Plasmonic Imaging of Dynamic Interactions between Membrane Receptor Clusters beyond the Diffraction Limit in Live Cells

As a general mechanism, ligand-induced receptor clustering on cell membrane plays determinative roles in pattern recognition and transmembrane signaling. Nevertheless, probing the dynamic characteristics for the complicated interactions between receptor clusters remains difficult because of the lack of strategy for receptor cluster labeling and long-term monitoring in live cells. Herein, we proposed a data-mining-integrated plasmon coupling microscopy to study the dynamic cluster-cluster interactions on cell surface. The receptor clusters were activated and labeled with multivalent plasmonic nanoprobes, which enables the real-time monitoring of individual receptor clusters and the measurement of cluster-cluster interactions from the analysis of plasmonic coupling for the nanoprobe pairs beyond the diffraction limit. Using this method, we found that the protease-activated receptor 1 (PAR1) clusters would experience an initial contact and then form a weakly bound cluster-cluster complex, followed by cluster fusion to generate large-sized signaling complexes. The underlying state transitions for the cluster-cluster fusion process were uncovered using a data-mining technique named the K-means-based hidden Markov model with the scattering intensity of coupled nanoprobe pairs as observations. All of the findings from single-particle analysis and bulk measurements suggested that the allosteric inhibitors could suppress the dynamic transitions from the weakly bound cluster-cluster complexes to fused signaling complexes, leading to the subsequent downregulation of intracellular calcium signaling pathways. We believe that this strategy is promising for imaging and monitoring receptor clustering as well as protein phase separation on the cell surface in various biological and physiological processes.