Research article 2565 A Werner syndrome homolog affects C. elegans development, growth rate, life span and sensitivity to DNA damage by acting at a DNA damage checkpoint Se-Jin Lee, Jong-Sung Yook, Sung Min Han and Hyeon-Sook Koo* Department of Biochemistry, College of Science, Yonsei University, Seoul 120-749, Korea *Author for correspondence (e-mail: [email protected])

Accepted 18 February 2004

Development 131, 2565-2575 Published by The Company of Biologists 2004 doi:10.1242/dev.01136

Summary A Werner syndrome protein homolog in C. elegans (WRN- irrespective of γ-irradiation, and pre-meiotic germ cells had 1) was immunolocalized to the nuclei of germ cells, an abnormal checkpoint response to DNA replication embryonic cells, and many other cells of larval and adult blockage. These observations suggest that WRN-1 acts as worms. When wrn-1 expression was inhibited by RNA a checkpoint protein for DNA damage and replication interference (RNAi), a slight reduction in C. elegans life blockage. This idea is also supported by an accelerated S span was observed, with accompanying signs of premature phase in wrn-1(RNAi) embryonic cells. wrn-1(RNAi) aging, such as earlier accumulation of lipofuscin and phenotypes similar to those of Werner syndrome, such as tissue deterioration in the head. In addition, various premature aging and short stature, suggest wrn-1-deficient developmental defects, including small, dumpy, ruptured, C. elegans as a useful model organism for Werner transparent body, growth arrest and bag of worms, were syndrome. induced by RNAi. The frequency of these defects was accentuated by γ-irradiation, implying that they were derived from spontaneous or induced DNA damage. Key words: Werner syndrome, Disease model, Aging, DNA damage, wrn-1(RNAi) worms showed accelerated larval growth Checkpoint

Introduction dependent pathway is involved in the accelerated cellular Werner syndrome (WS) is associated with a rapid acceleration caused by the absence of WRN. WS cells are also of aging, and is caused by mutations in the RecQ family DNA hypersensitive to certain DNA damaging agents, including the helicase wrn (Yu et al., 1996). wrn encodes a protein with chemical carcinogen 4-NQO (Ogburn et al., 1997), camptothecin (Poot et al., 1999), and DNA cross-linking agents a central domain of seven helicase motifs and two conserved (Poot et al., 2001). domains (RQC and HRDC) located C-terminal to the helicase WRN has been shown to form complexes with domain (Morozov et al., 1997). WRN differs from other involved in cellular responses to DNA damage and in DNA members of the RecQ family in that it possesses an unusual ′→ ′ replication. The identification of a functional interaction exonuclease domain homologous to the 3 5 exonuclease between WRN and the p53 tumor suppressor protein has domain of E. coli DNA polymerase I and of RNaseD (Moser emphasized the role of the RecQ family in maintaining et al., 1997; Huang et al., 1998; Shen et al., 1998; Suzuki et genomic stability (Spillare et al., 1999). WRN also al., 1999). Nevertheless, when mutated, WRN causes genomic dramatically stimulates the cleavage reaction catalyzed by the instability disorders associated with an elevated risk of cancer, human 5′ flap endonuclease/5′→3′ exonuclease FEN1 (Brosh short stature, and/or premature aging, like two other RecQ et al., 2001), a DNA structure-specific nuclease implicated family DNA helicases in humans: the Bloom syndrome (BS) in DNA replication and repair (Lieber, 1997). The ability and the Rothmund¯Thomson syndrome (RTS, RecQL4) of replication protein A (RP-A; RPA1 – Human Gene proteins (Mohaghegh and Hickson, 2001). Nomenclature Database) to stimulate the unwinding of long Several defects at the cellular level have been detected in stretches of DNA duplex by WRN helicase suggests that WRN WS. Cells cultured from WS patients have a reduced life span, may function in replication, a notion supported by interaction an extended S phase, and reduced RNA transcription by RNA of WRN with other replication proteins. Recent evidence polymerases I and II (Martin et al., 1970; Salk et al., 1985; points to a direct protein interaction between WRN and the Balajee et al., 1999; Shiratori et al., 2002). Although no direct Ku80/70 heterodimer implicated in non-homologous end- role of WRN has been established in metabolism, WS joining of double-strand breaks (Cooper et al., 2000). In fibroblasts expressing a transfected human telomerase (TERT) addition to these functional interactions, WRN has been gene have an increased life span and can be immortalized reported to physically interact with human polymerase delta (Wyllie et al., 2000). These results suggest that a telomerase- (Szekely et al., 2000), PCNA and DNA topoisomerase I (Label 2566 Development 131 (11) Research article et al., 1999). These interactions suggest that WRN is a central The overexpressed protein band was excised, crushed in PBS, mixed player in a macromolecular complex essential for DNA with Freund’s adjuvant, and then injected into Balb/c mice four times replication or repair. at weekly intervals (200 µg protein per injection). In C. elegans, four RecQ family proteins are predicted from Immunostaining the genomic DNA sequence. Of these proteins, the one encoded by the open reading frame (ORF) F18C5.2 is most C. elegans embryos were immunostained by a slightly modified version of the procedure of Crittenden and Kimble (Crittenden and homologous to human WRN. To understand the role of this C. Kimble, 1999). Embryos were freeze-cracked, fixed, incubated with elegans WRN homolog (WRN-1), we localized the protein in polyclonal mouse antiserum against the N-terminal 209 amino acids C. elegans and investigated the phenotypes arising from of WRN-1 (1:25 dilution), and then with FITC-conjugated goat anti- inhibited expression. mouse immunoglobulin G (1:500 dilution, Santa Cruz Biotechnology) pre-treated with C. elegans acetone powder. After being stained with Materials and methods DAPI (4,6-diamidino-2-phenylindole, 1 µg/ml), specimens were observed with a fluorescence microscope (DMR HC, Leica). Gonads Materials and intestines were extruded by decapitating adult C. elegans, fixed Bristol N2, as a standard wild-type strain, and div-1(or148ts) strains in 3% paraformaldehyde, then immunostained as described (Jones et were obtained from the C. elegans Genetics Center (St Paul, MN, al., 1996). Whole-worm staining was carried out by the collagenase USA). An EST clone of the Ce-wrn-1 gene (yk41c3) was provided method of Nonet et al. (Nonet et al., 1993). After fixation in 4% by Dr Y. Kohara (National Institute of Genetics, Japan). paraformaldehyde, worms were incubated in a reducing solution [5% Deoxynucleotide oligomers were synthesized at Genotech (Korea). β-mercaptoethanol, 1% Triton X-100, 0.1 M Tris-Cl (pH 6.9)] at 37°C overnight, and then reacted with collagenase (1000 units/ml, Sigma) ′ Cloning of a 5 cDNA fragment of C. elegans wrn-1 by RT- in buffer [0.1 M Tris-Cl (pH 7.5), 1 mM CaCl2] at 37°C for 5 hours. PCR and construction of full-length cDNA Subsequent reactions with primary and secondary antibodies were as The EST clone yk41c3 of the F18C5.2 ORF lacked the first exon (115 described above for embryos. nucleotides; nt) of the predicted ORF of 16 exons. Therefore, to obtain a 5′-terminal cDNA clone, we isolated C. elegans total RNA using Inhibition of wrn-1 expression by double-stranded RNA an RNeasy kit (Qiagen). cDNA synthesis progressed in a reaction microinjection mixture (50 µl) containing C. elegans total RNA (3 µg), a primer The pCeWRN recombinant plasmid was linearized with BamHI and (10 pmoles) of sequence 5′-GTGGACATAAGAACAAATTGGTC-3′ ApaI restriction enzymes at its multicloning site to prepare antisense (nt 752-729 in the ORF) from exon 3, and Superscript reverse and sense transcripts of wrn-1, respectively. Antisense RNA was transcriptase II (200 units, Stratagene), at 42°C for 1 hour. First cDNA synthesized using BamHI-digested plasmid DNA (2 µg), T7 RNA strand synthesis was terminated by heating at 70°C, and then template polymerase (5 units, MBI), ribonucleoside triphosphates (rNTPs, 0.4 RNA was degraded by RNase H (2 units, Takara). A cDNA fragment mM each) and RNase inhibitor (5 units, Takara) in buffer [40 mM Tris- was amplified from the first cDNA strand by PCR (polymerase chain Cl (pH 8.0), 8 mM MgCl2, 2 mM spermidine, 50 mM NaCl, 18 mM reaction) using the SL1 primer (5′-GGTTTAATTACCCAAGT- DTT; total 50 µl], at 37°C for 2 hours. Sense RNA was synthesized TTGAG-3′) and a primer of sequence 5′-CATTTCTGACAACATCC- under the same reaction conditions as described for antisense RNA, CACTG-3′ (nt 715-694 in the ORF) from exon 3. The amplified except for the use of ApaI-treated DNA (2 µl) and T3 RNA polymerase cDNA fragment was cloned into pGEM-T vector (Promega) and (5 units, MBI). After RNA synthesis, RNase-free DNase I (2 units) sequenced with an ABI PRISM Dye Terminator Cycle Sequencing was added to degrade template DNA, and then phenol (pH 4.5) Ready Reaction Kit (Perkin-Elmer). extraction and ethanol precipitation were carried out. An equivalent To obtain a full-length cDNA construct of wrn-1, PCR was carried amount of sense and antisense RNAs were mixed to a total out using two primers: 5′-CGCGGATCCATGATAAGTGATGAT- concentration of 1 µg/µl, and then microinjected into the intestines of GACGATC-3′, containing nt 1-22 of the ORF, and a BamHI young adult N2 worms. The worms were placed on an NGM plate with recognition sequence (underlined); and 5′-CATTTCTGACAACATC- an E. coli OP50 lawn, and were transferred to new plates after 12 hours. CCACTG-3′, corresponding to nt 715-694 of the ORF. The amplified DNA product was digested with BamHI and HindIII (nt 641-646 in the Measurement of life span ORF) restriction enzymes, electrophoresed on a 1% agarose gel, and Twelve hours after microinjection, the worms were allowed to lay then eluted from the gel using a DNA extraction kit (Intron embryos for 6 hours. F1 progeny (>100) were grown at 20°C or 25°C, Biotechnology, Korea). The purified cDNA fragment was inserted into and transferred to fresh plates every one or two days. Death was plasmid yk41c3, previously digested with BamHI and HindIII, to scored by absence of any movement after several light pokes with a yield pCeWRN. platinum wire. Antibody preparation Phenotypic analysis A 5′-terminal cDNA fragment was amplified from pCeWRN by PCR F1 progeny of the microinjected worms, designated Ce-wrn-1(RNAi) using primers: nt 1-22 of the ORF with a BamHI recognition worms, were grown at 20°C or 25°C. Over 500 F1 worms were sequence; and nt 627-609 with a SalI recognition sequence examined daily, with a stereomicroscope or with Nomarski optics (underlined), 5′-CAGCTGCGTCATTGATGCCCACTTC-3′. The (DMR HC, Leica), from the L1 stage. Worms with abnormal cDNA fragment was cloned into pGEM-T, excised from the phenotypes were counted to 8 days old. The wild-type N2 strain was recombinant T-vector, and then inserted between the BamH1 and SalI used as a control instead of a mock-RNAi strain, as the phenotypes sites of the pMAL-c2 overexpression vector (New England Biolabs). produced by microinjection of dsRNA derived from a 5′-upstream E. coli XL1-Blue cells harboring pMAL/CeWRN were cultured DNA sequence were the same as wild type. at 37°C to O.D.600nm 0.5, and isopropyl-thio-β-D-galactoside (Calbiochem) was added to 1 mM. After further incubation for 3 C. elegans sensitivity to DNA damage hours, the cells were harvested and sonicated in 10 ml PBS with 10 F1 larvae at the L1 stage, derived from the microinjected P0 137 mM Na2HPO4, 2 mM KH2PO4 (pH 7.4), 137 mM NaCl and 2.7 mM worms, were γ-irradiated with a Cs source (IBL 437C, CIS KCl. Cell lysate was microcentrifuged at 14,000 rpm for 5 minutes, Biointernational) at 60 Gray (Gy). After being kept at 20°C for 3 days, and the supernatant electrophoresed on a 7% SDS polyacrylamide gel. over 200 worms were examined under a stereomicroscope or with Werner syndrome in C. elegans 2567

