BSA aggregation

Rich Dudley rdudley+ at
Thu Sep 3 07:38:12 EST 1998

Are you sure that it's not your protein aggregating?  Ultrafiltration is
notorious for this kind of a problem.  There aren't any guarantees in
protein work, so you'll have to try a few guesses.  I suggest trying an
ammonium sulfate precipitation prior to ultrafiltration--maybe you can
separate enough BSA to prevent the agglomeration.  Is your protein an
integral membrane protein?  Or is is secreted into the medium?  If it's
secreted, I would try a synthetic medium with amino acids, rather than a
complex protein source.  If it's an integral membrane, you might want to
avoid concentration by ultrafiltration--that will remove the detergents to
below the CMC, and promote glopping of your protein (and I'd wash my cells
much better before I made the gemish).  Your best first step may not be
untrafiltration--in my experience, it was a better later step.

If you have to use ultrafiltration, is there a NMWCO between the size of
your protein and BSA? (probably not--that would be too easy, and protein
works can't be easy)

Hope this helps!


bigred wrote:

> A simple(?) protein purification question:
> As a first step, I've been trying to concentrate via ultrafiltration my
> protein of interest from conditioned mammalian cell culture media.  The
> BSA present in the medium appears to be aggregating, it's being retained
> even with a 500 kD cutoff membrane!  It also appears to be taking my
> stuff with it.  Aside from using serum-free/protein-free medium (and
> taking the BSA out of the equation), is there a known method for
> avoiding BSA aggregate formation (salt, pH, reducing agents) without
> mucking up most of the other proteins in the gemish?
> Thanks,
> J.P. McGrath
> Alkermes, Inc.
> Cambridge, MA 02139
> john_mcgrath at

--- --- --- -- -- -- --- --- ---
Richard J. Dudley (rdudley+ at
Research Specialist V
Dept. of Cell Biology and Physiology
University of Pittsburgh
---> search BIONET archives at <---

-------------- next part --------------
A non-text attachment was scrubbed...
Name: vcard.vcf
Type: text/x-vcard
Size: 463 bytes
Desc: Card for Richard Dudley
Url :

More information about the Methods mailing list