Dialysis and gel filtration have been mentioned. centrifugal concentrators are a third option. There are advantages and disadvantages to each of these methods.

1) Dialysis is probably the best method, particularly for sample 1. The main question is what size is your protein? "standard" dialysis tubing has a molecular weight cut off (MWCO) of about 8 kDa so it is suitable for most proteins but not if you are working with a small protein, less than 10-15 kDa in size. You can buy dialysis tubing with MWCO of less than 1 kDa if necessary although 3-5 kDa would be sufficient.

You need to think about what volumes you want to dialysis as this determines the width of tubing you need. A few ml will be lost in 10 cm diameter tubing while 100 ml will need too much 1 cm diameter tubing. You may have to buy two widths for small scale work and larger preps.

Another question here about what buffer you want to dialyze into is cost. To efficiently dialyze 100 ml of solution, you'll need to use 2-3 changes of 2-5 liters of buffer solution. When you calculate the cost of 50 g phosphate, Tris and MOPS you will favor phosphate and use it if it is compatible with what you want to do next.

The other big question about dialysis is when to stop. To be honest, I haven't found anyone with a scientific answer to that question. The standard approach is to "dialyze against another change of buffer" or "for a longer time than suggested by the manufacturer". If you have access to a decent NMR spectrometer then you can easily detect millimolar concentrations of buffers remaining in your solutions with just a few scans (note, you're doing small molecule NMR to detect imidazole and don't care if you see protein or not).

One question is about protein stability as it takes >24 hours to efficiently dialyze a protein (preferably in a cold room/fridge)

2) Gel filtration methods are very efficient if you buy a big, long column to attach to your FPLC equipment . The PD10 columns that Cedric mentioned are useful as they are small enough to use on the bench and can work under gravity. I agree they are very useful for small volumes of most proteins.

Note that the exclusion limit is 5 kDa so small proteins under 10 kDa might elute first with the unwanted salt (I have lost samples this way!). So save the first salty fractions as well as the second fraction in the new buffer and analyze by PAGE to make sure everything is OK.

The other note is that if the desalting capacity is 99% efficient, you can expect to have 2.5 mM imidazole left in your sample after PD10 treatment (although the treatment dilutes your sample by 40% so it will actually be ~1.5 mM imidazole). Such a concentration of imidazole should not affect binding to an IMAC column again but might interfere with your enzyme assays. Again, a decent NMR spectrometer will detect this concentration of a small molecule like imidazole with no problem.

You can resuse PD10 columns if you clean them well with buffer, water and then 20% ethanol. Our practice was to use the same PD10 column for only one protein.

3) protein concentrators are useful if you need to make small, concentrated protein samples (like for NMR protein structure studies). You need alot of centrifuge time if you have to use a small MWCO but standard sizes > 8kDa generally only need ~10 minutes to achieve a 10-fold concentration. The idea is to concentrate your protein then add the new buffer. You need to go through several cycles of concentration and dilution in order to make an efficient exchange of buffer. If you make sequential 10-fold concentrations and add the new buffer back to the same level then you can easily calculate the efficiency of buffer exchange (in this example, 90% after the first step, 99%, 99.9% and 99.99% if you repeat 1-3 times more).

I like to use the concentrators that fit in an eppendorf centrifuge and use it in the cold room. It is difficult to match a suitable centrifuge, rotor and length of concentrator for larger centrifugal devices greater than 1 ml. Also, if you are using a small MWCO device and need the centrifuge all day then your colleagues won't be very happy with you. If I have 20 ml to concentrate then I will balance four concentrators in the Eppendorf centrifuge and first top up with protein solution for the first runs before starting the buffer exchange.

There are two main problems with this method. The first is when the filter blocks up and then it stops concentrating (or takes a very long time). It is best to make sure you protein solution is very clean to start with (either filter through 0.22 micron filter or ultracentrifuge). The second is that proteins have a maximum solubility. If you overconcentrate them then the protein can precipitate and this will also block the filters. Protein solubility is the biggest question about selecting this method.

All three approaches are a similar price and easy to implement. I would be cautious about removing all 3M MgCl2 in one step as this might be a shock to the protein. If the protein precipitates then the answer could be to exchange to new buffer with 2M and then 1M MgCl2 added to it before stepping to pure new buffer.

Hi Bornali!

I can just tell you what buffer I use for my his-tagged protein: I changed my buffer to 50mM Tris pH=7.6 for my further in vitro kinase assays.. what I also know is that you can use HEPES. 

Which in vitro assays are you going to use your protein for? I would rather look up in the literature where they performed your assays and dialyse with those buffers they had the proteins inside ;)

Hi there,

I usually go for the same buffer composition minus imidazole of course. But just as Fatma mentioned, it also depends on what downstream experiment you intend to do with the sample...

Dialyze your protien sample in Tris-HCl buffer as Fatma said. I generally dialize my protien in 50 mM Tris-HCl pH 8.0, 25 mM NaCl  in step-wise manner. You can calculate the probable dilution of imidazole in the sample.  I used 2 L buffer for 10-20 ml of sample  and dialyzed 4-6 hours then transferred the sample to another 2 L same dialysis buffer for 4-6 hours and finally in 2 L buffer (for next step of purification such as Gel filtration chromatography) for overnight  (for my protein it was HEPES buffer).

You can try in sample buffer (that you need for further application) if you don't want to follow further purification process.

Best of luck

Thank you all for the suggestions..:)

Thank you Fatma and Mubinur...that sound worthwhile to dialyse in the same buffer i shal be using for downstream assays. I need to proceed with the subsequent assays using HEPES buffer. 

