Title: Werner syndrome protein interacts with human flap endonuclease 1 and stimulates its cleavage activity
Abstract: Article15 October 2001free access Werner syndrome protein interacts with human flap endonuclease 1 and stimulates its cleavage activity Robert M. Brosh Jr Corresponding Author Robert M. Brosh Jr Laboratory of Molecular Gerontology, National Institute on Aging, National Institutes of Health, 5600 Nathan Shock Drive, Baltimore, MD, 21224 USA Search for more papers by this author Cayetano von Kobbe Cayetano von Kobbe Laboratory of Molecular Gerontology, National Institute on Aging, National Institutes of Health, 5600 Nathan Shock Drive, Baltimore, MD, 21224 USA Search for more papers by this author Joshua A. Sommers Joshua A. Sommers Laboratory of Molecular Gerontology, National Institute on Aging, National Institutes of Health, 5600 Nathan Shock Drive, Baltimore, MD, 21224 USA Search for more papers by this author Parimal Karmakar Parimal Karmakar Laboratory of Molecular Gerontology, National Institute on Aging, National Institutes of Health, 5600 Nathan Shock Drive, Baltimore, MD, 21224 USA Search for more papers by this author Patricia L. Opresko Patricia L. Opresko Laboratory of Molecular Gerontology, National Institute on Aging, National Institutes of Health, 5600 Nathan Shock Drive, Baltimore, MD, 21224 USA Search for more papers by this author Jason Piotrowski Jason Piotrowski Laboratory of Molecular Gerontology, National Institute on Aging, National Institutes of Health, 5600 Nathan Shock Drive, Baltimore, MD, 21224 USA Search for more papers by this author Irina Dianova Irina Dianova MRC Radiation & Genome Stability Unit, Harwell, Oxfordshire, OX11 0RD UK Search for more papers by this author Grigory L. Dianov Grigory L. Dianov MRC Radiation & Genome Stability Unit, Harwell, Oxfordshire, OX11 0RD UK Search for more papers by this author Vilhelm A. Bohr Vilhelm A. Bohr Laboratory of Molecular Gerontology, National Institute on Aging, National Institutes of Health, 5600 Nathan Shock Drive, Baltimore, MD, 21224 USA Search for more papers by this author Robert M. Brosh Jr Corresponding Author Robert M. Brosh Jr Laboratory of Molecular Gerontology, National Institute on Aging, National Institutes of Health, 5600 Nathan Shock Drive, Baltimore, MD, 21224 USA Search for more papers by this author Cayetano von Kobbe Cayetano von Kobbe Laboratory of Molecular Gerontology, National Institute on Aging, National Institutes of Health, 5600 Nathan Shock Drive, Baltimore, MD, 21224 USA Search for more papers by this author Joshua A. Sommers Joshua A. Sommers Laboratory of Molecular Gerontology, National Institute on Aging, National Institutes of Health, 5600 Nathan Shock Drive, Baltimore, MD, 21224 USA Search for more papers by this author Parimal Karmakar Parimal Karmakar Laboratory of Molecular Gerontology, National Institute on Aging, National Institutes of Health, 5600 Nathan Shock Drive, Baltimore, MD, 21224 USA Search for more papers by this author Patricia L. Opresko Patricia L. Opresko Laboratory of Molecular Gerontology, National Institute on Aging, National Institutes of Health, 5600 Nathan Shock Drive, Baltimore, MD, 21224 USA Search for more papers by this author Jason Piotrowski Jason Piotrowski Laboratory of Molecular Gerontology, National Institute on Aging, National Institutes of Health, 5600 Nathan Shock Drive, Baltimore, MD, 21224 USA Search for more papers by this author Irina Dianova Irina Dianova MRC Radiation & Genome Stability Unit, Harwell, Oxfordshire, OX11 0RD UK Search for more papers by this author Grigory L. Dianov Grigory L. Dianov MRC Radiation & Genome Stability Unit, Harwell, Oxfordshire, OX11 0RD