Structure-function relationship of a novel fucoside-binding fruiting body lectin from Coprinopsis cinerea exhibiting nematotoxic activity.

Lectins are non-immunoglobulin-type proteins that bind to specific carbohydrate epitopes and play important roles in intra- and inter-organismic interactions. Here, we describe a novel fucose-specific lectin, termed CML1, which we identified from fruiting body extracts of Coprinopsis cinerea. For further characterization, the coding sequence for CML1 was cloned and heterologously expressed in Escherichia coli. Feeding of CML1-producing bacteria inhibited larval development of the bacterivorous nematode Caenorhabditis tropicalis, but not of C. elegans. The crystal structure of the recombinant protein in its apo-form and in complex with H type I or Lewis A blood group antigens was determined by X-ray crystallography. The protein folds as a sandwich of 2 antiparallel ß-sheets and forms hexamers resulting from a trimer of dimers. The hexameric arrangement was confirmed by small-angle X-ray scattering (SAXS). One carbohydrate-binding site per protomer was found at the dimer interface with both protomers contributing to ligand binding, resulting in a hexavalent lectin. In terms of lectin activity of recombinant CML1, substitution of the carbohydrate-interacting residues His54, Asn55, Trp94, and Arg114 by Ala abolished carbohydrate-binding and nematotoxicity. Although no similarities to any characterized lectin were found, sequence alignments identified many non-characterized agaricomycete proteins. These results suggest that CML1 is the founding member of a novel family of fucoside-binding lectins involved in the defense of agaricomycete fruiting bodies against predation by fungivorous nematodes.

Quaternary dynamics of the SecA motor drive translocase catalysis.

Most secretory preproteins exit bacterial cells through the protein translocase, comprising the SecYEG channel and the dimeric peripheral ATPase motor SecA. Energetic coupling to work remains elusive. We now demonstrate that translocation is driven by unusually dynamic quaternary changes in SecA. The dimer occupies several successive states with distinct protomer arrangements. SecA docks on SecYEG as a dimer and becomes functionally asymmetric. Docking occurs via only one protomer. The second protomer allosterically regulates downstream steps. Binding of one preprotein signal peptide to the SecYEG-docked SecA protomer elongates the SecA dimer and triggers the translocase holoenzyme to obtain a lower activation energy conformation. ATP hydrolysis monomerizes the triggered SecA dimer, causing mature chain trapping and processive translocation. This is a unique example of one protein exploiting quaternary dynamics to become a substrate receptor, a "loading clamp," and a "processive motor." This mechanism has widespread implications on protein translocases, chaperones, and motors.

Alternative folding to a monomer or homopolymer is a common feature of the type 1 pilus subunit FimA from enteroinvasive bacteria.

Adhesive type 1 pili from enteroinvasive, Gram-negative bacteria mediate attachment to host cells. Up to 3000 copies of the main pilus subunit, FimA, assemble into the filamentous, helical quaternary structure of the pilus rod via a mechanism termed donor-strand complementation, in which the N-terminal extension of each subunit, the donor strand, is inserted into the incomplete immunoglobulin-like fold of the preceding FimA subunit. For FimA from Escherichia coli, it has been previously shown that the protein can also adopt a monomeric, self-complemented conformation in which the donor strand is inserted intramolecularly in the opposite orientation relative to that observed for FimA polymers. Notably, soluble FimA monomers can act as apoptosis inhibitors in epithelial cells after uptake of type 1-piliated pathogens. Here, we show that the FimA orthologues from Escherichia coli, Shigella flexneri, and Salmonella enterica can all fold to form self-complemented monomers. We solved X-ray structures of all three FimA monomers at 0.89-1.69 Å resolutions, revealing identical, intramolecular donor-strand complementation mechanisms. Our results also showed that the pseudo-palindromic sequences of the donor strands in all FimA proteins permit their alternative folding possibilities. All FimA monomers proved to be 50-60 kJ/mol less stable against unfolding than their pilus rod-like counterparts (which exhibited very high energy barriers of unfolding and refolding). We conclude that the ability of FimA to adopt an alternative, monomeric state with anti-apoptotic activity is a general feature of FimA proteins of type 1-piliated bacteria.

Analyzing the symmetrical arrangement of structural repeats in proteins with CE-Symm.

