FIC proteins: from bacteria to humans and back again

FIC proteins: from bacteria to humans and back again Abstract During the last decade, FIC proteins have emerged as a large family comprised of a variety of bacterial enzymes and a single member in animals. The air de famille of FIC proteins stems from a domain of conserved structure, which catalyzes the post-translational modification of proteins (PTM) by a phosphate-containing compound. In bacteria, examples of FIC proteins include the toxin component of toxin/antitoxin modules, such as Doc-Phd and VbhT-VbhA, toxins secreted by pathogenic bacteria to divert host cell processes, such as VopS, IbpA and AnkX, and a vast majority of proteins of unknown functions. FIC proteins catalyze primarily the transfer of AMP (AMPylation), but they are not restricted to this PTM and also carry out other modifications, for example by phosphocholine or phosphate. In a recent twist, animal FICD/HYPE was shown to catalyze both AMPylation and de-AMPylation of the endoplasmic reticulum BIP chaperone to regulate the unfolded protein response. FICD shares structural features with some bacterial FIC proteins, raising the possibility that bacteria also encode such dual activities. In this review, we discuss how structural, biochemical and cellular approaches have fertilized each other to understand the mechanism, regulation and function of FIC proteins from bacterial pathogens to humans. FIC proteins, structural biology, biochemistry, toxins, cellular biology, pathogens FIC PROTEINS: FROM THE AIR DE FAMILLE TO FUNCTION FIC proteins are widespread in bacteria, with a single member found in animals. They have attracted considerable interest over the last decade for their functions in bacterial stress responses and in infections (reviewed in Garcia-Pino, Zenkin and Loris 2014; Roy and Cherfils 2015; Harms, Stanger and Dehio 2016; Casey and Orth 2017) and recently in proteostasis control in animal cells (reviewed in Truttmann and Ploegh 2017). The air de famille of FIC proteins stems from a domain of conserved 3D structure with a signature sequence (the FIC motif) that catalyzes the post-translational modification (PTM) of proteins by a phosphate-containing compound. This domain was called Fido (FIC domain) as an umbrella term to refer to FIC proteins and other proteins with the same overall fold (Kinch et al. 2009). In bacteria, genes coding for FIC proteins are often found in mobile genomic islands which may have facilitated their spread across bacterial species (Khater and Mohanty 2015b). A commonality of FIC proteins with known functions is that they carry out the PTM of target proteins by a phosphate-containing compound, and this enzymatic activity is entirely comprised in the FIC domain. This function was not known when the first fic gene was discovered in Escherichia coli 30 years ago, in which a point mutation was found to impair cell division under stress conditions (cyclic AMP in growth medium at high temperature), leading to filamentation (Kawamukai et al. 1988, 1989). This now anecdotal observation led to naming this gene fic-1 for filamentation induced by cAMP, and later to extend the name to the conserved FIC domain (Worby et al. 2009; Yarbrough et al. 2009) and eventually to the entire family (Kinch et al. 2009). The product of the fic-1 gene was recently characterized as the toxin component of a toxin-antitoxin (TA) module structurally related to the VbhT-VbhA TA module and has now been renamed EcFicT (Stanger et al. 2016a). The function of EcFicT has not yet been elucidated, but functions of related FIC proteins have begun to be unraveled. The related Pseudomonas fluorescens Fic-1 inhibits DNA replication by adding AMP to DNA gyrase (Lu et al. 2016), a reaction coined AMPylation (also referred to as adenylylation). Likewise, ectopic expression of Bartonella schoenbuchensis VbhT in E. coli disrupts DNA topology by AMPylation of DNA gyrase and DNA topoisomerase IV (Harms et al. 2015). Thus, it is likely that EcFicT also functions in controlling DNA replication. It took more than three decades for the FIC family to enter the limelight, following the landmark discovery that toxins carrying a FIC domain are secreted by bacterial pathogens to divert small GTPases of infected cells through AMPylation (Yarbrough et al. 2009). Vibrio parahaemolyticus VopS (Yarbrough et al. 2009) then Histophilus somni IbpA (Worby et al. 2009) were both shown to AMPylate small GTPases of the Rho family, which inactivates them leading to the collapse of the actin cytoskeleton. This reaction uses ATP as donor for AMP, which is transferred to hydroxyl-containing residues (tyrosine, serine or threonine) from the target protein. While AMPylation was briefly thought to be the reaction unifying the family (reviewed in Roy and Mukherjee 2009), alternative enzymatic activities were soon discovered, including phosphocholination of Rab small GTPases by the Legionella pneumophila type IV effector AnkX (Mukherjee et al. 2011), phosphorylation of the GTPase-related elongation factor Ef-Tu by the bacteriophage P1 TA Doc-Phd module in E. coli (Castro-Roa et al. 2013; Cruz et al. 2014) and addition of UMP to plant immune kinases by an FIC protein from the plant pathogen Xanthomonas campestris (Feng et al. 2012). The reaction landscape of FIC proteins was accordingly extended to the more general function of transfering a phosphate-containing compound to a target protein leading to signaling roadblocks (reviewed in Cruz and Woychik 2014; Roy and Cherfils 2015; Harms, Stanger and Dehio 2016). In an unexpected twist, the sole animal protein with a FIC domain, called HYPE or FICD, was recently shown to carry out both the addition (Ham et al. 2014; Preissler et al. 2015; Sanyal et al. 2015) and removal (Casey et al. 2017; Preissler et al. 2017) of AMP from the endoplasmic reticulum chaperone BIP. De-AMPylation was shown to depend on a glutamate located outside the catalytic FIC motif (Preissler et al. 2017), which is structurally conserved in FICD/HYPE and in a subset of bacterial FIC proteins where it is thought to serve an autoinhibitory function (Engel et al. 2012). These recent discoveries now raise new questions, and important issues remain uncompletely understood. For example, what is the protein substrate, the PTM donor (refered to as the co-factor hereafter) and the primary PTM carried out by individual FIC proteins? How are animal and bacterial FIC proteins regulated? Is de-AMPylation also found in bacterial FIC proteins? Ultimately, what are the functions of the still predominently uncharacterized bacterial FIC proteins? In this review, we discuss how complementary bioinformatics, structural, biochemical, cellular and in vivo approaches have fertilized each other to generate and test hypothesis about the mechanism, regulation and function of FIC proteins from bacterial pathogens to animals. BIOINFORMATICS: CLASSIFYING FIC PROTEINS AND ANALYZING PHYLOGENY High-throughput sequencing of bacterial genomes identified potential coding genes for proteins with FIC signatures in a large number of bacterial species, including free living, host-associated, commensal and pathogenic species. For example, as of 2017 the PFAM database reports about 7500 sequences that contain a FIC domain, organized in 96 architectures found in >2300 bacteria, 360 eukaryotes and 71 archae species (Fig. 1). In many cases, these genes were annotated in the databases as ‘filamentation induced by cAMP protein’ or ‘adenosine monophosphate-protein transferase’. While the first annotation is now mostly anecdotal, the second annotation may be somewhat misleading as enzymatic activities other than AMPylation continue to be discovered. We suggest that unknown genes should therefore better be annotated as ‘uncharacterized FIC family protein’ to reflect that not all FIC proteins carry out AMPylation. Sequence databases have provided a rich material to classify FIC proteins and investigate their phylogeny (Engel et al. 2012; Garcia-Pino, Zenkin and Loris 2014; Harms et al. 2015; Khater and Mohanty 2015a,b; Kinch et al. 2009; Worby et al. 2009; Yarbrough et al. 2009) and to forge hypotheses about regulatory mechanisms (Engel et al. 2012). However, predicting whether an FIC protein of unknown function is an actual AMPylator based on its sequence alone stills remains uncertain. We discuss below how structural biology highlighted functional diversity encoded in FIC sequences, while experiments using recombinant proteins and cellular assays have remained the primary approaches for discovering the enzymatic activities and functions of individual FIC proteins. Figure 1. View largeDownload slide A graphical representation of the distribution of Fic genes (PF02661) across taxonomic levels. Reproduced from the EMBL-EBI PFAM server (http://pfam.xfam.org/family/Fic#tabview = tab7) (Finn et al. 2016). The distribution among the eight major taxonomic levels is shown, starting with superkingdom in the center of the sunburst, and then concentric circles corresponding successively to kingdom, phylum, class, order, family, genus and species. Each circle is subdivided into colored arcs corresponding to superkingdoms and/or kingdoms. The longer the arc, the higher the number of species harboring at least one FIC gene. On the outermost circle, one line corresponds to one species, irrespectively of the number of FIC genes (one or more) in its genome. It should be noted that except for the Doc/Phd module harbored on a bacteriophage P1 plasmid in E. coli, FIC proteins have not been identified in viruses. Figure 1. View largeDownload slide A graphical representation of the distribution of Fic genes (PF02661) across taxonomic levels. Reproduced from the EMBL-EBI PFAM server (http://pfam.xfam.org/family/Fic#tabview = tab7) (Finn et al. 2016). The distribution among the eight major taxonomic levels is shown, starting with superkingdom in the center of the sunburst, and then concentric circles corresponding successively to kingdom, phylum, class, order, family, genus and species. Each circle is subdivided into colored arcs corresponding to superkingdoms and/or kingdoms. The longer the arc, the higher the number of species harboring at least one FIC gene. On the outermost circle, one line corresponds to one species, irrespectively of the number of FIC genes (one or more) in its genome. It should be noted that except for the Doc/Phd module harbored on a bacteriophage P1 plasmid in E. coli, FIC proteins have not been identified in viruses. STRUCTURAL BIOLOGY: DISSECTING MECHANISMS AND INSPIRING HYPOTHESES The first high-resolution structures of an FIC protein were those of single-domain FIC proteins from Helicobacter pylori and Neisseria meningitidis, which were determined as part of Structural Genomics Initiatives in 2006 (PDB entries 2G03 and 2F6S, which have no associated publications). However, the structure of FIC proteins only began to be discussed with the structure of the Doc/Phd TA complex (Garcia-Pino et al. 2008) and soon afterwards with the discovery of AMPylating FIC toxins from pathogenic bacteria (Worby et al. 2009; Yarbrough et al. 2009; Xiao et al. 2010). As of today, the structures of >20 FIC proteins have been determined, several of them in complex with a co-factor bound to the active site, while only one has a bound protein substrate (Table 1 and references therein). These structures established the remarkable conservation of the FIC domain and of the FIC motif (HPFx[D/E]GNGR, with variations) that carries the invariant catalytic histidine (Fig. 2A). Importantly, they provided a framework to identify conserved catalytic and substrate recognition features as well as divergences and they irrigated the field with fruitful hypotheses (reviewed in Garcia-Pino, Zenkin and Loris 2014; Roy and Cherfils 2015; Harms, Stanger and Dehio 2016). We present below some of the milestone structures that shaped our current understanding. Figure 2. View largeDownload slide Structural features of FIC proteins. (A) The canonical FIC domain, exemplified by the structure of N. meningitidis. The FIC motif is in pink and the catalytic histidine shown; the β-hairpin is in cyan. A bound ATP co-factor, of which only the ADP moiety is visible, is shown in stick representation. The same color coding is used in the other panels. All PDB entries and references are given in Table 1. (B) The structure of the H. somni IbpA toxin bound the human small GTPase Cdc42 (Xiao et al. 2010). Cdc42 is in purple and IbpA is in gray. The AMPylated tyrosine in the switch 1 of Cdc42 is indicated. (C) Examples of co-factors found in crystal structures of FIC proteins. AMP-Tyr is from Histophilus IbpA in complex with AMPylated Cdc42 (Xiao et al. 2010), ADP-Mg2+ is from human FICD/HYPE carrying a mutation of the inhibitory glutamate (Bunney et al. 2014), ATP-Mg2+ is from Bartonella VbhT in complex with the VbhA antitoxin