Ab purification problem

D.K. dk at no.email.thankstospam.net
Thu Oct 17 21:27:30 EST 2002

dk at no.email.thankstospam.net (D.K.) wrote:
>ljd3 at duke.edu (Lou D'Amico) wrote:
>>i used a 3mg/ml (little less) solution, i had about a (very) little
>>over 1ml of that, and i bound it to 1ml of column matrix.  It's the
>>same volume as the company i outsourced the first affinity
>>purification to.  Any help?
>Some. It would help if you could state how much of your antigen is
>bound to the matrix, not just how much you used. Depending 
>on many factors coupling efficiency can vary greatly from 
>as low as 10% to nearly 100. Still, lets do some rough numbers:
>Assume the antigen is actually coupled. Say, 75%  of it, 50% in 
>the worst scenario. So you have ~1 ml of ~ 2 mg/ml column. 
>Because it's serum, it is reasonable to expect more than 1:1 
>molar binding. Unlikely more than 3 because of steric 
>hindrances, etc. OK, so that means you can expect at least 
>2/2.5 mg Ab = 800 ug (150K = 2.5x60K). You say you actually 
>got 500 ug, which is not far off. In ideal world I would expect 
>about 3X of this. But since you did not select for IgG first, it 
>is quite possible that i) serum was collected prematurely and 
>has a lot of IgM ab, ii) you failed to elute those IgM (they are 
>harder to elutel; you don't mention how you elute and if you 
>checked on a gel what remains bound). All in all, your result does
>not seem to be impossibly wrong. 

Another possibility, although not very likely, is that epitopes 
normally hidden in the native protein are much more 
immunogenic than the ones exposed. In which case, particularly 
for late, mature bleeds, majority of Ab in the serum will not 
bind to the column made with native antigen. They will work 
perfectly fine on Western because antigene there is denatured 
and "crucified" on membrane. This kind of thing happens 
all the time with moAb. That kind of thing can account for your
low yeild, too. 


More information about the Proteins mailing list