Protocol Online logo
Top : Forum Archives: : Molecular Biology

EMSA probe design - (Oct/19/2005 )

Hi, I am just setting up for EMSA analysis, and I am wondering about the probe design. I want to investigate the binding of two factors to overlapping consensus sequences that together are 10bp long. I have designed 20bp oligos and plan to label the 5' end of the complementary oligo with biotin before annealing. Does this sound okay, any obvious problems??

Also, I saw a protocol where someone suggests putting "half-sites" on the ends of the oligo (they use BamHI and BglII) what is that for? They don't seem to use them in the protocol at all...

I think I am going to use your protocol, but I don't know about buying the pierce kit, I plan on ordering biotin labeled complementary oligos so I won't need labeling components... What do you suggest?

Thank you all for your help.


Becca- I don't blame you. We are a po' little lab or I would have ordered them that way myself rolleyes.gif i get a lot of mileage out of those labelling kits.

another thing to note...I only ever label one strand, and I get plenty of oligo for detection...I mean plenty. so I think you could order one strand labelled and one strand unlabelled and anneal them, if you are ordering the strands ss. I suppose it depends on your ordering options with your synthesizing company, maybe you'll just get them ds to begin with?

Oh, and hey, I would add BSA to 50ug/mL to the binding buffer; I started doing that a while back when I found that it generally gives a stronger supershift. I suppose I should look into editing the protocol for that step...

As far as the 1/2 site thing, I can't help you there. I don't know why you would do that?

I have looked at sites about 5 or 6 bp apart, but never contiguous sites. If you want to separate the sites in your analysis, I would add a couple bp mutation in one site on one oligo, then a couple bp mutation on the other site on a different oligo. or are you looking at the interaction of two TF's on what is essentially one site?

good luck, becca!




Thank you very much for your reply, I do plan on using mutations to investigate the binding, we did site directed mutagenesis of these two sites in luciferase constructs and got wierd results so I am trying to determine if the Sp1 "specific" mutation affects myc binding or vice versa with emsa analysis... I will take your advice and only label one strand (saves lots of money!) biggrin.gif Thank you again for your help...


I've spent all sorts of time wondering how much the adjacent strand sequence will affect TF binding. Someday when I'm bored maybe I'll plan some experiments. From what I have been able to glean in the literature, sometimes it can make a difference, but the rules don't seem very hard and fast here. there was a very interesting paper out earlier this year about NFkB subunit orientation on the strand and the effects on expression; there are just so many factors.

who knows? maybe you'll get some crazy synergistic effect dry.gif

let me know if it works for you when you get there?