I am performing ChIP on a DNA repair factor in budding yeast. We are interested in looking at the enrichment of this factor in WT and some mutant strains. And I am using a strain that can induce a single locus-specific DNA double strand break upon shifting the carbon source to galactose. I have my DNA repair factor tagged with 3xFLAG and I'm using EZ View preconjugated FLAG beads for my IP.
I collect cells before Galactose induction and at various time points after galactose induction, use IgG as mock IP and primers at various distances from the DSB sites to probe for enrichment.
So far, I have only performed ChIP in my WT strain and I'm getting some weird results. When I normalize my qPCR results, I usually get the highest enrichment from non-induced sample and the rest of the samples show some enrichment but are much lower than my non-induced (no DSB) which is odd to me as I expect higher enrichment when DSB is induced.
I am not sure if the problem is with more cell lysate in later time points (as cells still continue to grow for some time after DSB induction). Since what I do after lysis and sonication, is to save 1% as input and the rest of the lysate is divided equally between mock IP and FLAG IP for all time points (regardless of how much cells I actually collected). Does anybody have any recommendations on how to troubleshoot?
Please note that I am not an expert in research on yeast. I have experience with ChIP on brain and other tissues and cell cultures.
There can be several reasons for variations in ChIP results. One of the most important ones is the fragmentation of the cross-linked DNA. Did you check on an agarose gel or a bioanalyzer the degree of DNA fragmentation of your samples? Fragmentation needs to be roughly between 100 bp and 1kb (although opinions differ here) and, importantly, must be equivalent between samples else they are not comparable. If samples are not comparable then you would need to look into your sonication protocol.
Besides the problems with variation you could also consider to calculate your enrichment as Bound/Input. This may give you a better idea about your enrichment levels. For this you would need to do a high-sensitivity Qubit to determine dsDNA content before running a certain amount of DNA on the qPCR.
Good luck!
How are you normalizing qPCR results?
As said before, you should calculate what is called "%INPUT" recovered in each IP. If you still find that samples are not comparable, you can try to meassure INPUT and ChIP DNA concentration (for exmaple with QUBIT) and normalize with thesee values.
Also calulate relative binding values of regions were you expect no binding to see what is the background level = the zero of your experiment. It may happen that you are not detecting a real binding, if you measeure for a region a value of 0.1 vs 0.01 between samples, it seems a big difference....but maybe this is background and the real binding event would gave you a value of >1 for example.
- Badges
- Science method