Changes in the disposition of BiP induced by manipulation of conditions in the ER are conveniently tracked by native gel electrophoresis (native-PAGE) and immunoblotting (Freiden et al., 1992) coupled with site-directed proteolysis (Preissler et al., 2015) (Figure 1A). Inhibition of protein synthesis led to accumulation of a prominent high mobility species tentatively named ‘B’ form, whereas depletion of ER calcium progressively drew on this pool to promote the assembly of BiP oligomers (Preissler et al., 2015) (Figure 1B and C). The ‘B’ form of BiP was also noted for its relative resistance to cleavage in vitro by a bacterial protease, SubA, which cuts BiP within a highly conserved linker sequence connecting the nucleotide binding domain (NBD) and substrate binding domain (SBD) (Paton et al., 2006) (Figure 1A and D). Resistance to cleavage by SubA is also a feature of the acidic, modified form of BiP (Chambers et al., 2012), suggesting that the ‘B’ form observed on native-PAGE might be related to the covalently modified form of BiP (previously believed to be the ADP-ribosylated form).

Figure 1

Download asset Open asset

To examine the potential role of FICD in elaboration of the ‘B’ form of BiP, we inactivated both alleles of FICD in hamster CHO-K1 cells by CRISPR-Cas9-mediated genome editing, truncating the coding sequence N-terminal to the FIC-domain active site (Figure 2A). The SubA-resistant ‘B’ form, prominent in cycloheximide-treated wildtype cells, was conspicuously absent in FICD^-/- cells (compare lanes 2 and 5 in Figure 2B), whilst other aspects of BiP’s structural dynamism (such as oligomerization in response to ER calcium depletion) remained largely unchanged. FICD inactivation also led to disappearance of the acidic form of BiP, present in lysates of cycloheximide-treated wildtype cells resolved by isoelectric focusing PAGE (IEF-PAGE, Figure 2C). FICD-mediated AMPylation is strongly autoinhibited by intramolecular competition for ATP γ-phosphate binding by glutamate 234, whose mutation to glycine (FICD^E234G) de-represses FICD, whereas replacement of histidine 363, which is located in the highly conserved catalytic FIC motif, with alanine (FICD^H363A) abolishes AMPylation activity (Engel et al., 2012). In keeping with these observations, enforced overexpression of the constitutively active FICD^E234G led to the re-appearance of an acidic form of BiP in the FICD^-/- cells (Figure 2D). Wildtype FICD did not promote BiP’s acidic form, even when strongly overexpressed, suggesting the existence of robust inhibitory mechanisms restraining potentially deleterious effects of protein modification (Engel et al., 2012). Such tight regulation is a feature common to FIC enzymes (Garcia-Pino et al., 2014) but the signals leading to derepression under physiological circumstances are unknown so far.

Figure 2 with 1 supplement see all

Download asset Open asset

The FICD^-/- cells provided a convenient experimental reference against which to gage time dependent changes in the abundance of modified BiP following the imposition of ER stress and during its resolution in wildtype cells. A decline in ‘B’ form and an increase in susceptibility to digestion by SubA were noted within 2 hr of exposure of cells to the reversible ER stress inducing agent 2-deoxyglucose ( Figure 2—figure supplement 1A and B) and persisted for up to 8 hr thereafter. This decline in ‘B’ form occurred in the face of a marked increase in FICD mRNA (Figure 2—figure supplement 1D) (Ham et al., 2014; Sanyal et al., 2015), a finding consistent with regulatory mechanisms that contravene the increase in mRNA. Washout of 2-deoxyglucose was associated with progressive increase in ‘B’ form and the emergence of resistance to cleavage by SubA (Figure 2—figure supplement 1A and C).

These changes fit well the previously observed inverse correlation between levels of ER stress and the abundance of modified acidic BiP (Chambers et al., 2012; Laitusis et al., 1999). Together with evidence that FICD was both necessary and sufficient for elaboration of the acidic, modified form of BiP on IEF-PAGE, these observations lent strong support to the notion that the ‘B’ form of BiP, observed on native gels, reflects the same or a related species. The high mobility of the FICD-dependent ‘B’ form and its indifference to the effects of ATP (Figure 2B, lower panel) (which promotes dissociation of BiP from its client proteins in vitro) are consistent with a role for the FICD-mediated modification in BiP inactivation. To examine these relationships in further detail we measured the effects of FICD on BiP in vitro, in a system constituted of pure components.

Whether purified as a GST-fusion protein from bacteria or from overexpressing mammalian cells, active FICD^E234G promoted the ATP-dependent in vitro appearance of a ‘B’ form of recombinant BiP purified from bacteria. Like the endogenous ‘B’ form, found in cells, the ‘B’ form constituted in vitro was also partially resistant to cleavage by SubA (Figure 3A). Emergence of the ‘B’ form in vitro correlated with the acquisition of a faster migrating acidic form of BiP on IEF-PAGE (Figure 3B). FICD converted all the detectable BiP to a single acidic form in a time-dependent manner, which, like the endogenous acidic from of BiP, was also relatively resistant to cleavage by SubA (Figure 3B, lane 8). Likewise, addition of purified active FICD^E234G to lysates from FICD^-/- cells entirely converted endogenous BiP into the acidic form (Figure 3C). Of note, within the resolution of IEF-PAGE, there is no evidence for heterogeneity in BiP modification [also see (Carlsson and Lazarides, 1983; Laitusis et al., 1999)]. This observation is consistent either with processive modification of BiP or modification occurring on a single site in any given BiP molecule.

Figure 3 with 1 supplement see all

Download asset Open asset

FICD efficiently transferred the ³²P label from the alpha phosphate of ATP onto BiP, consistent with AMPylation (Figure 3D). Radiolabeled BiP too was resistant to cleavage by SubA indicating that BiP radiolabeled in vitro by FICD^E234G reports faithfully on the behavior of endogenous BiP (Figure 3—figure supplement 1). Interestingly, cleavage of radiolabeled BiP (by prolonged incubation with SubA) led to emergence of a radiolabeled substrate binding domain fragment (Leu⁴¹⁷-Leu⁶⁵⁴); the nucleotide binding domain fragment, though conspicuous on the Coomassie-stained gel, was entirely devoid of label (Figure 3D and Figure 3—figure supplement 1B). This in vitro labeling pattern fits well with the metabolic labeling of BiP by ³²P orthophosphate or ³H adenosine label donors in vivo, as the label was mapped to a single CnBr cleavage fragment spanning Thr⁴³⁴ to Met⁵⁴¹ (Chambers et al., 2012; Gaut, 1997). In contrast, FICD-mediated transfer of ³²P label from ATP to BiP’s substrate binding domain, observed here in vitro, can not be reconciled with reported AMPylation of Thr³⁶⁶ or Ser³⁶⁵ (Ham et al., 2014; Sanyal et al., 2015).

The migration of purified recombinant BiP on IEF and native gels suggests that most or all of the protein was modified in vitro with FICD^E234G, yielding homogenous preparations of modified BiP, whereas BiP treated with the inactive mutant FICD^E234G-H363A remained unmodified (Figure 3A and 3B). These preparations were examined by electrospray ionization mass spectrometry and produced molecular masses of 71,060.1 and 71,395.6 Da for unmodified and modified BiP, respectively (Figure 4A and Figure 4—figure supplement 1). These two species were the only proteinaceous ions detected in the samples; the trailing shoulder of the ion-current traces (Figure 4—figure supplement 1A) being comprised entirely of singly charged low molecular mass contaminants. The mass difference of 335 Da is consistent with a single covalent modification by AMP (329.12 Da) in modified BiP. The difference between the predicted mass of a molecule of AMP bound via an ester linkage (329.12 Da) and the observed mass difference between the modified and unmodified BiP (335 Da) is reflective of the broad peaks associated with these measurements. These observations indicate a single quantitative modification of BiP by AMPylation and also fit the simple pattern of modified BiP migration on the IEF-PAGE gel. An analogous mass spectrometry experiment with endogenous BiP immunopurified from cycloheximide-treated wildtype CHO-K1 cells also revealed only two molecular masses (70,538.3 Da and 70,866.1 Da) with a mass difference of 327.8 Da, consistent with modification of BiP by a single AMP moiety in vivo. BiP from AMPylation-deficient FICD^-/- cells gave rise to a single species with a molecular mass (70,532.2 Da) that is similar to the predicted mass of unmodified mature BiP (Figure 4—figure supplement 2).

