The resolution of gel electrophoresis is greater than the resolution of SEC, so a single SEC peak may actually contain several different proteins that are resolved as bands on BN-PAGE. Additionally, proteins elute from SEC based not just on their size but also by their shape, so the proteins in a single peak may actually have various molecular weights.
In my case I'm purifying a single protein and using sec to separate the protein in its in monomeric, multimeric or aggregates form. In this scenario it is possible for you that a multimerband aggregate ellute in the same peak? If is so sec looses any utility
That would depend on the size of the monomer and the grade of SEC matrix that you are using and possibly also whether the multimers are stable in the buffer system that you are using to separate. Details of the size and matrix would help
I'm using a superdex 200 pg. Protein multimer size arround 340 kD. And making the bn-page of the fractions I we this strange band in all of them. And the band is between the 720 and the 1040 kd reference ladder.
The superdex will include all sizes from 10,000 to 600,000 so the separation on size looks appropriate. Native PAGE relies on charge to mass ratio which may outweigh the size effect of electrophoresis. I wonder if the detergent effect of the Coomassie blue dye is causing the aggregate to denature and the individual molecules have a very strong charge that is hidden from the surface of the multimer in the multimeric state so the mobility of the monomer is unrepresentative of its size? Just a guess. Possibly also non globular protein has co purified with your protein but is actually a different size and this is now showing up on the more accurate PAGE as Adam says
If you are using an acidic dye like nigrosine it will be negatively charged so it will not stick to negatively charged protein. Perhaps it would not stain some of the negatively charged protein thus making your protein look purer. Would this fit the electrophoresis result?