there is no "magic" imidazole concentration - depending on the oligomeric state of your protein and the accessibility of the His-tag optimal binding, wash and elution concentration vary. My advice:

- use as little Ni-resin as possible to avoid impurities binding in the first place. It is better to lose some protein that can't bind than get less pure protein in the end. We usually use 0.5 ml resin suspension per litre of bacterial culture.

- do the binding step in batch, not in a column

- then pour in a column and elute with a step gradient, for example: 10, 20, 50, 100, 200, 500, 1000 mM imidazole. Run eluates on a gel and decide which fractions to pool.

Well, various cause for the lower protein binding/yield.

But more specifically, as you say 50mM imidazole is quite higher for binding!! please try within 10-20mM or simply skip imidazole on binding. I dont think you are going for a structure study from this purified protein, so presence of contaminants upto certain levels are agreeable for making assays. Also think about the expression levels of protein (if animal cells, check for optimum amount of DNA to be transfected).

Good luck.

Thank you for your reply. You're right, I'm not going for any structure study. I know transfection conditions has affect on the yield, but it should not have any effect on if the protein is recovered in the flowthrough or not:) But I am looking in to he transfection aswell, in order to increase the yield.

Nver heard of using a conc as high as 50 mM seems more than a bit excessive

Try to keep 10-20mM IMDZ during binding. Anyway, you apply gradient of IMDZ concentrations afterwards, so that impurities would be washed away while IMDZ concentration goes up. The other reason for low binding affinity could be accessibility of HIS-tag since you purify your protein under native conditions. If lowering IMDZ concentrations does not increase binding of your target protein, you may consider to put HIS-tag on the other terminus of your protein.

In general, you may increase NaCl up to 300mM in order to disrupt ionic protein-protein interactions, addition of 5-10% of glycerol may help to disrupt hydrophobic PPIs. I also wonder why you are using pH 7.4. If it is good for your protein, you may consider using TALON-resin or charge your HIS-trap with Co2+ which has optimum binding to HIS-tag at this pH range. It will reduce the amount of impurities bound to the matrix.

Good luck!

I support Ananda Ayyappan considerations.

At first you should focus on protein recovery.

So I would suggest to start with no imidazole in binding buffer (and elute it with 0.5M imidazole) and see if the protein binds and only later focus on increasing its purity by increase concentration imidazole (no more than 20 mM) in the binding buffer.

Poor binding could also result from too high contaminant protein concentration in the loaded sample (if this is the case, try to dilute it before loading).

Also consider that the His-Tag could be poorly exposed to the solvent due to the structure of the protein itself. In this case, the sole solution could be to change the position of the His-Tag (from C-term to N-term or viceversa).

All the best,

Gianluca

To get an idea of the binding, try a comparably big colum with no imidazole and see at which imidazole concentration the protein is eluted. Then you know the amount you might add whilst appying to the column. This will of course give no pure elution, but might be helpful. In a second step try to apply with that amount of imidazole and you should have a quite pure protein.

Another thing is to apply the supernatant more than once prior to elution.

The imidazol content can differ strongly due to template (e.g. membrane protein or not, expression level) and exposure of the tag.

For the optimization Konstantin Byrgazov gave a nice overview.

Anette,

I suggest you follow the below suggestions.

Before binding on Ni column, it is better to avoid Imidazole.

Let the protein bind then the nonspecifically bound proteins can easily be eluted using lower conc. of Imidazole (50-100mM). Elution can be achieved using higher concentrations of imidazole (200-250mM).

Surely do include salt (sodium chloride) in your equilibration, load and wash as well. Incase protein is not binding to the column at all, it means the HIS tag is not accessible, in that case you can use mild chaotrope (200-300mM Urea) in the load.

Hope that helps

Good luck

azim

It's all a matter of balance. The higher the imidazole concentration, the higher purity but probably with a lower yield. In the other hand, if you use low imidazol conc, other proteins interacts unspecifically with the nickel and the puriy will be lower, so you have to dedide your priority.

