I'm having issues regarding the morphology of the bands produced in my electrophoretic gel mobility shift assays (EMSAs). At high enough concentrations of protein, the bands begin to take the shape of a 'V' (see attached image). Has anyone seen this effect before? What causes it? These bands make it very hard to interpret the results.
Some information about my assay set up:
Protein concentration: 7.8 - 1000 nM
Binding buffer: 50 mM HEPES pH 7.5, 50 mM NaCl
Substrate: 250 bp, 0.2 nM in reaction
Gel: 5% polyacrylamide (19:1), 1X TGE running buffer, 105 V
This protein (histone-like nucleoid structuring protein, H-NS) is known to bridge as well as stiffen DNA so I wonder if that is the cause of these 'V' shapes on the gel. Other than that, my only troubleshooting ideas are to add glycerol to my binding buffer (to stabilize the protein-DNA complex and hope it runs better) and to change the NaCl to potassium glutamate (others have reported this is a better buffer component based on the intracellular concentrations of chloride ions).
Any feedback would be greatly appreciated. Thanks!
-Ryan
Hi, Ryan,
based on the lengthy smear in both lanes, I can imagine two possible reasons: since it is a histone-like protein, you may have multimers forming at higher protein concentrations that lead to lower mobility. Also, at higher protein concentrations, you may have significantly altered the salt concentration by the addition of the DNA-binding protein in higher volumes of its storage buffer, which could lead to either association-disassociation events and/or multimers of the protein forming.
I think you may have overloaded your gel. So, make a couple of dilutions of your protein bound samples, and run the samples at a lower voltage which may help. You may also want to use 0.5X TGE or 0.5X TG without EDTA. Finally, I would suggest running a 1% agarose gel instead of the polyacrylamide gel.
Sven Buelow - Interesting point, my protein is stored in 300 mM so that is definitely a plausible explanation. How would you circumvent this issue? Lower the ionic strength of my binding buffer to compensate? Repurify the protein and store in a lower salt concentration?
Hua Xiao - I really like this idea. I will try using 0.5X TGE. In order to use an agarose gel, I would have to dramatically increase substrate concentration in order to the DNA with ethidium bromide, correct? In my EMSAs I use 1 nM substrate in a 20 uL reaction which equates to ~ 3.4 ng of DNA. If I recall correctly, I don't think I would be able to see this on an agarose gel. What concentration of substrate would you recommend increasing to, 10 nM?
Thank you for the suggestions, this is great feedback!
-Ryan
Dear Ryan, I think you can use a more sensitive DNA stain such as SYBR Green I or SYBR safe DNA stain, and also you may want to use a little more substrate, for example, 2-3nM, which should be more than enough for SYBR Green stain. As a matter of fact, the gel and running buffer can be 0.25X TGE or TBE, which would be more suitable for low affinity protein-DNA complexes.
Hi, Ryan: my recommendation is to repurify, or if you have sufficient volume, even better: dialyse against a lower salt buffer. If you then still have multimerization due to high concentration of your protein of interest, you could go with a lower protein concentration and choose a more sensitive detection method or longer exposure times.
Sven Buelow - thank you for the feedback, I will keep this in mind!
Hua Xiao - I tried using a 0.5x TGE gel and running the gel at a lower voltage (50V) and while I must admit the band quality is improved (see attached image), I would like to improve them further. Based on these results, I would think running the gel overnight at 10V would allow me to obtain normal, linear bands. However, this seems like a sub-optimal solution. Other than trying an agarose gel, I'm not sure what else I can do to improve the situation. Thoughts?
Hua Xiao - After trying a myriad of protocol changes with no avail, I decided to try running my DNA binding reactions through an agarose gel. To my surprise, it worked fairly well! The bands aren't perfectly resolved but they look far better than anything else I have tried (attached). I should be able to easily interpret the results of my experiments now. Thank you so much for your advice!
In case this helps someone else, the experimental details are outlined below:
- 100 ng DNA substrate/reaction
- 1.65% agarose gel (poured with 1x TBE)
- Gel was run in 1x TBE at 120 V for ~ 70 minutes
- DNA was visualized using 1 µg/mL ethidium bromide (25 mL volume)
For anyone that this might help in the future, I found a better way of resolving large nucleoprotein complexes in native PAGE. It's called an agarose-polyacrylamide composite gel. To prepare this gel, make a solution of 1% agarose (melted in 1x running buffer) and a solution of 4% polyacrylamide (5/6% works too). To cast the gel, simply mix equal volumes of both solutions, vortex, and pour. The agarose functions to stabilize polyacrylamide gels that are
- Badges
- Science method