Characterization of FasG segments required for 987P fimbria-mediated binding to piglet glycoprotein receptors.

The 987P fimbriae of enterotoxigenic strains of Escherichia coli bind to both glycoprotein and glycolipid receptors on the brush borders of piglet enterocytes. A mutation in lysine residue 117 of the adhesive subunit FasG [fasG(K117A)] previously shown to abrogate 987P binding to the lipid receptor sulfatide did not affect the interaction with the glycoprotein receptors. Both the fimbriae and the FasG subunits of the wild type and the fasG(K117A) mutant bound to the glycoprotein receptors, confirming that lysine 117 was not required for binding to the glycoprotein receptors. Truncated FasG molecules were used to identify domains required for glycoprotein receptor recognition. At least two segments which did not include lysine117, namely, residues 211 (glutamine) to 220 (serine) and 20 (aspartic acid) to 41 (serine), were shown to be involved in the FasG-glycoprotein receptor interactions by ligand-blotting assays. Changing isoleucine 217 or leucine 215 of FasG to alanine abolished the property of a truncated FasG fusion protein to inhibit 987P recognition of its glycoprotein receptors. Thus, the K117 residue of FasG is required only for binding to the glycolipid receptor, whereas the newly identified hydrophobic residues of the FasG subunit are required specifically for the recognition of the glycoprotein receptor. Taken together, our data indicate that different residues of the FasG adhesin are important in 987P fimbrial binding to sulfatide and glycoprotein receptors, suggesting different mechanisms of interaction.

Enhanced immune responses to viral epitopes by combining macrophage-inducible expression with multimeric display on a Salmonella vector.

In this study, the immunogenicity of chimeric 987P fimbriae on a Salmonella vaccine strain was improved by optimizing fimbrial expression. The constitutive tetA promoter and the in vivo activated nirB and pagC promoters were evaluated for their use to express two epitopes of the transmissible gastroenteritis virus (TGEV) spike protein carried by fimbriae which were displayed on a Salmonella vaccine strain. Constructs with the pagC promoter were shown to drive increased expression of chimeric 987P fimbriae in macrophages as well as in Mg(2+)-poor media, mimicking a major environmental signal found in Salmonella-containing endocytic vacuoles of macrophages. Mice immunized orally with a Salmonella vaccine strain which expressed chimeric fimbriae from the pagC promoter elicited significantly higher mucosal and systemic immune responses to both the 987P fimbriae and the TGEV epitopes than mice immunized with the same strain hosting a tetA or nirB promoter-driven expression plasmid. Moreover, only the Salmonella vaccine strains harboring a plasmid with the pagC promoter, with or without an additional tetA promoter in tandem, elicited neutralizing antibodies to TGEV. This indicated that the pagC promoter can be used successfully to improve epitope-display by chimeric fimbriae on Salmonella vaccine strains for the induction of a desired immune response.

Mucosal and systemic immune responses to chimeric fimbriae expressed by Salmonella enterica serovar typhimurium vaccine strains.