carrying a mutation of the inhibitory glutamate (Goepfert et al. 2013), CDP-choline is from Legionella AnkX carrying a mutation of the catalytic histidine (Campanacci et al. 2013). The FIC domain (not drawn) is in the same orientation in all views. (D) The active site of Legionella AnkX (Campanacci et al. 2013) showing elements from outside the FIC motif that recognize the cytosine and choline moieties of CDP-choline. Below, the sequence of AnkX, highlighting these residues in blue. Residues from the β-hairpin, which is interrupted by an insertion in AnkX, are in cyan and the FIC motif is in red. (E) Comparison of the structures of the Bartonella VbhT-VbhA TA module (left panel, Goepfert et al. 2013) and Neisseiria FIC (right panel, PDB entry 2G03). The inhibitory glutamate is circled in red. The helix carrying the glutamate is located in the antitoxin in the VbhT-VbhA complex, and is in C-terminus in Nesseiria FIC (shown in orange). Figure 2. View largeDownload slide Structural features of FIC proteins. (A) The canonical FIC domain, exemplified by the structure of N. meningitidis. The FIC motif is in pink and the catalytic histidine shown; the β-hairpin is in cyan. A bound ATP co-factor, of which only the ADP moiety is visible, is shown in stick representation. The same color coding is used in the other panels. All PDB entries and references are given in Table 1. (B) The structure of the H. somni IbpA toxin bound the human small GTPase Cdc42 (Xiao et al. 2010). Cdc42 is in purple and IbpA is in gray. The AMPylated tyrosine in the switch 1 of Cdc42 is indicated. (C) Examples of co-factors found in crystal structures of FIC proteins. AMP-Tyr is from Histophilus IbpA in complex with AMPylated Cdc42 (Xiao et al. 2010), ADP-Mg2+ is from human FICD/HYPE carrying a mutation of the inhibitory glutamate (Bunney et al. 2014), ATP-Mg2+ is from Bartonella VbhT in complex with the VbhA antitoxin carrying a mutation of the inhibitory glutamate (Goepfert et al. 2013), CDP-choline is from Legionella AnkX carrying a mutation of the catalytic histidine (Campanacci et al. 2013). The FIC domain (not drawn) is in the same orientation in all views. (D) The active site of Legionella AnkX (Campanacci et al. 2013) showing elements from outside the FIC motif that recognize the cytosine and choline moieties of CDP-choline. Below, the sequence of AnkX, highlighting these residues in blue. Residues from the β-hairpin, which is interrupted by an insertion in AnkX, are in cyan and the FIC motif is in red. (E) Comparison of the structures of the Bartonella VbhT-VbhA TA module (left panel, Goepfert et al. 2013) and Neisseiria FIC (right panel, PDB entry 2G03). The inhibitory glutamate is circled in red. The helix carrying the glutamate is located in the antitoxin in the VbhT-VbhA complex, and is in C-terminus in Nesseiria FIC (shown in orange). Table 1. Crystallographic FIC structures available in the Protein Data Bank.     Toxins secreted from bacterial pathogens are in pink; toxin/antitoxin modules are in green; bacterial single-domain FIC proteins with a regulatory glutamate are in blue; other bacterial FIC proteins with a regulatory glutamate are in khaki; animal FICD/HYPE are in orange; other proteins with FIC domains are in white. WT: wild-type. AMPPNP and AMPPCP are non-hydrolyzable ATP analogs. View Large Table 1. Crystallographic FIC structures available in the Protein Data Bank.     Toxins secreted from bacterial pathogens are in pink; toxin/antitoxin modules are in green; bacterial single-domain FIC proteins with a regulatory glutamate are in blue; other bacterial FIC proteins with a regulatory glutamate are in khaki; animal FICD/HYPE are in orange; other proteins with FIC domains are in white. WT: wild-type. AMPPNP and AMPPCP are non-hydrolyzable ATP analogs. View Large How FIC proteins bind to their protein substrate was observed in the structure of H. somni IbpA bound to AMPylated Cdc42, a small GTPase that regulates actin cytoskeleton dynamics (Xiao et al. 2010); as of today, it remains the only structure of an FIC protein bound to a protein substrate (Fig. 2B). This structure revealed that the target residue (here, a tyrosine from the small GTPase Cdc42) is presented into the active site as an extented peptide that binds to a β-hairpin found in all FIC domain proteins so far, except Doc. In a related protein with an FIC fold, AvrB, a peptide derived from the protein substrate associated with the β-hairpin in a similar manner (Desveaux et al. 2007). No significant conservation has been identified in the sequence of the β-hairpin across FIC proteins, although sequences could be clustered into seven subgroups based on a four-residue stretch that was predicted to contribute to the specificity of substrate recognition (Khater and Mohanty 2015a). These studies suggest that FIC proteins with a β-hairpin use it to recognize their substrates. Interestingly, two studies point to a contribution of structural flexibility to substrate recognition. In Cdc42, the target peptide is the so-called switch 1, known to be intrinsically flexible; the switch 1 adopts a distorted conformation in the complex with IbpA, suggesting that flexibility facilitates its recognition by the β-hairpin (Xiao et al. 2010). In a different study, introducing flexibility in single-domain bacterial FIC by mutations increased their modification by auto-AMPylation (Engel et al. 2012). Besides the FIC motif and the β-hairpin, IbpA interacts with Cdc42 by regions located outside the FIC domain, which explains the narrow specificity of the toxin for Rho GTPases. Structures of other FIC proteins, such as the bacterial toxin VopS (Luong et al. 2010), display non-FIC domains in the vicinity of their catalytic site, which may also contribute to the specific recognition of their protein target. Conversely, single-domain bacterial FIC proteins, such as Neisseria FIC (Engel et al. 2012), lack such additional elements, raising the possibility that some of them may recognize their substrates mostly through the FIC motif and the β-hairpin. Accordingly, the possibility that some of these proteins modify peptides rather than proteins cannot be excluded. Domains appended to the FIC domain can also have localization functions. Using small angle X-ray scattering, a method to observe the structure of proteins in solution, we found that the AnkX toxin has a horseshoe shape that allows it to bind simultaneously to membrane-attached Rab GTPases by its FIC domain and to the membrane by its C-terminal domain (our unpublished observations). These findings are consistent with the observation that AnkX binds to membranes in cells by elements located in its C-terminal half (Allgood et al. 2017). Structures of FIC proteins bound to co-factors or reaction products revealed both common and unique features of co-factor recognition and processing (reviewed in Garcia-Pino, Zenkin and Loris 2014; Roy and Cherfils 2015; Harms, Stanger and Dehio 2016; Table 1 and Fig. 2C). Notably, the structure of L. pneumophila AnkX in complex with intact CDP-choline, the co-factor for the phosphocholination of the small GTPase Rab1, showed that the FIC motif recognizes solely the diphosphate moiety of the co-factor, while the cytidine and choline moieties are recognized by elements located both upstream and downstream the FIC motif (Fig. 2D) (Campanacci et al. 2013). Structures of AMPylating FIC proteins also highlighted significant variations in the binding site of the adenine moiety of ATP, making the prediction of the nature of the co-factor difficult based on sequences only. In the future, a complementary strategy to discover co-factors of FIC proteins could be through structure-based docking, an approach that docks and ranks small molecules into crystal structures or structural models and is widely used for drug discovery (reviewed in Cheng et al. 2012). In the case of FIC proteins, all co-factors identified to date contain a diphosphate moiety, suggesting that this approach could be restricted to a diphosphate-containing compound library. Because FIC proteins appear to carry out reactions that stall major cellular processes, an important issue is whether they are regulated, and in which case, what are the modalities of this regulation. The structures of plasmid prophage Doc/Phd (Garcia-Pino et al. 2008; Arbing et al. 2010) and Bartonella VbhT-VbhA (Engel et al. 2012), and more recently E. coli FicT-FicA (Stanger et al. 2016a), TA modules provided a general principle for the inhibition of FIC toxins by their cognate antitoxins, in which the catalytic site of the toxin is obstructed by the antitoxin and dissociation of the antitoxin releases inhibition. In Bartonella VbhT-VbhA, inhibition is mediated by a glutamate from the VbhA antitoxin that prevents productive binding of the ATP co-substrate (Engel et al. 2012). Remarkably, a number of FIC proteins feature a structurally equivalent glutamate that wedges into the catalytic site, among which are single-domain bacterial FIC proteins and human FICD/HYPE (Engel et al. 2012) (Fig. 2E). Mutation of the glutamate unleashed a potent AMPylation activity in these proteins, leading to the hypothesis that this residue is autoinhibitory for the AMPylation reaction (Engel et al. 2012). Consistently, several structures of glutamate-containing FIC proteins highlighted that ATP binds in a non-canonical manner (Goepfert et al. 2013; Dedic et al. 2016) or in a manner in which only the ADP moiety is visible (Engel et al. 2012), suggesting that the glutamate conflicts with the ATP binding in a catalytically competent conformation. This hypothesis fueled intense research to discover signals or interactions able to trigger inhibition release. In the case of single-domain bacterial FIC proteins, recurrent observation of oligomers in the crystal led to propose that they are regulated by oligomerization, thereby enabling auto-AMPylation as a means to displace the autoinhibitory glutamate (Stanger et al. 2016b). In an unexpected twist, animal HYPE/FICD proteins were recently shown to encode de-AMPylation of the BIP chaperone in cells and in vitro, and this was dependent on this glutamate (Casey et al. 2017; Preissler et al. 2017). This discovery raises the intriguing possibility that bacterial FIC proteins that possess this glutamate may also encode dual AMPylation and de-AMPYlation activities. Another major question remaining is the nature of the cellular cues able to undergo variations that are large enough to promote the switch from the AMPylation reaction to the de-AMPylation reaction. CELL-FREE ASSAYS WITH RECOMBINANT PROTEINS: CHARACTERIZING THE ACTIVITY LANDSCAPE OF FIC PROTEINS Purified recombinant proteins have played a central role in understanding the biochemical activities of FIC proteins. Radiolabeling, in which the FIC protein is incubated with radioactive co-factors followed by autoradiography, is the standard assay to report on the PTM activity of FIC proteins. For example; 32P-α-ATP was used to establish the AMPylation activities of VopS (Yarbrough et al. 2009) and IbpA (Worby et al. 2009) toward Rho GTPases (Fig. 3A) and the AMPylation of BIP by FICD/HYPE (Ham et al. 2014; Preissler et al. 2015). Surprisingly, recombinant FICD/HYPE is also capable of AMPylating purified Rho GTPases in vitro (Worby et al. 2009), but a phenotype suggesting that it inactivates Rho GTPases in cells has not been described. Phosphorylation of the elongation factor Ef-Tu by recombinant Doc was demonstrated by its radiolabeling by 32P-γ-ATP but not 32P-α-ATP, using either purified Ef-Tu or E. coli extracts (Castro-Roa et al. 2013; Cruz et al. 2014). As an alternative to autoradiography, an anti-phosphocholine antibody was used to detect phosphocholination of Rab1 and Rab35 GTPases by AnkX (Mukherjee et al. 2011; Campanacci et al. 2013) and an anti AMP-threonine antibody was used to detect AMPylation of BIP by Drosophila Fic (the ortholog of HYPE/FICD) (Casey et al. 2017). Mass spectrometry is another approach that was used to identify or confirm candidate targets, using recombinant FIC proteins incubated with cell extracts treated with ATP. This approach established that recombinant bacterial VopS AMPylates human Rho GTPases in cell extracts (Yarbrough et al. 2009), and that the chaperone BIP is the substrate of human HYPE/FICD using a hyperactive mutant in which the inhibitory glutamate had been mutated (Ham et al. 2014). Figure 3. View largeDownload slide In vitro characterization of FIC proteins activity. (A) In vitro AMPylation of Rac-family GTPases by recombinant VopS. Recombinant human GTPases (Rac, Rho, Cdc42), active VopS from V. parahaemolyticus and inactive VopS harboring a mutation of the catalytic histidine were expressed in E. coli and purified. VopS constructs lacked the N-terminal secretion signal. GTPases loaded with GTP were then incubated with recombinant VopS proteins in the presence of