How can I prevent protein aggregation during Ultrafiltration?

I was concentrating my protein in ultrafiltration amicon cell and I found aggregation of my protein which could decrease the overall yield of it after purification. Can anyone suggest how I can avoid this problem?


0 / 0  ·  30 Answers  ·  8340 Views

Popular Answers

All Answers (30)

  • Gaurav Bhardwaj · University of Greifswald
    Several times i also used the ultra-filtration amicon cell and never found this problem. what is the your protein sample and in which buffer it is reconstituted?
  • Xavier Kubiak · University of Copenhagen
    I had this problem more than once with ultrafiltration amicon cells and it always appeared at the same exact concentration (solubility limit) for my protein samples.

    It may depend on several different factors starting from the type of your protein (membrane, cytosolic ?)
    But I would start by doing a buffer exchange against several different buffer conditions (pH, salt concentrations, and ultimately type of buffering agent (TrisHCl, phosphate,...) if nothing imprives the solubility of your protein.
    You may also add some gycerol into the protein (5-20 %), but take care to your protein activity if it's en enzyme because it may abbrogate the activity.
    Some also add low concentrations of urea to avoid intramolecular contacts and thus aggregation to occur but I never tried that approach.
  • Myrto-Panagiota Zacharof · Swansea University
    What kind of membrane you are using in the cell? Wht's roughly to size of the protein?also do you stir the cell?might be an issue of the electric potential of the membrane , and stirring defineteluy will cause agreggation and foaming.
  • For prevent aggregation or precipitation,you must add to solution sodium chloride or other salts during Uf treat
  • Birendra Singh · Lund University
    If your protein is preciitating when you concentrate it, means it need to optimize the buffer conditions, pH, salts or other additives. You can simply take 4-5 different buffers such as tris, HEPES, MOPS, MES, PO4 and set pH accordingly. after dilysing your protein in those buffer and lower protein concentration, concentrate them in amicon. during concentrating just mix protein solution by pipetting up and down and see if it is precipitating! if it precipitates dont proceed. Now you can add NaCl to it if still precipitation happens. However addition of 100-200 mM NaCl is always good to add in protein solutions. If your protein is hydrophobic, such as membrane proteins, then you have to use detergents.
  • Basir Ahmad · Centre for Excellence in Basic Sciences
    Some proteins are more prone to aggregation at high concentrations. If this is the case you should not concentrate beyond a certain limit. One solution is engineering your solvent as suggest by Birendra Sing.
    Designing a solvent condition without knowing the nature of protein is very difficult. But one suggestion is that always keep your protein at pH=PI+_1. Because at PI (Isoelectric point) almost all proteins show high tendency of precipitation.
  • Anatoliy Zhivotchenko · Biovectra
    At high concentration any protein can precipitate therefore you should estimate what concentration you have reached and starting from this point you can continue UF as UF/DF to change buffer or remove the MW cut off impurities.
  • Ananda Ayyappan Jaguva Vasudevan · Heinrich-Heine-Universität Düsseldorf
    I had the same, but in my case the visible aggregation started from the process of elution. And I got the idea of reducing the NaCl concentration to 50mM during elution from a Prof. and it helped to solve this issue. Give it a try.
  • Archana Kumari · Indian Institute of Technology Kanpur
    Thank u all for replying me, i was using Amicon YM3 kDa and size of my protein is roughly 12 KDa.protein was already in KCl because i was concentrating the active fractions by looking enzyme activity after ion exhange chromatography.I had used 50mM MES buffer pH6.8 with 5% glycerol as a additive.
    and during UF it was in proper stirring condition.
    can anyone tell me what should be concentration limit of protein after UF followed by DF ?
    and is this possible to recover aggregated protein into its active enzymatic form.
    can i dissolve in some buffer n additive to get recovery back.
  • Mahendra Vikram Rajawat · Indian Agricultural Research Institute
    Yes, if your protein is not unfolded it should be dissolved. Try suitable buffer to dissolve it and check the activity.
  • Matthew Lalonde · University of Utah
    Unless someone here has worked with the same exact protein, no one will be able to tell you it's limit of solubility. Every protein is different, so protein solubility cannot be reliably predicted.

    I think your question is one of general solubility. You need to provide conditions in which the protein is more soluble.

    Whenever I want to increase my protein solubility, I found it most productive to optimize buffer conditions (as Birendra suggested). Before getting into glycerol and detergents, I like to start simple with three conditions: buffer pH, salt and reducing reagent. I noticed you didn't mention a reducing agent - you could start there. You might try DTT or BME or TCEP (making sure to re-adjust your pH after adding TCEP). Then you could try higher and lower pH and salt. You may find that, with optimal pH, your enzyme is 10-100 fold more soluble and it has much more specific activity than under non-optimal conditions. Once you get those conditions down, you might get into glycerol and detergents, or you might find that you don't need them.

