We use SEC HPLC to help us quantify the % high molecular weight species in a variety of our antibody and fusion protein products. We have found that for some of our proteins, the %HMW species increases over the lifetime of a column. In the most extreme case the %HMW species increases from 1% when the column is new to about 3%(after about 60 injections), then levels off. The absolute area recovery of the monomer stays relatively constant, while the HMW area recovery increases. We are confident that this is not a sample stability issue (when a fresh sample is prep'd the %HMW doesn't go back to 1%, it continues its steady increase). I suspect that we are dealing with a differential recovery of the HMW species vs the monomeric species. I am starting a search for others who have had similar experiences with their SEC and would appreciate any incite or actual literature references on the subject.

It is a known phenomenon in SEC of proteins. People have often attempted to fix it by injecting a large amount ot protein at the beginning of an assay. One suspects active sites. However, I do not have any practical experience with this myself.

What column are you using?

Are there peak width or shape changes during the change in recovery? What do you use as mobile phase?

In response to Uwe Neue: We have are aware that this phenomenon is common and we do inject large amounts of protein when the column is new. However, we only have large amounts of the monomer species. We don't have large amounts of the HMW species. The interesting part of this is that it seems that we need to "condition" the column with the HMW species. Is it possible that "active sites" on the column are specific for different species. We know that different proteins demonstrate differentlevels of recovery. Why not, differential recovery of HMW vs monomer?

In response to A. Mouse: We are using the TosoHaas G3000 SWxl. We have also seen the same phenomenon using a YMC Diol 200 column.

In response to HW Mueller: We don't see any peak width changes during the change in recovery. We have been doing overlay plots and peak width and resolution are very similar from injection 1 to injection 60. In the overlay plots you can see the HMW region "grow" over the course of the recovery change. The mobile phase is PBS, pH 6.5

It seems that you are not changing the surface, that is, it does look like active sites....
It would, therefore, be highly interesting whether this could be influenced via detergents, chaotropic substances..... But: I permanently changed a Super SW 3000 column doing just that(fluorine carboxylic acids caused the change, discussed this here earlier, still don´t know why...).

I am surprised, but the blockage of the sites may be specific. I would have thought that the low MW species will block every active site, and that the high MW species will be ok. But your experiments indicate that you may need to deactivate with the high MW species as well.

We recently also experienced this effect. Our method using TSK300swxl in vertually the shipping buffer worked faultlessly for years with albumin. When using the same setup for “other proteins” we observed varyation in oligomer area with no change in monomer area. This was most noticable on older columns where no oligomer was detected at all. we stabilised the results by modifying the buffer, we increased the concentration of salt untill a constant value was achieved across several buffer sytems.