Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2017 Jun 19;30(6):1317-1326.
doi: 10.1021/acs.chemrestox.7b00057. Epub 2017 May 22.

Reactivity and Cross-Linking of 5'-Terminal Abasic Sites Within DNA

Affiliations
Free PMC article

Reactivity and Cross-Linking of 5'-Terminal Abasic Sites Within DNA

Suzanne J Admiraal et al. Chem Res Toxicol. .
Free PMC article

Abstract

Nicking of the DNA strand immediately upstream of an internal abasic (AP) site produces 5'-terminal abasic (dRp) DNA. Both the intact and the nicked abasic species are reactive intermediates along the DNA base excision repair (BER) pathway and can be derailed by side reactions. Aberrant accumulation of the 5'-terminal abasic intermediate has been proposed to lead to cell death, so we explored its reactivity and compared it to the reactivity of the better-characterized internal abasic intermediate. We find that the 5'-terminal abasic group cross-links with the exocyclic amine of a nucleotide on the opposing strand to form an interstrand DNA-DNA cross-link (ICL). This cross-linking reaction has the same kinetic constants and follows the same pH dependence as the corresponding cross-linking reaction of intact abasic DNA, despite the changes in charge and flexibility engendered by the nick. However, the ICL that traps nicked abasic DNA has a shorter lifetime at physiological pH than the otherwise analogous ICL of intact abasic DNA due to the reversibility of the cross-linking reaction coupled with faster breakdown of the 5'-terminal abasic species via β-elimination. Unlike internal abasic DNA, 5'-terminal abasic DNA can also react with exocyclic amines of unpaired nucleotides at the 3'-end of the nick, thereby bridging the nick by connecting DNA strands of the same orientation. The discovery and characterization of cross-links between 5'-terminal abasic sites and exocyclic amines of both opposing and adjacent nucleotides add to our knowledge of DNA damage with the potential to disrupt DNA transactions.

Conflict of interest statement

Notes

The authors declare no competing financial interest.

