Enterobacterial Common Antigen and Capsular Polysaccharides
Chapter
9
PAUL D. RICK and RICHARD P. SILVER
The structural organization of the cell envelope of gram-negative enteric bacteria has been established in considerable detail (Fig. 1). The major components of the outer membrane include the lipopolysaccharide (LPS, endotoxin) molecule that is characteristic of these organisms as well as phospholipids and proteins (see chapter 5 of this volume). It is important to note that in addition to LPS, the outer membrane of gram-negative enteric bacteria contains other glycolipids whose occurrence is frequently overlooked. For example, the cell surface of many of these organisms is covered by a capsular polysaccharide (225). The individual polysaccharide chains of certain capsular polysaccharide molecules, such as the group II K antigens of Escherichia coli, appear to be covalently linked to a diglyceride moiety through a phosphodiester linkage. The phosphoglyceride moieties appear to be integrated into the outer leaflet of the outer membrane, and they presumably serve to anchor the polysaccharide chains to the cell surface (225). Gram-negative enteric bacteria also possess a unique cell surface glycolipid, the enterobacterial common antigen (ECA). ECA is possessed by all members of the family Enterobacteriaceae. However, in spite of its unique and universal occurrence within this family, ECA has not been studied extensively. Indeed, although ECA was discovered over 30 years ago, we are just now beginning to gain an understanding of the genetics and biochemistry of its assembly. The following discussion focuses primarily on the current status of our knowledge concerning the structure, biosynthesis, and genetics of ECA. The reader is referred to earlier reviews (96, 113, 118) for more in-depth discussions of the immunogenicity and antigenicity of ECA, its biological and clinical significance, and the properties of other, related bacterial cell surface antigens.
ECA was discovered by Kunin et al. in 1962 while they were characterizing E. coli strains responsible for causing urinary tract infections in humans (98, 100). During the course of these investigations, the serological cross-reactivities of 145 E. coli O serotypes were analyzed by a passive hemagglutination assay. Accordingly, rabbit antisera prepared against heat-killed cells of each O serotype were examined for their ability to agglutinate erythrocytes sensitized with heterologous and homologous bacterial antigens present in cell extracts. A cross-reactive determinant was found in almost all of these O serotypes when certain heterologous antisera were used in the hemagglutination assay. For example, rabbit antisera prepared against heat-killed cells of E. coli O14 were able to agglutinate erythrocytes sensitized with extracts obtained from all other O serotypes. Preabsorption of the antisera with heterologous extracts removed the cross-reactive antibodies, but the preabsorbed antisera were still able to agglutinate erythrocytes sensitized with extracts obtained from E. coli O14 and were unrelated to this LPS. Additional experiments revealed that heat-killed cells of several other O serotypes either failed to elicit cross-reactive antibodies in rabbits or were only weakly immunogenic, even though cell extracts prepared from them contained the cross-reactive property (100).
The above observations led to the conclusion that E. coli and other members of the Enterobacteriaceae possess a "common antigen" that occurs in two forms. Accordingly, certain organisms appeared to possess a so-called immunogenic form capable of eliciting antibodies when heat-killed cells were used to immunize rabbits. In contrast, other bacteria appeared to possess a so-called haptenic form that was recognized by antibodies elicited against the immunogenic form but failed to function as an immunogen when whole heat-killed cells were injected into rabbits. The difference between the two forms was believed to be due to the occurrence of an additional structural component in the immunogenic form that allowed it to induce antibody formation. The "common antigen" is now called the ECA, and, as described below, the basis for the two forms of ECA is now understood.
Early studies revealed the presence of ECA in crude extracts obtained from other members of the Enterobacteriaceae including Salmonella, Shigella, Proteus, and Aerobacter species; however, it was not detected in extracts obtained from various nonenteric gram-negative bacteria and gram-positive bacteria (100). More recent investigations established the occurrence of ECA in all wild-type strains of gram-negative bacteria belonging to the Enterobacteriaceae (96, 118, 159). ECA has also been found in Plesiomonas shigelloides, which is considered a member of the Vibrionaceae (14, 95). The occurrence of ECA in P. shigelloides may be explained by observations that this organism is more closely related to Shigella sonnei than to members of the Vibrionaceae (96). Thus, the taxonomic placement of P. shigelloides within the Vibrionaceae may be in error. It is interesting that ECA has not been detected in Erwinia chrysanthemi, even though it appears to be present in all other Erwinia species; it has been suggested that this discrepancy may reflect the pronounced genetic heterogeneity that apparently exists within this genus (96). Trace amounts of ECA have also been detected in extracts of certain Actinobacillus species by immunoassay (22); however, the presence of ECA in these organisms could not be verified by chemical analysis (159).
Structure of ECA Heteropolysaccharide Chains.
The demonstration of ECA by immunologic methods was followed by numerous attempts in several laboratories to purify and structurally characterize ECA (61, 77, 99, 116, 120). These early attempts employed a variety of methods to purify ECA from both immunogenic and nonimmunogenic organisms, but they resulted in only crude preparations, and compositional analyses of these preparations varied considerably. Nevertheless, these studies provided important insights into the relationship between the so-called immunogenic and nonimmunogenic forms of ECA. More specifically, several of the studies reported that a portion of the ECA from immunogenic strains could not be separated from LPS regardless of the extraction procedures employed, whereas the remainder of the ECA appeared to be free of LPS. In contrast, all of the ECA from nonimmunogenic strains appeared to be LPS free. Männel and Mayer (115) inally succeeded in obtaining pure preparations of ECA from a nonimmunogenic wild-type strain of Salmonella montevideo as follows. They first obtained an extract highly enriched for ECA by using a combination of the phenol-water and phenol-chloroform-petroleum ether extraction procedures originally developed for the isolation of "smooth" and "rough" LPS, respectively. Then purified ECA was obtained from the extract following DEAE-cellulose chromatography. Chemical analyses of the ECA revealed the occurrence of equimolar amounts of N-acetyl-d-glucosamine (d-GlcNAc) and N-acetyl-d-mannosaminuronic acid (d-ManNAcA); these analyses also indicated that the GlcNAc and ManNAcA residues were linked by 1→ 4 linkages. In addition, phosphate, O-acetyl groups, and ester-linked palmitic acid were detected in the purified material. Accordingly, it was concluded that the nonimmunogenic form of ECA consisted of linear polysaccharide chains composed of the disaccharide repeat unit → 4)–d-GlcNAc–(1→ 4)-d-ManNAcA-(1→ as well as an undefined phospholipid component. Similar results were obtained with ECA isolated from Shigella sonnei (106).
Lugowski et al. subsequently (108) clarified the structure of the polysaccharide component of ECA by demonstrating that ECA obtained from Shigella sonnei phase I contained 4-acetamido-4,6-dideoxy-d-galactose (d-Fuc4NAc) in addition to GlcNAc and ManNAcA. Small amounts of fatty acids (C16:0 and C18:1) and phosphate were also detected. These investigators established that ECA polysaccharide chains are composed of the trisaccharide repeat unit →3)–d-Fuc4NAc–(1→4)–β–d-ManNAcA–(1→ 4)–α-d-GlcNAc(1→ . The anomeric configuration of the d-Fuc4NAc was not determined; however, optical rotation data indicate that this residue is α linked. 13C nuclear magnetic resonance spectra of ECA and O-deacetylated ECA also revealed that approximately 70% of the GlcNAc residues are O-acetylated at C-6. Subsequent analyses of ECA obtained from a variety of organisms are in agreement with the structure of the trisaccharide repeat unit described above.
Phosphoglyceride-Linked ECA.
The lipid component of nonimmunogenic ECA was identified as a diacylglycerol moiety linked to the polysaccharide chains through phosphodiester linkage (96, 97). Structural characterization of the phosphoglyceride aglycone revealed palmitic acid (C16:0) as the O-fatty acyl substituent in the 1 position and either palmitic, palmitoleic (C16:1), cis-vaccenic (C18:1), or stearic (C18:0) acid as the O-fatty acyl substituent in the 2 position. The occurrence of palmitic and stearic acids as O-acyl substituents in the 2 position is somewhat surprising, since this position is normally occupied by unsaturated fatty acyl groups in bacterial phospholipids. The nature of the linkage of the phosphoglyceride moiety to the polysaccharide chains was established by treatment of phosphoglyceride-linked ECA with phospholipase D. Thus, incubation of phosphoglyceride-linked ECA with phospholipase D resulted in the release of the phosphoglyceride moiety from the polysaccharide chain with the concomitant appearance of reducing terminal GlcNAc (95). Thus, the polysaccharide chains of the nonimmunogenic form of ECA appear to be linked to the phosphoglyceride moiety by phosphodiester linkage of the terminal reducing GlcNAc of the first trisaccharide repeat unit to the diglyceride. Accordingly, phosphoglyceride-linked ECA is commonly referred to as ECAPG (96).
Early studies proposed a relative molecular weight of approximately 3,000 for ECAPG (115). However, subsequent examination of ECAPG obtained from crude extracts of E. coli and Salmonella typhimurium (official designation, Salmonella enterica serovar Typhimurium)—either by Western immunoblot analysis with mouse anti-ECA monoclonal antibody (immunoglobulin G2a [IgG2a]) or by sodium dodacyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE)/fluorography in the case of radiolabeled ECAPG—revealed a population of molecules arranged in a ladder-like pattern typical of polymers that make up a homologous series of repeat units (13, 162). These analyses indicated that individual ECA chains have an apparent M r of approximately 12,000 to 35,000, which corresponds to molecules with apparent degrees of polymerization ranging from 18 to 55. The structure of ECAPG is shown in Fig. 2.
The presence of ECAPG in cell extracts is routinely detected by a variety of serological techniques including passive hemagglutination and enzyme-linked immunosorbent assay (113, 150). These assays rely on the ability of ECAPG to coat erythrocytes or the wells of microtiter plates, respectively, and this ability is dependent on the structural integrity of the phosphoglyceride aglycone. Thus, the loss of fatty acyl groups from ECAPG following treatment with either mild alkali, phospholipase A2, or lipases present in culture fluids of Pseudomonas aeruginosa precludes the detection of ECAPG by these assays (95, 96, 97). The phosphoglyceride aglycone also appears to be essential for the ability of ECAPG to migrate in SDS-PAGE gels, since mild-alkali-treated ECAPG cannot be detected on gels by immunoblotting (95, 96) or by fluorography in the case of [3H]GlcNAc-labeled preparations (unpublished results).
LPS-Linked ECA.
As mentioned above, immunogenic strains were found to possess ECAPG as well as another form of ECA, which appeared to be associated with LPS (119, 130, 197, 215). The nature of the LPS-associated form of ECA and its relationship to immunogenicity later emerged as a result of studies which demonstrated that only rough mutants of E. coli and Shigella spp. that possessed either a complete R1 or R4 LPS core were immunogenic, whereas similar rough mutants that possessed other core structures were nonimmunogenic (180, 219). Later studies revealed that a number of E. coli K-12 isolates possessing a complete LPS core region were also immunogenic (96). In addition, the function of the rfaL locus, which encodes the O-translocase responsible for the transfer of O side chains from the lipid carrier to the LPS core, was found to be required for immunogenicity (180). Kiss et al. (87) subsequently obtained evidence for the covalent linkage of ECA polysaccharide chains to the R1 core of E. coli F470; however, specific details concerning the nature of the linkage between ECA chains and the LPS core were not determined. The association of ECA with LPS has also been demonstrated by immunochemical techniques. Western blot analyses of LPS from immunogenic strains, using a mouse anti-ECA monoclonal antibody (IgG2a), revealed the ladder-like pattern characteristic of ECA polysaccharide chains (96). In contrast, these polymers were not detected in LPS isolated from nonimmunogenic strains that were analyzed in the same manner. The available data support the conclusion that the immunogenic form of ECA consists of ECA chains linked to a complete R1, R4, or K-12 core region lacking O side chains (ECALPS) (Fig. 2). As discussed below, synthesis of ECALPS involves transfer of ECA chains from a lipid carrier to the appropriate LPS core, and this process requires the O-translocase or O-antigen ligase encoded by the rfaL gene. It appears that lipid-linked ECA chains are rather poor substrates for the ligase; accordingly, the substitution of LPS cores does not appear to be stoichiometric, and it has been estimated that only 5% of such cores are substituted with ECA polysaccharide chains (96).
The absence of O side chains in immunogenic strains conflicts with the original discovery of the immunogenic form of ECA in E. coli O14:K7. However, subsequent analyses of the E. coli isolate used in these early studies mistakenly classified it as possessing the O14 O side chains when in fact it was a rough strain (rfb) containing a complete R4 outer-core region (178).
Although there are structural similarities between the R1 and R4 outer-core regions, the structure of the K-12 outer-core region differs from these core structures substantially (Fig. 3). Accordingly, it is not clear what structural feature(s) of the outer core is necessary for it to function as an acceptor of ECA chains. Progress in this area is also hampered by the fact that the nature of the linkages between ECA chains and R1, R4, or K-12 outer-core regions have not yet been established.
Cyclic ECA.
A cyclic form of ECA (ECACYC) consisting of four to six trisaccharide repeat units and lacking a phospholipid component has been isolated from Shigella sonnei phase I (33) and Yersinia pestis (215) (Fig. 2). ECACYC has also been reported to occur in S. monetevideo (96, 215). Although ECACYC from Y. pestis and Shigella sonnei are very similar in structure, the cyclic tetramer predominates in Y. pestis whereas the cyclic polymer from Shigella sonnei is more heterogeneous in size. In addition, the GlcNAc residues in ECACYC from Y. pestis are N- and O-acetylated to a lesser degree than is the case with ECACYC from Shigella sonnei. Both of these organisms also contain ECAPG.
It is important to stress that ECALPS occurs infrequently within the Enterobacteriaceae, which accounts for the relatively small number of immunogenic strains. In contrast, ECAPG is present in all gram-negative enteric species including those that possess ECALPS. The terms "immunogenic" and "nonimmunogenic" have classically been used to describe the ability of ECAPG-containing heat-killed cells that either possess or lack ECALPS, respectively, to elicit anti-ECA antibodies following their injection into rabbits. However, the results of studies using either live bacterial cells or various cell-free preparations of ECA for immunizations indicate that the immunogenicity of ECALPS and ECAPG is somewhat more complex. Thus, in contrast to the results obtained with heat-killed cells, when rabbits are immunized with living cells that possess only ECAPG, they react similarly to the way they react to viable cells that possess both ECALPS and ECAPG (219). In addition, purified ECA polysaccharide chains per se appear to be poor immunogens; rather, anti-ECA antibodies are elicited only in response to ECA polysaccharide chains that are associated with other molecules. Accordingly, the immunogenicity of ECALPS appears to be dependent on its association with LPS, since purified ECAPG does not function as an immunogen (96). Partially purified ECAPG is immunogenic, and the immunogenicity of these preparations appears to be related to the association of ECAPG with cellular proteins (94, 96). Indeed, ECAPG has been demonstrated to form complexes with a variety of basic and hydrophobic proteins (94). It is possible that such complexes are destroyed by heating, thus accounting for the inability of heat-killed cells of so-called nonimmunogenic strains to elicit anti-ECA antibodies in rabbits.