UK Search for more papers by this author Vilhelm A. Bohr Vilhelm A. Bohr Laboratory of Molecular Gerontology, National Institute on Aging, National Institutes of Health, 5600 Nathan Shock Drive, Baltimore, MD, 21224 USA Search for more papers by this author Author Information Robert M. Brosh 1, Cayetano von Kobbe1, Joshua A. Sommers1, Parimal Karmakar1, Patricia L. Opresko1, Jason Piotrowski1, Irina Dianova2, Grigory L. Dianov2 and Vilhelm A. Bohr1 1Laboratory of Molecular Gerontology, National Institute on Aging, National Institutes of Health, 5600 Nathan Shock Drive, Baltimore, MD, 21224 USA 2MRC Radiation & Genome Stability Unit, Harwell, Oxfordshire, OX11 0RD UK *Corresponding author. E-mail: [email protected] The EMBO Journal (2001)20:5791-5801https://doi.org/10.1093/emboj/20.20.5791 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Werner syndrome (WS) is a human premature aging disorder characterized by chromosomal instability. The cellular defects of WS presumably reflect compromised or aberrant function of a DNA metabolic pathway that under normal circumstances confers stability to the genome. We report a novel interaction of the WRN gene product with the human 5′ flap endonuclease/5′–3′ exonuclease (FEN-1), a DNA structure-specific nuclease implicated in DNA replication, recombination and repair. WS protein (WRN) dramatically stimulates the rate of FEN-1 cleavage of a 5′ flap DNA substrate. The WRN–FEN-1 functional interaction is independent of WRN catalytic function and mediated by a 144 amino acid domain of WRN that shares homology with RecQ DNA helicases. A physical interaction between WRN and FEN-1 is demonstrated by their co-immunoprecipitation from HeLa cell lysate and affinity pull-down experiments using a recombinant C-terminal fragment of WRN. The underlying defect of WS is discussed in light of the evidence for the interaction between WRN and FEN-1. Introduction Werner syndrome (WS) is an autosomal recessive disorder that displays symptoms of premature aging after adolescence (Martin, 1978). WS cells exhibit replication defects (Martin et al., 1970; Takeuchi et al., 1982; Hanaoka et al., 1985; Salk et al., 1985; Poot et al., 1992), genomic instability (Salk et al., 1981, 1985; Fukuchi et al., 1989) and altered telomere dynamics (Hisama et al., 2000; Ouellette et al., 2000; Wyllie et al., 2000). An anti-recombinogenic role of WRN has been suggested (Yamagata et al., 1998; Constantinou et al., 2000); however, the basis for the genomic instability in WS is not understood. The gene defective in WS, designated WRN, encodes a nuclear (Matsumoto et al., 1997) 1432 amino acid protein with the seven conserved motifs found in the RecQ family of DNA helicases (Yu et al., 1996; for review see Mohaghegh and Hickson, 2001) (Figure 1). WRN is a DNA-dependent ATPase and 3′ to 5′ helicase (Gray et al., 1997; Suzuki et al., 1997). The protein sequence of WRN also contains a region of conserved exonuclease motifs (Moser et al., 1997) (Figure 1), and possesses a 3′ to 5′ exonuclease activity (Huang et al., 1998; Kamath-Loeb et al., 1998; Shen et al., 1998). The catalytic activities of WRN suggest that a pathway of DNA metabolism is defective in WS. A role of WRN in DNA replication and/or repair is suggested by a number of its protein interactions (Shen et al., 1998; Blander et al., 1999; Brosh et al., 1999, 2001; Lebel et al., 1999; Spillare et al., 1999; Cooper et al., 2000; Kamath-Loeb et al., 2000; Kawabe et al., 2000; Li and Comai, 2000; Szekely et al., 2000) (Figure 1A). Precisely how these protein interactions are important in a biological setting remains to be established. Figure 1.Structural map of WRN protein motifs and recombinant