An oligomeric state‐dependent switch in the ER enzyme FICD regulates AMPylation and deAMPylation of BiP

Abstract AMPylation is an inactivating modification that alters the activity of the major endoplasmic reticulum (ER) chaperone BiP to match the burden of unfolded proteins. A single ER‐localised Fic protein, FICD (HYPE), catalyses both AMPylation and deAMPylation of BiP. However, the basis for the switch in FICD's activity is unknown. We report on the transition of FICD from a dimeric enzyme, that deAMPylates BiP, to a monomer with potent AMPylation activity. Mutations in the dimer interface, or of residues along an inhibitory pathway linking the dimer interface to the enzyme's active site, favour BiP AMPylation in vitro and in cells. Mechanistically, monomerisation relieves a repressive effect allosterically propagated from the dimer interface to the inhibitory Glu234, thereby permitting AMPylation‐competent binding of MgATP. Moreover, a reciprocal signal, propagated from the nucleotide‐binding site, provides a mechanism for coupling the oligomeric state and enzymatic activity of FICD to the energy status of the ER.

A Summary of deAMPylation rates of wild-type and mutant FICD proteins. Shown are deAMPylation rates of BiP V461F -AMP FAM by the indicated FICD proteins (at 0.75 µM or 7.5 µM) as detected by a change in fluorescence polarisation. Mean decay rate constant values AE SD from normalised raw data fitted to monoexponential decay functions of at least four independent measurements are presented. B, C The effect of FICD overexpression on a UPR reporter. (B) Flow cytometry analysis of wild-type (wt) and FICD À/À CHO-K1 CHOP::GFP UPR reporter cells transfected with plasmids encoding wild-type or the indicated FICD derivatives and a mCherry transfection marker. Shown are the median values AE SD of the GFP fluorescence signal of mCherry-positive cells from three independent experiments (fold change relative to wild-type cells transfected with a plasmid encoding mCherry alone). Note that only Glu234Gly-containing, deAMPylation-deficient FICDs activate the reporter. (C) Flow cytometry raw data of a representative experiment. D AMP production by FICD dimer interface or relay mutants is BiP dependent. AMP production in the presence of [a-32 P]-ATP was measured by TLC and autoradiography (as in Fig 2B). Plotted below are mean AMP values AE SD from three independent experiments. E-G Characterisation of covalently linked S-S FICD A252C-H363A-C421S dimers-a trap for BiP-AMP.   Figure EV2.

The EMBO Journal
Luke A Perera et al

EV4
The EMBO Journal e102177 | 2019 ª 2019 The Authors ▸ Figure EV3. FICD dimer relay mutants produce a pool of AMPylated BiP in vitro, and FICD AMPylation activity correlates with increased flexibility.
A Radioactive in vitro AMPylation reactions with the indicated FICD proteins at the indicated concentrations, [a-32 P]-ATP, and BiP T229A-V461F were analysed by SDS-PAGE. The radioactive signals were detected by autoradiography and proteins were visualised by Coomassie staining. Note the enhanced production of AMPylated BiP in the presence of dimer relay mutants, FICD K256S and FICD E242A , relative to the wild-type protein and a further increase in the production of AMPylated BiP by the monomeric FICD K256S-L258D double mutant relative to monomeric FICD L258D . Also note the auto-AMPylation signals of the monomeric FICDs detectable at high enzyme concentration. B, C In vitro deAMPylation of fluorescent BiP V461F -AMP FAM by the indicated FICD proteins (at 7.5 µM) measured by fluorescence polarisation. A representative experiment (data points and fit curves) is shown and rates are presented in Fig EV2A. Note the impaired deAMPylation activity of the monomeric FICD K256S-L258D double mutant in (C). D DSF T m analysis of wild-type (wt) and mutant FICD proteins in absence (Apo) or presence of ATP or ADP. Nucleotide concentrations in mM are given in parentheses.
Non-oxidised and oxidised forms of FICD A252C-C421S were assayed in buffer lacking reducing agent (which did not affect the T m of wild-type FICD; see source data). Shown are the mean T m values AE SD from three independent experiments. Note that FICD K256A is more stable than FICD K256S but less than wild-type FICD. Furthermore, the stabilities of oxidised and non-oxidised FICD C421S-A252C relative to the wild-type correlate inversely with their AMPylation activities (Fig 3B). For the wild-type FICD, FICD E242A , FICD G299S , FICD L258D and FICD K256S-L258D , in the apo state, the same data are presented in Fig 4E.

