Uracil in duplex DNA is a substrate for the nucleotide incision repair pathway in human cells

Abstract

Spontaneous hydrolytic deamination of cytosine to uracil (U) in DNA is a constant source of genome instability in cells. This mutagenic process is greatly enhanced at high temperatures and in single-stranded DNA. If not repaired, these uracil residues give rise to C → T transitions, which are the most common spontaneous mutations occurring in living organisms and are frequently found in human tumors. In the majority of species, uracil residues are removed from DNA by specific uracil-DNA glycosylases in the base excision repair pathway. Alternatively, in certain archaeal organisms, uracil residues are eliminated by apurinic/apyrimidinic (AP) endonucleases in the nucleotide incision repair pathway. Here, we characterized the substrate specificity of the major human AP endonuclease 1, APE1, toward U in duplex DNA. APE1 cleaves oligonucleotide duplexes containing a single U⋅G base pair; this activity depends strongly on the sequence context and the base opposite to U. The apparent kinetic parameters of the reactions show that APE1 has high affinity for DNA containing U but cleaves the DNA duplex at an extremely low rate. MALDI-TOF MS analysis of the reaction products demonstrated that APE1-catalyzed cleavage of a U⋅G duplex generates the expected DNA fragments containing a 5'-terminal deoxyuridine monophosphate. The fact that U in duplex DNA is recognized and cleaved by APE1 in vitro suggests that this property of the exonuclease III family of AP endonucleases is remarkably conserved from Archaea to humans. We propose that nucleotide incision repair may act as a backup pathway to base excision repair to remove uracils arising from cytosine deamination.

Keywords: alternative excision repair; evolution; spontaneous DNA base deamination.

Fig. 1.

APE1-catalyzed deoxyuridine endonuclease activity. (A) Time- and condition-dependent cleavage of a DNA duplex containing a single uracil residue by APE1. 10 nM 3′-[³²P]–labeled 30-mer U⋅G duplex (dU-RT sequence context) was incubated for 15–180 min at 37 °C with 10 nM APE1 under the NIR and BER conditions. Lane 1, control U⋅G; lane 2, as 1 but incubated with 40 nM hTDG and 10 nM APE1 under the BER+Mg²⁺ reaction conditions. Lanes 3–14, as 1 but incubated with APE1 under the BER+Mg²⁺ or NIR conditions. (B) Action of various NIR endonucleases and APE1-D308A mutant toward the U⋅G duplex. 10 nM 3′-[³²P]–labeled 30-mer U⋅G, and αdA⋅T oligonucleotide duplexes (dU-RT sequence context) were incubated with either 10 nM wild-type (WT) APE1 or 10 nM APE1-D308A under the NIR conditions for 2 h or with a limited amount of Nfo and Apn1 in their respective reaction buffers for 30 min at 37 °C. The arrows denote the position of the noncleaved 31-mer substrate, the position of the 21-mer APE1-NIR product, and the position of the 20-mer DNA glycosylase product. For details, see Materials and Methods.

Fig. 2.

MALDI-TOF MS analysis of the reaction products generated by the incubation of a 17-mer oligonucleotide duplex containing a single dU residue with APE1 (A) and with TDG and APE1 (B). Typically, 10 pmol of the 17-mer U⋅G duplex (dU-RT17 sequence context) was incubated with either 10 nM APE1 under NIR conditions at 37 °C for 17 h or with 40 nM hTDG at 37 °C for 30 min and then with 10 nM APE1 at 37 °C for 30 min under the BER+Mg²⁺ reaction conditions. For details, see Materials and Methods.

Fig. 3.

In vitro reconstitution of APE1-catalyzed dU-endonuclease activities. (A) Action of APE1 on a sodium bisulfite-treated DNA duplex. NaHSO₃-treated and 3′-[³²P]–labeled 31-mer C⋅G duplex was incubated with either hTDG/APE1 under the BER+Mg²⁺ conditions or with 10 nM APE1 under the NIR conditions. Lane 1, control NaHSO₃-treated C⋅G duplex; lane 2, as 1 but hTDG/APE1; lanes 3–6, as 1 but APE1 alone for 0.5, 1, 2, and 3 h, respectively. (B) In vitro reconstitution of the NIR pathway for uracil residues. 10 nM nonlabeled 40-mer U⋅G oligonucleotide duplex (DL10 sequence context) was incubated for 1 h at 37 °C in the presence of DNA repair proteins, nonlabeled dNTPs, and [α-³²P]dCTP. Lanes 1–11, U⋅G incubated with repair proteins either under the BER or the NIR conditions; lane 12, 40-mer size marker. All incubations were performed in the presence of 1 unit of uracil-DNA glycosylase inhibitor. For details, see Materials and Methods.

Fig. 4.

Characterization of DNA substrate specificity of Mth212. 10 nM 3′-[³²P]–labeled 30-mer oligonucleotide duplexes (dU-RT sequence context) THF⋅T, U⋅G, αdA⋅T, and 5ohC⋅G were incubated either with Mth212 for 15 min at 55 °C or with 1 nM Nfo for 5 min at 37 °C under their respective reaction conditions. (A) Action of Mth212 on AP sites and uracil residues. Lane 1, THF⋅T incubated for 2.5 h at 55 °C; lane 2, nontreated THF⋅T; lanes 3–7, as 2 but 0.5, 2, 4, 10, and 20 nM Mth212; lane 8, 1 nM Nfo; lane 9, U⋅G incubated for 2.5 h at 55 °C; lane 10, nontreated U⋅G; lanes 11–15, as 10 but 0.5, 2, 4, 10, and 20 nM Mth212; lane 16, 1 nM Nfo. (B) Action of Mth212 on αdA and 5ohC residues. Lane 1, nontreated αdA⋅T; lanes 2–6, as 1 but 2, 4, 10, 20, and 100 nM Mth212; lane 7, 1 nM Nfo; lane 8, 5ohC⋅G incubated for 2.5 h at 55 °C; lane 9, nontreated 5ohC⋅G; lanes 10–14, as 9 but 2, 4, 10, 20, and 100 nM Mth212; lane 15, 1 nM Nfo. The arrow “s” denotes the position of the noncleaved 31-mer substrate, and the arrow “p” denotes the position of the 21-mer cleavage product. For details, see Materials and Methods.