Re: using SDS PAGE for purity

Harry Whatley (h_whatley@hotmail.com)
Thu, 15 Jan 1998 13:44:55 PST

There is another way to do this assay that is more quantitative (and
simpler) than scanning stained gels. If you have access to a capillary
electrophoresis system you can do your SDS separations that way. You
only run one "lane" at a time, but everything is automated and the
reagents are all in "kit" form. The detection is on-capillary by UV
absorbance at 214 nm. Because you are detecting the peptide bonds there
is much less interference from glycosylation and other modifications.

There is definitely a chance that the carbohydrate will interfere with
coomassie binding. There is also a great deal of variability in
Coomassie from lot to lot (it is a mix of dyes, not a pure substance),
and that can come back to bite you in a validated method that you plan
to run for years. I recall one protein (years ago) that stained bright
pink with some lots of dye.

I think there are already some validated methods using SDS gels on
capillary electrophoresis, so it can be done.

Harry

>From abrf-request@aecom.yu.edu Tue Jan 13 16:25:45 1998
>Received: (from daemon@localhost)
> by post.aecom.yu.edu (8.8.8/8.8.8) id MAA05983
> for abrf-out; Tue, 13 Jan 1998 12:12:01 -0500 (EST)
>Received: from t1.immunex.com (firewall-user@T1.IMMUNEX.COM
[198.178.217.1])
> by post.aecom.yu.edu (8.8.8/8.8.8) with ESMTP id MAA05967
> for <abrf@aecom.yu.edu>; Tue, 13 Jan 1998 12:12:00 -0500 (EST)
>Received: by t1.immunex.com; id JAA25267; Tue, 13 Jan 1998 09:08:10
-0800 (PST)
>Received: from radium.immunex.com(198.178.220.38) by t1.immunex.com via
smap (3.2)
> id xma025132; Tue, 13 Jan 98 09:07:47 -0800
>Received: from ccMail by radium.immunex.com (ccMail Link to SMTP
R8.10.00)
> id AA884711612; Tue, 13 Jan 98 09:13:35 -0800
>Message-Id: <9801138847.AA884711612@radium.immunex.com>
>X-Mailer: ccMail Link to SMTP R8.10.00
>Date: Tue, 13 Jan 98 09:11:31 -0800
>From: "Leslie J Novick"<novick@immunex.com>
>Old-To: <abrf@aecom.yu.edu>
>Subject: using SDS PAGE for purity
>MIME-Version: 1.0
>Content-Type: multipart/mixed; boundary="simple boundary"
>To: Recipients of ABRF List <abrf@aecom.yu.edu>
>Sender: Association of Biomolecular Resource Facilities
<abrf-request@aecom.yu.edu>
>Precedence: bulk
>Errors-To: abrf-request@aecom.yu.edu
>
>I am trying to quantitate impurities in "pure" protein by SDS PAGE
Commassie.
>I am using pharmacia scanning software (ImageMaster) with gels scanned
using
>LabScan in Transmittance mode. The 'pure' proteins are glycosylated
(thus not
>sharp bands), the gels are gradient and the impurities are <5%.
>
>Issues I am addressing include:
>1) Scanning parameters (300 vs 600 dpi)
>2) Lane drawing: Should the lane encompass the whole protein, or would
a thin
>lane through the middle work better?
>3) Coomassie staining: a) Do glycosylated proteins abs. coomassie
differently
>than non-glycosylated proteins? b) Why do I find that lower molecular
weight
>protein standards absorb coomassie MORE than higher molecular weight
proteins?
>c) If a gel is well destained, do I need to consider the acrylamide
gradient?
>4) Linear range: to get the contaminants within their linear range I
must
>overload the protein. By doing this, the calculated % purity is wrong.
>
>
>Whatever method I use would be for an FDA submission. Does anyone have
>any ideas or references about this?
>
>
>

______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com