Annette,

When I have to concentrate protein from my purifications it usually requires me to spin them for a long time as I have large volumes of protein from several liters of growth. It can be time consuming, but I spin my columns for just 2.5 mins to start so I can see about how fast the solution is flowing through the filter. Then from that I determine if I can go up to 5 mins or 10 mins. After each spin I 1) insert a 1000ul pipet tip into the top of the concreted protein and pull up 1ml and then stick the tip gentle down into the membrane area (be careful not to touch the membrane too much or too hard as you don't want to tear a hole in it) and then mix the solution that is in the top of the concentrator. You will be able to see your more viscous protein solution close to the membrane swirl up and diffuse into the less concentrated solution. I then mix it 2-3 times. 2) I then add back in the volume that went through the filter to keep my protein from being too concentrated and spin for another 5 mins (you have to judge the timing based on how concentrated you see the solution that is close to the membrane). If you keep doing this "washing" of the membrane during the concentration process you will prevent our protein concentration from getting too high too fast which may cause it to start to precipitate and it keeps it from building-up and blocking the membrane so the flow through the membrane stays high.

I realize it is a pain to have to stop and start spins every 2.5-5 mins, but if you are just starting to work with a protein and don't know a lot about it then it is better to be safe then sorry. It takes a lot of time and money to get to this point so you don't want to loose any of your hard work nor do you want to waste all the money spent in supplies. I have worked with some pretty picky proteins and I have found doing this slowly, mixing after each short spin and then adding more protein solution to be concentrated until it is full again works really well. This would also help if like Alvaro stated your protein has "a very high hydrophobicity and a tendency to form non-specific binding interactions with the membrane material". You should always mix the solution in the top of the concentrator between spins regardless to keep your protein happy.

Hi Anette,

I am using these device, there could be several possibilities

1. Protein has the tendency to aggregate at higher concentration

2. pH of the buffer is near pI of the protein

3. protein was unfolded

I do not know whether these membranes absorb protein

if the protein is unstable u may try some addictive.

protein precipitation is a common problem, even some stable proteins tend to precipitate when concentrated after a certain concentration.

best luck with experiment

Hi Anette

I am also using the Amicon ultracentrifugal device and had the same experience. In my case i am loosing 35-40% protein. this must be due to adsorption.

To minimize the loose you can increase the equilibration time of membrane.

Good luck

I am using Amicon Ultrafiltration Device instead of centrifugal device. And I obtain 100% recovery with this. Try that, it is faster and gives you better concentration. You need to use proper cut off membrane for the same.

good luck :)

Hi Satabdi

I dot think there is any difference between Ultrafiltration Device and centrifugal device. if you know, please let me know...

Hi Anette,

Unspecific adsorption of proteins to these kind of devices is a well known phenomenon. Although most companies have tried to reduce these losses to a minimum a certain percentage of your protein has to be "sacrificed" , this is the price to pay. But you can try to reduce it to a minimum:

- if you protein is too diluted, try to add a carrier (eg BSA) to minimize the loss, at least if your downstream application is OK with this

- keep you salt concentration close to physiological conditions

- be aware that some proteins are sticky per se, so you have to admit some losses

Depending of you applications some precipitation methods (ammoniun sulfate, PEG, ...) could be useful when handling large volumes.

Good luck.

Hi Gaurav

I agree the mechanism in both the cases are the same, but generally centrifugal apparatus face problems such as loss of protein. I have used both the devices and I use pressure based ultrafiltration for concentration.

:)

Hi, Anette!

I have had the same issues many years ago with one of my proteins. After much work, we discovered that it was caused by a very high hydrophobicity and a tendency to form non-specific binding interactions with the membrane material. Of course, the protein tended to precipitate when not stabilized, as Amit has pointed out.

