2012年3月14日星期三
记录
2012年3月10日星期六
Affinity Chromatography of Histag protein
Many biochemists purify their beloved proteins, whether at small or large scale and for different reasons (identification, enzymatic reactions, characterization etc.). There are several rules and tips that can facilitate the purification process and increase your chances to have a pure protein in your freezer at the end of the day.
Predict and plan
The first step is to perform some basic bioinformatics analysis. The most important ones are those that relate to the physio-chemical properties of your protein such as amino acid composition, extinction coefficient, secondary structure prediction and motif prediction (see here for a comprehensive list of bioinformatics tools). Another good prediction website worth mentioning is Xtalpred, which acts as a meta server for the most basic protein analysis.
Once you gathered some basic knowledge about the protein, you can now assess:
- Membrane or cytosolic
- Number of cystienes
- Predicted pI
- Expected extinction coefficient
- Motif or structural domain which necessitates the addition of ions or co-factors
All the above will affect your expression system approach and your subsequent purification process. Plan ahead when to do an expression tests (different hosts, different induction media, different temperature and induction times). Evaluate how much protein you should require and plan your up-scaling expression accordingly. Check to see that you have the required materials (DNAse, protease inhibitors, columns and FPLC reserved) and prepare your schedule (you should look into Labguru, a web-based laboratory management system, that can help you manage your experiments).
Expressing your protein – making the most of every ml
This is one of the crucial steps in a succesfull purification scheme – you can do 5 purification steps and still be left with a dirty prep or do just 2 steps and get 95% and higher purity level. It is THAT crucial. Getting protein highly expressed in E. Coli is not always trivial but with some initial screening you might struck a successful induction setting (see my earlier post on how to improve recombinant protein expression). What settings should be checked?
- Media – Don’t settle only for LB and IPTG induction; you should try theAutoinduction media, developed by William Studier, which can boost protein expression between 3-10 fold, depending on the protein.
- Temperatures & induction time – Check several temperature points: 18, 27, 30 and 37 degrees Celsius, while varying the induction time appropriately. Meaning, the lower the temperature the longer induction time (we routinely induce proteins at 20 or 27 degrees over weekend). Start with an overnight and modify accordingly.
- Strains – Try inducing with Rosetta, Codon+, Tuner and Rosetta-Gami strains.
All the above settings should be tested at small scale (i.e. 50ml culture) and once you found a successful combination go ahead and up scale. And when I talk about up scaling I am talking about a minimum of 4L culture or higher. You might be surprised at this volume scale, but it is better to get a bucket loads of protein and then throw away major chunks as you cut out dirty fractions. No problem to start with 200mg of protein and then be left with 10mg of super pure protein.
Breaking cells or cell lysis
Assuming most will purify a recombinant protein, your protein will be expected to contain an N or C-terminal tag.
- When breaking cells try to use five volumes of lysis buffer per one gram of cell pellet.
- I found it best to redissolve the frozen pellet with the aid of a stirrer – this way you can continue preparing for the coming purification while the pellet slowly dissolves into the lysis buffer
- If you’re not purifying a membrane protein you can add 0.02% Triton X-100 to aid with the lysis of the bacteria.
- If you’re using His tag, don’t forger to add 10-20 mM Imidzole to the buffer
- Use ice all the times! The sample should be at 4°C all the time.
- Pre-chill the French press in ice about 10 minutes before intial lysis. If you’re having many samples you might want to chill it again for 5-10 minutes per several samples to keep it cold.
Initial separation and affinity chromatography
- Centrifugation at 45,000 rpm for 30-45 minutes is sufficient for most proteins (you can live also with half of it using SS-34 rotor at 20,000 rpm for 2 hours).
- If using a dripping column, initially pack beads that will be sufficient to ~10mg protein/1L of induction. After the first purification it will be easier for you to adjust the amount of beads to get the best purification without getting too much junk.
- Remember that in most cases the affinity chromatography step will determine the fate of your sample in terms of purity levels. Do your best to calibrate this step to the optimal setting possible so you get the protein at the highest purity level. From my own experience a dirty sample post affinity will not improve tremendously after using other columns (again, in most cases).
- Perform extensive washes, especially if working with dripping columns, 20 times the amount of beads volume (20 CV), varying the amount of salts and Imidazole.