Immunization of rabbits with various preparations of ECA results primarily in the formation of IgM antibodies (164). However, conjugates of tetanus toxoid and de-O-acetylated ECA polysaccharide chains containing one to five trisaccharide repeat units were found to be highly immunogenic, and they elicited high titers of anti-ECA IgG antibodies when injected into rabbits (107). Both IgG and IgM mouse anti-ECA monoclonal antibodies have also been prepared (150), and in both cases the carboxyl group of ManNAcA appears to be involved in antibody recognition (95).
The occurrence and uniform distribution of ECALPS on the cell surface of immunogenic strains have been clearly demonstrated by a variety of immunological techniques including immunofluorescence and immunoelectron microscopy with either ECA-specific polyclonal antisera or mouse anti-ECA monoclonal antibody (3, 164). These techniques have been used to demonstrate that ECAPG is also localized in the outer leaflet of the outer membrane of nonimmunogenic strains (4, 164). Unlike ECALPS, the distribution of ECAPG in the outer membrane of nonimmunogenic strains is not uniform, and it appears to be localized to distinct regions or patches (3, 5, 164). Furthermore, the amount of ECAPG in the outer membrane of nonimmunogenic strains appears to be smaller than the amount of ECALPS present in genetically related immunogenic strains. Indeed, such differences have been observed in closely related E. coli mutants that differ primarily in their ability to transfer ECA chains to the R1 core. Thus, the amount of ECAPG in the outer membrane of an E. coli mutant able to synthesize only ECAPG (nonimmunogenic) was found to be smaller than that present in the outer membrane of a closely related E. coli strain able to synthesize both ECAPG and ECALPS (immunogenic) (164). Since both strains lacked O side chains and capsules, the observed difference does not appear to be a reflection of the decreased accessibility of ECAPG to anti-ECA antibodies that occurs in strains that possess these polymers (5).
It is interesting that ECA has also been detected by immunoelectron microscopy in ribosome-containing areas of the cytoplasm of E. coli and Y. enterocolitica (3). It is tempting to speculate that this observation reflects the cellular localization of ECACYC because of the hydrophilic nature of this form. However, ECACYC has not been shown to occur in E. coli and Y. enterocolitica. The cellular localization of ECACYC in organisms that contain it has not been established.
The ECA content of organisms appears to vary quite significantly from one species to another (96). Analyses of the ECA content of S. montevideo indicate that ECAPG accounts for approximately 0.3% of the dry weight of cells (115). It has been reported that S. montevideo may also contain ECACYC (96, 215), but the content of ECACYC has not been determined. Significant variations in the cellular content of ECA have also been observed within a single species (12, 96).
Although many of the steps involved in the biosynthesis of the ECA polysaccharide chain remain to be established, the mechanism of synthesis of the trisaccharide repeat unit of ECA polysaccharide chains in E. coli and S. typhimurium is known in considerable detail (11, 12, 162, 163). Both in vitro and in vivo studies have established that the repeat unit is assembled as a lipid-linked intermediate (lipid III) in a series of reactions that involve the polyisoprenoid lipid undecaprenyl monophosphate (Fig. 4). The initial step in the synthesis of the repeat unit is the transfer of GlcNAc-1-phosphate from UDP-GlcNAc to undecaprenyl monophosphate to yield GlcNAc-pyrophosphorylundecaprenol (lipid I). The mechanism of lipid I synthesis appears to be the same as that involved in the synthesis of GlcNAc-pyrophosphoryldolichol, the first intermediate in the synthesis of the asparagine-linked oligosaccharides of glycoproteins in eukaryotic cells (1, 90). In vitro and in vivo studies have clearly established that the E. coli and S. typhimurium UDP-GlcNAc:undecaprenylphosphate GlcNAc-1-phosphate transferases are extremely sensitive to the antibiotic tunicamycin (11, 162), which specifically inhibits enzymes that catalyze the transfer of GlcNAc-1-phosphate from UDP-GlcNAc to polyprenylphosphate acceptors (62, 110, 199, 201). The in vivo synthesis of ECA is totally abolished when tunicamycin is added to cultures to a final concentration of 3 μg/ml (162). In this regard, it is interesting that tunicamycin does not inhibit the in vivo synthesis of ECA in cells that possess a complete LPS outer-core region (162). The reason for this is unknown; however, it is possible that the inability to synthesize a complete outer-core structure is accompanied by changes in the composition and structure of the outer membrane which may increase the permeability of the outer membrane to tunicamycin. Indeed, this general mechanism is responsible for the increased sensitivity of deep-rough mutants to a variety of hydrophobic antibiotics and other agents (131).
The synthesis of lipid I is followed by the successive incorporation of ManNAcA and Fuc4NAc from the donors UDP-ManNAcA and TDP-Fuc4NAc, to form lipid II and lipid III, respectively (Fig. 4) (11, 12). Additional steps involve the elongation of ECA chains and their subsequent transfer to appropriate acceptors to form ECAPG and ECALPS or, alternatively, the cyclization of the polysaccharide chain to form ECACYC. The in vitro synthesis of ECA polysaccharide chains has been demonstrated by using isolated membranes as well as ether-treated cells (11). However, the mechanism of chain elongation remains unclear. Experiments with S. typhimurium Δ rfbA mutants suggest that chain elongation may occur as a result of the polymerization of trisaccharide repeat units by a mechanism similar to that described for the synthesis of O antigen in S. newington (11). Accordingly, S. typhimurium Δ rfbA mutants accumulate lipid II as a consequence of their inability to synthesize TDP-glucose, an obligatory precursor of TDP-Fuc4NAc (163). Addition of TDP-[3H]Fuc4NAc to ether-permeabilized cells of the mutants results in the incorporation of isotope into ECA polymers. This synthesis does not require an exogenous supply of UDP-GlcNAc or UDP-ManNAcA, since nucleotide sugars and other water-soluble metabolites are released during ether treatment (127, 216). The exogenously supplied TDP-Fuc4NAc presumably allows the conversion of accumulated lipid II to lipid III, and the lipid III is utilized for the subsequent polymerization of trisaccharide repeat units.
In vitro synthesis of ECA polymers has also been demonstrated by using E. coli membrane preparations obtained from cells not defective in ECA synthesis (163). Incubation of membranes with UDP-[3H]GlcNAc, UDP-ManNAcA, and TDP-Fuc4NAc resulted in the tunicamycin-sensitive incorporation of isotope into ECA polymers. However, unlike polymer synthesis in ether-treated cells of S. typhimurium Δ rfbA mutants, in vitro synthesis of radiolabeled polymer in membranes depends on the presence of UDP-ManNAcA and TDP-Fuc4NAc as well as UDP-[3H]GlcNAc in reaction mixtures. The requirement for all three nucleotide sugars most probably reflects the necessity for de novo synthesis of lipid III as a prerequisite for the polymerization of trisaccharide repeat units rather than the alternative mechanism of chain elongation, namely, the successive transfer of sugars to the nonreducing terminus of nascent polymer. The latter mechanism has been proposed for the growth of E. coli O8 and O9 side chains (70).
As stated above, the structural features of the R1, R4, and K-12 LPS core regions that enable them to serve as acceptors of ECA chains are not known. In addition, the acceptor of ECA chains for the synthesis of ECAPG remains to be determined. It is possible that the synthesis of ECAPG involves the transfer of ECA chains from lipid carrier to diglyceride; in this event, the transfer of polysaccharide-1-phosphate moieties to diglyceride acceptor would occur. Alternatively, the synthesis of ECAPG may involve the transfer of free polysaccharide chains from lipid carrier to phosphatidic acid. Nothing is known concerning the steps involved in ECACYC synthesis.
The mechanism involved in the translocation of ECAPG from the inner membrane to the outer membrane remains to be determined. However, it must be noted that although it seems likely that ECA chains are assembled at the inner or cytoplasmic membrane, there are no data to substantiate this assumption.
rfe
and rff Genes.
The genetic determinants of ECA biosynthesis include genes in the rfe-rff cluster located at min 83 and 85 of the S. typhimurium and E. coli chromosomes, respectively (111, 112, 122, 181). Nucleotide sequencing of the E. coli chromosome from 84.5 to 86.5 min has revealed that the rfe-rff gene cluster contains 11 open reading frames (32), and biochemical and genetic studies have thus far identified seven genes within this cluster that function in ECA synthesis (Fig. 5) (121, 123). The partial gene order of rfe-rffD/rffE-rffC-rffA-rffT-rffM has been established.
Both biochemical and genetic data have established that the rfe gene product is the tunicamycin-sensitive UDP-GlcNAc:undecaprenylphosphate GlcNAc-1-phosphate transferase that catalyzes the synthesis of lipid I (121, 133). The deduced amino acid sequence of Rfe indicates that it is a hydrophobic protein. The translational start for Rfe has not been determined directly; therefore, the molecular size of Rfe has been estimated to be either 29 kDa (121, 133) or 41 kDa (32), depending on the choice of possible translational initiation codons.
The rfe gene is also important for the synthesis of many O antigens; several of these are cited in Table 1. The available data support the conclusion that synthesis of an O antigen with a GlcNAc at the potential reducing terminus of the repeat unit is rfe dependent; this is in agreement with the GlcNAc-1-phosphate transferase function of Rfe. The structures of many of the O repeat units in Table 1 satisfy this requirement. In contrast, the role of rfe in the synthesis of several other rfe-dependent O antigens is unclear, since their respective O repeat units either lack GlcNAc completely or contain GlcNAc in a position that is incompatible with a requirement for lipid I in their biosynthetic pathway.
Table 1Structures of rfe-dependent O-antigen repeat unitsa |
The results of recent studies have provided an explanation for the rfe-dependent synthesis of the O8 antigen (161). These studies revealed that the synthesis of O8 side chains is tunicamycin sensitive; furthermore, lipid-linked O8 side chains have a single GlcNAc at the potential reducing terminus. Jann et al. (70) reported that the mechanism of O8 chain elongation is similar to that observed for the structurally related O9 antigen. Chain elongation of these mannans does not involve the polymerization of individual lipid-linked repeat units. Rather, growth of the mannan chain occurs by the sequential transfer of mannose residues from GDP-mannose to the nonreducing terminus of nascent chains. Accordingly, it appears that GlcNAc-pyrophosphorylundecaprenol functions as the initial acceptor of mannose residues in this process. The GlcNAc is presumably transferred to the LPS core along with the mannan chain; however, this has not been demonstrated. In any event, it seems likely that GlcNAc-pyrophosphorylundecaprenol may play a similar role in the rfe-dependent synthesis of the O9 antigen and other O antigens in which the role of rfe has been enigmatic.
The rffE and rffD genes are involved in the synthesis of UDP-ManNAcA, and they encode the enzymes UDP-GlcNAc-2-epimerase and UDP-ManNAc dehydrogenase, respectively (104). However, identification of the specific open reading frames which function as the rffE and rffD loci has not been unequivocally determined. Mutants possessing a defective rffC gene accumulate lipid III (121, 123). Although the specific role of rffC in ECA synthesis has not yet been determined, this observation suggests that rffC plays a role in the elongation ECA polysaccharide chains. However, as stated above, the mechanism of chain elongation remains to be determined. The rffA gene codes for the transaminase that catalyzes the conversion of TDP-4-keto-6-deoxy-d-glucose to TDP-d-fucosamine, the immediate precursor of TDP-Fuc4NAc. The rffT and rffM genes encode the transferases responsible for the transfer of ManNAcA from UDP-ManNAcA to lipid I and the transfer of Fuc4NAc from TDP-Fuc4NAc to lipid II, respectively.
Additional Genes Involved in Assembly.
In S. typhimurium (group B), genes in the his-linked rfb region are also required for ECA synthesis (112, 114). The rfbA and rfbB genes code for TDP-glucose pyrophosphorylase and TDP-glucose oxidoreductase, respectively. These enzymes are necessary for the synthesis of TDP-4-keto-6-deoxy-d-glucose, which is a precursor of TDP-Fuc4NAc and TDP-rhamnose, the donors of Fuc4NAc and rhamnose for ECA and group B O-antigen synthesis, respectively. S. typhimurium Δ rfbA mutants accumulate lipid II, which appears to render them hypersensitive to SDS (163). In addition, these mutants synthesize only trace amounts of ECA (114). Accordingly, it has been suggested that UDP-glucose pyrophosphorylase catalyzes the synthesis of sufficient amounts of TDP-glucose to support the synthesis of trace amounts of ECA (103). In salmonellae belonging to serotypes that do not contain rhamnose as a component of their O antigens, the genes responsible for TDP-Fuc4NAc synthesis appear to be located in the rff gene cluster (103).
The translocation of ECA polysaccharide chains from the lipid carrier to the LPS core region to form ECALPS requires an O-translocase (or ligase), which is encoded by the rfaL gene (96, 180). Thus, rfaL appears to be required for translocation of both ECA and O side chains to the complete LPS core. However, the actual transfer of ECA chains from lipid carrier to complete R1, R4, or K-12 core acceptors has not yet been experimentally demonstrated.
Much is known about the synthesis of ECA, but there are several aspects for which there is essentially no information. For example, the utilization of lipid-linked ECA chains for the synthesis of both ECAPG and ECACYC involves as yet unidentified mechanisms; accordingly, the genes involved in these processes remain to be established. Furthermore, nothing is known concerning either the possible regulation of ECA synthesis or the mechanism involved in the translocation of ECAPG from the cytoplasmic membrane to the outer membrane. Finally, the function(s) of ECA is unknown. The growth and physiology of mutants unable to synthesize ECA appear unaltered when the mutants are grown in culture. A role for ECA in the virulence of salmonellae toward mice was suggested by observations that ECA-negative mutants of S. typhimurium were found to be approximately 10-fold less virulent than the corresponding ECA-positive strains (207). However, subsequent studies indicated that the apparent greater virulence of ECA-positive strains was due to their increased survival in mice, and this could be only partially attributed to their ECA-positive character (132). In addition, ECA does not possess endotoxin-like activity (118), and passive immunization of mice with anti-ECA antibodies does not protect against salmonellosis (175). Thus, ECA does not appear to be a significant determinant of virulence. Nevertheless, the striking occurrence of ECA in all members of the Enterobacteriaceae and its absence in all other bacteria suggest that it serves a function unique to these organisms.
E. coli isolated from natural populations is an antigenically complex species. The full extent of this complexity was revealed by the classic studies of Kauffman and his colleagues more than 50 years ago (81, 206). They provided order based on the distribution of the three major surface antigens: the LPS somatic (O), flagellar (H), and capsular (K) antigens. These antigens occur in various combinations and provide the basis for the serological differentiation of E. coli (82). The definition of the O, K, and H antigens was based on bacterial agglutination in absorbed and nonabsorbed rabbit antisera and on the influence of heating the bacteria on agglutination and immunogenicity (82). To date, 173 distinct O antigens, 80 K antigens, and 56 H antigens have been described (134, 135).
