If your protein is rich in prolines, that very close other band may actually be a different isomer conformation, as those have a tendency to run at a slightly different rate on SDS-PAGE. A westernblot is the best place to start, to verify for certain the other band is also your protein.

The suggestions here are all great. I'll just add that if you are getting this degradation (sounds like if so it'd be near the C-terminus, and likely a carboxyl protease as the culprit), you could think about using a cell line with a chaperone that may protect your protein, such as the SHuffle line from NEB.

Finally, since EDTA can't be used to inactive metalloproteases, you could try instead citrate at 5x the molar concentration of the EDTA you are replacing. Citrate is mild and should not strip the nickel resin, but acts as a good buffer to grab up the accessible metals and sequester them from metalloproteases. This is a trick used for zinc finger protein purification, for instance.

Good luck and all the best.

Have you considered a Western-Blot (anti.His) and Peptide-Mass-Fingerprinting of your second band? Often they are not degradation but co-purified impurities (like chaperons) ... and sometimes they are bona-fida binding partners. Perhaps you could post gel-photo?

it may also be due to poor protein translation due to rare codons within the transcript. Rosetta strains should improve that.

It would be easier to give you some idea if we new little more about your work; such as ... (1) Do you purify a soluble or an integral membrane protein? (2) How high is the imidazole concentration during the binding and the washing steps? (3) As far as I know Roche has more than one kind of protease inhibitor tablets; which one are you using?

If your protein is rich in prolines, that very close other band may actually be a different isomer conformation, as those have a tendency to run at a slightly different rate on SDS-PAGE. A westernblot is the best place to start, to verify for certain the other band is also your protein.

The suggestions here are all great. I'll just add that if you are getting this degradation (sounds like if so it'd be near the C-terminus, and likely a carboxyl protease as the culprit), you could think about using a cell line with a chaperone that may protect your protein, such as the SHuffle line from NEB.

Finally, since EDTA can't be used to inactive metalloproteases, you could try instead citrate at 5x the molar concentration of the EDTA you are replacing. Citrate is mild and should not strip the nickel resin, but acts as a good buffer to grab up the accessible metals and sequester them from metalloproteases. This is a trick used for zinc finger protein purification, for instance.

Good luck and all the best.

according to above sugesstions, if u r sure that it is ur protein which is getting degraded during purification, then take these precautions while purification:

1. use E. coli BL21 /codon plus cells for overexpression

2. use 2X YT media (enriched one)

3. induce at 16C overnite (12-16hrs max).

4. as u r already performing all the steps at 4C, go ahead with that only.

5. use less resin (0.8-1ml resin for 1L culture)

6. in the columns, bind slowly by passing the cell lysate using flow-rate controllers. The whole binding step shud not be more than 1 hr (for 20ml lysate), at 4C.

7. add 10% glycerol to all the solutions including elution buffer.

8. do not perform washing step for more than 3-4hrs, with slow washing, but faster than that of binding step.

9. do not add detergents during washing.

10. after elution, aliquot ur protein (10-30ul), and immediately store at -80C.

Hope these precautions will help u.

You could try including 1% protease inhibitor cocktail in all your buffers.

Better to use protease inhibitor cocktails in buffers

Thank you everyone for suggestions. They have all been very helpful and I am definatly looking forward to trying some of these things.

Attached is a rough gel image of my latest purification. In this gel, you can see the lighter extra protein band below the darker one (the protein of interest).

I have previously performed a western blot using an antibody against the His tag and both of these bands showed up as being my protein. This is why I think this is a degredation product of my protein. Is there any other way to confirm?