Recombinant live oral vaccines expressing pathogen-derived antigens offer a unique set of attractive properties. Among these are the simplicity of administration, the capacity to induce mucosal and systemic immunity, and the advantage of permitting genetic manipulation for optimal antigen presentation. In this study, the benefit of having a heterologous antigen expressed on the surface of a live vector rather than intracellularly was evaluated. Accordingly, the immune response of mice immunized with a Salmonella enterica serovar Typhimurium vaccine strain expressing the Escherichia coli 987P fimbrial antigen on its surface (Fas(+)) was compared with the expression in the periplasmic compartment (Fas(-)). Orally immunized BALB/c mice showed that 987P fimbriated Salmonella serovar Typhimurium CS3263 (aroA asd) with pCS151 (fas(+) asd(+)) elicited a significantly higher level of 987P-specific systemic immunoglobulin G (IgG) and mucosal IgA than serovar Typhimurium CS3263 with pCS152 (fasD mutant, asd(+)) expressing 987P periplasmic antigen. Further studies were aimed at determining whether the 987P fimbriae expressed by serovar Typhimurium chi4550 (cya crp asd) could be used as carriers of foreign epitopes. For this, the vaccine strain was genetically engineered to express chimeric fimbriae carrying the transmissible gastroenteritis virus (TGEV) C (379-388) and A (521-531) epitopes of the spike protein inserted into the 987P major fimbrial subunit FasA. BALB/c mice administered orally serovar Typhimurium chi4550 expressing the chimeric fimbriae from the tet promoter in pCS154 (fas(+) asd(+)) produced systemic antibodies against both fimbria and the TGEV C epitope but not against the TGEV A epitope. To improve the immunogenicity of the chimeric fimbriae, the in vivo inducible nirB promoter was inserted into pCS154, upstream of the fas genes, to create pCS155. In comparison with the previously used vaccine, BALB/c mice immunized orally with serovar Typhimurium chi4550/pCS155 demonstrated significantly higher levels of serum IgG and mucosal IgA against 987P fimbria. Moreover, mucosal IgA against the TGEV C epitope was only detected with serovar Typhimurium chi4550/pCS155. The induced antibodies also recognized the epitopes in the context of the full-length TGEV spike protein. Hence, immune responses to heterologous chimeric fimbriae on Salmonella vaccine vectors can be optimized by using promoters known to be activated in vivo.

A role for Salmonella fimbriae in intraperitoneal infections.

Enteric bacteria possess multiple fimbriae, many of which play critical roles in attachment to epithelial cell surfaces. SEF14 fimbriae are only found in Salmonella enterica serovar Enteritidis (S. enteritidis) and closely related serovars, suggesting that SEF14 fimbriae may affect serovar-specific virulence traits. Despite evidence that SEF14 fimbriae are expressed by S. enteritidis in vivo, previous studies showed that SEF14 fimbriae do not mediate adhesion to the intestinal epithelium. Therefore, we tested whether SEF14 fimbriae are required for virulence at a stage in infection after the bacteria have passed the intestinal barrier. Polar mutations that disrupt the entire sef operon decreased virulence in mice more than 1,000-fold. Nonpolar mutations that disrupted sefA (encoding the major structural subunit) did not affect virulence, but mutations that disrupted sefD (encoding the putative adhesion subunit) resulted in a severe virulence defect. The results indicate that the putative SEF14 adhesion subunit is specifically required for a stage of the infection subsequent to transit across the intestinal barrier. Therefore, we tested whether SefD is required for uptake or survival in macrophages. The majority of wild-type bacteria were detected inside macrophages soon after i.p. infection, but the sefD mutants were not readily internalized by peritoneal macrophages. These results indicate that the potential SEF14 adhesion subunit is essential for efficient uptake or survival of S. enteritidis in macrophages. This report describes a role of fimbriae in intracellular infection, and indicates that fimbriae may be required for systemic infections at stages beyond the initial colonization of host epithelial surfaces.

Lysine residue 117 of the FasG adhesin of enterotoxigenic Escherichia coli is essential for binding of 987P fimbriae to sulfatide.

The FasG subunit of the 987P fimbriae of enterotoxigenic strains of Escherichia coli was previously shown to mediate fimbrial binding to a glycoprotein and a sulfatide receptor on intestinal brush borders of piglets. Moreover, the 987P adhesin FasG is required for fimbrial expression, since fasG null mutants are nonfimbriated. In this study, fasG was modified by site-directed mutagenesis to study its sulfatide binding properties. Twenty single mutants were generated by replacing positively charged lysine (K) or arginine (R) residues with small, nonpolar alanine (A) residues. Reduced levels of binding to sulfatide-containing liposomes correlated with reduced fimbriation and FasG surface display in four fasG mutants (R27A, R286A, R226A, and R368). Among the 16 remaining normally fimbriated mutants with wild-type levels of surface-exposed FasG, only one mutant (K117A) did not interact at all with sulfatide-containing liposomes. Four mutants (K117A, R116A, K118A, and R200A) demonstrated reduced binding to such liposomes. Since complete phenotypic dissociation between the structure and specific function of 987P was observed only with mutant K117A, this residue is proposed to play an essential role in the FasG-sulfatide interaction, possibly communicating with the sulfate group of sulfatide by hydrogen bonding and/or salt bridge formation. Residues K17, R116, K118, and R200 may stabilize this interaction.

