Molecular Characterization of theEscherichia coliFimH Adhesin

Bacterial adherence mediated by fimbriae is an essential prerequisite for colonization of the urinary tract. Uropathogenic Escherichia coli express a number of different adhesive organelles, including P, type 1, S, and F1C fimbriae [1]. Type 1, or mannose-sensitive, fimbriae are produced by >80% of all uropathogenic E. coli. It is now well established that the expression of type 1 fimbriae by E. coli is a virulence factor for pathogenesis of the urinary tract [2, 3] A typical type 1 fimbriated bacterium has 200–500 peritrichously arranged fimbriae on its surface. A single type 1 fimbria is a 7-nm-wide and ∼1-μm-long surface polymer. The bulk of the structure is made up of ∼1000 copies of the major subunit protein, FimA, polymerized into a right-handed helical structure. Small quantities of the minor components (FimF, FimG, and FimH) are also present. The FimH protein is the receptor-recognizing element of type 1 fimbriae. FimH is located at the organelle tip in a short fibrillum and perhaps additionally intercalated along the fimbrial shaft. The FimF and FimG components seem to be required for integration of the FimH adhesin into the fimbriae The components of the fimbrial organelle are encoded by the chromosomally located fim gene cluster. In addition to the structural components, this 9.5-kb DNA segment also encodes the fimbrial biosynthesis machinery as well as regulatory elements. The expression of type 1 fimbriae is phase variable, with individual cells switching between fimbriated and nonfimbriated states. This characteristic is due to inversion of a 314-bp DNA segment located immediately upstream of the fimA gene. A promoter residing in this phase switch drives the expression of the fim genes when the switch is in the on orientation but not when it is in the off orientation. Two recombinases, FimB and FimE, mediate inversion of the phase switch by binding to recognition sites that flank and overlap the left and right inverted repeats of the switch. The fimbrial organelle components (FimA, FimF, FimG, and FimH) are produced as precursors having an N-terminal signal sequence. This feature is subsequently removed during export across the inner membrane. Thus, the FimH protein is produced as a precursor of 300 amino acids and is processed into a mature form of 279 amino acids. Further export, from the periplasm and across the outer membrane, is dependent on a fimbriae-specific export and assembly system constituted by the FimC and FimD proteins (reviewed in [1]) Herein, we review our own work regarding the molecular characterization of the FimH adhesin. We have used three technology sets to study FimH receptor-ligand interactions: linker mutagenesis of the entire protein, construction of novel FimH-FocH hybrids, and random mutagenesis of the receptor-binding domain. The use of FimH for the display of heterologous sequences ranging from immune-relevant foreign epitopes to random peptide libraries is also discussed Linker mutagenesis of the fimH geneIn an initial approach to elucidate structure-function relationships of FimH, amino acid changes were introduced in a number of positions spanning the entire sequence [4]. Specifically, BglII restriction sites were introduced into the fimH gene at positions corresponding to residues 8, 56, 136, 225, 259, and 279 of the mature FimH protein. This resulted in the introduction of an arginine-serine sequence at each position The effects of each of the mutations were assessed in an E. coli K-12 fimH-null mutant strain. Alterations in FimH were shown in some cases to affect the number and morphology of fimbriae produced on the cell surface, consistent with the notion that FimH is also involved in initiation of fimbrial biosynthesis. To study FimH receptor-binding activity, each of the mutants was examined for its ability to cause mannose-sensitive hemagglutination of guinea pig erythrocytes. Mutations at positions 56, 136, and 279 completely abolished hemagglutination activity, while mutations at positions 8, 225, and 229 still gave wild-type–like titers (figure 1A). These preliminary data pinpointed the amino acid positions 56 and 136 as being either part of or proximal to the receptor-binding domain of FimH. The mutation at position 279 most likely affects integration of FimH into the fimbrial organelle A Linker mutagenesis of the fimH gene. Shown is an overview of the codon positions in the mature FimH protein that were mutated and the resulting agglutination phenotype. Relative positions of the signal peptide (black), receptor-binding domain (gray) and organelle integration domain (white) are indicated. B Summary of the phenotypes conferred by FimH-FocH hybrids. The extent of FimH in the hybrid constructs is indicated Probing the receptor-binding domain of FimH by fimbrial displayed FimH-FocH hybridsA more precise analysis of FimH structure function was done by exploiting the compatibility of type 1 fimbriae with another fimbrial system, namely F1C fimbriae. Unlike type 1 fimbriae, F1C fimbriae do not target d-mannosides but instead demonstrate high affinity binding to the GalNAcβ1-4Galβ sequence of glycolipids [5]. The structure of F1C fimbriae closely resembles that of type 1 fimbriae, and they contain minor components