Figure 2. Structure of the γ-Secretase E-S complex at the transition state and effects of FAD PSEN1 mutations.

(A) Schematic of substrate-based structural probe design and specific probe SB-250.

(B) The overall cryo-EM structure of SB-250 bound to WT γ-Secretase complex. SB-250 is shown in sticks.

(C) The conformation of active PSEN1 (cyan) in complex with SB-250 (red) shows high similarity with catalytically inactive PSEN1 (yellow) covalently cross-linked to APP-C83 (marine). Superimposition of these two PSEN1 molecules reveals a root-mean-square distance (RMSD) of 0.428 Å over 250 Ca atoms.

(D) The structural probe traps active γ-Secretase at the transition state of intramembrane proteolysis. Note the rotation about the Cα–Cβ bond of catalytic residue D257 in SB-250-bound active γ-Secretase compared with APP-C83-bound inactive enzyme (see curved arrow), positioning the aspartyl carboxylate for coordination with the transition-state-mimicking hydroxyl group of SB-250.

(E) Overlap of the cryo-EM structure of substrate mimetic probe bound to γ-Secretase with the activated E-S complex from GaMD simulations reveals close similarity of the alignment and orientation of the active-site aspartates, with respect to one another as well as with substrate/mimetic. Catalytic D257 and D385 are green for cryo-EM structure and yellow for GaMD simulation.

(F) GaMD simulations of γ-Secretase bound to C99 substrate show overall reduced flexibility with FAD PSEN1 mutations, implying complex stabilization. Root-mean-square fluctuations (RMSFs) of PSEN1 and C99 substrate within E-S complexes were calculated by averaging the RMSFs from individual MD simulations of each γ-Secretase system. Changes in RMSFs (ΔRMSF) from WT to FAD-mutant PSEN1 were calculated by subtracting the RMSFs of the WT E-S complexes from those with FAD-mutant PSEN1.

(G) Free-energy profiles of WT and six FAD PSEN1 mutations from GaMD simulations of the Aβ49 → Aβ46 trimming step.