radiolabeled ATP. Proteins were separated by SDS-PAGE and analyzed by autoradiography. Controls in which VopS was omitted were performed (lanes 1–3). Reproduced with permission from Yarbrough et al. (2009). (B) Kinetics of phosphocholination of human Rab1b by Legionella AnkX analyzed by intrinsic fluorescence measurement. Rab1b was loaded with GDP or with the non-hydrolysable GTP analog GppNHp. Addition of AnkX results in an exponential increase of tryptophane fluorescence (λexc = 297 nm and λem = 340 nm). The curves can be fitted by monoexponentials, giving apparent kcat/Km values of 9.8 × 104 M−1s−1 and 3.8 × 104 M−1s−1 respectively. Reproduced with permission from Goody et al. (2012). Figure 3. View largeDownload slide In vitro characterization of FIC proteins activity. (A) In vitro AMPylation of Rac-family GTPases by recombinant VopS. Recombinant human GTPases (Rac, Rho, Cdc42), active VopS from V. parahaemolyticus and inactive VopS harboring a mutation of the catalytic histidine were expressed in E. coli and purified. VopS constructs lacked the N-terminal secretion signal. GTPases loaded with GTP were then incubated with recombinant VopS proteins in the presence of radiolabeled ATP. Proteins were separated by SDS-PAGE and analyzed by autoradiography. Controls in which VopS was omitted were performed (lanes 1–3). Reproduced with permission from Yarbrough et al. (2009). (B) Kinetics of phosphocholination of human Rab1b by Legionella AnkX analyzed by intrinsic fluorescence measurement. Rab1b was loaded with GDP or with the non-hydrolysable GTP analog GppNHp. Addition of AnkX results in an exponential increase of tryptophane fluorescence (λexc = 297 nm and λem = 340 nm). The curves can be fitted by monoexponentials, giving apparent kcat/Km values of 9.8 × 104 M−1s−1 and 3.8 × 104 M−1s−1 respectively. Reproduced with permission from Goody et al. (2012). Autoradiography is also the standard assay to identify co-factors of FIC proteins. For example, it was used to show that both IbpA and VopS can use ATP and CTP to modify eukaryotic Cdc42 in vitro, while only VopS but not IbpA can use GTP (Mattoo et al. 2011). Another method to compare candidate co-factors is the thermal shift assay, which is based on the assumption that natural co-factors will bind more strongly to the FIC protein thereby increasing its thermal stability (Dedic et al. 2016). A shortcoming in characterizing the activity of FIC proteins is that the natural co-factor, the natural protein target or both are often not known. A convenient readout for activity has therefore been the analysis of automodifications by radiolabeled compounds (Kinch et al. 2009; Xiao et al. 2010; Mattoo et al. 2011; Engel et al. 2012; Feng et al. 2012; Goody et al. 2012; Mishra et al. 2012; Dedic et al. 2016; Preissler et al. 2017; our unpublished results). It should be emphasized that it has remained unclear whether automodifications are functionally important. In the case of single-domain FIC proteins, evidence based on mutants of the regulatory glutamate suggested that automodifications facilitate their conversion to an AMPylation-competent conformation (Engel et al. 2012; Stanger et al. 2016b). Likewise, mutation of auto-AMPylation sites in Pseudomonas Fic-1, which is related to the VbhT-VbhA TA module, impaired its ability to modify DNA gyrase (Lu et al. 2016). Conversely, study of autophosphoch olination of AnkX by time-resolved Fourier-transform infrared spectroscopy using photocleavable caged compound concluded that it is not relevant for catalysis (Gavriljuk et al. 2016) and some AnkX constructs with potent phosphocholination activity were not automodified (Campanacci et al. 2013). This body of observations points to potentially different usages of automodifications by FIC proteins. Gaining quantitative insight into the efficiency of FIC proteins is important to understand their regulation and this requires that their kinetics can be monitored accurately. Kinetics of phosphocholination of Rab GTPases by AnkX was measured using fluorescence, using either a coupled-enzyme assay, intrinsic tryptophan fluorescence or mant-guanine nucleotides fluorescence changes (Goody et al. 2012) (Fig. 3B). Likewise, the kinetics of de-AMPylation of the BIP chaperone by FICD was monitored by fluorescence anisotropy using an ATP derivative labeled by a fluorescent probe attached to the adenine base (Preissler et al. 2017). Semiquantitative kinetics can also be achieved under conditions where the reaction is sufficiently slow. For example, the kinetics of Cdc42 AMPylation by VopS was assayed by measuring the incorporation of radiolabeled AMP at several time points (Luong et al. 2010); similarly, the kinetics of Rab1 and Rab35 phosphocholination by AnkX and the stimulatory effect of membranes were determined using an anti-phosphocholine antibody (our unpublished results). In the future, a systematic determination of kinetics appears highly desirable to quantify the efficiency, specificity and regulation of FIC proteins. CELL-BASED AND IN VIVO ASSAYS: DISCOVERING THE FUNCTIONS OF FIC PROTEINS A primary use of cell-based assays is to test functional hypotheses regarding how FIC proteins manipulate cell signaling. In most cases, investigations of FIC functions have been carried out through ectopic expression in cells. For example, phenotypes resulting from ectopic expression of VopS (Yarbrough et al. 2009) and IbpA (Worby et al. 2009) in human cells guided the identification of their protein targets. Both toxins disrupted the actin cytoskeleton, a hallmark of the inactivation of Rho GTPases, which led to the discovery that they inactivate Rho GTPases by AMPylation (Fig. 4A). Deletion and ectopic expression of HYPE/FICD in human cells, which resulted in AMPylation patterns that could not be explained by a simple AMPylation function, was instrumental in guiding the discovery of irs de-AMPylation activity toward the BIP chaperone in the ER (Preissler et al. 2017) (Fig. 4B). In another study, heterologous expression of animal FICD in yeast induced an unexpected heat shock response; this guided the subsequent identification of a putative cytosolic activity of FICD in human cells which has yet to be characterized (Truttmann et al. 2017). For various bacterial FIC proteins that cannot be investigated directly in the species that express them, heterologous expression in E. coli has been used as a convenient surrogate to investigate their functions and identify candidate protein targets. For example, Bartonella VbhT was shown to modify E. coli DNA gyrase and DNA topoisomerase IV, suggesting that it also targets these enzymes in Bartonella (Harms et al. 2015). In another recent study, expression of Campylobacter fetus FIC proteins from in E. coli highlighted that the toxicity of Fic1 and its ability to induce a filamentous phenotype was neutralized by co-expression with Fic2, suggesting that Fic1 and Fic2 form a novel type of functional TA module (Sprenger et al. 2017). In these assays, it cannot be excluded that other substrates not present in E. coli remain to be discovered. Figure 4. View largeDownload slide In cellulo analysis of FIC proteins. (A) Ectopic expression of Vibrio VopS in human cells results in the collapse of the actin cytoskeleton, suggesting that VopS inactivates Rho family small GTPases. HeLa cells were transfected with an empty vector or with a recombinant vector driving heterologous expression of wild-type VopS or an inactive mutant (H348A) and then observed by confocal microscopy. Co-transfection with a vector expressing SFFV-eGFP fusion protein allows detection of transfected cells. Nuclei were identified by Hoechst stain. Scale bar, 10 μm. The abnormal rounded shape specifically observed upon expression of wild-type VopS is typical of actin cytoskeleton disorganization. Reproduced with permission from Yarbrough et al. (2009). (B) Cellular effect of FICD overexpression on de-AMPylation of the BIP chaperone. CHO-K1 cells, untransfected or transfected with increasing amounts of FICD expression vector, were exposed to cycloheximide (CHX, an inhibitor of protein synthesis) to promote BIP AMPylation. Proteins in lysates were then separated by native polyacrylamide gel electrophoresis and detected by western blot using an anti-FICD antibody. I, II and III indicate the different oligomerization states of BIP. The monomeric form (III) can be subdivided into unmodified (A) and AMPylated (B) forms. Increasing expression of FICD induces a decrease in BIP AMPylation, consistent with FICD acting as a de-AMPylating enzyme. Reproduced with permission from Preissler et al. (2017). Figure 4. View largeDownload slide In cellulo analysis of FIC proteins. (A) Ectopic expression of Vibrio VopS in human cells results in the collapse of the actin cytoskeleton, suggesting that VopS inactivates Rho family small GTPases. HeLa cells were transfected with an empty vector or with a recombinant vector driving heterologous expression of wild-type VopS or an inactive mutant (H348A) and then observed by confocal microscopy. Co-transfection with a vector expressing SFFV-eGFP fusion protein allows detection of transfected cells. Nuclei were identified by Hoechst stain. Scale bar, 10 μm. The abnormal rounded shape specifically observed upon expression of wild-type VopS is typical of actin cytoskeleton disorganization. Reproduced with permission from Yarbrough et al. (2009). (B) Cellular effect of FICD overexpression on de-AMPylation of the BIP chaperone. CHO-K1 cells, untransfected or transfected with increasing amounts of FICD expression vector, were exposed to cycloheximide (CHX, an inhibitor of protein synthesis) to promote BIP AMPylation. Proteins in lysates were then separated by native polyacrylamide gel electrophoresis and detected by western blot using an anti-FICD antibody. I, II and III indicate the different oligomerization states of BIP. The monomeric form (III) can be subdivided into unmodified (A) and AMPylated (B) forms. Increasing expression of FICD induces a decrease in BIP AMPylation, consistent with FICD acting as a de-AMPylating enzyme. Reproduced with permission from Preissler et al. (2017). Infection of cells by intracellular bacterial pathogens led to important functional findings. For example, reasoning that Legionella effectors interfere with the Rab GTPase machinery, Rab GTPases from macrophages infected by Legionella strains that do or do not secrete AnkX were analyzed by mass spectrometry, revealing that AnkX catalyzes their phosphocholination (Mukherjee et al. 2011). More recently, observation of ectopically expressed AnkX by immunogold transmission electron microscopy and confocal microscopy showed that it localizes on endosomal membranes of human cells, subsequently revealing that it disrupts endosomal recycling in the course of infection (Allgood et al. 2017). In another cell infection study, the Doc-Phd TA module of the pathogen Salmonella typhimurium was shown to be required for the formation of macrophage-induced non-replicating persisters (Helaine et al. 2014). Besides these infection studies, there has only been a few in vivo studies to date. In one study, deletion of Drosophila Fic (an HYPE ortholog) was shown to result in blindness in flies (Rahman et al. 2012). In another study, the Caenorhabditis elegans ortholog of FICD/HYPE was shown to AMPylate cytosolic proteins in vivo to modulate the immune response (Truttmann et al. 2016); paralleling these observations, expression in human cells of HYPE carrying an activating mutation resulted in the AMPylation of cytosolic chaperones (Truttmann et al. 2017). CONCLUDING REMARKS The field of FIC proteins continues to grow, with elegant experimental strategies revealing new surprises. New chemical tools are currently being devised, such as AMPylation-specific antibodies (Hao et al. 2011; Smit et al. 2011), camelid-derived nanobodies to modulate AMPylation (Truttmann et al. 2015), chemical reporters of AMPylation (Grammel et al. 2011; Truttmann et al. 2016) or of phosphocholination (Heller et al. 2015) that bind covalently to modified protein targets using click chemistry, or labeling with stable isosope-labeled ATP (Pieles et al. 2014). Some of these tools can be used in combination with mass spectrometry to determine the PTM profiles of FIC proteins in cells (Yu et al. 2014; Broncel et al. 2016). Together, this broad range of computational and experimental methods has allowed and cross-validated important discoveries and they should continue to fertilize each other to understand the biochemical activities, regulations and functions of FIC proteins. Likewise, discoveries in the bacterial and animal kingdoms should continue to contribute jointly to an integrated understanding of the functional landscape of this fascinating family. Ultimately, this knowledge may guide the discovery of novel therapeutic strategies in infections and proteostasis-related diseases. FUNDING This work was supported by grants from the Fondation pour la Recherche Médicale and from the Agence Nationale pour la Recherche to JC and by a grant from the DIM MALINF to SV. Conflict of interest. None declared. REFERENCES Allgood SC, Romero Duenas BP, Noll RR et al.   