Location of binding sites on common type 1 fimbriae from Escherichia coli.

Common type 1 fimbriae were isolated from Escherichia coli and their length distribution profile was determined before and after treatment with ultrasound. As fimbriae were shortened, so their haemagglutinating capacity decreased, but their ability to bind to erythrocytes did not decrease to the same extent. Isolated fimbriae did not agglutinate inside-out vesicles prepared from horse erythrocytes or liposomes, suggesting that the binding mechanism was not based on non-specific hydrophobic interactions. The results support a lateral rather than a terminal location for the fimbrial binding site responsible for haemagglutination.     

Fragmentation of Escherichia coli type 1 fimbriae exposes cryptic D-mannose-binding sites.

Cells of the gram-negative bacterium Escherichia coli are able to attach to various host cells by means of a mannose-specific adhesin associated with type 1 fimbriae. Here we show that fragmentation of type 1 fimbriae by freezing and thawing results in increased mannose-binding activity as demonstrated by increased hemagglutination, increased stimulation of human lymphocyte proliferation, and increased binding of the mannose-containing enzyme horseradish peroxidase. Increased activity in all three assays was mannose sensitive and was not exhibited by FimH- mutant type 1 fimbriae lacking the adhesin. Scatchard analysis of the data from peroxidase binding assays showed that unfrozen and frozen fimbriae contain binding sites displaying two classes of affinity. Frozen and thawed fimbriae expressed an increase in the number of high-affinity binding sites. These results show that fragmentation of the fimbrial structure exposes cryptic mannose-binding activity associated with type 1 fimbriae, presumably that of internally located adhesin molecules. Our data support earlier observations that adhesin moieties of type 1 fimbriae are located both at the tips and at intervals along the length of the fimbriae. In addition, our data suggest that only the adhesin moieties that are located at the fimbrial tips are functional in binding mannose. Adhesins located along the length of the fimbriae have their mannose-binding activity buried within the fimbrial structure and hence are not functional. We propose an updated model for the structure of type 1 fimbriae that is in agreement with the above observations.     

Immunoelectron microscopic analysis of elongation of type 1 fimbriae in Escherichia coli.

Using 10- and 20-nm-diameter gold particles conjugated to an antifimbrial monoclonal antibody, we analyzed the location of assembly of newly formed subunits on growing type 1 fimbriae of Escherichia coli. Fimbriae were removed from an E. coli K-12-derived strain, CSH50, by blending. Blended cells were allowed to regenerate their fimbriae in growth medium for approximately 25 min, after which they were labeled with a 20-nm-gold-monoclonal antibody probe. Continued outgrowth of these labeled fimbriae was allowed for additional time intervals, after which they were labeled with a 10-nm-gold-monoclonal antibody probe. The resulting fimbriae, double labeled with 10- and 20-nm-diameter gold particles, were examined in an electron microscope. The pattern of labeling on individual fimbrial organelles indicated morphologically that newly synthesized subunits are added to a growing organelle at its base.     

Glycerol-induced unraveling of the tight helical conformation of Escherichia coli type 1 fimbriae.

Glycerol was found to unravel the helical conformation of Escherichia coli type 1 fimbriae without appreciable depolymerization. The linearized fimbrial polymers have a diameter of 2 nm, react strongly with a monoclonal antibody directed at an inaccessible epitope on native fimbriae, and display greater mannose-binding activity and trypsin sensitivity than native fimbriae. Removal of glycerol by dialysis results in spontaneous reassembly of the linear polymers into structures morphologically, antigenically, and functionally indistinguishable from native fimbriae.     

FimH adhesin of Escherichia coli K1 type 1 fimbriae activates BV-2 microglia

The generation of intense inflammation in the subarachnoid space in response to meningitis-causing bacteria contributes to brain dysfunction and neuronal injury in bacterial meningitis. Microglia, the major immune effector cells in the central nervous system (CNS), become activated by bacterial components to produce proinflammatory immune mediators. In this study, we showed that FimH adhesin, a tip component of type 1 fimbriae of meningitis-causing Escherichia coli K1, activated the murine microglial cell line, BV-2, which resulted in the production of nitric oxide and the release of tumor necrosis factor-alpha. Mitogen-activated protein kinases, ERK and p-38, and nuclear factor-kappaB were involved in FimH adhesin-mediated microglial activation. These findings suggest that FimH adhesin contributes to the CNS inflammatory response by virtue of activating microglia in E. coli meningitis.     

