - human CDA promoter into a luciferase vector and showed that the ing the general population . BS is caused by mutations in the BLM CDA promoter was significantly less active in the absence of BLM gene, which encodes BLM, a RecQ 3′–5′ DNA helicase, indicating or of its helicase activity (Fig. 1b), demonstrating the requirement an essential role for BLM in maintaining genetic stability and pre- of BLM helicase activity for correct CDA expression and, thus, venting cancer . In vitro, BLM unwinds DNA structures mimick- the direct or indirect involvement of this activity in regulation of ing replication forks and homologous recombination intermedi- the CDA promoter. ates . The specific functions of BLM remain unclear, but it is widely CDA has an essential role in maintaining the cellular pool of thought that BLM is involved in restarting blocked replication nucleotides, which is crucial for genome stability. Indeed, an imbal- 4,5 forks . In the absence of BLM, cells display a high rate of sister ance in the nucleotide pools is highly mutagenic in mammalian 6 16 6,17 chromatid exchanges (SCEs), pathognomonic for BS . BS cells also cells , and BS cells display a mutator and hyper-Rec phenotype . display a slowing of replication fork progression associated with e Th purine pools were not ae ff cted in BS cells, whereas the pyrimi - 4,5 endogenous activation of the ATM-Chk2-γH2AX pathway . This dine pools were significantly unbalanced ( Fig. 1c). Indeed, BS cells pathway is activated in precancerous lesions and is thought to be had about twice as much dCTP and a slightly lower level of dTTP 7,8 a part of an antitumorigenesis barrier . This barrier is based on than GFP-BLM cells (Fig. 1c). Moreover, expression of the GFP- DNA replication stress, leading to activation of the DNA damage BLMI841T non-functional protein in BS cells, which was not able 7,8 checkpoint and, thus, apoptosis or cell cycle arrest . BS cells may to rescue CDA gene expression (Fig 1a), also failed to normalize the therefore be in a precancerous state. Our working hypothesis is dCTP pool size (Supplementary Fig. S1). that BLM deficiency results in a global ‘SOS-like’ cellular response, facilitating progression through the antitumorigenesis barrier CDA expression partially rescues cellular BS phenotype. Stable potentially involved in carcinogenesis in the general population . expression of the CDA gene in BS cells (BS-CDA) was sufficient u Th s, identification of the genes and pathways deregulated in the to restore a dCTP pool similar to that of GFP-BLM cells (Fig. 2a), absence of BLM may facilitate the establishment of a molecular demonstrating the direct responsibility of the CDA defect for signature of the precancerous state. the pyrimidine pool imbalance observed in BLM-deficient cells. Deoxyribonucleoside triphosphate (dNTP) levels are also of CDA expression in BS cells also decreased the frequency of SCE major importance for maintaining genomic integrity. Indeed, by 25% (P = 0.0015, Student’s t-test), reflected in a great decrease nucleotide pool imbalance, particularly for the pyrimidine pools, in the number of cells with high levels of SCE (Fig. 2b), suggest- is known to result in genetic instability and oncogenic transforma- ing a direct relationship between the high rate of SCE observed in 10,11 tion . Pyrimidine biosynthesis occurs through the de novo path- BS cells and nucleotide pool disequilibrium. Addition of 500 µM way, in which nucleotides are synthesized from small metabolites, or dC to the culture medium of GFP-BLM cells led to a 1.75 times the salvage pathway, which recycles nucleosides and nucleotides . increase in dCTP pool size (Fig. 2c) and ~42% increase in SCE In mammalian cells, dec fi iencies of many of the enzymes involved in frequency (P = 0.0327, Student’s t-test; Fig. 2d), confirming that the two pyrimidine biosynthesis pathways, such as thymidine kinase, nucleotide pool disequilibrium promotes an increase in SCE fre- dTMP synthase, dCTP kinase, CTP synthase and dCMP deami- quency. CDA expression in BS cells also increased colony-forming nase, have been associated with the induction of mutations, DNA efficiency significantly ( Fig. 2e). Moreover, CDA expression in BS breaks, sensitization to DNA-damaging agents, increase in the rate cells was sufficient to increase replication fork speed, by up to 10,11 of homologous recombination and chromosomal abnormalities . 