I'm immensely grateful for Gabe's mentorship throughout my postdoc, and for our colleagues at Scripps, especially the Kelly and Forli labs.

We hope our study can further the understanding of TTR's rich biological role, as well as bolster the relevance of structural studies to probe the paramount role of protein dynamics, even in the age of fast and accurate protein structure prediction.

The partially unfolded state we observe is also consistent with amyloid formation intermediates proposed to exist more than 30 years ago, and provides possible explanation to how certain common mutations lead to TTR amyloidosis.

We think the loss of entropy (ie, the partially unfolded state) in the double-bound state explains TTR's negative cooperativity.

We also found that when both binding sites are occupied, the partially unfolded state seems to disappear. When only one of the binding sites is occupied, the partially unfolded state comes back.

We started by looking at apo TTR, and found multiple conformations indicating TTR can go a sort of accordion motion and partial unfolding (beta sandwich "fraying").

All this work was carried out in a 200keV Arctica, on gold-substrate grids coated with a single layer of graphene. This support was crucial to maintain the integrity of the sample.

BONUS: TTR is a D2 tetramer, so ligand pose has been confounded by the overall symmetry of the molecule. With cryo-EM we were able to isolate a fully-asymmetric ligand pose, all this in a 55kDa protein!

We think these states explain some odd ligand-binding behaviors of TTR, and open a window to the beginning of the process of amyloid formation.