EMSA: non-specific binding is quite strong ??

From: Xun Lu
Date: 24 October 2011 20:37



Dear all,

       I have a protein-DNA crystal structure, and am doing EMSA to confirm the protein-DNA interactions that I am seeing in the structure.

       I am using IRD dyes, which can be analyzed by Odyssey system (LiCor). The Kd is estimated to be around 2nM (assayed with 0.3nM probe and 1-15nM protein). The transition is very steep, so the data do not seem to fit the Michaelis–Menten equation. But the most strange thing is that even the "random" DNA can compete for the binding very well (just 3 fold less well compared to the consensus binding site). I designed the "random" DNA so that it is of the same length with the labeled probe and does not contain the consensus binding site. Then, I tried dIdC. 30pg/ul of dIdC competed for about half of the binding (49% bound probe ---> 29% bound probe).

       I repeated EMSA many times, and my signals were quite good and consistent. Why is the non-specific binding so strong? I really don't have any clue. The DNA concentrations were measured by both nanodrop and normal Spectrophotometer.

       Any advice is welcome!


Thanks,
Xun

--
Department of Molecular and Structural Biochemistry
North Carolina State University