# Affinity, what is it?

William Reece w.h.h.reece at clinmed.gla.ac.uk
Thu Nov 16 13:19:15 EST 1995

```In article <47uqrl\$r95 at biovax.biobase.dk>, wind at biobase.dk9 says...
>
>Hi all,
>
>Heres a little theoretical question for you immunologists. Imagine, that two
>monoclonal Abs towards the SAME antigen are available. For simplicity, lets
>assume the antigen is monomeric, i.e. no epitopes exist in duplicate.
>
>If one uses these two Abs in ELISA towards the antigen under the same
>conditions (same coating, same concentration of Ab, same detection Abs
etc.),
>and one of them gives a higher signal than the other, is it then fair to say
>that it has a higher affinity than the other? If not, then WHAT is higher or
>better, yhat results in the better ELISA-signals?
>
>I would really appreciate any input on this!
>
>Troels Wind
>PhD-student
>University of Aarhus
>Denmark

It is not aboslutely true that the affinity must be higher.  I'll give you a
couple of instances where you could see a higher signal with a lower
affinity.

The affinity of an antibody depends on the ratio of its "on" rate to its
"off" rate.  In an ELISA, you let the system come to equilibrium with the
antibody and its antigen.  Then you do several washes, and measure how much
of the second antibody is bound.  How much is left depends entirely on the
"off" rate of the antibody, and very little on the "on" rate, since anything
that comes off will most likely be washed away.  You are therefore not
classically measuring the affinity of the antibody, but rather its "off"
rate.

Secondly, depending on where the antibody binds, it may alter the
conformation of the antigen so that the antigen can no longer be recognised
by the coating antibody.  The higher the affinity of the second antibody, the
greater the energy change when it binds its antigen, and therefore the
greater the chance that the antigen will change its tertiary structure to fit
the second antibody.  If the antigen can no longer be recognised by the
coating antibody, it will be washed off and not detected by the ELISA.

These are just a couple of thoughts that I had, but I bet if you think for
longer, you can come up with several other scenarios where ELISA gives a
lower signal for a higher affinity antibody (what about the affinity of the
detecting antibodies for the second antibody, etc.).

If you want to know the affinity of an antibody for its antigen, measure it
using a recognised technique for the purpose.

Love,
Will

```