Many proteins fold into highly regular and repetitive three dimensional structures. The analysis of structural patterns and repeated elements is fundamental to understand protein function and evolution. We present recent improvements to the CE-Symm tool for systematically detecting and analyzing the internal symmetry and structural repeats in proteins. In addition to the accurate detection of internal symmetry, the tool is now capable of i) reporting the type of symmetry, ii) identifying the smallest repeating unit, iii) describing the arrangement of repeats with transformation operations and symmetry axes, and iv) comparing the similarity of all the internal repeats at the residue level. CE-Symm 2.0 helps the user investigate proteins with a robust and intuitive sequence-to-structure analysis, with many applications in protein classification, functional annotation and evolutionary studies. We describe the algorithmic extensions of the method and demonstrate its applications to the study of interesting cases of protein evolution.

Structural basis for recognition of synaptic vesicle protein 2C by botulinum neurotoxin A.

Botulinum neurotoxin A (BoNT/A) belongs to the most dangerous class of bioweapons. Despite this, BoNT/A is used to treat a wide range of common medical conditions such as migraines and a variety of ocular motility and movement disorders. BoNT/A is probably best known for its use as an antiwrinkle agent in cosmetic applications (including Botox and Dysport). BoNT/A application causes long-lasting flaccid paralysis of muscles through inhibiting the release of the neurotransmitter acetylcholine by cleaving synaptosomal-associated protein 25 (SNAP-25) within presynaptic nerve terminals. Two types of BoNT/A receptor have been identified, both of which are required for BoNT/A toxicity and are therefore likely to cooperate with each other: gangliosides and members of the synaptic vesicle glycoprotein 2 (SV2) family, which are putative transporter proteins that are predicted to have 12 transmembrane domains, associate with the receptor-binding domain of the toxin. Recently, fibroblast growth factor receptor 3 (FGFR3) has also been reported to be a potential BoNT/A receptor. In SV2 proteins, the BoNT/A-binding site has been mapped to the luminal domain, but the molecular details of the interaction between BoNT/A and SV2 are unknown. Here we determined the high-resolution crystal structure of the BoNT/A receptor-binding domain (BoNT/A-RBD) in complex with the SV2C luminal domain (SV2C-LD). SV2C-LD consists of a right-handed, quadrilateral ß-helix that associates with BoNT/A-RBD mainly through backbone-to-backbone interactions at open ß-strand edges, in a manner that resembles the inter-strand interactions in amyloid structures. Competition experiments identified a peptide that inhibits the formation of the complex. Our findings provide a strong platform for the development of novel antitoxin agents and for the rational design of BoNT/A variants with improved therapeutic properties.

Automated evaluation of quaternary structures from protein crystals.

A correct assessment of the quaternary structure of proteins is a fundamental prerequisite to understanding their function, physico-chemical properties and mode of interaction with other proteins. Currently about 90% of structures in the Protein Data Bank are crystal structures, in which the correct quaternary structure is embedded in the crystal lattice among a number of crystal contacts. Computational methods are required to 1) classify all protein-protein contacts in crystal lattices as biologically relevant or crystal contacts and 2) provide an assessment of how the biologically relevant interfaces combine into a biological assembly. In our previous work we addressed the first problem with our EPPIC (Evolutionary Protein Protein Interface Classifier) method. Here, we present our solution to the second problem with a new method that combines the interface classification results with symmetry and topology considerations. The new algorithm enumerates all possible valid assemblies within the crystal using a graph representation of the lattice and predicts the most probable biological unit based on the pairwise interface scoring. Our method achieves 85% precision (ranging from 76% to 90% for different oligomeric types) on a new dataset of 1,481 biological assemblies with consensus of PDB annotations. Although almost the same precision is achieved by PISA, currently the most popular quaternary structure assignment method, we show that, due to the fundamentally different approach to the problem, the two methods are complementary and could be combined to improve biological assembly assignments. The software for the automatic assessment of protein assemblies (EPPIC version 3) has been made available through a web server at http://www.eppic-web.org.

Assessment of protein assembly prediction in CASP12.