consisting of FocF, FocG, and FocH. The FimH and FocH proteins, which show 36% sequence identity, can be readily exchanged between the two systems, resulting in hybrid organelles with receptor specificities defined by the adhesin [6]. This information permitted the novel approach of using FocH as a scaffold into which various parts of the FimH adhesin could be introduced to probe the extent of the receptor-binding domain in the context of intact fimbrial organelles The receptor binding of fimbriae-presented chimeric FimH-FocH hybrids was studied by examining d-mannose binding and agglutination of eukaryotic cells (figure 1B) [7]. FimH-FocH hybrids containing 72% of the N-terminus of FimH fused to the complementary sector of FocH conferred agglutination of erythrocytes and yeast cells at a comparable level to that of FimH. Similarly constructed fusions containing 56% and 66% FimH conferred binding to d-mannose but failed to agglutinate erythrocytes or yeast cells. This suggests that d-mannose binding and agglutination actually can be distinct phenotypes and that the two characteristics can be related to physically distinct regions in FimH. The receptor-binding capacity of hybrid fusions containing ⩽50% of the FimH N-terminal region was virtually abolished. Taken together, the data from this novel chimeric FimH-FocH fusion approach suggest that the minimal information for d-mannose recognition in FimH is located within the first 50% of the protein. Furthermore, this “core domain” can be complemented with steric information residing in the remaining part of FimH. Of note, amino acids 185–201 in the mature protein contain information that upgrades the d-mannose–binding core domain to additionally confer agglutination. The effect of this region in modulating FimH receptor recognition might only be observable in the context of complete fimbrial organelles presented on the bacterial surface A random mutagenesis approach to study FimH structure functionIn uropathogenic strains, FimH-mediated adhesion to mannosylated glycoproteins can vary due to minor structural alterations occurring naturally in the N-terminal half of the protein. Detailed analyses of such variants have revealed that these alterations lead to physiologically important changes in receptor recognition. In about 80% of fecal E. coli isolates, the FimH adhesin can bind only to trimannose receptors. In contrast, the FimH adhesins from ∼70% of urinary tract isolates carry minor mutations (compared with the fecal isolates) that enhance the ability to recognize monomannose receptors. The mutant alleles confer a significantly higher tropism for the uroepithelium and dramatically enhance the ability of E. coli to colonize the mouse urinary tract [3]. Some of the monomannose-binding E. coli can also recognize complex oligosaccharides with no terminally exposed mannose residues Naturally occurring fimH mutants adapted to enhance colonization of either commensal intestinal or pathogenic extra-intestinal niches constitute a relatively tight group of receptor affinities that can only be invoked by highly specific structural changes. In a study to probe the binding potential of the FimH adhesin more extensively, a random mutant library based on the fimH gene from E. coli K-12 strain PC31 was constructed and analyzed for the functional impact of non-selective mutations [8] The FimH mutant library was constructed by error-prone polymerase chain reaction and specifically targeted the receptor-binding region of the molecule (codons 8–225 of the mature protein). FimH-mediated binding of E. coli was assessed quantitatively to monomannose, oligomannose, and oligosaccharide substrates. Both the pathogenicity-adaptive monomannose- and the highly conserved oligomannose-binding phenotypes could be altered by minor amino acid changes in the FimH protein. The monomannose-binding phenotype was particularly sensitive to changes, with extensive differences in binding observed in comparison to wild-type FimH levels. Two FimH variants were also identified that could mediate adherence to oligosaccharide substrates. Different structural alterations caused similar functional changes in FimH, suggesting a high degree of flexibility to target recognition by this adhesin The three-dimensional structure of a FimH-FimC complex was resolved recently by x-ray crystallography [9]. According to this study, the FimH protein is folded into two domains, an N-terminal adhesive domain (residues 1–156) and a C-terminal organelle-integration domain (residues 160–279) linked by a tetra-peptide loop (figure 2). A carbohydrate-binding pocket was identified at the tip of the jelly roll–shaped adhesive (lectin) domain. An important pattern emerges from the analysis of the distribution of functional amino acid substitutions that have been identified in FimH (reviewed in [8]). Specific mutations that decrease the mannose-binding capability of FimH are caused by amino acid changes that occur either within or near the carbohydrate-interacting residues that form the FimH binding pocket. Conversely, amino acid changes that result in an enhanced mannose-binding capability of FimH do not interact directly with the mannose receptor site. Instead, these mutations occur as a mirror distribution in the “bottom” part of the lectin