I would recommend altering the pH of your buffer, and/or using some additive. The ones I recommend are those that decrease non-specific binding and stabilize proteins in their soluble folded state. We had good results with the inclusion of BSA in the mixture, but that of course carries the disadvantage of contaminating your protein with proteinaceous material. Other useful additives could be the use of some detergents, such as the Triton series, to inhibit non-specific hydrophobic interactions, or testing the effect of the presence or absence of divalent or multivalent cations. Unfortunately, you have to test whether any of these is relevant for your protein.

I would also recommend that you carry out your protein concentration as fast as possible, avoiding waiting times, and you could also try performing it at different temperatures. Temperature influences the precipitation kinetics of proteins, the entropic and enthalpic components of interactions and also the viscosity of water, which is determinant for the filtration speed. Just to point out a disadvantage of speeding up, if the ultrafiltration process is too fast, it could lead to polarization of the concentration of the protein around the membrane, which could lead to unfolding and increase of non-specific protein-protein and protein-membrane interactions.

Again, you have to test different conditions, but given that you only had the problem when concentrating larger volumes in larger times, I think it is worth a try to speed up the process in some way or to concentrate only small batches at a time, using temperature and additives as helpers to achieve this.

I hope this helps you to set up new experiments and solve this issue.

Good luck with your experiments!

I have used many concentration devices (centrifuge forcefield, filtration through applying N2 pressure like Amicon Ultrafiltration Device which are quite good systems by the way). What I can say is that it always depends on the protein. Some proteins like, other don't. If you want to concentrate you have not many choices. Filtration when it works. I have experienced better results for fragile proteins by adding glycerol (up to 20%) but it takes a very long time to filter because of viscosity. Another interesting way is to feel your dialysis tubing with the protein solution and lay it down on a succrose layer (PEG is good also but precipitation problems may occure and it is quite difficult to remove PEG if required). Sucrose quite good in fact. Protein solution concentrated in sucrose may be stored at 4°C with azide or -20°C also. Do not attempt to remove succrose from your concentrated protein by dialysis cause it may precipitate. When you need some of your protein just pick up the volume you want and disolve it in the appropriate buffer (a little bite of sucrose still remains of course but your diluted protein may be dialysed.

Thanks for all good hinst and tips. Fells good to know that I am not the only one encoutering this problem.

Hi Annette,

you should NEVER use spin concentrators and centrifuge consecutively for long periods of time, because the local concentration at the end part where the filter membrane is located increases strongly over time, especially if you don't mix the sample. There just isn't enough time for your macromolecules to diffuse into equilibrium. If the local concentration gets too high, you protein will form aggregates which may precipitate.

I frequently stop a concentration process and mix the concentrate gently using a pipette without introducing air or creating foam, usually every 15 minutes, but this may vary depending on the circumstances. I also like to use concentrators that cannot take my complete sample because I think the smaller the filter part is the lesser the unspecific binding of sample to the filter membrane will be. In case of using such setup, e.g. Concentrating a 15ml sample in a 5 ml concentrator, I usually start concentrating to 50% of the concentrator's volume and then pipetting the concentrate back into the tube that contains the rest of the sample and mixing its content before refilling the concentrator.

When it comes to concentrating proteins, speed isn't everything, it's rather essential not to overconcentrate the sample. The solubility of protein depends on many properties, including its hydrophobicity and isoelectric point so choose your buffer accordingly, never concentrate it in a buffer where the pH equals its isoelectric point, which you can estimate with Expasy's protparam program,etc. (google). Also, use a sufficient amount of salt and for most intracellular proteins, a reducing agent (DTT, beta-mercaptoethanol).

A slow down of the concentrating speed may indicate aggregation. Aggregates usually do not have biological activity any more. If your sample turns bright yellowish this may also indicate the formation of soluble aggregates which usually do not disassociate any more. Frequently check your sample during concentration with a photometer, ideally a nanodrop. If you see an increase at 320-350nm this is usually indicative of light-diffraction by aggregated particles. If you see this STOP concentrating.

