NEW YORK – Using a new single-cell amplification and sequencing strategy aimed at characterizing a cell's "damagenome," a Baylor College of Medicine-led team has identified parts of the genome that appear prone to spontaneous DNA damage in human brain cells — an analysis that highlighted "high-damage" genes that may contribute to disease.
"These high-damage genes essentially correspond to the Achilles' heels of our genome since DNA damage can not only cause somatic mutations, but frequent damage could also induce epigenetic changes, resulting [in] gene expression alterations and variations," senior author Chenghang Zong, a molecular and human genetics researcher at BCM, said in an email.
As they explained in a paper published in Science Advances on Friday, he and his colleagues came up with a "linear copy and split-based whole-genome amplification," or LCS-WGA, method for taking advantage of the base misincorporation events that can occur when whole-genome amplification is applied to DNA with spontaneous damage.
With a linear copying pre-amplification step using a proofreading-free DNA polymerase activity, followed by a split amplification to independently amplify and sequence each sample three times, the approach appeared to weed out most of the false-positive amplification errors that did not stem from DNA damage.
"Because we are confident that we have filtered out almost all amplification errors," Zong said, "we know these large number of variants must have a different source of origins, which we later confirmed … are associated with spontaneous DNA damage."
The authors noted that the approach is focused on the types of DNA damage that boost base misincorporation rates, such as DNA damage involving oxidized cytosine or 8-oxoguanine, but may not pick up changes that do not enhance misincorporation.
After validating the method in cells from a normal breast cell line grown in normal conditions or exposed to hydrogen peroxide, the team used LCS-WGA to search for damage-related single-nucleotide variants, or "damSNVs," in dozens of individual post-mortem neurons from the prefrontal cortex or hippocampus of young, middle-aged, or elderly individuals.
In those experiments, the researchers saw hints that DNA damage may be slightly higher in cortical and hippocampus neurons from the middle-aged individuals compared to their younger or older counterparts, though that damSNV difference was not statistically significant in the individuals considered.
The results also suggested that spontaneous DNA damage does not turn up uniformly across the genome. Instead, parts of the genome linked to certain genes or functional roles seemed to have distinct DNA damage profiles in analyses focused on the damagenome in relation to three-dimensional genomic features.
For example, the investigators described what they called "high-damage" genes that appear especially likely to become marred — a set that overlapped with genes that have been shown to have altered expression in brain samples from individuals with Alzheimer's disease or autism spectrum disorder in the previous single-cell RNA sequencing studies.
Their own LCS-WGA experiments suggested that DNA damage was significantly higher in individual neurons from Alzheimer's-related samples, suggesting that the damagenome may provide clues to the genes and functional pathways contributing to disease in different cell types.
"Overall, our result sheds new light on vulnerable genes and vulnerable functional pathways in the human genome and their potential roles in complex human diseases," Zong said.