24%, restoring fork velocity to levels similar to those in GFP-BLM We report here that BLM deficiency leads to cytidine deaminase cells (Fig. 2f ). These results suggest that the slowdown of replica - (CDA) downregulation, resulting in pyrimidine pool disequilib- tion speed in BS cells could be the consequence of the pyrimidine rium, accounting for the slowing of replication fork progression and pool imbalance. contributing to the increase in SCE frequency associated with the BS phenotype. CDA downregulation reproduces part of the BS phenotype. e Th short interfering RNA (siRNA)-mediated downregulation of Results CDA in GFP-BLM cells increased the frequency of SCE by 46% BLM deficiency is associated with a strong CDA defect . For the (P = 0.0134, Student’s t-test), reflecting a significant increase in the identification of genes specifically deregulated in the absence of number of cells with a high SCE frequency (Fig. 3a). By contrast, BLM, we performed a microarray analysis comparing the transient transfection of BS cells (with no detectable CDA expres- transcriptome of wild-type HeLa cells with that of HeLa cells in sion) with siRNAs specific for CDA had no ee ff ct on SCE level which BLM was downregulated , and the transcriptome of the (Supplementary Fig. S2). CDA downregulation in HeLa cells and GM8505B BS cell line (BS) with that of its BLM-complemented in BLM-downregulated HeLa cells (in which CDA downregulation counterpart (green uo fl rescent protein (GFP)-BLM). We identified was not complete as seen in Fig. 1a) also significantly increased 77 candidate genes for which regulation was potentially modified the number of cells with a high SCE frequency (Fig. 3b). Moreo- in the same way in both HeLa cells depleted of BLM and in BS ver, CDA downregulation significantly decreased replication fork cells (Supplementary Table S1). One of these genes, encoding CDA velocity in GFP-BLM cells, but not in BS cells (Fig. 3c). us Th , the (EC 3.5.4.5; locus 1p36.2-p35), particularly attracted our attention. level of CDA gene expression inu fl ences both SCE frequency and CDA is an enzyme of the pyrimidine salvage pathway catalysing the speed of the replication fork, CDA downregulation being asso- the hydrolytic deamination of cytidine (C) and deoxycytidine (dC) ciated with an increase in the level of SCE and a decrease in the to uridine (U) and deoxyuridine (dU), respectively . We confirmed replication fork velocity. that CDA mRNA and protein levels were strongly reduced in BS cells and in BLM-downregulated HeLa cells, and in HEK 293 human Addition of dU partially rescues BS phenotype. Finally, we found embryonic kidney cells depleted of BLM (Fig. 1a). e Th expression of that the addition of 25 µM dU (the final CDA product) or thymidine a functional BLM protein (GFP-BLM) in BS cells was sufficient to (T) to the culture medium led to ~40% decrease in SCE frequency in restore CDA expression (Fig. 1a). By contrast, expression of a BLM BS cells (Fig. 4a). As the addition of dU at this concentration had no protein with an inactive helicase domain but a functional DNA- effect on residual SCE frequency in GFP-BLM cells or on the qual - binding domain (GFP-BLMI841T) in BS cells failed to rescue ity of the differential labeling of sister chromatids in either cell line, nATuRE C ommunICATIons | 2:368 | DoI: 10.1038/ncomms1363 | www.nature.com/naturecommunications © 2011 Macmillan Publishers Limited. All rights reserved. n ATu RE Commun ICATIons | Do I: 10.1038/ncomms1363 ARTICLE P<0.0001 P<0.0001 P=0.0002 1.2 1.2 1.0 kDa 0.8 0.8 0.60 0.56 170 - BLM 0.6 16 - CDA 0.4 0.4 42 - β-Actin 0.2 0.008 0.0166 0 0 BS GFP-BLM GFP-BLMI841T GFP-BLM BS GFP-BLMI841T P=0.0069 P<0.0001 P=0.0299 1.2 7.45 1 9 60 1.0 46.85 kDa 38.75 0.8 6 40 170 - BLM 3.84 0.6 0.456 16 - CDA 0.4 3 20 0.2 β-Actin 42 - 0 0 0 HeLaV-siCtrl HeLash-siBLM BS GFP-BLM BS GFP-BLM P<0.0001 1.2 9 9 1.0 4.67 4.65 kDa 0.8 6 6 0.6 BLM 170 - 0.4 16 - 3 CDA 1.07 0.109 0.97 0.2 42 - β-Actin 0 0 0 HEK293V-siCtrl HEK293sh-siBLM BS GFP-BLM BS GFP-BLM Figure 1 | BLM dec fi iency is associated with CDA dec fi iency and pyrimidine pool imbalance. (a) CDA mRn A and protein levels determined by real- time quantitative PCR and western blotting, respectively, in Bs , GFP-BLm and GFP-BLm I841T cells (upper panels), in HeLash-siBLm and HeLaV-siCtrl cells (middle panels) and in human embryonic kidney HEK293sh-siBLm and HEK293V-siCtrl cells (lower panels). Error bars represent the s.e.m. of three independent experiments. The signic fi ance of differences was assessed by s tudent’s t-tests. (b) Analysis of CDA promoter activity based on the luciferase assay in Bs , GFP-BLm and GFP-BLm I841T cells. Firey fl luciferase activity was normalized with respect to that of the Renilla luciferase expressed from a control vector introduced by co-transfection. Error bars represent the s.e.m. of seven independent experiments. The signic fi ance of differences was assessed by s tudent’s t-tests. (c) Deoxycytidine triphosphate (dCTP) and deoxythymidine (dTTP) levels in Bs and GFP-BLm cells. Error bars represent the s.e.m. of seven independent experiments. The signic fi ance of differences was assessed by s tudent’s t-tests (upper panels). Deoxyadenosine