If you use detergents you frequently can usually not use a photometer to determine protein concentrations at 280 nm because many absorb light strongly at this wavelength. I try to avoid using them as they also tend to keep protein aggregates solubilized and they concentrate as well: A 0.5% Triton-X-100 solution will be a 10% solution after a 20-fold concentration as the micelles are typically larger than the pores in the concentrators. If you have to use Triton-X-100 plan accordingly, e.g starting with a 0.025% solution in that case to end up with a 0.5% solution. Also consider using reduced Triton-X-100, it has similar properties but its light absortion is much less at 280nm. To check for aggregates you can run a gel filtration column with parts of your sample. Optimize the concentration process for each protein you deal with and take notes. Check biological activity as well. Good luck, concentrating proteins can be very cumbersome.

To minimize protein adsorption, you should use PES membranes. I currently favor the VivaSpin products, because they have a much larger filter surface area and spin down more quickly than the Amicons. You can also minimize adsorption by maintaining a modest ionic strength in your buffer. Even 100 mM NaCl will often be helpful for both protein solubility and adsorption suppression.

An old, but still good method to concentrate proteins is to put them in a dyalisis tube, insert it in a flask and then apply vacuum inside. Water is removed from vacuum. You need of course to take care of your sample during time to avoid precipitation.

Annette,

When I have to concentrate protein from my purifications it usually requires me to spin them for a long time as I have large volumes of protein from several liters of growth. It can be time consuming, but I spin my columns for just 2.5 mins to start so I can see about how fast the solution is flowing through the filter. Then from that I determine if I can go up to 5 mins or 10 mins. After each spin I 1) insert a 1000ul pipet tip into the top of the concreted protein and pull up 1ml and then stick the tip gentle down into the membrane area (be careful not to touch the membrane too much or too hard as you don't want to tear a hole in it) and then mix the solution that is in the top of the concentrator. You will be able to see your more viscous protein solution close to the membrane swirl up and diffuse into the less concentrated solution. I then mix it 2-3 times. 2) I then add back in the volume that went through the filter to keep my protein from being too concentrated and spin for another 5 mins (you have to judge the timing based on how concentrated you see the solution that is close to the membrane). If you keep doing this "washing" of the membrane during the concentration process you will prevent our protein concentration from getting too high too fast which may cause it to start to precipitate and it keeps it from building-up and blocking the membrane so the flow through the membrane stays high.

I realize it is a pain to have to stop and start spins every 2.5-5 mins, but if you are just starting to work with a protein and don't know a lot about it then it is better to be safe then sorry. It takes a lot of time and money to get to this point so you don't want to loose any of your hard work nor do you want to waste all the money spent in supplies. I have worked with some pretty picky proteins and I have found doing this slowly, mixing after each short spin and then adding more protein solution to be concentrated until it is full again works really well. This would also help if like Alvaro stated your protein has "a very high hydrophobicity and a tendency to form non-specific binding interactions with the membrane material". You should always mix the solution in the top of the concentrator between spins regardless to keep your protein happy.

when concentrate samples

1) centrifuge the sample at high rpm before concentrate

2)interrupt the centrifugation at 15 minutes (if done for long time) and mix gently

3)use sartorius (vivaspin ) centrigugal concentrators

5) choose the appropriate MW cut off

6) If the eluate from a column is too large, use an affintiy step

to capture your protein (usually affinity resins will work at much less resin volumes, and thus less elution volumes, eg; His resin) and then concentrate.

2hours is a long time to spin on one concentrating device. It just increases the chances of your protein being adsorbed onto the membrane. I recommend changing the concentrator for every 30 min or 1 hour (at most) spin. If any of the suggested additives to your storage buffer are not compatible with your protein just spin for less longer and change concentrators for every 30 minute spin.