Exonuclease Acidic Helicase motif RQC HRDC NLS a.a. domain domain hWRN 1432

31% Identify 43% 25% Ce-WRN-1 1056 Fig. 1. Comparison of C. elegans WRN-1 with human Werner syndrome protein. WRN-1 shares 43% identity in amino acid sequence in the helicase motif and 25% identity in the RQC-HRDC region with human WRN. RQC, RecQ conserved domain; HRDC, helicase RNase D C- terminal conserved domain; NLS, nuclear localization signal.

Nomarski optics. Wild-type N2 worms were also γ-irradiated and their to human WRN; it has therefore been named as wrn-1 in phenotypes examined as a γ-irradiation control. In order to measure WormBase (http://www.wormbase.org; Fig. 1). As the yk41c3 larval growth rate at 20°C, over 150 L1 stage worms were γ-irradiated EST clone was shorter than the predicted ORF at its 5′-end, a at 0, 10 or 20 Gy. Subsequently, worms reaching the L4 stage, as 5′-end cDNA clone was amplified by reverse transcription- defined by vulva shape, were scored every 12 hours. polymerase chain reaction (RT-PCR) of C. elegans total RNA, Aging phenotypes induced by bacteria-mediated RNAi of using gene-specific primers and an SL1 primer. The trans- Ce-wrn-1 splice leader sequence of SL1 (22 nucleotides) is found in two- The EST clone yk41c3 of the F18C5.2 ORF was digested with NotI thirds of C. elegans mRNAs and indicates that precursor and ApaI enzymes, and inserted into the pPD129.36(L4440) vector mRNAs have been trans-spliced with SL1 RNA (Conrad et al., (Timmons and Fire, 1998), which contains two convergent T7 1995). The 5′-end cDNA clone was fused to yk41c3 at a polymerase promoters in opposite orientations, separated by the HindIII site (nt 641-646 in the ORF) to construct a full-length multicloning site. Plasmid DNA was transformed into E. coli cDNA clone (pCeWRN) encoding 1056 amino acids. However, HT115(DE3) (W3110, rnc14::∆Tn10) cells by electroporation two EST clones of wrn-1 (yk1276b08 and yk811e06), which (Invitrogen). Cells harboring plasmid DNA were directly applied onto have trans-splice leader sequences of SL2 RNA (5′- agar plates, composed of standard NGM/agar medium supplemented ′ µ µ GGTTTTAACCCAGTTACTCAAG-3 ) and a novel SL RNA with 100 g/ml ampicillin, 12.5 g/ml tetracycline and 0.4 mM IPTG, ′ ′ ′ and then cultured overnight at room temperature. L4 stage N2 worms (5 -GTTTTTAACCCAGTTAATTGAG-3 ) at the 5 -end, were grown to adults on the plate covered with E. coli cells producing respectively, have recently been reported in WormBase. dsRNA of wrn-1, allowed to lay embryos for 1 hour, and then removed These two EST clones together with our 5′-end cDNA clone from the plate. The embryos were incubated at 20°C until they indicate that the pre-mRNAs of wrn-1 are trans-spliced at the reached the L4 larval stage, followed by a temperature shift to 25°C. same splice junction with three different SL RNAs. Trans- Twenty-four hours later, autofluorescence of adult worms was splicing with SL2 RNA means that the pre-mRNA is co- photographed using a fluorescence microscope (525 nm filter), and transcribed from an upstream gene (Evans et al., 2001), most their heads were observed with Nomarski optics for 7 days. Control likely from the F18C5.3 gene, which has a mRNA worms were fed with non-transformed HT115(DE3) E. coli cells. polyadenylation site about 100 nucleotides away from the Time-lapse microscopy of embryonic cell divisions trans-splice site of wrn-1. The F18C5.3 ORF is highly The EST clone yk1302e07 of chk-1 (Y39H10A.7) was digested with homologous to human DRIM (down-regulated in metastasis) XhoI and inserted into the pPD129.36(L4440) vector. On the feeding- protein, the molecular function of which is not clearly known plate covered with E. coli cells producing dsRNA of wrn-1, chk-1, or (Schwirzke et al., 1998). both, wild-type N2 or div-1(or148ts) worms at the L4 stage were placed and grown to adults at 25°C. The adult worms were dissected C. elegans wrn-1 encodes a DNA helicase similar to to isolate 2-cell-stage F1 embryos, which were then observed human WRN microscopically with Nomarski optics at appropriate intervals. In All RecQ family proteins have a helicase domain of 200-300 order to measure the duration of S and M phase in the early embryos, amino acids that contains five to seven motifs that are embryos were photographed every 10 seconds by time-lapse conserved among all helicases. The helicase domain with a microscopy (Leica IM 1000). DEAH box in the WRN-1 ORF shows 43% identity in the Hydroxyurea treatment of the germ line amino acid sequence with human WRN (Fig. 1). In addition, On the feeding-plate covered with E. coli cells producing dsRNA of the C-terminal sequence succeeding the helicase domain, wrn-1, wild-type N2 worms were grown from L1 to L4 stages at 25°C. which consists of an RQC (RecQ helicase conserved) domain L4 stage worms were transferred to a new feeding-plate containing and an HRDC (Helicase and RNase D, C-terminal conserved) 25 mM hydroxyurea and were dissected to isolate the gonads 12 hour domain, share 25% sequence identity with human WRN. One later. After staining with DAPI, the gonads were observed using a major difference between the two WRN proteins is the absence fluorescence microscope. of an exonuclease domain in C. elegans WRN-1. The exonuclease domain in human WRN is most homologous to Results ZK1098.3 and ZK1098.8 (mut-7) of the C. elegans genomic ORFs, suggesting the possibility that one of these proteins is Construction of a full-length cDNA clone of associated with C. elegans WRN-1. C. elegans wrn-1 Of the four RecQ family proteins predicted by the C. elegans WRN-1 localization in C. elegans genomic DNA sequence, ORF F18C5.2 is most homologous WRN-1 was immunolocalized in C. elegans at various 2568 Development 131 (11) Research article developmental stages (Fig. 2). The protein was non-uniformly numerous cells, and the fraction of WRN-1-positive somatic distributed in nuclei from the early embryonic stage throughout cells was significantly lower in adults (Fig. 2B). In adult embryogenesis. In early embryos, the protein level was higher worms, the protein was expressed in hypodermal, intestinal and in mitotic cells than in interphase cells, as is clearly shown in germ cells, and the protein level decreased with age. When 6- and 8-cell stage embryos (mitotic cells marked with nuclei in the intestine were magnified, an uneven distribution arrowheads or an arrow). At metaphase, WRN-1 had an of the protein was observed (Fig. 2C). Both mitotic and meiotic unusual location, overlapping with the periphery of prophase germ cells in the L4 stage gonad contained WRN-1 equatorially aligned condensed facing the in the nuclei (Fig. 2C), but the level was significantly reduced spindle poles (see the cell marked with an arrow in Fig. 2A, in the meiotic prophase germ cells of adults (data not shown). and the metaphase cell in Fig. 2D), and also, less frequently, In the oocytes of adult gonads, WRN-1 localized to condensed overlapping with the mitotic spindles (Fig. 2A, arrow). This chromosomes (Fig. 2D). RNAi of wrn-1 eliminated localization to one side of a mitotic sister chromatid was immunologically detectable protein in embryos, supporting observed for CENP-A and CENP-C homologs in C. elegans effective suppression of the endogenous and embryos, which are centromere-binding proteins of holocentric no cross-reactivity of the antibody (Fig. 2A). By contrast, C. elegans chromosomes (Moore and Roth, 2001), and also for RNAi of a RecQ homolog (RCQ-5) encoded by the E03A35.2 SAN-1, which is involved in the spindle checkpoint (Nystul et ORF, with 39% amino acid identity in the helicase domain of al., 2003). WRN-1, did not affect WRN-1 expression, thus underlining the In larval stages WRN-1 was present in the nuclei of specificity of wrn-1 RNAi (S.-J.L., unpublished).

Fig. 2. Immunolocalization of WRN- 1 in C. elegans at various developmental stages. C. elegans embryos or worms were reacted with a mouse antiserum against WRN-1 (green) and then with an FITC- conjugated secondary antibody, followed by nuclear staining with DAPI (blue). (A) C. elegans embryos. Mitotic prophase nuclei, two nuclei in the 6-cell embryo and one nucleus in the 8-cell embryo, are marked with arrowheads; a metaphase nucleus in the 8-cell embryo is marked with an arrow. (B) L1 larval and adult stage worms. (C) Intestine and gonad from an L4 stage larva. (D) An oocyte and an embryonic metaphase cell. Scale bars: A, 10 µm; B, 50 µm; C, 50 µm; D, 5 µm for the oocyte and 1 µm for the metaphase cell. Werner syndrome in C. elegans 2569

