| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 280, Issue 7, 5391-5399, February 18, 2005
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
From the Department of Biochemistry and Biophysics, University of Rochester School of Medicine and Dentistry, Rochester, New York 14642
Received for publication, November 1, 2004 , and in revised form, November 30, 2004.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
-radiation (8). Chicken cells lacking FEN1 were viable but exhibited a lower proliferation rate than wild-type cells. A defect in chicken FEN1 also caused hypersensitivity to methylating agents and H2O2 (12). Collectively, these results indicate that FEN1 plays an essential role not only in DNA replication but also in repair.
Defects in RAD27, the FEN1 homologue in Saccharomyces cerevisiae, are not lethal (13, 14). However, growth of RAD27-null deletion mutants is restricted at 37 °C and reduced even at the permissive temperature of 30 °C. When transferred to the restrictive temperature, RAD27-null deletion mutants arrested with cell cycle morphology similar to that of cells containing mutations in DNA replication proteins, verifying a role for RAD27 in DNA replication (13, 14). In addition, it was demonstrated that both RAD27 and mammalian FEN1 are critical in suppressing mutations (9, 15, 16). RAD27 deletion mutants exhibit elevated sensitivity to methylmethanesulfonate but not x-rays or UV light, indicating that FEN1 is involved in DNA base excision repair (14). Increased homologous recombination and nonhomologous end joining events have been reported in RAD27 deletion mutants (17–19).
Importantly, several independent studies showed that duplication of sequences frequently occurs due to RAD27 mutation (18, 20) In addition, repetitive DNA is unstable in RAD27-null deletion mutants. RAD27 deletion leads to a dramatically increased expansion in sequences flanked by direct repeats (18) and trinucleotide repeats, and the expansion was biased to the 5'-end of the repeat tract (2, 21–23). A widely accepted model describes this process. During lagging strand synthesis, the displaced single strand can equilibrate to form foldback and bubble structures that inhibit FEN1. Ligation of a flap with a bubble to the upstream primer creates an expanded sequence (18, 24).
Biochemical studies further support this mechanism of repetitive sequence expansion. Flaps with increasing lengths of complementarity or CTG repeats are cleaved by FEN1 at reduced rates (25). We have shown previously that a flap in the bubble configuration can produce an expansion product (26). In addition, neither proliferating cell nuclear antigen nor RPA was found to assist FEN1 in cleaving hairpin substrates (25). These data imply that there are other accessory proteins assisting FEN1 in resolving secondary structures. Recently, it was reported that two members of the RecQ helicase family, BLM and WRN, interact with FEN1 and suppress the phenotypes of DNA2–1 mutants (27–29).
BLM is a 3'-5' DNA helicase that, when mutated, is responsible for Bloom's syndrome (30). Knocking out BLM caused embryonic lethality in mice (31). Defective BLM produced live mice prone to tumorigenesis (32). Humans with heterozygous BLM have an increased risk of developing colorectal cancer (33, 34). BLM-mutated mouse or chicken cells exhibit a high level of sister chromatid exchanges and chromosomal breaks as well as interexchanges as seen in human BS cells (31, 32, 35, 36). In addition, immunodepletion of Xenopus BLM inhibits the replication of DNA in reconstituted nuclei, providing direct evidence that BLM is involved in DNA replication (37). BS cells were found to grow slower than wild-type cells, elongation of nascent DNA strands was retarded, and abnormal replication intermediates were accumulated (38). It was suggested that BLM plays a role in recovering stalled replication forks (35, 39). It was reported in BS cells that relative amounts of total ribosomal DNA decreased, but telomere DNA increased. Both ribosomal DNA and telomere DNA contain repeated sequences (40). These data imply that BLM acts to stabilize repeated sequences.