Legionella Effector AnkX Disrupts Host Cell Endocytic Recycling in a Phosphocholination-Dependent Manner, Front Cell Infect Microbiol  2017; 7: 397. Google Scholar CrossRef Search ADS PubMed  Arbing MA, Handelman SK, Kuzin AP et al.   Crystal structures of Phd-Doc, HigA, and YeeU establish multiple evolutionary links between microbial growth-regulating toxin-antitoxin systems, Structure  2010; 18: 996– 1010. Google Scholar CrossRef Search ADS PubMed  Broncel M, Serwa RA, Bunney TD et al.   Global Profiling of Huntingtin-associated protein E (HYPE)-Mediated AMPylation through a Chemical Proteomic Approach, Molecular & cellular proteomics: MCP  2016; 15: 715– 25. Google Scholar CrossRef Search ADS PubMed  Bunney TD, Cole AR, Broncel M et al.   Crystal structure of the human, FIC-domain containing protein HYPE and implications for its functions, Structure  2014; 22: 1831– 43. Google Scholar CrossRef Search ADS PubMed  Campanacci V, Mukherjee S, Roy CR et al.   Structure of the Legionella effector AnkX reveals the mechanism of phosphocholine transfer by the FIC domain, EMBO J  2013; 32: 1469– 77. Google Scholar CrossRef Search ADS PubMed  Casey AK, Orth K. Enzymes Involved in AMPylation and deAMPylation, Chem Rev  2017b; 118: 1199– 215. Google Scholar CrossRef Search ADS   Casey AK, Moehlman AT, Zhang J et al.   Fic-mediated deAMPylation is not dependent on homodimerization and rescues toxic AMPylation in flies, J Biol Chem  2017; 292: 21193– 204. Google Scholar CrossRef Search ADS PubMed  Castro-Roa D, Garcia-Pino A, De Gieter S et al.   The Fic protein Doc uses an inverted substrate to phosphorylate and inactivate EF-Tu, Nat Chem Biol  2013; 9: 811– 7. Google Scholar CrossRef Search ADS PubMed  Cheng T, Li Q, Zhou Z et al.   Structure-Based Virtual Screening for Drug Discovery: a Problem-Centric Review, AAPS J . 2012; 14: 133– 41. Google Scholar CrossRef Search ADS PubMed  Cruz JW, Woychik NA. Teaching Fido New ModiFICation Tricks, PLoS Pathog  2014; 10: e1004349. Google Scholar CrossRef Search ADS PubMed  Cruz JW, Rothenbacher FP, Maehigashi T et al.   Doc toxin is a kinase that inactivates elongation factor Tu, J Biol Chem  2014; 289: 7788– 98. Google Scholar CrossRef Search ADS PubMed  Dedic E, Alsarraf H, Welner DH et al.   A Novel Fic (Filamentation Induced by cAMP) Protein from Clostridium difficile Reveals an Inhibitory Motif-independent Adenylylation/AMPylation Mechanism, J Biol Chem  2016; 291: 13286– 300. Google Scholar CrossRef Search ADS PubMed  Desveaux D, Singer AU, Wu AJ et al.   Type III effector activation via nucleotide binding, phosphorylation, and host target interaction, PLoS Pathog  2007; 3: e48. Google Scholar CrossRef Search ADS PubMed  Engel P, Goepfert A, Stanger FV et al.   Adenylylation control by intra- or intermolecular active-site obstruction in Fic proteins, Nature  2012; 482: 107– 10. Google Scholar CrossRef Search ADS PubMed  Feng F, Yang F, Rong W et al.   A Xanthomonas uridine 5'-monophosphate transferase inhibits plant immune kinases, Nature  2012; 485: 114– 8. Google Scholar CrossRef Search ADS PubMed  Finn RD, Coggill P, Eberhardt RY et al.   The Pfam protein families database: towards a more sustainable future, Nucleic Acids Res  2016; 44: D279– 85. Google Scholar CrossRef Search ADS PubMed  Garcia-Pino A, Zenkin N, Loris R. The many faces of Fic: structural and functional aspects of Fic enzymes, Trends Biochem Sci  2014; 39: 121– 9. Google Scholar CrossRef Search ADS PubMed  Garcia-Pino A, Christensen-Dalsgaard M, Wyns L et al.   Doc of prophage P1 is inhibited by its antitoxin partner Phd through fold complementation, J Biol Chem  2008; 283: 30821– 7. Google Scholar CrossRef Search ADS PubMed  Garcia-Pino A, Balasubramanian S, Wyns L et al.   Allostery and intrinsic disorder mediate transcription regulation by conditional cooperativity, Cell  2010; 142: 101– 11. Google Scholar CrossRef Search ADS PubMed  Gavriljuk K, Schartner J, Seidel H et al.   Unraveling the Phosphocholination Mechanism of the Legionella pneumophila Enzyme AnkX, Biochemistry  2016; 55: 4375– 85. Google Scholar CrossRef Search ADS PubMed  Goepfert A, Stanger FV, Dehio C et al.   Conserved inhibitory mechanism and competent ATP binding mode for adenylyltransferases with fic fold, PLoS One  2013; 8: e64901. Google Scholar CrossRef Search ADS PubMed  Goody PR, Heller K, Oesterlin LK et al.   Reversible phosphocholination of Rab proteins by Legionella pneumophila effector proteins, EMBO J  2012; 31: 1774– 84. Google Scholar CrossRef Search ADS PubMed  Grammel M, Luong P, Orth K et al.   A chemical reporter for protein AMPylation, J Am Chem Soc  2011; 133: 17103– 5. Google Scholar CrossRef Search ADS PubMed  Ham H, Woolery AR, Tracy C et al.   Unfolded Protein Response-regulated Drosophila Fic (dFic) Protein Reversibly AMPylates BiP Chaperone during Endoplasmic Reticulum Homeostasis, J Biol Chem  2014; 289: 36059– 69. Google Scholar CrossRef Search ADS PubMed  Hao YH, Chuang T, Ball HL et al.   Characterization of a rabbit polyclonal antibody against threonine-AMPylation, J Biotechnol  2011; 151: 251– 4. Google Scholar CrossRef Search ADS PubMed  Harms A, Stanger FV, Dehio C. Biological Diversity and Molecular Plasticity of FIC Domain Proteins, Annu Rev Microbiol  2016; 70: 341– 60. Google Scholar CrossRef Search ADS PubMed  Harms A, Stanger FV, Scheu PD et al.   Adenylylation of Gyrase and Topo IV by FicT Toxins Disrupts Bacterial DNA Topology, Cell Rep  2015; 12: 1497– 507. Google Scholar CrossRef Search ADS PubMed  Helaine S, Cheverton AM, Watson KG et al.   Internalization of Salmonella by macrophages induces formation of nonreplicating persisters, Science  2014; 343: 204– 8. Google Scholar CrossRef Search ADS PubMed  Heller K, Ochtrop P, Albers MF et al.   Covalent Protein Labeling by Enzymatic Phosphocholination, Angew Chem Int Ed Engl  2015; 54: 10327– 30. Google Scholar CrossRef Search ADS PubMed  Kawamukai M, Matsuda H, Fujii W et al.   Nucleotide sequences of fic and fic-1 genes involved in cell filamentation induced by cyclic AMP in Escherichia coli, J Bacteriol  1989; 171: 4525– 9. Google Scholar CrossRef Search ADS PubMed  Kawamukai M, Matsuda H, Fujii W et al.   Cloning of the fic-1 gene involved in cell filamentation induced by cyclic AMP and construction of a delta fic Escherichia coli strain, J Bacteriol  1988; 170: 3864– 9. Google Scholar CrossRef Search ADS PubMed  Khater S, Mohanty D. In silico identification of AMPylating enzymes and study of their divergent evolution, Sci Rep  2015a; 5: 10804. Google Scholar CrossRef Search ADS   Khater S, Mohanty D. Deciphering the Molecular Basis of Functional Divergence in AMPylating Enzymes by Molecular Dynamics Simulations and Structure Guided Phylogeny, Biochemistry  2015b; 54: 5209– 24. Google Scholar CrossRef Search ADS   Kinch LN, Yarbrough ML, Orth K et al.   Fido, a novel AMPylation domain common to fic, doc, and AvrB, PLoS One  2009; 4: e5818. Google Scholar CrossRef Search ADS PubMed  Lee CC, Wood MD, Ng K et al.   Crystal structure of the type III effector AvrB from Pseudomonas syringae, Structure  2004; 12: 487– 94. Google Scholar CrossRef Search ADS PubMed  Lu C, Nakayasu ES, Zhang LQ et al.   Identification of Fic-1 as an enzyme that inhibits bacterial DNA replication by AMPylating GyrB, promoting filament formation, Sci Signal  2016; 9: ra11. Google Scholar CrossRef Search ADS PubMed  Luong P, Kinch LN, Brautigam CA et al.   Kinetic and structural insights into the mechanism of AMPylation by VopS Fic domain, J Biol Chem  2010; 285: 20155– 63. Google Scholar CrossRef Search ADS PubMed  Mattoo S, Durrant E, Chen MJ et al.   Comparative analysis of Histophilus somni immunoglobulin-binding protein A (IbpA) with other fic domain-containing enzymes reveals differences in substrate and nucleotide specificities, J Biol Chem  2011; 286: 32834– 42. Google Scholar CrossRef Search ADS PubMed  Mishra S, Bhagavat R, Chandra N et al.   Cloning, expression, purification, and biochemical characterisation of the FIC motif containing protein of Mycobacterium tuberculosis, Protein Expr Purif  2012; 86: 58– 67. Google Scholar CrossRef Search ADS PubMed  Mukherjee S, Liu X, Arasaki K et al.   Modulation of Rab GTPase function by a protein phosphocholine transferase, Nature  2011; 477: 103– 6. Google Scholar CrossRef Search ADS PubMed  Palanivelu DV, Goepfert A, Meury M et al.   Fic domain-catalyzed adenylylation: insight provided by the structural analysis of the type IV secretion system effector BepA, Protein Sci  2011; 20: 492– 9. Google Scholar CrossRef Search ADS PubMed  Pieles K, Glatter T, Harms A et al.   An experimental strategy for the identification of AMPylation targets from complex protein samples, Proteomics  2014; 14: 1048– 52. Google Scholar CrossRef Search ADS PubMed  Preissler S, Rato C, Perera L et al.   FICD acts bifunctionally to AMPylate and de-AMPylate the endoplasmic reticulum chaperone BiP, Nature structural & molecular biology  2017; 24: 23– 29. Google Scholar CrossRef Search ADS PubMed  Preissler S, Rato C, Chen R et al.   AMPylation matches BiP activity to client protein load in the endoplasmic reticulum, eLife  2015; 4: e12621. Google Scholar PubMed  Rahman M, Ham H, Liu X et al.   Visual neurotransmission in Drosophila requires expression of Fic in glial capitate projections, Nat Neurosci  2012; 15: 871– 5. Google Scholar CrossRef Search ADS PubMed  Roy CR, Mukherjee S. Bacterial FIC Proteins AMP Up Infection, Sci Signal  2009; 2: pe14. Google Scholar CrossRef Search ADS PubMed  Roy CR, Cherfils J Structure and function of Fic proteins, Nat Rev Microbiol  2015; 13: 631– 40. Google Scholar CrossRef Search ADS PubMed  Sanyal A, Chen AJ, Nakayasu ES et al.   A Novel Link between Fic (Filamentation Induced by cAMP)-mediated Adenylylation/AMPylation and the Unfolded Protein Response, J Biol Chem  2015; 290: 8482– 99. Google Scholar CrossRef Search ADS PubMed  Smit C, Blumer J, Eerland MF et al.   Efficient synthesis and applications of peptides containing adenylylated tyrosine residues, Angew Chem Int Ed Engl  2011; 50: 9200– 4. Google Scholar CrossRef Search ADS PubMed  Sprenger H, Kienesberger S, Pertschy B et al.   Fic Proteins of Campylobacter fetus subsp. venerealis Form a Network of Functional Toxin-Antitoxin Systems, Front Microbiol  2017; 8: 1965. Google Scholar CrossRef Search ADS PubMed  Stanger FV, Harms A, Dehio C et al.   Crystal Structure of the Escherichia coli Fic Toxin-Like Protein in Complex with Its Cognate Antitoxin, PLoS One  2016a; 11: e0163654. Google Scholar CrossRef Search ADS   Stanger FV, Burmann BM, Harms A et al.   Intrinsic regulation of FIC-domain AMP-transferases by oligomerization and automodification, Proc Natl Acad Sci U S A  2016b; 113: E529– 37. Google Scholar CrossRef Search ADS   Truttmann MC, Ploegh HL. rAMPing Up Stress Signaling: Protein AMPylation in Metazoans, Trends Cell Biol  2017; 27: 608– 20. Google Scholar CrossRef Search ADS PubMed  Truttmann MC, Wu Q, Stiegeler S et al.   HypE-specific nanobodies as tools to modulate HypE-mediated target AMPylation, J Biol Chem  2015; 290: 9087– 100. Google Scholar CrossRef Search ADS PubMed  Truttmann MC, Cruz VE, Guo X et al.   The Caenorhabditis elegans Protein FIC-1 Is an AMPylase That Covalently Modifies Heat-Shock 70 Family Proteins, Translation Elongation Factors and Histones, PLoS Genet  2016; 12: e1006023. Google Scholar CrossRef Search ADS PubMed  Truttmann MC, Zheng X, Hanke L et al.   Unrestrained AMPylation targets cytosolic chaperones and activates the heat shock response, Proc Natl Acad Sci U S A  2017; 114: E152– E60. Google Scholar CrossRef Search ADS PubMed  Worby CA, Mattoo S, Kruger RP et al.   The fic domain: regulation of cell signaling by adenylylation, Mol Cell  2009; 34: 93– 103. Google Scholar CrossRef Search ADS PubMed  Xiao J, Worby CA, Mattoo S et al.   Structural basis of Fic-mediated adenylylation, Nat Struct Mol Biol  2010; 17: 1004– 10. Google Scholar CrossRef Search ADS PubMed  Yarbrough ML, Li Y, Kinch LN et al.   AMPylation of Rho GTPases by Vibrio VopS disrupts effector binding and downstream signaling, Science  2009; 323: 269– 72. Google Scholar CrossRef Search ADS PubMed  Yu X, Woolery AR, Luong P et al.   Copper-catalyzed azide-alkyne cycloaddition (click chemistry)-based detection of global pathogen-host AMPylation on self-assembled human protein microarrays, Mol Cell Proteomics  2014; 13: 3164– 76. Google Scholar CrossRef Search ADS PubMed  © FEMS 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Pathogens and Disease Oxford University Press

FIC proteins: from bacteria to humans and back again

Loading next page...
