Antibody purification protocol --any suggestions? - (Feb/02/2007 )
I am working on a protocol to purify an antibody from a hybridoma cell supernatant. I wanted to know if you guys have any suggestions, tips, or criticisms.
Here is my protocol right now:
1. I cultivate my hybridoma cells in a cell factory until I see the media change color from red to orange (~ 1 - 1.5 weeks).
2. Collect the supernatant ~2 liters. Spin at 500xg to remove cells. Pass through filter (0.2um PES).
3. Do a 40% Ammonium precipitation of the supernatant. Spin down the precipitated protein at 4000xg for 15min at room temp.
4. Decant, resuspend pellet with PBS (in about 20ml volume). Dialyse overnight at 4C against 2L of PBS (to remove excess ammonium sulfate).
5. Concentrate the protein using a Amicon Ultra 100kDa cutoff to a final volume of about 3mL.
6. Add about 4mL of the Protein A binding buffer to the volume above (7mL total).
7. Run through a Protein A / agarose beads column (biorad).
While running the crude antibody solution through the column the flowrate grows increasingly slower. When the flowrate becomes almost null, I attach a pump to force the
rest of the solution down. (at about 1.5 - 2ml / min). I need to keep using the pump until the very last step. The column just won't run anymore by gravity flow.
(If anyone knows why this is happening please let me know)
8. I elute the antibody using citric acid pH = 3 (10mL, 5 bed volumes) into a tube containing 2mL 1M Tris-HCL pH = 8.0 (for 12mL total).
9. Dialyse overnight at 4C against 2L of PBS
10. Concentrate to a final concentration of 1mg / ml using amicon column (see above)
Tipical yield = 6mg - 8mg of antibody
why bother concentrating before applying to the column? you may be getting aggregates (not necessarily of your antibody). just add your binding buffer (or dialyze against binding buffer) then clarify and flow into column. it may prevent the column from clogging.
I think you work with quantitives of Mabs min 20mg or more. So try to use as a first step after concentration DEAE IEC ( I think that you can use only ultrafiltration concentration step because a little number of balast proteins as you work with o-o,5% serum in media ). This step will be usefull to enrich your protein mixture and get 90% purity. The other advantage of this approach - you keep activity of your MAb ( because pH 3 elution is too harsh for this molecule ) Then polish with GF ( for ex Sephacryl S300HR, or HiLoad Superdex 200 to purify from dimers and other impurities.
I thought about directly resuspending the protein pellet in the binding buffer and dialyzing against 2L of the binding buffer. But this would consume alot of binding buffer.
I just tought PBS would be cheaper. But probably next time I will try to do it they way you suggested hopefully the membrane won't clog.
Somethings that I wonder about:
1. Should I leave the media in the cells longer? I could wait until the media turns close to yellow (But am afraid the antibody won't like the low pH).
2. Is 40% ammonium sulfate too much; should I try lower amounts (maybe 35%)?
3. Centrifugation after the ammonium sulfate precipitation; should I do longer and faster?
I just wanted to mention that the original protocol I was working with was running the whole 2L of cell supernatant (after being filtered) into the protein A column.
Since the column can only hold a max of 20mL we needed to use a pump to run the supernatant overnight.
The problem with this was: my first yield was about 6mg from doing this. Then I followed I followed a protocol similar to the one above for the supernatant that had already been run and I recovered 8mg more of antibody. But this fraction was slightly more unpure (you could see some bands higher than 50kDa (heavy chain) on the commassie blue staining
-most likely they were serum bands; BSA or such).
Then I tought about using 100kDa cutoff ultrafiltration columns (amicon) so that the most serum proteins run through the filter, leaving the antibody in the concentrated phase.
So this is how I came up with the above protocol, which still needs some improvement...
Hi Chicho!
About incubation time! I don't recommend you to incubate cells till very yellow!. Your prize will be minimal but of course your risk first of all Mab activity and it may be proteolitic digestion of your mab with protease which take freedom from died cells ( for ex lysosomes ). Typical conc of your Mabs by cultivation of hybridoma cells in standart culture flasks is between 10-60 mkg\ml. If your would like to rise amount of your Mabs you must change method for their production. I analysed literature about these approaches and stop my choice on Fiber cell technology. If you would like I can talk to you about this in new post.