Figure 4 with 3 supplements see all

Download asset Open asset

To identify the AMPylation site(s), samples of unmodified BiP or BiP modified in vitro with FICD^E234G were treated with several proteases and the peptide digests were analyzed by liquid chromatography and tandem mass spectrometry (LC-MS/MS). A unique peptide (doubly charged peak of 1374.15 m/z) corresponding in mass to an AMPylation of residues 511-532 was observed in samples digested with Arg-C, and another peptide (doubly charged peak of 916.94 m/z) corresponding to AMPylation of residues 515-528 was detected in the Asp-N digest (Figure 4B and C). Moreover, another doubly charged peak (1209.62 m/z) corresponding to the unmodified BiP 511-532 fragment was present in the Arg-C digests of unmodified BiP and absent from the spectrum of the modified form (Figure 4C, left panels) enhancing the confidence in these assignments.

No masses corresponding to peptides overlapping BiP 511-532 [the region identified in the Arg-C and Asp-N digests to contain AMPylation site(s)], were detected in either the chymotrypsin or trypsin digests of AMPylated BiP. However, all four digests contained peptides covering Ser³⁶⁵ and Thr³⁶⁶, but these were all unmodified peptides (Figure 4—figure supplement 3). In contrast to the unmodified 511–532 Arg-C fragment that was absent from the spectrum of the modified BiP (Figure 4C), in all four digests the relative intensity of the signals from peptides encompassing Ser³⁶⁵ and Thr³⁶⁶ was undiminished by modification with FICD^E234G (Figure 4—figure supplement 3B); this despite evidence that all the detectable BiP molecules in the sample were modified (Figure 3B and 4). The mass spectrometry-based evidence for non-modification of Ser³⁶⁵ or Thr³⁶⁶ also fits with the observation that under the in vitro conditions studied here, FICD^E234G (using α-³²P-ATP as a substrate) selectively modified a BiP fragment C-terminal to the SubA cleavage site (at Leu⁴¹⁶) with no evidence for modification of the N-terminal nucleotide binding domain (Figure 3D and Figure 3—figure supplement 1).

The LC-MS/MS fragment spectra of AMP modified peptides from Arg-C and Asp-N digests obtained by high energy collisional induced dissociation (HCD) or electron transfer dissociation (ETD) did not directly identify the modified residue(s). However, the specificity of FICD for AMPylation of hydroxyl side chains and the overlap of modified peptides from the two digests narrows the modification site to three residues: Thr⁵¹⁸, Thr⁵²⁵ and Thr⁵²⁷. Both BiP^T525A and BiP^T527A remained substrates for AMPylation in vitro, whereas the BiP^T518A mutation abolished all modification of BiP (Figure 5A and B).

Figure 5

Download asset Open asset

BiP^T366A and BiP^R470E or BiP^R492E [the latter are mutations that attenuate labeling of FLAG-tagged BiP in vivo by ³²P orthophosphate or ³H adenosine (Chambers et al., 2012)] were all modified by FICD^E234G in vitro with varying efficiency, as was the substrate binding-deficient mutant BiP^V461F (Figure 5C). The substitution of Thr⁵¹⁸ by either alanine or glutamic acid moderately impaired the ability of BiP to form oligomers and slightly modified the mobility of the monomeric (‘A’) form of BiP on native gels, whereas mutating Thr³⁶⁶ had no effect on oligomerization (Figure 5D). BiP^T518E could not be AMPylated in vitro (Figure 5C). Yet, whether exposed to FICD^E234G or not, BiP^T518E resembled AMPylated BiP by its migration on native-PAGE and by its resistance to SubA-mediated proteolytic cleavage (Figure 5D). These observations are consistent with the bulky and negatively charged side chain of glutamic acid mimicking some aspect of AMPylation at Thr⁵¹⁸, promoting a linker-protected conformation. Together these observations support Thr⁵¹⁸ as the only FICD-mediated AMPylation site in vitro and suggest that changes at Thr⁵¹⁸ introduced by AMPylation (or mutation) have significant structural and functional consequences.

To evaluate sites of FICD-mediated BiP modification in vivo, we took advantage of a CHO-K1 cell line in which the FICD gene had been inactivated by CRISPR-Cas9-mediated gene editing (Figure 2). Wildtype and FICD knockout CHO-K1 cells were cultured in media containing arginine with stable isotopes of either "light" or "heavy" nitrogen and carbon (resulting in a net mass difference of 10 Da per arginine residue) and exposed to cycloheximide (to enhance FICD-dependent modification of BiP). Endogenous BiP was recovered by immunoaffinity purification, digested with proteases and the resultant peptides were subjected to LC-MS/MS analysis. The SILAC (stable isotope labeling with amino acids in cell culture) procedure was designed to quantify relative differences in the abundance of unmodified and modified species in differentially labeled cultures of cells (Figure 6A).

Figure 6 with 2 supplements see all

Download asset Open asset

Peptides from Arg-C digests corresponding in mass to the unmodified and AMPylated BiP^511-532 were noted in spectra derived from both untreated and cycloheximide-treated wildtype cells. The unmodified species was 1.5-fold more abundant in the untreated sample, whereas the AMPylated species was 1.5-fold more abundant in the cycloheximide-treated sample (Figure 6B). Unmodified BiP^511-532 was 2.5-fold more abundant in the cycloheximide-treated FICD knockout sample than in the cycloheximide-treated wildtype sample, whereas no modified BiP^511-532 was detected in the cycloheximide-treated FICD knockout sample (Figure 6C). These figures attest to a large fraction of AMPylated BiP in CHO-K1 cells growing in SILAC media (especially when compared to cells growing in complete media, Figure 2C); perhaps a reflection of depletion wrought by dialysis of the serum.

The region encompassing Ser³⁶⁵ and Thr³⁶⁶ was well represented by peptides from the Arg-C digest. The quantification data showed no change in the abundance of the unmodified Arg-C peptide encompassing Ser³⁶⁵ and Thr³⁶⁶ (BiP^337-367, Figure 6—figure supplement 1). The contrast between the consistent cycloheximide-dependent and FICD genotype-dependent variation in abundance of the peptide containing Thr⁵¹⁸ and absence of change in the abundance of peptides containing Ser³⁶⁵ and Thr³⁶⁶, indicates that the latter residues are unlikely to be major sites of BiP AMPylation in CHO-K1 cells.

The mass spectra of the intact protein indicates that FICD modifies any given BiP molecule on a single site both in vitro and in vivo (Figure 4 and figure supplements) and the peptide fragment analysis indicates that the modification site(s) are all encompassed by peptides spanning 511-532 (Figure 6 and figure supplements). The fragmentation spectra of Arg-C peptides 511-532 from both endogenous BiP AMPylated in vivo, or recombinant BiP AMPylated in vitro, yielded a collection of y-ions and b-ions that included all the residues on the 511-532 peptide except Thr⁵¹⁸, whereas the latter residue was conspicuously represented in the ion series from the fragmentation spectrum of unmodified endogenous and recombinant BiP (Figure 6—figure supplement 2). The HCD procedure likely fragments the modified peptide in an unpredictable manner, rendering it unrecognizable, however, in relief, the fragmentation pattern reveals Thr⁵¹⁸ as the only modification site of BiP, in vitro and in vivo.

In vitro radiolabeling revealed that amino acid substitutions distant from the primary AMPylation site (Thr⁵¹⁸) notably influenced modification efficiency. Such point mutations might have indirectly influenced the accessibility of Thr⁵¹⁸ to FICD by altering the ability of BiP proteins to interact with other molecules or by changing their overall conformation. In particular, we observed stronger modification of BiP^V461F and BiP^T366A compared to wildtype BiP (Figure 5C). The V461F mutation may enhance access of FICD to BiP simply by reducing competing oligomerization interactions amongst BiP proteins. In contrast, Thr³⁶⁶ is located in the nucleotide binding domain and mutations at this position may interfere with ATP binding or hydrolysis and thus affect the conformation of BiP. The relationship between BiP’s conformation and the efficiency with which it is modified by FICD was systematically tested by analysis of a series of well-characterized BiP mutants that favor certain conformational states based on their altered abilities to interact with and hydrolyze nucleotides (Gaut and Hendershot, 1993; Petrova et al., 2008; Wei et al., 1995).

