Authors

S Routledge¹; R Brindley¹; M Azam¹; C Box¹; K Thompson¹; 
¹ Gifford Bioscience Ltd, UK

Discussion

Receptor-ligand interaction is traditionally modelled as single-site 1:1 binding. However, multimeric ligands can undergo multiple linked binding interactions where receptors are closer together in a cell membrane. Ligand dissociation is then dependent on the simultaneous dissociation of all binding interactions. These binding interactions appear to have a very high affinity - an effect called avidity. Bivalent antibodies exploit avidity and remain tightly bound to their antigens. Here we utilise two different biosensor approaches, Surface Plasmon Resonance (SPR) and Bio-Layer Interferometry (BLI) to measure multi-site avidity of trimeric SARS-CoV2 spike protein (spike trimer) binding to human lung cell angiotensin converting enzyme 2 (ACE2) protein. ACE2 was immobilised at a high density on the SPR and BLI biosensor surfaces, and spike trimer or monovalent spike binding domain (spike RBD) binding compared. Spike trimer binding demonstrated avidity, binding to adjacent ACE2 proteins with an apparent high affinity (K_D < 1 pM in SPR and BLI). In contrast, spike RBD binding to ACE2 does not demonstrate avidity, and responses fitted to a single-site binding model with kinetic affinity constants (K_D) of 6.1 nM (SPR) and 4.8 nM (BLI). We used alternative configurations to avoid avidity in the spike trimer-ACE2 interaction. Immobilising low levels of ACE2 (10-fold reduction) revealed single-site binding affinity at ACE2 of approximately 2 nM. Additionally, in the reverse orientation with spike trimer immobilised on the biosensor, binding of monovalent ACE2 to spike trimer gave K_D values of 0.3 nM (SPR) and 1.5 nM (BLI). Utilising SPR and BLI biosensor approaches, we were able to determine the titre of anti-COVID-19 antibodies in serum and demonstrate avidity in the spike trimer-ACE2 interaction.