Just because two things ultimately do the same thing doesn’t mean they’re equivalent. Take Facebook and MySpace, for example. Researchers from the University of Washington say that imprinting and random allele-specific expression have two very different mechanisms, even though the outcome—only one of the two alleles of a gene is translated—is the same.
H3K9me3 and CTCF binding play a big role in imprinting, so they performed ChIP-chip analysis and found 293 loci that could associate with both. Only eight of these were monoallelically expressed, according to a previous report. And when the researchers analyzed lymphoblastoid cell lines (LCLs), only one of these—the gene IGF2BP1—was expressed from one allele, and this was random. Already, the imprinting mechanism wasn’t holding up well. Here’s more info about the random process:
- “DNA methylation does not always lead to inhibition of CTCF binding. Methylation needs to be precisely targeted, and the motif ‘polarity’ or orientation is of major importance.” Says lead author Brandon Thomas
- Both H3K4me3 (active chromatin mark) and H3K27me3 (silent chromatin mark) coexisted on both the expressed and unexpressed alleles, so those mods didn’t seem to play a role. It also showed that genes of differentiated LCLs could have bivalent chromatin.
- RNA polymerase II bound to both the unexpressed and expressed IGF2BP1 alleles, so polymerase loading wasn’t the deciding factor, either.
- Surprisingly, transcription elongation may be the key. This is a well-known, major rate-limiting step in transcription, going back to work first reported nearly 20 years ago.
The authors note that IGF2BP1 is overexpressed in many cancers, so selective silencing of one allele could be one way that the body protects itself against disease.
Elongate your transcriptional knowledge by reading the full paper at Epigenetics & Chromatin, August 2011.