Your protein may have multiple possible aggregation states (e.g. monomer, dimer, hexamer, aggregate), with the distribution between them affected by the conditions. A great way to find out what is happening is to use a SEC-MALS system to measure the mass of the protein in each peak during elution. If a SEC-MALS system is not available, dynamic light scattering can be performed on each peak to measure the particle size, which can be used to estimate mass. Analytical ultracentrifugation is another approach to this problem.

We don't know the dimensions of your column. It would be better to tell us what size (kDa) relates to the Ve values. Run some standards on your column so you can tell us if Ve 9.8ml is twice or triple the size of Ve 11.35ml.  Also, did you analyze the peak on SDS-PAGE and native PAGE?

Your protein may have multiple possible aggregation states (e.g. monomer, dimer, hexamer, aggregate), with the distribution between them affected by the conditions. A great way to find out what is happening is to use a SEC-MALS system to measure the mass of the protein in each peak during elution. If a SEC-MALS system is not available, dynamic light scattering can be performed on each peak to measure the particle size, which can be used to estimate mass. Analytical ultracentrifugation is another approach to this problem.

Hello Cornelia In this case, it is too difficult to know the relation between the Ve and the weight because when you read the Superose 6 information  Thyroglobulin (Mr: 660.000 Da) is out of values to calibration curve ( Curve Ve/Mr). I think the differences between the peaks are big enough.

Hello Adam, it is a good idea to do de dynamic light scattering and I am going to do it. But I dont understand when i add additives like glycerol, arginine or increase the NaCl concentration change the column profile ( I said it in the introduction) and this peaks change along the time and i think the protein concentration is significant too.

Hi David,

I agree with Adam about possible multimers electrostaically linked, frequently occurring on very large proteins. Should be added the possibility of proteolytic degradation.

Considering retention, adding 10% glycerol may considerably change protein conformation and/or interaction with pattners and surrounding water,  and then affect retention. Also retention changes due to viscosity changes of the buffer has to be considered according to the physical Stokes law.

Good luck

Yes maybe there are differents oligomers in the solution but what can I do to stabilize one of them?. Thank you.

As Adam pointed out, the presence of two peaks is indicative of heterogeneity in the oligomeric state of your protein of interest. I presume the protein is very pure at this stage so the other peak is not a contaminant?  Alternatively, it could be a proteolytic fragment as Patrick suggests, SDS-PAGE would confirm this. Do you have a functional assay for your protein? I would recommend running both peak fraction through this assay and pooling fractions based on activity vs total yield.

Hello Neville, I have a typical contamination ( Arna gene) it appear in all  fractions ( Both peaks) and it has other peak in Ve=14 ml where it appear too ,but in more quantity. Maybe there is an interaction between both proteins but why do without additives appear one peak and with theirs appear two?. Thank you.

Hi David,

I definitively think you're analyzing a protein complex or agregate, seems to have arnA as patner.

Are you shure you are not in presence of inclusion bodies typical in EColi protein overexpression?

Oligomers may also be separated when using mercatan in buffers (such as glutathione or DTT frequently used as protease inhibition agent).

I post a Word document ( It is in Spanish language sorry) with several images.