triphosphate (dATP) and deoxyguanosine triphosphate (dGTP) levels in Bs -GFP and GFP-BLm cells. Error bars represent the s.e.m. of three independent experiments (lower panels). and did not modify the rate of 5-bromodeoxyuridine incorporation substrates for the recombinant BLM protein are G-quadruplex 18,19 into DNA (Supplementary Fig. S3), we concluded that the addition structures . However, we identified no potential intramolecular of dU or T mimicked the activity of CDA, thus accounting for the G-quadruplex-forming sequences or particular sequences likely observed decrease in SCE frequency in BS cells (Fig. 4a). Moreover, to form secondary structures in the CDA promoter (using Quad dU addition led to an 18.2% increase in fork velocity in BS cells Parser, website http://www.quadruplex.com), suggesting that CDA but had no effect on fork velocity in GFP-BLM cells ( Fig. 4b). e Th expression is not repressed by secondary structures formed in number of asymmetric replication bubbles in BS cells, reflecting the the absence of BLM. Furthermore, in our microarray study, none number of arrested forks , was not ae ff cted by dU addition ( Fig. 4c). of the genes with deregulated expression in the absence of BLM es Th e results show that the addition of dU efficiently decreased the encoded a known transcriptional factor (Supplementary Table S1). frequency of SCE and restored fork velocity in BS cells (although Another possibility is that BLM directly targets certain promoters, not to the same extent as CDA expression), demonstrating that the as reported for the WRN helicase . contribution of the CDA defect to the BS phenotype results from e Th key n fi dings of this work are that the defect in replication fork pyrimidine pool disequilibrium. es Th e results also make it pos - speed and, a fraction of the increase in SCE frequency in BS cells sible to distinguish between the replication fork velocity and fork are the consequences of a cascade reaction in which the defect in restart as two independent events, both ae ff cted by BLM helicase BLM helicase activity leads to a CDA expression defect that in turn deficiency, through nucleotide pool disequilibrium in the case of provokes a pyrimidine pool disequilibrium. Importantly, replication the replication fork velocity. fork progression defect can be rescued and the rate of SCEs reduced by 25–40% by acting on the nucleotide pool, without restoring BLM Discussion activity. Indeed, addition of dU, the n fi al CDA product, was found to This study demonstrates for the first time that BLM deficiency be suc ffi ient to restore the replication fork velocity and to decrease leads to CDA downregulation, resulting in pyrimidine pool dis- signic fi antly the level of SCEs in BLM-dec fi ient cells. e Th se data equilibrium and accounting, in part, for the BS phenotype. How- highlight the direct mechanistic connection between the replication ever, it remains unclear how BLM controls CDA expression. We defect, part of the increase in SCE and pyrimidine pool disequili- demonstrate here that the helicase activity of BLM regulates CDA brium in BS cells. Decrease in the fork velocity because of the nucleo- promoter function either directly or indirectly. The preferred tide pool disequilibrium may lead to replication stress and, thus, to n ATu RE Commun ICATIons | 2:368 | Do I: 10.1038/ncomms1363 | www.nature.com/naturecommunications © 2011 Macmillan Publishers Limited. All rights reserved. BS GFP-BLM GFP-BLMI841T GFP-BLM HeLaV-siCtrl HeLash-siBLM HEK293V-siCtrl HEK293sh-siBLM CDA mRNA relative quantity CDA mRNA relative quantity CDA mRNA relative quantity dATP pool dCTP pool 6 6 (pmol per 10 cells) (pmol per 10 cells) Normalized luciferase activity dGTP pool dTTP pool 6 6 (pmol per 10 cells) (pmol per 10 cells) ARTICLE n ATu RE Commun ICATIons | Do I: 10.1038/ncomms1363 50% 47.5% 45.2% a b 40% 70% 64.7% kDa 32.5% 60% 16 - CDA 30% 50% 44.8% 42 - β-Actin kDa 40% 20.0% 19.0% 19.0% 8 6.80 170- 20% 27.6% BLM 30% 20.6% 14.3% 20.7% 16- 6 CDA 20% 14.7% 10% 3.58 42- 6.9% β-Actin 4 10% 2.5% 0% 0% 0% BS-V BS-CDA BS-V BS-CDA GFP-BLM siCtrl GFP-BLM siCDA c d 84.4% 60% 80% 51.4% 50% 60.6% 40.5% 4.79 40% 60% 29.7% 30% 24.3% 2.74 18.9% kDa 40% 16.2% 18.9% 33.3% 20% 170- BLM 10% 16- CDA 15.6% 20% 0% 0 0% 42- β-Actin 6.1% GFP-BLM GFP-BLM+dC 0% GFP-BLM GFP-BLM+dC e f HeLaV-siCtrl/Ctrl HeLaV-siCtrl/CDA P = 0.0366 1.4 1.26 50% 45.8% 1.2 1 BS-V 0.885 41.2% 1.0 37.5% * 40% 0.8 1.100 BS-CDA 0.6 0.4 30% 26.5% GFP-BLM 1.066 0.2 17.6% 1.2 kDa 20% 0 0.2 0.4 0.6 0.8 1.0 BS-V BS-CDA 14.7% 14.6% –1 Replication fork velocity (kb min ) CDA 16- 10% 42- β-Actin 2.1% Figure 2 | CDA expression in BLM-deficient cells partially complements 0% BS phenotype. (a) dCTP pool level in Bs -V and Bs -CDA cells and CDA expression assessed by immunoblotting. Error bars represent