 
/lp/ou_press/fic-proteins-from-bacteria-to-humans-and-back-again-v5ZvZFuR9M
Publisher
Oxford University Press
Copyright
© FEMS 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
2049-632X
eISSN
2049-632X
D.O.I.
10.1093/femspd/fty012
Publisher site
See Article on Publisher Site

Abstract

Abstract During the last decade, FIC proteins have emerged as a large family comprised of a variety of bacterial enzymes and a single member in animals. The air de famille of FIC proteins stems from a domain of conserved structure, which catalyzes the post-translational modification of proteins (PTM) by a phosphate-containing compound. In bacteria, examples of FIC proteins include the toxin component of toxin/antitoxin modules, such as Doc-Phd and VbhT-VbhA, toxins secreted by pathogenic bacteria to divert host cell processes, such as VopS, IbpA and AnkX, and a vast majority of proteins of unknown functions. FIC proteins catalyze primarily the transfer of AMP (AMPylation), but they are not restricted to this PTM and also carry out other modifications, for example by phosphocholine or phosphate. In a recent twist, animal FICD/HYPE was shown to catalyze both AMPylation and de-AMPylation of the endoplasmic reticulum BIP chaperone to regulate the unfolded protein response. FICD shares structural features with some bacterial FIC proteins, raising the possibility that bacteria also encode such dual activities. In this review, we discuss how structural, biochemical and cellular approaches have fertilized each other to understand the mechanism, regulation and function of FIC proteins from bacterial pathogens to humans. FIC proteins, structural biology, biochemistry, toxins, cellular biology, pathogens FIC PROTEINS: FROM THE AIR DE FAMILLE TO FUNCTION FIC proteins are widespread in bacteria, with a single member found in animals. They have attracted considerable interest over the last decade for their functions in bacterial stress responses and in infections (reviewed in Garcia-Pino, Zenkin and Loris 2014; Roy and Cherfils 2015; Harms, Stanger and Dehio 2016; Casey and Orth 2017) and recently in proteostasis control in animal cells (reviewed in Truttmann and Ploegh 2017). The air de famille of FIC proteins stems from a domain of conserved 3D structure with a signature sequence (the FIC motif) that catalyzes the post-translational modification (PTM) of proteins by a phosphate-containing compound. This domain was called Fido (FIC domain) as an umbrella term to refer to FIC proteins and other proteins with the same overall fold (Kinch et al. 2009). In bacteria, genes coding for FIC proteins are often found in mobile genomic islands which may have facilitated their spread across bacterial species (Khater and Mohanty 2015b). A commonality of FIC proteins with known functions is that they carry out the PTM of target proteins by a phosphate-containing compound, and this enzymatic activity is entirely comprised in the FIC domain. This function was not known when the first fic gene was discovered in Escherichia coli 30 years ago, in which a point mutation was found to impair cell division under stress conditions (cyclic AMP in growth medium at high temperature), leading to filamentation (Kawamukai et al. 1988, 1989). This now anecdotal observation led to naming this gene fic-1 for filamentation induced by cAMP, and later to extend the name to the conserved FIC domain (Worby et al. 2009; Yarbrough et al. 2009) and eventually to the entire family (Kinch et al. 2009). The product of the fic-1 gene was recently characterized as the toxin component of a toxin-antitoxin (TA) module structurally related to the VbhT-VbhA TA module and has now been renamed EcFicT (Stanger et al. 2016a). The function of EcFicT has not yet been elucidated, but functions of related FIC proteins have begun to be unraveled. The related Pseudomonas fluorescens Fic-1 inhibits DNA replication by adding AMP to DNA gyrase (Lu et al. 2016), a reaction coined AMPylation (also referred to as adenylylation). Likewise, ectopic expression of Bartonella schoenbuchensis VbhT in E. coli disrupts DNA topology by AMPylation of DNA gyrase and DNA topoisomerase IV (Harms et al. 2015). Thus, it is likely that EcFicT also functions in controlling DNA replication. It took more than three decades for the FIC family to enter the limelight, following the landmark discovery that toxins carrying a FIC domain are secreted by bacterial pathogens to divert small GTPases of infected cells through AMPylation (Yarbrough et al. 2009). Vibrio parahaemolyticus VopS (Yarbrough et al. 2009) then Histophilus somni IbpA (Worby et al. 2009) were both shown to AMPylate small GTPases of the Rho family, which inactivates them leading to the collapse of the actin cytoskeleton. This reaction uses ATP as donor for AMP, which is transferred to hydroxyl-containing residues (tyrosine, serine or threonine) from the target protein. While AMPylation was briefly thought to be the reaction unifying the family (reviewed in Roy and Mukherjee 2009), alternative enzymatic activities were soon discovered, including phosphocholination of Rab small GTPases by the Legionella pneumophila type IV effector AnkX (Mukherjee et al. 2011), phosphorylation of the GTPase-related elongation factor Ef-Tu by the bacteriophage P1 TA Doc-Phd module in E. coli (Castro-Roa et al. 2013; Cruz et al. 2014) and addition of UMP to plant immune kinases by an FIC protein from the plant pathogen Xanthomonas campestris (Feng et al. 2012). The reaction landscape of FIC proteins was accordingly extended to the more general function of transfering a phosphate-containing compound to a target protein leading to signaling roadblocks (reviewed in Cruz and Woychik 2014; Roy and Cherfils 2015; Harms, Stanger and Dehio 2016). In an unexpected twist, the sole animal protein with a FIC domain, called HYPE or FICD, was recently shown to carry out both the addition (Ham et al. 2014; Preissler et al. 2015; Sanyal et al. 2015) and removal (Casey et al. 2017; Preissler et al. 2017) of AMP from the endoplasmic reticulum chaperone BIP. De-AMPylation was shown to depend on a glutamate located outside the catalytic FIC motif (Preissler et al. 2017), which is structurally conserved in FICD/HYPE and in a subset of bacterial FIC proteins where it is thought to serve an autoinhibitory function (Engel et al. 2012). These recent discoveries now raise new questions, and important issues remain uncompletely understood. For example, what is the protein substrate, the PTM donor (refered to as the co-factor hereafter) and the primary PTM carried out by individual FIC proteins? How are animal and bacterial FIC proteins regulated? Is de-AMPylation also found in bacterial FIC proteins? Ultimately, what are the functions of the still predominently uncharacterized bacterial FIC proteins? In this review, we discuss how complementary bioinformatics, structural, biochemical, cellular and in vivo approaches have fertilized each other to generate and test hypothesis about the mechanism, regulation and function of FIC proteins from bacterial pathogens to animals. BIOINFORMATICS: CLASSIFYING FIC PROTEINS AND ANALYZING PHYLOGENY High-throughput sequencing of bacterial genomes identified potential coding genes for proteins with FIC signatures in a large number of bacterial species, including free living, host-associated, commensal and pathogenic species. For example, as of 2017 the PFAM database reports about 7500 sequences that contain a FIC domain, organized in 96 architectures found in >2300 bacteria, 360 eukaryotes and 71 archae species (Fig. 1). In many cases, these genes were annotated in the databases as ‘filamentation induced by cAMP protein’ or ‘adenosine monophosphate-protein transferase’. While the first annotation is now mostly anecdotal, the second annotation may be somewhat misleading as enzymatic activities other than AMPylation continue to be discovered. We suggest that unknown genes should therefore better be annotated as ‘uncharacterized FIC family protein’ to reflect that not all FIC proteins carry out AMPylation. Sequence databases have provided a rich material to classify FIC proteins and investigate their phylogeny (Engel et al. 2012; Garcia-Pino, Zenkin and Loris 2014; Harms et al. 2015; Khater and Mohanty 2015a,b; Kinch et al. 2009; Worby et al. 2009; Yarbrough et al. 2009) and to forge hypotheses about regulatory mechanisms (Engel et al. 2012). However, predicting whether an FIC protein of unknown function is an actual AMPylator based on its sequence alone stills remains uncertain. We discuss below how structural biology highlighted functional diversity encoded in FIC sequences, while experiments using recombinant proteins and cellular assays have remained the primary approaches for discovering the enzymatic activities and functions of individual FIC proteins. Figure 1. View largeDownload slide A graphical representation of the distribution of Fic genes (PF02661) across taxonomic levels. Reproduced from the EMBL-EBI PFAM server (http://pfam.xfam.org/family/Fic#tabview = tab7) (Finn et al. 2016). The distribution among the eight major taxonomic levels is shown, starting with superkingdom in the center of the sunburst, and then concentric circles corresponding successively to kingdom, phylum, class, order, family, genus and species. Each circle is subdivided into colored arcs corresponding to superkingdoms and/or kingdoms. The longer the arc, the higher the number of species harboring at least one FIC gene. On the outermost circle, one line corresponds to one species, irrespectively of the number of FIC genes (one or more) in its genome. It should be noted that except for the Doc/Phd module harbored on a bacteriophage P1 plasmid in E. coli, FIC proteins have not been identified in viruses. Figure 1. View largeDownload slide A graphical representation of the distribution of Fic genes (PF02661) across taxonomic levels. Reproduced from the EMBL-EBI PFAM server (http://pfam.xfam.org/family/Fic#tabview = tab7) (Finn et al. 2016). The distribution among the eight major taxonomic levels is shown, starting with superkingdom in the center of the sunburst, and then concentric circles corresponding successively to kingdom, phylum, class, order, family, genus and species. Each circle is subdivided into colored arcs corresponding to superkingdoms and/or kingdoms. The longer the arc, the higher the number of species harboring at least one FIC gene. On the outermost circle, one line corresponds to one species, irrespectively of the number of FIC genes (one or more) in its genome. It should be noted that except for the Doc/Phd module harbored on a bacteriophage P1 plasmid in E. coli, FIC proteins have not been identified in viruses. STRUCTURAL BIOLOGY: DISSECTING MECHANISMS AND INSPIRING HYPOTHESES The first high-resolution structures of an FIC protein were those of single-domain FIC proteins from Helicobacter pylori and Neisseria meningitidis, which were determined as part of Structural Genomics Initiatives in 2006 (PDB entries 2G03 and 2F6S, which have no associated publications). However, the structure of FIC proteins only began to be discussed with the structure of the Doc/Phd TA complex (Garcia-Pino et al. 2008) and soon afterwards with the discovery of AMPylating FIC toxins from pathogenic bacteria (Worby et al. 2009; Yarbrough et al. 2009; Xiao et al. 2010). As of today, the structures of >20 FIC proteins have been determined, several of them in complex with a co-factor bound to the active site, while only one has a bound protein substrate (Table 1 and references therein). These structures established the remarkable conservation of the FIC domain and of the FIC motif (HPFx[D/E]GNGR, with variations) that carries the invariant catalytic histidine (Fig. 2A). Importantly, they provided a framework to identify conserved catalytic and substrate recognition features as well as divergences and they irrigated the field with fruitful hypotheses (reviewed in Garcia-Pino, Zenkin and Loris 2014; Roy and Cherfils 2015; Harms, Stanger and Dehio 2016). We present below some of the milestone structures that shaped our current understanding. Figure 2. View largeDownload slide Structural features of FIC proteins. (A) The canonical FIC domain, exemplified by the structure of N. meningitidis. The FIC motif is in pink and the catalytic histidine shown; the β-hairpin is in cyan. A bound ATP co-factor, of which only the ADP moiety is visible, is shown in stick representation. The same color coding is used in the other panels. All PDB entries and references are given in Table 1. (B) The structure of the H. somni IbpA toxin bound the human small GTPase Cdc42 (Xiao et al. 2010). Cdc42 is in purple and IbpA is in gray. The AMPylated tyrosine in the switch 1 of Cdc42 is indicated. (C) Examples of co-factors found in crystal structures of FIC proteins. AMP-Tyr is from Histophilus IbpA in complex with AMPylated Cdc42 (Xiao et al. 2010), ADP-Mg2+ is from human FICD/HYPE