Polymeric display of immunogenic epitopes from herpes simplex virus and transmissible gastroenteritis virus surface proteins on an enteroadherent fimbria.

The strong immunogenicity of bacterial fimbriae results from their polymeric and proteinaceous nature, and the protective role of these immunogens in experimental or commercial vaccines is associated with their capacity to induce antiadhesive antibodies. Fimbria-mediated intestinal colonization by enteropathogens typically leads to similar antibody responses. The possibility of taking advantage of these properties was investigated by determining whether enteroadhesive fimbriae, like the 987P fimbriae of enterotoxigenic Escherichia coli, can serve as carriers for foreign antigens without losing their adhesive characteristics. Random linker insertion mutagenesis of the fasA gene encoding the major 987P subunit identified five different mutants expressing wild-type levels of fimbriation. The linker insertion sites of these mutants were used to introduce three continuous segments of viral surface glycoproteins known to be accessible to antibodies. These segments encode residues 11 to 19 or 272 to 279 of herpes simplex virus type 1 (HSV-1) glycoprotein D [gD(11-19) and gD(272-279), respectively] or residues 379 to 388 of the transmissible gastroenteritis virus (TGEV) spike protein [S(379-388)]. Studies of bacteria expressing fimbriae incorporating mutated FasA subunits alone or together with wild-type FasA subunits (hybrid fimbriae) indicated that foreign epitopes were best exported and displayed on assembled fimbriae when they were inserted near the amino terminus of FasA. Fimbriated bacteria expressing FasA subunits carrying the HSV gD(11-19) or the TGEV S(379-388) epitope inserted between the second and third residues of mature FasA elicited high levels of foreign epitope antibodies in all rabbits immunized parenterally. Antibodies against the HSV epitope were also shown to recognize the epitope in the context of the whole gD protein. Because the 987P adhesive subunit FasG was shown to be present on mutated fimbriae and to mediate bacterial attachment to porcine intestinal receptors, polymeric display of foreign epitopes on 987P offers new opportunities to test the potential beneficial effect of enteroadhesion for mucosal immunization and protection against various enteric pathogens.

Improved allelic exchange vectors and their use to analyze 987P fimbria gene expression.

A series of vectors has been developed to provide improved positive and negative selection for allelic exchange. Based on homologous regions of DNA ranging in size from less than 200 bp to over 1 kb, we have successfully used these new plasmids to introduce or remove markers in chromosomal or plasmid DNA. Wild type fimbria genes were replaced both in Salmonella enteritidis (sefA, agfA and fimC) and Escherichia coli (fasA and fasH). Regulation of 987P fimbriation could be identified after replacement of fasA and fasH with allelic reporter fusions. The expression of fasA but not fasH is dependent upon the osmolarity of the growth medium in an HNS-dependent manner, but unlike some other fimbrial systems expression is not dependent on the exogenous iron concentration.

Differential regulation of fasA and fasH expression of Escherichia coli 987P fimbriae by environmental cues.

An early process in the pathogenesis of enteric bacteria is colonization of the intestinal epithelium leading to local multiplication, pathophysiological interactions with the host and further spreading. Attachment is typically mediated by bacterial fimbriae, which are selectively expressed during growth in the intestine. Here we report an analysis of the regulation of 987P fimbrial expression of enterotoxigenic Escherichia coli (ETEC). Expression of both fasH, the transcriptional activator of the 987P fimbrial genes, and fasA, the major fimbrial subunit, is regulated in response to a variety of environmental stimuli. We have found that expression of fasH is regulated in response to the carbon status of the growth medium by the cAMP-CRP complex. Moreover, fasH is regulated in response to both the nitrogen status of the growth medium and the external pH. Expression of fasA is activated by FasH, and is also selectively regulated in response to growth temperature by HNS. Regulation of fimbrial expression by carbon and/or nitrogen gradients is proposed to provide a mechanism that allows preferential colonization of different segments of the intestine by various enteropathogens, such as ETEC, enteropathogenic E. coli and Vibrio cholerae.

Invasion of chicken reproductive tissues and forming eggs is not unique to Salmonella enteritidis.