If you have very large volumes of protein (e.g., >60-70 mL), then you would might be well served by another concentration technique. For example, ammonium sulfate precipitation, or concentration by ion-exchange or hydrophobic interaction chromatography with step-gradient elution might be an efficient way to both polish and concentrate your protein to a reasonable volume for final concentration in an ultrafilter. I generally don't like extended centrifugal ultrafiltration as it increases the chances of filter clogging or protein adsorption.

I let the Amicon ultra filter incubate in PBS a repeated number of times and measured the asborbance at 280 nm in the PBS. I changed PBS untill the absorbance decreased. I thereafter pooled and concetrated the fractions of PBS. From this I obtained 10% "extra" protein. It is a bit frustrating and astonishing that 10% of the protein can have been absorbed by the filter and then "leaked" out of it again.

One way to avoid generating a protein gradient during concentration is to use a Amicon stirred cell concentrator (http://www.millipore.com/catalogue/module/c3259). It uses nitrogen gas to force your sample against a membrane while a stir bar sweeps the surface of the membrane preventing a protein gradient from forming and that prevents your protein from precipitating onto the membrane. The unit can be put in an ice bath to keep your sample cold (or in the cold room), and you can easily monitor the concentration of the sample through the clear container. A very fast way to concentrate a protein, so calculate the initial time it takes 1-5 mL of liquid to pass through the membrane and keep an eye on it! I have used it to concentrate proteins to 8-10 mg/mL with no precipitation problems.

Concentration of high volumes should never be performed on spin concentrator, the best option is to use a Tangential field concentrator such as Vivaflow 50 or Vivaflow 200 with the right membrane cutoff, the loss is much yield.

Need article "Protein concentration using Amicon Ultra spin column, adsorption on membrane" for reference

Hi Anette,

Centrifuge-based concentrators will result in protein gradients. I avoid centrifuge-based concentrators as much as possible. I'm with Marco Sette and Pius S. Padayatti on this one. Use a stirred-cell concentrator if you have access to one or an affinity purification step.

My favourite technique of protein concentration and something that hasn't been mentioned in this discussion is freeze-drying your protein and resuspending in a smaller volume (of water since the buffer components also remain). This sounds unusual but most proteins survive freeze-drying just fine in my experience. They retain their fold - the 15N HSQC spectra of most proteins look the same after freeze-drying. It also means that you can resuspend them at their solubility limit.

To test: Freeze-dry 100 uL of protein and resuspend in 10 uL. See if it is 10 times more concentrated by A280 nm.

Cheers,

Soumya

Hi Arshad,

tell me, I dont' make the point. How can you concentrate by dialysis ?

Unless of course your dialysis medium is highly concentrated in sucrose or glycerol or whatever (see below).

Although we have been successfull in the past with diaflow under nitrogen (which, like Gary, I also highly recommand), this no longer work since we now deal with very delicat proteins. We have had success by laying down the dilyisis tubing over a bed of solid sucrose. Concentration factor can be up to 5 time. (You might then store the protein at -20°C as succrose is a good preserver of proteins for long storage at low temp). Of course the next step is to perform a dialysis of your concentrateted protein against the buffer of choice to remove sucrose (this again dilute the protein a bit but not that much (a factor 2; result is 2.5 time concentration). Using this process, we usually lose from 20 to 50 % of the proteins but concentration obtained (up to 2.5 mg/mL) allow many experiments.

Hi everyone!

Just to add to the former answers, in my lab when we want to concentrate a large volume of protein without increasing the amount of solutes in the medium, we just use a dialysis bag immersed in a concentrated solution of high molecular weight PEG, or we even leave the bag directly in contact with solid PEG overnight (if buffer concentration is not an issue). This treatment is not as harsh as ultrafiltration with pressure and stirring and gives good results. Another advantage is that PEG does not diffuse through suitably chosen dialysis membranes, so it does not contaminate your protein solution. Always check the material of your membranes in order to prevent any unintended unspecific interaction.

Sometimes, we also usually use glycerol, sucrose or trehalose (much more expensive!) and let them diffuse into the protein buffer for the same reasons Thierry explained before. If pH or ions are a concern for the stability of your protein, I also recommend to use a PEG solution in a suitable buffer with the required counter ions.