BiP^E201G and BiP^T229A, mutants that bind ATP and undergo allosteric transitions but are defective in nucleotide hydrolysis (Gaut and Hendershot, 1993; Petrova et al., 2008; Wei et al., 1995), exhibited enhanced AMPylation rates compared to wildtype BiP (Figure 7A). By contrast, BiP^T37G, which is defective in ATP hydrolysis and in the allosteric transitions upon ATP binding, showed slightly reduced AMPylation, whereas mutants that are severely defective in adopting the ATP-bound conformation, such as the ATP binding-deficient BiP^G226D or BiP^ADDA, which is locked in the domain-uncoupled state due to a four residue substitution in its interdomain linker (Laufen et al., 1999; Preissler et al., 2015), remained unmodified by FICD (Figure 7A and B). Consistent with an important role for the conformation of BiP in determining its ability to serve as a substrate for FICD, the isolated BiP substrate binding domain (purified from bacteria as a fusion to yeast Smt3) was not modified by FICD^E234G (Figure 7—figure supplement 1).

Figure 7 with 2 supplements see all

Download asset Open asset

Kinetic differences in modification rates of mutant BiP molecules, suggested by the radiolabeling experiments, were further confirmed by tracking the time-dependent formation of the ‘B’ form of BiP following exposure to FICD^E234G (Figure 7C). The t_1/2max values for in vitro AMPylation of BiP^T229A (4.0 min) and BiP^E201G (2.3 min) were significantly shorter than for wildtype BiP (9.9 min) (Figure 7D). ATP is a substrate for FICD, nonetheless, these important kinetic differences in modification were unlikely to reflect the trivial consequence of varying contributions of ATP hydrolysis (by BiP) to the availability of substrate nucleotide for the AMPylation reaction. Accelerated modification was not shared by the ATP hydrolysis-defective BiP^T37G. Furthermore, BiP AMPylation was performed in the presence of millimolar ATP, whereas the affinity of FICD^E234G for nucleotide is measured in the 100 nM range (Bunney et al., 2014). Together, these observations indicate that modification of BiP is modulated by its conformation and BiP in the compact ATP-bound conformation is the preferred substrate for FICD. The recently determined crystal structure of human BiP substrate binding domain supports this conclusion in that the loop encompassing the Thr⁵¹⁸ AMPylation site is stabilized by six polar interactions present in the apo/ADP conformation that are lost in the ATP conformation (Yang et al., 2015), freeing the Thr⁵¹⁸ side-chain to react with the active site of FICD (Figure 7—figure supplement 2).

Impaired oligomerization of AMPylated BiP and its restricted conformational flexibility (reflected in resistance to cleavage by SubA) suggest that the modification affects BiP function and activity. Hsp70s have low basal ATP hydrolysis rates that are stimulated by J protein co-factors (Liberek et al., 1991). Therefore, we compared the basal and J protein-stimulated ATPase activity of unmodified BiP and BiP modified to completion. AMPylation lowered the basal ATP hydrolysis rate of BiP by ~50% (Figure 8A and B) and, while the ATPase activity of unmodified BiP was enhanced by the presence of a J-domain in a concentration-dependent manner, AMPylated BiP was almost entirely resistant to J-mediated stimulation of ATP hydrolysis (Figure 8C).

Figure 8

Download asset Open asset

Figure 8—source data 1

Data from three independent repeats (each performed in triplicates) of the experiment presented in Figure 8E are shown.

The insert on top of each graph shows the absolute fluorescence polarization (FP) signals in arbitrary units (A.U.) of a reference sample containing only free fluorescent substrate peptide, which were used to create normalized FP traces of samples containing BiP + peptide. The initial values (after reference signal subtraction) were set to 100%. The fit to a single phase decay curve (tabulated here) was better than to a two phase model. The fit values from the three experiments were used to calculate the average values for “k_off” and the half-lives. Experiment 3 is shown in Figure 8E.

https://doi.org/10.7554/eLife.12621.020

Download elife-12621-fig8-data1-v2.zip

AMPylation barely influenced steady-state binding of substrate peptide to BiP: The dissociation constants of the complex between unmodified BiP or AMPylated BiP and a well-characterized substrate peptide, HTFPAVLGSC, measured at equilibrium (in the presence of ADP) by fluorescence polarization, were similar (K_d^BiP 16.7 µM and K_d^BiP-AMP 11.5 µM; Figure 8D) and within the range previously reported (Chambers et al., 2012; Marcinowski et al., 2011). However, challenge with excess of unlabeled peptide revealed a markedly higher “off” rate of the complex between AMPylated BiP and the bound peptide (k_off^BiP 0.037 ± 0.006 min^-1 vs. k_off^BiP-AMP 0.212 ± 0.021 min^-1; Figure 8E).

These observations support the view that AMPylated BiP preferentially populated an ATP-like state with high substrate “on” and “off” rates, even when associated with ADP, whereas in the ATP-bound state AMPylated BiP was rendered inactive by its inability to respond to J protein co-factor.

AMPylation weakens the interaction of BiP with its substrates and was therefore hypothesized to reverse BiP’s ability to repress the UPR transducers - whether imposed directly by binding to their regulatory lumenal domains (Bertolotti et al., 2000) or indirectly by competing for unfolded protein substrates (Pincus et al., 2010), BiP repression of the UPR involves stable engagement of clients in its substrate binding domain. To test the derivative prediction that enforced expression of an active FICD would lead to UPR activation, wildtype FICD or a catalytically-dead or active FICD^E234G were introduced by transient transfection into CHO-K1 cells bearing an integrated CHOP::GFP UPR reporter. The dormant reporter was activated by tunicamycin-induced ER stress (Figure 9A, panels 1 and 2; a positive control), and by acquisition of a constitutively active FICD^E234G but not by a catalytically dead FICD^E234G-H363A nor by the regulated wildtype enzyme (Figure 9A, panels 3-5). The robustness of these observations and the independence of UPR activation from endogenous FICD – an important point given the evidence for FICD dimerization (Bunney et al., 2014) – are attested to by similar observations made in FICD knockout cells (Figure 9—figure supplement 1).

Figure 9 with 1 supplement see all

Download asset Open asset

The induction of the UPR reporter in CHO-K1 cells was mirrored by the behavior of endogenous BiP, whose levels increased in a time-dependent manner in human HEK 293 cells upon activation of a conditional allele encoding FICD^E234G (Figure 9B). Qualitatively similar observations were made in two separate clones of FICD^E234G-expressing cells [Figure 9C; the conspicuous basal (Dox-independent) SubA resistance of endogenous BiP, lanes 5 and 7, likely reflecting basal leakiness of the expression system]. These findings are readily explained by the engagement of a feedback mechanism (the UPR) that defends a level of active BiP and compensates for the accumulation of inert, AMPylated BiP (‘B’ form). As suggested previously (Sanyal et al., 2015), the capacity of the cells to compensate for the consequences of deregulated FICD activity is limited; reflected here by the adverse effect of sustained FICD^E234G expression on cell viability (Figure 9D).

The aforementioned observations support the inactivating nature of FICD-mediated BiP modification and suggest that in some circumstances cells deficient in FICD may have elevated levels of active BiP. Given the strong repressive role of BiP on the UPR, enhanced availability of active chaperone attendant upon the absence of FICD and the consequential enhanced capacity to buffer unfolded protein stress might attenuate UPR signaling. Strong feedback control operative in the UPR specifies that such attenuation, were it to occur, would be transient and so more likely to be detected by monitoring early events in the UPR, such as PERK and IRE1α autophosphorylation, than later events, such as reporter gene activity. Furthermore, “over-chaperoning” attendant upon the absence of FICD would likely be revealed by manipulating the flux of proteins into the ER of a secretory cell. These theoretical considerations led us to AR42j pancreatic acinar cells - a secretory cell type in which PERK and IRE1α autophosphorylation are easy to detect (Bertolotti et al., 2000).