the range of two independent experiments. (b) Percentage of metaphases with n s CE HeLash-siBLM/Ctrl HeLash-siBLM/CDA per chromosome from Bs -V and Bs -CDA cells. In all, 40–50 metaphases from three independent experiments were analysed for each condition. GFP-BLM (c) dCTP pool level in GFP-BLm cells untreated or treated with dC. Error 1.061 siCtrl bars represent the range of two independent experiments. (d) Percentage GFP-BLM 0.884 of metaphases with n s CE per chromosome from GFP-BLm cells untreated siCDA or treated with dC. In all, 40–50 metaphases from three independent 0.782 BS siCtrl experiments were analysed for each condition. (e) Bs -V or Bs -CDA cells were plated in triplicate at three dilutions (200, 400 and 600 cells). 0.770 BS siCDA Error bars represent the s.e.m. of three independent experiments. The significance of differences was assessed by s tudent’s t-tests. 0 0.2 0.4 0.6 0.8 1.0 1.2 –1 (f) Replication fork velocity in Bs -V, Bs -CDA and GFP-BLm cells. Error Replication fork velocity (kb min ) bars represent the range of two independent experiments. m ann–Whitney Figure 3 | siRNA-mediated CDA depletion partially reproduces BS statistical tests have been performed on the total number of fork analysed phenotype. (a) Percentage of metaphases with n s CE per chromosome from the two experiments (Bs -V, n = 880; Bs -CDA, n = 568; and GFP-BLm : from GFP-BLm cells transfected with the indicated siRn As. In all, 40–50 n = 800). Asterisk indicates a P-value < 0.00001. metaphases from three independent experiments were analysed for each condition. CDA levels were assessed by immunoblotting. (b) Percentage of an increase in replication-associated DNA breaks, accounting in metaphases with n s CEs per chromosome from HeLaV-siCtrl and HeLash- part for SCE formation. e Th se data indicate that the high frequency siBLm cells transiently transfected with the indicated siRn As. CDA protein of SCEs in BS cells has at least two die ff rent origins: 60–75% of levels were assessed by immunoblotting. About 40–50 metaphases SCEs may ree fl ct the restarting of blocked replication forks through from two independent experiments were analysed for each condition. RAD51-mediated homologous recombination , as proposed in (c) Replication fork velocity in GFP-BLm and Bs cells transfected with 9,21,22 several models . e Th remaining 25–40% of SCEs may ree fl ct the the indicated siRn As. Error bars represent the range of two independent repair of DNA breaks because of pyrimidine pool disequilibrium. i Th s experiments. m ann–Whitney statistical tests have been performed on second type of SCEs may also be dependent on RAD51-dependent the total number of forks analysed from the two experiments (GFP-BLm homologous recombination . i Th s study also demonstrates that siCtrl, n = 312; GFP-BLm siCDA, n = 452; Bs siCtrl: n = 459; and Bs siCDA, CDA downregulation in BLM-expressing cells leads to a signic fi ant n = 467). Asterisks indicate a P-value < 0.00001. increase in SCE frequency and replication stress, potentially confer- ring a predisposition to cancer. A previous French study reported low levels of CDA activity, rendering gemcitabine highly toxic . that 7% of adult patients with cancer treated with gemcitabine (5′- We therefore suggest that CDA dec fi iency may be a marker of azacytidine, 2′,2′-diu fl orodeoxycytidine), a CDA substrate, had very predisposition to cancer in the general population. n ATu RE Commun ICATIons | 2:368 | Do I: 10.1038/ncomms1363 | www.nature.com/naturecommunications © 2011 Macmillan Publishers Limited. All rights reserved. BS-V BS-CDA n < 1 1< n < 1.2 1.2 < n < 1.4 n > 1.4 n < 1 1 < n < 1.2 1.2 < n < 1.4 n > 1.4 GFP-BLM siCtrl GFP-BLM siCDA n < 0.3 0.3 < n < 0.4 0.4 < n < 0.5 n > 0.5 n < 0.3 0.3 < n < 0.4 0.4 < n < 0.5 n > 0.5 HeLaV-siCtrl/Ctrl 0 < n < 0.2 HeLaV-siCtrl/CDA 0.2 < n < 0.3 0.3 < n < 0.4 n > 0.4 0 < n < 0.2 0.2 < n < 0.3 0.3 < n < 0.4 n > 0.4 HeLash-siBLM/Ctrl HeLash-siBLM/CDA 0 < n < 0.2 0.2 < n < 0.3 0.3 < n < 0.4 0 < n < 0.2 0.2 < n < 0.3 0.3 < n < 0.4 0 < n < 0.2 0.2 < n < 0.3 0.3 < n < 0.4 n > 0.4 0 < n < 0.2 0.2 < n < 0.3 0.3 < n < 0.4 n > 0.4 dCTP pool dCTP pool Clonogenic survival 6 6 (pmol per 10 cells) (pmol per 10 cells) Percentage of metaphases Percentage of metaphases with n SCE per chromosome with n SCE per chromosome Percentage of metaphases Percentage of metaphases Percentage of metaphases with n SCE per chromosome with n SCE per chromosome with n SCE per chromosome n ATu RE Commun ICATIons | Do I: 10.1038/ncomms1363 ARTICLE P = 0.003 analysis of microarray expression experiments was performed as previously de- P = 0.0052 scribed using a P-value cutoff of 0.05. 1.6 1.415 Raw and normalized transcriptomic data are available at the Institut Curie GFP-BLM 1.066 microarray dataset repository (http://microarrays.curie.fr/. Direct link: http:// 1.2 0.879 GFP-BLM+dU 1.069 0.822 microarrays.curie.fr/publications/UMR3348/pyrimidine-bloom-syndrome/). 