Recognition of a Clickable Abasic Site Analog by DNA Polymerases and DNA Repair Enzymes

Error-Prone DNA Synthesis on Click-Ligated Templates

Click ligation is a technology of joining DNA fragments based on azide-alkyne cycloaddition. In the most common variant, click ligation introduces a 4-methyl-1,2,3-triazole (trz) group instead of the phosphodiester bond between two adjacent nucleosides. While this linkage is believed to be biocompatible, little is known about the possibility of its recognition by DNA repair systems or its potential for DNA polymerase stalling and miscoding. Here we report that trz linkage is resistant to several human and bacterial endonucleases involved in DNA repair. At the same time, it strongly blocks some DNA polymerases (Pfu, DNA polymerase β) while allowing bypass by others (phage RB69 polymerase, Klenow fragment). All polymerases, except for DNA polymerase β, showed high frequency of misinsertion at the trz site, incorporating dAMP instead of the next complementary nucleotide. Thus, click ligation can be expected to produce a large amount of errors if used in custom gene synthesis.

Repair and DNA Polymerase Bypass of Clickable Pyrimidine Nucleotides

Clickable nucleosides, most often 5-ethynyl-2'-deoxyuridine (EtU), are widely used in studies of DNA replication in living cells and in DNA functionalization for bionanotechology applications. Although clickable dNTPs are easily incorporated by DNA polymerases into the growing chain, afterwards they might become targets for DNA repair systems or interfere with faithful nucleotide insertion. Little is known about the possibility and mechanisms of these post-synthetic events. Here, we investigated the repair and (mis)coding properties of EtU and two bulkier clickable pyrimidine nucleosides, 5-(octa-1,7-diyn-1-yl)-U (C8-AlkU) and 5-(octa-1,7-diyn-1-yl)-C (C8-AlkC). In vitro, EtU and C8-AlkU, but not C8-AlkC, were excised by SMUG1 and MBD4, two DNA glycosylases from the base excision repair pathway. However, when placed into a plasmid encoding a fluorescent reporter inactivated by repair in human cells, EtU and C8-AlkU persisted for much longer than uracil or its poorly repairable phosphorothioate-flanked derivative. DNA polymerases from four different structural families preferentially bypassed EtU, C8-AlkU and C8-AlkC in an error-free manner, but a certain degree of misincorporation was also observed, especially evident for DNA polymerase β. Overall, clickable pyrimidine nucleotides could undergo repair and be a source of mutations, but the frequency of such events in the cell is unlikely to be considerable.

Probing the Conformational Restraints of DNA Damage Recognition with β-L-Nucleotides

The DNA building blocks 2'-deoxynucleotides are enantiomeric, with their natural β-D-configuration dictated by the sugar moiety. Their synthetic β-L-enantiomers (βLdNs) can be used to obtain L-DNA, which, when fully substituted, is resistant to nucleases and is finding use in many biosensing and nanotechnology applications. However, much less is known about the enzymatic recognition and processing of individual βLdNs embedded in D-DNA. Here, we address the template properties of βLdNs for several DNA polymerases and the ability of base excision repair enzymes to remove these modifications from DNA. The Klenow fragment was fully blocked by βLdNs, whereas DNA polymerase κ bypassed them in an error-free manner. Phage RB69 DNA polymerase and DNA polymerase β treated βLdNs as non-instructive but the latter enzyme shifted towards error-free incorporation on a gapped DNA substrate. DNA glycosylases and AP endonucleases did not process βLdNs. DNA glycosylases sensitive to the base opposite their cognate lesions also did not recognize βLdNs as a correct pairing partner. Nevertheless, when placed in a reporter plasmid, pyrimidine βLdNs were resistant to repair in human cells, whereas purine βLdNs appear to be partly repaired. Overall, βLdNs are unique modifications that are mostly non-instructive but have dual non-instructive/instructive properties in special cases.

Aberrant Repair of 8-Oxoguanine in Short DNA Bulges

The presence of DNA damage can increase the likelihood of DNA replication errors and promote mutations. In particular, pauses of DNA polymerase at the site of damage can lead to polymerase slippage and the formation of 1-2-nucleotide bulges. Repair of such structures using an undamaged DNA template leads to small deletions. One of the most abundant oxidative DNA lesions, 8-oxoguanine (oxoG), was shown to induce small deletions, but the mechanism of this phenomenon is currently unknown. We studied the aberrant repair of oxoG located in one- and two-nucleotide bulges by the Escherichia coli and human base excision repair systems. Our results indicate that the repair in such substrates can serve as a mechanism for fixing small deletions in bacteria but not in humans.

Bypass of Abasic Site-Peptide Cross-Links by Human Repair and Translesion DNA Polymerases

DNA-protein cross-links remain the least-studied type of DNA damage. Recently, their repair was shown to involve proteolysis; however, the fate of the peptide remnant attached to DNA is unclear. Particularly, peptide cross-links could interfere with DNA polymerases. Apurinuic/apyrimidinic (AP) sites, abundant and spontaneously arising DNA lesions, readily form cross-links with proteins. Their degradation products (AP site-peptide cross-links, APPXLs) are non-instructive and should be even more problematic for polymerases. Here, we address the ability of human DNA polymerases involved in DNA repair and translesion synthesis (POLβ, POLλ, POLη, POLκ and PrimPOL) to carry out synthesis on templates containing AP sites cross-linked to the N-terminus of a 10-mer peptide (APPXL-I) or to an internal lysine of a 23-mer peptide (APPXL-Y). Generally, APPXLs strongly blocked processive DNA synthesis. The blocking properties of APPXL-I were comparable with those of an AP site, while APPXL-Y constituted a much stronger obstruction. POLη and POLκ demonstrated the highest bypass ability. DNA polymerases mostly used dNTP-stabilized template misalignment to incorporate nucleotides when encountering an APPXL. We conclude that APPXLs are likely highly cytotoxic and mutagenic intermediates of AP site-protein cross-link repair and must be quickly eliminated before replication.