carrying a mutation of the inhibitory glutamate (Bunney et al. 2014), ATP-Mg2+ is from Bartonella VbhT in complex with the VbhA antitoxin carrying a mutation of the inhibitory glutamate (Goepfert et al. 2013), CDP-choline is from Legionella AnkX carrying a mutation of the catalytic histidine (Campanacci et al. 2013). The FIC domain (not drawn) is in the same orientation in all views. (D) The active site of Legionella AnkX (Campanacci et al. 2013) showing elements from outside the FIC motif that recognize the cytosine and choline moieties of CDP-choline. Below, the sequence of AnkX, highlighting these residues in blue. Residues from the β-hairpin, which is interrupted by an insertion in AnkX, are in cyan and the FIC motif is in red. (E) Comparison of the structures of the Bartonella VbhT-VbhA TA module (left panel, Goepfert et al. 2013) and Neisseiria FIC (right panel, PDB entry 2G03). The inhibitory glutamate is circled in red. The helix carrying the glutamate is located in the antitoxin in the VbhT-VbhA complex, and is in C-terminus in Nesseiria FIC (shown in orange). Figure 2. View largeDownload slide Structural features of FIC proteins. (A) The canonical FIC domain, exemplified by the structure of N. meningitidis. The FIC motif is in pink and the catalytic histidine shown; the β-hairpin is in cyan. A bound ATP co-factor, of which only the ADP moiety is visible, is shown in stick representation. The same color coding is used in the other panels. All PDB entries and references are given in Table 1. (B) The structure of the H. somni IbpA toxin bound the human small GTPase Cdc42 (Xiao et al. 2010). Cdc42 is in purple and IbpA is in gray. The AMPylated tyrosine in the switch 1 of Cdc42 is indicated. (C) Examples of co-factors found in crystal structures of FIC proteins. AMP-Tyr is from Histophilus IbpA in complex with AMPylated Cdc42 (Xiao et al. 2010), ADP-Mg2+ is from human FICD/HYPE carrying a mutation of the inhibitory glutamate (Bunney et al. 2014), ATP-Mg2+ is from Bartonella VbhT in complex with the VbhA antitoxin carrying a mutation of the inhibitory glutamate (Goepfert et al. 2013), CDP-choline is from Legionella AnkX carrying a mutation of the catalytic histidine (Campanacci et al. 2013). The FIC domain (not drawn) is in the same orientation in all views. (D) The active site of Legionella AnkX (Campanacci et al. 2013) showing elements from outside the FIC motif that recognize the cytosine and choline moieties of CDP-choline. Below, the sequence of AnkX, highlighting these residues in blue. Residues from the β-hairpin, which is interrupted by an insertion in AnkX, are in cyan and the FIC motif is in red. (E) Comparison of the structures of the Bartonella VbhT-VbhA TA module (left panel, Goepfert et al. 2013) and Neisseiria FIC (right panel, PDB entry 2G03). The inhibitory glutamate is circled in red. The helix carrying the glutamate is located in the antitoxin in the VbhT-VbhA complex, and is in C-terminus in Nesseiria FIC (shown in orange). Table 1. Crystallographic FIC structures available in the Protein Data Bank.     Toxins secreted from bacterial pathogens are in pink; toxin/antitoxin modules are in green; bacterial single-domain FIC proteins with a regulatory glutamate are in blue; other bacterial FIC proteins with a regulatory glutamate are in khaki; animal FICD/HYPE are in orange; other proteins with FIC domains are in white. WT: wild-type. AMPPNP and AMPPCP are non-hydrolyzable ATP analogs. View Large Table 1. Crystallographic FIC structures available in the Protein Data Bank.     Toxins secreted from bacterial pathogens are in pink; toxin/antitoxin modules are in green; bacterial single-domain FIC proteins with a regulatory glutamate are in blue; other bacterial FIC proteins with a regulatory glutamate are in khaki; animal FICD/HYPE are in orange; other proteins with FIC domains are in white. WT: wild-type. AMPPNP and AMPPCP are non-hydrolyzable ATP analogs. View Large How FIC proteins bind to their protein substrate was observed in the structure of H. somni IbpA bound to AMPylated Cdc42, a small GTPase that regulates actin cytoskeleton dynamics (Xiao et al. 2010); as of today, it remains the only structure of an FIC protein bound to a protein substrate (Fig. 2B). This structure revealed that the target residue (here, a tyrosine from the small GTPase Cdc42) is presented into the active site as an extented peptide that binds to a β-hairpin found in all FIC domain proteins so far, except Doc. In a related protein with an FIC fold, AvrB, a peptide derived from the protein substrate associated with the β-hairpin in a similar manner (Desveaux et al. 2007). No significant conservation has been identified in the sequence of the β-hairpin across FIC proteins, although sequences could be clustered into seven subgroups based on a four-residue stretch that was predicted to contribute to the specificity of substrate recognition (Khater and Mohanty 2015a). These studies suggest that FIC proteins with a β-hairpin use it to recognize their substrates. Interestingly, two studies point to a contribution of structural flexibility to substrate recognition. In Cdc42, the target peptide is the so-called switch 1, known to be intrinsically flexible; the switch 1 adopts a distorted conformation in the complex with IbpA, suggesting that flexibility facilitates its recognition by the β-hairpin (Xiao et al. 2010). In a different study, introducing flexibility in single-domain bacterial FIC by mutations increased their modification by auto-AMPylation (Engel et al. 2012). Besides the FIC motif and the β-hairpin, IbpA interacts with Cdc42 by regions located outside the FIC domain, which explains the narrow specificity of the toxin for Rho GTPases. Structures of other FIC proteins, such as the bacterial toxin VopS (Luong et al. 2010), display non-FIC domains in the vicinity of their catalytic site, which may also contribute to the specific recognition of their protein target. Conversely, single-domain bacterial FIC proteins, such as Neisseria FIC (Engel et al. 2012), lack such additional elements, raising the possibility that some of them may recognize their substrates mostly through the FIC motif and the β-hairpin. Accordingly, the possibility that some of these proteins modify peptides rather than proteins cannot be excluded. Domains appended to the FIC domain can also have localization functions. Using small angle X-ray scattering, a method to observe the structure of proteins in solution, we found that the AnkX toxin has a horseshoe shape that allows it to bind simultaneously to membrane-attached Rab GTPases by its FIC domain and to the membrane by its C-terminal domain (our unpublished observations). These findings are consistent with the observation that AnkX binds to membranes in cells by elements located in its C-terminal half (Allgood et al. 2017). Structures of FIC proteins bound to co-factors or reaction products revealed both common and unique features of co-factor recognition and processing (reviewed in Garcia-Pino, Zenkin and Loris 2014; Roy and Cherfils 2015; Harms, Stanger and Dehio 2016; Table 1 and Fig. 2C). Notably, the structure of L. pneumophila AnkX in complex with intact CDP-choline, the co-factor for the phosphocholination of the small GTPase Rab1, showed that the FIC motif recognizes solely the diphosphate moiety of the co-factor, while the cytidine and choline moieties are recognized by elements located both upstream and downstream the FIC motif (Fig. 2D) (Campanacci et al. 2013). Structures of AMPylating FIC proteins also highlighted significant variations in the binding site of the adenine moiety of ATP, making the prediction of the nature of the co-factor difficult based on sequences only. In the future, a complementary strategy to discover co-factors of FIC proteins could be through structure-based docking, an approach that docks and ranks small molecules into crystal structures or structural models and is widely used for drug discovery (reviewed in Cheng et al. 2012). In the case of FIC proteins, all co-factors identified to date contain a diphosphate moiety, suggesting that this approach could be restricted to a diphosphate-containing compound library. Because FIC proteins appear to carry out reactions that stall major cellular processes, an important issue is whether they are regulated, and in which case, what are the modalities of this regulation. The structures of plasmid prophage Doc/Phd (Garcia-Pino et al. 2008; Arbing et al. 2010) and Bartonella VbhT-VbhA (Engel et al. 2012), and more recently E. coli FicT-FicA (Stanger et al. 2016a), TA modules provided a general principle for the inhibition of FIC toxins by their cognate antitoxins, in which the catalytic site of the toxin is obstructed by the antitoxin and dissociation of the antitoxin releases inhibition. In Bartonella VbhT-VbhA, inhibition is mediated by a glutamate from the VbhA antitoxin that prevents productive binding of the ATP co-substrate (Engel et al. 2012). Remarkably, a number of FIC proteins feature a structurally equivalent glutamate that wedges into the catalytic site, among which are single-domain bacterial FIC proteins and human FICD/HYPE (Engel et al. 2012) (Fig. 2E). Mutation of the glutamate unleashed a potent AMPylation activity in these proteins, leading to the hypothesis that this residue is autoinhibitory for the AMPylation reaction (Engel et al. 2012). Consistently, several structures of glutamate-containing FIC proteins highlighted that ATP binds in a non-canonical manner (Goepfert et al. 2013; Dedic et al. 2016) or in a manner in which only the ADP moiety is visible (Engel et al. 2012), suggesting that the glutamate conflicts with the ATP binding in a catalytically competent conformation. This hypothesis fueled intense research to discover signals or interactions able to trigger inhibition release. In the case of single-domain bacterial FIC proteins, recurrent observation of oligomers in the crystal led to propose that they are regulated by oligomerization, thereby enabling auto-AMPylation as a means to displace the autoinhibitory glutamate (Stanger et al. 2016b). In an unexpected twist, animal HYPE/FICD proteins were recently shown to encode de-AMPylation of the BIP chaperone in cells and in vitro, and this was dependent on this glutamate (Casey et al. 2017; Preissler et al. 2017). This discovery raises the intriguing possibility that bacterial FIC proteins that possess this glutamate may also encode dual AMPylation and de-AMPYlation activities. Another major question remaining is the nature of the cellular cues able to undergo variations that are large enough to promote the switch from the AMPylation reaction to the de-AMPylation reaction. CELL-FREE ASSAYS WITH RECOMBINANT PROTEINS: CHARACTERIZING THE ACTIVITY LANDSCAPE OF FIC PROTEINS Purified recombinant proteins have played a central role in understanding the biochemical activities of FIC proteins. Radiolabeling, in which the FIC protein is incubated with radioactive co-factors followed by autoradiography, is the standard assay to report on the PTM activity of FIC proteins. For example; 32P-α-ATP was used to establish the AMPylation activities of VopS (Yarbrough et al. 2009) and IbpA (Worby et al. 2009) toward Rho GTPases (Fig. 3A) and the AMPylation of BIP by FICD/HYPE (Ham et al. 2014; Preissler et al. 2015). Surprisingly, recombinant FICD/HYPE is also capable of AMPylating purified Rho GTPases in vitro (Worby et al. 2009), but a phenotype suggesting that it inactivates Rho GTPases in cells has not been described. Phosphorylation of the elongation factor Ef-Tu by recombinant Doc was demonstrated by its radiolabeling by 32P-γ-ATP but not 32P-α-ATP, using either purified Ef-Tu or E. coli extracts (Castro-Roa et al. 2013; Cruz et al. 2014). As an alternative to autoradiography, an anti-phosphocholine antibody was used to detect phosphocholination of Rab1 and Rab35 GTPases by AnkX (Mukherjee et al. 2011; Campanacci et al. 2013) and an anti AMP-threonine antibody was used to detect AMPylation of BIP by Drosophila Fic (the ortholog of HYPE/FICD) (Casey et al. 2017). Mass spectrometry is another approach that was used to identify or confirm candidate targets, using recombinant FIC proteins incubated with cell extracts treated with ATP. This approach established that recombinant bacterial VopS AMPylates human Rho GTPases in cell extracts (Yarbrough et al. 2009), and that the chaperone BIP is the substrate of human HYPE/FICD using a hyperactive mutant in which the inhibitory glutamate had been mutated (Ham et al. 2014). Figure 3. View largeDownload slide In vitro characterization of FIC proteins activity. (A) In vitro AMPylation of Rac-family GTPases by recombinant VopS. Recombinant human GTPases (Rac, Rho, Cdc42), active VopS from V. parahaemolyticus and inactive VopS harboring a mutation of the catalytic histidine were expressed in E. coli and purified. VopS constructs lacked the N-terminal secretion signal. GTPases