The capsular K antigens (from the German word Kapsel, meaning capsule) of E. coli are acidic polymers that surround the bacterial cell. They consist of oligosaccharide repeating units that differ in constitutents, branching, and charge density (140, 225). Historically, the presence of K antigens was determined by means of bacterial agglutination tests; an E. coli strain that was inagglutinable by O antiserum but became agglutinable when the cells were heated was determined to have a K antigen (81). Classification of K antigens is now based on electrophoretic mobility, reflecting differences in charge and molecular size (134). A few K antigens can also be typed by capsule-specific bacteriophage (59, 134). This section is not intended to include a comprehensive review of the capsules of E. coli but provides an overview of the structural features, biosynthesis, genetics, and biological function of E. coli capsules. For a more detailed analysis of the subject, the reader is referred to the excellent review by Whitfield and Valvano (225) as well as several earlier reviews (24, 25, 68, 72, 73, 76, 165, 187).
The K antigens of E. coli were classified into two distinct groups, termed I and II, on the basis of physical, chemical, and genetic criteria (68, 72, 73). The group I K antigens, which are characterized by a high molecular mass (>100 kDa) and low charge density, have been further subdivided into polymers that contain amino sugars (group IB) and those that do not (group IA) (225). The latter resemble the capsules of Klebsiella spp. Group I K antigens are expressed at all growth temperatures. Moreover, coexpression of the group I K antigens and LPS is restricted to relatively few LPS serotypes, primarily O8, O9, and O20 (68, 73).
Many isolates of E. coli also synthesize the slime polysaccharide colanic acid (M antigen). Colanic acid and group IA K antigens have several features in common: their chemical structures are similar; their genetic determinants are allelic; and they share a common regulatory system (225). Moreover, colanic acid can be produced simultaneously with group IB and group II K antigens but not group IA K antigens (83). These observations are consistent with the view that colanic acid and group IA capsules are part of the same group (225).
Group II K antigens, in contrast to group I, have a lower molecular mass (∼50 kDa) and a higher charge density, are coexpressed with a variety of LPS serotypes, and are expressed only at physiological temperatures (21, 68, 73, 138). They are similar to capsules produced by Haemophilus influenzae and Neisseria meningitidis. Group II K antigens contain a variety of acidic components, including N-acetylneuraminic acid (sialic acid), 2-keto-3-deoxymanno-octonic acid (KDO), glucuronic acid, N-acetylmannosamine, and phosphate (68, 76, 140, 165, 225).
Sialic acids are essential constituents of many mammalian glycoconjugates, displaying a variety of biological functions, but are relatively rare among prokaryotes (171, 176). Sialic acids are, however, common components of capsular polysaccharides (68, 76, 165). The group II K1 capsular antigen is an α2-8-linked polymer of sialic acid (polysialic acid [PSA]) (reviewed in references 187 and 202). In addition to E. coli K1, α2-8 PSA is produced by N. meningitidis group B, the opportunistic pathogen Moraxella nonliquefaciens, and Pasteurella haemolytica A2, which causes pasteurellosis in preweaned lambs (6, 35, 80). In addition, α2-9-linked PSA is expressed by N. meningitidis group C, while an α2-8, α2-9 PSA copolymer is produced by E. coli K92 (reviewed in reference 202).
The addition of O-acetyl residues and other side groups can affect the serological classification and antigenicity of E. coli K antigens. For example, the antigenicity of the K1 capsule can be modulated by O-acetylation of sialic acid residues (137). The O-acetyltransferase responsible for polysialic acid acetylation has been partially purified and shown to use acetyl coenzyme A as an acetyl group donor and require continued binding of a sialic acid oligomer of more than 14 residues (63). Interestingly, acetylation of the K1 polymer can undergo a cyclical transition between acetylated and unacetylated states reminiscent of flagellar and other cell surface antigen variations (137).
E. coli group II K antigens terminate in a phosphodiester linkage to 1,2-diacylglycerol (57, 177). The termination of group II K antigens with phospholipid seems to be a general structural feature of group II K antigens as well as meningococcal and Haemophilus capsular polysaccharides (55, 68, 101, 117). Some group I polymers carry a lipid A substitution at the reducing end of the polymer (69, 223). The biological role of the lipid substitution is difficult to assess, although it has been speculated that lipids may serve to anchor the polysaccharide to the bacterial cell surface. This may be difficult to reconcile with the extreme lability of the polysaccharide lipid linkage and the observation that not all molecules carry the lipid substitution (57, 177). Alternatively, the lipid-substituted molecule may represent an intermediate in polymer synthesis. The glycerophosphatidyl substitution of the K5 polymer appears to be linked to the polysacharide by a KDO residue (48). KDO moieties have been detected at the reducing terminus of several group II polymers in which KDO is not part of the repeating unit (177). This correlates with the observation that strains synthesizing the group II antigens have increased levels of CMP-KDO synthetase activity when cells are grown at 37°C (49) and led Finke et al. (50) to propose that CMP-KDO synthetase may play a role in synthesis of group II K antigens.
The proteins necessary for synthesis and expression of the group I and II capsules are chromosomally encoded and are nonallelic (140). Genes for the biosynthesis and expression of the group IA antigens, and presumably the group IB antigens as well, are located near the his-linked rfb locus (74, 102, 140). A trp-linked locus (rfc) was postulated to also be involved in expression of the K27 polysaccharide (179). The genetic determinants for the group II K antigens are located near serA on the E. coli genetic map (23, 139, 141, 212).
Recent genetic and biochemical data indicate that several E. coli capsules, previously thought to be group II antigens, resemble more closely those of group I and perhaps represent a third group of E. coli K antigens, designated I/II (41). For instance, strains expressing the K2, K3, K10, K11, K19, and K54 capsules do not have elevated levels of CMP-KDO synthetase at 37°C, characteristic of most group II capsular strains (49). In addition, the K3, K10, K11, K54, and K98 antigens do not exhibit temperature-dependent regulation (138). The group I/II capsule genes, like group II, map near serA, but Southern blot analysis indicates they lack the group II capsule genes (41). These observations support the notion of three capsule gene clusters among E. coli isolates, two of which map near serA (41).
The biosynthesis of E. coli capsules is a complex process that involves the synthesis and polymerization of repeating units into a large polymer which must be transported through the lipid bilayer and anchored to the cell surface. There is little information on the biosynthetic steps involved in synthesis of the group I K antigens, although it has been speculated that the mechanism is similar to that described for the capsules of K. aerogenes (225). Capsules are polymerized in K. aerogenes by using undecaprenol-linked intermediates and a blockwise assembly mechanism (225). In contrast to the group I K antigens, the synthesis and expression of the group II capsules has been studied in great detail.
kps
Gene Cluster.
The kps cluster was first cloned from E. coli synthesizing the K1 PSA capsule (185). The kps gene cluster for E. coli K1 is 17 kb (26, 213). Subsequently, allelic kps loci were isolated from several other group II serotypes (42, 167). The genetic organization and most of the nucleotide sequence of the K1 and K5 kps clusters, the best-studied group II K antigens, have been determined (Fig. 6).
The kps gene cluster is functionally divided into three regions (26, 168) (Fig. 6). The central region, region 2, contains a biosynthetic cassette that is flanked on either side by genes that function in more general aspects of capsule biosynthesis. For the most part, cells with mutations in region 2 fail to produce polysaccharide (9, 25, 213). This region is unique for each E. coli serotype synthesizing a chemically distinct capsular polysaccharide (24, 25, 168). In contrast to region 2, genes in the flanking regions, regions 1 and 3, are conserved among E. coli species synthesizing serologically distinct capsules (25, 168, 183). Cells with mutations either in region 1 or region 3 tend to accumulate intracellular polysaccharide, suggesting that they are important for transport of polysaccharide to the cell surface (27, 48, 145, 213, 227). An additional phenotype of cells harboring mutations in either region 3 or region 1 is the reduction of endogenous glycosyltransferase activity (28, 213). Endogenous activity is measured by the transfer of labeled sugar onto preexisting acceptors within the membranes, while exogenous activity is measured by transfer of labeled sugar onto exogenously added polymer (203). The reason for reduced endogenous activity in region 1 or 3 mutants is not apparent, since the exogenous activity in these cells is normal, indicating that the transferase enzymes are not affected (213). The current view of capsule expression and transport assumes that the membrane-bound components of the cluster form a hetero-oligomeric complex involved with polymer synthesis and transport. The decrease in endogenous transferase activity, however, does not appear to be due to the lack of a component from such a complex but is more likely to be related to the inability of cells to transport polymer (145). If one assumes coupling between the biosynthetic machinery and the transport machinery, it is possible that a block in export in some way affects the nature or number of endogenous acceptors. In any event, these observations highlight the complexity of polymer synthesis and export and underscore the need to determine the nature of the endogenous acceptor within the cytoplasmic membrane.
Region 2: Biosynthetic Cassette.
The genes in region 2 contain information for specific sugar synthesis, activation, and polymerization functions (9, 25, 210, 213). The number of genes correlates with the size and complexity of the polymer repeating unit (, 25, 168). The genetic organization of region 2 and the steps involved in the biosynthetic process are best understood for the PSA capsule of E. coli K1 (187, 202, 214, 225). Sialic acid is synthesized in the cytoplasm of K1 organisms in a condensation reaction between N-acetylmannosamine and phosphenolpyruvate. The resulting monosaccharide is activated to a nucleotide derivative, CMP-NeuNAc, which is subsequently used as the sialic acid donor during polymerization (202). Polymerization is catalyzed by a sialyltransferase and proceeds processively at the nonreducing terminus by addition of activated NeuNAc, molecules to the growing polysaccharide chain (192, 202, 214). The functional domain of the sialyltransferase enzyme is located on the cytoplasmic surface of the inner membrane (67, 202). The growing polymer must therefore traverse the cytoplasmic membrane before being transported to the outer surface of the bacterial cell. Mature, surface-associated capsular polysaccharide is approximately 200 residues in length and may be inserted into the outer membrane via the phosphatidic acid moiety present at the reducing end of the polymer (202).
The polysialyltransferase enzyme cannot initiate polymer synthesis (192), and the mechanism responsible for chain initiation is not fully understood. Initiation is thought, however, to involve the attachment of polysialosyl monomers to an endogenous acceptor in the membrane (202). A molecule other than sialic acid is believed to serve as endogenous acceptor (169, 170, 202, 218) although the molecular nature of the acceptor moiety is not known. Polymerization of PSA does involve a lipid intermediate, NeuNAc-P-undecaprenol (204), but the amount of polymerization that occurs on the lipid carrier is uncertain. Full polymerization may occur on the lipid, with subsequent transfer of the mature chain to the acceptor, or the lipid may serve as an intermediate carrier of sialyl residues. Another intermediate consisting of a 20-kDa protein linked to PSA has also been detected, suggesting that a protein may play a role in synthesis or transport by functioning as an acceptor of partially or fully polymerized sialosyl chains (218). In contrast to synthesis of the K1 antigen, undecaprenol-linked intermediates were not detected in the biosynthesis of the K5 polymer (48).
Region 2 of the kps gene cluster from E. coli K1 is 5.7 kb long and encodes six proteins, NeuD, NeuB, NeuA, NeuC, NeuE, and NeuS (Fig. 6). The function of NeuD, encoded by the first gene in region 2, is not known, although NeuD is not required for sialic acid synthesis or activation (9, 210). NeuD shows significant similarity to a family of bacterial acetyltransferases (9) that includes LacA and CysE of E. coli, NodL of Rhizobium leguminosarum (40), and the Tn2424 Cat protein (144). The proteins in this family also share a novel region of homology with other acyl- and acetyltransferases, including the FirA (LpxD) and LpxA proteins, that has been termed the isoleucine patch (36, 205). It is unlikely that neuD encodes the O-acetyltransferase responsible for sialic acid acetylation (63), since cells harboring mutations in neuD have an acapsular phenotype and do not synthesize detectable polymer (9). In addition, only unacetylated polymer was detected in E. coli K-12 cells harboring the wild-type kps gene cluster (184, 185).
Two genes, neuB and neuC, are involved in sialic acid synthesis (9, 186, 210, 213, 230). Cells harboring mutations in either gene are defective in capsule synthesis but synthesize capsular polysaccharide when exogenous sialic acid is provided (186, 213, 230). NeuB has recently been shown to encode sialic acid synthase, the enzyme that catalyzes the condensation of N-acetylmannosamine and phosphoenolpyruvate to form sialic acid (9). NeuC is postulated to be involved in production of N-acetylmannosamine (198). Both NeuB and NeuC shows significant sequence homology with products involved in sialic acid synthesis that are encoded by the PSA gene cluster of N. meningitidis group B (9, 44, 56, 198). The 48.6-kDa product of the neuA gene, CMP-NeuNAc synthetase, catalyzes the activation of NeuNAc to the sugar nucleotide and has been purified to homogeneity (209, 231). The neuE gene is unique among region 2 genes in that cells with a mutation in neuE are still capable of producing intracellular polymer, leading to speculation that NeuE may be involved in coupling polymerization with transport (214). The 47-kDa product of the neuS gene has been identified as the sialyltransferase enzyme required for polymerization of NeuNAc (192, 193, 217). The polysialyltransferase enzyme (polyST) encoded by neuS elongates polysialosyl polymers within the membrane and can also transfer sialic acid residues from CMP-sialic acid to exogenously added PSA (192, 193, 203, 214).
Region 3: KpsMT Transporter.
Region 3 encodes two proteins, KpsM and KpsT (146, 189). KpsM is a 29-kDa hydrophobic integral membrane protein, while KpsT is a 25-kDa hydrophilic peripheral inner membrane protein that contains a consensus ATP-binding site (146, 189). KpsM and KpsT were proposed to function together as an ATP-dependent system involved in transporting capsular polysaccharide across the inner membrane via energy obtained from ATP hydrolysis (146, 189). They belong to the ATP-binding cassette superfamily of transporters (also referred to as traffic ATPases) (8, 20, 38, 46, 64). The members of the family have a common organization consisting of a hydrophobic membrane-associated component and a hydrophilic ATP-binding component (8, 64). Two of each type of component are believed to be required for the assembly of a functional transporter (8, 64). There are now more than 40 identified bac-terial exporters that transport proteins, peptides, and nonprotein substrates such as polysaccharides (46). The KpsMT transporter is the prototypical exporter of capsular polysaccharides in bacteria (46). BexAB and CtrCD, proteins homologous to KpsMT, were identified in H. influenzae type b and N. meningitidis group B, respectively, and are postulated to play analogous roles in transport of polysaccharide in both organisms (54, 91).