- Elution should be conducted at a relative small flow rate to get the sample concentrated at the end of this step; in certain cases you’d want to increase the flow rate to avoid high concentration elution in case the protein is prone for aggregation.
- Evaluate your sample purity through an SDS-PAGE analysis. In case you find it hard to identify your protein do a western blot for verification (blot directly from the coomassie-stained gel).
- Depending on the purity level required and the observed purity of the sample, assess whether you want to do a step of ion-exchange or size exclusion.
- In case you plan on doing a step of ion-exchange, you might consider eluting your protein with low salt concentration (maximum 50 mM ion strength) and continue directly with your post-affinity elution into the ion-exchange column.
Purify, purify and…purify more: tips for improving your protein purification capabilities – part 2

HisTrap Prepack column from GE
Nickel affinity column gradient elution – capturing and purifying
While batch binding and elution step can be performed on a faster time scale (hour to hour and half from start to finish including wash steps), using a prepack nickel affinity column (such as HisTrap from GE) can help ending your first chromatography with a higher level of purity. This is true for all samples but it is crucial when you find your nickel affinity purification relatively dirty.
Why not use the prepack column from the first place?
It depends on the amount of proteins your purifying at a single moment (in our lab we sometime purify two to four proteins in parallel) and of course, if your sample volume is large it will take more time to bind, wash and elute.
When I am using a prepack column with gradient I still get the protein dirty no matter how much wash steps I do – what can be done to improve the nickel affinity purification?
From my own experience, when you have a good overexpression (i.e. higher than 15mg/1L of induction) then you will have better purification because the dominant protein in the sample will be your protein. However, if you have a low expression from the first place then your protein will have to compete with other host proteins which have similar expression levels. In such cases you should first consider improving expression via Autoinduction media (Studier et al. 2005) which I found to increase the fold of protein expression by 3-10 fold, depending on the protein. See my previous note about increasing expression.
Assuming you’ve tried everything and your protein is still low expressing you can do the following combination strategies, separately or combined:
- Decrease beads volume – if using a prepack column, try loading the protein on a 1ml column and not a 5ml. Why do that? Well, let’s assume your protein is expressing at 5mg/1L of media and you induced 4L of culture, giving a total of 20mg of protein of interest. If you will load the lysate on a 1ml of prepack column, that can capture 20-30mg of tagged protein, between 67% to 90% of the column will be captured by your protein. On the other hand, if you load the 5ml prepack column, with a potential capacity of 200mg (!), all the rest of the host proteins will have plenty of room to bind to.
- Removal of periplasm – Another strategy is to do periplasm wash of the bacterial pellet before performing the bacterial cell lysis. That’s because the periplasm was shown to contain many proteins and enzymes that naturally bind divalent/metal ion. By removing these protein you lower the host protein’s that compete over binding to the beads.
- Guanidinium/Urea wash – This is a bit risky but it is worth the try. With this strategy cells are lysed and bound to the nickel column. Then a wash with 2M of Guanidinium or Urea is performed which usually removes most of the host proteins and afterwards a shallow gradient of decreasing amounts of Guanidinium/Urea brings back your protein to native state. Of course, you will need to check and see that your protein didn’t loss its fold and function (usually if the protein refolds improperly you will notice aggregates once you concentrate your protein). The advantage of this technique is the ability to get a highly pure protein within one chromatography step.
Purify, purify and…purify more: tips for improving your protein purification capabilities
After doing our best to purify our protein at the affinity chromatography step, we might decide we want an additional step of purification, which in many cases involves either an Ion-exchange or size exclusion chromatography. In this part I will focus on Ion-exchange chromatography and some tips on how to maximize your time and effort to get the most of this system.
Ion-exchange chromatography (IEC) binding step
Ion exchange chromatography is a powerful technique that is used on many cases at the middle of the purification process. I have discussed a method how to improve peak resolution by repetitive Ion-Exchange chromatography and today I would like to discuss general consideration when working with ion exchange columns.