Mitogenic stimulation of human B lymphocytes by the mannose-specific adhesin on Escherichia coli type 1 fimbriae

Escherichia coli type 1 fimbriae contain in association with the major structural protein a lectin-like adhesin moiety that mediates attachment of E. coli to mannose-containing receptors on the surface of host cells. We have investigated the lymphocyte mitogenic activity of this mannose-specific adhesin by comparing the ability of purified wild type type 1 fimbriae containing the adhesin and mutant type 1 fimbriae lacking the adhesin to stimulate proliferation in human lymphocytes. Both fimbriae stimulated a peak of proliferation at 8 days whereas only the wild type fimbriae stimulated an additional peak of proliferation occurring at 3 days. Proliferation at 3 days but not at 8 days could be blocked by the addition of alpha-methyl-D-mannoside. Neonatal lymphocytes from umbilical cord blood responded to both wild type and mutant fimbriae in a fashion similar to adult cells. Stimulation of separated T and non-T cell populations indicated that the proliferation seen at 3 days was solely due to non-T cells whereas the 8-day response was due to T cell proliferation. The addition of gamma-irradiated T cells did not appear to enhance the 3-day response of the non-T cells. However, the 8-day response by T cells was dependent on the presence of gamma-irradiated non-T cells. In cultures of unseparated cells, wild type fimbriae stimulated more than 75% of the B cells to enter the S and G2 phase at 3 days whereas at 8 days cycling T cells were present in both wild type and mutant fimbriae-stimulated cultures. Taken together, our observations suggest that the adhesin molecule stimulates a polyclonal mitogenic response in B cells that peaks at 3 days, and other structural components of the fimbriae are responsible for evoking an 8-day (probably immune) response in T cells.     

Presence of type I-like fimbriae and thin filaments on diarrhoeagenic Escherichia coli

Abstract Type I-like fimbriae were found on more serotypes of diarrhoeagenic Escherichia coli than exprcted from previous observations. The fimbriae had long 2 nm thick filaments that appeared to extend from their tips, but thus far the thin filaments have not been clearly implicated in adhesion to gut epithelial cells.     

Antigen 43 and type 1 fimbriae determine colony morphology of Escherichia coli K-12

Colony morphology has been used as an important identification and characterization criterion in bacteriology for many decades. However, the molecular mechanisms underlying the appearance of different colony types have been given little attention. The synthesis of O antigen is defunct in Escherichia coli K-12, and colonies should accordingly only appear to be rough. However, previous reports have noted the presence of different interchangeable colony morphology types. In this study we have addressed the influence of two phase-variable surface structures, antigen 43 and type 1 fimbriae, on colony morphology. Due to differential expression of these structures, four different colony phenotypes could be distinguished. By creating and studying defined mutants of the respective loci, i.e. , flu and fim, we conclude that the presence or absence of the corresponding gene products on the cells correlates with the observed colony morphology forms. Interestingly, the habitat specificity of bacteria under static liquid conditions seems to correlate with the colony phenotypes.     

Genetic analysis of the phase variation control of expression of type 1 fimbriae in Escherichia coli.

Expression of type 1 fimbriae in Escherichia coli exhibits phase variation, whereby individual cells can alternate between states of organelle expression (Fim+) and nonexpression (Fim-). Strains with a fimD-lac operon fusion, in which lac, rather than fimD, expression is under the control of the fimD promoter, undergo Lac+ in equilibrium Lac- phase variation, instead. After positioning a lambda prophage adjacent to the operon fusion, we were able to isolate specialized lambda phage carrying both the fimD-lac fusion and the phase variation control region. Introduction of such phage into an Fim+ strain resulted in construction of a strain with a double, independently switching phenotype (Fim+ in equilibrium Fim- and Lac+ in equilibrium Lac-), demonstrating that the region controlling phase variation is contiguous with the fimD-lac operon fusion and is cis acting. When the specialized lambda phage was propagated on a delta lac delta fim strain, phase variation occurred within the plaques, confirming that the phase variation control region is carried on the specialized transducing phage. All lysogens acquired the Lac+ in equilibrium Lac- phenotype, except for two nonswitching Lac+ recombinants, which acquired Lac+ in equilibrium Lac- phase variation only by trans complementation with fim. Phase variation of type 1 fimbriae, therefore, appears to involve both a cis-active element, which is cloned on a specialized lambda phage, and a trans-active permissive factor, which is not present on the phage, but rather must be supplied by the recipient strain in the transduction.     

Linker insertion analysis of the FimH adhesin of type 1 fimbriae in an Escherichia coli fimH-null background

Abstract The gene encoding the Escherichia coli FimH adhesin of type 1 fimbriae has been subjected to linker insertion mutagenesis. Amino acid changes were introduced at a number of positions spanning the entire sequence in order to probe the structure-function relationship of the FimH protein. The effect of these mutations on the ability of bacteria to express a D-mannose binding phenotype was assessed in a fimH null mutant (MS4) constructed by allelic exchange in the E. coli K-12 strain PC31. Mutations mapping at amino acid residues 36, 58 and 279 of the mature FimH protein were shown to completely abolish binding to D-mannose receptors. Differences in the level of fimbriation were also observed as a result of some of the mutations in the fimH gene. These mutants may prove useful in dissecting receptor-ligand interactions by defining regions of the FimH protein that are important in erythrocyte binding.     

