FEN1 functions in long patch base excision repair under conditions of oxidative stress in vertebrate cells

Abstract

From in vitro studies, flap endonuclease 1 (FEN1) has been proposed to play a role in the long patch (LP) base excision repair (BER) subpathway. Yet the role of FEN1 in BER in the context of the living vertebrate cell has not been thoroughly explored. In the present study, we cloned a DT40 chicken cell line with a deletion in the FEN1 gene and found that these FEN1-deficient cells exhibited hypersensitivity to H(2)O(2). This oxidant produces genotoxic lesions that are repaired by BER, suggesting that the cells have a deficiency in BER affecting survival. In experiments with extracts from the isogenic FEN1 null and wild-type cell lines, the LP-BER activity of FEN1 null cells was deficient, whereas repair by the single-nucleotide BER subpathway was normal. Other consequences of the FEN1 deficiency were also evaluated. These results illustrate that FEN1 plays a role in LP-BER in higher eukaryotes, presumably by processing the flap-containing intermediates of BER.

Figure 1. Sensitivity of FEN1 null cells to H₂O₂

(A) Survival curves of wild type (wt) and FEN1 null cells (FEN1^−/−), and (B) wild type (wt) and FEN1 null cells stably overexpressing FEN1 (FEN1^−/− + wtFEN1 exposed to H₂O₂. Mean data and S.D. (bars) were from means of duplicate experiments.

Figure 2. Kinetic analysis of BER for oxidative DNA damage using a 8oxoG-containing oligonucleotide duplex DNA substrate

(A) After separation of whole cell extract by SDS-PAGE, proteins were transferred to a nitrocellulose membrane that was then probed with anti-FEN1 (upper panel) and anti-G3PDH (lower panel) antibodies. (B-D) Incorporation of [α-³²P]dGMP was measured as a function of incubation time using various DT40 cell extracts. (B) Schematic diagram of a 100-bp oligonucleotide containing an 8oxoG residue. ‘X’ represents the position of 8oxoG. (C) Photographs of PhosphorImager analysis illustrating 8oxoG-DNA BER are shown. The left two lanes show the marker oligonucleotide after the nicking reaction without and with Ogg1 and APE. The nicked product with Ogg1 and APE yields the 23-bp product. The mobility of the 100-mer oligonucleotide was slightly faster than the ligated product due to the presence of a 5′-phosphate group. (D) The relative amount of ligated BER product formed during a 20 min incubation is represented. The experiments were repeated 3 times, and the initial rates were calculated by using a curve fit program as a function of time of incubation. The average initial rate of activity of each extract for the in vitro 8oxoG-DNA BER reaction is shown in a bar diagram.

Figure 3. Analysis of SN- and LP-BER capacities for oxidative DNA damage using an 8oxoG-containing oligonucleotide duplex DNA substrate

Incorporation of dGMP (G) was measured in the presence of ddCTP (ddC) to discriminate between SN-BER and LP-BER. (A) Schematic representation of the substrate DNA and predicted BER reaction products and intermediates. The sizes and intermediates were 1-nt addition, SN-BER (23-bp); 2-nt addition, LP-BER (24-bp); and complete BER product [ligated SN-BER (100-bp)]. A 100-bp oligonucleotide containing an 8oxoG residue at position 23 was utilized in the BER assay. In SN-BER, dGMP was incorporated in place of 8oxoG, and the intermediate was directly ligated to complete the repair. In LP-BER, ddCMP was incorporated following incorporation of dGMP in place of 8oxoG lesion, and the resulting dideoxy-terminated intermediate prevented subsequent primer extension DNA synthesis, as well as the ligation reaction. (B) Photograph of PhosphorImager image illustrating LP-BER analysis. A 100-bp duplex oligonucleotide containing an 8oxoG residue at position 23 was incubated with dGTP, ddCTP, and cell extracts. The incubations were performed with wild type (wt) or FEN1 null (FEN1^−/−) DT40 cell extracts. (C) The relative amount of LP-BER product (24-mer) formed is represented. The experiments were repeated 3 times. The initial rate of activity of each extract was calculated as described in Figure 2C and shown in a bar diagram.