Unlike affinity chromatography, the elution in IEC is conducted via increase in ion strength or gradual change in pH that may affect your protein behavior (structure and/or function). Since binding of the sample to the column should be at minimum salt interference and maximum surface charge, usually there will be a requirement to lower the salt concentration. If the previous step was a His tag Affinity chromatography, it will save time if the elution step will contain a minimal amount of salt (e.g., ~50mM NaCl) so there will be no need to exchange buffer. If you’re wondering about the effect of the Imidazole in the solution, it depends on the type of IEC column used. If you’re using a cationic exchanger or working under pH 7 you have no need to exchange the imidazole simply because the fact that imidazole has a pKa of ~7 for its first nitrogen, thus at higher pH the ring will not be charged and will not interfere with the binding of the protein to the matrix.
The beauty about IEC is the fact it is a capture method: you can load as much volume as you want as long as you don’t overload the column with protein mass. This means you can start from 50ml or even 500ml diluted solution assuming you load your sample though a buffer valve. One thing to remember, though, is that there are proteins which do not bind to the anionic or cationic (or both!) so make sure you start collection immediately upon injection. Even in case your protein was not bound to the column, the contaminants might bind to the column so it is important to step-elute the bound material and evaluate it against the flowthrough and the pre-injected sample.
IEC elution step
When considering the elution step, two primary factors should be evaluated: elution buffer composition and elution methodology. In regard to the elution buffer it is usually recommended to use high concentration of salt (1-2M). pH shifts is less common since proteins are sensitive to small changes to the buffer’s pH which might affect their structure. On the other hand, it should be considered that gradual increase in salt can shift the protein from one oligomeric form to another (i.e. monomer vs. dimer) depending on its inter-molecular surface characteristics. Exploring other salts than the common NaCl is recommended when the protein aggregates or not eluting from the column.
The most common elution methodology is gradual increase in eluting buffer concentration (commonly known as “B buffer” or %B). While requiring more time, this methodology is essentially “fail-proof” as long as the gradient is relatively shallow such as 50%/50 min or 50%/10 column volume (CV) at 1ml/min flow rate using a 5ml column. Essentially you’ll want to elute the protein across 10-20CV so you’ll have better separation among close peaks. The slow rate will generate a much more concentrated elution which can expedite downstream concentration step.
Usually it is recommended to do a pilot run on the IEC. Inject small amount of the sample so you won’t overload the column (depending on the amount at hand, try to aim at 5-10mg) and evaluate the separation profile established with the current elution buffer composition and gradient. It is essential to couple a SDS-PAGE analysis of selected peaks together with the chromatogram so you can evaluate the following parameters:
- Peak separation
- Purity of the selected fractions
- Amount of injected sample – should you increase/decrease the amount?
At this point you can decide on keeping or changing the current settings.
When analyzing the peaks, one should remember that proteins can bind to the matrix through differently charged surfaces, which means that the elution profile of a 95% protein can generate several elution peaks. That’s why it is crucial to perform an SDS-PAGE analysis of the major peaks to verify whether a certain elution contains the protein of interest or a contamination. One should also remember that in certain cases what looks like a monomer on the gel can well be a dimer or an oligomer which forms a new charged patch that doesn’t exist in the monomer state; in certain cases, like crystallography, sample heterogeneity at the oligomeric level can be fatal for successful crystallization effort.
Step or gradient?
Moving from the gradient methodology to the step elution is a bit more complicated than it seems, especially when trying to separate between closely related peaks. Lets take the following scenario on the right.

IEC Gradient elution
In this purification run the main peak is bordered by two relatively small peaks. This was done by using an anion exchange column; color codes are blue color stands for 280nm, red for 254mn and brown for the conductivity in mS. Even though I would not recommend a shift from gradient to step in this particular case, I will use it to explain the workflow to perform such a change. First you should determine the volume in which you will want to start the step elution according to the absorbance curve (in this case I chose the inflection point, red dash line). From this volume deduct the total volume of the column (in this case 5ml) + additional 10% overhead in case of measurement errors. An additional way to determine the exact volume is to use the “%concentration” curve and determine the difference in volume between the starting point of mixing buffer A and B and the point at which there is increase in conductivity. I have done so manually (without the %concentration function) and established the volume at ~115ml (blue dashed line) afterwhich I have deduced the conductivity value at this specific volume. This will be the first elution step that will remove the preceding contaminant peak. The same can be done to separate the main peak from the late contaminant peak. Once you figure this out, repeat the test run with this step elution and see if it fits with your needs.