NEW YORK (GenomeWeb) – Researchers at the Georgia Institute of Technology and two other groups have independently developed new techniques for mapping ribonucleotides that have become incorporated into genomic DNA.
The Georgia Tech group's method, called ribose-seq, is described in a paper in Nature Methods today. Along with the two other techniques, published elsewhere today, it could help researchers understand how ribonucleotides change the structure and function of DNA and chromatin, and whether ribonucleotide signatures could serve as biomarkers for certain human diseases.
It has been known for a while that DNA occasionally incorporates ribonucleotides as building blocks instead of the usual deoxyribonucleotides. The two differ by a single additional 2'-hydroxyl group that is present in the ribonucleotides, but there has been no method for mapping ribonucleotides on a genome-wide basis.
Ribonucleotides can get incorporated by several DNA polymerases during DNA replication or DNA repair, and they can also arise when DNA gets damaged. Cellular mechanisms exist that remove ribonucleotides from DNA because their highly reactive hydroxyl group can make the DNA unstable. Mutations in one such ribonucleotide-removing enzyme, called RNAse H2, are associated with a neurological disorder in humans called Aicardi-Goutieres syndrome.
For their new ribose-seq method, the Georgia Tech researchers first fragment genomic DNA and add a molecular barcode with a sequencing adaptor. They then treat the sample with alkali, which denatures the DNA and cleaves it at positions where a ribonucleotide has been incorporated. The cleavage products have a 2',3'-cyclic phosphate or a 2'-phosphate end, which, with the help of a tRNA ligase from Arabidopsis called AtRNL, connects with the other end of the fragment, resulting in single-stranded DNA circles. After destroying all remaining non-circular DNA using T5 exonuclease, the DNA circles are then PCR-amplified and sequenced. From the sequence reads, the position of the ribonucleotide in the genome can be inferred.
In their paper, the team applied ribose-seq to nuclear and mitochondrial DNA from a strain of budding yeast that is missing the RNase H2 enzyme. Overall, they found that C-ribonucleotides (rCMP) and rGMPs were more frequently incorporated into DNA than rAMP and rUMP, and that the upstream base following the ribonucleotide is most often an A, and least often a G. Ribonucleotides appear to be more frequently incorporated into the leading strand than in the lagging strand of newly-synthesized nuclear DNA. The researchers also discovered that in the mitochondrial genome, ribonucleotides are preferentially incorporated upstream of GC-rich regions.
There were several hotspots for ribonucleotides, all in DNA regions with multiple copies per cell: mitochondrial DNA, ribosomal DNA repeats, and yeast retrotransposons.
Separately from the Georgia Tech group, researchers at the University of Edinburgh in Scotland developed a related method called embedded ribonucleotide sequencing, or emRiboSeq, which they published in Nature today as part of a study of lagging strand replication.
Their method calls for treating genomic DNA first with recombinant RNase H2, which generates a nick on the 5' side of a ribonucleotide in the DNA. They then ligate a sequencing adaptor to the 3'-hydroxyl group of the deoxynucleotide right next to the ribonucleotide and sequence the resulting fragments.
Another method to map the genome-wide distribution of ribonucleotides, called hydrolytic end sequencing or HydEn-seq, was published today in Nature Structural & Molecular Biology by a group led by the National Institute of Environmental Health Sciences researchers.
All three approaches "should allow us to better understand the impact of rNMPs on the structure and function of DNA and chromatin, and specific rNMP signatures may represent new biomarkers for human diseases such as Aicardi-Goutieres syndrome, cancer, and other degenerative disorders," the Georgia Tech researchers concluded.