Finally, I also recommend using a buffer pH at least 1-1.5 unit above or below your protein isoelectric point in order to minimize the possibility of molecular aggregation during purification.

BW,

Alvaro

the safest way and cheapest way for concentration of large volumes is by putting the solution you want to concentrate inside a dialysis tube with suitable cut off size then you connect the top of the tube to a Buchnner funnel whicht then is inserted inside the rubber stopper and then inserted into the Buchnner flask ,by doing this you will have a close system which when you connect it to a pump (even a small water pump which we always use in  the organic laboratory  ,you will see that when the vacuum will start to  built inside the flask the dialysis tube will start to swell and when it reaches its maximum level ( as judge by your eye ) you now need to put  a small clump ( like that we usually use for the column cinematography system) to clump the tubing that is connecting to the water pump to keep the vacuum in side the flask ,after a while you will see that the water carrying  the small molecular weight substances will diffuse out of the dialysis tube and you can keep the closed system in refrigerator until you get the volume you need and by this you concentrate the protein solution. You can find a draw explaining this method in a book entitled An introduction to practical biochemistry by David T. Plummer.Remind you you have to use a clean flask & funnel so if for any thing go wrong you will not lose your sample.I have used this method several time & i think it is excellent for large volume solutions.  Of course you can use  the dialysis tubes immersed in sucrose or sephedax or polyethylene glycol mentioned above ,but this will be more expensive methods.

i am also having a similar problem with amicon concentrators my protein molecular weight is 13.4kda i use a 3 k cut off filter but there is protein loss when the volume to be concentrated is more please suggest an alternate dialysis causes my protein to aggregate so i dont want to take that risk

The Amicon Ultra series of centrifugal concentrators have the filters on the sides, instead of the bottom, and work great with no significant loss of protein. I just used Amicon Ultra-4; 4mL capacity; 10k NMWL. 

I tried using 15ml centrifugal concentrators for just 1ml of purified protein which i need to remove the imidazole, and also do a buffer exchange to PBS. However, the protein was loss at more than 50% and when I "rinse" the filters again with PBS, there are still detectable protein... is it because the protein aggregate or bind to the membrane, leading to the low recovery rate?

or it is not recommended to use such a large filter column for small volume of sample? Would a small size (2ml) desalting column, protein concentrator or dialysis be more appropriate and helps to improve recovery? Thanks !

I am curious if anyone has tried the vacuum dialysis methods described in the thread.   I have been trying to set this up but the dialysis tubing immediately expands and pops the moment I apply vacuum.   Can anyone point me to a link/reference with a image or video of the method.   Many thanks

Hi Gaurav,

Can you please explain what do you mean by adjusting the equilibration time of membrane???

I have used the vacuum dialysis methods,& found it very useful especially if you have large volumes to be concentrated.You have to be careful with two things The first one is to use a funnel with very smooth ends so it will not cause  any damage in the dialysis tubes when you pushed into the tubes.The other important thing is when you apply the vacuum ,you have to do that carefully ,you should watch the tubes & let it to expand  till it is about to pops ,at this moment you have to switch the vacuum of of course after you clamp the tubes tightly to keep the build up vacuum in.then you can simply can put the whole system in the frig. and watch it between time to time.

You can have a look at the method in" An introduction to practical biochemistry" By Plummer

At what speed does one centrifuge protein sample in Amicon filter? Is there any relation between the membrane cut-off and the speed selection? For instance, concentrating 50 kDa protein using 15KDa cut-off filters and on other hand using 30 kDa filters?

prefill your ultrafiltration device with glycerol 20% for 24 h, then rince. This will minimize protein loss for very diluted samples

Good question… Please check references below.. hope it will be helpful for you…

Regards…

http://www.ictsl.net/downloads/proteinsapplicationguidepurificationconcentrat.pdf