determination of binding constant of protein-ligand

David Martin damartin at
Wed Feb 14 08:22:24 EST 1996

In article <4fs0g7$agm at>,
   Philippe Lefebvre <plefebvre at> wrote:
-->Whatever method you are using to separate bound from free ligand, you have 
to test 
-->ligand concentrations from 1/10 Kd to 5 Kd, keeping in mind that the data 
-->at very low ligand concentrations will be critical for determining the Kd. 

Good quality data all round is important. For Kd the lower end of the range is 
important. For Bmax the higher end of the range is important.
Better to stretch the range to more than 1 order of magnitude (10X or 1/10X 
Kd), collect lots of data with many repeats and see what it looks like.

 Then you 
-->have to plot the data according to Scatchard.

Scatchard is a linear representation of the binding equation

Bound = ([ligand] x Bmax) / (Kd + [ligand])

This looks very much like the michaelis menten equation.

It is better to fit the equation directly to the data rather than enhance 
errors in one dimension by transforming to the scatchard plot. There are 
plenty of free programs that will do curve fitting for you. Try the EMBL 
software archive.

(the shape of the curve and how well it fits will also help you to determine 
where the deficiencies in your method are)

hope this helps.

...david Martin

* David Martin, PhD -  Post-Doctoral Research Fellow    *
* Atherosclerosis and Thrombosis research group         *
* Biotechnologisenteret i Oslo                          *
* Gaustadalleen 21  (Postboks 1125 Blindern)            *
* N-0316 Oslo                                           *
* Norway                                                *
* Tel: +47 22 95 84 54  Fax: +47 22 69 41 30            *
* email: david.martin at                     *

More information about the Methods mailing list