Capsules of Escherichia coli, expression and biological significance
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by UNIVERSITY OF OREGON on 12/10/17 For personal use only.
KLAUSJANNAND BARBARA JANN Max-Planck-Institut fur Immunbiologie, Stiibeweg 51, Freiburg-Zahringen, Germany Received May 19, 1991 Revision received December 4, 1991 Accepted January 3, 1992 JANN,K., and JANN,B. 1992. Capsules of Escherichia coli, expression and biological significance. Can. J. Microbiol. 38: 705-710. Escherichia coli may cause intestinal or extraintestinal infections. Generally, extraintestinalE. coli are encapsulated. The capsules are important virulence determinants, which enable the pathogenic bacteria to evade or counteract the unspecific host defense during the early (preimmune) phase of infection. They interfere with the action of complement and phagocytes. This effect is generally transient and overcome by capsule-specific antibodies in the immune phase of the host defense. In some cases, capsules are not or only poorly immunogenic, as a result of structural relationship or identity with host material. Strains with such capsules (e.g., K1 or K5) are very virulent. Bacterial capsules consist of acidic polysaccharides, which are made up from oligosaccharide repeating units. The capsules of E. coli are divided into two groups, which differ in chemistry, biochemistry, and genetic organization. All capsular polysaccharides are chromosomally determined: those of group I close to his and those of group I1 close to serA. The biosynthesis and surface expression have been extensively studied with representatives of group I1 capsular polysaccharides. It could be shown that their biosynthesis is directed from a gene block that determines the synthesis of the polysaccharide, its translocation across the cytoplasmic membrane, as well as its surface expression in a coordinate process. The chemical nature of group I1 capsular polysaccharides, as well as the mechanism(s) of their biosynthesis and expression, is presented. Key words: Escherichia coli, capsular polysaccharides, structure, genetics, biology. JANN,K., et JANN,B. 1992. Capsules of Escherichia coli, expression and biological significance. Can. J. Microbiol. 38 : 705-710. Escherichia coli est responsable d'infections intestinales et extraintestinales. En gCnCral, les souches extraintestinales d'E. coli sont capsulCes. Les capsules ont des facteurs de virulence importants qui permettent aux bactCries pathogknes d'kchapper ou de s'opposer aux dCfenses non spkcifiques de l'h8te durant les premittres phases (prCimmunes) de l'infection. La capsule nuit a l'action du complCment et des cellules phagocytaires. Cet effet est gknkralement temporaire et surmontC par les anticorps capsulaires produits dans la phase immune de dCfense de ]'hate. Dans certains cas les capsules sont peu ou pas immunogknes a cause de leur identit6 ou de leur structure apparentke a l'h8te. Les souches qui posskdent de telles capsules (p. ex. Kl ou K5) sont hautement virulentes. Les capsules bactkriennes contiennent des polysaccharides acides formCs d'unitCs rCpCtitives d'oligosaccharides. Les capsules d'E. coli appartiennent a deux groupes selon leurs propriCtCs chimiques et biochimiques et leur organisation gCnCtique. Tous les polysaccharides capsulaires sont sous contr8le chromosomique : ceux du groupe I sont prtts de his et ceux du groupe I1 prtts de serA. La biosynthttse et l'expression de surface ont CtC CtudiCes de faqon extensive avec des polysaccharides reprksentatifs du groupe 11. I1 a CtC dCmontrC que leur biosynthkse est dirigCe par un groupe de genes qui contralent selon un processus coordonnC la synthttse du polysaccharide, sa translocation a travers la membrane cytoplasmique et son expression a la surface bactkrienne. La nature chimique des polysaccharides capsulaires du groupe I1 ainsi que le ou les mkcanismes de leur biosynthttse et de leur expression sont discutCs dans la prCsente Ctude. Mots clks : Escherichia coli, polysaccharides capsulaires, structure, gCnCtique, biologie. [Traduit par la rkdaction]
Introduction All living cells are surrounded by complex carbohydrates. Eukaryotic cells express glycoproteins and glycolipids, which play a role in cell-cell or cell-ligand interactions and which are frequently also interacting with the extracellular matrix. Gram-negative bacteria contain in their outer membrane lipopolysaccharides, which are absent from Gram-positive bacteria. Both, Gram-negative and Gram-positive bacteria frequently express capsules, which consist of polysaccharides. Lipopolysaccharides and capsular polysaccharides are important virulence determinants, which interfere with the host defense against pathogenic bacteria. Encapsulation is especially important in the virulence of extraintestinal and invasive bacteria. For this reason, bacterial capsules are the subject of extensive studies. Clinical, biological, biochemical, and genetic investigations have been performed for many years with encapsulated bacteria and with the isolated capsular polysaccharides. In this article, which deals with the chemistry, expression, and biological significance Printed in Canada / lmprime au Canada
of capsules, we restrict ourselves to the capsules of
Escherichia coli.