proteins used in this study. (A) Conserved RecQ ATPase/helicase and exonuclease motifs, RecQ C-terminal region (RQC), helicase-related domain (HRDC), nuclear localization sequence (NLS) and known regions for WRN protein interactions are designated. Positions of site-directed mutations in motif I of the helicase domain (WRN-K577M) and motif I of the exonuclease domain (WRN-E84A) are indicated by asterisks. GST–WRN recombinant proteins are shown with subscripted numbers to designate the WRN amino acid sequences. (B) Purified full-length recombinant proteins (1 μg) and GST–WRN recombinant fragments (2–4 μg) used in this study were resolved on 10% polyacrylamide SDS gels run and stained with Coomassie Brilliant Blue. The 66 kDa band in the WRN protein preparations is BSA (100 μg/ml) (Orren et al., 1999). Bands migrating below GST–WRN949–1432 (∼85 kDa) were determined to be degradation products by western blot analysis using anti-GST antibody (Santa Cruz). Download figure Download PowerPoint The evidence implicating 5′ flap endonuclease/5′–3′ exonuclease (FEN-1) in DNA replication, repair and the maintenance of genomic stability suggested that WRN may interact with FEN-1 to facilitate its function. FEN-1 is required during Okazaki fragment processing (Bambara et al., 1997) and long patch base excision repair (BER) (Klungland and Lindahl, 1997; Kim et al., 1998). FEN-1 is a DNA structure-specific nuclease that cleaves 5′ flap single-stranded (ss)DNA at the single strand–double strand junction (reviewed in Lieber, 1997). FEN-1 is active as a 5′ to 3′ exonuclease at nicks in duplex DNA and also catalytically removes the 5′ terminal RNA mononucleotide, the latter process thought to be important in Okazaki fragment processing (Waga and Stillman, 1998). The FEN-1 homolog RAD27 in Saccharomyces cerevisiae plays an important role in the maintenance of genome stability (Johnson et al., 1995; Sommers et al., 1995; Vallen and Cross, 1995; Tishkoff et al., 1997; Freudenreich et al., 1998; Kokoska et al., 1998; Schweitzer and Livingston, 1998; Gary et al., 1999a), telomere stability (Parenteau and Wellinger, 1999), response to DNA damage (Reagan et al., 1995; Sommers et al., 1995) and non-homologous end-joining (NHEJ) (Wu et al., 1999). Thus genomic instability persists in vivo when FEN-1 is either absent or its cleavage activity is blocked by DNA secondary structure. Proliferating cell nuclear antigen (PCNA) binds FEN-1 (Li et al., 1995; Wu et al., 1996) and stimulates FEN-1 nuclease activity (Tom et al., 2000). Elimination of PCNA binding by a site-specific mutation in RAD27 did not significantly increase genetic instability, recombination or methyl methane sulfate sensitivity (Gary et al., 1999b), suggesting that redundant protein interactions/enzyme activities may be important in vivo. Dna2, a DNA helicase and endonuclease that physically and genetically interacts with RAD27, may play a direct role in Okazaki fragment maturation in conjunction with FEN-1 (Bae and Seo, 2000). However, no functional homologs for Dna2 have been identified in mammalian systems. We report that WRN interacts physically with FEN-1 and stimulates FEN-1 cleavage. The functional interaction is independent of WRN catalytic activities and mediated by a C-terminal region of WRN. WRN and FEN-1 are likely to act together during DNA replication and/or repair. Defects in the WRN–FEN-1 interaction may contribute to the genomic instability of WS. Results Physical interaction between WRN and FEN-1 The C-terminal domain of WRN does not have any known catalytic activities, but binds to some of the WRN-interacting proteins (Figure 1). We established a pull-down assay to determine a