About ammonium sulfate conc.
Make saturated solution of AS at rom temp - ( as i know 76g\100ml, dissolve at heating and store overnight at room temp) and add this solution slowly till 45%(V\V).
About centrifugation: better conditions 20000g or more for 20 min at 4C).
I think that before apply to column you should concentrate your super on Amicon. The binding activity depends on concentration ( if the concentration of your mab will be near 1 mg\ml it will be better and of course time consuming). When I've used protein A purification protocol I concentrated super by Amicon, then apply to SephadexG25 equlibrated in binding buffer and then apply protein to the protein A column.
About 100kDa ultrafiltration. In protein isolation and purification strategy it is not a method of purifing but only concentration method. But of course in some cases you can achive purification for a little extent. I recommend to use 50kDa UF membrane, because you can lose some amount of your mAbs on 100kDa membrane. But the best choice for polishing is SEC.
About impurity of your Mabs after purification - I always make SEC on Sephacryl S300HR to remove such residual impurities like Albumin and others.
It seems you have a low yield (6-8 mg for 2 liters of culture). However, it depends on the hybridoma.
usually I get 5-10 mg from 300 mL culture.
I culture in optimem medium, without FCS (easier to purify), I let it around 10-15 days, until the medium is yellow, but the cells do no start to die.
you can try to increase the time of culture, but as previously said, be careful the cells do not die, unless the yield decrease dramatically.
Then I centrifuge, filter the supernatant, and load directly on a protein A sepharose column. with a pump. (1 mg of protein A sepharose). The pump is after the column, so there is no pressure on the column.
I elute with citrate buffer, and load again the column with the flowthrough, in case of my column wath saturated the first time.
I don't think you need to concentrate before the purification.
1. Should I leave the media in the cells longer? I could wait until the media turns close to yellow (But am afraid the antibody won't like the low pH).
2. Is 40% ammonium sulfate too much; should I try lower amounts (maybe 35%)?
3. Centrifugation after the ammonium sulfate precipitation; should I do longer and faster?
1. i don't do cell culture work so i can't answer this question, but, others have.
2. not necessarily. some proteins come out at lower concentrations, others at higher. you need to find the right concentration for your protein.
3. i agree with circlepoint, more rcf and maybe more time may be necessary. also, i would spin the sample at 4C rather than room temperature. ammonium sulfate precipitation is more effective at cold temperatures (in fact, you should incubate your mixture on ice for a period of time prior to spinning).
So for concentration of antibody before applying to the column there are 2 votes against it and 1 vote for it.
I originally also tought that it might be good to do this, but now that I did I noticed that the concentrated protein solution had a much higher density.
I also ran some of this protein solution in a SDS-PAGE gel. I noticed that after adding the loading buffer some protein precipitated, after boiling and centrifugation I
had a blue pellet and pink-purple supernatant. I separated these two phases and added more loading buffer to each, followed by more boiling + centrifugation.
After staining with commassie blue I noticed that the pellet had a bing quantity of probably albumin.
I think I shouldn't be running such a high density protein solution in the column.
I see circlepoint at least put his through a SephadexG25 for a more pure antibody solution to add to the protein A column.
A question about this kind of column: do I have to pack it myself or can I buy it prepacked? Is it cheap?
The step with Sephadex G-25 commonly used as alternative for dialysis. Is is time saving procedure. Yoy can change buffer system for your protein for 20-60 min. against 24 hour for dialyze procedure. So GF on Sephadex-G25 may be so called as purification of your protein from low molecular substances ( Exclusion limit for this resin - 5kDa) or simply change one buffer to another (another call desalting procedure ). But you should Know to make good conversion of your protein in another buffer the volume of applied sample should not exceed 25-30% of volume of the column. So you should make concentration step before. Sephadex g25 - one of relatively cheap resin. I recommend you to buy Sephadex G25 fine ( but not superfine, because it can be easy clogged). You can make a column yorself. There is no high demands and sophisticated procedure for packing the column. The dry powder dissolve as recommended in instructions ( dont use magnetic stirrer) and then degass well your swelling gel. And pack easy. To desalt 30 ml I use 100ml column.