Recently, human BLM was reported to restore the phenotypes of the yeast Dna2-1 mutant, but not the FEN1 (RAD27) deletion mutant (27). The complementation was dependent on BLM helicase activity. Dna2p has been suggested to participate with FEN1 in Okazaki fragment maturation (41, 42). Dna2p is a 5'-3' helicase and 5'-3' nuclease. It has been shown that the helicase activity assists the nuclease activity in cleaving structured flaps (43). BLM interacts with both Dna2p and FEN1 (27). These facts suggest that BLM participates in Okazaki fragment maturation. In fact, it has been demonstrated that both BLM and WRN stimulate FEN1 by interaction through their C-terminal domains (RQC domains) (28, 44). This stimulation was independent of BLM helicase activity. However, analysis of BS cells showed that the BLM lacking helicase activity could not suppress elevated sister chromatid exchanges in BS cells, indicating that helicase activity is an important BLM function. Although BLM Q672R and C1055S are not mutated within the helicase domain of BLM protein, these two mis-sense mutants lost their helicase activity, similar to K695T, which is mutated in the first helicase motif (45). The relationship between the RQC domain of BLM and sister chromatid exchanges is still unclear. In vitro, BLM was shown to be a potent helicase that can unwind many kinds of double-stranded DNA including bubbles and G4 quadraplexes (46–48).
Suppression of the Dna2-1 mutant but not Rad27 mutants by BLM and suppression of the Dna2-1 mutant by overexpression of FEN1 suggest that Dna2p acts in a different pathway than BLM and FEN1. We considered the possibility that BLM uses its helicase activity to aid FEN1 in resolving structured flaps. Here we examine whether BLM promotes FEN1 cleavage of foldbacks and bubbles associated with DNA expansion (26). Our results reveal that activation of helicase function allows BLM to stimulate the cleavage of foldback flaps, bubbles, or triplet repeats by FEN1. Cleavage after addition of BLM and ATP is distinctly greater than that after addition of only BLM. These results suggest that BLM plays a role in preventing repeat expansion by aiding FEN1 cleavage.
|
EXPERIMENTAL PROCEDURES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
-32P]ATP (3000 Ci/mmol) and [
-32P]dCTP (6000 Ci/mmol) were bought from PerkinElmer Life Sciences. The T4 polynucleotide kinase (labeling grade), the Klenow fragment of DNA polymerase I, and ATP were from Roche Applied Science. RNase-free solutions and reagents were from Ambion, Inc. (Austin, TX). All the other reagents were the best available commercial grade. Enzyme Expression and Purification—Recombinant human FEN1 was expressed with the T7 expression plasmid pET-FCH and purified (49). Recombinant human BLM was expressed in yeast strain JEL1 transformed with pJK1. BLM was purified as described previously (46)
Oligonucleotide Substrates—Oligomer sequences are listed in Table I. They were designed to form a series of flap or bubble substrates. Each substrate consisted of a downstream primer (D), a template (T), and an upstream primer (U). Substrates were generated including different foldback structures or CTG repeats on the downstream primer. For flap substrates, the 3'-end of each downstream primer was complementary to the 5'-end of its appropriate template. The 5'-end of the downstream primer formed the unannealed 5'-tail or flap. The 5'-end of downstream primers containing secondary structure formed a hairpin loop or foldback. The annealing length of the foldback was 18 nucleotides. Upstream primers were annealed to the 3'-end of the template, forming a 1-nucleotide 3'-flap at the annealing point of the downstream primer. Bubble substrates contained 1–15 nucleotides at the 5'-end of primer and 12–28 nucleotides at 3'-end of the primer, which were complementary to the template. Nucleotides in the middle region were not complementary to the template, allowing the formation of the bubble. Upstream primers were annealed to the 3'-end of the template in most experiments, forming a nick at the 5'-end of the bubble. Substrates were constructed as described in the figure legends. The equilibrating substrates were generated by annealing the downstream primer to the appropriate template and then annealing an upstream primer that contained a 30-nucleotide sequence overlapping with the 5'-end of the downstream primer.