My current procedure consists of using BL21-DE3 lys cells expressing a pET28a vector with my protein containing a His-tag at the C-terminus. I grow 250mL of cells to an OD600 of 0.6 and induce expression with 1mM IPTG for 16hrs at 22C. Previous experiments have suggested that my protein is a soluble one. To purify, I first use a Ni-NTA resin column (0.5mL column volume). I perform washing steps with an imidazole concentration of 20mM. I then use buffers containing 60mM and 80mM of immidazole to elute. I then take this eluate (~10mL) and put it through an ion exchange column (Q Sepharose Fast Flow). In this column I wash with 0.2M NaCl and Elute with 5 mL of 0.4M NaCl. This second column's elutions is what you see on the gel. My protein elutes during the first two mililiters along with the other lighter protein band.

I use a Roche complete, ULTRA, Mini, EDTA-free, protease inhibitor tablet in my lysis buffer only, but all my other buffers contain 1mM PMSF.

Does this information change anything?

Thank you again for all your help!

just one thing....after washing with 20mM imidazole, wash with 10ml of buffer containing 50mM imidazole and add 200mM imidazole in final elution buffer. since u are anyways performing gel exclusion, high imidazole will not hamper ur protein quality but improves ur quantity of protein. u can have a look at the protocols (see the link) that we usually follow and we do get very good yield.

and I cant see ur gel image....

http://www.ncbi.nlm.nih.gov/pubmed/19826007

You might consider expressing your protein in a Lon and OmpT protease-deficient E. coli strain. Such strains are commercially available.

I would perform mass spectrometry analysis for the degraded protein band and find out where the cleavage occurs. If the cleavage occurs at N or C-terminal region of protein, I would create a new construct with exclusion of that region. However, it all depends on your purpose of studies if you need the full length protein for your experiments suggestions are provided here good.

Good luck

As a couple people mentioned above, I'd recommend trying a couple other/additional protease inhibitors. I had a smilar problem a while back, and adding PMSF to my protease inhibitor cocktail did the trick.

Are you sure proteolysis is in play? True degradation needs to show other products over time; have you tried to really degrade it and take the breakdown to completion, or at least show an enrichment/ depletion exchange between the bands?

Without knowledge of size and other protein characteristics, nevertheless it’s important to fully characterize your expression target and be sure it’s pure. Two possible glitches could be a gel migration problem, and the other a true close expression mixture. For gel migration, presence of salts in the load mixture can cause fuzzy bands, and thorough dialysis (in cleaned dialysis tubing vs EDTA) is used to draw them out; this is also needed for isoelectric focusing, which should be done stat if not already tried. If you have two different bands in IEF, you can distinguish two polypeptides comigrating at the same molecular weight.

Another common gel problem, if your MW is near 68KD, is albumin—binds so many things, and doesn’t migrate sharply. Culture media with serum or embryo extract are a pain, but excess albumin can be extracted from a lysate with an Affi-Gel Blue column. Brzeski et al. (Biotechniques 35:1128) describe using a lectin affinity column for depleting human serum albumin, and there are other means for prefractionation. Try dialysis and a blue column separately against your stuff, then do a quick assay to be sure you still have the activity—they are simple steps.

You can also go directly to 2D-PAGE Western blot—if done in your lab, this would be my vote. The IEF step will desalt and separate by isoelectric point, and you might get two spots far apart on the final blot. Then you have to explain how you labeled two separate proteins! But maybe it’s as simple as some of your target was making friends with another molecule like albumin and got distracted; IEF might resolve that.

Ya I experienced similar things and found the additional band was a chaperone and not the part of my protein of interest! so better to make a mass spec of your both bands and dont spend too much time on thinking about the additional band!

Hi Arianna

I encountered similar problems. You can try the followings:

1. Protease + Phosphatase inhibitor(de-phosphorylation of proteins sometimes make them prone to degradation). just add a Phosphatase inhibitor(Sod Ortho Vand etc). Thermo scientific sells these .

2. Use proteosomal inhibitors too (people hardly use them but some case they are very effective to stop proteolysis)

3. if you deal with purifying a protein complexes, its very prone to degradation, Simply the different components of the complex separates out if there is change in PH, viscosity or ionic strength. Make sure the bio-physical parameters are maintained and you in all case prevent FREEZE THAW.