0.8 * BS 0.839 0.350 0.349 0.4 0.284 * Plasmid construction and site-directed mutagenesis. Site-directed mutagenesis 0.992 BS + dU was carried out in the EGFP-C1 vector containing the full-length BLM cDNA: 0 0.2 0.4 0.6 0.8 1.0 1.2 Ile-841 was mutated with the QuikChange XL site-directed mutagenesis kit (Strata- –1 Replication fork velocity (kb min ) gene) according to the manufacturer’s instructions. Symmetric replication bubble BS GFP-BLM Primer 1 (5′-CCCAGGGTACAGAAGGACACCCTGACTCAGCTGAAG-3′) IdU incorporation and primer 2 (5′-CTTCAGCTGAGTCAGGGTGTCCTTCTGTACCCTGGG-3′) 40% CldU incorporation 31.7% 30.9% were used. 30% 17.7% 18.5% 20% Reverse transcription and real-time quantitative PCR. Total RNA was extracted Asymmetric replication bubble with RNeasy Mini kit (Qiagen) including a DNAse digestion step. RNA quality was 10% IdU incorporation assessed with the Experion system (BioRad), and cDNAs were synthesized with CldU incorporation 0% 250 ng of random hexamers (Invitrogen), 2 µg of RNA and Superscript II reverse transcriptase (Invitrogen). Quantitative PCR experiments were performed accord- ing to the MIQE Guideline . Amplification mixtures contained the cDNA template (1/100 dilution), SYBR Green Supermix 1× (BioRad) and 300 nM forward and reverse primers. Amplification was performed with the CFX96 detection system Figure 4 | The BS phenotype is partially complemented by dU or T. (BioRad). The relative quantities of the CDA and BLM cDNAs were normalized (a) sCE frequencies in Bs and GFP-BLm cell lines untreated or treated with du against four reference genes (RPL32, HPRT1, HMBS and SDHA) chosen on the basis or T (about 1,800 chromosomes analysed for each condition). Error bars of their low M-value . The primer sequences for BLM, RPL32, HPRT1, HMBS and 28–30 represent the s.e.m. of three independent experiments. The significance of SDHA have been described in previous publications . CDA primer sequences: differences was assessed by s tudent’s t-tests. (b) Replication fork velocity primer 1 (5′-CCCTACAGTCACTTTCCTG-3′) and primer 2 (5′-CGGGTAGCAGG CATTTTCTA-3′). in GFP-BLm and Bs cell lines untreated or treated with du . Error bars represent the range of three independent experiments. m ann–Whitney Western blot analysis and antibodies. Cells were lysed in 350 mM NaCl, 50 mM statistical tests have been performed on the total number of forks analysed Tris–HCl, pH7.5, 1% NP-40 and protease inhibitors (Roche), sonicated and heated from the three experiments (GFP-BLm , n = 530; GFP-BLm + du , n = 642; Bs , at 70 °C for 10 min. Samples equivalent to 25 µg of protein were subjected to electrophoresis in NuPAGE Novex 4–12% Bis-Tris pre-cast gels (Invitrogen). The n = 858; and Bs + du , n = 440). Asterisks indicate a P-value < 0.00001. procedures used for gel electrophoresis and immunoblotting have been described (c) Percentage of asymmetric replication bubbles in Bs and GFP-BLm elsewhere . Primary and secondary antibodies were used at the following concen- cell lines untreated or treated with du , and examples of symmetric and trations: rabbit anti-BLM antibody (1:1,000; ab476 from Abcam); rabbit anti-CDA asymmetric bubbles (70–130 replication bubbles analysed for each antibody (1:500; ab56053 from Abcam); rabbit anti-β-actin antibody (1:10,000; condition). Cldu , chlorodeoxyuridine; Idu , iododeoxyuridine. Error bars Sigma); and horseradish peroxidase-conjugated goat anti-rabbit IgG (1:5,000; Santa Cruz Biotechnology). represent the range of two independent experiments. dNTP assays. Assays were performed as previously described . Briefly, dNTPs were extracted from fresh cell pellets (10–40×10 cells) in 0.6 M trichloroace- Methods Cell culture and treatments. Cell lines were cultured in DMEM supplemented tic acid and 15 mM MgCl . Extracts were neutralized and used immediately or stored at − 80 °C. For each dNTP assay, an oligonucleotide template primer was with 10% FCS. BS and GFP-BLM cells were obtained by transfecting BS GM08505B designed , and DNA polymerization reactions were carried out with Sequenase cells with the EGFP-C1 vector alone (Clontech) or with the same vector containing the full-length BLM cDNA , respectively, using JetPEI reagent (Ozyme). Ae ft r 48 h, (USB Corporation) and a radiolabeled nucleotide (Perkin Elmer) on samples or on − 1 dNTP standard solutions (0–4 pM). Samples were spotted on to DE81 membranes selection with 800–1,600 µg ml G418 (PAA) was applied. Individual colonies were − 1 (Whatman), which were washed with 5% (w/v) Na HPO (Sigma) and counted in a isolated and cultured in medium containing 500 µg ml G418. 