Ramdom competitor in Gel Shift Assay - EMSA expert, please help! (Apr/20/2006 )

using PIERCE EMSA kit to do my gel shift. The probe and competitors are working pretty well. To show the binding and competition specificity, I designed a negative control competitor with irrelevant sequence that was not labelled, and of the same length as my probe. Theoretically, with the random competitor, the band should remain the same as those without any competitor. But my randome competitor made my bind get every stronger. When I put more, it gave a dark, smeared band.

well, first thing's first....if you're going to make a mutant competitor, it is a good idea to be as close to your sequence-of-interest as possible, destroying only the binding motif, and LABEL this probe and run it separately from your primary labeled probe (do you see why?)

also, if you want a nonspecific DNA competitor to be sure it's not just random binding to nucleotides, isn't there a binding buffer component (such as poly dI.dC) that will act as non-specific competitor in every reaction? these are important to be sure you are not picking up any old random DNA-BP with your emsa

OK, the next thing...yeah, your results are whacked. I would write it off as non-specific binding, except that your mutant probe isn't labeled. so, that doesn't make much sense, unless your mutant probe somehow acts to stabilize the binding reaction between your primary probe and TF?
or perhaps your 'irrelevant' sequence accidentally contains another binding motif of which you were unaware (another pitfall when using a random primer-approach to competition)

if possible, could you post a picture of your gel?