Fig. 7: Quantification of intermediate populations in ex vivo AAT variants.

a Overlays of ¹H,¹³C SOFAST-gHMQC spectra (isoleucine region) for free M (blue), Z (magenta) and S (green) AAT, and 716-bound M (cyan) and Z (orange) AAT, with selected resonance attributions and colinear chemical shift perturbations indicated. Contour levels are normalised for concentration and acquisition time. b Schematic illustration of the origin of collinear chemical shift perturbations, arising from a rapid equilibrium between native (N) and intermediate (I) states. c Chemical shift perturbations of AAT resonances relative to wild-type M AAT, plotted against chemical shift changes in M AAT upon 716 binding. Lines of best fit through the origin are shown. ¹H and ¹³C chemical shift differences are plotted independently and ¹³C measurements have been scaled by a factor of 0.25. Resonances near sites of mutation or 716 binding showing evidence of local perturbations to their chemical environment have been excluded from this analysis. d The conformation selection reaction scheme. A dotted box highlights the intramolecular equilibrium, while the apparent dissociation constant of the ligand is determined relative to the total concentration of unbound species, \(K_{\mathrm{{d,app}}}=([\mathrm{N}]+[\mathrm{I}])/[\mathrm{B}]\). e Free-energy diagram of AAT showing native (N), intermediate (I) and 716-bound (B) states under standard conditions. f Populations of the intermediate state in wild-type and variant AAT. Error bars indicate the standard error determined from fits. g Cartoon illustration of the inhibition of AAT misfolding and polymerisation (P) via the binding and stabilisation of an on-pathway intermediate state by a small-molecule ligand. Source data are provided as a Source Data file.