Figures

Figure 1
Figure 1
Base excision repair pathway for DNA. Internal abasic DNA and 5′-terminal abasic DNA intermediates and their reactive aldehyde forms are boxed. The numbered reactions correspond to the following enzyme activities: 1, DNA glycosylase; 2, AP endonuclease; 3, dRp lyase; 4, DNA polymerase; 5, DNA ligase. Internal abasic DNA and 5′-terminal abasic DNA are also known as apurinic/apyrimidinic (AP) DNA and 5′-terminal deoxyribose phosphate (dRp) DNA, respectively.
Figure 2
Figure 2
Structural representations of internal and 5′-terminal abasic substrates used in this study. The DNA oligonucleotides annealed to construct each substrate are shown above or below the structures, and complete sequences are listed in Table 1. Abbreviations: ab, abasic site; P, 2-aminopurine; X = P or A; *, fluorescein label linked to the 3′-terminus of DNA via a phosphorothioate group.
Figure 3
Figure 3
Internal and 5′-terminal abasic sites both cross-link with an opposing P. A. Slowly migrating cross-linked DNA forms in reactions of 2P (left) and 5P (right). Reactions contained 0.25 μM DNA and were carried out at 37 °C and pH 6.47. Samples were reduced with NaBH4 prior to separation on the same 15% denaturing polyacrylamide gel (intervening lanes have been omitted for clarity). The 5′-P 17mer that is visible at the bottom of all lanes arises from nonenzymatic β-elimination of the abasic sites prior to or during the reaction time course. The inset for the 5P reaction (right) is the gel at lower contrast, showing that the 5′-P 17mer is resolved from the 5′-terminal abasic strand. B. Fraction of 5′-P 17mer, the product of β-elimination, at time points in A, in duplicate reactions, and in reactions of related abasic substrates at pH 6.47 is shown. Some β-elimination occurs during preparation of the abasic DNA strands, so at time zero reactions of the internal abasic DNA substrates contain 3% 5′-P 17mer and reactions of the 5′-terminal abasic DNA substrates contain 14% 5′-P 17mer (see Experimental Section). Curves are linear fits to the initial rate data for the internal abasic DNA substrates (squares) and exponential fits to the data for the 5′-terminal abasic DNA substrates (circles). C. Fraction of cross-linked DNA at time points in A, in duplicate reactions, and in reactions of related abasic substrates at pH 6.47 is shown. This cross-linked DNA does not form when the opposing strand is absent (1, 3) or contains A in place of P (2A, 5A). D. To account for the loss of the abasic substrates due to β-elimination during the time courses, the ratio of cross-linked to unlinked substrate is shown for the same reactions of 2P and 5P as in B and C (see Experimental Section). End point ratios of 0.036 and 0.033 and kobs of 0.019 min−1 and 0.014 min−1 for the approach to the end point were obtained from exponential fits to the data for 2P and 5P, respectively. All error bars represent one standard deviation from the mean of at least two independent data points.
Figure 4
Figure 4
Effect of pH on reactivity of abasic DNA species. A. The pH dependence of β-elimination for internal (squares) and 5′-terminal (circles) abasic DNA substrates. Non-linear least squares fits to a single ionization event give overlapping curves with pKa values of 7.0, 7.0, and 7.1 for 3, 5A, and 5P, respectively. B. Analogous chemical mechanisms for β-elimination of internal (left) and 5′-terminal (right) abasic DNA substrates. Protonation of the 5′-phosphate of the terminal abasic DNA substrate makes it less susceptible to β-elimination. C. The pH dependence of cross-linking for internal (squares) and 5′-terminal (circles) abasic DNA substrates, in the absence (black) or presence (red) of 5 mM MgCl2. A line with a slope of −1 is drawn through the data. D. General mechanism for imine formation. The step depicting acid-catalyzed protonation of oxygen, which promotes subsequent dehydration, is italicized. Imine formation in the cross-linking reaction involves attack of the exocyclic amine of P on the aldehyde moiety of an internal or 5′-terminal abasic group of DNA.
Figure 5
Figure 5
Effect of DNA context on reactions of 5′-terminal abasic DNA variants at pH 6.47. A. The fraction of 5′-P 17mer, the product of β-elimination, in reactions is shown. Curves are exponential fits to the data for each substrate and give similar kelim values (0.0011–0.0013 min−1; Table S1). B. The ratio of cross-link to unlinked abasic DNA is plotted for each substrate. End point ratios and klink values were obtained from exponential fits to each data set (Table S1). C. Magnified version of Figure 5B, with the scale of the y-axis adjusted and the outlier substrate 4P omitted for clarity. All reactions were performed in duplicate, and error bars represent one standard deviation from the mean value at each time point.
Figure 6
Figure 6
Exocyclic amines of upstream nucleotides cross-link with 5′-terminal abasic DNA. A. Cross-links form in reactions that have an upstream strand with a 3′-overhanging A, C, or G, when templated by RAA (lanes 2–5). For size comparison, the end lanes (S) contain the 35 nt oligonucleotide pre-Fint and a 49 nt oligonucleotide of similar sequence. Reactions contained 0.25 μM DNA and were incubated at 37 °C and pH 6.47 for 12 hrs, at which time samples were reduced with NaBH4 and separated on a 15% denaturing polyacrylamide gel. Only the upper portion of the gel where cross-linked species migrate is shown. B. Structural representation of the 18+18 nt cross-linked species visible in lanes 2, 4, and 5 of Figure 6A, where V is A, C, or G, respectively. The cross-linked species is shown annealed to its template strand. The analogous 25+18 nt cross-link in lane 3 has an upstream 5′-extension of 7 nt. C. The ratio of 18+18 nt cross-link to unlinked abasic DNA is plotted for each substrate. Curves are linear fits to each data set and give klink values of 1.4×10−5 min−1 for 6A, 6.5×10−6 min−1 for 7A, and 6.3×10−6 min−1 for 8A. Error bars represent one standard deviation from the mean of at least two independent data points.
Figure 7
Figure 7
Possible variation in reaction order in BER. Intermediates capable of cross-linking with the 5′-terminal abasic site will form if DNA polymerase activity precedes dRp lyase activity (red arrows).

Similar articles

See all similar articles

Cited by 2 articles

Publication types

Feedback