We present the results of the first independent assessment of protein assemblies in CASP. A total of 1624 oligomeric models were submitted by 108 predictor groups for the 30 oligomeric targets in the CASP12 edition. We evaluated the accuracy of oligomeric predictions by comparison to their reference structures at the interface patch and residue contact levels. We find that interface patches are more reliably predicted than the specific residue contacts. Whereas none of the 15 hard oligomeric targets have successful predictions for the residue contacts at the interface, six have models with resemblance in the interface patch. Successful predictions of interface patch and contacts exist for all targets suitable for homology modeling, with at least one group improving over the best available template for each target. However, the participation in protein assembly prediction is low and uneven. Three human groups are closely ranked at the top by overall performance, but a server outperforms all other predictors for targets suitable for homology modeling. The state of the art of protein assembly prediction methods is in development and has apparent room for improvement, especially for assemblies without templates.

Biological and functional relevance of CASP predictions.

Our goal is to answer the question: compared with experimental structures, how useful are predicted models for functional annotation? We assessed the functional utility of predicted models by comparing the performances of a suite of methods for functional characterization on the predictions and the experimental structures. We identified 28 sites in 25 protein targets to perform functional assessment. These 28 sites included nine sites with known ligand binding (holo-sites), nine sites that are expected or suggested by experimental authors for small molecule binding (apo-sites), and Ten sites containing important motifs, loops, or key residues with important disease-associated mutations. We evaluated the utility of the predictions by comparing their microenvironments to the experimental structures. Overall structural quality correlates with functional utility. However, the best-ranked predictions (global) may not have the best functional quality (local). Our assessment provides an ability to discriminate between predictions with high structural quality. When assessing ligand-binding sites, most prediction methods have higher performance on apo-sites than holo-sites. Some servers show consistently high performance for certain types of functional sites. Finally, many functional sites are associated with protein-protein interaction. We also analyzed biologically relevant features from the protein assemblies of two targets where the active site spanned the protein-protein interface. For the assembly targets, we find that the features in the models are mainly determined by the choice of template.

Understanding the fabric of protein crystals: computational classification of biological interfaces and crystal contacts.

Modern structural biology still draws the vast majority of information from crystallography, a technique where the objects being investigated are embedded in a crystal lattice. Given the complexity and variety of those objects, it becomes fundamental to computationally assess which of the interfaces in the lattice are biologically relevant and which are simply crystal contacts. Since the mid-1990s, several approaches have been applied to obtain high-accuracy classification of crystal contacts and biological protein-protein interfaces. This review provides an overview of the concepts and main approaches to protein interface classification: thermodynamic estimation of interface stability, evolutionary approaches based on conservation of interface residues, and co-occurrence of the interface across different crystal forms. Among the three categories, evolutionary approaches offer the strongest promise for improvement, thanks to the incessant growth in sequence knowledge. Importantly, protein interface classification algorithms can also be used on multimeric structures obtained using other high-resolution techniques or for protein assembly design or validation purposes. A key issue linked to protein interface classification is the identification of the biological assembly of a crystal structure and the analysis of its symmetry. Here, we highlight the most important concepts and problems to be overcome in assembly prediction. Over the next few years, tools and concepts of interface classification will probably become more frequently used and integrated in several areas of structural biology and structural bioinformatics. Among the main challenges for the future are better addressing of weak interfaces and the application of interface classification concepts to prediction problems like protein-protein docking.

The use of ene adducts to study and engineer enoyl-thioester reductases.

An improved understanding of enzymes' catalytic proficiency and stereoselectivity would further enable applications in chemistry, biocatalysis and industrial biotechnology. We use a chemical probe to dissect individual catalytic steps of enoyl-thioester reductases (Etrs), validating an active site tyrosine as the cryptic proton donor and explaining how it had eluded definitive identification. This information enabled the rational redesign of Etr, yielding mutants that create products with inverted stereochemistry at wild type-like turnover frequency.

Functional roles of the hexamer organization of plant glutamate decarboxylase.

