If it was me, I would take what you boiled off the column and run it out on a PAGE to determine whether it was relatively pure or dirty. If it is dirty, you may need to mix up some higher concentration elution buffers and run aliquots consecutively in a step-wise gradient, if it's pretty clean you can make one extra strong elution buffer to clear it out (after you know everything else is washed out and it's stuck), you'll have to run it through a desalting column or dialyze it after. The salt/maltose concentration is not enough to disrupt the MBPs binding interactions with the column, in order to overcome these you can either increase the ionic strength of the buffer, increase the maltose in order to competitively inhibit those binding interactions, or both. [or mess with pH, temperature, etc, don't do any of that]

how are you eluting your protein? For tag like mbp, gst etc. the binding and dissociation kinetics are slow. You might want to incubate your resin with maltose for a while (30 minuets to an hour) to elute your protein. Alternatively, keep the flow rate low.

another alternative is to add his tag besides mbp and purify your protein by NiNTA. It’s a lot easier to work with.

Firstly are you sure that your protein is soluble in the loading buffer, spin at high speed, >20k g then check supernatant? High hydrophobicity of sample may contribute to non specific binding and the high salt will enhance these interactions. You could run buffer with detergent to check

Good suggestion above to leave elution buffer on column for longer

Add His-tag to the MBP. Then you can purify it on Ni-NTA. That is tolerant to aggregation suppresing additives like glycerol and detergents. These additives could then help your protein behave nicely.

Thanks all. A lot of good suggestions to try, luckily I have frozen samples to work with to try out some of the approaches you kind folks have put forward. I'll definitely increase the time in my maltose elution buffer, that hadn't even crossed my mind. Luckily we actually have a His tag upstream of the MBP tag so I'll try running some with an NiNTA column too. I was hesitant to try this as we need to cut the MBP tag with a His-tagged HRV 3C protease, but I could try and purify, desalt, cut and then run an Ni-NTA column to remove the fragment and protease. Ian, is 200 mM considered high salt in this kind of situation? The elution buffer I used was the standard one suggested by NEB, so I guess my protein needs a little optimization. Again, thanks for your suggestions everyone.

Increase the salt to 500mM NaCl. Increase the Maltose to 20mM. Add reducing agents like BME or DTT. Add solubilizing agents like Glycerol. If it has cofactor, like Mg or Ca add chloride salts of the cofactor. What is the PI of the Fusion protein?

Do you have aggregates? Run analytical gel filtration to check. I have seen a few examples of aggregation-prone proteins that were kept in solution when expressed as MBP fusions, but they were still large aggregates. So large that they did not enter the pores in the purification resins and I had very low capacity on the column. A sign of large aggregates is getting significantly larger capacity on resin with smaller beads. Adding 2 M Urea solved the problem for me.

Instead of boiling the resin you can clean it with 0.1% SDS.

When equilibrating the Amylose Agarose, you could have washed the resin with the Elution Buffer and then brought it back to the Loading buffer, This will lock up any site on the matrix that would cause non specific binding.

Probably besides the affinity interaction, you have another kind of interaction between the protein and the matrix. If you say that your protein is composed of 44% hydrophobic residues and you use 200mM NaCl in your elution buffer you could be promoting some kind of unspecific interaction. Have you tried the use of some detergent in your elution buffer?

Your protein has high percentage of hydrophobic residues and it is being routed from column using Tris buffer. You need look at the elation buffer options and also is your protein extract clear or turbid that you are loading on the column.

Hi all, I tried another run yesterday incorporating some of the above mentioned suggestions.

Namely:

0.05% TX-100 in all buffers

100 mM NaCl instead of 200 mM

50 mM HEPES instead of 20 mM Tris

15 mM Maltose instead of 10 mM (and letting it sit in the column for 20 mins before eluting the 1st fraction)

I did increase my yield but not by a huge amount. There is definitely a lot of protein still in the column when I look at it by PAGE/Coomassie.

In parallel, I ran half of my samples through an Ni-NTA column to see if I'd get more protein that way. While I definitely get more of my fusion protein, I get a lot of other bands eluting too. We're hoping to generate a nano body against the protein of interest (using the Kruse lab's yeast nano body library) so it seems much too dirty for us to be able to do a double His purification unfortunately.

Amarjit K Nath, yes my lysate was a little cloudy and not clear which I was a little bit suspect of. I spun it a second time @ 15,000 x g for 20 mins (after the initial spin to remove the debris from lysis )and nothing pelleted at the bottom though.

I really appreciate all of the suggestions folks. Thank you.

Is it absolutely necessary to have MBP tag? How dirty is the protein after NiNTA? often time, you get less cleaner protein after NiNTA and you have to do ion exchange or other purification step to get cleaner protein.

Are you sure your protein is folded properly? I had mixed experience with MBP fusion. Atleast in 2 cases, I was getting huge expression with MBP tag but downside was my protein was not folded properly and was getting aggregated after cutting the MBP tag off. often time, if your protein is not folded properly, you can pull out lots of impurity. can you share gel picture after NiNTA? Whatever you got out of NiNTA, you can try to pass through analytical S200 with and without cutting MBP tag off and see if your protein is folded properly. its worth doing before putting effort in making nano body as making nano body is time consuming and you dont want to invest your time on construct which is not going to work.

When I start a new project I test many conditions to solubilize the protein. I also try different tags on N and/or C terminus. I have also bad experience with MBP. After cleavage I got aggregates of some of my proteins.

For sure I would try detergents as tween 20, chaps, NP-40. Hepes is often better than Tris. I would go first for NiNTA than ion exchange chromatography. Size exclusion chromatography will show the quality of your protein.

Sometimes helps 10-20%glycerol and for NINTA I put in lysis buffer 15-20mM Imidazol to get rid of unspecific bound proteins.

Did you try to express the protein in Hi5 cells? Sometimes I have better expression in Hi5 than in S9 cells.

I wish you good luck.

Does your protein need to be native? If you simply need short epitopes for Ab generation, I would move to a denaturing pipeline. Perform His purification using urea containing buffers: will definitely help with your solubility and recovery problems. Simply remove urea slowly (dialysis) at the end. I would use cobalt based resins which are typically cleaner than nickel based (e.g. Talon resins; see attached). Procedure should be optimized a bit but you should be able to get a pretty clean prep. If you need it cleaner, you may want to end it with an additional chromatography step that will enrich for hydrophobic proteins (e.g. HIC).

Good luck!