2 4 liquid scintillation counter. GFP-BLMI841T cells were obtained by transfecting BS GM08505B cells as dTTP and dCTP pools were determined in GFP-BLM cells treated with 2 mM described above, with the EGFP-C1 vector containing the full-length BLM cDNA hydroxyurea (Sigma) for 2 h as a positive control for the dNTP assay (Supplemen- mutated at codon 841. tary Fig. S4). HeLaV and HeLashBLM cells, HEK293V and HEK293shBLM cells were obtained by transfecting cells with an empty pSM2 vector or with the same vector encoding a short hairpin RNA sequence directed against BLM (Open Biosystems, SCE assays. Cells were transferred to slides and cultured in the presence of 10 µM 5-bromodeoxyuridine (BrdU) (Sigma) during two cell divisions. Aer 40–48 ft h, clone V2HS-89234), respectively, using JetPEI reagent. Aer 48 ft h, selection with − 1 − 1 1–5 µg ml puromycin (Invivogen) was done. Individual colonies were isolated colchicine (Sigma) was added (0.1 µg ml ) and the cells were incubated for 1 h. − 1 and cultured in medium containing 0.5 µg ml puromycin. Cells were then incubated in hypotonic solution (1:5 (vol/vol) FCS-distilled water) and fixed with a 3:1 (vol/vol) mixture of methanol-acetic acid. Cells were then BS-V and BS-CDA cell lines were obtained by transfecting BS cells with an 25 − 1 stained by incubation with 10 µg ml Hoechst 33258 (Sigma) in distilled water for empty pCI-puro vector , or with the same vector containing the full-length CDA cDNA (NM001785), using JetPEI reagent. Aer 48 ft h, selection was carried out with 20 min, rinsed with 2×SSC (Euromedex), exposed to ultraviolet light at 365 nm at a − 1 − 1 distance of 10 cm for 105 min, rinsed in water, stained by incubation with 2% Giemsa 0.2 µg ml puromycin (Invivogen) and 500 µg ml G418 (Invitrogen). Individual − 1 solution (VWR) for 16 min, rinsed in water, dried and mounted. Chromosomes colonies were isolated and maintained in culture with 0.1 µg ml puromycin and − 1 500 µg ml G418. were observed with a Leica DMRB microscope at ×100 magnification. Metaphases 5 5 5 For siRNA transfection assays, 3×10 to 4×10 HeLa or HEK cells or 8×10 to were captured with a SONY DXC 930 P camera and SCEs were analysed. 9×10 BS or GFP-BLM cells were used to seed the wells of a six-well plate. Cells were transfected with an siRNA specic fi for BLM or CDA (ON-TARGETplus SMART - DNA molecular combing. Asynchronous populations of cells were labeled by pool, Dharmacon) or negative control siRNAs (ON-TARGETplus siCONTROL Non incubation for 40 min with 100 µM IdU (Sigma-Aldrich), washed with medium Targeting Pool, Dharmacon; 100 nM n fi al concentration) for 48 h for BLM, or twice at 37 °C and then labeled by incubation for 40 min with 100 µM CldU (Sigma- Aldrich). The DNA solution was prepared as previously described , and DNA was successively, for a total of 120 h for CDA using DharmaFect (Dharmacon). Deoxyribonucleosides (Sigma) were added to the cell culture medium at a final combed on silane-treated coverslips with a combing apparatus (Genomic Vision). concentration of 500 µM (dC) and 25 µM (dU, T) for 96 h (2×48 h). Coverslips with combed DNA were baked overnight at 60 °C and incubated in 0.5 M NaOH–1 M NaCl solution for 15 min, with gentle shaking, for DNA Transcriptomic analysis with Affymetrix microarrays. Analysis was performed denaturation. Coverslips were washed several times in PBS and dried by successive on BS, GFP-BLM, HeLaV-siCtrl and HeLash-siBLM cell lines. For each cell line, incubations, for 5 min each, in 70, 90 and 100% ethanol. They were then incubated three independent total RNA samples were extracted using the RNeasy system with the primary antibodies. All antibodies were diluted in BlockAid (Invitrogen). (Qiagen). Fragmented biotinylated antisense RNA was generated and hybridized to Coverslips were incubated at room temperature, first with 1/5 mouse anti-BrdU Affymetrix Human Genome U133 Plus 2.0 Arrays. Aer h ft ybridization, arrays were (BD Biosciences) + 1/25 rat anti-CldU (ABserotec) for 1 h, then with 1/25 goat anti- scanned following guidelines from Affymetrix (http://www.affymetrix.com). These mouse 488 + 1/25 Alexa Fluor goat anti-rat 555 antibodies (Invitrogen) for 40 min, arrays contain ~54,000 probesets, representing ~47,000 transcripts. Statistical followed by 1/50 mouse anti-ssDNA antibody (Millipore) for 40 min, then 1/25 n ATu RE Commun ICATIons | 2:368 | Do I: 10.1038/ncomms1363 | www.nature.com/naturecommunications © 2011 Macmillan Publishers Limited. All rights reserved. GFP-BLM GFP-BLM+dU BS BS+dU Untreated dU 25µM T 25µM Untreated dU 25µM T 25µM Percentage of No. of SCE per asymmetric bubbles chromosome ARTICLE nATuRE C ommunICATIons | DoI: 10.1038/ncomms1363 rabbit anti-mouse 350 antibody (Invitrogen) for 30 min and finally with 1/25 goat 19. Sun, H., Karow, J. K., Hickson, I. D. & Maizels, N. The Bloom’s syndrome anti-rabbit 350 antibody (Alexa Fluor, Invitrogen) for 30 min. Between incuba- helicase unwinds G4 DNA. J. Biol. Chem. 273, 27587–27592 (1998). tions, coverslips were washed three times, for 5 min each, in PBS. Coverslips were 20. Lachaud, A. A. et al. Werner’s syndrome helicase participates in transcription mounted in Prolong Gold antifade reagent (Invitrogen). Images were acquired with of phenobarbital-inducible CYP2B genes in rat and mouse liver. Biochem. a Leica DM RXA microscope equipped with a motorized XY stage, using a ×40 Pharmacol. 