The major subunit of Escherichia coli type 1 fimbriae is not required for D-mannose-specific adhesion

Type 1 fimbriae are surface organelles on Escherichia coli, which mediate specific binding to D-mannose-containing structures. These fimbriae are heteropolymers composed of a major building element, the FimA protein, and small amounts of the FimF, FimG and FimH proteins. The FimH protein is uniquely responsible for the D-mannose receptor binding. In this work data are presented which indicate that the major subunit of type 1 fimbriae is dispensable for D-mannose-specific binding. A recombinant strain was studied which harboured an insertional deletion in the fimA gene, and was thereby unable to produce type 1 fimbriae; however, it was still able to express a D-mannose-binding phenotype. However, the deletion resulted in a 25-fold reduction of the adhesive potential, as measured by binding to D-mannose-coated Sepharose beads. Serological and specific receptor binding evidence is presented that suggests that the FimH adhesion is capable of being exposed on the bacterial surface without being an integral part of the fimbriae.     

Role of type I fimbriae in the aggregative adhesion pattern of enteroaggregative Escherichia coli

Enteroaggregative Escherichia coli (EAEC) is distinguished by its characteristic aggregative adherence (AA) pattern to cultured epithelial cells. In this study we investigated the role of type I fimbriae (TIF) in the AA pattern to HEp-2 cells and in biofilm formation. Accentuation of this pattern was observed when the adherence assay was performed in the absence of mannose. This effect was observed in the prototype EAEC strain 042 (O44:H18), O128:H35 strains and for other EAEC serotypes. Antiserum against TIF decreased AA by 70% and 90% for strains 042 and 18 (O128:H35 prototype strain), respectively. A non-polar knockout of fimD, the TIF usher, in strains 042 and 18 resulted in inhibition of the accentuated AA pattern of approximately 80% and 70% respectively, and biofilm formation diminution of 49% for 042::fimD and 76% for 18::fimD. Our data evidence a role for TIF in the AA pattern and in EAEC biofilm formation, demonstrating that these phenotypes are multifactorial.     

Escherichia coli fimbriae recognizing sialyl galactosides

Fimbriae recognizing sialyl galactosides (S fimbriae) were purified from an Escherichia coli strain. The S fimbriae were morphologically identical to type 1 and P fimbriae of E. coli and showed a hemagglutination that was abolished when erythrocytes were treated with neuraminidase. Hemagglutination by the purified fimbriae was inhibited by orosomucoid but not by its desialylated derivative. Of the oligosaccharides tested, sialyl-(alpha 2-3)-lactose and sialyl-(alpha 2-3)-N-acetyllactosamine had the strongest inhibitory activities. It was concluded that S fimbriae have the strongest affinity for (alpha 2-3)-linked sialyl galactosides. In the enzyme-linked immunosorbent assay, the hyperimmune serum to the S fimbriae reacted strongly with the homologous antigen but not with type 1, P, or nonhemagglutinating KS71C fimbriae of E. coli. Analogously, the hyperimmune sera to the other E. coli fimbriae did not react with the purified S fimbriae. The immunoprecipitation assay showed that S fimbriae on different E. coli serotypes shared immunological cross-reactivity.     

Regulatory cross-talk between adhesin operons in Escherichia coli: inhibition of type 1 fimbriae expression by the PapB protein

Pathogenic Escherichia coli often carry determinants for several different adhesins. We show a direct communication between two adhesin gene clusters in uropathogenic E.coli: type 1 fimbriae (fim) and pyelonephritis-associated pili (pap). A regulator of pap, PapB, is a key factor in this cross-talk. FimB recombinase turns on type 1 fimbrial expression, and PapB inhibited phase transition by FimB in both off-to-on and on-to-off directions. On-to-off switching requiring FimE was increased by PapB. By analysis of FimB- and FimE-LacZ translational fusions it was concluded that the increase in on-to-off transition rates was via an increase in FimE expression. Inhibition of FimB-promoted switching was via a different mechanism: PapB inhibited FimB-promoted in vitro recombination, indicating that FimB activity was blocked at the fim switch. In vitro analyses showed that PapB bound to several DNA regions of the type 1 fimbrial operon, including the fim switch region. These data show that Pap expression turns off type 1 fimbriae expression in the same cell. Such cross-talk between adhesin gene clusters may bring about appropriate expression at the single cell level.     

Assembly of a chemically synthesized peptide of Escherichia coli type 1 fimbriae into fimbria-like antigenic structures.

Escherichia coli type 1 fimbriae are composed of subunits, each of which comprises 158 amino acids. We synthesized a copy of a 13-residue peptide, located near the NH2 terminus of the fimbrial subunit, that assumed some of the properties of type 1 fimbriae. At pH 5.5 the synthetic peptide autoassembled into fibrillar structures that resembled type 1 fimbriae except that they appeared less rigid and rodlike. A quaternary structure-specific monoclonal antibody against type 1 fimbriae recognized the synthetic peptide in the assembled but not the unassembled state. Furthermore, when the synthetic peptide was injected in its fimbrial conformation into rabbits, it evoked antibodies that reacted with type 1 fimbriae isolated from E. coli.