Life span is reduced in wrn-1(RNAi) worms small body size, bag of worms, ruptured body, dumpy shape, To assess the in vivo function of WRN-1, RNA interference growth arrest at larval stages and transparent body (Fig. 5A). (RNAi) was carried out by microinjecting double-stranded These abnormal phenotypes were about three-fold more RNA (dsRNA) of the wrn-1 gene into young adult worms (P0). frequent in wrn-1(RNAi) worms than in wild-type worms (Fig. When the progeny (F1) wrn-1(RNAi) worms were grown at 5C). Some phenotypes were induced by RNAi even at 20°C, 20°C, their life span was not affected: wild type and wrn- but at a much lower frequency than at 25°C (Fig. 5B). 1(RNAi) worms lived for 17.1 (±0.2) and 16.8 (±0.2) days after Small body, dumpy and growth arrest are similar to the short birth, respectively (J.-S.Y., unpublished). By contrast, RNAi stature characteristics of human Werner syndrome. Small body significantly reduced their life span at 25°C: the life span was was designated as the adult body length less than 70% of wild 11.0 (±0.2) days for wrn-1(RNAi) worms and 13.6 (±0.1) days type. Dumpy worms had short and fat bodies. Ruptured body, for wild-type worms (P<0.001; Fig. 3). In addition, the brood with internal organs such as the intestine and gonad bursting size of wrn-1(RNAi) worms was reduced to 94% of wild type out of the vulva, followed on from a protruding vulva. This at 20°C, and to 84% at 25°C (J.-S.Y., unpublished), whereas phenotype usually occurs because of defects in vulval embryonic hatching was unaffected at either temperature. morphogenesis (Hurd and Kemphues, 2003). The bag of worms phenotype, i.e. an egg-laying defect resulting in many wrn-1(RNAi) worms have progeroid tissue hatched larvae remaining inside the worm, was probably phenotypes caused by defects in HSN neurons producing serotonin, or in To determine whether the reduced life span of wrn-1(RNAi) chemosensory neurons, as the defect was relieved by serotonin worms was due to progeria or to sickness, symptoms of normal treatment (S.-J.L., unpublished). However another possibility aging, such as lipofuscin accumulation and tissue deterioration is a partial defect in other egg-laying systems, such as vulva in the head (Garigan et al., 2002), were probed for the and sex muscles. In addition, transparent body lacking gut worms fed with E. coli cells producing dsRNA of wrn-1. granules (Fitzgerald and Schwarzbauer, 1998) was induced at Autofluorescence from intestinal lipofuscin increased with age a very low frequency by the RNAi. As the small or dumpy in adult worms, as shown in Fig. 4A, and wrn-1(RNAi) worms aspect of wrn(RNAi) worms was not inherited by their had stronger autofluorescence than wild type at the same adult progeny, these phenotypes probably were not caused by germ- stage. When the two strains were compared for overall line mutations. The combined frequency of bag of worms, autofluorescence intensity in a worm, wrn-1(RNAi) adults ruptured body and growth arrest in wrn-1(RNAi) worms was advanced in lipofuscin accumulation by about 2 days relative 6% higher than in wild type and may have contributed to the to wild type. Adult C. elegans heads were photographed as in 19% reduction in life span at 25°C (Fig. 3). Nevertheless, the Fig. 4B, in order to assign scores of 1-5 depending on the small or dumpy phenotype was generally not accompanied by extent of tissue damage, such as cavity formation and premature death, suggesting that the reduced life span mainly pharyngeal clogging due to bacterial packing. From the scatter resulted from accelerated aging. diagrams of Fig. 4C, representing tissue deterioration in the head, wrn-1(RNAi) worms are estimated to age about one day The RNAi phenotypes of wrn-1 were enhanced by faster than wild type. ionizing radiation L1 stage larvae were γ-irradiated and their growth to adults was wrn-1(RNAi) phenotypes are similar to symptoms of examined after 3 days. As shown in Fig. 5D, γ-radiation greatly human Werner syndrome increased the frequency of the phenotypes listed in Fig. 5A Besides reduced life span, wrn-1(RNAi) worms at 25°C even at 20°C, both in wild type and wrn-1(RNAi) strains. showed an increased frequency of abnormal features such as Among these phenotypes, small body and bag of worms appeared in 40% and 35% of wrn-1(RNAi) worms, respectively, much higher than the values (close to 10% for 100 N2 each) for wild type. These large increases in the frequency of abnormal phenotypes must be due to enhancement of the RNAi wrn- 1 (RNAi ) effects by γ-radiation, given the much lower frequency (<2%) ) 80

% of these phenotypes in the absence of ionizing radiation (at (

l 60 20°C). This finding suggested that WRN-1 participates in a

v cellular responses to DNA damage and that the exacerbated i

v 40

r developmental defects in its absence probably resulted from u defective cellular signaling or/and DNA repair. The fact that S 20 the dumpy phenotype, unlike the other phenotypes listed in Fig. 5, was not significantly increased by ionizing radiation is 0 0 5 10 15 20 25 very intriguing, as it points to a distinctive role of elevated metabolic rate at a higher temperature in inducing dumpiness. Days The rapid larval growth of wrn-1(RNAi) worms is not Fig. 3. Life span of F1 progeny produced by P0 C. elegans worms affected by ionizing radiation microinjected with double-stranded RNA of wrn-1. The life span of wrn-1(RNAi) worms was reduced by 2.6 (P<0.001) days compared As the enhancement of RNAi phenotypes by ionizing radiation with wild-type N2 strain at 25°C. wrn(RNAi), 11.0±0.2 (s.e.m.) suggested that WRN-1 played a role in cellular responses to days; N2(control), 13.6±0.1 days. Over 100 F1 worms were used per DNA damage, larval growth rate was measured after irradiating single data point, and each experiment was repeated three times. L1 stage worms with lower doses of γ-ray than those worms 2570 Development 131 (11) Research article

Fig. 4. Premature aging of wrn- 1(RNAi) worms. Wild-type N2 and wrn-1(RNAi) worms at 1 to 7 days of adulthood were photographed on the same day under identical conditions. wrn- 1(RNAi) worms were prepared by feeding on E. coli cells producing dsRNA of wrn-1, and growth temperature was shifted from 20°C to 25°C at the L4 stage. (A) Accumulation of lipofuscin autofluorescence. Ten worms of each strain were photographed every day, and one worm with an averaged intensity of autofluorescence is shown. (B) C. elegans heads were photographed with a Nomarski optics microscope. Photographs of heads were given a score of 1-5, with 1 representing a youthful appearance and 2-5 denoting increasing orders of overall deterioration in the tissue. (C) Scatter diagram of the values assigned to tissue deterioration in the head. Each dot corresponds to a single animal. Scale bars: A, 500 µm; B, 50 µm. shown in Fig. 5D. Even in the absence of γ-irradiation, wrn- with Nomarski optics, the amount of time elapsing between the 1(RNAi) larvae surprisingly grew faster than wild-type N2 2-cell and the 4-cell stage was 16.1 (±0.3) minutes (n=5) for larvae, as shown in Fig. 6. Although wrn-1(RNAi) larvae wild type, whereas the corresponding time was 11.9 (±0.2) reached the adult stage about 6 hours earlier than wild type at minutes for the wrn-1(RNAi) strain (P<0.001). Ninety minutes 20°C, this was much less than the difference of 2.6 days after reaching the 2-cell stage, wild-type and wrn-1(RNAi) (P<0.001) in life span that was observed between the two strains reached the 35-cell and 50-cell stages, respectively, strains at 25°C. Therefore, there remains a substantial clearly demonstrating a difference between cell division rates reduction in life span (Fig. 3) even when the difference in larval in the two strains (Fig. 7A). In order to determine which phase growth rate is taken into account. Another striking aspect of of the cell cycle was accelerated in wrn-1(RNAi) embryos, the growth rate of wrn-1(RNAi) larvae was its independence time-lapse micrographs were taken of the early embryos from from ionizing radiation, which contrasted with the growth 2- to 4-cell stages. S-phase duration was measured from retardation of wild-type larvae with increasing γ-ray dose (Fig. cytokinesis of the P0 cell to nuclear envelope breakdown 6). The fast growth of wrn-1(RNAi) larvae and its insensitivity (NEBD) of the AB or P1 cell (Fig. 7B), as described by to DNA damage suggested that the strain probably was Brauchle et al. (Brauchle et al., 2003). As plotted in Fig. 7C, defective in a cell cycle checkpoint responding to DNA S-phase duration was shortened by 2.3 minutes (P<0.001) in damage. the AB cell and by 2.5 minutes (P<0.001) in the P1 cell, by the RNAi, whereas M-phase duration was not affected (data S-phase acceleration in wrn-1(RNAi) embryos not shown). chk-1(RNAi) was as effective as wrn-1(RNAi) in Fast development of wrn-1(RNAi) strain was observed during reducing the S-phase duration of the AB cell (P=0.44), and was embryogenesis as well as at the larval stage, as shown in Fig. more effective in shortening the same phase of the P1 cell 7. When 2-cell embryos were observed under a microscope (P=0.005), whereas simultaneous RNAi of both had no Werner syndrome in C. elegans 2571

