Problems Fitting Data

Hi everyone,

I've been investigating protein-protein binding using an NTA chip and have found these two proteins in question seem to bind fairly tightly with a Kd ~ 5 nM (1:1 binding), and whilst I can get consistent data (Kon/Koff etc) across repeats, and have relatively low Chi squared values (~3), visual inspection shows the biacore evaluation software does not fit my SPR trace too well. I've tried changing the buffers (particularly the salt concentration has been increased to 500 mM) but I still get traces like the one I've attached, has anyone seen anything like this before and does anyone have any suggestions as to ways I can change the modelling or the experiment to improve this?

Thanks!!!!!!!

Hi and welcome,

I have some questions/remarks:
1) is the trace reference subtracted? The fit fails because of high bulk jumps (RI).
2) although the fit seems to use RI in the highest conc. it does not in the lowest. Are the fitting settings the same for each concentration?
3) can you match sample and buffer better or is the sample in a special buffer? what is you highes concentration?

Kind regards
Arnoud

Hi Arnoud,

Thanks for your reply! I agree with what you say about the high bulk jumps, although the evaluation software does not think there is any problem with this bit so maybe this is something I need to adjust more manually.

To answer your questions, the trace is double reference subtracted with the same fitting settings for each concentration. I have tested a 3-fold dilution series with 7 concentrations going from 750 nM to about 1 nM of protein, and I have made up the samples by diluting 5 uL of protein stock with 295 uL of the running buffer then diluting samples of this further with the running buffer for lower concentrations. Since the protein stock uses a different buffer I guess the samples will be in a more different buffer for the highest concentrations so perhaps this slight buffer difference is enough to cause this problem...

Nevermind, matching the buffer does not help the machine model the bulk jumps unfortunately,