Chemical characterization About 70 distinct capsular (K) antigens are known today. They are acidic polysaccharides, which consist of oligosaccharide repeating units and differ in constituents, branching, and charge density. We have classified the capsular polysaccharides o n the basis of the nature of the acidic component, charge density, substitution with lipid at the reducing end, and the mode of expression at different growth temperatures. The differentiation was also based on the coexpression of the capsular polysaccharides (K antigens) with certain lipopolysaccharides (0 antigens). In these studies, two groups of capsular polysaccharides could be defined (Table 1) (Jann and Jann 1990). Group I can be divided further into polysaccharides containing amino sugars and those that do not contain amino sugars and thus resemble the capsular polysaccharides of Klebsiella. Similarly, poly-
CAN. J. MICROBIOL. VOL. 38, 1992
TABLE1. Grouping of capsular polysaccharide antigens of E. coli Capsular polysaccharide group
I
Property
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by UNIVERSITY OF OREGON on 12/10/17 For personal use only.
Acidic component Expressed below 20" C Coexpression with: Lipid at the reducing end Removal of lipid at pH 5-6, 100°C Chromosomal determination at (close to) CMP-KDO synthetase activity elevateda Intergeneric relationship with:
Glucuronic acid, galacturonic acid (pyruvate) Yes 08, 09, 020 Core-lipid A No rfl(his), rfc(trp) No Klebsiella
I1 Glucuronic acid, NeuNAc, KDO, ManNAcA, phosphate No many 0 antigens Phosphatidic acid Yes kpsA (serA ) Yes H. influenzae, N. meningitidis
'In the cytoplasmic fraction of capsular-expressing E. coli.
saccharides of group I1 were divided into groups according to the nature of their acidic components. Some examples are given in the following: A number of group I1 polysaccharides contain 2-keto-3deoxymanno-octonic acid (KDO). Of 13 of these polysaccharides, 9 consist only of KDO and ribose. Two of the KDO-containing capsular polysaccharides (K 14 and K 15) (Jann and Jann 1990) are structurally related to capsular polysaccharides of Neisseria meningitidis. Several capsular polysaccharides are polysialic acids. The Kl polysaccharide contains N-acetylneuraminic acid in a-2,8 linkage and is identical with the capsular polysaccharide of N. meningitidis b. The K92 polysaccharide contains N-acetylneuraminic acid in alternating a-2,8 and a-2,9 linkages. A capsular polysaccharide containing only a-2,9 sialyl linkages, such as the capsular polysaccharide of N. meningitidis c, has not been found in E. coli. A large number of group I1 capsules contain phosphate (Jann and Jann 1990). Some are poly-N-acetylamino sugar phosphates, related to the capsular polysaccharide of N. meningitidis a; others contain polyol phosphate. The K18 and KlOO antigens are polyribosyl ribitol phosphates (Rodriguez et al. 1988a) and are closely related to the capsular polysaccharide of Haemophilus influenzae b. Molecular modelling of these polymers (M .-L. Rodriguez, K. Himmelspach, K.D. Kroncke, A. Kmety, K. Jann, B. Jann, and A. Neszmelyi, to be published) indicated that partial structures, which appear very similar on inspection of the primary structure, expose different epitopes in the helical secondary structure. Thus, computer-aided modelling may be an important tool for the analysis of the biological significance of polysaccharides. Two interesting group I1 capsular polysaccharides are the K4 antigen and the K5 antigen. Like the K1 antigen they are related to structures expressed by the host and play an important role in the infection immunity against the respective bacteria (see section Biological significance). The structures of these polysaccharides are given in Table 2. A characteristic feature of group I1 polysaccharides is their substitution with phosphatidic acid at the reducing end (Gotschlich et al. 1981; Schmidt and Jann 1982). Not all molecules of a given polysaccharide preparation are substituted, and the rest of the molecules have a free reducing end. The fatty acid composition of the phosphatidic acid substituent does not differ from E. coli membrane
phospholipids. There is evidence that (at least some of the) group I capsular polysaccharides are substituted at the reducing end with core-lipid A (Jann et al. 1992). Also in these cases, the substituent is not present on all molecules. It is assumed that the lipid substitutions may serve to anchor the capsular polysaccharides to the outer membrane of the bacteria with hydrophobic interactions. Thus, the lipid moiety may be important in the maintenance of the capsule.