loaded with GTP were then incubated with recombinant VopS proteins in the presence of radiolabeled ATP. Proteins were separated by SDS-PAGE and analyzed by autoradiography. Controls in which VopS was omitted were performed (lanes 1–3). Reproduced with permission from Yarbrough et al. (2009). (B) Kinetics of phosphocholination of human Rab1b by Legionella AnkX analyzed by intrinsic fluorescence measurement. Rab1b was loaded with GDP or with the non-hydrolysable GTP analog GppNHp. Addition of AnkX results in an exponential increase of tryptophane fluorescence (λexc = 297 nm and λem = 340 nm). The curves can be fitted by monoexponentials, giving apparent kcat/Km values of 9.8 × 104 M−1s−1 and 3.8 × 104 M−1s−1 respectively. Reproduced with permission from Goody et al. (2012). Figure 3. View largeDownload slide In vitro characterization of FIC proteins activity. (A) In vitro AMPylation of Rac-family GTPases by recombinant VopS. Recombinant human GTPases (Rac, Rho, Cdc42), active VopS from V. parahaemolyticus and inactive VopS harboring a mutation of the catalytic histidine were expressed in E. coli and purified. VopS constructs lacked the N-terminal secretion signal. GTPases loaded with GTP were then incubated with recombinant VopS proteins in the presence of radiolabeled ATP. Proteins were separated by SDS-PAGE and analyzed by autoradiography. Controls in which VopS was omitted were performed (lanes 1–3). Reproduced with permission from Yarbrough et al. (2009). (B) Kinetics of phosphocholination of human Rab1b by Legionella AnkX analyzed by intrinsic fluorescence measurement. Rab1b was loaded with GDP or with the non-hydrolysable GTP analog GppNHp. Addition of AnkX results in an exponential increase of tryptophane fluorescence (λexc = 297 nm and λem = 340 nm). The curves can be fitted by monoexponentials, giving apparent kcat/Km values of 9.8 × 104 M−1s−1 and 3.8 × 104 M−1s−1 respectively. Reproduced with permission from Goody et al. (2012). Autoradiography is also the standard assay to identify co-factors of FIC proteins. For example, it was used to show that both IbpA and VopS can use ATP and CTP to modify eukaryotic Cdc42 in vitro, while only VopS but not IbpA can use GTP (Mattoo et al. 2011). Another method to compare candidate co-factors is the thermal shift assay, which is based on the assumption that natural co-factors will bind more strongly to the FIC protein thereby increasing its thermal stability (Dedic et al. 2016). A shortcoming in characterizing the activity of FIC proteins is that the natural co-factor, the natural protein target or both are often not known. A convenient readout for activity has therefore been the analysis of automodifications by radiolabeled compounds (Kinch et al. 2009; Xiao et al. 2010; Mattoo et al. 2011; Engel et al. 2012; Feng et al. 2012; Goody et al. 2012; Mishra et al. 2012; Dedic et al. 2016; Preissler et al. 2017; our unpublished results). It should be emphasized that it has remained unclear whether automodifications are functionally important. In the case of single-domain FIC proteins, evidence based on mutants of the regulatory glutamate suggested that automodifications facilitate their conversion to an AMPylation-competent conformation (Engel et al. 2012; Stanger et al. 2016b). Likewise, mutation of auto-AMPylation sites in Pseudomonas Fic-1, which is related to the VbhT-VbhA TA module, impaired its ability to modify DNA gyrase (Lu et al. 2016). Conversely, study of autophosphoch olination of AnkX by time-resolved Fourier-transform infrared spectroscopy using photocleavable caged compound concluded that it is not relevant for catalysis (Gavriljuk et al. 2016) and some AnkX constructs with potent phosphocholination activity were not automodified (Campanacci et al. 2013). This body of observations points to potentially different usages of automodifications by FIC proteins. Gaining quantitative insight into the efficiency of FIC proteins is important to understand their regulation and this requires that their kinetics can be monitored accurately. Kinetics of phosphocholination of Rab GTPases by AnkX was measured using fluorescence, using either a coupled-enzyme assay, intrinsic tryptophan fluorescence or mant-guanine nucleotides fluorescence changes (Goody et al. 2012) (Fig. 3B). Likewise, the kinetics of de-AMPylation of the BIP chaperone by FICD was monitored by fluorescence anisotropy using an ATP derivative labeled by a fluorescent probe attached to the adenine base (Preissler et al. 2017). Semiquantitative kinetics can also be achieved under conditions where the reaction is sufficiently slow. For example, the kinetics of Cdc42 AMPylation by VopS was assayed by measuring the incorporation of radiolabeled AMP at several time points (Luong et al. 2010); similarly, the kinetics of Rab1 and Rab35 phosphocholination by AnkX and the stimulatory effect of membranes were determined using an anti-phosphocholine antibody (our unpublished results). In the future, a systematic determination of kinetics appears highly desirable to quantify the efficiency, specificity and regulation of FIC proteins. CELL-BASED AND IN VIVO ASSAYS: DISCOVERING THE FUNCTIONS OF FIC PROTEINS A primary use of cell-based assays is to test functional hypotheses regarding how FIC proteins manipulate cell signaling. In most cases, investigations of FIC functions have been carried out through ectopic expression in cells. For example, phenotypes resulting from ectopic expression of VopS (Yarbrough et al. 2009) and IbpA (Worby et al. 2009) in human cells guided the identification of their protein targets. Both toxins disrupted the actin cytoskeleton, a hallmark of the inactivation of Rho GTPases, which led to the discovery that they inactivate Rho GTPases by AMPylation (Fig. 4A). Deletion and ectopic expression of HYPE/FICD in human cells, which resulted in AMPylation patterns that could not be explained by a simple AMPylation function, was instrumental in guiding the discovery of irs de-AMPylation activity toward the BIP chaperone in the ER (Preissler et al. 2017) (Fig. 4B). In another study, heterologous expression of animal FICD in yeast induced an unexpected heat shock response; this guided the subsequent identification of a putative cytosolic activity of FICD in human cells which has yet to be characterized (Truttmann et al. 2017). For various bacterial FIC proteins that cannot be investigated directly in the species that express them, heterologous expression in E. coli has been used as a convenient surrogate to investigate their functions and identify candidate protein targets. For example, Bartonella VbhT was shown to modify E. coli DNA gyrase and DNA topoisomerase IV, suggesting that it also targets these enzymes in Bartonella (Harms et al. 2015). In another recent study, expression of Campylobacter fetus FIC proteins from in E. coli highlighted that the toxicity of Fic1 and its ability to induce a filamentous phenotype was neutralized by co-expression with Fic2, suggesting that Fic1 and Fic2 form a novel type of functional TA module (Sprenger et al. 2017). In these assays, it cannot be excluded that other substrates not present in E. coli remain to be discovered. Figure 4. View largeDownload slide In cellulo analysis of FIC proteins. (A) Ectopic expression of Vibrio VopS in human cells results in the collapse of the actin cytoskeleton, suggesting that VopS inactivates Rho family small GTPases. HeLa cells were transfected with an empty vector or with a recombinant vector driving heterologous expression of wild-type VopS or an inactive mutant (H348A) and then observed by confocal microscopy. Co-transfection with a vector expressing SFFV-eGFP fusion protein allows detection of transfected cells. Nuclei were identified by Hoechst stain. Scale bar, 10 μm. The abnormal rounded shape specifically observed upon expression of wild-type VopS is typical of actin cytoskeleton disorganization. Reproduced with permission from Yarbrough et al. (2009). (B) Cellular effect of FICD overexpression on de-AMPylation of the BIP chaperone. CHO-K1 cells, untransfected or transfected with increasing amounts of FICD expression vector, were exposed to cycloheximide (CHX, an inhibitor of protein synthesis) to promote BIP AMPylation. Proteins in lysates were then separated by native polyacrylamide gel electrophoresis and detected by western blot using an anti-FICD antibody. I, II and III indicate the different oligomerization states of BIP. The monomeric form (III) can be subdivided into unmodified (A) and AMPylated (B) forms. Increasing expression of FICD induces a decrease in BIP AMPylation, consistent with FICD acting as a de-AMPylating enzyme. Reproduced with permission from Preissler et al. (2017). Figure 4. View largeDownload slide In cellulo analysis of FIC proteins. (A) Ectopic expression of Vibrio VopS in human cells results in the collapse of the actin cytoskeleton, suggesting that VopS inactivates Rho family small GTPases. HeLa cells were transfected with an empty vector or with a recombinant vector driving heterologous expression of wild-type VopS or an inactive mutant (H348A) and then observed by confocal microscopy. Co-transfection with a vector expressing SFFV-eGFP fusion protein allows detection of transfected cells. Nuclei were identified by Hoechst stain. Scale bar, 10 μm. The abnormal rounded shape specifically observed upon expression of wild-type VopS is typical of actin cytoskeleton disorganization. Reproduced with permission from Yarbrough et al. (2009). (B) Cellular effect of FICD overexpression on de-AMPylation of the BIP chaperone. CHO-K1 cells, untransfected or transfected with increasing amounts of FICD expression vector, were exposed to cycloheximide (CHX, an inhibitor of protein synthesis) to promote BIP AMPylation. Proteins in lysates were then separated by native polyacrylamide gel electrophoresis and detected by western blot using an anti-FICD antibody. I, II and III indicate the different oligomerization states of BIP. The monomeric form (III) can be subdivided into unmodified (A) and AMPylated (B) forms. Increasing expression of FICD induces a decrease in BIP AMPylation, consistent with FICD acting as a de-AMPylating enzyme. Reproduced with permission from Preissler et al. (2017). Infection of cells by intracellular bacterial pathogens led to important functional findings. For example, reasoning that Legionella effectors interfere with the Rab GTPase machinery, Rab GTPases from macrophages infected by Legionella strains that do or do not secrete AnkX were analyzed by mass spectrometry, revealing that AnkX catalyzes their phosphocholination (Mukherjee et al. 2011). More recently, observation of ectopically expressed AnkX by immunogold transmission electron microscopy and confocal microscopy showed that it localizes on endosomal membranes of human cells, subsequently revealing that it disrupts endosomal recycling in the course of infection (Allgood et al. 2017). In another cell infection study, the Doc-Phd TA module of the pathogen Salmonella typhimurium was shown to be required for the formation of macrophage-induced non-replicating persisters (Helaine et al. 2014). Besides these infection studies, there has only been a few in vivo studies to date. In one study, deletion of Drosophila Fic (an HYPE ortholog) was shown to result in blindness in flies (Rahman et al. 2012). In another study, the Caenorhabditis elegans ortholog of FICD/HYPE was shown to AMPylate cytosolic proteins in vivo to modulate the immune response (Truttmann et al. 2016); paralleling these observations, expression in human cells of HYPE carrying an activating mutation resulted in the AMPylation of cytosolic chaperones (Truttmann et al. 2017). CONCLUDING REMARKS The field of FIC proteins continues to grow, with elegant experimental strategies revealing new surprises. New chemical tools are currently being devised, such as AMPylation-specific antibodies (Hao et al. 2011; Smit et al. 2011), camelid-derived nanobodies to modulate AMPylation (Truttmann et al. 2015), chemical reporters of AMPylation (Grammel et al. 2011; Truttmann et al. 2016) or of phosphocholination (Heller et al. 2015) that bind covalently to modified protein targets using click chemistry, or labeling with stable isosope-labeled ATP (Pieles et al. 2014). Some of these tools can be used in combination with mass spectrometry to determine the PTM profiles of FIC proteins in cells (Yu et al. 2014; Broncel et al. 2016). Together, this broad range of computational and experimental methods has allowed and cross-validated important discoveries and they should continue to fertilize each other to understand the biochemical activities, regulations and functions of FIC proteins. Likewise, discoveries in the bacterial and animal kingdoms should continue to contribute jointly to an integrated understanding of the functional landscape of this fascinating family. Ultimately, this knowledge may guide the discovery of novel therapeutic strategies in infections and proteostasis-related diseases. FUNDING This work was supported by grants from the Fondation pour la Recherche Médicale and from the Agence Nationale pour la Recherche to JC and by a grant from the DIM MALINF to SV. Conflict of interest. None declared. REFERENCES Allgood SC, Romero Duenas BP, Noll RR et al.   