Being average isn’t all that bad, but it’s far from good. Since its very beginning, chromatin immunoprecipitation (ChIP) has been plagued with the problem of averages: chromatin profiles based on ChIP data are an average of the many thousands of cells required to perform this experiment.
This gives us a rough overview of the chromatin landscape, but no idea of the amount of variation in cell populations. However, a new technique reported by Bradley Bernstein’s laboratory has now enabled ChIP to finally catch up with RNA seq and DNA methylation analysis and provide chromatin data at single cell resolution.
The high levels of background noise encountered with small sample sizes in ChIP has prevented its use in rare cell populations and at the single cell level. Taking on this challenge, Bernstein and his team reasoned that if chromatin from single cells were labelled before immunoprecipitation, then chromatin from multiple cells could be combined, thus avoiding this background noise.
To achieve this, the team turned to drop-based microfluidics, in which individual cells are encapsulated in micron-sized aqueous drops that can be split, combined, detected or sorted at a rate of several hundred Hz in a microfluidics device.
Here’s a breakdown of their system:
- A suspension of cells is mixed with detergent and micrococcal nuclease to fragment chromatin just milliseconds before being encapsulated into drops containing one individual cell each.
- In parallel, a barcode library containing a pool of drops is engineered in which each drop contains a distinct oligonucleotide adaptor.
- Each barcode-containing drop is then fused to a single nucleosome-containing drop in the microfluidics device. Following ligation, the chromatin content of each individual cell is labeled with a unique barcode.
- Chromatin from many drops is combined and immunoprecipitated with an antibody of interest. After DNA sequencing, chromatin profiles can be established from single cells by partitioning sequencing reads by their barcode sequence.
Using this aptly-named “Drop-ChIP” technique, Bernstein’s team investigated H3K4 dimethylation, a marker of promoters and enhancers, in mouse embryonic stem cells (ESCs). Their analysis identified three distinct subpopulations of ESCs, differing in terms of H3K4 methylation at pluripotency and differentiation genes.
The coupling of this new technique with single cell RNA seq and methylation data may enable a deeper understanding of cell populations and the epigenetic control of gene expression.
For all the details, “drop” by Nature Biotechnology, October 2015.