|
-32P]dCTP. The radiolabeled primers were purified on 15–18% denaturing polyacrylamide gels containing 7 M urea. Annealing conditions were the same as those described previously (20), except that the annealing ratio was 1:2:5 (downstream/template/upstream). Enzyme Assays—Reactions were performed in 30 mM HEPES (pH 7.5), 5% glycerol, 40 mM KCl, 0.1 mg/ml bovine serum albumin, and 8 mM MgCl2 with or without 4 mM ATP. Enzyme stocks were diluted in 30 mM HEPES (pH 7.5), 5% glycerol, 40 mM KCl, and 0.1 mg/ml bovine serum albumin. Each reaction contained 5 fmol of substrates in a 20-µl reaction mix with different amounts of the enzymes as indicated in the figure legends. All the assays, except helicase assays, were preincubated at 37 °C for 5 min with BLM, incubated at 37 °C for 10 min after addition of FEN-1, and then stopped by the addition of 20 µl of 2x termination dye (95% formamide (v/v) with bromphenol blue and xylene cyanole). BLM helicase assays were performed at 37 °C for 20 min and then immediately run on a native 12% polyacryamide gel. The denatured reactions were resolved on 15% polyacrylamide, 7 M urea denaturing gels. Each gel was quantitated using a PhosphorImager (Amersham Biosciences) and analyzed using ImageQuant version 1.2 software from Amersham Biosciences. In all studies, the quantitated amounts of substrates and products were utilized to calculate the percentage of product formation from the product/(substrate + product) ratio. This method allows for the correction of any loading errors among lanes. All assays were performed at least in triplicate, and representative assays are shown.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
Ability of FEN1 to Cleave Flaps with Double-stranded Regions Was Increased by BLM—FEN1 has previously been shown to load onto flap substrates by entering from the 5'-end and then tracking to the base of the flap for cleavage (16). Flaps with annealed primers or foldback structures have been shown to inhibit FEN1 loading, decreasing the efficiency of cleavage (25). We examined whether BLM helicase activity increased FEN1 cleavage of foldback flaps. We designed substrates with an 18-nucleotide inverted repeat sequence on the flap. The inverted 18 nucleotides annealed with their complementary sequence to form a foldback stem-loop structure. The loop contained 6 nucleotides. One had a single-stranded region at the 5'-end of the flap, and one lacked this region. We designed another substrate with a 37-nucleotide flap that had an 18-nucleotide blocking primer annealed at its 5'-end. This effectively creates a stem structure without the loop. The same 37-nucleotide flap substrate, but without the 18-nucleotide blocking primer, was used as a control for comparison of cleavage efficiency. All of the downstream primers were labeled at the 3'-end with [-32P]dCTP. Upstream primers (26 nucleotides in length) designed to form a 1-nucleotide flap at the 3'-end were used to make double flap substrates.
Cleavage of these substrates generates a 26-nucleotide labeled product (Fig. 1A). 1 fmol of FEN1 was used for the unstructured flap, and 5 fmol of FEN1 was used for the flaps with structure. Even with the higher amount of nuclease, substrates with blocking primer or foldback structure were cleaved less efficiently than the unstructured flap. In the absence of ATP, the addition of BLM increased amounts of product formation with all substrates (Fig. 1A, lanes 4, 9, 14, and 19). This observation is consistent with the ATP-independent stimulation of FEN1 observed previously (44). In the presence of ATP, an additional increment of cleavage was observed with the structured flap substrates that was not observed with the unstructured flap substrate (Fig. 1A, lanes 5, 10, 15, and 20). We also tested 1-, 4-, and 11-nucleotide unstructured flap substrates. In each case, BLM stimulated cleavage by FEN1, but there was no further increment of stimulation when ATP was added to activate helicase function (data not shown). These results suggest that BLM uses its helicase activity to unwind structured flaps. This allows FEN1 to load more effectively, resulting in more efficient cleavage.