possible association of glutathione S-transferase (GST)–WRN949–1432 (Figure 1) with FEN-1 from human nuclear extract (NE). FEN-1 was coprecipitated with GST–WRN949–1432 (Figure 2A, lane 3) by comparison with the input (lane 1). In the absence of NE input, a specific band migrating at the position of FEN-1 was not detected (Figure 2A, lane 5). Both GST and GST–WRN1072–1236 were unable to coprecipitate FEN-1 (Figure 2A, lanes 2 and 4). There was some cross-reactivity of the anti-FEN-1 antibody to bacterial proteins migrating at higher molecular weight (Figure 2A, lanes 3–6). These results demonstrate that FEN-1 can be specifically precipitated by the GST–WRN949–1432 fragment. To address the possibility that the interaction between GST–WRN949–1432 and FEN-1 may have been indirectly mediated by other NE proteins, purified recombinant FEN-1 was incubated with the GST–WRN949–1432 affinity resin. The results demonstrate that FEN-1 is precipitated by the GST–WRN949–1432 affinity beads (Figure 2B, lane 2), but not by GST resin (Figure 2B, lane 1). The FEN-1–GST–WRN949–1432 interaction was resistant to DNase I (10 μg/ml), indicating that binding was not DNA mediated (data not shown). Figure 2.WRN and FEN-1 interact physically. (A) Beads with GST (lane 2), GST–WRN949–1432 (lane 3) or GST–WRN1072–1236 (lane 4) were incubated with 750 μg of HeLa NE. In control experiments, NE was omitted during binding with GST–WRN949–1432 (lane 5) or GST–WRN1072–1236 (lane 6). Western blotting was performed with anti-FEN-1 antibodies. The NE input (lane 1) corresponds to 10 μg. (B) Beads with GST (lane 1) or GST–WRN949–1432 (lane 2) were incubated with 1.2 μg of recombinant FEN-1 purified from E.coli. In control experiments, FEN-1 was omitted during binding with GST–WRN949–1432 (lane 3). Western blotting was performed with anti-FEN-1 antibodies. The FEN-1 input (lane 4) corresponds to 20 ng. (C) FEN-1–Sepharose specifically binds recombinant WRN. Purified WRN and FEN-1 were judged to be pure of DNA by analysis using SYBR Green Stain (FMC Products). Western blot of the third wash (lanes 1 and 2) or eluted fractions (lanes 3 and 4) from binding reaction of WRN to FEN-1–Sepharose (lanes 2 and 4) or BSA–Sepharose (lanes 1 and 3) is shown. (D) WRN and FEN-1 coprecipitate from HeLa cell lysate using anti-WRN antibody as demonstrated by western blotting. Top panel, blot was probed with anti-FEN-1 antibody. Bottom panel, blot was probed with anti-WRN antibody. Lane 1, control precipitate from HeLa cell lysate in which WRN antibody was omitted from immunoprecipitation; lane 2, HeLa cell lysate input; lane 3, immunoprecipitate from HeLa cell lysate using WRN antibody; lane 4, AG11395 (WS−/−) cell lysate input; lane 5, immunoprecipitate from AG11395 cell lysate using WRN antibody; lane 6, purified His-tagged FEN-1, which migrates slightly higher than NE FEN-1. Download figure Download PowerPoint To determine whether full-length WRN interacts with FEN-1, we tested for full-length recombinant WRN to bind FEN-1–Sepharose beads. As shown in Figure 2C, WRN remained bound through successive washes (lane 2) and was eluted from the FEN-1–Sepharose beads (lane 4). Twenty percent of the purified WRN input was bound by the FEN-1–Sepharose beads. WRN failed to effectively bind to the bovine serum albumin (BSA)–Sepharose beads as detected in the eluted fraction (Figure 2C, lane 3). These studies indicate that full-length WRN binds directly to FEN-1. To determine whether endogenous WRN and FEN-1 interact in vivo, co-immunoprecipitation experiments from HeLa whole-cell extracts were performed using a polyclonal antibody against the WRN N-terminus (Figure 2D). FEN-1 was effectively precipitated