I've purified some proteins with a similar problem and in fact they eluted at 50-100mM imidazole but I'm pretty sure you would not have loss of yield if you wash at 20mM and you would improve your protein purity.

In any case, in order to minimize the unspecific interactions, you can use a higher conc. of salts (200mM NaCl is OK).

if you use high concentration of imidazole you will get low loading of protein on the IMAC support. Tris strategy is often used in shielding chromatography, but when you are woraking whith complexidad mixtures of proteíns or you need to decrease concentration of low Molecular weight proteíns. I do not know what type of proteín you need to purify. The other aspect that you ned to consider it is type of metal that you have in your chromatographic resin. Nickel is highly selective and you should use low imidazole concentration in orden to get loading of your proteíns. I recommend that you try to purify firstly in batch mode, previously to go with the chromatographic column. Use a little quantities of your support loading proteíin extracts at different imidazole concentraciones and subsequently measuring proteíin adsorbed in the support by Bradford and then characterizing proteíns by SDS-PAGE in both supernatant and support.

I agree with everything has been said previously : increase your NaCl concentration (we go upto 400 mM), decrease til 20 mM the concentration of imidazole in your equilibration/purification buffer, maybe another clue : try to load your supernatant at lower flow. And finally if nothing change the idea of Konstantin might be good : use IMAC with Cobalt as ligand.

Another point check if you have any kind of chelating agent in your culture medium...

Good luck !

Hi! The imidazole concentration in the washing buffer is best at 50mM if your protein binds well. The most probable reason why your protein doesn't do that is that the his-tag is not free and is partially or totally hidden by the protein itself. My suggestions are: 1) Increase your salt concentration 2) try a different metal (e.g.: with a His-Trap Talon column (Co-IMAC) or by loading Co2+ instead of Ni2+ in your resin. 3) Try to fuse your protein with a His-tag on the other side!

If you are seeing additional binding with your flow through then you are either overloading the binding capacity of the column resin or the binding of your His-tagged protein is weak for some reason (His residues not easily accessible in fold) and being out-competed by NS interactions with other cell proteins for the resin. I would consider what Rodrigo mentioned above. If time is less of an issue and you are not dealing with a very large number of different samples, you may want to consider switching to a batch mode that would allow for prolonged incubation with the resin with mild agitation.

You can try at higher pH and salt concentration, decrease the imidazole concentration and add 5% glycerol. You can try this buffer: 0.02M phosphate, 0.25M NaCl, 10 mM imidazole and 5% glycerol. Adjust the final pH to 8.0. If still doesn't binds, try batch method.

For your case, try to use low imidazol concentration 50mM Imidazol. you can wash several times by 20mM imidazol buffer, then use 40mM, 60mM, 80mM and 100mM imidazol buffer to wash and track how pure are your proteins. Then pure protein can be eluted at 500mM imidazol buffer.

If isolating from cells it is in troublesome cases helpful to remove the periplasma preparation before lysis. In your case (isolating from cell supernatant) you can not get rid of the bad stuff that easily, extensive dialysis may help.

Anette,

I have been trying to optimize the purification of my His-tag protein for months now, so let me just start off by saying that I know how frustrating it can be. Nonetheless, little by little you will get there!

I can never attach files on this website for some reason but below is the title of a manual I read and helped me so much in understanding Ni-NTA Resin purification and allowed me to get a lot better results than what I was getting before. The title of the manual is:

A handbook for

high-level expression and purification

of 6xHis-tagged proteins

by Qiagen

I hope you can find it since I can't seem to attach it here.

Short advice to your question though, I would use 20mM immidazole in washing buffer and no higher since you can see your protein in the FT. I would also increase NaCl concentration to 300mM. For elution, I would use 2.5mL of increasing concentrations of immidazole buffer. I used 40mM, 60, 80, 100, 200, and 1000mM, my protein eluting at about 60-80mM. Collect 0.5mL fractions and analyze with SDS-PAGE gel to see where your protein is eluting.