Glutamate decarboxylase (GAD) is a pyridoxal 5'-phosphate (PLP)-dependent enzyme that catalyzes the α-decarboxylation of glutamate to γ-aminobutyrate. A unique feature of plant GAD is the presence of a calmodulin (CaM)-binding domain at its C-terminus. In plants, transient elevation of cytosolic Ca²âº in response to different types of stress is responsible for GAD activation via CaM. The crystal structure of GAD isoform 1 from Arabidopsis thaliana (AtGAD1) shows that the enzyme is a hexamer composed of a trimer of dimers. Herein, we show that in solution AtGAD1 is in a dimer-hexamer equilibrium and estimate the dissociation constant (Kd) for the hexamer under different conditions. The association of dimers into hexamers is promoted by several conditions, including high protein concentrations and low pH. Notably, binding of Ca²âº/CaM1 abolishes the dissociation of the AtGAD1 oligomer. The AtGAD1 N-terminal domain is critical for maintaining the oligomeric state as removal of the first 24 N-terminal residues dramatically affects oligomerization by producing a dimeric enzyme. The deleted mutant retains decarboxylase activity, highlighting the dimeric nature of the basic structural unit of AtGAD1. Site-directed mutagenesis identified Arg24 in the N-terminal domain as a key residue since its mutation to Ala prevents hexamer formation in solution. Both dimeric mutant enzymes form a stable hexamer in the presence of Ca²âº/CaM1. Our data clearly reveal that the oligomeric state of AtGAD1 is highly responsive to a number of experimental parameters and may have functional relevance in vivo in the light of the biphasic regulation of AtGAD1 activity by pH and Ca²âº/CaM1 in plant cells. This article is part of a special issue titled "Cofactor-Dependent Proteins: Evolution, Chemical Diversity and Bio-applications."

Accelerating the Association of the Most Stable Protein-Ligand Complex by More than Two Orders of Magnitude.

The complex between the bacterial type 1 pilus subunit FimG and the peptide corresponding to the N-terminal extension (termed donor strand, Ds) of the partner subunit FimF (DsF) shows the strongest reported noncovalent molecular interaction, with a dissociation constant (KD ) of 1.5×10(-20) m. However, the complex only exhibits a slow association rate of 330 m(-1) s(-1) that limits technical applications, such as its use in affinity purification. Herein, a structure-based approach was used to design pairs of FimGt (a FimG variant lacking its own N-terminal extension) and DsF variants with enhanced electrostatic surface complementarity. Association of the best mutant FimGt/DsF pairs was accelerated by more than two orders of magnitude, while the dissociation rates and 3D structures of the improved complexes remained essentially unperturbed. A KD  value of 8.8×10(-22) m was obtained for the best mutant complex, which is the lowest value reported to date for a protein/ligand complex.

How periplasmic thioredoxin TlpA reduces bacterial copper chaperone ScoI and cytochrome oxidase subunit II (CoxB) prior to metallation.

Two critical cysteine residues in the copper-A site (Cu(A)) on subunit II (CoxB) of bacterial cytochrome c oxidase lie on the periplasmic side of the cytoplasmic membrane. As the periplasm is an oxidizing environment as compared with the reducing cytoplasm, the prediction was that a disulfide bond formed between these cysteines must be eliminated by reduction prior to copper insertion. We show here that a periplasmic thioredoxin (TlpA) acts as a specific reductant not only for the Cu(2+) transfer chaperone ScoI but also for CoxB. The dual role of TlpA was documented best with high-resolution crystal structures of the kinetically trapped TlpA-ScoI and TlpA-CoxB mixed disulfide intermediates. They uncovered surprisingly disparate contact sites on TlpA for each of the two protein substrates. The equilibrium of CoxB reduction by TlpA revealed a thermodynamically favorable reaction, with a less negative redox potential of CoxB (E'0 = -231 mV) as compared with that of TlpA (E'0 = -256 mV). The reduction of CoxB by TlpA via disulfide exchange proved to be very fast, with a rate constant of 8.4 × 10(4) M(-1) s(-1) that is similar to that found previously for ScoI reduction. Hence, TlpA is a physiologically relevant reductase for both ScoI and CoxB. Although the requirement of ScoI for assembly of the Cu(A)-CoxB complex may be bypassed in vivo by high environmental Cu(2+) concentrations, TlpA is essential in this process because only reduced CoxB can bind copper ions.

A type IV translocated Legionella cysteine phytase counteracts intracellular growth restriction by phytate.