Electron microscopy was used to examine the location of polymer in cells carrying mutations in region 3 of E. coli K1 and K5 (92, 93, 145, 148). These studies confirmed that cells with mutations in either kpsM or kpsT accumulate intracellular polymer in the cytoplasm. Moreover, these mutants also accumulated polymer at discrete sites around the periphery of the cell (145). This material was reported to be shorter than surface polysaccharide, and it lacked the phosphatidic acid substitution (92). Studies using the energy-uncoupling agent carbonyl cyanide m-chlorophenylhydrazone also showed polysaccharide only in the cytoplasm of wild-type K5 organisms (93). These observations are consistent with the view that region 3 products function in transport of polymer across the inner membrane in an energy-dependent process.
Pigeon and Silver (153) studied the topology of the KpsM protein within the cytoplasmic membrane by using β-lactamase fusions and alkaline phosphatase sandwich fusions. Their analysis provides evidence for a model of KpsM having six membrane-spanning regions, with the N- and C-terminal domains facing the cytoplasm and with a short domain within the third periplasmic loop, referred to as the SV-SVI linker, localized in the membrane (153). Protease digestion studies were consistent with regions of KpsM exposed to the periplasmic space, and in vivo cross-linking studies provided support for the dimerization of KpsM within the cytoplasmic membrane (153). Linker insertion and site-directed mutagenesis defined the N terminus, the first cytoplasmic loop, and the SV-SVI linker as regions that are important for the function of KpsM in polymer transport (153).
Results obtained from site-directed and saturation mutagenesis of the ATP-binding consensus sequence of KpsT are consistent with the view that ATP binding and, presumably, hydrolysis are important to KpsT function and capsule transport (145). In addition, Pavelka et al. (145) obtained biochemical evidence in 8-azido-ATP photolabeling assays that KpsT binds ATP. They also proposed a secondary- and tertiary-structure model of the KpsT protein based on a model developed for other ATP-binding cassette transporter proteins (126, 145). The results obtained from chemical mutagenesis of KpsT are consistent with the model and revealed characteristics particular to the capsule transporters (145).
The kpsM and kpsT genes from both the E. coli K5 and K1 serotypes have been cloned and sequenced (146, 189). The kpsM gene products are essentially identical. In contrast, the kspT gene products are 72% identical and 84% similar (146). The two proteins are very similar at their amino termini, but the homology decreases at their carboxy termini (146). The K5 protein is 5 amino acids longer at the C terminus than the K1 protein is (145). The 3' ends of the genes are located at the junction between region 3 and region 2 of the gene cluster, an area that appears to have been the site of DNA recombination (24, 168). The proximity of kpsT to the junction may explain the divergence in the C termini of the proteins.
Region 1 Genes.
Region 1 of the kps gene cluster encodes six proteins, KpsF, KpsE, KpsD, KpsU, KpsC, and KpsS, that are involved in transport of polymer to the cell surface (Fig. 6) (27, 29, 147, 213, 227). KpsD is a 60-kDa protein localized in the periplasmic space (183). kpsD mutants and kpsD kpsE double mutants accumulate intracellular polymer (27, 227). Biochemical studies and electron microscopy indicated that polysaccharide accumulated in the periplasmic space of mutant cells (27, 227). Polymer was apparently full length and carried a phosphatidic acid substitution at the reducing end (27, 92). Not all mutations in region 1, however, result in the periplasmic accumulation of polymer. Mutations in kpsS and kpsC resulted in the accumulation of polysaccharide in the cytoplasm of mutant K5 organisms (27). The functions of KpsC and KpsS are unknown, but they may be involved in the coupling of the biosynthetic and transport machinery resulting in the assembly of polysaccharide into a transport-competent form. Cells harboring an insertion mutation in kpsF also accumulate polymer in the cytoplasm of mutant cells (29). The kpsF gene encodes a 35-kDa protein with significant homology to GutQ, an ATP-binding protein encoded by gutQ in the glucitol utilization operon of E. coli K-12 (29). kpsF appears to be the first gene in region 1, and an insertion mutation might exert polar effects on region 1 gene expression. The product of the kpsU gene is CMP-KDO synthetase, which explains the increased CMP-KDO synthetase activity in group II isolates (27, 49). KpsU, however, does not appear essential for polymer transport (27).
Newly synthesized polymer seems to appear as patches that are randomly distributed on the cell surface (93). The polysaccharide eventually covers the entire surface, but the exact mechanism of polymer translocation and growth is not known. However, the isolation of polysaccharide from the periplasm of cells harboring mutations in region 1 does not necessarily imply that polysaccharide passses through the periplasm during the translocation process. It is possible that region 1 gene products somehow directly connect the transport machinery of the inner membrane with a protein in the outer membrane, allowing the polysaccharide to bypass the periplasmic space. Interestingly, expression of capsules requires a functional porin in the outer membrane (J. Foulds and W. Aaronson, Abstr. Annu. Meet. Am. Soc. Microbiol. 1984, abstr. D-21, 1984). A particular porin, protein K, has been correlated with capsular polysaccharide production in most encapsulated strains of E. coli (142, 196, 221) and there appears to be a temporal correlation between the insertion of protein K into membranes and the expression of polymer on the cell surface (226). It has also been postulated that the translocation of polymer occurs at sites of adhesion between the inner and outer membranes (Bayer junctions) (15, 16, 17, 93). Although the existence of adhesion zones has been challenged (85), studies report the isolation of membrane fractions containing endogenous sialyltransferase activity apparently enriched in adhesion sites (220). Membranes from E. coli K1 grown at 15°C that are unable to synthesize polymer have been shown to undergo a time-dependent activation of endogenous polymer synthesis at 33°C (170). Activation was localized to a low-density vesicle fraction, and its presence coupled obligatorily to protein synthesis (220, 224). These low-density vesicles contain proteins found in both the inner and outer membranes and have been interpreted as representing zones of adhesion between the inner and outer membranes (220).
The current view of capsule synthesis in E. coli postulates that all required components for biosynthesis and transport are produced in equimolar amounts and are assembled at a defined locus as part of a multicomponent complex. It is therefore important to determine how the cell coordinates synthesis and expression of the polypeptides that function as part of such a complex. It is also important to focus on how changes in the environment elicit modulations in capsule gene expression. Encapsulated E. coli and other enteric pathogens must adapt to different conditions while moving from a free-living state to a host-associated state and must also respond quickly to changing host microenvironments encountered in vivo. For example, capsule expression may be down-regulated in the intestinal tract, facilitating the interaction of E. coli with epithelial cells, while it is up-regulated in the blood, where it is known that capsules are important for the survival of the organism. While there is no single mechanism by which these goals are achieved in bacteria, several themes are repeated (124). Histone-like proteins of bacteria have been implicated in regulation of a number of virulence genes, especially genes that are up-regulated in response to temperature shifts from 25 to 37°C (39). There are also several reports describing genes encoding bacterial exopolysaccharide antigens that are regulated by two-component sensory systems (34, 58, 152).
Expression of some group I K antigens is regulated by the rcs (regulator of capsule synthesis) system (75, 84, 225) described by Gottesman and her colleagues for the transcriptional regulation of colanic acid in E. coli K-12 (reviewed in reference 58). This capsule is made in small amounts by most strains, but certain conditions, such as growth at temperatures below 25°C or in the presence of high concentrations of phosphate, can lead to dramatic increases in synthesis. Regulation involves products of the rcsABC genes (58). The Lon ATP-dependent protease also participates in regulation of the rcs system. RcsB and RscC are positive and negative regulators, respectively, and are members of a two-component regulatory system (58). RcsA is an additional positive regulator postulated to form a complex with RcsB. RscA is subject to degradation by the Lon protease (58). Rcs homologs are also involved in regulating the expression of surface polysaccharides in Klebsiella spp. and the plant-pathogenic Erwinia spp., as well as the Vi antigen of Salmonella spp. and Citrobacter freundii (reviewed in reference 225).
While information is emerging on the regulation systems that control expression of group I K antigens, little is known of the regulatory elements involved in the expression of group II K antigens. Stevens et al. (194) studied the ability of mutations in several known regulatory genes in E. coli to inhibit K5 capsule expression. They showed that mutations in the rfaH gene abolished capsule expression at 37°C (194). RfaH, which regulates virulence and fertility genes in E. coli and Salmonella spp., was postulated to be a transcriptional antiterminator (19). More recently, a role for RfaH as a positive regulator was suggested (10, 157). Other global regulators, including the histone-like protein, H-NS, integration host factor (IHF), and the leucine-responsive regulatory gene had no apparent effect on polymer synthesis, nor did the rcs system (194). None of the mutations induced capsule expression at 18°C (194).
Transcription of region 3 genes in the kps gene cluster of E. coli K1 initiates from a promoter located 745 bp upstream of the kpsM initiation codon (R. P. Silver, L. F. Wright, and Q. Zhao, unpublished observations). The 745-bp distance between the 5' end of the kpsMT transcript and the kpsM start codon is one of the longest RNA leader sequences described in prokaryotes. Several potential regulatory sequences were identified within this long untranslated region, including a cyclic AMP (cAMP) receptor protein (CRP)-binding site and an IHF-binding site (Silver et al., unpublished). CRP is a DNA-binding protein that functions to activate or repress transcription of many operons in E. coli (89). IHF is a sequence-specific DNA-binding protein encoded by the himA and hip genes (52, 53). This histone-like protein is involved in a wide variety of processes in E. coli including regulation of gene expression (52, 53). Introduction of a himA allele into a strain harboring a kpsM ' -lacZ transcriptional fusion in the chromosome resulted in a 50% decrease in β-galactosidase activity (Silver et al., unpublished). Both IHF and CRP have been shown to bend the DNA to which they bind (149), and regulation of region 3 transcription may involve interaction of regulatory proteins with RNA polymerase through DNA looping to form a complex required for initiation of transcription at the kpsMT promoter.
Region 3 and region 2 of the kps gene cluster are transcribed in the same direction (Fig. 6), resulting in distinct transcripts (Silver et al., unpublished). The mRNA encoded by region 3 had a half-life of approximately 2 min, while the region 2 mRNA was much more stable, with a half-life greater than 10 min (Silver et al., unpublished). Analysis of neuD ' -lacZ transcriptional fusions suggests, however, that region 3 and region 2 may be part of the same transcriptional unit (Silver et al., unpublished), which raises the possibility that region 2 mRNA is the processing product of a larger precursor mRNA whose transcription is initiated from the kpsMT promoter. Indeed, many genes in bacteria are organized into large, polycistronic operons that allow gene regulation to be coordinated efficiently. However, the constituent genes of large transcriptional units often need to be expressed at different levels. Prokaryotes have evolved various methods to solve this problem, including differential RNA stability (18).
Translational coupling is another posttranscriptional mechanism that helps ensure balanced production of polypeptides that function as part of a multicomponent complex (see chapter 60 of this volume). Interestingly, the chain termination codon of kpsM overlaps the initiation codon of kpsT by two nucleotides (146, 189). Experiments comparing gene expression distal to the wild-type and a nonsense mutant allele of kpsM indicated that translation of kpsT is coupled, to some degree, to the translation of kpsM (184). A similar overlap of proximal and distal genes is common in region 2 of the E. coli K1 kps gene cluster and has been observed between neuD and neuB, neuB and neuA, neuA and neuC, and neuE and neuS (9, 193, 230, 231).
The K1 capsule is not synthesized at temperatures below 30°C (21). Region 3-specific transcripts were not detected with RNA isolated from cells grown at 25°C (Silver et al., unpublished), suggesting that thermoregulation of region 3 of the kps cluster operates at the level of transcription. In contrast, region 2 genes do not appear to be thermoregulated. Merker and Troy (125) reported that sialic acid synthase, the product of the neuB gene, was synthesized at low temperatures but was inactive. Enzymatic activity could be reactivated by shifting the cells to 37°C (125). Neither transcription nor translation seemed to be necessary for reactivation, which led the authors to conclude that the enzyme was cold sensitive (125). These observations are difficult to reconcile with the idea that region 3 and region 2 may be part of the same transcriptional unit. However, all the regulatory signals and promoters within the kps gene cluster have yet to be determined. It seems apparent that synthesis and expression of E. coli capsules are complex processes, and all transcriptional and posttranscriptional events must be fully examined to understand the sophisticated regulatory mechanisms involved.
It is important to realize that the structures that constitute the bacterial cell envelope are not unrelated entities. A bacterium must coordinate synthesis of distinct surface polysaccharides to ensure that membranes retain their normal composition and permit orderly growth in vivo. The noncoding region upstream of several bacterial gene clusters involved in the production of various polysaccharide antigens contains a 39-bp sequence (65). These antigens include the outer core (rfa) and O-antigen (rfb) components of LPS, the Vi antigen (viaB) of S. typhi, and a sequence upstream of kpsM in the kps gene cluster (65). This highly conserved sequence (the most divergent sequence has only 6 of 39 mismatches) has been referred to as the JUMPstart sequence (just upstream of many polysaccharide-associated gene starts) (65). There are two occurrences of a 6-bp motif (5'-GGTAGC-3'), and the second half of the sequence is extremely GC rich. The orientation of this element with respect to the downstream polysaccharide genes is the same in each instance, implying a conserved function in polysaccharide gene expression (65). A bacterial cell must coordinately regulate the expression of the polysaccharide structures that it produces, and the JUMPstart sequence has been postulated to be involved in the transcriptional regulation of bacterial gene clusters encoding surface polysaccharides (65).
E. coli is a common inhabitant of the intestinal tract of most warm-blooded animals. Although most strains are harmless commensals, certain isolates are primary pathogens that possess specific virulence determinants that allow the organism to evade host defenses, leading to infection and overt disease (see chapter 149 of this volume). Capsules have long been recognized as important virulence determinants of E. coli that cause extraintestinal infections. In 1927, Smith and Bryant observed that naturally occurring unencapsulated mutants of E. coli, isolated from bacteremic calves, were less pathogenic in guinea pigs and more susceptible to phagocytosis than was the wild-type organism (191). Capsules are the outermost structures on an E. coli cell, and they play a critical role in the interaction between the bacterium and its environment. Capsules provide E. coli with mechanisms to avoid nonspecific host defenses (31, 128, 190, 200). Although the best-known habitat for E. coli is the intestinal tract, part of its life cycle is also spent as an autonomous free-living organism. Capsules may also be important in the ability of E. coli to survive in the extraorganismic environment.
The immunological defense systems in the bloodstream of the vertebrate host present a considerable obstacle to bacterial survival. Activation of the antibody-independent alternative-complement pathway, leading to both bacterial lysis by membrane attack complexes and opsonophagocytosis, is the first line of defense against bacterial infections in the nonimmune host (47, 78). The production of a polysaccharide capsule is a strategy commonly used by invasive bacteria to surmount this challenge (31, 72, 128, 190, 200). The capsule masks deeper, immunogenic cell wall constituents that would normally activate the complement pathway, thereby protecting the organism from both complement-mediated lysis and opsonophagocytosis (31, 66, 128, 200). In addition, the capsule may act as a physical barrier to phagocytosis by preventing contact between the bacterium and the phagocytic cell, a result of the anionic and hydrophilic nature of its constituent polysaccharide (30, 66, 128).
