Overexpression of PCNA Attenuates Oxidative Stress-Caused Delay of Gap-Filling during Repair of UV-Induced DNA Damage

Abstract

UVC irradiation-caused DNA lesions are repaired in mammalian cells solely by nucleotide excision repair (NER), which consists of sequential events including initial damage recognition, dual incision of damage site, gap-filling, and ligation. We have previously shown that gap-filling during the repair of UV-induced DNA lesions may be delayed by a subsequent treatment of oxidants or prooxidants such as hydrogen peroxide, flavonoids, and colcemid. We considered the delay as a result of competition for limiting protein/enzyme factor(s) during repair synthesis between NER and base excision repair (BER) induced by the oxidative chemicals. In this report, using colcemid as oxidative stress inducer, we showed that colcemid-caused delay of gap-filling during the repair of UV-induced DNA lesions was attenuated by overexpression of PCNA but not ligase-I. PCNA knockdown, as expected, delayed the gap-filling of NER but also impaired the repair of oxidative DNA damage. Fen-1 knockdown, however, did not affect the repair of oxidative DNA damage, suggesting repair of oxidative DNA damage is not of long patch BER. Furthermore, overexpression of XRCC1 delayed the gap-filling, and presumably increase of XRCC1 pulls PCNA away from gap-filling of NER for BER, consistent with our hypothesis that delay of gap-filling of NER attributes the competition between NER and BER.

Figure 1

(a) Top: schematic illustration of experimental protocol for studying the effect of chemicals on gap-filling during repair of UV-induced DNA lesions. Bottom: colcemid delays the gap-filling. Initial level of the gaps at the time zero after H/A removal was taken as 100%. (b) Similar to (a), AGS cells in log phase were UV irradiated (3 J/m²) and H/A treated for 2 h. After H/A was removed, cells were treated with or without colcemid or 18 μM β-carotene. Cells were harvested for comet assay at the indicated time. (c) Similar to (b), cells were harvested at 6 h after H/A was removed for comet assay.

Figure 2

Top: western analysis of PCNA protein levels in total cell extracts of the CL1-0 cells stably transfected with wild type PCNA (EGFP-wt. PCNA) or truncated PCNA (EGFP-ΔPCNA) expression plasmids or EGFP-vector alone. Bottom: overexpression of PCNA attenuates colcemid-caused delay of gap-filling. CL1-0 cells stably transfected with the indicated plasmids were treated with the procedures illustrated in Figure 1 for studying the effect of PCNA overexpression on colcemid-caused delay of gap-filling during repair of UV-induced DNA lesions.

Figure 3

Flow cytometric analysis of nucleoid size. AGS cells were treated similarly as those described in Figure 1. (a) Cont. for control, that is, cells without treatment. UV and UV → col. for UV alone and UV then colcemid, respectively. Time points: 0–8 h after H/A removal. (b) Similar to (a), yet the cells were transiently transfected with the indicated expression plasmids. (Histogram: x-axis: forward scattering; y-axis: counts.)

Figure 4

The effect of PCNA knockdown on gap-filling during repair of UV-induced DNA lesions. Similar to the procedures described in Figure 1, yet colcemid was excluded to avoid “overdelay.” AGS cells were stably transfected with pLKO-1 shPCNA to knock down PCNA (closed circles) or control vector (open circles).

Figure 5

The effect of Lig-1 overexpression on colcemid-caused delay of gap-filling during repair of UV-induced DNA lesions. (a) Top: western analysis of Lig-1 protein in CL1-0 cells stably transfected with expression plasmid of Lig-1 (RFP-Lig-1) or empty vector. Bottom: similar to Figure 1, yet the CL1-0 cells stably transfected with expression plasmid of Lig-1 (filled circle) or empty vector (open circle) were used. (b) Flow cytometric analysis of nucleoid size. Dose dependence. Cells were transiently transfected with various amounts of PCNA or ligase-I expression plasmids.

Figure 6

The effect of PCNA knockdown on the repair of oxidative DNA damage. (a) Comet assay. AGS cells transfected with pLKO-1 shPCNA to knock down PCNA or control vector were treated with 20 μM H₂O₂ for indicated periods of time before being harvested for comet-Fpg/Endo III assay to monitor the levels of H₂O₂-induced DNA lesions in cells. Bottom panel: western analysis of PCNA protein with levels of β-actin protein as a loading control. (b) Immunostaining analysis and (c) ELISA. Cells transfected with pLKO-1 shPCNA or control vector were treated with 1 mM H₂O₂ for 1 and 4 h before being harvested for immunofluorescence and ELISA assays of 8-oxo-dG. Typical images were shown. N for control, that is, cells without treatment. ∗∗ for p < 0.005.

Figure 7

The effect of Fen-1 knockdown. (a) On repair of oxidative DNA damage. Top: western analysis of Fen-1 protein in AGS cells transfected with pLKO-1 sh Fen-1 to knock down Fen-1 or control vector; levels of PCNA protein were used as a loading control. Bottom: comet assay. AGS cells were treated with 20 μM H₂O₂ for indicated periods of time before being harvested for comet-Fpg/Endo III assay. (b) On DNA replication. S-phase cells were marked.

Figure 8

The effect of XRCC1 overexpressed on the gap-filling during repair of UV-induced DNA lesions. ((a) and (b)) Similar to the procedures described in Figures 1 and 3, yet colcemid was excluded. (a) AGS cells were transiently transfected with pCMV XRCC1 plasmids (open circle) or not (closed circle). (b) Cont. for control, that is, cells without treatment. AGS cells were transfected with wt. PCNA or pCMV-XRCC1 or empty vector. Time points: 0–8 h after H/A removal. Histograms of both panels: x-axis: forward scattering (relevant to particle size); y-axis: cell counts. (c) Similar to procedures described in Figure 1, colcemid was present, and AGS cells transfected with empty vector or plasmid EGF-PCNA with or without pCMV XRCC1 were used.

Figure 9

PCNA is important to the gap-filling during the repair of UV-induced DNA injuries and is also essential to the repair of oxidative DNA lesions. When oxidative DNA damage increases or XRCC1 overexpresses, more PCNA is recruited to BER, which may leave more gaps unfilled.