NEW YORK (GenomeWeb) – Scientists from the Broad Institute have used antibody-based detection coupled with single-molecule sequencing to begin decoding a dual-marking system in modified histones that signals for a gene to be activated or repressed.
Using their method, the researchers have been able to determine both the genomic locations and modification states of individual nucleosomes. Early results, published today in Science, suggest differentiated cells exhibit different patterns of "bivalent" markings than embryonic cells.
"This single-molecule technology has the potential to address fundamental questions in chromatin biology and epigenetic regulation," the authors, led by Bradley Bernstein of Massachusetts General Hospital and Harvard Medical School, wrote. "We have taken a critical step toward defining the nature and importance of combinatorial chromatin modifications."
The assay captures nucleosomes from cells and attaches them to fluorescent biotinylated oligonucleotide adaptors, which are in turn fixed to streptavidin-coated slides. An imaging step notes the nucleosomes' positions on the slides and the fluorophore is cleaved. Then, fluorescent antibodies targeting a particular histone modification are unleashed on the nucleosomes and repeated binding and dissociation events are logged and used to score modified histones.
As a proof of concept, the scientists used nucleosomes from HEK 293 cells and antibodies targeting histone H3 lysine 9 acetylation, lysine 4 trimethylation (H3K4me3), lysine 27 trimethylation (H3K27me3), lysine 27 dimethylation, and lysine 27 acetylation.
While H3K4me3 is thought to be an activating histone modification and H3K27me3 is thought to be repressive, developmental gene promoters in embryonic stem cells (ESCs) have been shown to be marked by both. "This bivalent chromatin state has been suggested to poise these genes for alternate fates," the authors said, noting that the idea is somewhat controversial. While chromatin immunoprecipitation and IP mass spectrometry can reveal both histone modifications, the two methods cannot definitively identify if a nucleosome exhibits such a bivalent status.
But using their antibody assay, the researchers were able to determine if nucleosomes from pluripotent ESCs, intermediately differentiated embryoid cells, and differentiated lung fibroblasts carried both markings.
Looking at different combinations of the several histone modifications revealed that there were "considerable differences between cell types," the authors wrote. They found that 0.5 percent of ESC nucleosomes were bivalent for H3K4me3 and H3K27me3, much higher than expected, and the more differentiated cells exhibited bivalent markings at a rate less than the expected background.
The authors also used single-molecule DNA sequencing to read each molecule to determine both the modification state and genomic location of each nucleosome. Of the nucleosomes that scored positively for H3k27me3, 45 percent aligned to genomic regions known to be enriched for that modification, comparable to results from ChIP-seq experiments.
The study represents the early results from a collaboration between Bernstein and single-molecule sequencing firm SeqLL, a Broad spokesperson told GenomeWeb in an email. SeqLL President and CEO Daniel Jones was a co-author on the Science paper. As the collaboration progresses, the partners could "help uncover new rules underlying cell fate," the spokesperson said.