Structural Insights into the Processing of Nucleobase-Modified Nucleotides by DNA Polymerases

The DNA polymerase-catalyzed incorporation of modified nucleotides is employed in many biological technologies of prime importance, such as next-generation sequencing, nucleic acid-based diagnostics, transcription analysis, and aptamer selection by systematic enrichment of ligands by exponential amplification (SELEX). Recent studies have shown that 2'-deoxynucleoside triphosphates (dNTPs) that are functionalized with modifications at the nucleobase such as dyes, affinity tags, spin and redox labels, or even oligonucleotides are substrates for DNA polymerases, even if modifications of high steric demand are used. The position at which the modification is introduced in the nucleotide has been identified as crucial for retaining substrate activity for DNA polymerases. Modifications are usually attached at the C5 position of pyrimidines and the C7 position of 7-deazapurines. Furthermore, it has been shown that the nature of the modification may impact the efficiency of incorporation of a modified nucleotide into the nascent DNA strand by a DNA polymerase. This Account places functional data obtained in studies of the incorporation of modified nucleotides by DNA polymerases in the context of recently obtained structural data. Crystal structure analysis of a Thermus aquaticus (Taq) DNA polymerase variant (namely, KlenTaq DNA polymerase) in ternary complex with primer-template DNA and several modified nucleotides provided the first structural insights into how nucleobase-modified triphosphates are tolerated. We found that bulky modifications are processed by KlenTaq DNA polymerase as a result of cavities in the protein that enable the modification to extend outside the active site. In addition, we found that the enzyme is able to adapt to different modifications in a flexible manner and adopts different amino acid side-chain conformations at the active site depending on the nature of the nucleotide modification. Different "strategies" (i.e., hydrogen bonding, cation-π interactions) enable the protein to stabilize the respective protein-substrate complex without significantly changing the overall structure of the complex. Interestingly, it was also discovered that a modified nucleotide may be more efficiently processed by KlenTaq DNA polymerase when the 3'-primer terminus is also a modified nucleotide instead of a nonmodified natural one. Indeed, the modifications of two modified nucleotides at adjacent positions can interact with each other (i.e., by π-π interactions) and thereby stabilize the enzyme-substrate complex, resulting in more efficient transformation. Several studies have indicated that archeal DNA polymerases belonging to sequence family B are better suited for the incorporation of nucleobase-modified nucleotides than enzymes from family A. However, significantly less structural data are available for family B DNA polymerases. In order to gain insights into the preference for modified substrates by members of family B, we succeeded in obtaining binary structures of 9°N and KOD DNA polymerases bound to primer-template DNA. We found that the major groove of the archeal family B DNA polymerases is better accessible than in family A DNA polymerases. This might explain the observed superiority of family B DNA polymerases in polymerizing nucleotides that bear bulky modifications located in the major groove, such as modification at C5 of pyrimidines and C7 of 7-deazapurines. Overall, this Account summarizes our recent findings providing structural insight into the mechanism by which modified nucleotides are processed by DNA polymerases. It provides guidelines for the design of modified nucleotides, thus supporting future efforts based on the acceptance of modified nucleotides by DNA polymerases.