    As for recovery of enzymatic activity after aggregation - that's always a long shot. There are refolding screens you can do, but my lab (which is a hard core protein purification lab) saves refolding as a last resort. Take my advice and optimize your buffers first because, even if you CAN refold your protein, you're again faced with the challenge of keeping it soluble.
  • Alajos Bérczi · Hungarian Academy of Sciences
    To expect more useful information we would need more specific information of (1) the protein in question (soluble? integral? pI value?) and (2) of the IEC (cation exchange or anion exchange? elution conc. of salt? purity fo the fraction?) you performed before UF. All prevoius suggestions are general but very useful suggestions.
  • Sharoar MG · University of Rajshahi, Bangladesh/Cleveland Clinic Lerner Research Institute, Cleveland, OH, USA
    I completely agree with Matthew Lalonde comments and suggestions.Also I want to add the importance of temperature. If you are keeping your protein at room temperature during concentration, keep some ice around the Amicon filtering instrument or perform it at 4C. Though pH 6.8 are used for many enzymes, some needs little higher pH. I used ~7.4 for most of the caspases. Also you need to use more reducing agent then normal during concentration if it needs longer time to concentrate .
  • Justin Ludeman · Monash University (Australia)
    Everything that has been suggested here is excellent. One thing I have noticed myself when using centrifugal concentration devices is that the protein becomes locally very highly concentrated at the membrane interface. This is not because the protein itself is being pelleted but rather is due to the column of solvent being pushed through the membrane leading to local high concentration at the filtration interface, the protein "left behind" so to speak and it is at these very high local concentrations that aggregation can be more pronounced. Try using shorter spins and gentle inversion/pasteur pipetting to redistribute the protein to equal concentration between spins. Analytical size exclusion analysis will tell you if you have soluble aggregate in your sample. Unfortunately, if you do indeed have aggregate present it will only act as a seed for further concentration dependent aggregation. Remove particulate matter by 0.2 micron filters and you can prevent membrane clogging and pronounced seeding effect mentioned above. Furthermore I sometimes find the pressure driven stirred amicon cells work really well for proteins that are prone to aggregation (which can be an intrinsic property of some proteins even if they are correctly folded) as the concentration can be perfomed more slowly. The use of reducing agents or redox couples can also reduce aggregation due to unfavourable reactions between unpaired cysteinyl groups as well and can even lead to productive "reshuffling" of disulphides to their correct pairings. Again analytical SEC can tell you if this treatment leads to greater monodispersion of your protein, at this point you may choose further preparative SEC prior to concentrating your resolved monomer or LMW peak/tailing shoulder. Careful inspection of reducing vs non-reducing SDS-PAGE gels can tell you quickly if abberrant disulphide-mediated aggregation is an issue. Finally, if you are doing buffer exchange concominantly with concentration try not to dilute your protein in vast increments as this sudden change in aqueous environment (pH, [salt], other chemicals present) can "shock" your protein and lead to your protein doing weird things, be gentle with your dilution into the new buffer and you may find you get less aggregation/precipitation.
  • Richard Heath · St. Jude Children's Research Hospital
    My general rule of thumb here is if it tends to precipitate, don't concentrate it in a stirred cell! Spin concentrators seem to work a bit better for some proteins, if you don't have too much volume. Just watch out for the solubility limits, as mentioned. Some proteins can go to 100 mg/ml or more in just about any buffer. Others might crash out at 1 mg/ml even in the most optimal buffer.

    You say it has just come off an IEX column - if it isn't concentrated enough at this point, you could try a smaller column, or shorted your gradient (or use step elution). Or add another concentrative step like HIC, or a different IEX (on a small column). If concentrating is vital, and a stirred cell is the only way you can do it, you will have to optimize the buffer, using the suggestions given. But this might only help a bit! Sometimes, you just have to go with what you can get.
  • Archana Kumari · Indian Institute of Technology Kanpur
    @ matt- -thanx 4 ur suggestion i had added DTT (0.5mM ) during IEX. i will try to optimize buffer.
  • Archana Kumari · Indian Institute of Technology Kanpur
    @justin Ludeman- It cud the reason behind aggregation which i missed, i dont no what should be rpm for concentrating protein in ultrafiltration cell because i hv observed foaming of it if i speed up rotation.
    after one step i hd filtered it through 0.2um membrane but on buffer exchage this again aggregates.
    so i hv 2 look at buffer 4 avoiding this problem.