Figure 4. Kinetic analysis of BER using a uracil-containing oligonucleotide duplex DNA substrate

(B and C) Incorporation of [α-³²P]dCMP was measured as a function of incubation time using various DT40 cell extracts. (B) Sequence of a 35-bp oligonucleotide containing a uracil residue. (C) Photographs of PhosphorImager analysis illustrating uracil-DNA BER are shown. (D) The relative amount of ligated uracil-BER product is represented.

Figure 5. Analysis of SN- and LP-BER capacities using a uracil-containing oligonucleotide duplex DNA substrate

Incorporation of [α-³²P]dCMP (*C) was measured in the presence of ddTTP (ddT) to discriminate between SN-BER and LP-BER. (A) Schematic representation of the substrate DNA and predicted BER reaction products and intermediates. The sizes and intermediates were 1-nt addition, SN-BER (15-bp); 2-nt addition, LP-BER (16-bp); and complete BER product (ligated SN-BER [35-bp]). A 35-bp oligonucleotide containing a uracil residue at position 15 was utilized in the BER assay. In SN-BER, [α-³²P]dCMP was incorporated in place of uracil and directly ligated to complete the repair. In LP-BER, ddTMP was incorporated following incorporation of [α-³²P]dCMP in place of uracil, and the resulting dideoxy-terminated intermediate prevented subsequent primer extension DNA synthesis, as well as the ligation reaction. The asterisks designate the position of the radiolabeled CMP group. (B and C) Photographs of PhosphorImager image illustrating LP-BER analysis. A 35-bp duplex oligonucleotide containing a uracil residue at position 15 was incubated with [α-³²P]dCTP, ddTTP, and cell extracts. The incubations were performed with wild type (wt) or FEN1 null (FEN1^−/−) DT40 cell extracts. (C) The reaction mixtures with FEN1 null extract were supplemented with 20 and 100 nM purified FEN1 as shown above the gel. (D) The relative amount of LP-BER product (16-mer) was plotted against incubation time.

Figure 6. Caspase-dependent cell death in FEN1-deficient cells exposed to H₂O₂

(A) DNA fragmentation in wild type (wt) and FEN1 null (FEN1^−/−) cells at 3 hrs after H₂O₂ (10 μM) treatment in the absence or presence of the caspace inhibitor BAF. (B) Frequency of apoptotic cells was expressed as the percentage of early-stage apoptosis (annexin V-positive/PI-negative) cells in the total number of cells assayed (e.g., sum of all quadrants).

Figure 7. IdU incorporation in FEN1-deficient cells exposed to H₂O₂

(A) IdU Immuno-slot-blot X-ray film of blotted DNA samples extracted from cells exposed to PBS or H₂O₂ at 20 μM for 30 min. (B) Amount of IdU incorporated into genomic DNA in cells expressed as a percentage of the wild type (wt) control. The amount of IdU was quantified from multiple blots. Mean data and S.D. (bars) were from triplicate samples.

Figure 8. Visualization of various stages of replication by DNA fiber combing and immunostaining

(A) Schematic illustration of results expected when genomic DNA from asynchronous cells (pulsed for 10 min with IdU followed by 20 min pulse with CldU) is aligned and straightened on a glass slide (fiber spread); the incorporated halogenated nucleotides are visualized by immunofluorescence. The various stages of DNA synthesis can be inferred by the presence and relative position of single and/or double-labeling in continuous replication tracks. The small letters in the schematic refer to the actual tracks illustrated in B. (B) Field of replication tracks as detected by fluorescence microscopy, containing representative examples of labeling. (C) Illustration of possible oxidative DNA damage effects on the various stages of replication.

Figure 9. H₂O₂-induced damage influences replication dynamics in wild type and full complement FEN1 null cells

(A) Bar graph showing the increase in premature termination of replication tracks (labeled with first pulse (red) only, white bars) relative to the PBS-treated cells (dark bars) in wild type, FEN1 null, and FEN1 null cells with ectopic expression of FEN1. (B) Bar graph of the amount of tracks showing only the second pulse (green), indicating replication units initiated after H₂O₂ treatment relative to controls (dark bars) in the two cell types. At least 50 red-green tracks were analyzed per experimental condition.