Do you have access to a fluorescence scanner? If so, you can use fluorescent labelled oligos, the sensitivity is quite good there as well. Its also possible to use SybrGreen for this.
If you search for non-radioactive EMSA or fluorescent EMSA with Google, you will find a lot of protocols and companies selling kits.
You can also use toluidine staining. The disadvantage of this solution is that you have to use greater amount of RNA than in the case of radioactive or fluorescent labeling.
As far as I know EMSA is used to study DNA-protein interactions. For what concerns the use of toluidine staining, it is used to stain nucleic acids in blue and polysaccharides in purple to sharpen the images from histological tissue preparations. For evaluation of DNA-protein interactions (i.e. potential binding of a transcription factor with a specific protein of interest) you should perform supershift analyses. For instance, let's say you want to know if the protein BDNF activates the transcription factor CREB. Then you should construct a radiolabelled-oligonucleotide sequence corresponding to the promoter region of CREB and run it in the absence or presence of a small amount of the antibody for BDNF. If the interaction exists then you will observe that the band will be shifted upwards since it will run slower on the gel due to the formation of a bond between the two.
Alessandro, thank you for your comment! What you say is exactly how I have imagined EMSA (let's say a standard EMSA), but from Leszek's message above I've got an impression that 1.) toluidine blue might be used, instead of radioactive labelling to detect nucleic acid bands in gel, and so to find out if their electrophoretic mobility has shifted or not, after they had been incubated with proteins; 2.) I understood that some people use EMSA to study RNA - protein interactions which should theoretically be possible, though I'd be more interested in the DNA-protein binding question. You suggested the supershift analysis, but I am interested in a somewhat different question: does a 13 bp long promoter element interact specifically with any protein? This is why I am thinking of EMSA, though it seems some researchers prefer yeast one hyrid system for this aim. It seems to me much more elaborate, however, than EMSA, though as you see, I'd be happy to see that even EMSA can be done some simpler (e.g. nonradioactive) way :)
Your impression is absolutely right and many people use EMSA to study RNA/protein interaction:) You can detect with toluidine both DNA and RNA but as I said toluidine sensitivity is low (compared to radioactivity). You can overcome this by using grater amount of DNA/RNA. I attached typical toluidine staining of RNA. Of course you can use EMSA to study promoter/protein interaction. Your promoter is short so you can obtain it by chemical synthesis. If you have a protein then EMSA is the simplest way to see an interaction:)
Hi Leszek, many thanks! By the way, are you doing this sort of analyses these days? Could it be possible to visit you for a couple of days and see it or even try it myself?