One more thing, how much resin are you using? (What is your column volume?)

I use 0.5mL column volume for 250mL-500mL of cells in LB.

Good luck!

Thanks for your advice and compasion Arianna

I found the handbook from Qiagen. I am now trying to apply my sample in buffer without imidzol and wash with 0 mM imidazol followed by elution with 500 mM imidazol but only 150 mM NaCl. I get less of my protein in flowthrough, but am still not pleased with the outcome. I have some optimizing left to do.

I use 1 ml resin (pree packed column from GE healthcare) because I apply up to 1 L cell supernatant.

I wish you good luck with your purifications aswell.

Co+2 column instead of Ni+2 would give much better His-Tag protein. Possible best binding buffer is 20 mM Tris-HCl, 500 mM NaCl, 25 mM imidazol and pH 8.0 and similar buffer with 300 mM imidazol would elute the bound protein. Look at the His-tag terminal end C or N- which also poses such problems in certain cases.

Based on my experience Co+2 resin give better result (pure protein) compared to Ni+2 resin.

Suggesting to try denature method (urea), because as mention by Marco Salomone-Stagni, the His tag located at the core of the protein.

Hello there,

I do not know if you did solve your issue with the low binding capacity of your protein. I did have the same issue in the past for a protein. In fact, for some proteins, the Tag might be a bit masked due to the folding around it and it thus becomes less accesible for the Ni2+ of your resin..

In that case, do not use GE column (they are prepacked with a definite volume, e.g. 1 or 5 mL) but a resin from Qiagen (e.g. fast flow Ni-resin) that will allow you to pack your own column. You might thus want to increase the column volume (10 to 50 mL) but also use as less imidazole in your lysis buffer as possible (pH 8 is important for binding, 5-10 mM Im, 0.5 M salt [very important to get rid of unspecific interaction] and 50 mM NaPi [or Tris-HCl or Hepes if your protein cannot handle NaPi]). Wash first with 10 col. volume containing 5-10 mM Im, then 10 col. volume (at least) 20 mM then perform a slow gradient from 20 to 250 mM Im (should be around 15-20 col. vol.). This should work as it did for me. If you are impatient and your overexpression is fabulous, just elute directly with 125-150 mM imidazole and perform a second column (monoQ or monoS). We increased by 20/30-fold the yield of this purification step by doing so and the protein was extra-pure.

Hi Anette,

sometimes, a simple ammonium sulfate precipitation (or better: fractionation) before loading on the nickel column does the trick. This step removes low molecular weight compounds from TC medium which compete for the metal ions:

Wu CH, Balasubramanian WR, Ko YP, Hsu G, Chang SE, Prijovich ZM, Chen KC, Roffler SR. A simple method for the production of recombinant proteins from mammalian cells. Biotechnol Appl Biochem. 2004 Oct;40(Pt 2):167-72.

http://www.ncbi.nlm.nih.gov/pubmed/14725509

HTH

Wo

We use zero imidalzole in the inital equilibration and wash buffers (with 50 mM Hepes, pH7.5ish, 500 mM NaCl and 10% glycerol). Sometimes we do another wash with 10 mM imidazle (plus the rest) but almost always we just elute with 250 mM imidazole (in the Hepes etc). It works. Usually.

Hi Anette,

As a first step I would definitely increase the volume of the resin. Non-specific binding of other proteins will mask most of the binding capacity of the resin. Low concentrations of imidazole will decrease or eliminate non-specific binding (only the strongest binding His-tag will bind) but it does not work for all proteins. There are proteins that bind so weakly that will fall off at the slightest addition of imidazole (even 5mM). Another way to optimize binding is buffer exchanging your extract into the column binding buffer. This can be achieved with a large-scale pump if you have liters of volume (like secreted protein) or diluting your protein and adjusting the final salt concentration if you have small starting volumes.

If none of these methods work, perhaps your tag is too weak or not freely accessible. You can create a construct with a stronger His-10x tag and/or introduce additional linker residues between your protein and the His-tag.

