You may collect concentrate and rerun same stuff again and do the wb. Most likely your protein is degrading and by rerunning after few hours you may see more degradation. If so then using other resin wont help instead working upstream and fixing degradation issue will help

Dear Thomas

i think that the best solution is run a supedex 75 coluomn. (the size 16/60 or 26/60) depends from the protein amount that giù would like to purify.

you cannot run it with gravity but you do not need necessarelly an akta sistem, you can connect it to a peristaltic pump. (for all low pressure columns as 16/60 and 26/60 Sephacryl and Supedex it works). Of course you need to pay attention to the bubble and verify the effective flow that you pump provide once that the columns it is connected, because the counter-pressure can reduce the real flow.

You can simply set a theoretically flow, measure the real flow by measuring the volume collected at the out of the columns in 10 minutes and do this until you found the theoretical flow that guarantee a real flow close to the one suggested for the columns from GE.

Several years ago before acquire the first AKTA, I setted-up a simple manual system using:

A GE P1 peristaltic pump connected with an Injection valve (manual – 6 port 2 position – 1 port for the siringe, 2 port for the loop, 1 port for the column, 2 port for the waste - https://kinesis-usa.com/rheodyne-idex-health-science-medium-pressure-valves-medium-pressure-injection-valve-6-port-2-position-bulkhead-manual-v-451.html) that allow to perform sample injection manually, connected with the coloumn which was also connected with a Frac-920 fraction collector which was manually started at the moment of sample injection.

This simple and cheap system was be used several times to perform semi preparative SEC purification that do not require gradients. Its work very well. The main drawback was related to the absence of the 280UV detector and therefore you have to run in the SDS page more or less all the fraction that you collect.

A possible solution to decrease the fraction number to be loaded is perform a fa rapid 96 well colorimetric assay with Bradford reagent 1X (eg. 100ul od Bradford reagent 1X + 10ul of protein from a single fraction) and check which fraction provide to you a blue coloration that indicate protein presence. However, you can do it, if you work with concentrated proteins that you are sure that provide a good signal with Bradford, because coupling the low sensitivity of Bradford with the protein dilution (5-10 times) that may happen during the run, you can risk to do not consider fraction that contains the protein but are negative to the Bradford. Therefore also if you reduce the fraction number to be loaded in the gel on the basis of the colorimetric assay result, do not through away the other fractions until you do not see the SDS-page ans you are sure that your protein is present in the loaded fractions

Sorry for the long message.

Good luck

Manuele Martinelli

p.s: if you are interested I’m building a blog (http://proteocool.blogspot.com/) with some suggestions regarding protein production and purification.

good luck

Manuele

Dear Thomas, we ran through the same problem for our 61.5kDa protein using a 6 His tag. We had to use a three step purification method to get >95% purity. We used affinity versus His-tag on Ni-sepharose column, Hydroxylapatite column and finished with Superose 12. You can read all about it in here:

Article Expression of novel fusion antiviral proteins ricin a chain-...

hope this helps,

Hi Thomas,

I agree with your conclusion that this is probably a degradation product. This is quite a common issue. Assuming that you want the entire protein intact as your final product I agree with Mahesh that you would be best to start by improving the upstream process to eliminate degradation as much as possible. Without knowing your process it is hard to advise, but some general tips follow.

1. Include protease inhibitors in your lysis buffer There are EDTA free cocktails available that are compatible with Ni purification, I’ve mostly used these:

https://www.sigmaaldrich.com/catalog/product/roche/coedtafro?lang=en®ion=GB

2. Keep everything ice cold (or as cold as possible) throughout your process. Pre-chill your buffers, work in a cold room where possible, keep samples on ice, and snap freeze your product if possible (preferably in liquid Nitrogen) but for sure freeze it as soon as possible.

3. Work fast! As soon as you’ve lysed your cells (whatever they are – I don’t know the source of your material) the clock is ticking. The longer the process takes the more degradation you’ll get. This means planning ahead for everything.

4. Avoid using Mg or Mn in your solutions (sometimes people add these as cofactors for DNAse etc). These are also co-factors for proteases and will accelerate any degradation. In fact the only reason I haven’t advocated adding EDTA is that you wish to use a Ni IMAC column. That said, you could still have EDTA in your lysis buffer if you then used an intervening step (ion exchange for instance) before your Ni IMAC and left EDTA OUT of you elution buffers.

5. If there is a dialysis step (for example for refolding? Or buffer exchange?) consider whether you need it. For example would a desalting column work for you? The reason being that dialysis is slow and gives proteases more time to act. If you have a dialysis step and it is essential to the process then ensure that you dialysis buckets are clean (wipe them out with 70% ethanol), cover them during dialysis (foil or cling film) to prevent cold room moulds falling in, obviously keep it cold throughout, and try to add inhibitors and / or EDTA (as appropriate given the IMAC use).

6. Ensure that your resins (whether you’re working in batch and reuse it, or you’re working with columns) are clean by employing a regular Clean In Place (CIP) protocol. See the manufacturer’s instructions in each case for this, but generally a wash step with 500 mM NaOH, followed by removal of the NaOH with a water wash, and then storage in a biocidal solution (can be 20% ethanol, or 0.01M NaOH). Also if you are using an FPLC or other system / pumps (sounds like you are not) make sure they are cleaned via a similar process. This point is especially important where resins and pump systems are communally used!!

The above points all suppose that you want the un-degraded full length protein, which brings us to the question of why you’re making it. If you are intending to crystallise it for example you may find that the truncated version crystallises while the full length does not. So consider why you are making it and whether or not you might want to clone a shorter version. There are various tools available that could help you predict where to make your truncations

(e.g. see https://www.expasy.org/proteomics/protein_structure)

and from these you could clone versions that do not encode the labile terminus that is being processed. This would allow you to express a more stable, proteolytically resistant product that may still meet your end goals. Better still, if you have access to a mass spec you could use your degraded prep to identify the region that is being clipped. Remember, you may be losing both ends!

Of course it may be that a labile N or C –terminus also turns out to be functionally significant, in which case you will want to retain it and the precautions against proteolysis listed above will help.

Finally, if you do still need to separate a labile fragment from the main body of protein, then yes Size Exclusion may do it – so long as the labile fragment is not associating with the main body of protein in solution. Remember, you are seeing these products on a denaturing gel, but on the column they may be stuck together. In that case consider your salt content and try raising it a bit to see if you can disrupt any interaction. Also consider ion exchange or hydrophobic interaction as purification options – in fact HIC may be particularly useful in this case. The other option you might consider is switching your tag to the other end of the protein - if only one end is being processed. However, that may give you the problem of a doublet as your main product (part processed, part not). A last gasp might be to tag both ends (different tags – say for instance His and Myc or Flag) and purify sequentially starting with the tag at the non-processed end.

Hope that’s given you something to think about. Good luck.

Hi Thomas,

I agree with Sourav Roy and suggest that you try the gel filtration method first. Superdex 75 has a higher resolution than Superdex 200. If your target protein with high concentration, you can also try a long column, depending on the characters of your protein.

Hope it helps!