additive effect [wrn-1 versus wrn-1/chk-1, P=0.60 (AB cell) and 0.93 (P1 cell); chk-1 versus wrn-1/chk-1, P=0.24 (AB) and 0.002 (P1)]. Thus, WRN-1 appears to act in the same checkpoint pathway as CHK-1. When wrn-1(RNAi) was performed in div-1(or148ts) C. elegans, which is mutated in the B subunit of the DNA polymerase α-primase complex (Encalada et al., 2000), S-phase was shortened by 2.6 minutes (P<0.001) in the AB cell and by 4.1 minutes (P<0.001) in the P1 cell. These changes of S-phase duration were more or less similar to the corresponding changes in wild type. By contrast, a greater reduction of S-phase duration was obtained by chk-1(RNAi) in the div-1(or148ts) mutant: 5.8 minutes (P<0.001) in the AB cell and 7.7 minutes (P<0.001) in the P1 cell. Double RNAi of chk-1 and wrn-1 in the div-1(or148ts) strain was no more effective than single RNAi of chk-1 [P=0.76 (AB) and 0.49 (P1) for chk-1 versus wrn-1/chk-1]. Therefore, C. elegans chk-1 is involved in the DNA replication checkpoint pathway induced by inefficient priming for Okazaki fragments, as first demonstrated by Brauchle et al. (Brauchle et al., 2003), and wrn-1 appeared to be less efficient than chk-1 at this checkpoint. The effect of wrn-1(RNAi) on S-phase duration was significantly enhanced in the P1 cell of the div-1(or148ts) mutant relative to that of wild type (P=0.01), whereas the effect was similar in the AB cells of the two strains (P>0.50). WRN-1 functions at the DNA replication checkpoint in the germ-line In order to probe the role of wrn-1 in the DNA replication checkpoint, C. elegans gonads were treated with hydroxyurea to interrupt DNA replication by depleting deoxynucleotides, and were then observed by fluorescence microscopy. As shown in Fig. 8, pre- meiotic nuclei in the wild-type gonad were enlarged and reduced in number because of cell cycle arrest combined with continued nuclear growth. By contrast, small nuclei were observed in the wrn-1(RNAi) gonad even after hydroxyurea treatment, and a fraction of the nuclei appeared to be fragmented, suggesting an ineffective checkpoint for replication blockage.

Discussion Werner syndrome is accompanied by genomic Fig. 5. Developmental abnormalities of wrn-1(RNAi) worms and their instability, leading to DNA deletions and translocations penetrance influenced by growth temperature and ionizing radiation. (Salk et al., 1981; Fukuchi et al., 1989) resulting from (A) Morphological phenotypes of F1 progeny produced by P0 C. elegans enzymatic malfunction of WRN as a helicase and worms microinjected with dsRNA of wrn-1. Small body, shorter than 70% exonuclease. The sequence similarity between C. of the wild-type body length and thin; dumpy, short and fat; ruptured elegans and human WRN suggests that these two body, gonad and intestine bursting out of the worm; bag of worms, proteins possess similar functions in maintaining hatched worms inside the adult worm due to a blockage of egg-laying; genomic stability, although C. elegans WRN-1 lacks growth arrest, at various larval stages. Scale bars: 100 µm. the exonuclease domain of human WRN. (B-D) Frequency of wrn-1(RNAi) phenotypes. Wild-type N2 and F1 wrn- 1(RNAi) worms were grown at (B) 20°C and (C) 25°C, and phenotypes WRN-1 in C. elegans is distinct in subcellular were scored up to the 8-day adult stage. (D) Wild-type N2 and F1 wrn- γ localization from its human and mouse 1(RNAi) worms were -irradiated at 60 Gy and 20°C and their phenotypes homologs scored after 3 days. Over 500 F1 worms were used in each measurement for B and C, and over 200 worms for D. Each experiment was repeated C. elegans WRN-1 is diffusely distributed in the three times and standard errors (s.e.m.) are shown by error bars. Note that nucleoplasm during interphase as is mouse WRN, the % phenotype scale in D is different from that in B and C. whereas human WRN is predominantly localized to the 2572 Development 131 (11) Research article

nucleolus (Marciniak et al., 1998; Suzuki et al., 2001). 100 NN22 0Gy0 Gy N2 10Gy However, human WRN relocates from the nucleolus to

) N2 10 Gy NN22 20Gy 20 Gy % nucleoplasmic foci upon induction of DNA damage with

( wWRNrn- 0Gy1 (RNAi) 0 Gy 80 WRN 10Gy UV, ionizing radiation, camptothecin, etoposide or 4- e wrn- 1 (RNAi) 10 Gy

g wWRNrn- 20Gy1 (RNAi) 20 Gy nitroquinoline-1-oxide (Gray et al., 1998; Sakamoto et al.,

a

t

s 60 2001; Blander et al., 2002). The WRN foci overlap with the

4 foci of RP-A almost completely, and overlap with those of

L RAD51 partially, suggesting that human WRN cooperates with

o

t 40 RP-A and RAD51 in response to DNA damage (Sakamoto et

h t al., 2001). Similarly, interruption of DNA synthesis by

w

o depleting deoxynucleotides with hydroxyurea results in the

r 20

G movement of WRN from nucleoli to distinct nuclear foci that co-localize with RP-A (Constantinou et al., 2000). However, 0 the localization of WRN-1 to the poleward periphery of 24 36 48 60 72 metaphase chromosomes in the early C. elegans embryo Time after γ-irradiation (hr) has not been observed in other organisms. This peculiar localization may be needed to ensure equal partition of the Fig. 6. The rapid growth to L4 stage of wrn-1(RNAi) worms is protein into daughter cells, or may mean that WRN-1 functions unaffected by γ-irradiation. Wild-type N2 worms were microinjected with dsDNA derived from Ce-wrn-1 cDNA. F1 progeny were in a spindle checkpoint involving holocentric chromosomes irradiated with different dose (0, 10, 20 Gy) of γ-rays at the L1 larval during mitosis. stage and their growth to L4 stage was measured every twelve hours at 20°C. Over 150 F1 worms were used per measurement with two wrn-1(RNAi) phenotypes resemble the symptoms of additional repetitions, and standard errors (s.e.m.) are shown by error human WRN syndrome bars. RNAi of C. elegans wrn-1 did not cause discernable