by the WRN antibody (Figure 2D, lane 3) and failed to be detected when the WRN antibody was omitted from the immunoprecipitation (lane 1). As a negative control, similar experiments were performed using WS cells (AG11395). FEN-1 failed to be immunoprecipitated by the WRN antibody (Figure 2D, lane 5) despite its presence in the AG11395 cell lysate (lane 4). These results demonstrate that WRN and FEN-1 can be co-immunoprecipitated from HeLa NE. WRN stimulates the FEN-1 cleavage reaction The physical interaction between WRN and FEN-1 suggested that the two proteins might modulate the catalytic activity of each other. No significant effect of FEN-1 on WRN ATPase or helicase activity was detected (data not shown). To characterize the effect of WRN on FEN-1 cleavage, we utilized a 19 bp DNA substrate with a single unannealed 5′ nucleotide adjacent to an upstream 25 bp duplex (1 nt 5′ flap). The 1 nt 5′ flap substrate was susceptible to FEN-1 cleavage that generated 1 and 2 nt products (Figure 3A, lane 2), as previously published (Tom et al., 2000). In the presence of 100 fmol of purified FEN-1, 5% of the substrate was incised (Figure 3A, lane 2, and B). In the presence of WRN (75 fmol), FEN-1 incised 55% of the flap substrate molecules (Figure 3A, lane 3, and B). Thus, at nearly equimolar concentrations of WRN and FEN-1, FEN-1 cleavage is stimulated 11-fold. Importantly, WRN alone did not catalyze significant cleavage of the 1 nt flap DNA substrate (Figure 3A, lane 4). Figure 3.WRN stimulates FEN-1 cleavage activity. (A) Reactions (20 μl) containing 10 fmol of a 1 nt 5′ flap DNA substrate, 100 fmol of FEN-1 and 75 fmol of WRN were incubated at 37°C for 15 min under standard conditions as described in Materials and methods. A phosphorimage of a typical gel is shown. Substrate and cleavage products are as indicated. Lane 1, no enzyme; lane 2, FEN-1; lane 3, FEN-1 + wild-type WRN; lane 4, wild-type WRN. (B) % incision from (A) (mean value of three experiments) with standard deviation (SD) indicated by error bars. (C) Reactions (20 μl) containing 10 fmol of a 1 nt 5′ flap DNA substrate, 10 fmol of FEN-1 and the indicated amounts of WRN were incubated in the absence or presence of 40 mM KCl as indicated at 37°C for 15 min. A phosphorimage of a typical gel is shown. Lane 1, no enzyme; lane 2, 80 fmol of WRN; lane 3, FEN-1; lane 4, 80 fmol of WRN + FEN-1; lane 5, 40 fmol of WRN + FEN-1; lane 6, 20 fmol of WRN + FEN-1; lane 7, 10 fmol of WRN + FEN-1; lane 8, 5 fmol of WRN + FEN-1. (D) % incision from (C) (mean value of three experiments) with SD. Filled circles, no KCl; open circles, 40 mM KCl. Download figure Download PowerPoint We subsequently analyzed FEN-1 cleavage as a function of WRN concentration under standard conditions (40 mM KCl) or in the absence of KCl, a more optimal condition for FEN-1 incision (Harrington and Lieber, 1994). A limiting amount of FEN-1 (10 fmol) was used such that cleavage of the 1 nt 5′ flap is very low (∼1%, 40 mM KCl; ∼3%, no KCl) (Figure 3C, lane 3, and D). FEN-1 cleavage was reproducibly stimulated ∼2-fold at a WRN amount of 20 fmol (Figure 3C, lane 6, and D) in the presence or absence of KCl. In the presence of 40 fmol of WRN, FEN-1 cleavage increased to 17 and 31% incision in the presence and absence of KCl, respectively (Figure 3C, lane 5, and D). At 80 fmol of WRN, product formation began to plateau at 20% for the FEN-1 cleavage reaction conducted in the presence of 40 mM KCl, whereas cleavage continued to increase to nearly 60% in the absence of KCl (Figure 3C, lane 4, and D). Kinetic analysis of the FEN-1-catalyzed cleavage reaction on the 1 nt 5′ flap substrate demonstrated a dramatic stimulation of the rate