Two analytes similar kon, koff, KD despite vastly different responses

Hello all.

First time poster, and novice SPR users.

In this set of experiments I have the same ligand, but two different analytes. Analyte 1 (sensogram 1) should bind fairly tightly to the ligand, which we do see. Analyte 2 (sensogram 2) should not bind, or maybe very weakly and nonspecifically.

If you look at the responses between the two and the fits, they look drastically different, but the kinetics data for the two are fairly close. I was just wondering how this is possible given the obvious differences between the two.

Data were fit using Scrubber 2.0c.

Thanks!

Welcome to the forum.
Yes indeed, this is the point where comparing just kinetics can give you some wrong answers. As you noticed the measured curves are totally different and more obvious the response level. There are several depencies ( www.sprpages.nl/sensorgram-tutorial/a-curve ) which lead to response level differences. Since we don't know the ligand immobilization level, size or anything about the analytes it is impossible to say where the differences come from. So please give some more information ( www.sprpages.nl/kunena/new-forum/72-read-this-first ) so we help you better.

Kind regards
Arnoud

Thanks for the reply. Some more info:

The instrument used is a Biacore 3000. The chip used was an SA chip and 250 RU of the biotinylated ligand were put down on the chip surface. Run buffer was 1x HBS-EP+, regen was 1 M NaCl + 20 mM NaOH.

Ligand = 9731 Daltons
Analyte 1 (sensogram 1) = 4725 Daltons
Analyte 2 (sensogram 2) = 4918 Daltons
Concentrations ran were 20/10/5/2.5/1.25 uM in triplicate

Experimentals were 30 uL/min flow rate, 3 minute on time, 30 minute off time. 1 minute regen

Kinetics
Sensogram 1: ka=1372, kd=6.56e-5, KD = 48.11 nM, Rmax 285.06
Sensogram 2: ka = 4.99e3, kd = 2.46e-4, KD = 49.3 nM, Rmax 17.47

Thank you for the help.

From the capture level and ligand and analyte size you can calculate the theoretical Rmax. Since both analytes are almost the same size it is around 125 RU. Interestingly analyte 1 has a calculated Rmax of 285 RU which is more than twice the expected value. Can the ligand bind two analytes?
Analyte 2 only gives an Rmax of 17 RU which is < 15% of the theoretical Rmax. This can indicate that there is no interaction, especially because of the high analyte concentrations (see below). And therefore, although the fitting is reporting a value, this can misleading.

Next is the analyte concentrations. In general we prefer an analyte concentration range between 0.1 and 10 times the expected K_D. With a K_D of around 50 nM this would be between 5 and 500 nM which is lower than your lowest concentration. Too high concentrations can add some extra non-specific interactions to the sensorgrams (analyte 2?). My recommendation is to repeat the experiment with a lower and broader concentration range. For instance start at 2 µM and make a dilution series of step 3 (2, 0.67, 2.2, 0.74, 0.25, 0.08 µM).
The problem with the slow association rate is that it takes a long time to get some curvature in the lines at lower concentrations. Simulations show that for analyte 1 it can take up to 20 minutes to get to steady state for a 2 µM concentration (2' at 20 µM). Unfortunately the maximum injection time at 30 µl/min is around 8'. Since mass transport seems not an issue here you can lower the flow rate and see what you get.

Kind regards
Arnoud