The arrow indicates the inner cell mass in the blastocyst.
If all life gives you is one lemon, can you still make lemonade? Traditional chromatin immunoprecipitation (ChIP) requires roughly 10 million cells. However, many regions of interest, including brain nuclei, contain far fewer cells. Pooling samples increases variability, and adding additional tissue often results in a heterogeneous population of cells. O'Neill et al. modify standard ChIP protocols to accommodate as few as 50 cells in a recent article in Nature Genetics.
The authors used 'carrier' chromatin to bulk up the small amount of experimental chromatin, supplementing 50–10,000 experimental cells with 50,000,000 Drosophila SL2 cells. They named their technique CChIP to differentiate it from the parent 'native' ChIP (NChIP) technique. The authors chose PCR primer pairs that did not hybridize with Drosophila DNA and used radioactive PCR instead of real-time PCR to quantify histone-bound DNA so that they could verify the quality of the resulting PCR product on agarose gels.
Acetylation of histone 4 (H4) and trimethylation of lysine 4 on histone 3 (H3K4) associate with transcriptional activity, whereas dimethylation of lysine 9 on histone 3 (H3K9) associates with transcriptional silencing. Using either NChIP or CChIP with 10,000,000 or 1,000 embryonic stem cells, respectively, the authors found low levels of H3K9 dimethylation and high levels of H4 acetylation of the Nanog gene, which is active in undifferentiated embryonic stem cells. Conversely, they found low levels of H4 acetylation of Cdx2, which is silent in undifferentiated stem cells. NChIP data generally agreed with CChIP data, and CChIP data did not vary extensively between experiments.
The authors then analyzed blastocysts containing approximately 50 cells. One-third of the blastocyst cells are the embryonic stem cells of the inner cell mass (ICM), and the remainder are differentiated trophoblast cells. The authors pooled 3 ICM samples to yield roughly 50 cells and did CChIP analysis. H4 acetylation was elevated in Nanog relative to Cdx2, and conversely H3K9 dimethylation was elevated in Cdx2 relative to Nanog. In similar CChIP analysis of trophoblast cells, the reverse was seen: H4 acetylation was reduced and H3K9 dimethylation was increased in Nanog relative to Cdx2.
These data suggest that CChIP can be performed with as few as 50 cells. The authors believe that this may be the lower limit of CChIP because analysis of a single ICM sample containing roughly 16 cells was unsuccessful.