domain. We therefore speculate that mutations that result in enhanced FimH recognition of mannose may alter the conformational stability of the protein loops that carry the receptor-interacting residues. It is possible that this phenomenon also exists for other fimbrial adhesins ß-sheet topology diagram of the lectin domain of FimH [9] indicating structural changes that alter FimH function. Though the length of the sheets and loops does not reflect the actual size, the relative positions of the labeled residues are indicated accurately. The data include functional amino acid changes identified by ourselves and others. Residues are indicated as substitutions causing enhanced monomannose binding (•), substitutions destroying monomannose binding (▪), and individually neutral substitutions that act in concert to enhance monomannose binding (X). The C-HEGA–interacting residues in the FimH carbohydrate-binding pocket are also indicated (○) Overexpression and purification of the FimH receptor-binding domainThe purification of minor adhesive components of fimbriae has proven to be extremely difficult due to the inherent stable structure of these organelles. Furthermore, the adhesins are highly susceptible to periplasmic proteases when not complexed with their cognate chaperone proteins. On this background, we engineered a version of FimH consisting of the receptor-binding domain (i.e., the signal peptide and residues 1–156 of the mature protein) fused to a histidine tag [10]. This construct was soluble, highly stable, and easy to purify. Most important, the construct was functionally active, as demonstrated by its ability to bind to α-d-mannosylated bovine serum albumin. It is possible that this approach to purify the receptor-binding domain of FimH may be applied to purify other fimbrial adhesins for structure-function studies and identification of their cognate receptor targets FimH-based vaccinesFimbriae are normally very good immunogens both in the context of live vaccines and as purified proteins. They are particularly attractive candidates for epitope display because of their high numbers on the cell surface, their inherent adhesive properties, and the ease with which they can be purified. Type 1 fimbriae have been used to display heterologous peptides in connection with the development of vaccine systems. Various heterologous sequences, representing immune-relevant sectors of foreign proteins, have been authentically displayed on the bacterial surface in both FimA and FimH [11]. In this case, FimA and FimH represent high- and low-valency display units, a feature unique to the fimbrial surface display system More recently, the possibility of making binary adhesions has been explored. To this end, random peptide libraries were inserted in a permissive position in the C-terminal domain of FimH. From such libraries, specific binders to, for example, certain heavy metals were isolated and characterized. Not only was it possible to engineer new receptor specificities into the C-terminal domain of the protein, but the natural binding domain residing in the N-terminal part of the protein was unaffected, resulting in molecules with two distinct binding sites and related binding phenotypes [12]. Furthermore, the natural binding site could be modified to yield new and interesting specificities [8, 12] The identification of the FimH adhesin as a key player in E. coli–mediated urinary tract infections has spurred significant efforts into the development of a FimH-based anti–urinary tract infection vaccine. Indeed, the feasibility of this approach has shown highly promising results in both mouse and monkey urinary tract models. In this respect, it is interesting to note that our histidine-tagged FimH protein construct could be used in the development of such a vaccine. Furthermore, the ability of FimH to display heterologous epitopes might pave the way for FimH-based, multivalent, single-protein vaccines Our understanding of the structure-function relationship of the FimH protein, together with its role in virulence, has increased significantly in recent years. Molecular manipulation of FimH has demonstrated that this adhesin exhibits a high degree of flexibility toward target recognition. From our current knowledge of the x-ray structure of FimH, it is difficult to speculate on how this protein interacts with larger receptor compounds, such as trimannose, or with other known oligosaccharide inhibitors that exhibit a binding affinity 10- to 30-fold higher than that of monomannose. It is very likely that a thorough understanding of the structural and functional basis of the natural adaptability of FimH will require co-crystallization of different FimH variants with various types of receptor molecules. Furthermore, these studies may require FimH analyses in the context of intact fimbrial organelles. In this respect, the FimH variants identified in the course of our studies constitute a well-defined group of variants based on the K-12 fimH allele. Studies on the adaptability of the E. coli FimH protein may serve as a paradigm for the analysis of other bacterial adhesins that can be functionally modified by naturally occurring mutations to result in enhanced virulence We thank David Hasty (University of Tennessee) for many fruitful discussions