Dear Bonsor! You have very good suggestions from Amit with multiple solutions you could try. In addition, I just want to add few more experiences facing with low expression yield and/or no expression:
1. For protein with high number of predicted disulfide bonds, you may want to try Origami strain which is specialized for rich disulfide protein expression because it maintains the reduction environment in its cytoplasm. 2. Another possibility you could try is to express your protein in a fusion form with a signal peptide (for example pelB in pET22b(+)) to secrete the expression protein into the periplasm, that could help in several cases. Fusion expression with some other protein may also help. 3. mRNA could be the cause of the problem. Did you check the overall GC content in the mRNA and in some specific regions? If you find very low GC content, i.e below 20%, in some regions, that could be the reason why the mRNA was not stable and tended to degrade quickly after production. 4. If you do not have any problem with mRNA, then you could probably check the secondary structure prediction. If the prediction suggests your protein could have more than one domain, so protein may get degraded into different part after expression. It happened to at least two proteins we were working with that the expression did not yield the right size band on the gel. Then the solution is to re-clone it with different truncated version (if possible) to see if they behave better. 5. Try to gain as much as possible the biochemical properties of your protein to let you identify (by different methods) a possible little expression on the gel that you may have missed. I hope this could give you some ideas for your problem. Thanh N On Tue, Nov 22, 2016 at 1:38 AM, Amit Meir <a.m...@mail.cryst.bbk.ac.uk> wrote: > Hi, > Well, sometimes proteins don't express even though they are from the same > family and their "cousin" behaved very well. > I found the followings helpful: > 1. forget about the concept "this special strain *should* work because it > is a Rosseta/ pLysS etc." Sometimes it's a long mRNA, sometimes a rare > codon, sometimes it's an additional disulfide bond. If you are happy with > the inclusion bodies protocol (and you know your protein is active later), > just scan additional strains that you have in the lab (I like the BL21 DE3 > BLR, which are specialized for long mRNA, but I found them useful in > "regular" genes as well). > 2. Scan "high" temperature (37C) in different time scales - maybe your > cells need overnight expression instead of 3/4/5hrs? Or maybe the protein > undergoes degradation overnight, so it is better to grow it for 3hrs only? > 3. Change the medium to 2XYT or TB, and increase IPTG concentration if you > work with a pET vector. > 4. Try inducing in OD=1 instead of 0.6. > 5. Add glucose (0.1%, but you can optimize) to the media to repress > leakiness of a toxic protein. > 6. Look for "autoinduced media" protocol. The additives there can make a > difference between none to a few milligrams/liter. > 7. Some mutations occurred in the expression vector. Re-clone to a fresh > plasmid (same/ another pET vector/ vector with a different inducer, e.g. > arabinose). > 8. From my experience, sometimes E.coli get along with non-optimized > eukaryotic genes even better than the optimized sequence. I would go to > additional optimization as one of the last options. > 9. A mammalian homolog? Or try expressing in lower temperature (17/20/23C) > to get a soluble protein? (I know sometimes this is not an option). > > > A starting protocol would be several different strains (everything you and > your neighbors have in the labs, basically), in 2XYT + 0.1% glucose, > induction in 0.5-1mM IPTG, then grow 3hrs/5hrs/overnight in 37C. > Alternatively, 2XYT+autoinduced media + 0.1-0.2mM IPTG when OD=0.6 to > "boost" expression even further (and different times as well). > > Good luck :) > > Amit > > > > > Dr. Amit Meir > Institute of Structural and Molecular Biology > Department of Biological Sciences > Birkbeck College > Malet Street > London WC1E 7HX > UK > -- ============================ Nguyen Hong Thanh, Ph.D. student Lab 20B. Macromolecular Structures Department Centro Nacional de Biotecnologia, CSIC C/ Darwin 3, Campus de Cantoblanco 28049, Madrid, Spain ============================ Genetic Engineering Laboratory Institute of BioTechnology Viet Nam Academy of Science and Technology 18 Hoang Quoc Viet, Cau Giay, Hanoi, Vietnam