As expected, FICD inactivation by CRISPR-Cas9-mediated gene editing in AR42j cells eliminated both the acidic form of BiP from IEF-PAGE and the ‘B’ form from native-PAGE (Figure 10A-C and Figure 10—figure supplement 1B). Wildtype and FICD-lacking AR42j cells were exposed to cycloheximide to build up pools of modified BiP (in the wildtype cells) and then to the ER stress-inducing agent DTT, whose ability to activate the UPR does not require ongoing protein synthesis (Bertolotti et al., 2000). FICD knockout led to a conspicuous temporal delay in PERK and IRE1α autophosphorylation, observed in two independently derived AR42j FICD^-/- clones (Figure 10D). Sluggish activation of the UPR in the FICD^-/- cells was also reflected in attenuated PERK-dependent repression of protein synthesis imposed by DTT on cells pre-treated with cycloheximide. In this experiment, wildtype or FICD^-/- AR42j cells were pre-exposed to cycloheximide for 3 hr followed by brief washout to allow recovery of protein synthesis, which was measured by incorporation of puromycin label into newly synthesized proteins. In wildtype cells, DTT led to marked attenuation of protein synthesis, regardless of pre-exposure to cycloheximide (Figure 10E, compare lanes 1 and 2 and 3 and 4), however pre-exposure to cycloheximide attenuated subsequent translational repression by DTT-mediated PERK activation in FICD^-/- cells (Figure 10E, lanes 5-8).

Figure 10 with 2 supplements see all

Download asset Open asset

Figure 10—source data 1

Data from three independent repeats used for quantification shown in the graph in Figure 10E.

https://doi.org/10.7554/eLife.12621.024

Download elife-12621-fig10-data1-v2.zip

Figure 10—source data 2

Source file of the flow cytometry data used to generate the plot in Figure 10—figure supplement 2D.

https://doi.org/10.7554/eLife.12621.025

Download elife-12621-fig10-data2-v2.zip

The aforementioned defect in UPR activation in FICD^-/- cells was both fleeting and dependent on the pre-imposition of a low incoming protein flux regime on the ER (attained experimentally by cycloheximide pre-treatment), as FICD deletion had no measureable effect on the kinetics of CHOP::GFP expression induced by treatment with tunicamycin (Figure 10—figure supplement 2). These observations are inconsistent with the role proposed for FICD in signaling by the PERK branch of the UPR (Sanyal et al., 2015) and suggest instead that FICD-mediated BiP inactivation adjusted the level of active BiP to transient fluctuations in unfolded protein flux into the ER. In the absence of this mechanism, FICD knockout cells experienced an excess of functional chaperone, which was revealed here as attenuated activation of the earliest steps of the UPR.

Supplementary file 1 lists the plasmids used, their lab names, description and notes their first appearance in the figures and their corresponding label, and provides a published reference, where available.

A combination of PCR-based manipulations, restriction digests and site-directed mutagenesis procedures were used to mobilize the coding sequences and produce in-frame fusions with the affinity tags [GST, hexahistidine (His6) and His6-Smt3 epitopes] or mCherry fluorescent marker, and to create the point mutations indicated in the text.

All cells were grown in tissue culture dishes or multi-well plates (Corning) at 37°C and 5% CO₂ and the following cell line-specific media were used:

CHO-K1 cells (ATCC CCL-61) were phenotypically validated as proline auxotrophs and their Cricitulus griseus origin was confirmed by genomic sequencing. The cells were cultured in Nutrient mixture F-12 Ham (Sigma, UK) supplemented with 10% (v/v) serum (FetalClone II; HyClone, South Logan, UT), 1 x Penicillin-Streptomycin (Sigma) and 2 mM L-glutamine (Sigma).

AR42j cells (ATCC CRL-1492) were phenotypically validated by documenting inducibility of amylase expression by dexamethasone and their Rattus sp. origin was confirmed by genomic sequencing. They were cultured in DMEM (Sigma) supplemented with 10% (v/v) serum (FetalClone II; HyClone), 1 x Penicillin-Streptomycin (Sigma), 2 mM L-glutamine (Sigma), 1 x non-essential amino acids (Sigma) and 50 µM β-mercaptoethanol (Gibco; Life Technologies, UK).

HEK293T cells (ATCC CRL-3216) were cultured in DMEM (Sigma) supplemented with 10% (v/v) serum (FetalClone II; HyClone), 1 x Penicillin-Streptomycin (Sigma) and 2 mM L-glutamine (Sigma). Flp-In T-REx 293 cells (Invitrogen, UK) were cultured in the same medium except that 10% (v/v) tetracycline-free serum (FBS Premium; PAN-Biotech, Germany) was used.

All experiments with untransfected cells were performed at cell densities of 60-90% confluence. For pharmacological treatments the drugs were first diluted into pre-warmed culture medium, mixed and immediately applied to the cells by medium exchange. Unless indicated otherwise the following final concentrations were used: 100 µg/ml cycloheximide (Sigma), 0.5 µM thapsigargin (Calbiochem, Germany), 1 mM DTT, 2.5 µg/ml tunicamycin (Melford, UK), 10 µg/ml puromycin (Calbiochem), and 3 mM 2-deoxy-D-glucose (ACROS Organics, Belgium).

Three single guide RNA sequences (plasmids UK1448, UK1449 and UK1450; Supplementary file 1) for targeting the third exon of Cricetulus griseus (Chinese hamster) FICD were selected from the CRISPy database [URL: http://staff.biosustain.dtu.dk/laeb/crispy/, (Ronda et al., 2014)] and duplex DNA oligonucleotides of the sequences were inserted into the pSpCas9(BB)-2A-GFP plasmid (plasmid UK1359; Supplementary file 1) following published procedures (Ran et al., 2013). 2 x 10⁵ CHO-K1 or CHO-K1 CHOP::GFP reporter cells (Novoa et al., 2001) were plated in 6-well plates. Twenty-four hr later the cells were transfected with a combination of guide RNA/Cas9 plasmids UK1448 and UK1449 or UK1448 and UK1450 (2 µg total plasmid DNA per transfection) using Lipofectamine LTX (Invitrogen). Thirty-six hr after transfection the cells were washed with PBS, resuspended in PBS containing 4 mM EDTA and 0.5% (w/v) BSA, and GFP-positive cells were individually sorted by fluorescence-activated cell sorting (FACS) into 96-well plates using a MoFlo Cell Sorter (Beckman Coulter). Clones were then analyzed by a PCR-based assay to detect FICD mutations as described (Klampfl et al., 2013). Briefly, primers were designed for the region encompassing the FICD RNA guide target sites and the reverse primer was labeled with 6-carboxyfluorescein (6-FAM) on the 5’ end. A PCR reaction was set up using 5 µl of AmpliTaq Gold 360 Master Mix (Applied Biosystems, UK), 0.6 µl of a mix of 10 µM forward and labeled reverse primers, 3.4 µl H₂O and 1 µl genomic DNA (approximately 10 ng/µl). PCR was performed as follows: 95°C for 10 min, 10 x (94°C for 15 s, 59°C for 15 s, 72°C for 30 s), 20 x (89°C for 15 s, 59°C for 15 s, 72°C for 30 s), 72°C for 20 min. PCR products were diluted 1:100 in water and fragment length was determined on a 3130xl Genetic Analyzer (Applied Biosystems) and the data were analyzed using the Gene Mapper software (Applied Biosystems). Clones for which frameshift-causing insertions or deletions were detected for both alleles were sequenced to confirm the FICD mutations.

FICD knockouts in the AR42j cell line were created as described above, using two single guide RNA sequences (plasmids UK1503 and UK1504; Supplementary file 1) for targeting the third exon of Rattus norvegicus (rat) FICD selected using the CRISPR Design tool [URL: http://crispr.mit.edu/ (Zhang Lab)]. These cells were electroporated using the Neon transfection system (Life Technologies, UK) as described (Tsunoda et al., 2014).

