Back

Click-chemistry-aided quantitation and sequencing of oxidized guanines and apurinic sites uncovers their transcription-linked strand bias in human cells

Takhaveev, V.; Püllen, N. J. L.; Singh, N. K.; Huber, S. M.; Schauer, S.; Gahlon, H. L.; Poetsch, A. R.; Sturla, S. J.

2024-07-23 genomics
10.1101/2024.07.21.604463 bioRxiv
Show abstract

DNA modifications drive aging, neurodegeneration, carcinogenesis, and chemotherapy drug action. To understand the functional genomic roles of DNA modifications, it is critical to accurately map their diverse chemical forms with single-nucleotide precision in complex genomes, but it remains challenging. Click-code-seq is a click-chemistry-aided single-nucleotide-resolution strategy for guanine-oxidation mapping, used in yeast DNA but having poor applicability to human genomes. Here, we upgraded click-code-seq to enable its first application for sequencing DNA oxidation and depurination in human genomes. For this, we developed a companion fluorescence assay, click-fluoro-quant, to rapidly quantify different common DNA modifications, and devised novel adapters to minimize false modification detection and assess modification frequency in cell populations. We uncovered that endogenous DNA oxidation in a human cell line has a highly similar pattern to cancer mutational signatures associated with reactive oxygen species. We established that the DNA-alkylating chemotherapy drug irofulven preferentially induces depurination in ApA dimers and promoter regions. Intriguingly, we revealed that oxidized guanines and apurinic sites, both irofulven-induced and endogenous, are depleted in gene transcribed strands, and the strand bias widens with increasing gene expression. This work substantially advances click-code-seq for deciphering the impacts of key modifications in human DNA on cellular physiology and toxicological responses.

Published in Nature Chemical Biology (predicted rank #29) · training set

Matching journals

The top 1 journal accounts for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.