Experiments were conducted in which Salmonella enteritidis Phage Type 8, Phage Type 2, and RDNC (reaction does not conform) or three isolates of Salmonella typhimurium of diverse origin were fed to adult laying hens to determine if S. enteritidis has a selective advantage over S. typhimurium, which is now rarely isolated from chicken eggs, in its capacity to invade reproductive tissues. The results revealed that S. enteritidis and S. typhimurium may be equal in their potential to colonize the tissues of the reproductive tract and eggs that are forming in the oviduct prior to oviposition. S. enteritidis, but not S. typhimurium, was isolated from egg contents after oviposition. The degree to which intestinal, hepatic, splenic, or reproductive tissues were colonized by either serotype was not seen to affect the rate of colonization of eggs forming in the oviduct or the contamination of eggs after oviposition. Virulence factors related to the difference in the association of S. enteritidis and S. typhimurium with egg-borne salmonellosis remain to be defined.

Porcine 987P glycolipid receptors on intestinal brush borders and their cognate bacterial ligands.

Certain strains of enterotoxigenic Escherichia coli adhere to piglet intestinal epithelial cells by means of the 987P fimbriae. The 987P fimbrial structure consists of a helical arrangement of three fimbrial proteins, namely, the major subunit FasA and two minor subunits, FasF and FasG. FasG, which is located at the fimbrial tip and at various positions along the fimbriae, mediates 987P binding to glycoprotein receptors. In this study, we isolated and analyzed the structure of piglet glycolipid brush border receptors and characterized their cognate ligands on the 987P fimbriae. Two major glycolipid bands recognized by 987P fimbrial probes in thin-layer chromatography overlay assays were further purified by high-performance thin-layer chromatography and shown to comigrate with control galactosylceramide containing hydroxylated fatty acids and with sulfatide. Their structures were confirmed by fast atom bombardment mass spectrometry, which detected homologous series of ceramide monohexoside and sulfatide with hydroxylated fatty acyl chains ranging from h16:0 to h24:0. Assembled 987P fimbriae, pre- and postassembly dissociated fimbrial subunits, and Fab fragments of specific anti-FasG, -FasF, and -FasA were used to inhibit 987P-mediated bacterial binding to the two identified piglet glycolipids and corresponding isoreceptor controls. Only assembled fimbriae and anti-FasG Fab fragments were significantly able to inhibit bacterial binding to sulfatide, indicating that in addition to glycoproteins, FasG recognizes a specific glycolipid of piglet brush borders. In contrast, only anti-FasA Fab fragments were significantly able to inhibit bacterial binding to galactosylceramide with hydroxylated fatty acids and piglet hydroxylated ceramide monohexoside, indicating that FasA may determine a third type of ligand-receptor interaction in the piglet intestines. Since these bacterial adhesins recognize their respective glycolipid receptors only after being assembled in their final fimbrial quaternary structure, adhesin binding may involve cooperative interactions and the subunits by themselves may have very low binding affinities. Alternatively, conformation-sensitive domains of these subunits present in the assembled fimbriae may be required for glycolipid binding.

Identification of major and minor chaperone proteins involved in the export of 987P fimbriae.

The 987P fimbriae of Escherichia coli consist mainly of the major subunit, FasA, and two minor subunits, FasF and FasG. In addition to the previously characterized outer membrane or usher protein FasD, the FasB, FasC, and FasE proteins are required for fimbriation. To better understand the roles of these minor proteins, their genes were sequenced and the predicted polypeptides were shown to be most similar to periplasmic chaperone proteins of fimbrial systems. Western blot (immunoblot) analysis and immunoprecipitation of various fas mutants with specific antibody probes identified both the subcellular localizations and associations of these minor components. FasB was shown to be a periplasmic chaperone for the major fimbrial subunit, FasA. A novel periplasmic chaperone, FasC, which stabilizes and specifically interacts with the adhesin, FasG, was identified. FasE, a chaperone-like protein, is also located in the periplasm and is required for optimal export of FasG and possibly other subunits. The use of different chaperone proteins for various 987P subunits is a novel observation for fimbrial biogenesis in bacteria. Whether other fimbrial systems use a similar tactic remains to be discovered.