The causative agent of Legionnaires' pneumonia, Legionella pneumophila, colonizes diverse environmental niches, including biofilms, plant material, and protozoa. In these habitats, myo-inositol hexakisphosphate (phytate) is prevalent and used as a phosphate storage compound or as a siderophore. L. pneumophila replicates in protozoa and mammalian phagocytes within a unique "Legionella-containing vacuole." The bacteria govern host cell interactions through the Icm/Dot type IV secretion system (T4SS) and ∼300 different "effector" proteins. Here we characterize a hitherto unrecognized Icm/Dot substrate, LppA, as a phytate phosphatase (phytase). Phytase activity of recombinant LppA required catalytically essential cysteine (Cys(231)) and arginine (Arg(237)) residues. The structure of LppA at 1.4 Å resolution revealed a mainly α-helical globular protein stabilized by four antiparallel ß-sheets that binds two phosphate moieties. The phosphates localize to a P-loop active site characteristic of dual specificity phosphatases or to a non-catalytic site, respectively. Phytate reversibly abolished growth of L. pneumophila in broth, and growth inhibition was relieved by overproduction of LppA or by metal ion titration. L. pneumophila lacking lppA replicated less efficiently in phytate-loaded Acanthamoeba castellanii or Dictyostelium discoideum, and the intracellular growth defect was complemented by the phytase gene. These findings identify the chelator phytate as an intracellular bacteriostatic component of cell-autonomous host immunity and reveal a T4SS-translocated L. pneumophila phytase that counteracts intracellular bacterial growth restriction by phytate. Thus, bacterial phytases might represent therapeutic targets to combat intracellular pathogens.

Conformational transitions driven by pyridoxal-5'-phosphate uptake in the psychrophilic serine hydroxymethyltransferase from Psychromonas ingrahamii.

Serine hydroxymethyltransferase (SHMT) is a pyridoxal-5'-phosphate (PLP)-dependent enzyme belonging to the fold type I superfamily, which catalyzes in vivo the reversible conversion of l-serine and tetrahydropteroylglutamate (H4PteGlu) to glycine and 5,10-methylenetetrahydropteroylglutamate (5,10-CH2-H4PteGlu). The SHMT from the psychrophilic bacterium Psychromonas ingrahamii (piSHMT) had been recently purified and characterized. This enzyme was shown to display catalytic and stability properties typical of psychrophilic enzymes, namely high catalytic activity at low temperature and thermolability. To gain deeper insights into the structure-function relationship of piSHMT, the three-dimensional structure of its apo form was determined by X-ray crystallography. Homology modeling techniques were applied to build a model of the piSHMT holo form. Comparison of the two forms unraveled the conformation modifications that take place when the apo enzyme binds its cofactor. Our results show that the apo form is in an "open" conformation and possesses four (or five, in chain A) disordered loops whose electron density is not visible by X-ray crystallography. These loops contain residues that interact with the PLP cofactor and three of them are localized in the major domain that, along with the small domain, constitutes the single subunit of the SHMT homodimer. Cofactor binding triggers a rearrangement of the small domain that moves toward the large domain and screens the PLP binding site at the solvent side. Comparison to the mesophilic apo SHMT from Salmonella typhimurium suggests that the backbone conformational changes are wider in psychrophilic SHMT.

A PDB-wide, evolution-based assessment of protein-protein interfaces.

BACKGROUND: Thanks to the growth in sequence and structure databases, more than 50 million sequences are now available in UniProt and 100,000 structures in the PDB. Rich information about protein-protein interfaces can be obtained by a comprehensive study of protein contacts in the PDB, their sequence conservation and geometric features. RESULTS: An automated computational pipeline was developed to run our Evolutionary Protein-Protein Interface Classifier (EPPIC) software on the entire PDB and store the results in a relational database, currently containing > 800,000 interfaces. This allows the analysis of interface data on a PDB-wide scale. Two large benchmark datasets of biological interfaces and crystal contacts, each containing about 3000 entries, were automatically generated based on criteria thought to be strong indicators of interface type. The BioMany set of biological interfaces includes NMR dimers solved as crystal structures and interfaces that are preserved across diverse crystal forms, as catalogued by the Protein Common Interface Database (ProtCID) from Xu and Dunbrack. The second dataset, XtalMany, is derived from interfaces that would lead to infinite assemblies and are therefore crystal contacts. BioMany and XtalMany were used to benchmark the EPPIC approach. The performance of EPPIC was also compared to classifications from the Protein Interfaces, Surfaces, and Assemblies (PISA) program on a PDB-wide scale, finding that the two approaches give the same call in about 88% of PDB interfaces. By comparing our safest predictions to the PDB author annotations, we provide a lower-bound estimate of the error rate of biological unit annotations in the PDB. Additionally, we developed a PyMOL plugin for direct download and easy visualization of EPPIC interfaces for any PDB entry. Both the datasets and the PyMOL plugin are available at http://www.eppic-web.org/ewui/\#downloads. CONCLUSIONS: Our computational pipeline allows us to analyze protein-protein contacts and their sequence conservation across the entire PDB. Two new benchmark datasets are provided, which are over an order of magnitude larger than existing manually curated ones. These tools enable the comprehensive study of several aspects of protein-protein contacts in the PDB and represent a basis for future, even larger scale studies of protein-protein interactions.