To be effective as a protective shield, the amount of capsule on the cell surface has to attain a certain level. Vermeulen et al. (211) showed that the amount of K1 polymer that was cell associated was greatest during the early phase of growth and correlated with increased virulence. Not all encapsulated E. coli strains are pathogenic. The structure, chemical composition, and conformation of the polymer are important in determining the pathogenic potential of a microorganism (76, 165). Only a few of the 80 distinct K antigens isolated from natural populations are associated with invasive E. coli disease (135). E. coli strains producing group II but not group I capsules are commonly associated with invasive disease. For example, strains expressing the group II K1, K2, K3, K5 and K12, and K13 capsular serotypes are commonly isolated from patients with urinary tract infection (79). Strains that produce the K1 polysialic acid capsule account for 80% of the isolates from patients with E. coli neonatal meningitis; they also form the majority of isolates from patients with neonatal septicemia without meningitis and from patients with acute pediatric pyelonephritis (166, 187). The distribution of somatic O antigens among the disease isolates is also not random, and relatively few serotypes predominate (135). For example, only four O antigens account for the majority of K1 strains isolated from infected neonates (2, 187). Ørskov et al. (136) first postulated the concept that isolates of the same serotype from a defined disease may be descendents of one or a few "clones" of bacteria. It is now well established that most populations of bacteria are discrete clonal lineages with preserved genotypes (182). Moreover, the majority of cases of serious disease, including disease caused by encapsulated E. coli (2), are caused by a small proportion of the total number of existing clones that make up a pathogenic species (2, 182).
The concept of clonality implies that strains that produce the same K antigen do not all have the same pathogenic potential. For example, E. coli K1 strains producing the O18 LPS serotype are found in 49% of meningeal isolates and account for 21% of the E. coli stool isolates from healthy people (156). On the other hand, K1 strains coexpressing the O1 LPS serotype represent the major isolates from stools of healthy individuals yet form only 15% of meningeal isolates (156). The difference in the virulence of the two strains appears to be a function of their O serotypes. While both the O18:K1 and O1:K1 strains were able to colonize the gut of infant rats and reach the mesenteric lymph nodes, only the O18:K1 strain caused bacteremia (156). These results, which can be understood by the observation that the O1 LPS is capable of activating the complement cascade via the classical pathway (154, 155), underscore the importance of other surface structures, the O antigen in particular, in the ability of E. coli to cause invasive disease (72, 151, 190).
The polysialic acid capsule of E. coli K1 has unique properties that contribute to the bloodstream survival of this organism (187). The predominant, preimmune defense mechanism for clearance of bacteria from the blood is the alternative complement pathway (47). In the absence of antibody, the K1 polysaccharide is a poor activator of the alternative complement pathway (187). Activation of this pathway is dependent upon the complement protein C3, which undergoes a low level of cleavage in the fluid phase, resulting in C3a and C3b (reviewed in reference 78). In the fluid phase, C3b is inactivated by binding to factor H and subsequent proteolytic cleavage by factor I. Binding of C3b to a cell surface results in a decrease in the affinity for factor H, leading to stabilization of C3b. Once C3b is bound and protected from inactivation, it can form C3 convertase (composed of C3b, factor B, factor D, and properidin), beginning the complement cleavage cascade which results in the formation of membrane attack complexes and, ultimately, lysis of the bacterial cell (78). In addition, deposition of C3b on the surface of bacteria leads to opsonophagocytosis by phagocytes possesing C3b receptors (66, 78, 128). Mammalian cell surfaces can be protected from this pathway by the presence of glycoconjugates containing sialic acid (47, 78). Surfaces with sialic acid have a high affinity for factor H, promoting the inactivation of C3b (47, 78, 143). This down-regulation of C3 activity by cell surfaces containing sialic acid is thought to be the mechanism that E. coli K1 and other sialic acid-containing capsules use to circumvent alternative complement activation (31, 78, 128, 200).
In the absence of complement activity, encapsulated bacteria can be cleared from the bloodstream via opsonic antibodies directed against the capsule. Several E. coli capsules, however, are also poorly immunogenic (73, 228), a property attributed to the identity of the polymers to structures found on host tissue. The most striking examples are the K1 and K5 polysaccharides. The PSA capsule of E. coli K1 is identical to PSA moieties on the embryonic form of the neural cell adhesion molecule, N-CAM (51). N-CAM is an integral membrane glycoprotein involved in homophilic neuronal cell adhesion, and its activity is modulated by alteration of the PSA content (43, 174). The PSA portion of N-CAM is not only developmentally regulated but is also an oncodevelopmental antigen (172, 173). The K5 capsular polysaccharide, a polymer of 4-β-glucuronic acid and 4-α-N-acetylglucosamine, is identical to the first intermediate in the synthesis of heparin (129, 208). The molecular mimicry of host antigens by the K1 and K5 capsules is probably responsible for the poor immunogenicity of these organisms and suggests that immune tolerance may be important to the pathogenesis of diseases caused by K1 and K5 organisms.
The work described here originating from our laboratories was supported by grant R07382 from the Uniformed Services University of the Health Sciences (P.D.R.) and by U.S. Public Health Service grants AI21309 (P.D.R.) and AI26655 (R.P.S.), a predoctoral training grant in microbial pathogenesis (5-T32-A107362) (R.P.S.), and an interdepartmental predoctoral training grant in genetics and regulation (5-T32-GM07102–17) (R.P.S.) from the National Institutes of Health. We are grateful to Henry Wu and Willie F. Vann for their critical reading of the manuscript and for helpful suggestions.
References
1. Abeijon, C., and C. B. Hirschberg. 1992. Topography of glycosylation reactions in the endoplasmic reticulum. Trends Biochem. Sci. 17:32–36.
2. Achtman, M., and G. Pluschke. 1986. Clonal analysis of descent and virulence among selected Escherichia coli. Annu. Rev. Microbiol. 40:185–210.
3. Acker, G., D. Bitter-Suermann, U. Meier-Dieter, H. Peters, and H. Mayer. 1986. Immunocytochemical localization of enterobacterial common antigen in Escherichia coli and Yersinia enterocolitica cells. J. Bacteriol. 168:348–356.
4. Acker, G., W. Knapp, K. Wartenberg, and H. Mayer. 1981. Localization of enterobacterial common antigen in Yersinia enterocolitica by the immunoferritin technique. J. Bacteriol. 147:602–611.
5. Acker, G., G. Schmidt, and H. Mayer. 1982. Accessibility of enterobacterial common antigen to antibodies in encapsulated and non-encapsulated S and R forms of Escherichia coli. J. Gen. Microbiol. 128:1577–1583.
6. Adlam, C., J. M. Knights, A. Mugridge, J. M. Williams, and J. C. Lindon. 1987.Production of colominic acid by Pasteurella haemolytica serotype A2 organisms. FEMS Microbiol. Lett. 42:23–25.
7. Alexander, D. C., and M. A. Valvano. 1994. Role of the rfe gene in the biosynthesis of the Escherichia coli O7-specific lipopolysaccharide and other O-specific polysaccharides containing N-acetylglucosamine. J. Bacteriol. 176:7079–7084.
8. Ames, G. F.-L., C. S. Mimura, S. R. Holbrook, and V. Shyamala. 1992. Traffic ATPases: a superfamily of transport proteins operating from Escherichia coli to humans. Adv. Enzymol. 65:1–47.
9. Annunziato, P., L. Wright, W. F. Vann, and R. P. Silver. 1995. Nucleotide sequence and genetic analysis of the neuD and neuB genes in region 2 of the polysialic acid gene cluster of Escherichia coli K1. J. Bacteriol. 177:312–319.
10. Bailey, M., V. Koronakis, T. Schmoll, and C. Hughes. 1992. Escherichia coli HlyT protein, a transcriptional activator of haemolysin synthesis and secretion, is encoded by the rfaH (sfrB) locus required for expression of sex factor and lipopolysaccharide genes. Mol. Microbiol. 6:1003–1012.
11. Barr, K., P. Nunes-Edwards, and P. D. Rick. 1989. In vitro synthesis of a lipid-linked trisaccharide involved in synthesis of enterobacterial common antigen. J. Bacteriol. 171:1326–1332.
12. Barr, K., and P. D. Rick. 1987. Biosynthesis of enterobacterial common antigen in Escherichia coli. In vitro synthesis of lipid-linked intermediates. J. Biol. Chem. 262:7142–7150.
13. Barr, K., S. Ward, U. Meier-Dieter, H. Mayer, and P. D. Rick. 1988. Characterization of an Escherichia coli rff mutant defective in transfer of N-acetylmannosaminuronic acid (ManNAcA) form UDP-ManNAcA to a lipid-linked intermediate involved in enterobacterial common antigen synthesis. J. Bacteriol. 170:228–233.
14. Basu, S., H.-M. Kuhn, A. Neszmelyi, K. Himmelspach, and H. Mayer. 1987. Chemical characterization of enterobacterial common antigen isolated from Plesiomonas shigelloides ATCC 14029. Eur. J. Biochem. 162:75–81.
15. Bayer, M. E. 1979. The fusion sites between outer membrane and cytoplasmic membrane of bacteria: their role in membrane assembly and virus infection, p. 167–202. In M. Inouye (ed.), Bacterial Outer Membranes: Biogenesis and Function. John Wiley & Sons, Inc., New York.
16. Bayer, M. E. 1991. Zones of adhesion in the cryofixed envelope of Escherichia coli. J. Struct. Biol. 107:268–280.
17. Bayer, M. E., and H. Thurow. 1977. Polysaccharide capsule of Escherichia coli: microscope study of its size, structure, and sites of synthesis. J. Bacteriol. 130:911–936.
18. Belasco, J. G. 1993. mRNA degradation in prokaryotic cells: an overview, p. 3–12. In J. G. Belasco and G. Brawerman (ed.), Control of Messenger RNA Stability. Academic Press, Inc., San Diego, Calif.
19. Beutin, L. P. Manning, M. Achtman, and N. Willets. 1981. sfrA and sfrB products of Escherichia coli K-12 are transcriptional control factors. J. Bacteriol. 145:840–844.
20. Blight, M. A., and I. B. Holland. 1990. Structure and function of haemolysin B, P-glycoprotein, and other members of a novel family of membrane translocators. Mol. Microbiol. 4:873–880.
21. Bortolussi, R., P. Ferrieri, and P. Quie. 1983. Influence of growth temperature of Escherichia coli K1 on capsular antigen production and resistance to opsonization. Infect. Immun. 39:1136–1141.
22. Böttger, E. C., M. Jürs, T. Barrett, K. Wachsmuth, S. Metzger, and D. Bitter-Suermann. 1987. Qualitative and quantitative determination of enterobacterial common antigen (ECA) with monoclonal antibodies. Expression of ECA by two Actinobacillus species. J. Clin. Microbiol. 25:377–382.
23. Boulnois, G., R. Drake, R. Pearce, and I. Roberts. 1992. Genome diversity at the serA-linked capsule locus in Escherichia coli. FEMS Microbiol. Lett. 100:121–124.
24. Boulnois, G. J., and K. Jann. 1989. Bacterial polysaccharide capsule synthesis, export, and evolution of structural diversity. Mol. Microbiol. 3:1819–1823.
25. Boulnois, G. J., and I. S. Roberts. 1990. Genetics of capsular polysaccharide production in bacteria. Curr. Top. Microbiol. Immunol. 150:1–18.
26. Boulnois, G. J., I. S. Roberts, R. Hodge, K. R. Hardy, K. Jann, and K. N. Timmis. 1987. Analysis of K1 capsule biosynthesis genes of Escherichia coli: definition of three functional regions for capsule production. Mol. Gen. Genet. 208:242–246.
27. Bronner, D., V. Sieberth, C. Pazzani, I. S. Roberts, G. J. Boulnois, B. Jann, and K. Jann. 1993. Expression of the capsular K5 polysaccharide of Escherichia coli: biochemical and electron microscopic analyses of mutants with defects in region 1 of the K5 gene cluster. J. Bacteriol. 175:5984–5992.
28. Bronner, D., V. Sieberth, C. Pazzani, A. Smith, G. Boulnois, I. Roberts, B. Jann, and K. Jann. 1993. Synthesis of the K5 (group II) capsular polysaccharide in transport-deficient recombinant Escherichia coli. FEMS Microbiol. Lett. 113:279–284.
29. Cieslewicz, M. J., S. M. Steenbergen, and E. R. Vimr. 1993. Cloning, sequencing, expression, and complementation analysis of the Escherichia coli K1 kps region 1 gene, kpsE, and identification of an upstream open reading frame encoding a protein with homology to GutQ. J. Bacteriol. 175:8018–8023.
30. Clarke, B. R., and C. Whitfield. 1992. Molecular cloning of the rfb region of Klebsiella pneumoniae serotype O1:K2O: the rfb gene cluster is responsible for synthesis of the d-galactan I O polysaccharide. J. Bacteriol. 174:4614–4621.
31. Cross, A. S. 1990. The biologic significance of bacterial encapsulation. Curr. Top. Microbiol. Immunol. 150:87–95.
32. Daniels, D. L., G. Plunkett III, V. Burland, and F. R. Blattner. 1992. Analysis of the Escherichia coli genome: DNA sequence of the region from 84.5 to 86.5 minutes. Science 257:771–778.
33. Dell, A., J. Oates, C. Lugowski, E. Romanowska, L. Kenne, and B. Lindberg. 1984. The enterobacterial common-antigen. A cyclic polysaccharide. Carbohydr. Res. 133:95–104.
34. Deretic, V. C., C. D. Mohr, and D. W. Martin. 1991. Mucoid Pseudomonas aeruginosa in cystic fibrosis: signal transduction and histone-like elements in the regulation of bacterial virulence. Mol. Microbiol. 5:1577–1583.
35. Devi, S. J. N., R. Schneerson, W. Egan, W. F. Vann, J. B. Robbins, and J. Shiloach. 1991. Identity between the polysaccharide antigens of Moraxella nonliquefaciens, group B Neisseria meningitidis, and Escherichia coli K1 (non-O acetylated). Infect. Immun. 59:732–736.
36. Dicker, I. B., and S. Seetharam. 1992. What is known about the structure and function of the Escherichia coli protein FirA? Mol. Microbiol. 6:817–823.
37. Dmitriev, B. A., K. A. Knirel, N. K. Kochetkov, and I. L. Hofman. 1976. Somatic antigens of Shigella. Structural investigations on the O-specific polysaccharide chain of Shigella dysenteriae type 1 lipopolysaccharide. Eur. J. Biochem. 66:559–566.
