Usual DNA-protein interaction studies use ChIP/ arrays /EMSA to identify the interaction and in most cases we know the protein of interest. What can be done to identify unknown protein interacting with a known, small sequence of DNA?
you can have one labeled oligo that is your bait and then another which is scrambled as negative control: you can do EMSA and look for shifted band in positive control DNA sequence. Then the area of the gel can be digested and treated with DNase and submitted for Mass spec. Good luck!
If you are interested in general in the proteins which may be binding the sequence, you could use a pull-down approach followed by mass spec. For example, apply cell lysate (from whichever cells you are interested in) to biotinylated DNA probes (sequence of interest as well as a control "scrambled" probe as mentioned above) which have been pre-conjugated to streptavidin coated beads (agarose or magnetic beads are common choices). Incubate for a period of time, do a number of washes with a stringency you find appropriate (DNA-protein interactions are generally electrostatic, so washes with an intermediate amount of salt will be essential to release more loosely bound protiens), and then elute. You can elute in SDS-page sample buffer with boiling, after which you can compare the samples (your seq. vs. scramble) on an SDS-PAGE gel and look for bands present with your sequence but not in the scramble. You can then excise these bands and send for mass spec. Obviously, there are many possible variations on this basic approach that could be useful.
Thank you Eric. I was thinking along the same lines too, but I would want to try a method that doesnt involve mass spec analysis.
I think you have excellent suggestions above. Have a look to this paper (Tacheny, et al. Nucl. Acids Res. (2012) 40 (21): e168). I don’t know why you don’t want to use Mass Spec. Some companies can do it for you for less than $ 400. Alternatively, seek collaboration with some on campus teams
Thank you Eric, Bharat and Abdelhalim for your suggestions.. I was trying to find alternative methods to mass spec but it seems like that is the best way to tackle the problem. :)
If you had some idea of what protein it MIGHT be, you could run a series of western blots. Without a very focused list of candidates, however, mass spec is definitely more cost effective due to the expense of antibodies.
These answers are not incorrect but you need to realize that any signal on an EMSA is likely to reflect a very small amount of protein, on the order of picograms. A good mass spec facility may need 1-10 nanograms to get some peptide sequence. Identifying a nuclear protein bound to a specific sequence involves making nuclear extracts from a large number of cells. You need to remove the non-specific DNA binding proteins, of which there are many, by passing lysates over DNA that does not bind your protein. A mutant version of your sequence would be best. Then use use your oligo fixed to a column to try and pull out your protein. You will likely need a 10,000 fold enrichment or so, meaning that you will need to start with milligram amounts of nuclear lysate. There is no quick and dirty way to get results here.
Dear Eric A Simko and others,
Im doing a pulldown essay now to find a regulator of an operon that upregulated after UV induction. I also used a biotinylated promoter of this operon and the Strep tactin (engineered of streptavidin) in a column istead of beads. After eluting, ran on SDS PAGE there were many bands in both control and experiments lane. I used salmon sperm DNA as the nonspecific inhibitor. Do you have any suggestion for my case?
- Badges
- Science topic