|
BLM Increased the Ability of FEN1 to Cleave Bubble Substrates—Another abnormal flap intermediate that could form in regions with sequence flanked by a direct repeat or short sequence repeats is the bubble. Natural bubbles would form by misaligned annealing of the 5'-end region of the flap. It has also been proposed that foldback structures of triplet repeats form bubble structures, which are sealed into the double helix to produce sequence expansion (24). Bubble structures have been shown to be effective inhibitors of FEN1 (25, 50). A 24-nucleotide oligo(dT) bubble was designed into the downstream primer. 7 nucleotides were annealed to the template at the 5'-end, followed by the bubble and a 28-nucleotide region annealed at the 3'-end. The reaction was carried out under the same conditions used with the structured flaps, with similar results. BLM stimulated FEN1 without ATP. However, more products were generated in the presence of ATP (Fig. 2A, lane 25) compared with the absence of ATP (Fig. 2A, lane 24). This result suggests that BLM loads onto the single-stranded region of the bubble in the downstream primer and moves toward the 5'-end, ultimately displacing the 5' annealed region. This would produce an unstructured flap suitable for maximum efficiency of FEN1 cleavage. Expecting that additional BLM would produce more stimulation, we titrated the BLM level in the reaction (Fig. 2B). Unexpectedly, high concentrations of BLM actually suppressed cleavage in the presence of ATP (Fig. 2B). Maximal stimulation occurred at levels of BLM from 2.4 to 12 fmol, at which 4.5x and 4.6x stimulation were obtained. Even when 24 fmol of BLM was used, 3.5x stimulation was still observed. But 120 fmol of BLM inhibited FEN1 cleavage. In these experiments, about 5 fmol of substrates was used. The ratio of BLM to substrates was in the range of 1:2 to 5:1 for stimulation. A ratio of BLM to substrate over 20:1 was inhibitory. Moreover, in the absence of ATP, increasing the BLM concentration augmented the ATP-independent form of stimulation (data not shown). With ATP, BLM was most effective over an intermediate concentration range. It is possible that excess BLM binds the flaps in a way that blocks FEN1 entry.
|
BLM Stimulates FEN1 Irrespective of Flap Bubble Size— Mohaghegh et al. (47) pointed out that double-stranded DNA containing a bubble shorter than 4 nucleotides was a poor substrate for BLM unwinding. To examine the effects of bubble size, we synthesized substrates with bubble lengths of 3, 6, 12, 18, and 24 nucleotides. Assay results showed that BLM was able to stimulate FEN1 cleavage on all of the bubble sizes tested (Fig. 2A). In all cases, both the basic level of stimulation and an increment produced by helicase function were observed. However, small bubbles were more resistant to cleavage than large bubbles. This was true for FEN1 only, after the addition of BLM or BLM plus ATP. It is likely that bigger bubbles are more naturally prone to spontaneous 5' unannealing in both the absence and presence of BLM and BLM helicase (Fig. 2A) (50).
BLM Stimulates FEN1 Irrespective of Template Bubble Size—Asa3' to 5' helicase, BLM can also bind single-stranded regions of the template and unwind in the direction of the downstream primer. This type of activity would not stimulate and could inhibit FEN1. To examine such effects, we created substrates with template single-strand sizes of 1, 3, 7, 12, and 15 nucleotides. No significant difference in ATP-dependent stimulation, compared with the base level stimulation without ATP, was observed with the change in template (Fig. 3). ATP-dependent stimulation is presumed to be derived from the BLM helicase function. This indicates that the primary action of BLM is to track on the downstream primer and unwind the 5'-end region of the flap. It is possible that the 28-nucleotide annealed region of the downstream primer is long enough to effectively resist the helicase action of BLM.
