Stalling of Eukaryotic Translesion DNA Polymerases at DNA-Protein Cross-Links

Abstract

DNA-protein cross-links (DPCs) are extremely bulky adducts that interfere with replication. In human cells, they are processed by SPRTN, a protease activated by DNA polymerases stuck at DPCs. We have recently proposed the mechanism of the interaction of DNA polymerases with DPCs, involving a clash of protein surfaces followed by the distortion of the cross-linked protein. Here, we used a model DPC, located in the single-stranded template, the template strand of double-stranded DNA, or the displaced strand, to study the eukaryotic translesion DNA polymerases ζ (POLζ), ι (POLι) and η (POLη). POLι demonstrated poor synthesis on the DPC-containing substrates. POLζ and POLη paused at sites dictated by the footprints of the polymerase and the cross-linked protein. Beyond that, POLζ was able to elongate the primer to the cross-link site when a DPC was in the template. Surprisingly, POLη was not only able to reach the cross-link site but also incorporated 1–2 nucleotides past it, which makes POLη the most efficient DNA polymerase on DPC-containing substrates. However, a DPC in the displaced strand was an insurmountable obstacle for all polymerases, which stalled several nucleotides before the cross-link site. Overall, the behavior of translesion polymerases agrees with the model of protein clash and distortion described above.

Nucleophilic Thiol Proteins Bind Covalently to Abasic Sites in DNA

“…This number seems unusually high, 2 orders of magnitude higher than any low M r DNA adducts, e.g., 8-oxo-7,8-dihydro-dGuo at 0.1–1/10 6 bases and even AP sites (values varying from 0.31 to 9/10 6 bases). − With low M r thiols, the questions of polymerase retardation and miscoding remain to be addressed. Other questions are how a DNA polymerase can bypass a very large entity, i.e., a DNA-protein cross-link, ,,− , as well as what determines the binding to the AP sites: thiol p K a , nucleophilicity, DNA affinity, concentration, or a combination of all of these. Further studies are in progress.…”

“…35−37 With low M r thiols, the questions of polymerase retardation and miscoding remain to be addressed. Other questions are how a DNA polymerase can bypass a very large entity, i.e., a DNA-protein cross-link, 13,14,[19][20][21]38 as well as what determines the binding to the AP sites: thiol pK a , nucleophilicity, DNA affinity, concentration, or a combination of all of these. Further studies are in progress.…”

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

“…It was almost always comparable with, and often even superior to, the displacement of the strand containing an AP site. These observations strongly contrast the blocking ability of full-length proteins cross-linked to a downstream strand [ 91 , 92 ]. Presumably, cross-linked DNA-binding proteins strongly stabilize the duplex through multiple contacts with DNA, which are lost upon proteolysis.…”

“…APPXL-containing substrates were assembled by annealing the 40-mer template strand with the 11-mer [ 32 P]-labeled primer (run_pri) and, if necessary, with the downstream strand (28down), as described in [ 91 , 92 ]. The cross-link was placed in either the template or the downstream strand.…”

Mechanisms and Regulation of DNA-Protein Crosslink Repair During DNA Replication by SPRTN Protease

“…Induction of topoisomerase-covalent crosslinks (TOP-ccs) and methyltransferase DPCs in Escherichia coli caused replication fork stalling ( Hong and Kreuzer, 2000 ; Pohlhaus and Kreuzer, 2005 ; Kuo et al, 2007 ). Several studies show that TLS polymerases are stalled by DPCs in vitro ( Chválová et al, 2007 ; Nakano et al, 2013 ; Yeo et al, 2014 ; Yudkina et al, 2022 ). TLS polymerases have more flexible active sites than canonical replicative polymerases, allowing for nascent DNA synthesis opposite a DNA lesion and subsequent bypass of the lesion by the replication fork ( Sale et al, 2012 ).…”