Hello.
I have been trying to determine dissociation constant for a monoclonal antibody in solution using an ELISA-based protocol first described by Friguet et al.. The same protocol is also described in the "Antibody Engineering, A practical approach" handbook and I believe it is quite often used as a simple technique of obtaining antibody dissociation constants.
Basically, indirect ELISA is used to determine concentration of the free antibody in a series of solutions, pre-incubated with varying concentrations of the antigen. An initial experiment has to be performed beforehand to ascertain that: (a) the antibody concentration used is in the linear range of the ELISA response (so that the absorbance is proportional to the Ab concentration); (b) only a small fraction of the total free antibody in solution is retained on the plate (so that the measurement does not significantly affect the Ab-Ag equillibrium in solution). For (b), two plates are coated with Ag and a series of Ab solutions with different concentrations is added to the wells. After incubation, antibody solutions are transferred to the second plate. Special care is taken that all the incubation and detection steps are the same for both plates. By comparing the signals from both plates, you can determine the fraction of the free antibody that is retained on the first plate (by comparing the slopes of absorbance vs. concentration graphs). No more than 10% of the antibody should be retained by the plate, if it is more, you should coat less antigen and repeat the experiment. I am having problems with this initial experiment, as the slopes differ by more than 10%. I lowered the concentration of the antigen twice (now down to 15 nM), but I see no improvement. In the last experiment I even got a bigger difference in slopes than when using double concentration of antigen and now I am losing my confidence in the reliability of the assay. It seems to me that the method is too sensitive to even very slight variations in the treatment of the two plates and that the differences can not be reliably attributed to the amount of the retained primary Ab. Is there anybody out there with experience with this method? Did it work for you? Do you have any suggestions or do you recommend any other approach of determining antibody Kd (ELISA-based would be favored; I will be doing Biacore analysis also, but that can be fishy, too). Any help will be gladly appreciated. Thank you in advance.
Some years ago, we published a paper, in which we used a quite simple method for the determination of Kd via ELISA. Unfortunately, we never published the protocol itself in more detail. The method is based on the fact that Kd can be calculated from the test midpoint (IC50 in mol/L) if two conditions are met: antibody concentration and enzyme conjugate concentration should approach zero, shown by dilution series. If Kd is reached, the IC50 does not change anymore. This is known as "affinity limitation". Only the concentration of the antigen needs to be known. If you need support for the application of this method, send me a message. https://link.springer.com/content/pdf/10.1007/s002160050479.pdf
Article Development of a high sensitive enzyme-immunoassay for the d...
Hi Jeonga, check this link. We use this method for the measurement of affinity constant (Kaff) of several antibodies developed in my lab. It is very easy to perform, good luck!!
Oscar
Article Measurement of monoclonal antibody affinity by non-competiti...
Given the limitations of solid-phase assays in general and the assumption that antibody binding to both the soluble and immobilized forms of the antigen is the same, determining the Kd becomes a fairly simple and frequently performed experiment.
1. Immobilize the antigen on the ELISA plates, wash, and block.
2. Prepare serial dilutions of the antigen in tubes (or even in the wells of the plate itself).
3. Add an equal amount of antibody to each tube and transfer to the plate (or to each well of the plate).
Ideally, the antibody concentration should be kept as low as possible but you need to use enough antibody to get a reliable signal in the absence of any competing antigen. Typically, an antibody concentration at or just below the Kd is optimal. Most antibodies have Kds in the nanomolar range so that's a good place to start. I recommend using this method primarily for comparative purposes and not for any definitive affinity determination.
Please look at my paper on "Researchgate" under title "Factors Governing the Sensitivity of Non-Competitive Immunometric Assays: A Theoretical Analysis"
It seems that the Elisa assay that you perform suffers from not meeting the requirements for Kd determination, either because the concentration of your antibody is significantly reduced during the assay (strong binding which means that you have to use relative low concentrations) and/or not reaching equilibrium during incubation (for strong binders this can take up to days at lower concentrations). Biacore will suffer less from these limitations but even there it can be tricky to monitor the slow dissociation rate. An alternative and cheaper method is provided by LigandTracer (www.ligandtracer.com), originally developed for measuring affinities and rate constants for ligands (antibodies) that bind to cellular receptors but that is also suitable for measuring affinities to antigens absorbed on polystyrene dishes like ELISA.
Don Kaiser Thanks for your advice. I tried ELISA but most serial dilutios of antigen looks like saturation, so I can't calculate Kd value. How can I do?
Jeonga. You had a great collection of answers by experts that will certainly help you. Just in case, there is a simple alternative one. Cleave Fab, purify it by a single SEC step (S75) and use that for binding in solution. You can either do ITC, which can be tricky and uses substantial amounts, or label it with a fluorophore (i.e. FITC, or anything around). You will have a simple binding model, likely 1:1 stoichiometry. If you are lucky enough, a tryptophan change may monitor binding and you won't need the fluorophore, but by the high affinity you mention, it might be necessary to go for the fluorophore in order to get into dissociation regime.
It is certainly a optimization problem. You can do the kinetics study to determine dissociation constant.
Please see the link, it'll help you:
https://www.ias.ac.in/article/fulltext/jbsc/021/05/0641-0651
Jeonga Min It's possible that your antigen is not completely soluble or that it forms oligomers or aggregates. Aggregates or precipitates could cause what you described for sure. IgM isotypes can also be problematic. I assume your antibody is not an IgM? Can you tell us your antigen?
- Badges
- Science topic
- Similar topics
- Philosophy
- Moral Philosophy
- Free Will