It is also worth trying few different pH values for the buffer... Higher pH may be better for binding, but will depend on protein. By the way, what is pI of this protein?

Use 1 mM imidazole for the binding. Make sure the pH of your buffer is correct. I tend to go to pH 8.0 (20 mM Tris). Then wash with 10 mM imidazole. If your protein comes off in the 10 mM wash then wash extensively with 1 mM instead and then elute as a final step.

The affinity of His proteins to bind Ni resin is reduced in Tris buffers when compared to phospahte buffers. As a rule we load proteins in PBS containing 20mM immidazole. The column is washed in 40mM imidazole and then the recombinant protein eluted with 250mM imidazole (hexahis tag).

Don't forget your column regeneration. Column regeneration enhances the column activity.

there is no "magic" imidazole concentration - depending on the oligomeric state of your protein and the accessibility of the His-tag optimal binding, wash and elution concentration vary. My advice:

- use as little Ni-resin as possible to avoid impurities binding in the first place. It is better to lose some protein that can't bind than get less pure protein in the end. We usually use 0.5 ml resin suspension per litre of bacterial culture.

- do the binding step in batch, not in a column

- then pour in a column and elute with a step gradient, for example: 10, 20, 50, 100, 200, 500, 1000 mM imidazole. Run eluates on a gel and decide which fractions to pool.

I took the supernatant to the Ni-Agarose (Quiagen) column mixing for 12 hours at 4ºC. This step allowed me to elute the isolated protein with a buffer of 500 mM Imidazol.

Hi,

wash and equilibrate the column with 5mM Imidazole and elute the protein with 250 - 500mM Imidazole.

NiNTA - Quiagen - perfect - high purity with good quantity.

The answer to this question is always on a case by case basis. I usually use 40mM in the binding and wash buffer for His Trap GE it is what they recommend and years of experience purifying GFP tagged proteins has shown me that nothing comes of in the 40mM wash. I used 0.5M for elution. Your protein may be an exception but it depends on what cells you are using for expression. If you are purifying from insect of mammalian cell culture you will find that the media can interfere with Hist tag binding to the affinity resin. Other things effecting binding include short linker sequences between the His-tag and protein too much reducing agents ie DTT or beta-ME. Also chelators such as EDTA can strip the resin.

Also keep in mind that imidazole is a buffer. You must pH your wash and elution buffers after preparing them. The pH will change if you are using stock buffers and diluting them.

I agree with Mark about your low binding problem, if the imidazole concentration was too high your protein would not bind at all. It could be that you did not use enough resin, but it has a very high binding capacity so unless you are loading 20-40 mg of "His-tagged protein" per mL of resin, that is not the issue. Try doing a batch binding for 1 hr at 4oC with gentle rocking to keep the resin suspended, then gently centrifuge the lysate/resin to collect it and then either wash it in batch mode with wash buffer (but will wash away protease inhibitors too), or resuspend resin in a small amount of lysis buffer with PIs and load onto a column for washing and elution. I have found with Talon resin and my "current" protein that I need to do a batch binding to get good binding. Using more resin than needed only increases non-specific protein binding.

Your protein could bind better at low immidazole concentrations for a number of reasons: large protein/small His-tag, sterically unfavorable for His-tag to interact with resin; His-tag interacting with the protein which blocks His-tag; both issues will be solved by longer incubation time with resin to increase chance it will bind to nickel.

You can increase the salt concentration (I have used up to 300 mM KCl), but I do not think that is the problem.

My 2 cents-- increasing resin volume, increased salt concentration and +/- detergent like CHAPS might help. Good luck.

You should tray pH 8 tris -NaCl buffer and low Molarity immidazol and increase for washing and than for ellution 100 -200-250 mM immidazol frection and then seen in SDS-PAGE .