Best of luck

If you want to use EDTA in your homogenization/loading buffer, you may reverse the purification scheme, use ion-exchange first, wash of EDTA using wash buffer and next elute bound proteins with high salt and load directly to Ni-NTA. His-tag affinity to metal ion on gel (Ni, Co, Cu) is still high enough to allow binding even at 1M NaCl, so you also save time this way.

Second using EDTA at 1-5 mm in the lysis buffer. Then you have to either dialyze it away or dilute it before going to the nickle column, or use the ion exchange column first. Metaloproteases will bind to the nickle column and if you have a broad elution concentration they will co-elute with your protein. I have found Talon resin to be good at not binding proteases since it is more specific for His-tagged proteins (Clontech will send a free sample). The other thing I saw is that you are not using gradient elutions. I would start using them, find the concentration needed to elute the protein, and only then go to a step gradient (but you may have to stick with a gradient elution). You could take fractions from the gradient elutions and incubate at 37oC to see where the protease is co-purifiying with your protein, then either modify the elution or just throw away those fractions.

To prevent proteolysis, I typically do all of the purification steps in one day. A benzamidine or gelatin column can remove proteinase impurities. And adding EDTA and EGTA is a good idea if your protein is not a metalloproteinasae. Just be careful of the EDTA concentration when using A Ni column as described above. One way around the proteolysis problem, is to make the protein go into inclusion bodies, purify the IBs, solubilize in Urea, and refold on the Ni Nta column. I have done this and the technique works great, especially for partially soluble proteins to begin with. Working fast is typically the key for me.

Hope this helps,

Marcia

I would suggest using PMSF strating from the lysis buffer and add a commercial protease inhibitor just before loading to the nickel column.

I have added PMSF but still my protein degrading and showing 3 bands other then protein band...... What should I do.....

First, you can now use 5 mM EDTA with His-tag purification and that should stop the metaloproteases (Roche cOmplete His-tag resin, they offer free samples).

Second, a band right below your protein can be due to old SDS/PAGE sample loading buffer, old gels, and not heating the protein sample at >90oC for 3 min. Try making up fresh sample loading buffer, heat your sample up to >90oC, and us a fresh gel (if you make them yourself, make sure it is fully polymerized).

I think it is first important to check if the band beneath is also your protein . I would recommend doing a Mass spec analysis of both the bands. 

The other question is  Can your protein cleave itself ? 

I generally use PMSF in my purification and it helps. You could also try adding excess amounts of one of the components of the cocktail(check catalog). Like adding pefabloc helps in some purifications. 

How many tablets of the cocktail inhibitor do you use ? and how much cell pellet is being processed ? 

If you feel that protein is getting degrades, instead of columns , you do gel extraction. Over express the protein and run the sonicated sample dissolved in PBS and then load it in SDS PAGE. Cut the overexpressed desired protein and reextract using 5% SDS solution. This has been pretty good. If you need, i can give you the detailed protocol. Yield is also pretty good.

Here are some suggestions, but if you could post a picture of a representative gel, that would be very helpful for trouble shooting the issue.

As I mentioned before, you can add EDTA to you protein extraction buffer if you use Roche cOmplete His tag resin, also if you extract the protein in 6M GuHCl, bind to column, wash to remove GuHCl "and" refold the protein on the column, then elute with imidazole, the two combined my get rid of the truncated protein "if it is actually due to a protease".

What is your loading buffer composition? I have had samples that had a doublet that was due to the 4X loading dye/buffer, I now make the 4X loading buffer using Tris HCl, and not Trizma, and pH with NaOH, and along with high purity SDS that solved the problem (read about this known issue online). For some reason the SDS no longer precipitates at 5oC which is the storage temp. I use because it contains BME. That is good because I do not have to warm it up to get the SDS back into solution.