Fig. 7. Reduction of S-phase duration by wrn-1(RNAi) during embryogenesis. (A) Accelerated cell division of a wrn-1(RNAi) embryo observed with Nomarski optics at 25°C. (B) Time-lapse DIC microscopy of 2- to 4-cell embryos. At 0 minutes, AB (right) and P1 (left) cells of a wild-type N2 embryo immediately after cytokinesis are shown. Nuclear envelope breakdown (NEBD) of AB and P1 cells occurs at 12 and 14 minutes, respectively. (C) Average duration of S phase in AB and P1 determined by timing cytokinesis and NEBD. The bar graphs correspond to the time separating cytokinesis of P0 from NEBD of either the AB or the P1 cell. Standard errors (s.e.m.) are shown by error bars. Ten embryos were observed for each estimate of S-phase duration. Scale bars: A, 20 µm; B, 10 µm. Werner syndrome in C. elegans 2573

Fig. 8. Morphological changes of pre- meiotic germ cell nuclei induced by hydroxyurea. Images are DAPI-stained nuclei in untreated (–HU) gonads and in those exposed to hydroxyurea (+HU). Upon HU treatment, pre-meiotic nuclei in the wild-type N2 gonad were substantially enlarged and reduced in number, but those in the wrn-1(RNAi) gonad were much smaller. Scale bar: 50 µm. phenotypes at 20°C, whereas at 25°C the worms had a reduced pathway. In germ cells treated with hydroxyurea, WRN-1 was life span and showed significant increases in such required to activate the DNA replication checkpoint, which developmental abnormalities as small body, bag of worms, agreed well with the role of a Saccharomyces cerevisiae ruptured body, dumpy shape, growth arrest and transparency homolog, Sgs1, at the same checkpoint (Frei and Gasser, 2000; (Fig. 5). These developmental abnormalities have been induced Myung and Kolodner, 2002). In addition, S phase was by deficiency of ATR and RAD51 homologs in C. elegans, and accelerated in wrn-1(RNAi) embryos, indicating a role of their frequency was increased by ionizing radiation (Aoki et WRN-1 as a checkpoint protein. Nevertheless, wrn-1(RNAi) al., 2000; Rinaldo et al., 2002). Therefore, the developmental was not as potent as chk-1(RNAi) in reducing the extended S- defects probably resulted from DNA damage during phase of div-1(or148ts) embryos, in which priming of Okazaki development. Some of these phenotypes resemble the fragments is inefficient. And, double RNAi of wrn-1 and chk- symptoms of human Werner syndrome, such as short stature 1 was no more effective than single RNAi of chk-1. This and premature aging. The enhanced phenotypes of wrn- suggests that WRN-1 works in a sub-pathway that diverges 1(RNAi) at 25°C was very likely due to the fact that the from CHK-1 in the DNA replication checkpoint pathway, and metabolic rate of C. elegans is 1.2 times higher at 25°C than that is either up- or downstream of CHK-1. In sgs1∆ S. at 20°C (Van Voorhies and Ward, 1999), thus generating a cerevisiae cells, S phase proceeds faster than in wild-type cells, greater oxidative stress. The slight reduction in life span was but the termination stage of S phase takes longer, so that the probably due to premature aging, as was demonstrated by total length of S phase is unchanged (Versini et al., 2003). The faster lipofuscin accumulation in the C. elegans intestine and fast progression of S phase in S. cerevisiae and C. elegans tissue deterioration in the head, although other phenotypes resulting from the absence of a RecQ homolog contrasts with such as bag of worms, ruptured body and growth arrest the extended S phase in human WRN– cells (Martin et al., 1970; certainly contributed to the reduction. Many small C. elegans Salk et al., 1985). Recently, human WRN– cells were found to mutants are known to have defects in the TGF-β signaling be defective in the chromosomal decatenation checkpoint in pathway (Gumienny and Padgett, 2003), and others are G2 phase (Franchitto et al., 2003), suggesting the possibility mutated in the spectrin β chain (sma-7), a MAP kinase (sma- that C. elegans WRN-1 may also participate in a DNA damage 5), a fatty acid elongation enzyme (elo-2) or a basement checkpoint during G2 phase. membrane protein (SPARC). Dumpiness in C. elegans is There have been conflicting reports concerning the generally due to mutation of genes participating in collagen sensitivity of wrn-deficient cells to DNA damaging agents. production or in dosage compensation. The phenotype bag of Human WS patient cells were not hypersensitive to UV or X- worms is thought to be caused by defects in HSN or ray (Fujiwara et al., 1977), but were sensitive to camptothecin, chemosensory neurons, or to partial defects in other egg-laying 4-nitroquinoline-N-oxide and DNA-crosslinking agents (Poot systems, based on its sensitivity to serotonin. A C. elegans line et al., 1999; Poot et al., 2001; Poot et al., 2002). The embryonic harboring human WRN cDNA linked to a strong C. elegans stem cells of wrn-knockout mice are hypersensitive to was prepared, but the occurrence of wrn-1(RNAi) camptothecin and etoposide, which are inhibitors of DNA phenotypes was not reduced in the transgenic line (S.M.H., topoisomerases I and II, respectively, but were not sensitive unpublished). The failure to rescue the phenotypes could be to γ-radiation, UV or mitomycin (Lebel and Leder, 1998). due to non-equivalence of human and C. elegans WRNs, or to WRN mutants of the chicken B-cell line DT-40 were inefficient expression of the exogenous gene. sensitive to various DNA damaging agents, such as Ionizing radiation increased the frequency of wrn-1(RNAi) methylmethanesulfonate, 4-nitroquinoline-N-oxide, etoposide phenotypes synergistically, suggesting that C. elegans WRN-1 and camptothecin (Imamura et al., 2002). The insensitivity of is involved in the DNA damage response (Fig. 5D). WRN-1 human and murine WRN– cells to ionizing radiation (Fujiwara may be involved in non-homologous end-joining by interacting et al., 1977; Lebel and Leder, 1998) is in contrast to the with /80 (Cooper et al., 2000), or may be involved in enhancement of the C. elegans wrn-1(RNAi) phenotypes by recombinational repair. WRN-1 may also play a role in DNA ionizing radiation demonstrated in this study. damage signaling, as in the control of p53-mediated Although a clear explanation for their pleiotropic transcriptional activation by human WRN (Spillare et al., phenotypes cannot be provided, the similarity in phenotypes 1999). Indeed, the fast larval growth of the wrn-1(RNAi) strain, between wrn-1(RNAi) worms and Werner syndrome patients especially its insensitivity to ionizing radiation, strongly suggests that the RNAi worm could be a useful model for suggests that WRN-1 acts in a DNA damage checkpoint the syndrome. The fact that Wrn knockout mice did not 2574 Development 131 (11) Research article recapitulate Werner syndrome phenotypes such as premature Gray, M. D., Wang, L., Youssoufian, H., Martin, G. M. and Oshima, J. aging, small stature, developmental abnormalities and tumor (1998). Werner helicase is localized to transcriptionally active nucleoli of formation (Lebel and Leder, 1998; Lombard et al., 2000) cycling cells. Exp. Cell Res. 242, 487-494. Gumienny, T. and Padgett, R. W. (2003). A small issue addressed. BioEssays further emphasizes the importance of wrn-deficient C. elegans 25, 305-308. as a potential model for Werner syndrome. Huang, S., Li, B., Gray, M. D., Oshima, J., Mian, I. S. and Campisi, J. (1998). The premature ageing syndrome protein, WRN, is a 3′→5′ We thank Dr Yuji Kohara (National Institute of Genetics, Japan) for exonuclease. Nat. Genet. 