I would like to know if there is a quantitative justification for using a specific threshold to calculate steady state Kd affinity using surface plasmon resonance (SPR). Some people say 10% but I don't know that this is backed by any data or simulations.
My goal is to be able to establish a cutoff for which we can confidently calculate steady state Kd for weak interactions. For example, you have a protein-protein interaction that has a Kd of 200uM and the highest you can go due to analyte supply is 2uM, then you won’t reach saturation. However, you can still fit the isotherm to get a Kd steady state value. In this case the final response will be much lower than the fitted Rmax.
Do you know of any papers that look into assigning a Kd value based on the noise in the system and possibly some simulations? For example, is a cutoff of 10% (final response over predicted Rmax) sufficient to assign a steady state Kd based on any models?
Thanks
Not sure if understood the questions fully.. However, you can calculate the theoretical Rmax followed by experimental Rmax upon ligand immobilisation..
This way you can then determine how active the sensor chip is following analyte injection after subtracting the reference sensor surface...
Chip type used, how much ligand immobilise and running buffer can influence Rmax - thus optimisation needed in some cases to obtain analyte saturation to work out the KD value..
In the past I optimisation experiments to obtain saturation and export the fitted Req from the evaluation software to extrapolate KD in prism by forming a hill plot..
Bubacarr G Kaira Thank you for your input. Suppose the predicted Rmax is 100RU for a weak interaction and due to analyte limitations (e.g. solubility, availability, etc.) you are only able to go high enough such that the highest response is 10RU. A control analyte was used to show the captured ligand on the surface is 100% active. Is this 10% saturation enough to get an accurate Rmax and KD fit? I don't think it is and haven't found much in the literature to support this.
@Rm Mal it has been recommended to keep the Rmax around 100-150 for reliable kD if the evaluation software is being used. However, the minimum Rmax I am not sure about but I think the noise level is around 2 RU when in running buffer - and may vary depending on buffer type..
Anyway, 15 Rmax has been reported before and see no reason why 10 RU should be reliable.. Of course the higher the better as long as there is no mass transport and the signal to noise ratio is reasonable to get a signal to fit the data..
Question : Can't you immobilise the analyte and use the ligand as "analyte" if the ligand has a bigger mass to increase the RU...?
By the way to get a realistic KD try 10 times less and 10 times more than the KD for the analyte concentrations for the binding..
- Badges
- Science topic
- Similar topics
- Biological Science
- Immunology
- Antibodies