Protocol Online logo
Top : New Forum Archives (2009-): : Protein and Proteomics

Help! Purifying native protein-2 years and counting... - (Dec/19/2012 )

I am at a loss as to what to do at this point. I do not really know what I am doing as far as this project goes, and was hoping someone here could help!

I am trying to express and purify my protein so we can use it for either functional studies, or for crystallography of some sort (both, ideally). I started out by inserting my gene into a His-tag expression vector (cloned at N-terminus). After fighting with that for a year, I finally got the protein to express in e. coli, was able to use a Cobalt column to (mostly) purify some of the protein, but then started having problems with either protein stability, or specificity. When I would take my sample and run it for SDS-PAGE and blot with the proper antibody, I would get a nice band at my expected size, but also some almost equally prominent smaller bands. Then I tried running the same sample again, and would get nothing. This happened even after adding protease inhibitors. I had eluted the protein into the suggested buffer of (50 mM Sodium Phosphate, 300 mM NaCl, 150 mM Imidazole, pH 7.0). It was suggested that I try running a gel and staining with coomassie to see if these other bands were prominent, and to get an idea of how much protein I had present. I did this with a freshly purified batch of protein, and got no bands. Even on a western. Repeated and got many bands, not even at good sizes for my protein. At this point, I had to go on leave for 3 months.

When I came back, my PI suggested I try ordering a pre-cloned mammalian expression vector with a FLAG tag (also, a different/more relevant isoform). I did this, did a transient transfection into 293T cells, and have tried using the ANTI-FLAG M2 Affinity gel to pull down/purify the protein in smaller batches. Again, western blotting appears to have good expression but other, non-specific bands.

I think the biggest problem I am having is that I know next to nothing about protein expression and purification. Nor can I find the desire to become an expert on it... I'm a lab manager/technician who was given this project because it was supposed to be easy, two months, tops! And now nobody else wants to/has time to/has the desire to take it over. Any suggestions would be greatly appreciated! Thanks in advance!


cobalt binds strongly than Ni so you should elute with more conc. of imidazole. Try 200 or 250mM.

Stick to protocol. check for pH of solutions.


It is not uncommon to have more bands than what you want when doing western blots. Your antibody could be a little unspecific and is causing this. I agree that you should just stain with coomassie and see if the other bands are there. The western blot will detect the smallest of contaminants and with coomassie it may not even be a problem. There are ways to clean up proteins when you have a mixture of contaminants. Size-exclusion chromatography will give you your desired band.

Also, prabhubct is right that you should try eluting with 250mM imidazole. Try washing with 10mM imidazole about 5 times before adding the 250mM imidazole, this will get rid of the non-specific binding to the cobalt.

The last thing, if you have truncated proteins that have the his6 tag and are the contaminating bands you can try using an expression cell with a plasmid for rare codons like Rosetta pLysS cells. Here, it supplys rare codon tRNA's so you can get full-length read through during translation. But first you need to determine if these contaminant bands have a His6 tag or not.


Thank you for your suggestions. I will look into these for my His-tagged protein. I have a few more questions:

Do you have any suggestions as to how to increase my protein stability during and after purification? What is the best lysis buffer for my cells (both bacterial and mammalian)?

How do I determine if my contaminant bands have a His tag or not? By blotting it with a His antibody? I did do that, and found bands at the same size as my protein of interest. I assumed that these were either incomplete proteins or degraded bands. I also have some of these bands at higher MWs than my target protein, though, so I wasn't sure what was happening. One colleague suggested that I try changing my culturing conditions during expression, especially for my mammalian FLAG-tagged vector (i.e. incubate at 30 degrees instead of 37 degrees). Does this sound right? I would prefer to purify this protein isoform, although I am getting a sinking feeling that if I want to get the amount of protein I want, I may need to figure out if I can clone it into my His-tagged vector, which took me like 6 months to do last time.

One other question, when running the SDS-PAGE for coomassie staining, what amount of protein do you usually run per lane? I was trying 40 ug, but am not sure if this is correct.

Thank you again for all your help!


there are a lot of ways to increase protein stability (depending on the protein). some are more stable within threshold concentrations (low and high); some with additives (bsa, glycerol, peg, detergents,...); pH, cations, chelators, salt concentration, buffer salt,...

blotting with anti-his will tell you if your contaminants have his tag.

for coomassie stained sds-page, if your protein is "pure" then 40ug is a lot more than you need (we use 40-50 ug for partially purified, still on the crude side or to check purity by overload).