The effect of FICD overexpression was analyzed in mammalian cell clones carrying stable transgenes that encode doxycycline-inducible mutant versions of FICD. To generate stable cell lines, plasmid DNA encoding FICD^E234G (UK1440) and FICD^E234G-H363A (UK1446) were introduced into the Flp-In T-REx 293 cell line (Invitrogen) as described by the manufacturer's protocol. Briefly, HindIII-XhoI DNA fragments containing the coding sequences for FICD^E234G and FICD^E234G-H363A were sub-cloned into pcDNA5/FRT/TO (Invitrogen). The resulting expression plasmids were individually co-transfected along with the Flp-recombinase expression vector pOG44 (Invitrogen) using Lipofectamine 2000 (Invitrogen) into Flp-In T-REx 293 cells. Isogenic clones expressing the transgene under a doxycycline-inducible promoter were selected for resistance to blasticidin (3 µg/ml; Thermo Fisher, UK) and hygromycin (250 µg/ml; Invitrogen) and sensitivity to zeocin (50 µg/ml; Invitrogen).

Flp-In T-REx 293 cells with stable transgenes encoding FICD^E234G or FICD^E234G-H363A were plated at a density of 8 x 10³ cells per well on a 24-well tissue culture plate and grown for 24 hr. The cells were then treated with 0.1 µg/ml doxycycline (Melford) for the indicated times. Afterwards, the cells were washed once in regular medium and then maintained in regular medium for three days. Following the recovery period, the medium was replaced with fresh medium containing 0.02 mM WST-1 (Dojindo, Germany) and 0.02 mM 1-methoxy phenazine methosulfate (Sigma), and the cells were incubated for 60 min at 37°C before absorbance was measured at 440 nm. Each experiment was performed in duplicates and repeated three times.

The effect of FICD overexpression (wildtype, FICD^E234G and FICD^E234G-H363A mutant versions) on the unfolded protein response was studied by transient transfection of wildtype and FICD^-/- CHO-K1 CHOP::GFP UPR reporter cell lines with plasmids UK1397, UK1398 or UK1443 (see Supplementary file 1) using Lipofectamine LTX. Where indicated, cells were treated 24 hr after transfection with the ER stress-inducing agent tunicamycin (2.5 μg/µl) for 16 hr to activate CHOP::GFP (which results in enhanced GFP production). Cells were analyzed by dual-channel flow cytometry with an LSRFortessa cell analyzer (BD Biosciences) as described previously (Tsunoda et al., 2014). GFP (excitation laser 488 nm, filter 530/30) and mCherry signals (excitation laser 561, filter 610/20) were detected. The data were processed using FlowJo and median reporter values were plotted using GraphPad Prism (GraphPad Software).

Cell lysis was performed as described (Preissler et al., 2015). Mammalian cells were grown on 10 cm dishes and treated as indicated. At the end of each experiment the dishes were placed on ice and washed twice with ice-cold PBS. The cells were detached with a cell scraper in PBS containing 1 mM EDTA, transferred to a 1.5 ml reaction tube and centrifuged for 5 min at 370 g at 4°C. The cells were lysed in HG lysis buffer [20 mM HEPES-KOH pH 7.4, 150 NaCl, 2 mM MgCl₂, 10 mM D-glucose, 10% (v/v) glycerol, 1% (v/v) Triton X-100] supplemented with protease inhibitiors [1 mM phenylmethylsulfonyl fluoride (PMSF), 2 µg/ml pepstatin, 2 µg/ml leupeptin, 4 µg/ml aprotinin] for 10 min on ice and centrifuged at 21,000 g for 10 min at 4°C. The protein concentrations of the cleared lysates were determined using the Bio-Rad protein assay reagent (Bio-Rad, Germany) and normalized with lysis buffer (usually the protein concentration was adjusted to 1.2 µg/µl). Where indicated the lysates were treated with SubA protease as described. For SDS polyacrylamide gel electrophoresis (SDS-PAGE) analysis, the proteins were denatured by addition of SDS sample buffer and heating for 10 min at 75°C followed by separation on 12% SDS polyacrylamide gels.

For detection of endogenous BiP by native-PAGE (see below) the cells were lysed in HG lysis buffer containing 2 x protease inhibitors (2 mM PMSF, 4 µg/ml pepstatin, 4 µg/ml leupeptin, 8 µg/ml aprotinin) with or without 100 U/ml hexokinase (Type III from baker’s yeast; Sigma) as described above and samples were loaded immediately on native or SDS polyacrylamide gels. Phosphatase inhibitors (10 mM tetrasodium pyrophosphate, 100 mM sodium fluoride, 17.5 mM β-glycerophosphate) were added to the lysis buffer in experiments where phosphorylated proteins were detected.

Native-PAGE was performed as described (Preissler et al., 2015). Tris-glycine polyacrylamide gels consisting of a 4.5% stacking gel and a 7.5% separation gel were used to separate purified protein or proteins from mammalian cell lysates under non-denaturing conditions to detect BiP oligomers. The gels were run in a Mini-PROTEAN electrophoresis chamber (Bio-Rad) in running buffer (25 mM Tris, 192 mM glycine, pH ~8.8) at 120 V for 2 hr when cell lysates were applied or for 1:45 hr when His6-tagged purified BiP proteins were analyzed. The proteins were then visualized by staining with InstantBlue Coomassie solution (expedeon, UK) or transferred for immunodetection to a polyvinylidene difluoride (PVDF) membrane in blotting buffer (48 mM Tris, 39 mM glycine, pH ~9.2) containing 0.04 (w/v) SDS for 16 hr at 30 V. The membrane was washed after the transfer for 20 min in blotting buffer supplemented with 20% (v/v) methanol before blocking. Seven µg of purified BiP protein was loaded per lane on a native gel to detect BiP oligomers by Coomassie staining and volumes of lysates corresponding to 30 µg of protein (CHO-K1 and AR42j cells) or 90 µg protein (Flp-In T-REx 293 cells) were loaded per lane.

Proteins were separated by SDS-PAGE or native-PAGE and transferred to PVDF membranes. The membranes were blocked with 5% (w/v) dried skimmed milk in TBS (25 mM Tris-HCl pH 7.5, 150 mM NaCl) and probed with primary antibodies followed by IRDye fluorescently labeled secondary antibodies (Li-Cor, UK). The membranes were scanned with an Odyssey near-infrared imager (Li-Cor) and where indicated densitometric quantification of the immunoblot signals was performed with ImageJ (NIH). Primary antibodies and antisera against hamster BiP [chicken anti-BiP; (Avezov et al., 2013)], eIF2α [mouse anti-eIF2α; (Scorsone et al., 1987)], phosphorylated eIF2α [rabbit anti-eIF2α Phospho (pS51); Epitomics cat. # 1090-1, UK], and FICD [rabbit anti-FICD; LifeSpan BioSciences cat. # LS-C80941, UK or chicken anti-FICD (see below)] were used.

To separate unmodified IRE1α and phosphorylated IRE1α lysates from AR42j cells were loaded on SDS polyacrylamaide gels (Mini-PROTEAN system; Bio-Rad) consisting of a 4% polyacrylamide stacking gel and a 7% polyacrylamide separation gel. The separation gel contained 50 µM MnCl₂ and 25 µM Phos-tag Acrylamide AAL-107 (NARD Institute Ltd., Japan). The gels were run for 3 hr at 300 V, incubated for 10 min in immunoblot transfer buffer containing 1 mM EDTA and proteins were transferred to PVDF membranes.

Wildtype and mutant versions of N-terminally hexahistidine- (His6-) tagged hamster BiP proteins (see Supplementary file 1 and “Plasmid construction”) were expressed in M15 E. coli cells (Qiagen, Germany) as described (Preissler et al., 2015). Bacterial cultures were grown at 37°C in LB medium containing 100 µg/ml ampicillin and 50 µg/ml kanamycin to an optical density (OD_{600 nm}) of 0.8 and expression was induced with 1 mM isopropylthio β-D-1-galactopyranoside (IPTG). After incubation for 6 hr at 37°C the cells were harvested by centrifugation and lysed with a high-pressure homogenizer (EmulsiFlex-C3; Avestin) in buffer A [50 mM Tris-HCl pH 7.5, 500 mM NaCl, 1 mM MgCl₂, 0.2% (v/v) Triton X-100, 10% (v/v) glycerol, 20 mM imidazole] containing protease inhibitors (2 mM PMSF, 4 µg/ml pepstatin, 4 µg/ml leupeptin, 8 µg/ml aprotinin) and 0.1 mg/ml DNaseI. The lysates were cleared by centrifugation (30 min at 25,000 g) and incubated with 1 ml nickel affinity matrix (Ni-NTA agarose; Qiagen) per 1 l of expression culture for 2 hr at 4°C. The beads were washed five times with 20 bed volumes of buffer A sequentially supplemented with (i) 30 mM imidazole, (ii) 1% (v/v) Triton X-100, (iii) 1 M NaCl, (iv) 5 mM Mg²⁺-ATP, or (v) 0.5 M Tris-HCl pH 7.5. Bound BiP proteins were eluted in buffer B [50 mM Tris-HCl pH 7.5, 500 mM NaCl, 1 mM MgCl₂, 10% (v/v) glycerol, 250 mM imidazole] and dialyzed against HKM buffer (50 mM HEPES-KOH pH 7.4, 150 mM KCl, 10 mM MgCl₂). The purified proteins were concentrated using centrifugal filters (Amicon Ultra, 30 kDa MWCO; Merck Millipore, UK), snap-frozen in liquid nitrogen and stored at -80°C. The purification of mature hamster BiP (19-654) was described in detail in Preissler et al., 2015.