E
Plot of the increase in FICD melting temperature (ΔT m ) against ATP concentration as measured by DSF (derived from Fig 4F). Note the similarity in the K 1/2 s of ATP-induced T m increase (annotated) between FICD L258D (mFICD) and the wild-type dimer (dFICD). Shown are mean ΔT m values AE SD of three independent experiments with the best-fit lines for a one-site binding model. ▸ Figure EV4. Monomerisation allows ATP to bind to FICD in a mode conducive to BiP AMPylation.
A Mutation of the dimer relay residue Lys256 does not result in large conformational changes in FICD. Shown is the alignment (residues 213-407) of the molecules in the asymmetric unit. Structures are coloured as indicated. Glu234, ATP and Mg (where applicable) are shown as sticks. The inhibitory alpha helix (a inh ) and gross domain architecture is annotated. The FICD:Apo structure is from PDB: 4U04. B Electron density of both MgAMPPNP and the inhibitory Glu234, from monomeric FICD L258D co-crystallised with MgAMPPNP. Unbiased polder (OMIT) maps are shown in blue and purple meshes, contoured at 3.0 and 5.0r, respectively. All residues and water molecules interacting with MgAMPPNP are shown as sticks and coloured by heteroatom. Mg 2+ coordination complex pseudo-bonds are show in purple dashed lines. C Unlike wild-type FICD, monomeric FICD L258D binds ATP and ATP analogues in an AMPylation-competent conformation. The indicated structures and distances are shown as in Fig 5C, with ATP interacting residues shown as sticks and annotated. The position of the a-phosphate relative to Val316 in the FICD:ATP structure (see distances in right-hand side panel) would preclude in-line nucleophilic attack (see D, E). The inset is a blow-up displaying distances (i-iv) between the c-phosphates and Glu234 residues. A potentially significant difference in the Glu234 position between the FICD L258D :MgAMPPNP and FICD:ATP structures is apparent: hypothetical distance (ii) (2.68 Å, between Glu234 of FICD:ATP and AMPPNP c-phosphate of FICD L258D ) is less favourable than the observed distance (iii) (2.94 Å, between the AMPPNP c-phosphate and Glu234 of FICD L258D ). Note, His363 of FICD:ATP is in a non-optimal flip state to facilitate general base catalysis (see Fig 5B). D (i) The mode of ATP binding in wild-type dimeric FICD sterically occludes the nucleophilic attack required for AMPylation. Shown are semi-opaque 3 Å centroids centred on Pa and Val316 (Cc1). The putative BiP Thr518 nucleophile (depicted by the cross) is positioned in-line with the scissile phosphoanhydride (parallel to the plane of the paper) and 3 Å from Pa. This nucleophile position lies within the Val316 centroid (indicating a steric clash). For clarity, the FICD:ATP structure is overlaid with a thin slice of the FICD:ATP structure in the plane of the Pa-O3a bond. (ii) In the monomeric AMPylation-competent FICD L258D :ATP structure, the nucleophile lies outside the Val316 centroid in proximity to His363 (the general base). E The ATP a-phosphates of monomer or dimer relay mutants are in the same position as that competently bound to the AMPylation unrestrained dimeric FICD E234G .
Shown are all AMPylation-competent MgATP structures overlaid as in (C) and Fig 5C. The dimeric FICD E234G :MgATP (dark blue, PDB: 4U07) is also included as a reference for an active AMPylating enzyme.    Figure EV5. AMPylation activity correlates with enhanced flexibility of the dimer interface and Glu234.
The residue average B-factors, for the four FICD complexes co-crystallised with ATP, are shown [in (i-iv)] with a cold to hot colour code. They display a trend of increasing Bfactors in the dimer interface and in the inhibitory glutamate region. This increase in B-factor is indicative of increasing flexibility and correlates with greater AMPylation activity of the corresponding FICD. All of these structures have almost identical dimer packing in their respective crystals and limited crystal contacts around the inhibitory helix (see Appendix Fig S2). Note, structure averaged B-factors are comparable (see Table 1). For clarity, the TPR domain (up to residue 182) is not shown.
A Immobilised BiP responds allosterically to, is saturated by and retains ATP for the duration of BLI kinetic assays. BLI traces of the interaction between FICD L258D-H363A and immobilised biotinylated BiP T229A-V461F in different nucleotide states. Before exposure to FICD L258D-H363A immobilised BiP:Apo was subjected to two consecutive incubation steps (activation and wash) in the presence or absence of ATP as indicated. FICD association and dissociation steps (shown) were then conducted in a nucleotide (Nt.)-free solution. Note that BiP only interacts with FICD L258D-H363A when pre-saturated with ATP. Importantly, ATP pre-bound BiP retains its affinity for FICD L258D-H363A even if subsequently washed in a buffer lacking ATP (compare red and green traces). Thus, the majority of BiP retains its bound ATP for the duration of the kinetic experiment, experimentally uncoupling the effect of nucleotide on the FICD analyte from its effect on the immobilised BiP ligand. B Cartoon schematic of the BLI assays presented in Fig 6A and B. The pre-AMPylation complex is formed between the immobilised BiP:ATP "ligand" and the FICD "analyte". C The BLI association and dissociation traces from Fig 6B are shown. The immobilised biotinylated BiP T229A-V461F was saturated with ATP and then exposed to nucleotide-free FICDs. Dissociation was performed in absence or presence of ATP, as indicated. [mFICD H363A : FICD L258D-H363A ; dFICD H363A : FICD H363A ]. D Quantification of the biphasic exponential decay fitting of dissociation traces shown in Fig 6B. Relative ATP-induced changes of these kinetic parameters are given in Fig 6D. Shown are mean values AE SD from three independent experiments. Note the greater relative contribution of fast dissociation of mFICD in presence of ATP versus absence. E Representative BLI traces of an FICD dimer dissociation experiment. The legend indicates the form of unlabelled FICD incubated with the N-terminally biotinylated FICD (at a 100-fold molar excess, prior to biosensor loading) and also the ligand present in the dissociation buffer (at 5 mM) if applicable. F Representative dissociation data derived from (E). Probes loaded with biotinylated FICD incubated with mFICD H363A act as controls for non-specific association and dissociation signals, these were subtracted from the respective dFICD H363A traces in (E). Mono-exponential decay best-fit lines are also displayed; resulting off rates are shown in Fig 6E(ii). Step  None ATP ADP Figure EV6.