38. Doige, C. A., and G. F-L. Ames. 1993. ATP-dependent transport systems in bacteria and humans: relevance to cystic fibrosis and multidrug resistance. Annu. Rev. Microbiol. 47:291–319.
39. Dorman, C. J., and N. N. Bhriain. 1993. DNA topology and bacterial virulence gene regulation. Trends Microbiol. 1:92–99.
40. Downie, J. A. 1989. The nodL gene of Rhizobium leguminosarum is homologous to the acetyl transferases encoded by lacA and cysE. Mol. Microbiol. 3:1649–1651.
41. Drake, C. R., G. J. Boulnois, and I. S. Roberts. 1993. The Escherichia coli serA-linked capsule locus and its flanking sequences are polymorphic, genetic evidence for the existence of more than two groups of capsule gene clusters. J. Gen. Microbiol. 139:1701–1714.
42. Drake, C. R., I. S. Roberts, B. Jann, K. Jann, and G. J. Boulnois. 1990. Molecular cloning and expression of the genes encoding the Escherichia coli K4 capsular polysaccharide, a fructose substituted chondroitin. FEMS Microbiol. Lett. 66:227–230.
43. Edelman, G. M. 1985. Cell adhesion and the molecular processes of morphogenesis. Annu. Rev. Biochem. 54:135–169.
44. Edwards, U., A. Müller, S. Hammerschmidt, R. Gerardy-Schahn, and M. Frosch. 1994. Molecular analysis of the biosynthesis pathway of the α-2,8 polysialic acid capsule by Neisseria meningitidis serogroup B. Mol. Microbiol. 14:141–149.
45. Erbing, C., L. Kenne, B. Lindberg, and S. Hammarström. 1978. Structure of the O-specific side-chains of the Escherichia coli 075 lipopolysaccharide: a revision. Carbohydr. Res. 60:400–403.
46. Fath, M. J., and R. Kolter. 1993. ABC-transporters: the bacterial exporters. Microbiol. Rev. 57:995–1017.
47. Fearson, D. T., and K. F. Austen. 1980. The alternative pathway of complement: a system for host resistance to microbial infection. N. Engl. J. Med. 303:254–263.
48. Finke, A., D. Bronner, A. V. Nikolaev, B. Jann, and K. Jann. 1991. Biosynthesis of the Escherichia coli K5 polysaccharide, a representative of group II capsular polysaccharides: polymerization in vitro and characterization of the product. J. Bacteriol. 173:4088–4094.
49. Finke, A., B. Jann, and K. Jann. 1990. CMP-KDO synthetase activity in Escherichia coli expressing different capsular polysaccharide. FEMS Microbiol. Lett. 69:129–134.
50. Finke, A., I. S. Roberts, G. J. Boulnois, C. Pazzani, and K. Jann. 1989. Activity of CMP-2-keto-3-deoxyoctulosonic acid synthetase in Escherichia coli strains expressing the capsular K5 polysaccharide: implications for K5 polysaccharide biosynthesis. J. Bacteriol. 171:3074–3075.
51. Finne, J., M. Leinonen, and P. H. Mäkelä. 1983. Antigenic similarities between brain components and bacteria causing meningitis; implications for vaccine development. Lancet ii:355–357.
52. Freundlich, M., N. Ramani, E. Mathew, A. Sirko, and P. Tsui. 1992. The role of integration host factor in gene expression in Escherichia coli. Mol. Microbiol. 6:2557–2563.
53. Friedman, D. I. 1988. Integration host factor: a protein for all reasons. Cell 55:484–489.
54. Frosch, M., U. Edwards, K. Bousset, B. Krausse, and C. Weisgerber. 1991. Evidence for a common molecular origin of the capsule gene loci in Gram-negative bacteria expressing group II capsular polysaccharides. Mol. Microbiol. 5:1251–1263.
55. Frosch, M., and A. Müller. 1993. Phospholipid substitution of capsular polysaccharides and mechanism of capsule formation in Neisseria meningitidis. Mol. Microbiol. 8:483–493.
56. Ganguli, S., G. Zapata, T. Wallis, C. Reid, G. Boulnois, W. F. Vann, and I. S. Roberts. 1994. Molecular cloning and analysis of genes for sialic acid synthesis in Neisseria meningitidis group B and purification of the meningococcal CMP-NeuNAc synthetase enzyme. J. Bacteriol. 176:4583–4589.
57. Gotschlich, E. C., B. A. Fraser, O. Nishimura, J. B. Robbins, and T.-Y. Liu. 1981. Lipids on capsular polysaccharides of gram-negative bacteria. J. Biol. Chem. 256:8915–8921.
58. Gottesman, S., and V. Stout. 1991. Regulation of capsular polysaccharide synthesis in Escherichia coli K-12. Mol. Microbiol. 5:1599–1606.
59. Gross, R. J., T. Cheasty, and B. Rowe. 1977. Isolation of bacteriophage specific for the K1 polysaccharide antigen of Escherichia coli. J. Clin. Microbiol. 6:548–550.
60. Gupta, D. S., B. Jann, and K. Jann. 1984. Escherichia coli 1 O8ac antigen: structure of the O-specific polysaccharide moiety. Infect. Immun. 45:203–209.
61. Hammerström, S., H. E. Carlsson, P. Perlmann, and S. Svensson. 1971. Immunochemistry of the common antigen of Enterobacteriaceae (Kunin). J. Exp. Med. 134:565–576.
62. Heifitz, A., R. W. Keenan, and A. D. Elbein. 1979. Mechanism of action of tunicamycin on the UDP-GlcNAc:dolichyl-phosphate GlcNAc-1-phosphate transferase. Biochemistry 18:2186–2192.
63. Higa, H. H., and A. Varki. 1988. Acetyl-coenzyme A:polysialic acid O-acetyltransferase from K1-positive Escherichia coli. J. Biol. Chem. 263:8872–8878.
64. Higgins, C. F. 1992. ABC transporters: from microorganisms to man. Annu. Rev. Cell Biol. 8:67–113.
65. Hobbs, M., and P. R. Reeves. 1994. The JUMPstart sequence: a 39 bp element common to several polysaccharide gene clusters. Mol. Microbiol. 12:855–856.
66. Horowitz, M., A. 1982. Phagocytosis of microorganisms. Rev. Infect. Dis. 4:104–123.
67. Janas, T., and F. A. Troy. 1989. The Escherichia coli K1 poly-α-2,8-sialosyl sialyltransferase is topologically oriented toward the cytoplasmic face of the inner membrane, p. 207–208. In N. Sharon, H. Lis, D. Duskin, and I. Kahane (ed.), Proceedings of the Xth International Symposium on Glycoconjugates. Organizing Committee Press, Jerusalem, Israel.
68. Jann, B., and K. Jann. 1990. Structure and biosynthesis of the capsular antigens of Escherichia coli. Curr. Top. Microbiol. Immunol. 150:19–42.
69. Jann, K., T. Dengler, and B. Jann. 1992. Core-lipid A on the K40 polysaccharide of Escherichia coli O8:K40:H9, a representative group I capsular polysaccharide. Zentralbl. Bakteriol. 276:196–204.
70. Jann, K., G. Goldemann, C. Weisgerber, C. Wolf-Ullisch, and S. Kanegasaki. 1982. Biosynthesis of the 09 antigen of Escherichia coli. Eur. J. Biochem. 127:157–164.
71. Jann, K., and B. Jann. 1984. Structure and biosynthesis of O-antigens, p. 138–186. In E. T. Rietschel (ed.), Handbook of Endotoxin. vol. 1. Chemistry of Endotoxins. Elsevier Science Publishing, Amsterdam.
72. Jann, K., and B. Jann. 1985. Cell surface components and virulence: Escherichia coli O and K antigens in relation to virulence and pathogenicity, p. 156–176. In M. Sussman (ed.), The Virulence of Escherichia coli. Academic Press, Inc., New York.
73. Jann, K., and B. Jann. 1992. Capsules of Escherichia coli, expression and biological significance. Can. J. Microbiol. 38:705–710.
74. Jayaratne, P., D. Bronner, P. R. MacLachlan, C. Dodgson, N. Kido, and C. Whitfield. 1994. Cloning and analysis of duplicated rfbM and rfbK genes involved in the formation of GDP-mannose in Escherichia coli O9:K30 and participation of rfb genes in the synthesis of the group I K30 capsular polysaccharide. J. Bacteriol. 176:3126–3139.
75. Jayaratne, P., W. Keenleyside, P. MacLachlan, and C. Whitfield. 1993. Characterisation of rcsB and rcsC from Escherichia coli O9:K30:H12 and examination of the role of the rcs regulatory system in expression of group I capsular polysaccharides. J. Bacteriol. 175:6725–6730.
76. Jennings, H. J. 1983. Capsular polysaccharides as human vaccines. Adv. Carbohydr. Chem. Biochem. 41:155–208.
77. Johns, M. A., R. E. Whiteside, E. E. Baker, and W. R. McCabe. 1973. Common enterobacterial antigen. I. Isolation and purification from Salmonella typhosa O:901. J. Immunol. 110:781–790.
78. Joiner, K. A. 1988. Complement evasion by bacteria and parasites. Annu. Rev. Microbiol. 42:201–230.
79. Kaijser, B., L. A. Hanson, V. Jodal, G. Linden-Johnson, and J. B. Robbins. 1977. Frequency of Escherichia coli K antigen in urinary tract infection in children. Lancet i:664–666.
80. Kasper, D. L., J. L. Winkelhake, W. D. Zollinger, B. L. Brandt, and M. S. Artenstein. 1973. Immunochemical similarity between polysaccharide antigens of Escherichia coli O7:K1L:NM and group B Neisseria meningitidis. J. Immunol. 110:262–268.
81. Kauffman, F. 1944. Zur Serologie der Coli-Gruppe. Acta Pathol. Microbiol. Scand. 21:20–45.
82. Kauffman, F. 1966. The Bacteriology of Enterobacteriaceae. Munksgaard, Copenhagen.
83. Keenleyside, W., D. Bronner, K. Jann, B. Jann, and C. Whitfield. 1993. Coexpression of colanic acid and serotype-specific capsular polysaccharides in Escherichia coli strains with group II K antigens. J. Bacteriol. 175:6725–6730.
84. Keenleyside, W., P. Jayaratne, P. MacLachlan, and C. Whitfield. 1992. The rcsA gene of Escherichia coli O9:K30:H12 is involved in the expression of the serotype-specific group I K (capsular) antigen. J. Bacteriol. 174:8–16.
85. Kellenberger, E. 1990. The ‘Bayer bridges’ confronted with results from improved electron microscopy methods. Mol. Microbiol. 4:697–705.
86. Kenne, L., and B. Lindberg. 1983. Bacterial polysaccharides, p. 287–363. In G. O. Aspinall (ed.), The Polysaccharides, vol. 2. Academic Press, Inc., New York.
87. Kiss, P., J. Rinno, G. Schmidt, and H. Mayer. 1978. Structural studies on the immunogenic form of the enterobacterial common antigen. Eur. J. Biochem. 88:211–218.
88. Klena, J. D., and C. A. Schnaitman. 1993. Function of the rfb gene cluster and the rfe gene in the synthesis of O antigen by Shigella dysenteriae 1. Mol. Microbiol. 9:393–402.
89. Kolb, A., S. Busby, H. Buc, S. Garges, and S. Adhya. 1993. Transcriptional regulation by cAMP and its receptor protein. Annu. Rev. Biochem. 62:749–795.
90. Kornfeld, R., and S. Kornfeld. 1985. Assembly of aparagine-linked oligosaccharides. Annu. Rev. Biochem. 54:631–664.
91. Kroll, J. S., B. Loynds, L. N. Brophy, and E. R. Moxon. 1990. The bex locus in encapsulated Haemophilus influenzae: a chromosomal region involved in capsule polysaccharide export. Mol. Microbiol. 4:1853–1862.
92. Kroncke, K. D., G. Boulnois, I. Roberts, D. Bitter-Suermann, J. R. Golecki, B. Jann, and K. Jann. 1990. Expression of the Escherichia coli K5 capsular antigen: immunoelectron microscopic and biochemical studies with recombinant E. coli. J. Bacteriol. 172:1085–1091.
93. Kroncke, K. D., J. R. Golecki, and K. Jann. 1990. Further electron microscopic studies on the expression of Escherichia coli group II capsules. J. Bacteriol. 172:3469–3472.
94. Kuhn, H.-M., G. Adamus, E. Romanowska, and H. Mayer. 1981. Effect of proteins on the immunogenicity of enterobacterial common antigen. Infect. Immun. 34:373–377.
95. Kuhn, H.-M., S. Basu, and H. Mayer. 1987. Comparison of enterobacterial common antigen from different species by serological techniques. Eur. J. Biochem. 162:69–74.
96. Kuhn, H.-M., U. Meier-Dieter, and H. Mayer. 1988. ECA, the enterobacterial common antigen.FEMS Microbiol. Rev. 54:195–222.
97. Kuhn, H. M., E. Neter, and H. Mayer. 1983. Modification of the lipid moiety of the enterobacterial common antigen by the "Pseudomonas" factor. Infect. Immun. 40:696–700.
98. Kunin, C. M. 1962. Antibody distribution against nonenteropathic E. coli. Arch. Intern. Med. 110:676–686.
99. Kunin, C. M. 1963. Separation, characterization, and biological significance of a common antigen in Enterobacteriaceae. J. Exp. Med. 118:565–586.
100. Kunin, C. M., M. V. Beard, and N. E. Halmagyi. 1962. Evidence for a common hapten associated with endotoxin fractions of E. coli and other Enterobacteriaceae. Proc. Soc. Exp. Biol. Med. 111:160–166.
101. Kuo, J. S.-C., V. W. Doelling, J. F. Graveline, and D. W. McCoy. 1985. Evidence for covalent attachment of of phospholipid to the capsular polysaccharide of Haemophilus influenzae type B. J. Bacteriol. 163:769–773.
102. Laasko, D. M., M. K. Homonylo, S. J. Wilmot, and C. Whitfield. 1988. Transfer and expression of the genetic determinants for O and K antigen synthesis in Escherichia coli O9:K(A)30 and Klebsiella sp. O1:K20 in E. coli K12. Can. J. Microbiol. 34:987–992.
103. Lew, H. C., P. H. Mäkelä, H.-M. Kuhn, H. Mayer, and H. Nikaido. 1986. Biosynthesis of enterobacterial common antigen requires dTDP-glucose pyrophosphorylase determined by a Salmonella typhimurium rfb gene and a Salmonella montevideo rfe gene. J. Bacteriol. 168:715–721.
104. Lew, H. C., N. Nikaido, and P. H. Mäkelä, 1978. Biosynthesis of uridine diphosphate N-acetylmannosaminuronic acid in rff mutants of Salmonella typhimurium. J. Bacteriol. 136:227–233.
