I have carried out a native PAGE with 4 reaction mixtures. To each I had added an equal volume of EDTA (1 µl/1mM) to sequester any divalent ions and an equal volume of calmodulin (5 µl/0.5 mg/ml). I added 2 µl of calcium (2 mM), magnesium (2 mM) and manganese (2 mM) ions respectively to three reaction mixtures and had a control that was made up to an equal volume with Tris-HCl. All reaction mixtures were made up to 10 µl with Tris-HCl. Here I ran into a problem with the waterbaths and the reactions couldn't be incubated at 37 degrees Celsius.

After I added the loading buffer to each, I added the 4 reaction mixtures to the polyacrylamide gel and electrophoresed for 45 minutes at 150 V. After this I placed the gel in Coomassie Blue stain for 20 minutes.

After the destaining process, I reviewed the gel to find that the calmodulin in the 4 lanes had migrated the same distance. I had expected the EDTA control to have migrated the furthest on the gel, with Ca2+ and Mn2+ migrating similar distances due to their similar ionic radii; making it a good substitute for the binding to the calmodulin. I was unsure of how far the Mg2+ would travel as I thought it would have a lower affinity to the calmodulin due to its binding to the N-terminus rather than the C-terminus. However, I was wondering if this was negligible in the absence of Ca2+, although now I can't be sure due to the results of the gel.

I was wondering what errors in the experimental design or process could have resulted in this? And if the experiment was run correctly, what results would be expected?

The first image is the result I was expecting. The second is the results from my gel electrophoresis.

This was the result I was expecting

This is the result I got

  • $\begingroup$ Why add EDTA if you want calmodulin to bind ions? Are you expecting just ion binding to influence migration rate? $\endgroup$
    – canadianer
    Mar 7, 2015 at 19:37
  • 2
    $\begingroup$ Binding of a cation would not change the mobility much so as to be detectable by electrophoresis. $\endgroup$
    Mar 7, 2015 at 19:57
  • $\begingroup$ If you want to sequester the bivalent ions, you need ways more EDTA. It complexes an equivalent amount of ions, so 1mM EDTA complexes 1mM of calcium (for example). If you EDTA solution is 1mM, but your ion solution is 2mM, this will never work. $\endgroup$
    – Chris
    Mar 7, 2015 at 21:47
  • $\begingroup$ Apologies, edits have been made to include the concentrations. EDTA was added under the instruction of my supervisor, which I did question at the time. I was under the impression that the migration would be detectable through the existing literature. $\endgroup$
    – user14821
    Mar 7, 2015 at 21:49
  • 1
    $\begingroup$ What article is that figure from? Also, you should destain your gel. $\endgroup$
    – canadianer
    Mar 7, 2015 at 22:49

1 Answer 1


The change in gel migration distance is due to the conformational changes due to the binding of the ion, and have nothing (detectable) to do with the ionic radius of the ion.

The left image shows calmodulin with no ligand bound, and the right image shows it with calcium and peptide bound to it.

enter image description here enter image description here

As you can clearly see, the unbound calmodulin is significantly thinner, and therefore can migrate faster through the gel. The change in charge on the protein due to cation binding in native PAGE may also affect the result.

Since the calmodulin conformational change is enhanced by both a peptide and calcium cations, as well as its phosphorylation state, it would be reasonable to investigate the contributions of these factors to your result. Reading the original paper again to see what you missed the first time may also be helpful.

  • $\begingroup$ Yes the left panel is apo-calmodulin, but Ca(II)-calmodulin also resembles structure in the left panel (with minor structural adjustment). The major conformational change on the right panel is a result of peptide binding, which is not part of experiment in the question. $\endgroup$
    – SYK
    Mar 19, 2015 at 5:12
  • $\begingroup$ @SYK I am pretty sure that peptide cannot bind without calcium also binding. $\endgroup$
    – March Ho
    Mar 19, 2015 at 5:35

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .