Fluorescence Binding Titration Experimenti Design

How to design a fluorescence binding titration experiment:

When you perform a binding experiment in fluorescence mode, the object is to follow the labeled component and measure the concentration of the free and complexed fluorescently labeled component. Of course, in a fluorescence experiment you cannot see the unlabeled component, and this makes interpretation a lot easier. The unlabeled component only becomes visible indirectly, when the labeled component is complexed with it. So we will design the experiment to see varying amounts of labeled free and complexed material.

The first thing to keep in mind is that you need sufficient signal. The more signal, the better the analysis. One of the nice things about the fluorescence detector is its sensitivity, so you often don't need very much material to get a good signal. Depending on your instrument, 50-200 nM of label could be a good range to work in. Keep in mind that you will always have a bit of free label present, even if you dialyze out the free label. In most cases, this is not even that important to do, and you can tolerate the presence of baseline contributions from free label, since you can subtract them out during analysis in UltraScan.

The second important element is which component needs to be labeled. Let's say you want to measure binding of a 5 kDa peptide to a 40 kDa protein. In such a case you will need to label the 5 kDa peptide, since there is a much bigger difference in sedimentation between a free 5 kDa species and a 45 kDa complex, than between a 40 kDa free species and a 45 kDa complex. So you always want to label the slower sedimenting species. You should also always perform a control experiment for both proteins, perhaps in unlabeled state using UV absorbance so you know which species sediments slower.

The third important element is the correct concentration of labeled and unlabeled material. The binding reaction will be somewhere between 100% free and 100% bound, depending on the Kd of the reaction and on the relative amounts of each species. The easiest ratio to see in the AUC is a situation where you have 50% bound and 50% free. To determine the appropriate concentrations, you need to have an idea about the concentrations of your two binding partner's stock solution, and it helps greatly to have a good estimate of where you should be looking for the Kd. Let's assume we have a labeled component A* and an unlabeled component B, forming a complex AB. Then the molar concentrations of those three species are given by [A*], [B], and [A*B]. At equilibrium, their concentrations can be used to get a Kd:

[A*][B]/[A*B] = Kd

To get a good signal, we want to figure out how much of A* and B to use so we can get about 10%, 30%, 50%, 70% and 90% binding, plus include a control of an unbound [A*]. Let's say we have a pretty good idea that the Kd is about 150 nM. Here is how you calculate your concentrations: First, determine the concentration of labeled A you will be using in each experiment. Let's assume we use 200 nM. This amount stays constant throughout the titration. For each titration you will now calculate the concentrations like this:

For the 30% binding titration point, you know that [A*B] will be 30% of [A*] and that there will be 70% of free [A*], so at equilibrium we will have 200*0.7 nM free [A*] and 200*0.3 nM complex [A*B]. Since we assume a Kd of 150 nM, we can write:

150 nM = [0.7*200] [B]/[0.3*200]


[B] = (150 * 0.3 * 200)/(0.7 * 200) = 64.28 nM

Of course, this is only the free [B] concentration, and we also have to add the amount of [B] that is going to be complexed. However, this amount is the same as the complex concentration we already know from the [A] concentration, so we need to add 0.3 * 200 = 60 nM to our 64.28, so the amount of [B] that needs to be added to achieve a 30% binding would be 124.28 nM when we start with a 200 nM amount of labeled [A*]. You can repeat the other ratios as an excercise.