20, 114-116. the yk41c EST clone. N2 and div-1(or148ts) C. elegans strains were Hurd, D. D. and Kemphues, K. J. (2003). PAR-1 is required for obtained from the C. elegans Genetics Center (St Paul, MN, USA), morphogenesis of the Caenorhabditis elegans vulva. Dev. Biol. 253, 54-65. which is supported by the National Center for Research Resources. Imamura, O., Fujita, K., Itoh, C., Takeda, S., Furuichi, Y. and Matsumoto, This work was supported by a Life Phenomena and Function Research T. (2002). Werner and Bloom helicases are involved in DNA repair in a complementary fashion. Oncogene 21, 954-963. Grant [01-J-LF-01-B-83] from the Korean Ministry of Science and Jones, A. R., Francis, R. and Schedl, T. (1996). GLD-1, a cytoplasmic Technology, and by the Brain Korea 21 Project in 2003 to H.-S.K. protein essential for oocyte differentiation, shows stage- and sex-specific expression during Caenorhabditis elegans germline development. Dev. Biol. 180, 165-183. References Label, M., Spillare, E. A., Harris, C. C. and Leder, P. (1999). The Werner Aoki, H., Sato, S., Takanami, T., Ishihara, T., Katsura, I., Takahashi, H. syndrome gene product co-purifies with the DNA replication complex and and Higashitani, A. (2000). Characterization of Ce-alt-1, an ATM-like gene interacts with PCNA and topoisomerase I. J. Biol. Chem. 274, 37795-37799. from Caenorhaditis elegans. Mol. Gen. Genet. 264, 119-126. Lebel, M. and Leder, P. (1998). A deletion within the murine Werner Balajee, A. S., Machwe, A., May, A., Gray, M. D., Oshima, J., Martin, G. syndrome helicase induces sensitivity to inhibitors of topoisomerase and M., Nehlin, J. O., Brosh, R., Orren, D. K. and Bohr, V. A. (1999). The loss of cellular proliferation capacity. Proc. Natl. Acad. Sci. USA 95, 13097- Werner syndrome protein is involved in RNA polymerase II transcription. 13102. Mol. Biol. Cell 10, 2655-2668. Lieber, M. R. (1997). The FEN-1 family of structure-specific nucleases in Blander, G., Zalle, N., Daniely, Y., Taplick, J., Gray, M. D. and Oren, M. eukaryotic DNA replication, recombination and repair. BioEssays 19, 233- (2002). DNA damage-induced translocation of the Werner helicase is 240. regulated by acetylation. J. Biol. Chem. 277, 50934-50940. Lombard, D. B., Beard, C., Johnson, B., Marciniak, R. A., Dausman, J., Brauchle, M., Baumer, K. and Gönczy, P. (2003). Differential activation of Bronson, R., Buhlmann, J. E., Lipman, R., Curry, R., Sharpe, A., the DNA replication checkpoint contributes to asynchrony of cell division Jaenisch, R. and Guarente, L. (2000). Mutations in the WRN gene in mice in C. elegans embryos. Curr. Biol. 13, 819-827. accelerate mortality in a p53-null background. Mol. Cell. Biol. 20, 3286- Brosh, R. M., Jr, von Kobbe, C., Sommers, J. A., Karmakar, P., Opresko, 3291. P. L., Piotrowski, J., Dianova, I., Dianova, G. L. and Bohr, V. A. (2001). Marciniak, R. A., Lombard, D. B., Johnson, F. B. and Guarente, L. (1998). Werner syndrome protein interacts with human flap endonuclease 1 and Nucleolar localization of the Werner syndrome protein in human cells. Proc. stimulates its cleavage activity. EMBO J. 20, 5791-5801. Natl. Acad. Sci. USA 95, 6887-6892. Conrad, R., Lea, K. and Blumenthal, T. (1995). SL1 trans-splicing specified Martin, G. M., Sprague, C. A. and Epstein, C. J. (1970). Replicative life by AU-rich synthetic RNA inserted at the 5′ end of Caenorhabditis elegans span of cultivated human cells: effects of donor’s age, tissue, and genotype. pre-mRNA. RNA 1, 164-170. Lab. Investig. 23, 86-92. Constantinou, A., Tarsounas, M., Karow, J. K., Brosh, R. M., Jr, Bohr, V. Mohaghegh, P. and Hickson, I. D. (2001). DNA helicase deficiencies A., Hickson, I. D. and West, S. C. (2000). Werner’s syndrome protein associated with cancer predisposition and premature ageing disorders. Hum. (WRN) migrates Holliday junctions and co-localizes with RPA upon Mol. Genet. 10, 741-746. replication arrest. EMBO J. 1, 80-84. Moore, L. L. and Roth, M. B. (2001). HCP-4, a CENP-C-like protein in Cooper, M. P., Machwe, A., Orren, D. K., Brosh, R. M., Jr, Ramsden, D. Caenorhabditis elegans, is required for resolution of sister centromeres. J. and Bohr, V. A. (2000). complex interacts with and stimulates the Cell Biol. 153, 1199-1208. Werner protein. Genes Dev. 14, 907-912. Morozov, A. R., Mushegian, E. V., Koonin, E. V. and Bork, P. (1997). A Crittenden, S. L. and Kimble, J. (1999). Confocal methods for putative nucleic acid-binding domain in Bloom’s and Werner’s syndrome Caenorhabditis elegans. Methods Mol. Biol. 122, 141-151. helicases. Trends Biochem. Sci. 22, 417-418. Encalada, S. E., Martin, P. R., Phillips, J. B., Lyczak, R., Hamill, D. R., Moser, M. J., Holley, W. R., Chatterjee, A. and Mian, I. S. (1997). The Swan, K. A. and Bowerman, B. (2000). DNA replication defects delay cell proofreading domain of Escherichia coli DNA polymerase and other DNA division and disrupt cell polarity in early Caenorhabditis elegans embryos. and/or RNA exonuclease domains. Nucleic Acids Res. 25, 5110-5118. Dev. Biol. 228, 225-238. Myung, K. and Kolodner, R. D. (2002). Suppression of genome instability Evans, D., Perez, E. D., MacMorris, M., Leake, D., Wilusz, C. J. and by redundant S-phase checkpoint pathways in Saccharomyces cerevisiae. Blumenthal, T. (2001). A complex containing CstF-64 and the SL2 snRNP Proc. Natl. Acad. Sci. USA 99, 4500-4507. connects mRNA 3′ end formation and trans-splicing in C. elegans operons. Nonet, M., Grundahl, K., Meyer, B. J. and Rand, J. B. (1993). Synaptic Genes Dev. 15, 2562-2571. function is impaired but not eliminated in C. elegans mutants lacking Fitzgerald, M. C. and Schwarzbauer, J. E. (1988). Importance of the synaptotagmin. Cell 73, 1291-1305. basement membrane protein SPARC for viability and fertility in Nystul, T., Goldmark, J. P., Padilla, P. A. and Roth, M. B. (2003). Caenorhabditis elegans. Curr. Biol. 8, 1285-1288. Suspended animation in C. elegans requires the spindle checkpoint. Science Franchitto, A. Oshima, J. and Pichierri, P. (2003). The G2-phase 302, 1038-1041. decatenation checkpoint is defective in Werner syndrome cells. Cancer Res. Ogburn, C. E., Oshima, J., Poot, M., Chen, R., Hunt, K. E., Gollahon, K. 63, 3289-3295. A., Rabinovitch, P. S. and Martin, G. M. (1997). An apoptosis-inducing Frei, C. and Gasser, S. M. (2000). The yeast Sgs1p helicase acts upstream of genotoxin differentiates heterozygote carriers for Werner helicase mutations Rad53p in the DNA replication checkpoint and colocalizes with Rad53p in from wild-type and homozygous mutants. Hum. Genet. 101, 121-125. S-phase-specific foci. Genes Dev. 14, 81-96. Poot, M., Gollahon, K. A., Emond, M. J., Silber, J. R. and Rabinovitch, P. Fujiwara, Y., Higashikawa, T. and Tatsumi, M. (1977). A retarded rate of S. (2002). Werner syndrome diploid fibroblasts are sensitive to 4- DNA replication and normal level of DNA repair in Werner’s syndrome nitroquinoline-N-oxide and 8-methoxypsoralen: implication for the disease fibroblasts in culture. J. Cell. Physiol. 92, 365-374. phenotype. FASEB J. 16, 757-758. Fukuchi, K., Martin, G. M. and Monnat, R. J., Jr (1989). Mutator Poot, M., Gollahon, K. A. and Rabinovitch, P. S. (1999). Werner syndrome phenotype of Werner syndrome is characterized by extensive deletions. lymphoblastoid cells are sensitive to camptothecin-induced apoptosis in S- Proc. Natl. Acad. Sci. USA 86, 5893-5897. phase. Hum. Genet. 104, 10-14. Garigan, D., Hsu, A.-L., Fraser, A. G., Kamath, R. S., Ahringer, J. and Poot, M., Yom, J. S., Whang, S. H., Kato, J. T., Gollahon, K. A. and Kenyon, C. (2002). Genetic analysis of tissue aging in Caenorhabditis Rabinovitch, P. S. (2001). Werner syndrome cells are sensitive to DNA elegans: a role for heat-shock factor and bacterial proliferation. Genetics cross-linking drugs. FASEB J. 15, 1224-1226. 161, 1101-1112. Rinaldo, C., Bazzicalupo, P., Ederle, S., Hilliard, M. and La Volpe, A. Werner syndrome in C. elegans 2575