Quality control of disulfide bond formation in pilus subunits by the chaperone FimC.

Type 1 pili from uropathogenic Escherichia coli are filamentous, noncovalent protein complexes mediating bacterial adhesion to the host tissue. All structural pilus subunits are homologous proteins sharing an invariant disulfide bridge. Here we show that disulfide bond formation in the unfolded subunits, catalyzed by the periplasmic oxidoreductase DsbA, is required for subunit recognition by the assembly chaperone FimC and for FimC-catalyzed subunit folding. FimC thus guarantees quantitative disulfide bond formation in each of the up to 3,000 subunits of the pilus. The X-ray structure of the complex between FimC and the main pilus subunit FimA and the kinetics of FimC-catalyzed FimA folding indicate that FimC accelerates folding of pilus subunits by lowering their topological complexity. The kinetic data, together with the measured in vivo concentrations of DsbA and FimC, predict an in vivo half-life of 2 s for oxidative folding of FimA in the periplasm.

(4R)- and (4S)-fluoroproline in the conserved cis-prolyl peptide bond of the thioredoxin fold: tertiary structure context dictates ring puckering.

Fine-tuning protein stability: The non-natural amino acids (2S,4R)- and (2S,4S)-fluoroproline modulate protein stability by biasing the proline ring pucker and the cis/trans equilibrium of prolyl peptide bonds. We incorporated both fluoroproline stereoisomers at the invariant cis-proline residue of the thioredoxin fold. The results show that tertiary structure context overrules the conformational preferences of fluoroprolines.

An analysis of oligomerization interfaces in transmembrane proteins.

BACKGROUND: The amount of transmembrane protein (TM) structures solved to date is now large enough to attempt large scale analyses. In particular, extensive studies of oligomeric interfaces in the transmembrane region are now possible. RESULTS: We have compiled the first fully comprehensive set of validated transmembrane protein interfaces in order to study their features and assess what differentiates them from their soluble counterparts. CONCLUSIONS: The general features of TM interfaces do not differ much from those of soluble proteins: they are large, tightly packed and possess many interface core residues. In our set, membrane lipids were not found to significantly mediate protein-protein interfaces. Although no G protein-coupled receptor (GPCR) was included in the validated set, we analyzed the crystallographic dimerization interfaces proposed in the literature. We found that the putative dimer interfaces proposed for class A GPCRs do not show the usual patterns of stable biological interfaces, neither in terms of evolution nor of packing, thus they likely correspond to crystal interfaces. We cannot however rule out the possibility that they constitute transient or weak interfaces. In contrast we do observe a clear signature of biological interface for the proposed dimer of the class F human Smoothened receptor.

Stochastic chain termination in bacterial pilus assembly.

Adhesive type 1 pili from uropathogenic Escherichia coli strains are filamentous, supramolecular protein complexes consisting of a short tip fibrillum and a long, helical rod formed by up to several thousand copies of the major pilus subunit FimA. Here, we reconstituted the entire type 1 pilus rod assembly reaction in vitro, using all constituent protein subunits in the presence of the assembly platform FimD, and identified the so-far uncharacterized subunit FimI as an irreversible assembly terminator. We provide a complete, quantitative model of pilus rod assembly kinetics based on the measured rate constants of FimD-catalyzed subunit incorporation. The model reliably predicts the length distribution of assembled pilus rods as a function of the ratio between FimI and the main pilus subunit FimA and is fully consistent with the length distribution of membrane-anchored pili assembled in vivo. The results show that the natural length distribution of adhesive pili formed via the chaperone-usher pathway results from a stochastic chain termination reaction. In addition, we demonstrate that FimI contributes to anchoring the pilus to the outer membrane and report the crystal structures of (i) FimI in complex with the assembly chaperone FimC, (ii) the FimI-FimC complex bound to the N-terminal domain of FimD, and (iii) a ternary complex between FimI, FimA and FimC that provides structural insights on pilus assembly termination and pilus anchoring by FimI.