Sincere thanks for the kind suggestions.

If your eluted fraction is less than 3 mL, you can use easy-to-handle PD-10 size exclusion column (e.g. from GE Healthcare) to remove small molecules as imidazole from your purified protein.

"Whatever buffer your protein is happy in" - exactly, we dialyse a recombinant glycosyltransferase in TBS (50 mM Tris-HCl, pH 7.4, 150 mM NaCl) before final concentration and addition of glycerol for long-term storage and downstream enzyme assays.

Dialysis and gel filtration have been mentioned. centrifugal concentrators are a third option. There are advantages and disadvantages to each of these methods.

1) Dialysis is probably the best method, particularly for sample 1. The main question is what size is your protein? "standard" dialysis tubing has a molecular weight cut off (MWCO) of about 8 kDa so it is suitable for most proteins but not if you are working with a small protein, less than 10-15 kDa in size. You can buy dialysis tubing with MWCO of less than 1 kDa if necessary although 3-5 kDa would be sufficient.

You need to think about what volumes you want to dialysis as this determines the width of tubing you need. A few ml will be lost in 10 cm diameter tubing while 100 ml will need too much 1 cm diameter tubing. You may have to buy two widths for small scale work and larger preps.

Another question here about what buffer you want to dialyze into is cost. To efficiently dialyze 100 ml of solution, you'll need to use 2-3 changes of 2-5 liters of buffer solution. When you calculate the cost of 50 g phosphate, Tris and MOPS you will favor phosphate and use it if it is compatible with what you want to do next.

The other big question about dialysis is when to stop. To be honest, I haven't found anyone with a scientific answer to that question. The standard approach is to "dialyze against another change of buffer" or "for a longer time than suggested by the manufacturer". If you have access to a decent NMR spectrometer then you can easily detect millimolar concentrations of buffers remaining in your solutions with just a few scans (note, you're doing small molecule NMR to detect imidazole and don't care if you see protein or not).

One question is about protein stability as it takes >24 hours to efficiently dialyze a protein (preferably in a cold room/fridge)

2) Gel filtration methods are very efficient if you buy a big, long column to attach to your FPLC equipment . The PD10 columns that Cedric mentioned are useful as they are small enough to use on the bench and can work under gravity. I agree they are very useful for small volumes of most proteins.

Note that the exclusion limit is 5 kDa so small proteins under 10 kDa might elute first with the unwanted salt (I have lost samples this way!). So save the first salty fractions as well as the second fraction in the new buffer and analyze by PAGE to make sure everything is OK.

The other note is that if the desalting capacity is 99% efficient, you can expect to have 2.5 mM imidazole left in your sample after PD10 treatment (although the treatment dilutes your sample by 40% so it will actually be ~1.5 mM imidazole). Such a concentration of imidazole should not affect binding to an IMAC column again but might interfere with your enzyme assays. Again, a decent NMR spectrometer will detect this concentration of a small molecule like imidazole with no problem.

You can resuse PD10 columns if you clean them well with buffer, water and then 20% ethanol. Our practice was to use the same PD10 column for only one protein.

3) protein concentrators are useful if you need to make small, concentrated protein samples (like for NMR protein structure studies). You need alot of centrifuge time if you have to use a small MWCO but standard sizes > 8kDa generally only need ~10 minutes to achieve a 10-fold concentration. The idea is to concentrate your protein then add the new buffer. You need to go through several cycles of concentration and dilution in order to make an efficient exchange of buffer. If you make sequential 10-fold concentrations and add the new buffer back to the same level then you can easily calculate the efficiency of buffer exchange (in this example, 90% after the first step, 99%, 99.9% and 99.99% if you repeat 1-3 times more).

I like to use the concentrators that fit in an eppendorf centrifuge and use it in the cold room. It is difficult to match a suitable centrifuge, rotor and length of concentrator for larger centrifugal devices greater than 1 ml. Also, if you are using a small MWCO device and need the centrifuge all day then your colleagues won't be very happy with you. If I have 20 ml to concentrate then I will balance four concentrators in the Eppendorf centrifuge and first top up with protein solution for the first runs before starting the buffer exchange.

There are two main problems with this method. The first is when the filter blocks up and then it stops concentrating (or takes a very long time). It is best to make sure you protein solution is very clean to start with (either filter through 0.22 micron filter or ultracentrifuge). The second is that proteins have a maximum solubility. If you overconcentrate them then the protein can precipitate and this will also block the filters. Protein solubility is the biggest question about selecting this method.

All three approaches are a similar price and easy to implement. I would be cautious about removing all 3M MgCl2 in one step as this might be a shock to the protein. If the protein precipitates then the answer could be to exchange to new buffer with 2M and then 1M MgCl2 added to it before stepping to pure new buffer.

Hi,

If your protein is stable at lower pH. you can try to elute it at acidic pH from your His-trap column without using imidazole. Try in the range pH 3-5 and adjust pH to neutral  after elution.

Hi,

that depends on your sample size.

We have been using as a routine the Penefsky method for desalting small volume samples. You can desalt 500 microliters per columnin a matter of minutes.(Penefsky, H.S., 1977. Reversible binding of Pi by beef heart mitochondrial adenosine triphosphatase. J Biol Chem 252, 2891-2899.) If this is your case and you have any doubts regarding the method you can contact me whenever you want.

Higher volumes will require the columns mentioned above or one you can prepare yourself with Sephadex G50 or G25, depending on the size of your protein. The higher the G number the faster you´ll go. A pressurized membrane concentartor will do the job as well. I prefer any of these methods to dyalisis.

Good luck!