105. Lüderitz, O., O. Westphal, A. M. Staub, and H. Nikaido. 1971. Isolation and chemical and immunological characterization of bacterial lipopolysaccharides, p. 145–233. In G. Weinbaum, S. Kadis, and S. Ajl (ed.), Microbial Toxins, vol. 4. Bacterial Endotoxins. Academic Press, Inc., New York.
106. Lugowski, C., and E. Romanowska. 1978. Enterobacterial common antigen: isolation from Shigella sonnei, purification and immunochemical characterization. Eur. J. Biochem. 91:89–97.
107. Lugowski, C., and E. Romanowska. 1991. Characterization of an enterobacterial common antigen (ECA) epitope recognized by anti-ECA-tetanus toxoid conjugate serum. FEMS Microbiol. Lett. 77:315–318.
108. Lugowski, C., E. Romanowska, L. Kenne, and B. Lindberg. 1983. Identification of a trisaccharide repeating-unit in the enterobacterial common antigen. Carbohydr. Res. 118:173–181.
109. L’vov, V. L., A. S. Shashkov, B. A. Dmitriev, N. K. Kochetkov, B. Jann, and K. Jann. 1984. Structural studies of the O-specific side of the lipopolysaccharide from Escherichia coli O7. Carbohydr. Res. 126:249–259.
110. Mahoney, W. C., and D. Duskin. 1979. Biological activities of the two major components of tunicamycin. J. Biol. Chem. 254:6572–6576.
111. Mäkelä, P. H., M. Jahkola, and O. Lüderitz. 1970. A new gene cluster rfe concerned with the biosynthesis of Salmonella lipopolysaccharide. J. Gen. Microbiol. 60:91–106.
112. Mäkelä, P. H., and H. Mayer. 1974. Participation of lipopolysaccharide genes in the determination of the enterobacterial common antigen: analysis in Salmonella groups B and C1. J. Bacteriol. 119:765–770.
113. Mäkelä, P. H., and H. Mayer. 1976. Enterobacterial common antigen.Bacteriol. Rev. 40:591–632.
114. Mäkelä, P. H., G. Schmidt, H. Mayer, N. Nikaido, Y. Whang, and E. Neter. 1976. Enterobacterial common antigen in rfb deletion mutants of Salmonella typhimurium. J. Bacteriol. 127:1141–1149.
115. Männel, D., and H. Mayer. 1978. Isolation and chemical characterization of the enterobacterial common antigen. Eur. J. Biochem. 86:364–370.
116. Marx, A., and M. Petcovici. 1975. Immunochemical studies on purified common enterobacterial antigen. Zentralbl. Bakteriol. Parasitenkd. Infectionskr. Hyg. Abt. I Orig. 233:486–494.
117. Masson, L., and B. E. Holbein. 1985. Role of lipid intermediate(s) in the synthesis of serogroup B Neisseria meningitidis capsular polysaccharide. J. Bacteriol. 161:861–867.
118. Mayer, H., and G. Schmidt. 1979. Chemistry and biology of the enterobacterial common antigen (ECA).Curr. Top. Microbiol. Immunol. 85:99–153.
119. Mayer, H., G. Schmidt, H. Y. Whang, and E. Neter. 1972. Biochemical basis of the immunogenicity of the common enterobacterial antigen. Infect. Immun. 6:540–544.
120. McLaughlin, J. C., and G. J. Domingue. 1974. The immunologic role of the ethanol-soluble enterobacterial common antigen versus experimental renal infection. Immunol. Commun. 3:51–75.
121. Meier-Dieter, U., K. Barr, R. Starman, L. Hatch, and P. D. Rick. 1992. Nucleotide sequence of the Escherichia coli rfe gene involved in the synthesis of enterobacterial common antigen. Molecular cloning of the rfe-rff gene cluster. J. Biol. Chem. 267:746–753.
122. Meier-Dieter, U., and H. Mayer. 1985. Genetic location of genes encoding enterobacterial common antigen. J. Bacteriol. 163:756–762.
123. Meier-Dieter, U., R. Starman, K. Barr, H. Mayer, and P. D. Rick. 1990. Biosynthesis of enterobacterial common antigen in Escherichia coli. Biochemical characterization of Tn10 insertion mutants defective in enterobacterial common antigen synthesis. J. Biol. Chem. 265:13490–13497.
124. Mekalanos, J. J. 1991. Environmental signals controlling expression of virulence determinants in bacteria. J. Bacteriol. 174:1–7.
125. Merker, R. I., and F. A. Troy. 1990. Biosynthesis of the polysialic acid capsule in Escherichia coli K1. Cold inactivation of sialic acid synthase regulates capsule expression below 20°C.Glycobiology 1:93–100.
126. Mimura, C. S., S. R. Holbrook, and G. F.-L. Ames. 1991. Structural model of the nucleotide-binding conserved component of periplasmic permeases. Proc. Natl. Acad. Sci. USA 88:84–88.
127. Mirelman, D., Y. Yashouv-Gan, and U. Schwartz. 1976. Peptidoglycan biosynthesis in a thermosensitive division mutant of Escherichia coli. Biochemistry 15:1781–1790.
128. Moxon, E. R., and J. S. Kroll. 1990. The role of bacterial polysaccharide capsules as virulence factors. Curr. Top. Microbiol. Immunol. 150:65–85.
129. Navia, J. L., J. Riesenfeld, W. F. Vann, U. Lindahl, and L. Roden. 1983. Assay of N-acetylheparosan deacetylase with a capsular polysaccharide from Escherichia coli K5 as substrate. Anal. Biochem. 35:134–140.
130. Neter, E., and H. Y. Whang. 1972. The common antigen of gram-negative bacteria, p. 14–21. In A. Nowotny (ed.), Cellular Antigens. Springer-Verlag, New York.
131. Nikaido, H., and M. Vaara. 1987. Outer membrane, p. 7–22. In F. C. Neidhardt, J. L. Ingraham, K. B. Low, B. Magasanik, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella typhimurium: Cellular and Molecular Biology, vol. 1. American Society for Microbiology, Washington, D.C.
132. Nnalue, N. A., and B. A. D. Stocker. 1987. The effects of O-antigen character and enterobacterial common antigen content on the in vivo persistence of aromatic-dependent Salmonella sp. live-vaccine strains. Microb. Pathog. 3:31–44.
133. Ohta, M., K. Ina, K. Kusuzaki, N. Kido, Y. Arakawa, and N. Kato. 1991. Cloning and expression of the rfe-rff gene cluster of Escherichia coli. Mol. Microbiol 5:1853–1862.
134. Ørskov, F., and I. Ørskov. 1984. Serotyping of Escherichia coli. Methods Microbiol. 14:43–112.
135. Ørskov, F., and I. Ørskov. 1992. Escherichia coli serotyping and disease in man and animals. Can. J. Microbiol. 38:699–704.
136. Ørskov, F., I. Ørskov, D. J. Evans, R. P. Sack, and D. A. Wadstrom. 1976. Special Escherichia coli serotypes among enterotoxigenic strains from diarrhaea in adults and children. Med. Microbiol. Immunol. 162:73–80.
137. Ørskov, F., I. Ørskov, A. Sutton, R. Schneerson, W. Lin, W. Egan, G. E. Hoff, and J. B. Robbins. 1979. Form variation in Escherichia coli K1: determined by O-acetylation of the capsular polysaccharide. J. Exp. Med. 149:669–685.
138. Ørskov, F., V. Sharma, and I. Ørskov. 1984. Influence of growth temperature on the development of Escherichia coli polysaccharide K antigens. J. Gen. Microbiol. 130:2681–2684.
139. Ørskov, I., and K. Nyman. 1974. Genetic mapping of the antigenic determinants of two polysaccharide K antigens, K10 and K54, in Escherichia coli. J. Bacteriol. 120:43–51.
140. Ørskov, I., F. Ørskov, B. Jann, and K. Jann. 1977. Serology, chemistry, and genetics of O and K antigens of Escherichia coli. Bacteriol. Rev. 41:667–710.
141. Ørskov, I., V. Sharma, and F. Ørskov. 1976. Genetic mapping of the K1 and K4 antigens (L) of Escherichia coli. Acta Pathol. Microbiol. Scand. Sect. B 84:125–131.
142. Paakkanen, J., E. C. Gotschlich, and P. H. Mäkelä. 1979. Protein K: a new major outer membrane protein found on encapsulated Escherichia coli. J. Bacteriol. 139:835–841.
143. Pangburn, M. K., and H. J. Muller-Eberharg. 1978. Complement C3 convertase: Cell surface restriction of β1H control and generation of restriction on neuraminidase-treated cells. Proc. Natl. Acad. Sci. USA 75:2416–2420.
144. Parent, R., and P. H. Roy. 1992. The chloramphenicol acetyltransferase gene of Tn2424: a new breed of cat. J. Bacteriol. 174:2891–2897.
145. Pavelka, M. S., Jr., S. F. Hayes, and R. P. Silver. 1994. Characterization of KpsT, the ATP-binding component of the ABC-transporter involved with the export of capsular polysialic acid in Escherichia coli K1. J. Biol. Chem. 269:20149–20158.
146. Pavelka, M. S., Jr., L. F. Wright, and R. P. Silver. 1991. Identification of two genes, kpsM and kpsT, in region 3 of the polysialic acid gene cluster of Escherichia coli K1. J. Bacteriol. 173:4603–4610.
147. Pazzani, C., C. Rosenow, G. J. Boulnois, D. Bronner, K. Jann, and I. S. Roberts. 1993. Molecular analysis of region 1 of the Escherichia coli K5 antigen gene cluster: a region encoding proteins involved in cell surface expression of capsular polysaccharide. J. Bacteriol. 175:5978–5983.
148. Pelkonen, S. 1990. Capsular sialyl chains of Escherichia coli K1 mutants resistant to K1 phage. Curr. Microbiol. 21:23–28.
149. Perez-Martin, J., F. Rojo, and V. deLorenzo. 1994. Promoters responsive to DNA bending: a common theme in prokaryotic gene expression. Microbiol. Rev. 58:268–290.
150. Peters, H., M. Jürs, B. Jann, K. Jann, K. N. Timmis, and D. Bitter-Suermann. 1985. Monoclonal antibodies to enterobacterial common antigen and to Escherichia coli lipopolysaccharide outer core: demonstration of an antigenic determinant shared by enterobacterial common antigen and E. coli K5 capsular polysaccharide. Infect. Immun. 50:459–466.
151. Peterson, P. K. and P. G. Quie. 1981. Bacterial surface components and the pathogenesis of infectious diseases. Annu. Rev. Med. 32:29–43.
152. Pickard, D., M. Roberts, D. Maskell, D. Hone, M. Levine, G. Dougan, and S. Chatfield. 1994. Characterization of defined ompR mutants of Salmonella typhi: ompR is involved in regulation of Vi polysaccharide expression. Infect. Immun. 62:3984–3993.
153. Pigeon, R. P., and R. P. Silver. 1994. Topological analysis of KpsM: the membrane component of an ABC transporter involved in export of polysialic acid in Escherichia coli K1. Mol. Microbiol. 14:871–881.
154. Pluschke, G., and M. Achtman. 1984. Degree of antibody-independent activation of the classical complement pathway by K1 Escherichia coli differs with O antigen type and correlates with virulence of meningitis in newborns. Infect. Immun. 43:684–692.
155. Pluschke, G., J. Mayden, M. Achtman, and R. P. Levine. 1983. Role of the capsule and the O antigen in resistance of O18:K1 Escherichia coli to complement-mediated killing. Infect. Immun. 42:907–913.
156. Pluschke, G., A. Mercer, B. Kusecek, A. Pohl, and M. Achtman. 1983. The induction of bacteremia in newborn rats by Escherichia coli K1 is correlated with only certain O (lipopolysaccharide) antigens. Infect. Immun. 39:599–608.
157. Pradel, E., and C. Schnaitman. 1991. Effect of rfaH (sfrB) and temperature on expression of rfa genes of Escherichia coli K-12. J. Bacteriol. 173:6428–6431.
158. Prehm, P., B. Jann, and K. Jann. 1976. The 09 antigen of Escherichia coli. Structure of the polysaccharide chain. Eur. J. Biochem. 67:53–56.
159. Ramia, S., E. Neter, and D.J. Brenner. 1982. Production of enterobacterial common antigen as an aid to classification of newly identified species of the families Enterobacteriaceae and Vibrionaceae. Int. J. Syst. Bacteriol. 32:395–398.
160. Reske, K., and K. Jann. 1972. The O8 antigen of Escherichia coli. Structure of the polysaccharide chain. Eur. J. Biochem. 31:320–328.
161. Rick, P. D., G. L. Hubbard, and K. Barr. 1994. Role of the rfe gene in the synthesis of the O8 antigen in Escherichia coli K-12. J. Bacteriol. 176:2877–2884.
162. Rick, P. D., H. Mayer, B.A. Neumeyer, S. Wolski, and D. Bitter-Suermann. 1985. Biosynthesis of enterobacterial common antigen. J. Bacteriol. 162: 494–503.
163. Rick, P. D., S. Wolski, K. Barr, S. Ward, and L. Ramsay-Sharer. 1988. Accumulation of a lipid-linked intermediate involved in enterobacterial common antigen synthesis in Salmonella typhimurium mutants lacking dTDP-glucose pyrophosphorylase. J. Bacteriol. 170:4008–4014.
164. Rinno, J., J. R. Golecki, and H. Mayer. 1980. Localization of enterobacterial common antigen: immunogenic and nonimmunogenic enterobacterial common antigen-containing Escherichia coli. J. Bacteriol. 141:814–821.
165. Robbins, J. B. 1978. Vaccines for the prevention of encapsulated bacterial disease. Current status, problems and prospects for the future. Immunochemistry 15:839–854.
166. Robbins, J. B., G. H. McCracken, Jr., E. C. Gotschlich, F. Ørskov, and L. A. Hanson. 1974. Escherichia coli K1 capsular polysaccharide associated with neonatal meningitis. N. Engl. J. Med. 290:1216–1220.
167. Roberts, I., R. Mountford, N. High, D. Bitter-Suermann, K. Jann, K. Timmis, and G. Boulnois. 1986. Molecular cloning and analysis of genes for production of K5, K7, K12, and K92 capsular polysaccharide in Escherichia coli. J. Bacteriol. 168:1228–1233.
168. Roberts, I. S., R. Mountford, R. Hodge, K. B. Jann, and G. J. Boulnois. 1988. Common organization of gene clusters for the production of different capsular polysaccharides (K antigens) in Escherichia coli. J. Bacteriol. 170:1305–1310.
169. Rodriquez-Aparicio, L. B., A. Reglero, A. I. Ortiz, and J. M. Luengo. 1988. A protein-sialyl polymer complex involved in colominic acid biosynthesis. Biochem. J. 251:589–596.
170. Rohr, T. E., and F. A. Troy. 1980. Structure and biosynthesis of surface polymers containing polysialic acid in Escherichia coli. J. Biol. Chem. 255:2332–2342.