79, 463–470 (2010). PlanApo N.A. 1.25 objective and a CoolSNAP HQ interline CCD camera (Photo- 21. Bachrati, C. Z. & Hickson, I. D. RecQ helicases: guardian angels of the DNA metrics). For each slide, a mosaic of 10×10 partly overlapping images was collected replication fork. Chromosoma 117, 219–233 (2008). with a Metamorph software (Molecular Devices) routine developed in-house. 22. Wu, L. Role of the BLM helicase in replication fork management. DNA Repair Image collections were assembled into a mosaic with the ‘Stitching 2D/3D’ plugin (Amst) 6, 936–944 (2007). (available from http://fly.mpi-cbg.de/~preibisch/software.html) for ImageJ software 23. Ciccolini, J. et al. Cytidine deaminase residual activity in serum is a (Rasband, W.S., ImageJ, US National Institutes of Health, Bethesda, Maryland, predictive marker of early severe toxicities in adults aer g ft emcitabine-based USA, http://rsb.info.nih.gov/ij/, 1997–2009.) chemotherapies. J. Clin. Oncol. 28, 160–165 (2010). 24. Eladad, S. et al. Intra-nuclear trafficking of the BLM helicase to DNA damage- CDA promoter cloning and luciferase reporter assays. e Th CDA promoter induced foci is regulated by SUMO modification. Hum. Mol. Genet. 14, was inserted into a pGL4.14 vector (Promega), upstream from the r fi ey fl luciferase 1351–1365 (2005). gene. Cells were transfected by incubation for 72 h with 2.5 µg of the CDA 25. Girard, P. M., Kysela, B., Harer, C. J., Doherty, A. J. & Jeggo, P. A. Analysis of promoter-containing plasmid and 0.25 µg of pGL4.74 vector (Promega, internal DNA ligase IV mutations found in LIG4 syndrome patients: the impact of two control, Renilla luciferase) in the presence of JetPEI reagent. Firey fl luciferase and linked polymorphisms. Hum. Mol. Genet. 13, 2369–2376 (2004). Renilla luciferase activities were assessed with the Dual-Luciferase Reporter Assay 26. Irizarry, R. A. et al. Exploration, normalization, and summaries of high density System protocol (Promega). oligonucleotide array probe level data. Biostatistics 4, 249–264 (2003). 27. Bustin, S. A. et al. The MIQE guidelines: minimum information for publication Clonogenic survival assays. Cells were plated in a drug-free medium at three of quantitative real-time PCR experiments. Clin. Chem. 55, 611–622 (2009). different densities, in triplicate, for the counting of 30–300 clones, depending 28. Vandesompele, J. et al. Accurate normalization of real-time quantitative RT- on expected survival. Aer 14–21 d ft ays of incubation, colonies were fixed and PCR data by geometric averaging of multiple internal control genes. Genome − 1 stained with methylene blue (5 g l in 50% water and 50% methanol) and scored. Biol. 3, RESEARCH0034 (2002). Only experiments giving a linear correlation between the different dilutions were 29. Bischof, O. et al. Regulation and localization of the Bloom syndrome protein in considered. Colony-forming efficiency was estimated by dividing the number of response to DNA damage. J. Cell. Biol. 153, 367–380 (2001). colony-forming units by the number of cells plated. 30. Viegas, M. H., Gehring, N. H., Breit, S., Hentze, M. W. & Kulozik, A. E. The abundance of RNPS1, a protein component of the exon junction complex, Statistical analysis. The significance of differences was assessed with Student’s can determine the variability in efficiency of the Nonsense Mediated Decay t-test or Mann–Whitney test. P < 0.05 was considered statistically significant. pathway. Nucleic Acids Res. 35, 4542–4551 (2007). 31. Sherman, P. A. & Fyfe, J. A. Enzymatic assay for deoxyribonucleoside References triphosphates using synthetic oligonucleotides as template primers. Anal. 1. German, J. Bloom’s syndrome. XX. The first 100 cancers. Cancer Genet. Biochem. 180, 222–226 (1989). Cytogenet. 93, 100–106 (1997). 32. Preibisch, S., Saalfeld, S. & Tomancak, P. Globally optimal stitching of tiled 3D 2. Ellis, N. A. et al. The Bloom’s syndrome gene product is homologous to RecQ microscopic image acquisitions. Bioinformatics 25, 1463–1465 (2009). helicases. Cell 83, 655–666 (1995). 33. Fitzgerald, S. M. et al. Identification of functional single nucleotide 3. Bennett, R. J. & Keck, J. L. Structure and function of RecQ DNA helicases. polymorphism haplotypes in the cytidine deaminase promoter. Hum. Genet. Crit. Rev. Biochem. Mol. Biol. 39, 79–97 (2004). 