NEW YORK – Researchers at the University of Washington School of Medicine have developed a new targeted single-molecule sequencing method to profile protein occupancy of DNA and chromatin accessibility with near single-nucleotide resolution.
Named Deaminase-Assisted single-molecule chromatin Fiber sequencing (DAF-seq), the method, described in a BioRxiv preprint published in November, is compatible with amplification and can be used in single cells, opening the door for researchers to study gene regulation heterogeneity between individual cells.
"Historically, there is really no good way of doing a targeted [chromatin accessibility] assay," said Andrew Stergachis, a genomics professor at UW whose lab invented DAF-seq. "The genesis of the method really comes from a very big need in the scientific community as well as the clinical community."
Previously, Stergachis and his collaborators developed Fiber-seq, a method that can map open chromatin regions using methyltransferases to create a stencil on the chromosome. However, since methylation events are not preserved during PCR amplification, Fiber-seq — and similar methyltransferase-based stenciling methods — typically require direct sequencing of the modified DNA, leading to a high sample input requirement.
DAF-seq, on the other hand, marks protein occupancy along a chromatin fiber via selective deamination, a signal that is preserved when the DNA is amplified.
"Once you enable amplification, it enables you to work with much lower cell volumes, it enables you to work with rarer clinical tissues or cell types, it also enables you to more efficiently get deeper coverage at a much cheaper cost," Stergachis noted.
Mechanistically, DAF-seq leverages SsDddA, a variant of the double-stranded cytidine deaminase toxin A (DddA) from the bacterial species Simiaoa sunii, to stencil chromatin protein occupancy in the form of deaminated cytidines.
SsDddA modifies accessible cytidine to uridine, which becomes thymine during DNA amplification. The PCR amplicons can then be analyzed by short-read or long-read DNA sequencing.
"The entire premise of DAF-seq is that we are using the deaminase as a spray paint," Stergachis explained. "We're trying to spray paint chromatin fibers such that where proteins are physically bound, this enzyme is not modifying them, and where physical proteins are not bound, it is modifying them."
To use DAF-seq, researchers first permeabilize the cells, which are then treated with SsDddA. The deamination reaction is halted by adding a detergent, and the DNA is isolated for PCR amplification of the target region, followed by sequencing.
Despite the relatively straightforward concept, Stergachis said his team spent more than five years developing and optimizing the method and tested hundreds of different enzymes before settling on SsDddA.
For a cytidine deaminase to be optimal for chromatin stenciling, he said, the enzyme must have minimal sequence biases, be highly catalytically active, and be highly specific for accessible DNA. Just as important, the enzyme has to be suitable for large-scale production, he added.
"The last thing you want is an assay where you have to spend six months actually making a protein," Stergachis said. "We wanted, for this assay, an enzyme that was robust and that we could make in large quantities, such that it actually could be adopted by the field."
In their study, the UW researchers demonstrated that DAF-seq can be performed on frozen primary human tissues as well as cultured cells.
They also showed that DAF-seq can enrich the target loci by around 200,000-fold, enabling chromatin stenciling with single-molecule and near single-nucleotide precision.
This resolution, in conjunction with high sequencing depth, allows researchers to characterize the architecture of regulatory elements of interest, revealing transcription factor occupancy and co-occupancy, chromatin actuation transition states, and the functional impact of rare variants with a low allele fraction.
They also tested DAF-seq on single cells, or scDAF-seq. For that, they treated four permeabilized GM24385 (HG002) cells with SsDddA and isolated them into individual wells.
They then carried out whole-genome amplification (WGA) on each cell using primary template-directed amplification (PTA), an isothermal amplification method developed by BioSkryb Genomics that claims to capture more than 95 percent of the single-cell genome. The sample is subsequently sequenced using HiFi long reads from Pacific Biosciences.
Overall, the researchers concluded that scDAF-seq can accurately reconstruct a haplotype-phased diploid genome from a single cell at the chromosomal scale and can evaluate 96 percent of each cell’s mappable haploid genome.
By comparison, previous cleavage-based chromatin mapping methods, such as single-cell ATAC-seq, can typically only cover 0.01 percent of a cell's mappable genome, the authors noted.
The single-cell capabilities of DAF-seq also allowed the researchers to gain insights into gene regulation. For instance, they observed that some of the four cells' accessible regulatory elements diverged by as much as 63 percent while still retaining the same cellular identity.
"The power of DAF-seq is that we can do strong enrichment of specific regions [of the chromatin] with amplification using targeted primers," said Winston Timp, a biomedical engineering professor at Johns Hopkins University who was not involved in the study. "It is compatible with PCR, meaning we can use lower inputs and enrich dramatically for regions using the standard amplicon methods, and even going down to the single cell."
Although other methods have been developed for single-cell chromatin analysis, such as single-cell ATAC-seq, these methods currently have lower resolution and "leave you with gaps," Timp noted.
DAF-seq's single-cell capabilities could help researchers obtain more chromatin epigenome information in a physiologically relevant context from individual cells, Timp said, though it remains to be seen how much the experiments cost, an important factor for the method's adoption.
Moving forward, Stergachis said his team will continue to optimize DAF-seq for both targeted and single-cell amplification. He and his coauthors also filed provisional patents pertaining to the technology, though he did not say whether there are any plans to commercialize the method.
Stergachis said his team has already received a lot of interest in DAF-seq from other researchers and is trying to figure out the best ways to disseminate the SsDddA deaminase.
"I'm very excited about this concept that we are getting nearly complete genomes and epigenomes from single cells," he said. "It just really opens up a whole lot of possibilities in terms of the questions, both scientifically and clinically, that we can start to ask and go after."