With Santa’s yearly gift giving spree fast approaching, kids everywhere are on their best behavior to avoid ending up on the “naughty list.” A consortium of chromatin researchers found out which ChIP antibodies are naughty or nice, by testing a ton of commercial antibodies raised against 3 histones with 57 different modifications in flies, worms, and humans.
Given the considerable time and expense of ChIP-chip and ChIP-seq experiments, the last thing you want to worry about is having an antibody that’s a dud. Maybe they get too cozy with other histones, modifications, or nonhistone proteins, or just stubbornly refuse to pull down chromatin. Either way, a misbehaving antibody could spell big trouble for your analysis.
As part of their activities in the NIH modENCODE and Roadmap Reference Epigenome initiatives, scientists from UC Santa Cruz checked the behavior of 246 commercially available antibodies in western blots, dot blots, and ChIP-chip or ChIP-seq experiments. The project found that while most antibodies behaved themselves, there are still several that acted up. Here’s the report card:
- Western blot analysis found 63% of the antibodies bound specifically to the appropriate modified histone, 26% showed cross-reactivity with other proteins, and 11% produced no signal.
- Dot blots showed 73% of antibodies were specific for a single type of modified histone peptide, 13% were cross-reactive with other modifications, 11% yielded no signal, and 3% had a low signal.
- In ChIP-chip or ChIP-seq experiments 78% of tested antibodies produced reproducible ChIP results, whereas 22% failed.
These results are in line with another recent study that tested the specificity of histone modification antibodies using peptide arrays.
The findings on specific antibodies are posted in a searchable Antibody Validation Database, and other researchers are encouraged to upload their own validation results. But the authors don’t recommend just taking their database’s word for it. They say that although the database will be a great resource for the chromatin community, you’ll probably still need to validate your own antibodies since lot-to-lot variation can be substantial.
Make EpiGenie’s “nice list” by reading this paper at Nature Structural & Molecular Biology, December 2010.