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
The C-terminal domain (RQC domain) of BLM and WRN has been found to associate with FEN1 (27, 28, 44). The RQC domain (which does not harbor the helicase function) alone was enough to stimulate both the endo- and exonuclease activity of FEN1 (28, 44). This kind of stimulation occurred by protein-protein interaction and was not caused by creating FEN1 substrates (44). However, genetic studies revealed that BLM helicase activity is required to complement the phenotypes of BS cells (45). This led us to hypothesize that the BLM helicase has an important role in Okazaki fragment maturation. In this study, we demonstrate that the ability of BLM helicase to relieve foldback and bubble structures in flaps improves FEN1 cleavage efficiency. The key difference from the previous study was the demonstration that BLM can use its helicase activity to stimulate FEN1 endonuclease activity. Moreover, this capacity is evident on substrates with structured flaps, but not on substrates with unstructured flaps. As shown in Figs. 1 and 2, although BLM stimulated FEN1 cleavage of foldback or bubble substrates in the absence of ATP, a greater stimulation was observed after ATP was added. The additional increment of stimulation presumably derives from the greater access for FEN1 loading provided by the disruption of secondary structure. In vivo, this would make the subset of natural flaps that have structure more accessible to FEN1.
These results are consistent with the finding that only wild-type BLM, not a BLM mutant that lacks helicase activity, complemented DNA2-1 mutant phenotypes (27). However, the additional observation that the WRN RQC domain, lacking helicase, can substantially complement DNA2-1 mutant phenotypes (27, 29) suggests differences in the mechanisms by which BLM and WRN support FEN1 function in vivo. It has been shown that BLM without helicase activity cannot complement elevated sister chromatid exchange and chromosome abnormalities (45, 56). Defects in RAD27 cause increased mitotic recombination. In yeast, the double mutants of RAD27 and the hBLM homologue Sgs1 were lethal (57). However, this point is still controversial (58). Because of the role of FEN1 in removing Okazaki fragment flaps, it was proposed that FEN1 defects result in double-strand breaks because of more extensive strand displacement (18). In the absence of BLM, the secondary structure of some flaps could inhibit FEN1 until those flaps have been extensively displaced. This situation would not only lead to foldbacks and bubbles but would also promote strand breakage. This suggests the connection between loss of BLM helicase activity and genomic instability.
The current model of Okazaki fragment processing proposes that the flaps displaced by polymerase 
normally are rapidly cleaved by FEN1 before they are more than a few nucleotides long (59). The presence of structure allows the flaps to inhibit FEN1 and thereby achieve a greater length. In these cases, the helicase/nuclease Dna2p has been proposed to aid flap processing (59, 60). Flaps in the 30-nucleotide size range have been shown to be coated by RPA. The RPA stimulates the nuclease action of Dna2p and inhibits FEN1 (60). Bae et al. (43) have proposed that this orders the action of the two nucleases. The RPA-coated flaps are cleaved by Dna2p, making them too short to retain RPA. The shorter flaps are then susceptible to cleavage by FEN1 to form a nick for ligation. Dna2p is also a helicase. However, our analyses of Dna2p in vitro suggest that its helicase activity is too weak to resolve some secondary structures in flaps, even in the presence of RPA (61). We suggest BLM as a likely candidate for this role. This suggestion is based on the results presented here and the demonstration that BLM interacts with Dna2p (27). We suggest that BLM assists Dna2p in resolving flaps containing secondary structures. Such an interpretation is also consistent with the observation that overexpression BLM or FEN1 suppresses the phenotypes of the DNA2-1 mutant (27, 62).
Our results showed that annealing length strongly affects the efficiency of FEN1 stimulation by BLM. Stimulation was most effective when the 5' annealed region of the bubble was short, making the bubble structure relatively unstable. When the 5'-end annealing length of the bubble was longer, the amount of stimulation originating from both the RQC domain activity and helicase activity diminished. The weakening of helicase activity with annealing length is consistent with an earlier report showing that it becomes difficult for BLM to unwind longer double strands (51). Aside from the annealing length, the size of the single-stranded region of the bubble also influences bubble stability. We noticed previously that a large single-stranded region destabilizes a bubble so that it is more likely to equilibrate to a flap sensitive to FEN1 (63). This destabilization effect also allows BLM to be a highly effective stimulator. On the other hand, decreasing the annealing length beyond the bubble on the downstream primer abolished stimulation by BLM helicase. This was interpreted as indicating that BLM can load and track on the template strand and displace the downstream primer. However, the disruptive effects of template loading are not likely to be significant in vivo, where Okazaki fragments are much longer than our downstream primers. Overall, these results show that BLM helicase can be a very effective remover of structures that inhibit FEN1 and can lead to duplications, expansions, and other genome disruption. However, there are limitations imposed by very stable structures that even the BLM helicase cannot alleviate. It is possible that the BLM helicase has evolved a capacity that matches the structures likely to be encountered in vivo. It may not have evolved to be more potent so that it would not displace the entire Okazaki fragment.
