Consistent with this, we found that a constitutionally active ADGRE1 (ADGRE1-b – lacking most of the extracellular domain but maintain the tethered agonist sequence) promoted condensates, unlike the apo receptor (ADGRE1 FL) or a receptor without tethered agonist (ADGRE-1 bT).

While the beta-arrestins in this 2:2 complex were not directly interacting, they were in a conformation that could promote an N-N interaction that was identified through metadynamics. This suggested that receptor activation could promote beta-arrestin oligomerization.

Serendipitously, Peng Xiao and Jinpeng Sun solved the structure of an adhesion receptor that forms tight complexes with beta-arrestins. Surprisingly, this structure displayed a 2:2 complex with the beta-arrestin at a unique angle compared to other GPCR:beta-arrestin structures.

Previous work from the Marullo, Benovic, and Gurevich labs identified specific sites in beta-arrestins that bind inositol hexaphosphate (IP6) to regulate beta-arrestin oligomerization. We found those mutations impacted beta-arrestin-dependent internalization and GPCR signaling.

We then used an optogenetic approach to test if these were consistent with condensates. When Cry2 undergoes light-induced oligomerization, the presence of specific sequences (such as IDRs) leads to condensate formation (see FUSN). Beta-arrestin 1 had a similar effect!

We performed FRAP that demonstrated that proteins in these puncta exchanged with solution, although with slightly slower kinetics than beta-arrestin 2 in the cytosol.

It has been known for decades that beta-arrestins oligomerize, but the relevance of this in a cellular context is unknown. Labeling beta-arrestin 2 with GFP results in a cytoplasmic pattern, while labeling with components of split GFP results in punctae, consistent with oligomers.