It is not quite clear to me if you refer to cell culture supernatant with secreted His-tagged protein. If so, the main problem for use of IMAC could be that most cell culture media for eukaryotic cells contain chelators that cause stripping of metal ions from all "conventional" IMAC resins. Stripping will of course give protein-binding problems - low or no binding. The stripping may be difficult to detect as a bleaching of the original column/resin color caused by Nickel or Cobalt ions, if the applied sample cause a dis-coloration by itself, or if using black magnetic IMAC beads. I have not seen the stripping aspect discussed earlier in this thread, but I may have missed it.

(Of course, the low binding could also be due to a problematic His-tagged protein.)

There are now some new IMAC resins on the market that bind Ni ions much more strongly than conventional resins, and do not suffer from stripping problems, even if applying EDTA solutions(!) Working for GE Healthcare, I hope that it is not improper to inform that we have such a new IMAC resin, as a complement to our older resins. (Having worked with IMAC for 10+ years, it would be strange if I would not share my experience.) Similar resins are avialable from at least two other vendors.

Just be aware that the stronger a resin binds nickle, the weaker it will bind a His-tagged protein, and a lower concentration of imidazole will have to be used to prevent non-specific proteins from binding and still allow a His-tagged protein to bind to the resin (since it binds weaker). At least one resin I have looked at (Roche Complete) requires that the binding buffer be at pH 8, with no binding occurring at pH 7.4 in my hands for my His-tagged protein.

Correct, Gary. Affinity for His-tag will decrease with a strongly Ni-binding ligand. So if the main problem is a tricky His-tagged protein, binding may get even poorer with the new resins. Partly, the inherent low affinity for His-tag of a strongly bound Ni may be counteracted with a high ligand density = high amount of ligand and Ni/ml resin.

Thanks Lars for the information, the other thing I have learned about the new resins is that you want to use the minimal amount of resin needed to bind your His-tagged protein, this allows your His-tagged protein to compete off other weakly binding proteins and so reduce the background of non-target proteins. Was counter intuitive for me at first, less resin is better.

Hi Annette,

I get the feeling it is a His tagged protein from cell supernatant, as Lars said.

Many of the previous responses to your question are insightful to finding a solution to your problem.

I agree with low imidazole to prevent non-specific protein binding whilst not being too strong to prevent your own protein binding.

A reason for weak binding might be breakdown of your histag from proteases (expecially if there is a cleavage site) therefore increasing your protease inhibitors can prevent this problem, if this is the problem. The easiest to test this is to perform western blot with anti-His antibody, and also to use an antibody for your protein of interest.

The other issue might be a buried histag, as stated before. If there is a buried tag then the only solution might be to try ‘loosening’ the protein structure under slight denaturing conditions i.e urea. However use this with caution as there is no guarantee your protein of interest will refold back properly and be active (if an enzyme). Otherwise try placing your tag on the other terminus.

Nickel usually produces greater binding for His tags but also more background therefore an idea would be to use Cobalt instead which may reduce nonspecific protein binding. I cannot say for sure whether this would work as it is very much protein dependent, but worth a try if you haven’t already.

Playing around with other additives such as glycerol can help stabilize your protein and help prevent possible protein interactions of your protein of interest with other proteins. NaCl concentration, pH of buffer can also have affects – but these would be final optimisations to your purification set up and tested later after your current problem is resolved.

Best of luck.

First of all 0.02M phospate is very low, Increase it to 50mM at least . When protein bind weekly or elute in less concentration of Imidazole, you can omit it in binding/sonication buffer as it realy helps and wash with the same. You can too use 1M urea as it does not denatures your protein but helps in the access of His tag to the matrix if your protein has a buried his tag.

Increase pH to 8. That worked for us. That should do the job while still keeping Imidazole low. As already said increasing Imidazole to avoid non specific binding will also reduce binding of your protein to the resin.

The secret is to make a stock solution of  1M imidazol that should be pH adjusted to pH 7.5 (or 8.0). This way it will not change the pH of your phosphate buffer when you add it.

The pH of your protein solution should be 8.0-8.5. We use 20 mM imidazole in the binding buffer, 50 mM imidazole in the wash buffer and 250 mM imidazole in the elution buffer.