Mammalian expressed GST-tagged recombinant FICD^E234G was produced in HEK293T cells. The cells were grown on 10 cm tissue culture dishes to a density of 80% and transfected either with plasmid encoding only GST or GST-FICD^E234G (UK1415) using the polyethylenimine (PEI) method with the following adaptations: 14 µg plasmid DNA was added to 700 µl Opti-MEM (Gibco; Life Technologies, UK) mixed with 56 µl of PEI (1 mg/ml stock) and added dropwise to the cells after incubation for 20 min at room temperature (ten dishes were usually transfected per preparation). Thirty-six hr after transfection the cells were lysed as described above with Triton lysis buffer [20 mM HEPES-KOH pH 7.4, 150 mM NaCl, 1 mM EDTA, 10% (v/v) glycerol, 1% (v/v) Triton X-100] containing protease inhibitors. The combined lysate (from ten dishes) was incubated with 200 µl GSH-Sepharose beads (Glutathione Sepharose 4B; GE Healthcare, UK) for 3 hr at 4°C. The beads were recovered by centrifugation for 5 min at 500 g, washed twice with Triton lysis buffer containing 300 mM NaCl and twice with AMPylation buffer [25 mM HEPES-KOH pH 7.4, 100 mM KCl, 4 mM MgCl₂, 1 mM CaCl₂, 0.1% (v/v) Triton X-100]. The beads with bound protein were stored in AMPylation buffer containing 50% (v/v) glycerol at -20°C and used for radioactive in vitro AMPylation reactions.

For bacterial expression of N-terminally GST-tagged FICD proteins and ERdj6 J-domains, plasmid DNA encoding active GST-FICD^E234G (UK1479), inactive GST-FICD^E234G-H363A (UK1480), GST-J (UK185) or mutant GST-J* (UK186) was transformed into BL21 T7 Express lysY/Iq E. coli cells (New England BioLabs cat. # C3013, UK). Bacterial cultures were grown at 37°C in LB medium containing 100 µg/ml ampicillin to OD_{600 nm} 0.8, shifted to 20°C and expression was induced with 0.1 mM IPTG for 16 hr. Afterwards, the cells were harvested and lysed as described above in lysis buffer [50 mM Tris-HCl pH 7.5, 500 mM NaCl, 1 mM MgCl₂, 2 mM dithiothreitol (DTT), 0.2% (v/v) Triton X-100, 10% (v/v) glycerol] containing protease inhibitors and DNaseI. The lysates were cleared by centrifugation (30 min at 25,000 g) and incubated with 0.7 ml GSH-Sepharose beads per 1 l of expression culture for 3 hr at 4°C. The beads were transferred to a column and washed with 20 ml wash buffer B [50 mM Tris-HCl pH 7.5, 500 mM NaCl, 1 mM DTT, 0.2% (v/v) Triton X-100, 10% (v/v) glycerol] containing protease inhibitors, 20 ml wash buffer C [50 mM Tris-HCl pH 7.5, 300 mM NaCl, 10 mM MgCl₂, 1 mM DTT, 0.1% (v/v) Triton X-100, 10% (v/v) glycerol] containing protease inhibitors and 20 ml wash buffer C sequentially supplemented with (i) 1% (v/v) Triton X-100, (ii) 1 M NaCl, (iii) 3 mM ATP, (iv) or 0.5 M Tris-HCl pH 7.5. Bound BiP proteins were then eluted in elution buffer [50 mM HEPES-KOH pH 7.4, 100 mM KCl, 4 mM MgCl₂, 1 mM CaCl₂, 0.1% (v/v) Triton X-100, 10% (v/v) glycerol, 40 mM reduced glutathione] and concentrated protein solutions were frozen in liquid nitrogen and stored at -80°C.

N-terminally His6-Smt3 tagged wildtype mouse FICD lumenal domain (L104-P458) (UK1564) was encoded on a pET28b plasmid (Novagen, UK) and expressed in E. coli BL21 T7 Express lysY/I^q cells. Bacterial cultures (3 liters) were grown at 37°C in LB medium containing 100 µg/ml ampicillin to OD_{600 nm} 0.8 and expression was induced with 0.1 mM IPTG. After incubation for 16 hr at 20°C the cells were sedimented by centrifugation and pellets were lysed as described above in lysis buffer [50 mM Tris-HCl pH 7.4, 500 mM NaCl, 1 mM MgCl₂, 10% (v/v) glycerol, 30 mM imidazole, 1 mM tris(2-carboxyethyl)phosphine (TCEP)] supplemented with protease inhibitors and DNaseI. The lysates were cleared by centrifugation for 30 min at 33,000 g, followed by addition of 0.2% (v/v) Triton X-100 and 0.002% (v/v) sodium deoxycholate, and incubated with 1 ml Ni-NTA agarose (Qiagen) for 2.5 hr at 4°C. The beads were washed with 25 bed volumes of wash buffer [50 mM Tris-HCl pH 7.4, 500 mM NaCl, 10% (v/v) glycerol, 0.2% (v/v) Triton X-100, 1 mM TCEP, protease inhibitors], once with buffer D [50 mM Tris-HCl pH 7.4, 300 mM NaCl, 10% (v/v) glycerol, 0.1% (v/v) Triton X-100, 10 mM MgCl₂, 1 mM TCEP, and protease inhibitors], and then with buffers E_i-iv [buffer D supplemented with (i) 1% (v/v) Triton X-100, (ii) 1 M NaCl, (iii) 3 mM ATP, or (iv) 0.5 M Tris-HCl pH 7.4]. Bound mFICD protein was eluted in buffer F [50 mM Tris-HCl pH 7.4, 100 mM NaCl, 4 mM MgCl₂, 1 mM CaCl_2,10% (v/v) glycerol, 250 mM imidazole, 1 mM TCEP), and dialyzed against 50 mM HEPES-NaOH pH 7.4, 100 mM NaCl, 0.5 mM 2-mercaptoethanol in the presence of Ulp1 (2 µg per mg of eluted protein) for 16 hr at 4°C to cleave the His6-Smt3 tag. The dialyzed and cleaved protein was then diluted 1:2 in 25 mM HEPES-NaOH pH 8.1 to reduce the ionic strength, loaded on a anion-exchange chromatography column (Mono Q 5/50 GL; GE Healthcare) and eluted with a linear salt gradient from 50 to 500 mM NaCl in 25 mM HEPES-NaOH pH 7.4. The mFICD-containing fractions were pooled and further purified by size-exclusion chromatography (Superdex 200 10/300; GE Healthcare) in gel filtration buffer (25 mM HEPES-NaOH pH 7.4, 150 mM NaCl). The eluted fractions were combined and provided to Aves Labs (Tigard, OR) as an immunogen for production of chicken anti-FICD antibodies.