of FEN-1 incision in the presence of WRN (Figure 4A). In these experiments, an amount of WRN was used that we previously determined to achieve maximum stimulation of FEN-1 cleavage (Figure 3D). The level of FEN-1 used resulted in a low, but reproducibly detectable incision of 2% of the 10 fmol of DNA substrate in a 15 min reaction incubated at 37°C (in the absence of WRN) (Figure 4A). Stimulation of FEN-1 incision by WRN was detected at time points as short as 0.5–1 min (Figure 4B). Up to 3 min, FEN-1 cleavage in the absence of WRN was below 1%; however, in the presence of WRN, FEN-1 cleaved 12% of the DNA substrate (Figure 4B). FEN-1 cleavage in the presence and absence of WRN was linear with respect to time from 0.5–3 min (R2 = 1.0 and 0.98, respectively). Linear regression analyses yielded reaction rates of 4.5 and 0.06 fmol product/min for the WRN + FEN-1 and FEN-1 only reactions, respectively. This represents an 82-fold rate increase when WRN is present. At 12 and 15 min, the FEN-1 cleavage reaction conducted in the presence of WRN achieved a plateau of 30–35% substrate incised. In contrast, FEN-1 alone only cleaved 1.5% of the substrate by the end of 15 min (Figure 4B). Figure 4.Kinetics of FEN-1 cleavage in the presence or absence of WRN. Reactions (160 μl) containing 80 fmol of 1 nt 5′ flap DNA substrate and 160 fmol of FEN-1 were incubated at 37°C under standard reaction conditions, and 20 μl aliquots were terminated at 0, 0.5, 1, 3, 6, 9, 12 and 15 min. The reactions in the presence of WRN contained 600 fmol of WRN. (A) Phosphorimage of a typical gel from a kinetic experiment. Increasing times of incubation (0–15 min) for the FEN-1 cleavage reactions conducted in the absence of WRN (lanes 3–9) or the presence of WRN (lanes 10–16) are indicated by the triangle. Fifteen minute incubations conducted in the absence of WRN + FEN-1 or FEN-1 are shown in lanes 1 and 2, respectively. (B) % incision for the reactions. Filled circles, FEN-1; open circles, FEN-1 + WRN. Download figure Download PowerPoint WRN stimulates FEN-1 cleavage more effectively than either PCNA or replication protein A Since replication protein A (RPA) and PCNA interact physically with WRN (Brosh et al., 1999; Lebel et al., 1999) and are capable of stimulating the FEN-1 cleavage reaction (Biswas et al., 1997; Tom et al., 2000), we tested WRN for the presence of either RPA or PCNA by western blot analysis. We examined the purity of recombinant WRN and it did not contain RPA or PCNA from insect cells (data not shown). We then tested purified PCNA and RPA for stimulation of FEN-1 cleavage on the 1 nt 5′ flap substrate. At molar amounts of PCNA homotrimer or RPA heterotrimer equal to WRN monomer (80 fmol) that displayed a 10-fold stimulation of FEN-1 cleavage, we did not detect any significant stimulation of FEN-1 cleavage by either PCNA or RPA (Figure 5). Stimulation of FEN-1 incision was detected at much higher amounts of PCNA (900 fmol homotrimer) (data not shown). These findings are consistent with a previous report (Tom et al., 2000) that PCNA can stimulate FEN-1 cleavage of the same 1 nt 5′ flap substrate used in this study. However, a large amount of PCNA (25 000 fmol) was used in that study because FEN-1 stimulation by PCNA is a diffusion-limited process that requires a large stoichiometric excess of PCNA (Burgers and Yoder, 1993; Wu et al., 1996). On a per mole basis, WRN is a significantly more effective stimulator of FEN-1 cleavage on the 1 nt 5′ flap substrate than either human PCNA or RPA. Figure 5.WRN stimulates FEN-1 cleavage much more effectively than either PCNA or RPA. Reactions (20 μl) containing 10 fmol of a 1 nt 5′ flap DNA substrate, 10 fmol of FEN-1 