171. Rosenberg, A., and C. Schengrand. 1976. Biological Roles of Sialic Acids. Plenum Publishing Corp., New York.
172. Roth, J., C. Zuber, P. Komminoth, E. P. Scheidegger, M. J. Warhol, D. Bitter-Suermann, and P. U. Heitz. 1993. Expression of polysialic acid in human tumors and its significance for tumor growth, p. 335–348. In J. Roth, U. Rutishauser, and F. A. Troy II (ed.), Polysialic Acid: from Microbes to Man. Birkhauser Verlag, Basel.
173. Roth, J., C. Zuber, P. Wagner, D. J. Taatjes, C. Weisgerber, P. U. Heits, C. Goridis, and D. Bitter-Suermann. 1988. Reexpression of poly(sialic acid) units of the neural cell adhesion molecule in Wilms tumor. Proc. Natl. Acad. Sci. USA 85:2999–3003.
174. Rutishauer, U. 1989. Polysialic acid as a regulator of cell interactions, p. 367–382. In R. U. Margolis and R. K. Margolis (ed.), Neurobiology of Glycoconjugates. Plenum Press, New York.
175. Saxen, H., and M. Hovi. 1983. The effect of antibodies to the enterobacterial common antigen (ECA) on experimental mouse salmonellosis. FEMS Microbiol. Lett. 27:307–312.
176. Schauer, R. 1982. Chemistry, metabolism, and biological functions of sialic acids. Adv. Carbohydr. Chem. 40:131–194.
177. Schmidt, A. M., and K. Jann. 1982. Phospholipid substitution of capsular (K) polysaccharide antigens from Escherichia coli causing extraintestinal infections. FEMS Microbiol. Lett. 14:69–74.
178. Schmidt, G., B. Jann, and K. Jann. 1974. Genetic and immunochemical studies on Escherichia coli O14:K7:H–. Eur. J. Biochem. 42:303–309.
179. Schmidt, G., B. Jann, and K. Jann. 1977. Genetic detrminants of the synthesis of the polysaccharide capsular antigen K27(A) of Escherichia coli. J. Gen. Microbiol. 100:355–361.
180. Schmidt, G., D. Männel, H. Mayer, H. Y. Whang, and E. Neter. 1976. The role of a lipopolysaccharide gene for immunogenicity of the enterobacterial common antigen. J. Bacteriol. 126:579–586.
181. Schmidt, G., H. Mayer, and P. H. Mäkelä. 1976. Presence of rfe genes in Escherichia coli: their participation in biosynthesis of O antigen and enterobacterial common antigen. J. Bacteriol. 127:755–762.
182. Selander, R. K., and J. M. Musser. 1990. Population genetics of bacterial pathogenesis, p. 11–36. In B. Iglewski and V. Clark (ed.), The Bacteria, vol. XI. Molecular Basis of Bacterial Pathogenesis. Academic Press, Inc., New York.
183. Silver, R. P., W. Aaronson, and W. F. Vann. 1987. Translocation of capsular polysaccharides in pathogenic strains of Escherichia coli requires a 60-kilodalton periplasmic protein. J. Bacteriol. 169:5489–5495.
184. Silver, R. P., P. Annunziato, M. S. Pavelka, R. P. Pigeon, L. F. Wright, and D. E. Wunder. 1993. Genetic and molecular analyses of the polysialic acid gene cluster of Escherichia coli K1, p. 59–71. In J. Roth, U. Rutishauser, and F. A. Troy II (ed.), Polysialic Acid: from Microbes to Man. Birkhauser Verlag, Basel.
185. Silver, R. P., C. W. Finn, W. F. Vann, W. Aaronson, R. Schneerson, P. J. Kretschmer, and C. F. Garon. 1981. Molecular cloning of the K1 capsular polysaccharide genes of Escherichia coli. Nature (London) 289:696–698.
186. Silver, R. P., W. F. Vann, and W. Aaronson. 1984. Genetic and molecular analyses of Escherichia coli K1 antigen genes. J. Bacteriol. 157:568–575.
187. Silver, R. P., and E. R. Vimr. 1990. Polysialic acid capsule of Escherichia coli K1, p. 39–60. In B. Iglewski and V. Clark (ed.), The Bacteria, vol. XI. Molecular Basis of Bacterial Pathogenesis. Academic Press, Inc., New York.
188. Simmons, D. A. R., and E. Romanowska. 1987. Structure and biology of Shigella flexneri O antigens. J. Med. Microbiol. 23:289–302.
189. Smith, A. N., G. J. Boulnois, and I. S. Roberts. 1990. Molecular analysis of the Escherichia coli K5 kps locus: identification and characterization of an inner-membrane capsular polysaccharide transport system. Mol. Microbiol. 4:1863–1869.
190. Smith, H. 1983. The elusive determinants of bacterial interference with non-specific host defences. Philos. Trans. R. Soc. London Ser. B 303:99–113.
191. Smith, T., and G. Bryant. 1927. Studies on pathogenic B. coli from bovine sources. II. Mutations and their immunological significance. J. Exp. Med. 46:133–140.
192. Steenbergen, S. M., and E. R. Vimr. 1990. Mechanism of polysialic chain elongation in Escherichia coli K1. Mol. Microbiol. 4:603–611.
193. Steenbergen, S. M., T. J. Wrona, and E. R. Vimr. 1992. Functional analysis of the sialyltransferase complex in Escherichia coli K1 and K92. J. Bacteriol. 174:1099–1108.
194. Stevens, M. P., P. Hänfling, B. Jann, K. Jann, and I. S. Roberts. 1994. Regulation of Escherichia coli K5 capsular polysaccharide expression: evidence for involvement of RfaH in expression of group II capsules. FEMS Microbiol. Lett. 124:93–98.
195. Stevenson, G., B. Neal, D. Liu, M. Hobbs, N. H. Packer, M. Batley, J. W. Redmond, L. Lindquist, and P. Reeves. 1994. Structure of the O antigen of Escherichia coli K-12 and the sequence of its rfb gene cluster. J. Bacteriol. 176:4144–4156.
196. Sutcliffe, J., R. Blumenthal, A. Walter, and J. Foulds. 1983. Escherichia coli outer membrane protein K is a porin. J. Bacteriol. 156:867–872.
197. Suzuki, T., E. A. Gorzynski, and E. Neter. 1964. Separation by ethanol of common and somatic antigens of Enterobacteriaceae. J. Bacteriol. 88:1240–1243.
198. Swartley, J. S., and D. S. Stephens. 1994. Identification of a genetic locus involved in the biosynthesis of N-acetyl-d-mannosamine, a precursor of the (α2–8)-linked polysialic acid capsule of serogroup B Neisseria meningitidis. J. Bacteriol. 176:1530–1534.
199. Takatsuki, A., K. Kohno, and G. Tamura. 1975. Inhibition of biosynthesis of polyisoprenol sugars in chick embryo microsomes by tunicamycin. Agric. Biol. Chem. 39:2089–2091.
200. Timmis, K. N., G. J. Boulnois, D. Bitter-Suermann, and F. C. Cabello. 1985. Surface components of Escherichia coli that mediate resistance to the bactericidal activities of serum and phagocytes. Curr. Top. Microbiol. Immunol. 118:197–218.
201. Tkacz, J. S., and J. O. Lampen. 1975. Tunicamycin inhibition of polysioprenyl N-acetylglucosaminyl pyrophosphate formation in calf-liver microsomes. Biochem. Biophys. Res. Commun. 65:248–257.
202. Troy, F. A. 1992. Polysialylation: from bacteria to brains. Glycobiology 2:5–23.
203. Troy, F. A., I. K. Vijay, M. A. McCloskey, and T. E. Rohr. 1982. Synthesis of capsular polymers containing polysialic acid in E. coli O7-K1. Methods Enzymol. 83:540–554.
204. Troy, F. A., I. K. Vijay, and N. Tesche. 1975 Role of undecaprenyl phosphate in synthesis of polymers containing sialic acid in Escherichia coli. J. Biol. Chem. 250:156–163.
205. Vaara, M. 1992. Eight bacterial proteins, including UDP-N-acetylglucosamine acyltransferase (LpxA) and three other transferases of Escherichia coli, consist of a six-residue periodicity theme. FEMS Microbiol. Lett. 97:249–254.
206. Vahlne, G. 1945. Serological typing of the colon bacteria. Acta Pathol. Microbiol. Scand. Suppl. 62:1–172.
207. Valtonen, M. V., U. M. Larinkari, M. Plosila, V. V. Valtonen, and P. H. Mäkelä. 1976. Effect of enterobacterial common antigen on mouse virulence of Salmonella typhimurium. Infect. Immun. 13:1601–1605.
208. Vann, W. F., M. A. Schmidt, B. Jann, and K. Jann. 1981. The structure of the capsular polysaccharide (K5 antigen) of urinary tract infective Escherichia coli O10:K5:H4. A polymer similar to desulfo-heparin. Eur. J. Biochem. 116:359–364.
209. Vann, W. F., R. P. Silver, C. Abeijon, K. Chang, W. Aaronson, A. Sutton, C. W. Finn, W. Lindner, and M. Kotsatos. 1987. Purification, properties, and genetic location of Escherichia coli cytidine 5'-monophosphate N-acetylneuraminic acid synthetase. J. Biol. Chem. 262:17556–17562.
210. Vann, W. F., G. Zapata, I. S. Roberts, G. J. Boulnois, and R. P. Silver. 1993. Structure and function of enzymes in sialic acid metabolism in polysialic acid producing bacteria, p. 125–136. In J. Roth, U. Rutishauser, and F. A. Troy II (ed.), Polysialic Acid: from Microbes to Man. Birkhauser Verlag, Basel.
211. Vermeulen, C., A. Cross, W. R. Byrne, and W. Zollinger. 1988. Quantitative relationship between capsular content and killing K1-encapsulated Escherichia coli. Infect. Immun. 56:2723–2730.
212. Vimr, E. R. 1991. Map position and genomic organization of the kps cluster for polysialic acid synthesis in Escherichia coli K1. J. Bacteriol. 173:1335–1338.
213. Vimr, E. R., W. Aaronson, and R. P. Silver. 1989. Genetic analysis of chromosomal mutations in the polysialic acid gene cluster of Escherichia coli K1. J. Bacteriol. 171:1106–1117.
214. Vimr, E. R., and S. Steenbergen. 1993. Mechanisms of polysialic acid assembly in Escherichia coli K1, p. 73–92. In J. Roth, U. Rutishauser, and F. A. Troy II (ed.), Polysialic Acid: from Microbes to Man. Birkhauser Verlag, Basel.
215. Vinogradov, E. V., Y. A. Knirel, J. E. Thomas-Oates, A. S. Shashkov, and V. L. L’vov. 1994. The structure of the cyclic enterobacterial common antigen (ECA) from Yersinia pestis. Carbohydr. Res. 258:223–232.
216. Vosberg, H.-P., and H. Hoffmann-Berling. 1971. DNA synthesis in nucleotide-permeable Escherichia coli cells. I. Preparation and properties of ether-treated cells. J. Mol. Biol. 58:739–753.
217. Weisgerber, C., A. Hansen, and M. Frosch. 1991. Complete nucleotide and deduced amino acid sequence of CMP-NeuAc:poly-α-2,8-sialosyl sialyltransferase of Escherichia coli K1. Glycobiology 1:357–363.
218. Weisgerber, C., and F. A. Troy. 1990. Biosynthesis of the polysialic acid capsule in Escherichia coli K1. The endogenous acceptor of polysialic acid is a membrane protein of 20 kDa. J. Biol. Chem. 265:1578–1587.
219. Whang, H. Y., H. Mayer, G. Schmidt, and E. Neter. 1972. Immunogenicity of the common enterobacterial antigen produced by smooth and rough strains. Infect. Immun. 6:533–539.
220. Whitfield, C., D. A. Adams, and F. A. Troy. 1984. Biosynthesis and assembly of the polysialic acid capsule in Escherichia coli K1. Role of a low density vesicle fraction in activation of the endogenous synthesis of sialyl polymers. J. Biol. Chem. 259:12769–12775.
221. Whitfield, C., R. E. W. Hancock, and J. W. Costerton. 1983. Outer membrane protein K of Escherichia coli: purification and pore forming properties in lipid membranes. J. Bacteriol. 156:873–879.
222. Whitfield, C., J. C. Richards, M. B. Perry, B. R. Clarke, and L. L. MacLean. 1991. Expression of two structurally distinct D-galactan O antigens in the lipopolysaccharide of Klebsiella pneumoniae serotype O1. J. Bacteriol. 173:1420–1431.
223. Whitfield, C., G. Schoenhals, and L. Graham. 1989. Mutants of Escherichia coli O9:K30 with altered synthesis and expression of capsular K antigen. J. Gen. Microbiol. 135:2589–2599.
224. Whitfield, C., and F. A. Troy. 1984. Biosynthesis and assembly of the polysialic acid capsule in Escherichia coli K1. Activation of sialyl polymer synthesis in inactive sialyltransferase complexes requires protein synthesis. J. Biol. Chem. 259:12776–12780.
225. Whitfield, C., and M.A. Valvano. 1993. Biosynthesis and expression of cell-surface polysaccharides in gram-negative bacteria.Adv. Microb. Physiol. 35:135–246.
226. Whitfield, C., E. R. Vimr, J. W. Costerton, and F. A. Troy. 1985. Membrane proteins correlated with expression of the polysialic acid capsule in Escherichia coli K1. J. Bacteriol. 161:743–749.
227. Wunder, D. E., W. Aaronson, S. F. Hayes, J. M. Bliss, and R. P. Silver. 1994. Nucleotide sequence and mutational analysis of the gene encoding KpsD: a periplasmic protein involved in transport of polysialic acid in Escherichia coli K1. J. Bacteriol. 176:4025–4033.
228. Wyle, F. A., M. S. Artenstein, B. L. Brandt, E. C. Tramont, D. L. Kasper, P. L. Altieri, S. L. Berman, and J. P. Lowenthal. 1972. Immunologic response of man to group B meningococcal polysaccharide vaccines. J. Infect. Dis. 126:514–522.
229. Yao, Z., and M. A. Volvano. 1994. Genetic analysis of the O-specific lipopolysaccharide biosynthesis region (rfb) of Escherichia coli K-12 W3110: identification of genes that confer group 6 specificity to Shigella flexneri serotypes Y and 4a. J. Bacteriol. 176:4133–4143.
230. Zapata, G. M., M. Crowley, and W. F. Vann. 1992. Sequence and expression of the Escherichia coli K1 neuC gene product. J. Bacteriol. 174:315–319.
231. Zapata, G., W. F. Vann, W. Aaronson, M. S. Lewis, and M. Moos. 1989. Sequence of the cloned Escherichia coli K1 CMP-N-acetylneuraminic acid synthetase gene. J. Biol. Chem. 264:14769–14774.
Chapter9
