Purification of a very small, highly hydrophobic membrane protein. Any sugg

2 posts / 0 new
Last post
bslfw's picture
Purification of a very small, highly hydrophobic membrane protein. Any sugg

Dear all,

I have just started to try to purify a small, very hydrophobic viral membrane protein (the protein is GST and flag tagged as there is no antibody to the endogenous protein) from BL21DE3 cells. This protein seems to inhibit/kill the bacteria and after induction at O.D 0.7, 1.0 or 1.3 the O.D ceases to increase, flatlining in each case at approx 1.2. I'm just about to start some viability assays to determine whether this protein is preventing bacterial growth or indeed killing the cells.
Upon lysis and susequent french press i spin my lysates at 7000g for 10 minutes to pellet the inclusion bodies. I can detect my viral protein in this fraction but i am losing a lot of my protein in the supernatent of this spin. When i spin this supernatent at 20000g i can detect my protein.

 Following these spins the pellets are resuspended in cleavage buffer and the GST is cleaved from the fustion protein leaving my flag tagged viral protein.

From the inclusion body pellet i can only detect the monomer of my protein when i solublise it with 1% sarcosyl (it appears that this protein oligomerises even in denaturing and reducing conditions) and cleavage practically failed (didn't work nearly as well) when i resuspended the pellet which was a result of the 20000g spin.

When i run each sample on a HPLC C4 column..... i do elute my viral protein along with some of the fusion protein but in very very small amounts.  

To get to the point!.... I can't see the fusion protein by coomassie from a 0.5ml pelleted sample of my induced bacterial culture, but i can detect it by western blot. I need to increase the amount of fusion protein that i have to begin with in order to increase the amount of final protein i have! I've tried growing the bugs at different temperatures and also inducing at varying O.Ds. My next step was to just increase the number of bugs i use! Bacterial expression is new to me and i wondered if there is anything glaringly obvious/tricks of the trade that i could try to increase my inital amounts of fusion protein? I'm guessing that cleavage of the fusion protein when its pelleted with all the other cellular debris isn't working because there's too much lipid and junk around?? Is pelleting the inclusion bodies the best option?!

Any help/suggestions would be much appreciated.

Thank you,

2nd year PhD student

Sami Tuomivaara
Sami Tuomivaara's picture


As you know, membrane proteins are generally a tough to overexpress in high yield. The decision with biggest impact in overexpression is generally the choice of vector. In which terminus the GST tag is relative to your protein of interest? This can make a difference also, since it might dictate whether your protein is directed to the membrane or not. Translational machinery can get clogged if hydrophobic protein is being translated but not inserted to the mebranne.

My understanding is that it would be easier to purify from membrane (or periplasmic space) rather than from inclusion body. Note that your protein of interest is naturally probably highly glycosylated (depending on the source) and it might not just be stable without proper glycosylation. In that case, you need to use other expression system with proper post-translational modifications.

OD flatlining at 1.2 might not have anything to do with your overexpression, just natural plateau in the growth curve, especially when all three conditions you tried plateau at the same OD. How long do you induce? Otherwise it looks like you tried the obvious, temperature and induction OD.

Also, it looks like the GST tag renders the fusion protein (at least partially) soluble. Why not to utilize that and, and you hinted, try to get in solution rather than pellet it. The cleavage won't work efficiently (if at all) for non-soluble substsrates.