and 80 fmol of WRN, hRPA or hPCNA were incubated under standard reaction conditions at 37°C for 15 min. % incision (mean value of three experiments) with SD. Download figure Download PowerPoint WRN may serve to shield the negative charge on the DNA, thereby permitting FEN-1 to bind, and thus cleave the substrate more easily. If so, then another DNA-binding protein such as Ku might also stimulate FEN-1 cleavage. We tested the effect of Ku86/70 heterodimer on FEN-1 cleavage and did not detect any stimulation at a range of Ku concentrations (10–500 fmol) under the same conditions that WRN effectively stimulated FEN-1 cleavage (data not shown). Catalytic activities of WRN are not required for stimulation of FEN-1 incision The ATPase, helicase or exonuclease activities of WRN may play a role in the functional interaction with FEN-1. To address this, we tested the effects of full-length recombinant WRN proteins (Figure 1) with site-directed mutations in the active sites of its catalytic domains on FEN-1 cleavage. The WRN-K577M mutant protein, devoid of ATPase or helicase activity (Gray et al., 1997; Brosh et al., 1999), was capable of stimulating the FEN-1 cleavage reaction similarly to wild-type WRN (Figure 6A, lanes 3 and 4). In control reactions, WRN-K577M alone did not yield the products (Figure 6A, lane 6). Thus, ATP hydrolysis/DNA unwinding are dispensable for the WRN–FEN-1 functional interaction, consistent with our finding that WRN stimulates FEN-1 cleavage of the 1 nt flap substrate in the absence or presence of ATP (data not shown). Figure 6.Catalytic activities of WRN are not required for stimulation of the FEN-1 cleavage reaction. Reactions (20 μl) containing 10 fmol of 1 nt 5′ flap DNA substrate and 40 fmol of FEN-1 were incubated at 37°C for 15 min under standard conditions. The reactions in the presence of either wild-type or mutant WRN contained 38 fmol of WRN. (A) WRN-K577M stimulates FEN-1 cleavage. Lane 1, no enzyme; lane 2, FEN-1; lane 3, FEN-1 + wild-type WRN; lane 4, FEN-1 + WRN-K577M; lane 5, wild-type WRN; lane 6, WRN-K577M. (B) WRN-E84A stimulates FEN-1 cleavage. Lane 1, FEN-1; lane 2, no enzyme; lane 3, FEN-1 + wild-type WRN; lane 4, FEN-1 + WRN-E84A; lane 5, wild-type WRN; lane 6, WRN-E84A. Download figure Download PowerPoint To address the importance of WRN exonuclease activity in the WRN–FEN-1 functional interaction, we tested the ability of a WRN exonuclease defective mutant WRN-E84A (Huang et al., 1998; Cooper et al., 2000) to stimulate FEN-1 cleavage. As shown in Figure 6B, WRN-E84A retained the ability to stimulate FEN-1 incision (lane 4). In control reactions, WRN-E84A alone did not yield the cleavage products (Figure 6B, lane 6). These results indicate that the exonuclease activity of WRN is not required for the functional interaction between WRN and FEN-1. Mapping of the WRN domain that is important for stimulation of FEN-1 cleavage The ability of the catalytically defective mutant WRN proteins to stimulate FEN-1 cleavage suggested that the functional interaction is mediated by a direct protein interaction. The physical interaction between FEN-1 and the WRN C-terminus (Figure 2) raised the possibility that this region may stimulate FEN-1 cleavage by itself. The results shown in Figure 7 provide evidence that indeed this is the case. Using a limiting amount of FEN-1 (5 fmol), the cleavage reaction was stimulated 8.5-fold by GST–WRN949–1432 (Figure 7A, lanes 3 and 4, and B). Importantly, the GST–WRN1072–1236 fragment that did not bind to FEN-1 failed to stimulate FEN-1 incision (Figure 7A, lane 5). Using a 2-fold higher level of FEN-1 (10 fmol), 6% of the DNA substrate was cleaved (Figure 7A, lane 6, and B). At