Measuring PROTAC Ternary Complex Formation by SPR

You are here:

Measuring binary interactions (Ligase-warhead, Protein of Interest-warhead) alone is not satisfactory for progressing PROTAC drug discovery projects. Being able to measure ternary complex formation (Ligase-PROTAC-Protein of Interest) gives a much wider appreciation of the complexities and subtleties that influence PROTAC optimization.

In recent years several papers1-3 have described an SPR method to study the kinetics and affinity of ternary complex formation. They expanded upon the important concept of ‘cooperativity’ (ratio of a PROTAC’s ternary complex affinity over its binary affinity).

It has been demonstrated that a PROTAC’s SPR-derived cooperativity factor can correlate very highly with degradation of the target protein in the cell. This raises the exciting prospect of using a high-throughput, quick-to-develop SPR assay as a predictive tool for future PROTAC cellular activity.

At Charnwood Discovery we have successfully developed a similar SPR-based ternary complex assay to that described in literature. We used the E3 ligase VHL and the well-known target protein BRD4 to validate our ternary complex assay. Read on to discover how.

Challenges

A main challenge with SPR assay development is identifying protein constructs that remain functional and folded after being purified with relevant affinity tags. 

This difficulty is especially acute with E3 ligases as, to remain active, they often rely on a number of different binding partners bound in complex (e.g., Cereblon:DDB1, VHL:ElonginB:ElonginC).  

What We Did...

We immobilized VHL ligase onto the chip surface using a different immobilization strategy than previously described:

  • Measured binary binding affinity of PROTAC MZ1 against this surface
  • ~ 70 nM affinity and reasonably high surface activity – Comparable to literature values
Ternary complex assays_SPR- Binding curve

Figure 1. Multi-cycle kinetics sensorgrams showing the binding of MZ1 to our VHL surface.

Then re-tested binding of MZ1 in near-saturating concentrations of protein BRD4. 

  • BRD4 contains two different bromodomains – BD1 and BD2 – each were tested separately
  • MZ1 : BRD4 BD1 had ~ 30 nM affinity – BD1 offers no significant synergistic cooperativity to MZ1 / VHL interaction
  • MZ1 : BRD4 BD2 had ~ 2 nM affinity – BD2 adds large synergistic cooperativity effect
Ternary complex assays_SPR_MZ1 vs VHL
Ternary complex assays_SPR_MZ1 vs VHL_1

Figure 2. Single cycle kinetics (SCK) sensorgrams showing the binding of the MZ1 and BRD4 BD1 (left) or BD2 (right) complex against our VHL surface.

This experiment demonstrates that our assay format can match the data previously reported in the literature for the ternary complex between MZ1, VHL and BRD4.

We proceeded to test the binding of a different – but similar – PROTAC ARV771 against this same VHL surface. As expected we saw similar binding response level, sensorgram quality and binary affinity of ~ 60 nM as compared to the MZ1 experiment above (data not shown).

As with MZ1, we re-tested the binding of ARV771 in near-saturating concentrations of protein BRD4

  • ARV771 : BRD4 BD1 had ~ 30 nM affinity – BD1 offers no significant synergistic cooperativity to ARV771 / VHL interaction
  • ARV771 : BRD4 BD2 had ~ 4 nM affinity – BD2 adds large synergistic cooperativity effect
Ternary complex assays_SPR_ARV771 vs VHL_BR4 BD1
Ternary complex assays_SPR_ARV771 vs VHL_BR4 BD2

Figure 3. Single cycle kinetics (SCK) sensorgrams showing the binding of the ARV771 and BRD4  BD1 (left) or BD2 (right) complex against our VHL surface.

This experiment demonstrates that our assay format works well with a different PROTAC (ARV771) – not just MZ1.

Finally we tested the binding of an unrelated PROTAC that should not bind to BRD4 – called here ‘VP1’. VHL is still the target ligase of VP1 but uses a different warhead to that of MZ1 and ARV771.

The binary interaction between VP1 and our VHL surface was revealed to be much weaker than seen above – an affinity of around 440 nM was measured.

Ternary complex assays_sensorgrams VP1 to VHL

Figure 4. Multi-cycle kinetics sensorgrams showing the binding of VP1 to our VHL surface.

Again, we re-tested the binding of VP1 in excess concentrations of protein BRD4.

  • VP1 : BRD4 BD1 had ~ 870 nM affinity – We see no evidence of ternary complex formation
  • VP1 : BRD4 BD2 had ~ 1000 nM affinity – We see no evidence of ternary complex formation
Ternary complex assays_SCK sensorgrams VP1_BRD2

Figure 5. Single cycle kinetics (SCK) sensorgrams showing the binding of VP1 in the presence of excess BRD4 (left) or BD2 (right) complex against our VHL surface.

Note the same binding response level in Figure 4 and Figure 5 – indicating no ternary complex formation.

This experiment demonstrates that our assay format can distinguish negative as well as positive hits with regards to ternary complex formation.

Summary

Measuring binary interactions (Ligase-warhead, Protein of Interest-warhead) alone is somewhat simplistic for progressing PROTAC drug discovery projects – being able to measure ternary complex formation (Ligase-PROTAC-Protein of Interest) gives a much wider appreciation of the complexities and subtleties that influence PROTAC optimization.

At Charnwood Discovery we have successfully developed a similar SPR-based ternary complex assay to that described previously in literature. We used the E3 ligase VHL and the well-known target protein BRD4 to validate our approach. We were able to:

  • Measure ternary complex formation between VHL, BRD4 and PROTAC MZ1 – comparable to the previously-reported affinity values and cooperativity factors
  • Distinguish between no cooperativity and positive cooperativity for BRD4 BD1 and BD2, respectively
  • Measure the same as above for an alternative, related PROTAC ARV771
  • Correctly identify no ternary complex formation for a VHL-targeting PROTAC developed against a protein entirely different to BRD4

References

  1. Zorba A, Nguyen C, Xu Y, et al. Delineating the role of cooperativity in the design of potent PROTACs for BTK. Proc Natl Acad Sci U S A. 2018;115(31):E7285-E7292, doi.org/10.1073/pnas.1803662115
  2. Roy MJ, Winkler S, Hughes SJ, et al. SPR-Measured Dissociation Kinetics of PROTAC Ternary Complexes Influence Target Degradation Rate. ACS Chem Biol. 2019;14(3):361-368, doi.org/10.1021/acschembio.9b00092
  3. Wurz RP, Rui H, Dellamaggiore K, et al. Affinity and cooperativity modulate ternary complex formation to drive targeted protein degradation. Nat Commun. 2023;14(1):4177, doi.org/10.1038/s41467-023-39904-5