119, 276–283 (2006). 4. Davies, S. L., North, P. S. & Hickson, I. D. Role for BLM in replication-fork restart and suppression of origin firing aer r ft eplicative stress. Nat. Struct. Mol. Acknowledgements Biol. 14, 677–679 (2007). We thank N. Ellis for providing us with the GFP-BLM construct, F. Cordelières for help 5. Rao, V. A. et al. Endogenous gamma-H2AX-ATM-Chk2 checkpoint activation in automating DNA combing analysis, B. Albaud and D. Gentien from the Affymetrix in Bloom’s syndrome helicase deficient cells is related to DNA replication Platform for microarray experiments, P. Hupé and N. Servant from the Bioinformatic arrested forks. Mol. Cancer Res. 5, 713–724 (2007). Platforms for helping us to analyse transcriptomic data. We also thank P-M. Girard and 6. Chaganti, R. S., Schonberg, S. & German, J. A manifold increase in sister D. Graindorge for advice about molecular combing and the members of UMR 3348 chromatid exchanges in Bloom’s syndrome lymphocytes. Proc. Natl Acad. Sci. for helpful discussions. This work was supported by grants from the Institut Curie, the USA 71, 4508–4512 (1974). CNRS, the Ligue contre le Cancer (Comité de l’Essonne) and by a fellowship awarded 7. Bartkova, J. et al. DNA damage response as a candidate anti-cancer barrier in to P.C. by the CNRS (BDI), the Ligue contre le Cancer (Comité de l’Essonne) and early human tumorigenesis. Nature 434, 864–870 (2005). Association pour la Recherche sur le Cancer. 8. Gorgoulis, V. G. et al. Activation of the DNA damage checkpoint and genomic instability in human precancerous lesions. Nature 434, 907–913 (2005). 9. Amor-Gueret, M. Bloom syndrome, genomic instability and cancer: the SOS- Author contributions like hypothesis. Cancer Lett. 236, 1–12 (2006). P.C. performed the transcriptomic analysis and all experiments, unless otherwise 10. Meuth, M. The molecular basis of mutations induced by deoxyribonucleoside indicated, participated in the design of experiments and data analysis and generated the triphosphate pool imbalances in mammalian cells. Exp. Cell. Res. 181, 305–316 (1989). figures. G.B.L. performed the work presented in Figures 2b,d and 4a. R.O.D. performed 11. Kunz, B. A. et al. International Commission for Protection Against the work presented in Figure 1b and in Supplementary Figure S1, CDA cloning and Environmental Mutagens and Carcinogens. Deoxyribonucleoside triphosphate established GFP-BLM, GFP-BLMI841T, BS-CDA and BS-V cell lines, and participated levels: a critical factor in the maintenance of genetic stability. Mutat. Res. 318, in the work in Figure 2c. S.L. and M.D. contributed to data analysis, O.B. contributed to 1–64 (1994). the work presented in Figure 1c and in Supplementary Figure S1. M.A.-G. designed the 12. Evans, D. R. & Guy, H. I. Mammalian pyrimidine biosynthesis: fresh insights experiments, analysed the data and wrote the manuscript. into an ancient pathway. J. Biol. Chem. 279, 33035–33038 (2004). 13. Lahkim Bennani-Belhaj, K et al. The Bloom syndrome protein limits the lethality associated with RAD51 deficiency. Mol. Cancer Res. 8, 385–394 (2010). Additional information 14. Nygaard, P. On the role of cytidine deaminase in cellular metabolism. Adv. Exp. Supplementary Information accompanies this paper on http://www.nature.com/ Med. Biol. 195 (Pt B), 415–420 (1986). naturecommunications 15. Guo, R. B. et al. Structural and functional analyses of disease-causing missense Competing financial interests: M.A.-G is the sole inventor of patent application mutations in Bloom syndrome protein. Nucleic Acids Res. 35, 6297–6310 (2007). ‘Low levels of cytidine deaminase as a marker for predisposition to develop cancer’, 16. Kunz, B. A. Mutagenesis and deoxyribonucleotide pool imbalance. Mutat. Res. PCT/EP2011/050784. The remaining authors declare no competing financial interests. 200, 133–147 (1988). 17. Warren, S. T., Schultz, R. A., Chang, C. C., Wade, M. H. & Trosko, J. E. Elevated Reprints and permission information is available online at http://npg.nature.com/ spontaneous mutation rate in Bloom syndrome fibroblasts. Proc. Natl Acad. Sci. reprintsandpermissions/ USA 78, 3133–3137 (1981). 18. Mohaghegh, P., Karow, J. K., Brosh, R. M. Jr., Bohr, V. A. & Hickson, I. D. The How to cite this article: Chabosseau, P. et al. Pyrimidine pool imbalance induced by Bloom’s and Werner’s syndrome proteins are DNA structure-specific helicases. BLM helicase deficiency contributes to genetic instability in Bloom syndrome. Nucleic Acids Res. 29, 2843–2849 (2001). Nat. Commun. 2:368 doi: 10.1038/ncomms1363 (2011). nATuRE C ommunICATIons | 2:368 | DoI: 10.1038/ncomms1363 | www.nature.com/naturecommunications © 2011 Macmillan Publishers Limited. All rights reserved.

Journal

Nature Communications – Springer Journals

Published: Jun 28, 2011

There are no references for this article.