(2002). Roles for Caenorhabditis elegans rad-51 in meiosis and in Suzuki, T., Shiratori, M., Furuichi, Y. and Matsumoto, T. (2001). Diverged resistance to ionizing radiation during development. Genetics 160, 471-479. nuclear localization of Werner helicase in human and mouse cells. Oncogene Sakamoto, S., Nishikawa, K., Heo, S. J., Goto, M., Furuichi, Y. and 20, 2551-2558. Shimamoto, A. (2001). Werner helicase relocates into nuclear foci in Suzuki, N., Shiratori, M., Goto, M. and Furuichi, Y. (1999). Werner response to DNA damaging agents and co-localizes with RPA and Rad51. syndrome helicase contains a 5′→3′ exonuclease activity that digests DNA Genes Cells 6, 421-430. and RNA strands in DNA/DNA and RNA/DNA duplexes dependent on Salk, D., Au, K., Hoehn, H. and Martin, G. M. (1981). Cytogenetics of unwinding. Nucleic Acids Res. 27, 2361-2368. Werner’s syndrome cultured skin blasts: variegated translocation mosaicism. Szekely, A. M., Chen, Y. H., Zhang, C., Oshima, J. and Weissman, S. M. Cytogenet. Cell Genet. 30, 92-107. (2000). Werner protein recruits DNA polymerase delta to the nucleolus. Salk, D., Bryant, E., Hoehn, H., Johnston, P. and Martin, G. M. (1985). Proc. Natl. Acad. Sci. USA 97, 11365-11370. Growth characteristics of Werner syndrome cells in vitro. Adv. Exp. Med. Timmons, L. and Fire, A. (1998). Specific interference by ingested dsRNA. Biol. 190, 305-311. Nature 395, 854. Schwirzke, M., Gnirke, A., Bork, P., Tarin, D. and Weidle, U. H. (1998). Van Voorhies, W. A. and Ward, S. (1999). Genetic and environmental Differential gene expression in mammary carcinoma cell lines: identification conditions that increase longevity in Caenorhabditis elegans decrease of DRIM, a new gene down-regulated in metastasis. Anticancer Res. 18, metabolic rate. Proc. Natl. Acad. Sci. USA 96, 11399-11403. 1409-1421. Versini, G., Comet, I., Wu, M., Hoopes, L., Schwob, E. and Pasero, P. Shen, J. C., Gray, M. D., Oshima, J., Kamath-Loeb, A. S., Fry, M. and Loeb, (2003). The yeast Sgs1 helicase is differentially required for genomic and L. A. (1998). Werner syndrome protein. I. DNA helicase and DNA ribosomal DNA replication. EMBO J. 22, 1939-1949. exonuclease reside on the same polypeptide. J. Biol. Chem. 273, 34139-34144. Wyllie, F. S., Jones, C. J., Skinner, J. W., Haughton, M. F., Wallis, C., Shiratori, M., Suzuki, T., Itoh, C., Goto, M., Furuichi, Y. and Matsumoto, Wynford-Thomas, D., Faragher, R. G. and Kipling, D. (2000). T. (2002). WRN helicase accelerates the transcription of ribosomal RNA as Telomerase prevents the accelerated cell aging of Werner syndrome a component of an RNA polymerase I-associated complex. Oncogene 21, fibroblasts. Nat. Genet. 24, 16-17. 2447-2454. Yu, C. E., Oshima, J., Fu, Y. H., Wijsman, E. M., Hisama, F., Alisch, R., Spillare, E. A., Robles, A. I., Wang, X. W., Shen, J. C., Yu, C. E., Matthews, S., Nakura, J., Miki, T., Ouais, S., Martin, G. M., Mulligan, Schellenberg, G. D. and Harris, C. C. (1999). p53-mediated apoptosis is J. and Schellenberg, G. D. (1996). Positional cloning of the Werner’s attenuated in Werner syndrome cells. Genes Dev. 13, 1355-1360. syndrome gene. Science 272, 258-262.