Repeat regions in the genome are thought to expand because they are facile at forming foldback and bubble intermediates. Increasing the efficiency of FEN1 suppresses the formation of these intermediates by rapidly resolving flap structures (50, 55). Stimulation of FEN1 by proliferating cell nuclear antigen is one mechanism employed by cells for rapid flap resolution. Our results also show that the presence of BLM promotes FEN1 cleavage of flaps that contain triplet repeats. We propose that the presence of BLM in vivo constantly resolves secondary structure in flaps to allow better accessibility of FEN1. This capacity applies to a variety of structures including triplet repeats.
It is not clear how the RQC domain of BLM works with the helicase domain to stimulate FEN1. FEN1 interacts directly with and is stimulated by the RQC region of BLM (44). However, BLM is a 3'-5' helicase, and FEN1 is a 5'-3' endonuclease. We could envision the BLM protein moving toward the 5'-end of the flap and then exiting to allow FEN1 to track over the 5'-end and then down to the point of cleavage. However, the RQC domain apparently interacts directly with FEN1 to stimulate the cleavage event. If both proteins remain associated with each other, BLM would carry FEN1 away from the cleavage site. If the BLM associated with FEN1 tracked down the flap to the cleavage site, it would be moving in the opposite direction as its helicase function. Why are these proteins not antagonistic? Are two different BLM proteins needed for maximum stimulation? Does a single BLM track up the flap for structure resolution and then back down with the FEN1? These difficult questions are the material of future experimentation.
|
FOOTNOTES |
|---|
To whom correspondence should be addressed: Dept. of Biochemistry and Biophysics, Box 712, University of Rochester School of Medicine and Dentistry, 601 Elmwood Ave., Rochester, NY 14642. E-mail: robert_bambara{at}urmc.rochester.edu.
1 The abbreviations used are: FEN1, flap endonuclease 1; BLM, Bloom syndrome mutated protein; RPA, replication protein A.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  E. Larsen, L. Kleppa, T. J. Meza, L. A. Meza-Zepeda, C. Rada, C. G. Castellanos, G. F. Lien, G. J. Nesse, M. S. Neuberger, J. K. Laerdahl, et al. Early-Onset Lymphoma and Extensive Embryonic Apoptosis in Two Domain-Specific Fen1 Mice Mutants Cancer Res., June 15, 2008; 68(12): 4571 - 4579. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. Chen, S. Zhong, X. Zhu, B. Dziegielewska, T. Ellenberger, G. M. Wilson, A. D. MacKerell Jr., and A. E. Tomkinson Rational Design of Human DNA Ligase Inhibitors that Target Cellular DNA Replication and Repair Cancer Res., May 1, 2008; 68(9): 3169 - 3177. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
3
indicates that the length of the unannealed region on the template is 3 nucleotides. nt, nucleotides. B, excessive BLM inhibited FEN1 cleavage of bubble substrates (D8:T2:U2). ATP was added in each reaction. Lanes 1–6 contained substrates and BLM (0, 2.4, 12, 24, 120, and 240 fmol, respectively). Lanes 7–12 contained substrates, BLM (0, 2.4, 12, 24, 120, and 240 fmol, respectively), and 5 fmol of FEN1. The labeled products of FEN1 cleavage were 28 nucleotides long. 24 represents a 24-nucleotide bubble in the downstream primer. 