Legionella Effector AnkX Disrupts Host Cell Endocytic Recycling in a Phosphocholination-Dependent Manner, Front Cell Infect Microbiol  2017; 7: 397. Google Scholar CrossRef Search ADS PubMed  Arbing MA, Handelman SK, Kuzin AP et al.   Crystal structures of Phd-Doc, HigA, and YeeU establish multiple evolutionary links between microbial growth-regulating toxin-antitoxin systems, Structure  2010; 18: 996– 1010. Google Scholar CrossRef Search ADS PubMed  Broncel M, Serwa RA, Bunney TD et al.   Global Profiling of Huntingtin-associated protein E (HYPE)-Mediated AMPylation through a Chemical Proteomic Approach, Molecular & cellular proteomics: MCP  2016; 15: 715– 25. Google Scholar CrossRef Search ADS PubMed  Bunney TD, Cole AR, Broncel M et al.   Crystal structure of the human, FIC-domain containing protein HYPE and implications for its functions, Structure  2014; 22: 1831– 43. Google Scholar CrossRef Search ADS PubMed  Campanacci V, Mukherjee S, Roy CR et al.   Structure of the Legionella effector AnkX reveals the mechanism of phosphocholine transfer by the FIC domain, EMBO J  2013; 32: 1469– 77. Google Scholar CrossRef Search ADS PubMed  Casey AK, Orth K. Enzymes Involved in AMPylation and deAMPylation, Chem Rev  2017b; 118: 1199– 215. Google Scholar CrossRef Search ADS   Casey AK, Moehlman AT, Zhang J et al.   Fic-mediated deAMPylation is not dependent on homodimerization and rescues toxic AMPylation in flies, J Biol Chem  2017; 292: 21193– 204. Google Scholar CrossRef Search ADS PubMed  Castro-Roa D, Garcia-Pino A, De Gieter S et al.   The Fic protein Doc uses an inverted substrate to phosphorylate and inactivate EF-Tu, Nat Chem Biol  2013; 9: 811– 7. Google Scholar CrossRef Search ADS PubMed  Cheng T, Li Q, Zhou Z et al.   Structure-Based Virtual Screening for Drug Discovery: a Problem-Centric Review, AAPS J . 2012; 14: 133– 41. Google Scholar CrossRef Search ADS PubMed  Cruz JW, Woychik NA. Teaching Fido New ModiFICation Tricks, PLoS Pathog  2014; 10: e1004349. Google Scholar CrossRef Search ADS PubMed  Cruz JW, Rothenbacher FP, Maehigashi T et al.   Doc toxin is a kinase that inactivates elongation factor Tu, J Biol Chem  2014; 289: 7788– 98. Google Scholar CrossRef Search ADS PubMed  Dedic E, Alsarraf H, Welner DH et al.   A Novel Fic (Filamentation Induced by cAMP) Protein from Clostridium difficile Reveals an Inhibitory Motif-independent Adenylylation/AMPylation Mechanism, J Biol Chem  2016; 291: 13286– 300. Google Scholar CrossRef Search ADS PubMed  Desveaux D, Singer AU, Wu AJ et al.   Type III effector activation via nucleotide binding, phosphorylation, and host target interaction, PLoS Pathog  2007; 3: e48. Google Scholar CrossRef Search ADS PubMed  Engel P, Goepfert A, Stanger FV et al.   Adenylylation control by intra- or intermolecular active-site obstruction in Fic proteins, Nature  2012; 482: 107– 10. Google Scholar CrossRef Search ADS PubMed  Feng F, Yang F, Rong W et al.   A Xanthomonas uridine 5'-monophosphate transferase inhibits plant immune kinases, Nature  2012; 485: 114– 8. Google Scholar CrossRef Search ADS PubMed  Finn RD, Coggill P, Eberhardt RY et al.   The Pfam protein families database: towards a more sustainable future, Nucleic Acids Res  2016; 44: D279– 85. Google Scholar CrossRef Search ADS PubMed  Garcia-Pino A, Zenkin N, Loris R. The many faces of Fic: structural and functional aspects of Fic enzymes, Trends Biochem Sci  2014; 39: 121– 9. Google Scholar CrossRef Search ADS PubMed  Garcia-Pino A, Christensen-Dalsgaard M, Wyns L et al.   Doc of prophage P1 is inhibited by its antitoxin partner Phd through fold complementation, J Biol Chem  2008; 283: 30821– 7. Google Scholar CrossRef Search ADS PubMed  Garcia-Pino A, Balasubramanian S, Wyns L et al.   Allostery and intrinsic disorder mediate transcription regulation by conditional cooperativity, Cell  2010; 142: 101– 11. Google Scholar CrossRef Search ADS PubMed  Gavriljuk K, Schartner J, Seidel H et al.   Unraveling the Phosphocholination Mechanism of the Legionella pneumophila Enzyme AnkX, Biochemistry  2016; 55: 4375– 85. Google Scholar CrossRef Search ADS PubMed  Goepfert A, Stanger FV, Dehio C et al.   Conserved inhibitory mechanism and competent ATP binding mode for adenylyltransferases with fic fold, PLoS One  2013; 8: e64901. Google Scholar CrossRef Search ADS PubMed  Goody PR, Heller K, Oesterlin LK et al.   Reversible phosphocholination of Rab proteins by Legionella pneumophila effector proteins, EMBO J  2012; 31: 1774– 84. Google Scholar CrossRef Search ADS PubMed  Grammel M, Luong P, Orth K et al.   A chemical reporter for protein AMPylation, J Am Chem Soc  2011; 133: 17103– 5. Google Scholar CrossRef Search ADS PubMed  Ham H, Woolery AR, Tracy C et al.   Unfolded Protein Response-regulated Drosophila Fic (dFic) Protein Reversibly AMPylates BiP Chaperone during Endoplasmic Reticulum Homeostasis, J Biol Chem  2014; 289: 36059– 69. Google Scholar CrossRef Search ADS PubMed  Hao YH, Chuang T, Ball HL et al.   Characterization of a rabbit polyclonal antibody against threonine-AMPylation, J Biotechnol  2011; 151: 251– 4. Google Scholar CrossRef Search ADS PubMed  Harms A, Stanger FV, Dehio C. Biological Diversity and Molecular Plasticity of FIC Domain Proteins, Annu Rev Microbiol  2016; 70: 341– 60. Google Scholar CrossRef Search ADS PubMed  Harms A, Stanger FV, Scheu PD et al.   Adenylylation of Gyrase and Topo IV by FicT Toxins Disrupts Bacterial DNA Topology, Cell Rep  2015; 12: 1497– 507. Google Scholar CrossRef Search ADS PubMed  Helaine S, Cheverton AM, Watson KG et al.   Internalization of Salmonella by macrophages induces formation of nonreplicating persisters, Science  2014; 343: 204– 8. Google Scholar CrossRef Search ADS PubMed  Heller K, Ochtrop P, Albers MF et al.   Covalent Protein Labeling by Enzymatic Phosphocholination, Angew Chem Int Ed Engl  2015; 54: 10327– 30. Google Scholar CrossRef Search ADS PubMed  Kawamukai M, Matsuda H, Fujii W et al.   Nucleotide sequences of fic and fic-1 genes involved in cell filamentation induced by cyclic AMP in Escherichia coli, J Bacteriol  1989; 171: 4525– 9. Google Scholar CrossRef Search ADS PubMed  Kawamukai M, Matsuda H, Fujii W et al.   Cloning of the fic-1 gene involved in cell filamentation induced by cyclic AMP and construction of a delta fic Escherichia coli strain, J Bacteriol  1988; 170: 3864– 9. Google Scholar CrossRef Search ADS PubMed  Khater S, Mohanty D. In silico identification of AMPylating enzymes and study of their divergent evolution, Sci Rep  2015a; 5: 10804. Google Scholar CrossRef Search ADS   Khater S, Mohanty D. Deciphering the Molecular Basis of Functional Divergence in AMPylating Enzymes by Molecular Dynamics Simulations and Structure Guided Phylogeny, Biochemistry  2015b; 54: 5209– 24. Google Scholar CrossRef Search ADS   Kinch LN, Yarbrough ML, Orth K et al.   Fido, a novel AMPylation domain common to fic, doc, and AvrB, PLoS One  2009; 4: e5818. Google Scholar CrossRef Search ADS PubMed  Lee CC, Wood MD, Ng K et al.   Crystal structure of the type III effector AvrB from Pseudomonas syringae, Structure  2004; 12: 487– 94. Google Scholar CrossRef Search ADS PubMed  Lu C, Nakayasu ES, Zhang LQ et al.   Identification of Fic-1 as an enzyme that inhibits bacterial DNA replication by AMPylating GyrB, promoting filament formation, Sci Signal  2016; 9: ra11. Google Scholar CrossRef Search ADS PubMed  Luong P, Kinch LN, Brautigam CA et al.   Kinetic and structural insights into the mechanism of AMPylation by VopS Fic domain, J Biol Chem  2010; 285: 20155– 63. Google Scholar CrossRef Search ADS PubMed  Mattoo S, Durrant E, Chen MJ et al.   Comparative analysis of Histophilus somni immunoglobulin-binding protein A (IbpA) with other fic domain-containing enzymes reveals differences in substrate and nucleotide specificities, J Biol Chem  2011; 286: 32834– 42. Google Scholar CrossRef Search ADS PubMed  Mishra S, Bhagavat R, Chandra N et al.   Cloning, expression, purification, and biochemical characterisation of the FIC motif containing protein of Mycobacterium tuberculosis, Protein Expr Purif  2012; 86: 58– 67. Google Scholar CrossRef Search ADS PubMed  Mukherjee S, Liu X, Arasaki K et al.   Modulation of Rab GTPase function by a protein phosphocholine transferase, Nature  2011; 477: 103– 6. Google Scholar CrossRef Search ADS PubMed  Palanivelu DV, Goepfert A, Meury M et al.   Fic domain-catalyzed adenylylation: insight provided by the structural analysis of the type IV secretion system effector BepA, Protein Sci  2011; 20: 492– 9. Google Scholar CrossRef Search ADS PubMed  Pieles K, Glatter T, Harms A et al.   An experimental strategy for the identification of AMPylation targets from complex protein samples, Proteomics  2014; 14: 1048– 52. Google Scholar CrossRef Search ADS PubMed  Preissler S, Rato C, Perera L et al.   FICD acts bifunctionally to AMPylate and de-AMPylate the endoplasmic reticulum chaperone BiP, Nature structural & molecular biology  2017; 24: 23– 29. Google Scholar CrossRef Search ADS PubMed  Preissler S, Rato C, Chen R et al.   AMPylation matches BiP activity to client protein load in the endoplasmic reticulum, eLife  2015; 4: e12621. Google Scholar PubMed  Rahman M, Ham H, Liu X et al.   Visual neurotransmission in Drosophila requires expression of Fic in glial capitate projections, Nat Neurosci  2012; 15: 871– 5. Google Scholar CrossRef Search ADS PubMed  Roy CR, Mukherjee S. Bacterial FIC Proteins AMP Up Infection, Sci Signal  2009; 2: pe14. Google Scholar CrossRef Search ADS PubMed  Roy CR, Cherfils J Structure and function of Fic proteins, Nat Rev Microbiol  2015; 13: 631– 40. Google Scholar CrossRef Search ADS PubMed  Sanyal A, Chen AJ, Nakayasu ES et al.   A Novel Link between Fic (Filamentation Induced by cAMP)-mediated Adenylylation/AMPylation and the Unfolded Protein Response, J Biol Chem  2015; 290: 8482– 99. Google Scholar CrossRef Search ADS PubMed  Smit C, Blumer J, Eerland MF et al.   Efficient synthesis and applications of peptides containing adenylylated tyrosine residues, Angew Chem Int Ed Engl  2011; 50: 9200– 4. Google Scholar CrossRef Search ADS PubMed  Sprenger H, Kienesberger S, Pertschy B et al.   Fic Proteins of Campylobacter fetus subsp. venerealis Form a Network of Functional Toxin-Antitoxin Systems, Front Microbiol  2017; 8: 1965. Google Scholar CrossRef Search ADS PubMed  Stanger FV, Harms A, Dehio C et al.   Crystal Structure of the Escherichia coli Fic Toxin-Like Protein in Complex with Its Cognate Antitoxin, PLoS One  2016a; 11: e0163654. Google Scholar CrossRef Search ADS   Stanger FV, Burmann BM, Harms A et al.   Intrinsic regulation of FIC-domain AMP-transferases by oligomerization and automodification, Proc Natl Acad Sci U S A  2016b; 113: E529– 37. Google Scholar CrossRef Search ADS   Truttmann MC, Ploegh HL. rAMPing Up Stress Signaling: Protein AMPylation in Metazoans, Trends Cell Biol  2017; 27: 608– 20. Google Scholar CrossRef Search ADS PubMed  Truttmann MC, Wu Q, Stiegeler S et al.   HypE-specific nanobodies as tools to modulate HypE-mediated target AMPylation, J Biol Chem  2015; 290: 9087– 100. Google Scholar CrossRef Search ADS PubMed  Truttmann MC, Cruz VE, Guo X et al.   The Caenorhabditis elegans Protein FIC-1 Is an AMPylase That Covalently Modifies Heat-Shock 70 Family Proteins, Translation Elongation Factors and Histones, PLoS Genet  2016; 12: e1006023. Google Scholar CrossRef Search ADS PubMed  Truttmann MC, Zheng X, Hanke L et al.   Unrestrained AMPylation targets cytosolic chaperones and activates the heat shock response, Proc Natl Acad Sci U S A  2017; 114: E152– E60. Google Scholar CrossRef Search ADS PubMed  Worby CA, Mattoo S, Kruger RP et al.   The fic domain: regulation of cell signaling by adenylylation, Mol Cell  2009; 34: 93– 103. Google Scholar CrossRef Search ADS PubMed  Xiao J, Worby CA, Mattoo S et al.   Structural basis of Fic-mediated adenylylation, Nat Struct Mol Biol  2010; 17: 1004– 10. Google Scholar CrossRef Search ADS PubMed  Yarbrough ML, Li Y, Kinch LN et al.   AMPylation of Rho GTPases by Vibrio VopS disrupts effector binding and downstream signaling, Science  2009; 323: 269– 72. Google Scholar CrossRef Search ADS PubMed  Yu X, Woolery AR, Luong P et al.   Copper-catalyzed azide-alkyne cycloaddition (click chemistry)-based detection of global pathogen-host AMPylation on self-assembled human protein microarrays, Mol Cell Proteomics  2014; 13: 3164– 76. Google Scholar CrossRef Search ADS PubMed  © FEMS 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com

Journal

Pathogens and DiseaseOxford University Press

Published: Mar 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off