this FEN-1 level, GST–WRN949–1432 stimulated FEN-1 cleavage to 46% of the substrate incised (Figure 7A, lane 7, and B). GST–WRN1072–1236 again failed to stimulate FEN-1 cleavage (Figure 7A, lane 6), indicating that the functional interaction is specific to WRN residues 949–1432. Similar results were obtained using 20 fmol of FEN-1 (lanes 9–11), although the level of stimulation was not as great because a plateau of incision activity (∼58%) was approached in the reactions containing FEN-1 and GST–WRN949–1432 (Figure 7B). A highly purified WRN fragment without the GST moiety (His-WRN940–1432) (Cooper et al., 2000) (Figure 1B) is also competent to physically and functionally interact with FEN-1 (see Supplementary figure 1, available at The EMBO Journal Online), suggesting that the interaction is not associated with GST and is specific to the WRN sequence 940–1432. Figure 7.A C-terminal fragment of WRN retains the ability to stimulate the FEN-1 cleavage reaction. Reactions (20 μl) containing 10 fmol of 1 nt 5′ flap DNA substrate and the indicated amounts of FEN-1 were incubated at 37°C for 15 min under standard conditions. The reactions in the presence of GST–WRN fragments contained 75 fmol of WRN fragment. (A) Phosphorimage from a typical gel. Lane 1, no enzyme; lane 2, GST–WRN949–1432; lane 3, 5 fmol of FEN-1; lane 4, GST–WRN949–1432 + 5 fmol of FEN-1; lane 5, 5 fmol of FEN-1 + GST–WRN1072–1236; lane 6, 10 fmol of FEN-1; lane 7, 10 fmol of FEN-1 + GST–WRN949–1432; lane 8, 10 fmol of FEN-1 + GST–WRN1072–1236; lane 9, 20 fmol of FEN-1; lane 10, 20 fmol of FEN-1 + GST–WRN949–1432; lane 11, 20 fmol of FEN-1 + GST–WRN1072–1236. (B) % incision (mean value of three experiments) with SD. Filled circles, FEN-1; open circles, FEN-1 + GST–WRN949–1432; open squares, FEN-1 + GST–WRN1072–1236. Download figure Download PowerPoint To further map the domain of WRN that mediates the functional interaction with FEN-1, several additional recombinant GST–WRN fragments were tested. As shown in Figure 8A, lane 4, GST–WRN949–1236, a shortened version of GST–WRN949–1432 that lacks 196 amino acids at the extreme C-terminus (Figure 1), stimulated FEN-1 incision of the 1 nt 5′ flap substrate 5-fold compared with the reaction containing FEN-1 only (Figure 8A, lane 2). In control reactions, GST–WRN949–1236 alone did not produce the FEN-1 incision products (Figure 8A, lane 9). The level of FEN-1 stimulation by GST–WRN949–1236 was comparable to that of GST–WRN949–1432 (Figures 7A, lane 7, and 8B), suggesting that the last 196 amino acids of GST–WRN949–1432 are dispensable for stimulation of FEN-1 cleavage. GST–WRN1072–1236 (Figures 8A, lane 5, and B, and 7) or GST (data not shown) failed to stimulate FEN-1 cleavage, attesting to the specificity of GST–WRN949–1236 in the functional interaction with FEN-1. Figure 8.Mapping of the FEN-1 interaction domain that mediates the functional interaction between WRN and FEN-1. Reactions (20 μl) containing 10 fmol of 1 nt 5′ flap DNA substrate, 10 fmol of FEN-1 and 75 fmol of the indicated GST–WRN fragment were incubated at 37°C for 15 min under standard conditions. (A) Phosphorimage of a typical gel. Lane 1, no enzyme; lane 2, FEN-1; lane 3, FEN-1 + GST–WRN949–1432; lane 4, FEN-1 + GST–WRN949–1236; lane 5, FEN-1 + GST–WRN1072–1236; lane 6, FEN-1 + GST–WRN949–1092; lane 7, FEN-1 + GST–WRN239–499; lane 8, GST–WRN949–1432; lane 9, GST–WRN949–1236; lane 10, GST–WRN1072–1236; lane 11, GST–WRN949–1092; lane 12, GST–WRN239–499. (B) % incision (mean value of three experiments) with SD. Download figure Download PowerPoint Since GST–WRN949–1236 was able to stimulate FEN-1 incision whereas GST–WRN1072–1236 failed, the domain of WRN necessary