  • Justin Ludeman · Monash University (Australia)
    Richard Heath makes a fair point as stirred cells can be deleterious if incorrectly used. Try low gas pressure and slower rotor rpm (also nitrogen is better than air or CO2 when disulphide exchange is to be minimised as dissolved O2 and divalent cations can perturb cysteinyl redox equilibria where disulphide formation is necessary for correct fold and function and CO2 can challenge acid-base equilibria and therefore pH) to minimise bubbling.
    Your protein solution can easily be degassed at a later stage if necessary.

    One problem is that stabilising a protein can mean stabilising the aggregate already present preferentially over the active - say monomeric - form of the protein. Moreover, some stabilisers or additives may promote the formation of aggregate so be careful!!! - as always it depends on your specific protein. Always confirm the homogeneity of your protein preparation by analytical SEC to confirm that you do not have species eluting in the void volume of the system employed indicating soluble aggregate. You may find it useful to conduct your own monodispersion expt, ie at what concentration does monomeric/LMW protein become unstable - form aggregate - and can it be reversed by dilution. If the answer is no to dilution then other additives may be useful to try and stabilise the monomeric form, as long as they don't interfere with downstream processes. All this dribble assumes you have sufficiently pure protein in the first place to assess homogeneity over other proteins in your mixture.

    Demonstrate monomer ---> concentrate & SEC assay for monomer %.
    If aggregation apparent dilute ---> concentrate & SEC assay for monomer %
    If aggregation irreversible (which it generally is compared to biologically relevant oligomerisation) take new starting material and concentrate less by different techniques and keep repeating this procedure with different stabilisers etc (detergents, salt, Arg, TMAO, sucrose, glycerol... endless list!!!). Also confirm that aberrant disulphide pairing is no issue by comparative reducing/non-reducing SDS-PAGE. Perform analytical SEC between each step to ensure homogeneity of the size you think you are after (or lack of change is a good sign). I hope you have assayed SEC fractions or Native gel slices to narrow the expected molecular weight of your enzymea/ctivity as this will be invaluable insight.

    All the best of luck and please keep me posted mate. What a challenging project.

  • Abdelhalim Boukaba · Chinese Academy of Sciences
    Has anyone tried reverse dialysis against solid strong hydrophilic polymers? I was once told to do it to concentrate a mononucleosomes preparation after ultracentrifugation in sucrose gradient. It did work but I am not sure if it is worth trying it with a protein preparations. It takes quite long to complete.
  • Matthew Lalonde · University of Utah
    Abdelhalim that's a good question - I have a colleague who used that technique regularly to concentrate a really problematic protein. You should post your question as a new thread so it may be more visible.
  • Mark Van Raaij · Spanish National Research Council
    Sometimes, using a concentrator with a sideways membrane instead of a bottom one helps minimise aggregation (in case you do not use this kind of concentrator already).
  • Abdelhalim Boukaba · Chinese Academy of Sciences
    Thanks Matthew! I may do but right now I think I will restrain myself. It could have negative impact on companies selling ultrafiltration filters. There are jobs in the equation
  • Alexander Golovanov · The University of Manchester
    We often have experienced the problem described, and all the previous suggestions are very good - what exactly will work for your particular protein should be tried experimentally. However one thing that often helped us in this situation previously is using 50 mM Arg + 50 mM Glu as buffer additives (so-called "Magic mixture"), see for protocol details: A simple method for improving protein solubility and long-term stability. JACS (2004). 126(29), 8933-9. | PMID:15264823 | DOI:10.1021/ja049297h
    Does not help all the time (eg is less effective with hydrophobic proteins, or ones which unfold easily), but often does.
  • Mark Van Raaij · Spanish National Research Council
    I can second that, we have had three proteins that were cold-sensitive, i.e. we had to concentrate at room temperature
  • Branka Salopek-Sondi · Ruđer Bošković Institute
    adding of 0.5 to 1mM DTT helps me a lot
  • Didier Fesquet · French National Centre for Scientific Research
    yes, apparently ARG/GLU may work to concentrate your prot w/O precipitation
  • Sanjay Mishra · IFTM University
    In fact, the nature of enzymatic protein should be taken into consideration prior to starting isolation and purification of the same. To check the aggregation, the suitable and sufficient amount of detergent is mandatory to add to the buffer, especially in case of membrane bound enzymes.
  • Yasar Kemal Erdem · Hacettepe University
    If you can tell me what your protein is, I can say something about your problem.
  • Sanjay Mishra · IFTM University
    I hereby give further remark that the statement given by Yasar may be right upto some extent.

Question Followers (48) See all