Assay setup for cooperativity assessment (e.g. PROTACs, molecular glues)
This guide focuses on how to use ternary complex formation to detect the cooperativity of binding by changes to the binding affinity of a ternary compared to a binary interaction. The underlying assumption is the presence of one central molecule which binds both outer molecules. These two molecules do not directly interact with each other.
One main application for this setup are PROTACs, like used in this BRD2-MZ1-VHL protocol. But the assay can be used for all interactions where the binding of a third molecule is suspected to non-competitively change the affinity of the initial binary interaction.
Step 1: Measuring the binary interaction (AB)
Throughout the whole assay, the central binding molecule (further referred to as B) is titrated and either one of the outer molecules can be labeled (further referred to as labeled target A).
What needs to be studied first, is if and how the labeled molecule (A) binds molecule (B). For this, a constant concentration of (A) is added to a dilution series of (B) to generate a dose response curve as seen in Figure 1. The binding information is vital for later cooperativity assessment. Moreover, the affinity should be sufficiently high to reach a saturated bound state at the highest available ligand concentration.
Figure 1. Spectral shift dose response curve of the binary interaction of the fluorescently labeled molecule A with B (Kd(AB) = 61 nM).
Step 2: Repeat the binding experiment of B to A in the presence of a constant concentration of ternary ligand (C)
In the second step, the same constant concentration of labeled protein (A) and dilution series of binary ligand (B) as for the binary assay are used, while an additional constant concentration of the third interaction partner (C) is added. A and C can be mixed beforehand and can be treated as the combined target. Since the independent variable is the same concentration series of binary ligand (B) in both setups, the two obtained curves can be overlayed to create a comparison graph (i.e. binary and ternary interaction plotted over the same axis). This enables to easily visualize differences, like in the dose response plot below (Figure 2).
Figure 2. Dose response curve of binary interaction (AB) in blue and ternary interaction in magenta. By using the same titration scheme for B, both curves can be overlayed and compared. The changes in affinity are caused by cooperative binding. (Kd(A-B) = 11 nM, Kd(A-BC) = 1 nM à Cooperativity α = 11). Note how at low B concentrations the ratio values are similar, which indicates no interaction between A and C (negative control).
If the availability of the third interaction partner (C) is not limiting, a very high concentration (even higher than the highest B concentration) can be used. These saturating conditions will ensure that there is always enough ternary ligand (C) to form the ternary complex.
Since ligand (B) typically is diluted to much lower concentrations as A or C in the assay, this assay setup also serves as a negative control (Figure 3) to exclude the interaction between labeled target (A) and third interaction partner (C). If, however, the ratio values between binary and ternary interaction change at low concentrations of B, then this is an indication for an interaction of A and C.
Figure 3. Ternary interaction where A and C are kept at constant concentration while B is titrated. The constant concentration of C is equal or higher than the highest B concentration. This is visualized in the graphs below the pictogram. At high B concentrations a ternary complexed is formed. Low B concentration data can be used as a negative control experiment for an interaction between A and C.
If a lower concentration of the third molecule is chosen, one might be able to observe the Hook effect (Figure 4). Here, the reporter molecule (A) starts out in the unbound state, forms a ternary complex at medium concentration of the binary ligand (B) and then mostly ends up in a binary complex, when B is present in excess. This can cause a biphasic dose response curve if the spectral shift ratio of ABC is larger than the ratio of AB.
Figure 4. Dose response curve of binary interaction (AB) with Kd-fit in blue. Ternary interaction (A-BC) with 500nM C in magenta and ternary interaction (A-BC) with 25nM C in yellow where both fitted with Hook-model. The directional change of the blue dose response curve observed around 50nM B is called “Hook-Effect”. It is caused as the observed target molecule A ensemble transitions from a majority of ternary molecules (ABC) to the binary complex AB once most of C is saturated.
Be aware that the Hook-Effect is normally not the main desired experimental outcome. Figure 5 should mainly highlight a possible setup which can lead to this outcome and may sometimes not be preventable.
Figure 5. Ternary interaction where A and C are kept at constant concentration while B is titrated. The constant concentration of C is much lower than the highest B concentration. This is visualized in the graphs below the pictogram. At medium B concentrations most of A is part of ternary complexes (not visualized). At increasing concentrations of B, the equilibrium shifts to binary complexes of AB and BC. This is called Hook-Effect.