The absence of FICD protein in AR42j FICD^-/- cells was confirmed by immunoaffinity purification as the endogenous protein could not be detected by immunoblotting of total lysates. Protein extracts from wildtype and FICD^-/- AR42j cells were prepared in HG lysis buffer containing protease inhibitors, cleared, normalized and equal volumes of the lysates (4 mg total protein) were incubated with 15 µl UltraLink Hydrazine Resin (Pierce cat. # 53149, UK) on which FICD-specific chicken IgY antibodies have been covalently immobilized according to the manufacturer’s instructions (chicken anti-FICD) or rabbit polyclonal IgG antibodies against human FICD (rabbit anti-FICD; LifeSpan BioSciences) bound to Protein A Sepharose 4B beads (Invitrogen cat. #10-1041), for 16 hr at 4°C. The beads were then recovered by centrifugation for 1 min at 1,000 g and washed three times with HG lysis buffer. Bound proteins were eluted in 40 µl 2 x SDS sample buffer for 10 min at 70°C and equal volumes of the samples were loaded on a 12% SDS polyacrylamide gel, and endogenous FICD was detected by immunoblotting with rabbit anti-FICD or chicken anti-FICD antibodies. Samples of the normalized lysates (25 µg) were loaded and BiP was detected as an “input” control.

Unless indicated otherwise, in vitro AMPylation was performed in AMPylation buffer (described above) by incubation of purified wildtype or mutant BiP proteins at 20 µM with bacterially expressed GST-FICD^E234G (or as a control with inactive GST-FICD^E234G-H363A) at 0.8 µM in presence of 1.5 mM ATP for 45 min at 30°C. The final volume of a typical reaction was 15 µl. The samples were then treated with or without the BiP linker-specific protease SubA at 30 ng/µl for 10 min at room temperature. Volumes corresponding to 7 µg BiP protein were loaded per lane on native or SDS polyacrylamide gels and visualized by Coomassie staining. The in vitro AMPylation time-course experiments presented in Figure 7C and D were performed likewise but after incubation for the indicated times at 30°C the reactions were stopped on ice by addition of 40 mM EDTA and samples were run immediately on native gels.

Radioactive in vitro AMPylation was performed by mixing 7.5 µg of purified wildtype or mutant BiP proteins with 10 µM non-radiolabeled ATP and 0.185 MBq α-³²P-ATP (EasyTide; Perkin Elmer, UK) in a total volume of 30 µl in AMPylation buffer. The reactions were incubated with 5 µl GSH-Sepharose beads coupled with mammalian expressed GST-FICD^E234G (or GST only) in 1.5 ml reaction tubes at 30°C on a thermomixer (Eppendorf) while shaking at 800 rpm. After 45 min [in these experiments a time before the reaction went to completion to reveal differences in AMPylation amongst BiP mutants (Figure 5A and C and Figure 7A)] the beads were sedimented by centrifugation and the reaction was terminated by addition of SDS loading buffer to the supernatants and heating for 5 min at 75°C. Equal volumes of the samples were loaded on 12% SDS polyacrylamide gels. Proteins were visualized by Coomassie staining and radioactive signals were detected with a Typhoon Trio imager (GE Healthcare) upon overnight exposure of the dried gels to a storage phosphor screen. To test the sensitivity of AMPylated BiP towards cleavage by SubA ( Figure 3—figure supplement 1), 6 µg of purified BiP^T229A were labeled in 60 µl AMPylation buffer containing 6.7 MBq of α-³²P-ATP and 10 µl of GST-FICD^E234G-coupled beads as described above. After incubation for 30 min at 30°C non-radioactive ATP was added at 5 µM and the reaction was allowed to proceed for another 60 min. The beads were then sedimented by centrifugation and 1.5 mM ATP was added to the supernatant. To remove nucleotides the sample was passed through a Sephadex G-50 MicroSpin column (illustra AutoSeq G-50; GE Healthcare) equilibrated with AMPylation buffer. 1.5 µg of the recovered protein was mixed with 60 µg unmodified BiP^T229A (at 0.6 mg/ml final) in presence of 1.5 mM ATP and the combined sample was divided into fractions to which SubA (from a dilution series) was added to yield final concentrations between 0.08 and 120 ng/µl. After incubation for 30 min at room temperature the proteins were denatured with SDS sample buffer and equal volumes were applied to SDS-PAGE, Coomassie-stained and signals were detected by autoradiography.

Twenty mg of purified wiltype BiP or BiP^V461F protein were in vitro AMPylated for 2 hr at 30°C with 1 mg bacterially expressed GST-FICD^E234G in presence of 1.5 mM ATP in AMPylation buffer in a final volume of 12 ml. BiP proteins were then bound to 400 µl nickel affinity matrix, washed with AMPylation buffer and eluted in AMPylation buffer containing 250 mM imidazole. The concentrated eluate was divided into aliquots and immediately frozen in liquid nitrogen and stored at -80°C. For applications that required complete absence of detergent or imidazole the affinity chromatography eluate was passed through a Centri•Pure P25 desalting column (emp BIOTECH, Germany) equilibrated in HKM buffer followed by concentration of the proteins using centrifugal filters (Amicon Ultra, 30 kDa MWCO; Merck Millipore) before freezing. The in vitro modified proteins were used for mass spectrometry analyses and functional assays (see below). Unmodified BiP from parallel mock in vitro AMPylation reactions, to which no enzyme has been added, served as controls in all assays.

The experimental strategy for the stable isotope labeling by amino acids in cell culture (SILAC) experiment is outlined in Figure 6A and samples were prepared as follows: Wildtype and FICD^-/- CHO-K1 cells were adapted to Ham’s F12 medium minus L-arginine and L-lysine for SILAC (Pierce cat. # 88424) supplemented with 10% (v/v) dialyzed fetal bovine serum (Gibco; Life Technologies cat. # 26400-044), 1 x Penicillin-Streptomycin (Sigma), 2 mM L-glutamine (Sigma), 280 mg/l L-proline (Sigma cat. # P5607-25G), 62.5 mg/l “light” L-lysine monohydrochloride (Sigma cat. # L8662-25G) and 60.5 mg/l “light” L-arginine monohydrochloride (Sigma cat. # A6969-25G) and incubated as described above. Once adapted, the cells were cultured in SILAC medium containing either 60.5 mg/l “light” (as above) or “heavy” L-arginine monohydrochloride (R10; U-13C6, U-15N4; Cambridge Isotope Laboratories, Inc. cat. # CNLM-539, Andover, MO) for several passages (> 15 cell divisions) before expansion in five 10 cm dishes per sample. The cells were grown to ~70% confluence and medium was exchanged 14 hr before treatment with or without 100 µg/ml cycloheximide for 3 hr. After washing with ice-cold PBS the cells were collected in PBS containing 1 mM EDTA and lysed in Triton lysis buffer [20 mM HEPES-KOH pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% (v/v) Triton X-100, 10% (v/v) glycerol] containing protease inhibitors. The protein concentrations of the cleared lysates were normalized and equal volumes were then mixed (1:1 ratio) as indicated. Each combined sample was pre-cleared for 15 min at 4°C with 30 µl blocked UltraLink Hydrazine Resin followed by immunoaffinity purification for 2 hr at 4°C with 30 µl of the same resin to which BiP-specific IgY antibodies were covalently coupled (Preissler et al., 2015). The beads were washed twice for 10 min with lysis buffer, once with lysis buffer containing 350 mM NaCl and once again with lysis buffer, before elution in 2x non-reducing SDS sample buffer for 10 min at 70°C. Afterwards, the beads were removed by centrifugation and the supernatants were supplemented with 125 mM DTT, heated for 5 min at 70°C and loaded on a 12% SDS polyacrylamide gel. The gel was stained with Coomassie and bands corresponding in size to BiP were cut out for in-gel digest with Arg-C endopeptidase, and analysis by mass spectrometry.

Wildtype and FICD^-/- CHO-K1 cells were grown in standard medium (ten 10 cm dishes per sample) and treated for 3 hr with 100 µg/ml cycloheximide before cell lysis with Triton lysis buffer supplemented with protease inhibitors and BiP was immunoaffinity purified as described above with the following modifications: The lysates (~4 mg protein per sample) were pre-cleared with 40 µl blocked UltraLink Hydrazine Resin for 30 min at 4°C followed by incubation for 2 hr with 60 µl of the resin carrying covalently attached BiP-specific IgY antibodies. The beads were washed sequentially for 10 min with lysis buffer, lysis buffer containing 350 mM NaCl, HG buffer (see above) containing 1 mM ATP, and once again with lysis buffer. Bound protein was eluted in 150 µl of 0.2 M glycine-HCl pH 2.5 for 10 min at 16°C followed by neutralization of the recovered sample supernatants with 1/10 volume of 1.5 M Tris-HCl pH 8.8. Afterwards, protein was precipitated with isopropanol and analyzed by mass spectrometry.

