Martin Offterdinger a8803349.nospam at
Wed Sep 15 01:13:48 EST 1999

On 14 Sep 1999 18:49:51 -0400, iayork at (Ian A. York) wrote:

>In article <37DEBAFB.2DB43E07 at>,
>Caroline Szymeczek-Seay  <css at> wrote:
>>label the cells, immunoppt with anti-protein X, and run the reaction on
>>a gel, flurograph, and expose.  This should give me, if protein X
>>interacts with X, Y, and Z, three different bands with intensities equal
>>to their affinity for X.  It won't tell me what proteins those bands
>If I follow you, you can't use this system to measure affinity.  The
>intensities of the ands will be related to the proteins' affinities, and
>also (and probably much more importantly) to their concentrations.  To
>measure affinity you need to know the concentration--even if protein Z had
>a much lower affinity for X than Y, if it is present at 100 times the
>concentration of Y it may still look like the dominant binding partner.
>As you described the experiments, you have no idea of the relative
>concentrations of each protein, and determining this is not a trivial
>task.  (Keep in mind, too, questions of subcompartmentalization and
>concentrations in a microenvironment.)
>If you really need to know affinities, you probably need to do the
>experiment in vitro in a defined system.

I  have to agree with Ian; in addition labeling of the cells is not
really better than Western Blotting; in this case you have to take in
consideration a possible different content of Methionine ( if you are
using 35-S-methionin for labeling) in the Proteins x, y and z which
will result in different band intensities after fluorography even if
the same amounts of x,y and z are present in the IPtates.
Martin Offterdinger
Internal Med.I,Dept. Oncology
University of Vienna
E-Mail:a8803349.nospam at
(remove .nospam before mailing)

More information about the Methods mailing list