Fig. 8: Model of FICD AMPylation and deAMPylation of BiP.

FICD’s TPR domain and catalytic Fic domain recognise, respectively, the linker-docked NBD and the ℓ7,8 region of the SBDβ of either AMPylated or unmodified BiP. Simultaneous engagement of both interfaces is only possible when BiP is in a domain-docked ATP-like state. Dimeric FICD has a relatively rigid gatekeeper Glu234 that facilitates efficient alignment of an attacking water for BiP deAMPylation and inhibits AMPylation competent binding of ATP. Enhanced flexibility of monomeric FICD’s Glu234 decreases deAMPylation efficiency whilst permitting AMPylation competent binding of MgATP by monomeric FICD. We speculate that the FICD monomer-dimer equilibrium is adjusted in response to changing levels of unfolded proteins within the ER by processes which may include a direct response to changes in the ER energy status (ATP/ADP ratio)19. Further details of the FICD-catalysed deAMPylation reaction are presented in Supplementary Fig. 10.