Molecular masses of intact BiP proteins were measured by electrospray ionization mass spectrometry using a Q-Trap 4000 (ABSciex) after ‘on-line’ reverse-phase chromatographic purification of samples as described previously (Carroll et al., 2009). The instrument was operated in MS mode, and was calibrated with a mixture of myoglobin and trypsinogen. Reconstructed molecular masses were calculated with Bioanalyst (ABSciex) and masses from amino acid sequences with MassLynx (Waters).

To analyze masses of BiP peptides, proteins were reduced, alkylated and digested “in-gel”. In vitro AMPylated BiP protein was digested with the proteolytic enzymes as indicated and SILAC samples were digested using Arg-C. The resulting peptides were analyzed by LC-MS/MS using either an Orbitrap XL coupled to a nanoAcquity UHPLC or a Q Exactive coupled to a RSLC3000 UHPLC. Data was processed in Proteome Discoverer 1.4 using Sequest to search a UniProt E. coli database (downloaded 29/04/15, 4,378 entries) with the sequence for Chinese hamster (Cricetulus griseus) BiP added or a Chinese Hamster database (downloaded 19/02/14, 23,884 entries). Oxidation (M), deamidation (N/Q) and AMPylation (S/T) were set as variable modifications and carbamidomethylation (C) as a fixed modification. FDR calculations were performed by Percolator and peptides were filtered to 1%. Peak areas were determined using the Precursor Ion Quantifier node in Proteome Discoverer.

Purified untagged hamster BiP^19-654 at 15 µM was in vitro AMPylated with 0.75 µM GST-FICD^E234G (or as a control with inactive GST-FICD^E234G-H363A) in presence of 1.5 mM ATP for the indicated times at 30°C as described above. The reactions were then treated without or with 30 ng/µl SubA for 10 min at room temperature in a final volume of 20 µl and analyzed by IEF as described previously (Chambers et al., 2012; Laitusis et al., 1999) with modifications: The samples were diluted with 180 µl of IEF sample buffer [8 M urea, 5% (w/v) CHAPS, 50 mM DTT, 2% (v/v) Pharmalyte (pH 4.5-5.4; GE Healthcare)] and loaded on a 3.75% polyacrylamide gel containing 8.8 M urea, 1.25% (w/v) CHAPS and 5% (v/v) Pharmalyte. The loaded samples were overlaid with 0.5 M urea and 2% (v/v) Pharmalyte solution before separation. 10 mM glutamic acid and 50 mM histidine were used as anode and cathode buffers, respectively. After the run (100V, 10 min; 250V, 1 hr; 300V, 1 hr; 500V, 30 min) proteins were transferred to a nitrocellulose membrane in blotting buffer [25 mM Tris-HCl pH 9.2, 190 mM glycine, 0.01% (w/v) SDS, 10% (v/v) methanol] for 3 hr at 300 mA and immunodetected with BiP-specific antibodies as described above.

Lysate samples from mammalian cells (CHO-K1 and AR42j) were prepared for analysis by IEF as follows: Cells were grown in 10 cm dishes to approximately 90% confluence, washed with ice-cold TBS, resuspended in 1 ml TBS, sedimented by centrifugation, and lysed in 30 x its packed cell volume of IEF lysis buffer [8.8 M urea, 5% (w/v) CHAPS, 1 µM sodium pyrophosphate, 2 mM imidodiphosphate, 50 mM DTT and 2% (v/v) Pharmalyte] at room temperature for 5 min. The lysates were then centrifuged at 20,238 g for 10 min at room temperature, the supernatants were transferred to a new tube and then centrifuged again for 60 min. The cleared lysates (50 µl) were then purified using Sephadex G-50 MicroSpin columns equilibrated with IEF sample buffer, diluted 1:2 in IEF sample buffer and 15 µl were loaded on a 3.75% polyacrylamide gel followed by BiP immunoblotting as described above.

Malachite green (MG) reaction solution was prepared freshly by mixing stock solutions of MG dye [0.111% (w/v) malachite green, 6 N sulphuric acid], 7.5% (w/v) ammonium molybdate and 11% (v/v) Tween 20 in a 10:2.5:0.2 ratio. Samples of purified BiP proteins (5 µM) were incubated in HKM with 3 mM ATP in a final volume of 20 µl for 60 min at 30°C. 15 µl of each sample were then diluted with 135 µl water on a 96-well plate, mixed with 50 µl MG reaction solution and incubated for 2 min at room temperature. 20 µl of a 34% (w/v) sodium citrate solution were added to quench the reactions and after incubation for further 30 min the absorbance was measured at 612 nm with a plate reader (Infinite F500; Tecan). Standard curves were prepared with serial dilutions of KH₂PO₄ and served as a reference to calculate V_max values. Statistical analysis by unpaired t-test was performed using Graphpad Prism version 6.

Binding of substrate peptide by purified BiP proteins was measured as previously described (Chambers et al., 2012; Marcinowski et al., 2011) with modifications: Unmodified or in vitro AMPylated BiP proteins were incubated at the indicated concentrations (between 0 and 40 µM) with 1 µM lucifer yellow (LY)-labeled substrate peptide (HTFPAVLGSC) in presence of 1 mM ADP for 24 hr at 30°C in FP buffer [37.5 mM HEPES-KOH pH 7.4, 125 mM KCl, 7 mM MgCl₂, 0.5 mM CaCl₂, 125 mM imidazole, 0.05% (v/v) Triton X-100]. 20 µl of each sample were transferred to a 386-well polystyrene microplate (µClear, black; greiner bio-one cat. # 781096, UK) and fluorescence polarization of the lucifer yellow fluorophore was measured (excitation λ = 430 nm; emission λ = 535 nm) at room temperature in a plate reader. For analysis of competitive substrate dissociation, BiP proteins (40 µM) were incubated with LY-labeled substrate peptide (1 µM) as described above. Afterwards, 400-fold molar excess of unlabeled substrate peptide (HTFPAVL) was added to the pre-formed BiP:substrate complexes and the change in fluorescence polarization was measured over time. From each value the background signal of a reference sample containing only LY-labeled peptide (without BiP protein) were subtracted.

RNA from CHO-K1 cells was extracted using RNA Stat-60 (Amsbio, UK) according to the manufacturer’s instructions. 1 µg RNA was reverse transcribed with Oligo(dT)18 primer using RevertAid Reverse Transcriptase (Thermo Scientific, UK). Quantitative PCR analysis was performed using the Power SYBR Green PCR Master Mix (Applied Biosystems) according to the manufacturer’s instructions on a 7900HT Fast Real-Time PCR System (Applied Biosystems). FICD, RPL27 (60S ribosomal protein L27) and PPIA (cyclophilin A) were amplified using the following primers: FICD-F 5’-GGGCTGCTGACTGTTAAGACTA-3’, FICD-R 5’-CCTTGACGCTTCATCTCCA-3’, RPL27-F 5’-ACAATCACCTCATGCCCACAAG-3’, RPL27-R 5’-GCGTTTCAGGGCTGGGTCTC-3’, PPIA-F 5’-TTCCTCCTTTCACAGAATTATCCC-3’, and PPIA-R 5’-CTGCCGCCAGTGCCATTATG-3’. Relative quantities of amplified cDNAs were then determined using SDS 2.4.1 software (Applied Biosystems) and normalized to PPIA mRNA.

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

We thank the CIMR flow cytometry core facility team (Reiner Schulte, Michal Maj and Chiara Cossetti) for assistance with FACS, Heather P. Harding, Niko Amin-Wetzel and Joseph E. Chambers (all from CIMR, Cambridge) for technical advice and critical input, Qinlian Liu from Virginia Commonwealth University for sharing unpublished structural data, and Claudia Flandoli (www.claudiaflandoli.com) for preparation of the cartoon images.

Supported by grants from the Wellcome Trust (Wellcome 084812/Z/08/Z) the European Commission (EU FP7 Beta-Bat No: 277713), a Wellcome Trust Strategic Award for core facilities to the Cambridge Institute for Medical Research (Wellcome 100140). DR is a Wellcome Trust Principal Research Fellow.

© 2015, Preissler et al.

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.