SEC separates according to (apparent) size and does not have a very high resolving power. It is rarely sufficient for a single step purification. Normally it is used as a last, "polishing" step, to get rid not only of remaining contaminant proteins, but also of oligomers and aggregates of the protein of interest, proteolytic fragments and of unwanted small-molecule components such as buffers and salts. Even if you only see a single band in SDS electrophoresis, you still have to prove that this is the protein of interest ,e.g. by mass spec. If SDS page shows the presence of impurities, you may consider another purification step. If you do not have any tags to aid affinity purification, Ion exchange chromatography is frequently a good method to separate components that cannot be separated based on size alone.

The first thing I'd try is SDS-PAGE, perhaps on a shallow gradient gel to get a wider range of molecular mass. Run several lanes with different amount of proteins. With the lowest amount, your protein of interest should be well visible. Then increase step-wise until the band of your protein is clearly overloaded. That gives you a very good impression of purity by just eyeballing the gel, but you can also quantify the impurities using NIH-Image (https://imagej.nih.gov/nih-image/).

The one thing this would not show is oligomerisation state of your protein, you'd need a low angle laser light scattering (LALLS) instrument or an analytical ultracentrifuge for that.

have you a run a SDS-page of the fractions from your SEC step, to see if there are other bands in your peak of interest?

Dear Jong

is it your protein contain a his or step tag and you performed other affinity purification step before sec or did you performed directly the sec?

the simplest way is it certanilly load the purified protein sample in different amount in sds-page and when you found an amount where the signal is not saturated, you can estimate the purity by use a programm as image j or image analysis supplied with the instrumemt that are you using for and acquisition to perfprm a densitometric analisys where purity is the intensity of your band vs the Total intensity of the bands on the gel line.

of course if you performed directly the sec is it possible that you have some inpurity at the same mw that you cannot distinguish from your protein and your value will be nor true.

one other approach could be load the purified sample on a second step based on different properties eg ionic exchange or reverse phase hplc which can separate your protein from impurities and estimate its purity from chronatogram.

however could not be simple to undertand which is your protein and which are the impurities of you did not perfom a preliminary step that enrich your protein.

ciao

Manuele

Dear Jong,

you need to perform SDS-PAGE with silver staining to reveal impurities. And you have to load at lot of protein. Based on peak area from SEC, you should load at least 10 µg, maybe 20 µg. This would give you a good idea about the purity. Be sure to distinguish product related impurities from host cell proteins. There are several methods / kits available for measuring CHO host cell proteins. Be aware that for material that will be used in clinical trials there are clear guidelines regarding acceptable levels of host cell proteins (see https://en.wikipedia.org/wiki/Host_cell_protein).

Cheers Markus

@Engelbert Buxbaum

Many thanks to your answer. I used 'ImageJ' tool for numerical analysis, but I recognized that this tool shows large variation depending on the resolution of images. Also cut-off value that I set to remove the background spot affect the purity result I get. It seems that >95% purity is impossible to get with 'ImageJ' tool. Any further advice would be appreciated.

@Camilla Faoro

Thank you for your comment. I run SDS-PAGE after SEC step and I got a clear result with the naked eye (Unfortunately, I don't have a Gel-doc machine to capture and analyze the image).

However, I need numerical analytic data to prove my fraction is highly purified (>95%).

@ Manuele Martinelli

Thank you for your comment.

As you mentioned, I just performed directly the sec and I am concerning some impurity that I couldn't distinguish. Using reverse phase HPLC as a second step is appealing!! This way might help me to get clear result with my protein. Thank you.

Markus Sack

Many thanks to your comment. I totally forgot the HCP. Let me try the SDS-PAGE followed by ImageJ analysis. And I will do the HCP analysis. Your answer really helped me. Thanks.

For quantitative analysis of gels you should first destain the gel carefully, to keep the background as low as possible. In my hands, fluorescent staining with RuBPS (doi:10.1007/978-1-4939-8745-0_12) leads to cleaner gels than silver.

Next you need a good scan (for fluorescent gels, use an orange filter to block out UV light), which you should save without lossy compression (tiff, not jpeg). Quantitation is best done in the middle of a lane, and should be done in an identical manner for all lanes. Adjusting the background line is necessarily subjective, but necessary for good results.

Engelbert Buxbaum

Many thanks to your reply. This thing is exactly what I was in trouble. From your advice, I become clear that I did work in the wrong way. Removal of background and save as tiff would improve my result. Thanks!

SEC separates according to (apparent) size and does not have a very high resolving power. It is rarely sufficient for a single step purification. Normally it is used as a last, "polishing" step, to get rid not only of remaining contaminant proteins, but also of oligomers and aggregates of the protein of interest, proteolytic fragments and of unwanted small-molecule components such as buffers and salts. Even if you only see a single band in SDS electrophoresis, you still have to prove that this is the protein of interest ,e.g. by mass spec. If SDS page shows the presence of impurities, you may consider another purification step. If you do not have any tags to aid affinity purification, Ion exchange chromatography is frequently a good method to separate components that cannot be separated based on size alone.