Novel function of the flap endonuclease 1 complex in processing stalled DNA replication forks

Abstract

Restarting stalled replication forks partly depends on the break-induced recombination pathway, in which a DNA double-stranded break (DSB) is created on the stalled replication fork to initiate the downstream recombination cascades. Single-stranded DNA gaps accumulating on stalled replication forks are potential targets for endonucleases to generate DSBs. However, it is unclear how this process is executed and which nucleases are involved in eukaryotic cells. Here, we identify a novel gap endonuclease (GEN) activity of human flap endonuclease 1 (FEN-1), critical in resolving stalled replication fork. In response to replication arrest, FEN-1 interacts specifically with Werner syndrome protein for efficient fork cleavage. Replication protein A facilitates FEN-1 interaction with DNA bubble structures. Human FEN-1, but not the GEN-deficient mutant, E178A, was shown to rescue the defect in resistance to UV and camptothecin in a yeast FEN-1 null mutant.

Figure 1

Identification of GEN from HeLa cell nuclear extracts. (A) Purification scheme for the GEN activity. The specified fraction in each step contained the GEN activity. (B) GEN cleavage of 5′ (left) or 3′ (right) DNA double-stranded flap substrates (oligos 1, 2, 4, 5) by HiTrap-SP-HP fractions 12–17. A 500 ng portion of each fraction was assayed and the reaction was analysed with sequencing PAGE. Lanes 2–7 in both panels correspond to fractions 12–17, respectively. (C) Detection of FEN-1 protein in GEN fractions by western blotting with purified anti-FEN-1 antibody. (D) FEN-1 accounts for the identified GEN activity. FEN-1 was depleted from F17 with anti-FEN-1-bound protein A-agarose. Nuclease activities were assayed with 500 ng F17, 500 ng FEN-1-depleted F17 and 500 ng FEN-1-depleted F17 supplemented with 100 ng of purified recombinant human (h)FEN-1. NE, nuclear extract.

Figure 2

Cleavage of DNA bubble and flap substrates by FEN-1. (A) DNA flap (oligos 7, 9, 15) and bubble substrates (oligos 7, 8, 13). (B) Analysis of FEN-1 cleavage of DNA substrates with native PAGE. A 2 pmol portion of human FEN-1 was incubated with 1 pmol of indicated DNA substrates. DNA substrates and products were analysed with native PAGE. Lanes 1 and 2 are ssDNA (oligo 16) and dsDNA (oligos 7, 8, 11, 12) markers, respectively. (C) Mapping of cleavage site of DNA substrates by FEN-1. Similar reactions to those in (B) were analysed with sequencing gel. Lane 1 is the molecular marker.

Figure 3

WRN interacts with FEN-1 and stimulates its GEN activity. (A) Immunoprecipitation of FEN-1. Nuclear extracts from CPT-treated or untreated HeLa S3 cells were incubated with purified polyclonal anti-FEN-1-bound protein A–agarose. The precipitated FEN-1, WRN and PCNA were detected with monoclonal anti-FEN-1, anti-WRN and anti-PCNA, respectively. The input levels of FEN-1, WRN and PCNA were also examined. The purified recombinant FEN-1, WRN and PCNA were included as a control. (B) WRN and PCNA effects on GEN and FEN activities of FEN-1. The GEN activity was assayed by incubation of 1 pmol of FEN-1 with 1 pmol of 5′ and 3′ double-stranded flap (lagging (oligos 1, 2, 4, 6) and leading (oligos 1, 3, 4, 5) substrates, respectively) or bubble substrates in the absence or presence of 5 pmol of PCNA, WRN, WRNC or 50 U RFC. FEN activity was assayed by incubation of 0.1 pmol of FEN-1 with 1 pmol of normal flap and double flap (d-flap, oligos 7, 10, 15) substrates in the absence or presence of 5 pmol of PCNA, WRN, WRNC or 50 U RFC. The DNA substrates and products were resolved with sequencing gel and quantified with the Imagine Quante program. Values are an average of three independent experiments.

Figure 4

RPA effects on FEN-1 bubble DNA substrate binding and cleavage. (A) RPA stimulates FEN-1 binding to DNA bubble structures. DNA bubble substrate-bound agarose beads were left uncoated or coated with RPA or E. coli SSB. The beads were then incubated with 5 and 10 pmol of human (h)FEN-1. Binding of FEN-1 to the agarose beads was evaluated with western blotting using antibody against FEN-1. (B) RPA and p-RPA concentration effect on the GEN activity of FEN-1. A 2 pmol portion of FEN-1 was incubated with 1 pmol of DNA substrates in the presence of 4, 10 and 20 pmol of RPA or p-RPA for 30 min at 30°C. GEN cleavage was quantified (lower panel).

Figure 5

E178A is a GEN-deficient FEN-1 mutant. (A,B) Cleavage of DNA double flap and bubble substrates by wt FEN-1 and E178A. A 0.5 or 5 pmol portion of FEN-1 proteins was incubated with 2 pmol of DNA double flap (A) or bubble substrates (B), respectively, for 2, 10, 20, 40 and 60 min at 37°C. GEN and FEN cleavages were quantified (lower panel). (C) WRN effect on wt FEN-1 and E178A cleavage of DNA bubble structures. A 2 pmol portion of FEN-1 was incubated with 1 pmol of bubble structure DNA in the absence or presence of 10 pmol WRN. (D) DNA substrate binding by wt FEN-1 and E178A. Agarose-bound biotinylated DNA flap or bubble substrate was incubated with 5 or 10 pmol of FEN-1 proteins. FEN-1 bound to agarose was evaluated with western blotting.

Figure 6

Complementation study of wt FEN-1 and the GEN-deficient mutant E178A. (A) Temperature sensitivity of different yeast strains. The temperature sensitivity was expressed as relative colony-forming units (CFU) grown at 37°C over those grown at 30°C. (B) Lys+ reverse mutation rate of different yeast strains. Values represent an average of the mutation rate for 28 independent colonies of each strain. (C) CPT sensitivity of different yeast strains. The CPT sensitivity was measured as survival of cells treated with indicated concentration of CPT over survival of untreated cells. (D) UV irradiation sensitivity of different yeast strains. UV sensitivity was measured as survival of cells exposed to the indicated intensity of UV irradiation over nonexposure survival of cells.