Ordered translocation of 987P fimbrial subunits through the outer membrane of Escherichia coli.

The 987P fimbria of enterotoxigenic Escherichia coli is a heteropolymeric structure which consists essentially of a major FasA subunit and a minor subunit, the FasG adhesin. The latter harbors the binding moiety for receptor molecules on piglet intestinal epithelial cells. In this study, anti-FasF antibody probes were developed and used to demonstrate that the FasF protein represents a new minor fimbrial component. FasF was identified in highly purified fimbriae, and its sequence demonstrated significant levels of similarity with that of FasA. Immune electron microscopy localized both the FasG and FasF proteins at the fimbrial tip as well as at broken ends and at various intervals along the fimbrial length. The presence of these minor proteins in purified 987P fimbriae was corroborated by enzyme-linked immunosorbent assay inhibitions. Finally, the use of nonfimbriated fasG, fasF, and fasA mutants indicated that subunit translocation through the outer membrane follows a specific order, FasG being the first, FasF being the second, and FasA being the third type of exported subunit. Since fimbriae are thought to grow from the base, FasG is proposed to be a tip adhesin and FasF is proposed to be a linker molecule between the adhesin and the fimbrial shaft. Moreover, export of FasG (or FasF) in the absence of FasF (or FasA) indicates that during the process of fimbrial biogenesis in the outer membrane, translocating events precede the initiation of subunit heteropolymerization.

A minor 987P protein different from the structural fimbrial subunit is the adhesin.

The 987P fimbriae produced by enterotoxigenic strains of Escherichia coli isolated from piglets mediate bacterial attachment to intestinal epithelial cells. These fimbriae consist essentially of a tight helical arrangement of one structural protein subunit encoded by fasA. Fimbriation and specific adhesion requires the expression of seven additional genes (fasB to fasH). In this study, we investigated whether FasA or another Fas protein, e.g., a potential minor fimbrial component, harbors the binding moiety for the pig 987P receptor glycoproteins. Fas proteins, specifically radiolabeled with an in vivo T7 expression system, were isolated from the periplasm and incubated with receptor-containing brush borders isolated from piglet intestinal epithelial cells. FasG bound best to brush borders, whereas no FasA adhered to them. Additional evidence that FasG, and not FasA, is the 987P adhesin was provided by ligand blotting inhibition assays indicating that FasG alone inhibited fimbrial binding to 987P receptors and that in the absence of FasG, other Fas proteins were not inhibitory. FasG was identified in purified fimbrial preparations with a specific anti-FasG antibody probe. Moreover, FasG was shown to be tightly associated with the fimbrial structure, since it was released only after disassembling fimbriae by heat and sodium dodecyl sulfate treatments. The primary structure of FasG, deduced from the DNA sequence, exhibited 19.1 to 24.4% similarity to FasA and large minor components and/or adhesins of other fimbriae. FasG is the first-described minor fimbrial subunit shown to be essential for both fimbrial biogenesis and specific adhesion.

Permissive linker insertion sites in the outer membrane protein of 987P fimbriae of Escherichia coli.

The FasD protein is essential for the biogenesis of 987P fimbriae of Escherichia coli. In this study, subcellular fractionation was used to demonstrate that FasD is an outer membrane protein. In addition, the accessibility of FasD to proteases established the presence of surface-exposed FasD domains on both sides of the outer membrane. The fasD gene was sequenced, and the deduced amino acid sequence was shown to share homologous domains with a family of outer membrane proteins from various fimbrial systems. Similar to porins, fimbrial outer membrane proteins are relatively polar, lack typical hydrophobic membrane-spanning domains, and posses secondary structures predicted to be rich in turns and amphipathic beta-sheets. On the basis of the experimental data and structural predictions, FasD is postulated to consist essentially of surface-exposed turns and loops and membrane-spanning interacting amphipathic beta-strands. In an attempt to test this prediction, the fasD gene was submitted to random in-frame linker insertion mutagenesis. Preliminary experiments demonstrated that it was possible to produce fasD mutants, whose products remain functional for fimbrial export and assembly. Subsequently, 11 fasD alleles, containing linker inserts encoding beta-turn-inducing residues, were shown to express functional proteins. The insertion sites were designated permissive sites. The inserts used are expected to be least detrimental to the function of FasD when they are inserted into surface-exposed domains not directly involved in fimbrial export. In contrast, FasD is not expected to accommodate such residues in its amphipathic beta-strands without being destabilized in the membrane and losing function. All permissive sites were sequenced and shown to be located in or one residue away from predicted turns. In contrast, 5 of 10 sequenced nonpermissive sites were mapped to predicted amphipathic beta-strands. These results are consistent with the structural predictions for FasD.

Genetic analysis of 987P adhesion and fimbriation of Escherichia coli: the fas genes link both phenotypes.

The 987P fimbrial gene cluster has recently been shown to contain eight genes (fasA to fasH) clustered on large plasmids of enterotoxigenic Escherichia coli and adjacent to a Tn1681-like transposon encoding the heat-stable enterotoxin STIa. Different genetic approaches were used to study the relationship between 987P fimbriation and adhesion. TnphoA mutagenesis, complementation assays, and T7 RNA polymerase-promoted gene expression indicated that all of the fas genes were involved in fimbrial expression and adhesion. In contrast to other fimbrial systems, the lack of expression of any single fas gene never resulted in the dissociation of fimbriation and adhesion, indicating that the adhesin is required for fimbrial expression and suggesting that FasA, the fimbrial structural subunit itself, is the adhesin. In addition, fimbrial length was shown to be modulated by the levels of expression of different fas genes.

987P fimbrial gene identification and protein characterization by T7 RNA polymerase-induced transcription and TnphoA mutagenesis.

The 987P fimbrial gene cluster has been previously cloned as a 12 kb fragment from prototype strain 987. Gene products encoded by the whole clone were analysed by utilizing an in vivo system based on the induction of transcription by T7 RNA polymerase. The sensitivity of this technique permitted us to identify new proteins involved in 987P fimbriation. In total, eight proteins were detected, their genes (fasA to fasH) were mapped and their orientation of transcription determined. Several of the gene products demonstrated typical properties of exported proteins. Precursor and processed forms could be correlated after inhibiting protein transport with ethanol. The detection of enzymatically active fusion proteins after TnphoA (Tn5IS50L::phoA) mutagenesis supported and complemented these results. One protein encoded by the 12kb fragment was found not to be related to fimbriation but rather the product of the STla gene, identified as a component of a Tn1681-like transposon.

The 987P fimbrial gene cluster of enterotoxigenic Escherichia coli is plasmid encoded.

A clone containing the 987P fimbrial gene cluster was selected from a cosmid library of total DNA of the prototype Escherichia coli strain 987 by using 987P-specific antiserum. A subclone of 12 kilobases containing all of the genes required for fimbrial expression on a nonfimbriated K-12 strain of E. coli and a DNA fragment internal to the fimbrial subunit gene were used to probe the prototype strain and various isolates of 987P-fimbriated enterotoxigenic E. coli. All strains had several plasmids, as shown by agarose gel electrophoresis, and each of five strains which expressed 987P fimbriae showed a plasmid of 35 to 40 megadaltons (MDa) hybridizing to both 987P-specific probes. Hybridization to restricted DNA of strain 987 supported a plasmid origin for the cloned 987P gene cluster. Moreover, an isogenic strain which had lost its 35-MDa plasmid was no longer capable of synthesizing fimbrial subunits, but regained fimbrial expression after reintroduction of the TnphoA (Tn5 IS50L::phoA)-tagged 35-MDa plasmid. Absence of fimbrial subunit synthesis in K-12 strains transformed with the 35-MDa plasmid alone suggested the requirement of regulatory elements existing in strain 987 but missing in K-12 strains. A probe for the heat-stable enterotoxin STIa hybridized in each of the 987P-fimbriated strains to the plasmid containing the 987P genes and in most of these strains to an additional plasmid which contained the gene for the heat-stable enterotoxin STII. Occurrence of the 987P and STIa genes on the same replicon correlates with epidemiological observations, STIa being the most prevalent toxin produced by 987P-fimbriated E. coli.