Biosynthesis and surface expression Genetic background The biosynthesis (polymerization) of capsular polysaccharides has been studied with Klebsiella, E. coli, and N. meningitidis (Troy et al. 1971, 1975; Rohr and Troy 1980; Masson and Holbein 1985; Vann et al. 1978). As shown in Table 1, the genes determining the expression of E. coli group I and group I1 capsular polysaccharides are located at different (not allelic) sites on the bacterial chromosome (Qrskov et al. 1976; Vimr 1990). It was shown that these gene clusters do not hybridize in vitro (I.S. Roberts, personal communication). Whereas little is known about the genetics of group I capsules, recent advances in the genetic analysis of group I1 and related capsules, especially with E. coli, N. meningitidis, and H. influenzae (Boulnois and Roberts 1990; Boulnois and Jann 1989; Silver et al. 1984; Kroll et al. 1989, 1990; Vimr et al. 1989; Steenbergen and Vimr 1990) have revealed the organization and function of these capsule genes. It was found that the capsule genes of E. coli are organized in three regions. The central region 2 directs the polymerization of the respective capsular polysaccharide; it is thus specific for a given capsule; region 2 is flanked by region 3, which directs the translocation of the polysaccharide across the cytoplasmic membrane, and region 1, which directs the transport through the periplasm and the outer membrane into the capsular compartment. Both regions 1 and 3, which serve general functions, can be exchanged between different E. coli with group I1 capsules. It was found later that region 1 encodes also other functions than those of transport (see below). The following discussion of biosynthesis and cellular expression of E. coli capsules will be set against this genetic background. Polymerization For the synthesis of a regular polysaccharide built from repeating units, the assembly of preformed oligosaccharides
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by UNIVERSITY OF OREGON on 12/10/17 For personal use only.
JANN A,ND JANN
by specific polymerases seems a reasonable proposition. Such a mechanism had been found for the biosynthesis of a group I related capsular polysaccharide of Klebsiella (Troy et al. 1971) and was also reported for certain Salmonella lipopolysaccharides (for a review see Jann and Jann, 1984). These polymerization reactions utilized undecaprenol phosphate as an intermediary acceptor. Little is known about the polymerization of group I capsules. There are indications of .the participation of lipid-linked oligosaccharides in the synthesis of the K30 polysaccharide (D.F. Chandler and K. Jann, unpublished). A trp-linked polymerase with a map position equivalent to that of the polymerase gene rfc in Salmonella was postulated for the K27 polysaccharide (Schmidt et al. 1977). This would indicate a participation of undecaprenyl pyrophosphoryl oligosaccharides in the synthesis of this polysaccharide. Such a polymerase seems to have a different chromosomal location in E. coli K30 (Laakso et al. 1988). The biosynthesis of group I1 polysaccharides has been studied with the K1 and K5 polysaccharides. Polymerization with participation of undecaprenol phosphate was found for the Kl polysaccharide (Troy et al. 1975; Rohr and Troy 1980) and for the structurally identical capsular polysaccharide of N. meningitidis (Masson and Holbein 1985). In contrast, no lipid participation and no intermediary formation of oligosaccharides could be detected in the polymerization of .the K5 polysaccharide (Finke et al. 1992). Here, the direct transfer of the sugar constituents (glucuronic acid and N-acetylglucosamine) was postulated. These findings show that the polymerization mechanism of polysaccharides does not seem to be reflected in their gene organization. Interestingly, both the K1 and K5 polysaccharides are elongated at their nonreducing ends. In an analysis of the capsular K5 polysaccharide we found that its reducing end is KDO (Finke et al. 1992). KDO had previously been found as reducing end of several other group I1 polysaccharides (Schmidt and Jann 1982), which however, had this constituent also in their repeating units (see section Chemical characterization). These findings are correlated with the recent observation that E. coli strains expressing the K5 polysaccharide have an elevated activity of CMPKDO synthetase, as compared with uncapsulated strains and rough strains. In recombinant E. coli K5 this enzyme activity is even further elevated. A strict correlation could be found between the activity of CMP-KDO synthetase and capsule expression and the polymerizing activity of the respective membrane preparations. Interestingly the CMP-KDO synthetase had an elevated activity only at capsule-permissive temperatures and not below (Finke et al. 1989). The respective gene was located to the general region 1 of the capsule genes and, in conjunction with this finding, all E. coli with group I1 capsules have an elevated activity of CMP-KDO synthetase (Finke et al. 1990). This indicated that CMPKDO is engaged in a reaction essential for the biosynthesis of group I1 capsules. We have recently isolated a CMPKDO synthetase from recombinant E. coli harboring the CMP-KDO gene in the capsule gene region 1 (C. Pazzani, I.S. Roberts, D. Bronner, C. Rosenow, K. Jann, and G.T. Boulnois, unpublished) and found (C. Rosenow and K. Jann, unpublished) that it is distinct from the enzyme that is engaged in lipid A biosynthesis, which had been described before (Goldman et al. 1986). Thus, E. coli expressing group I1 capsules contain two versions of CMP-
707
KDO synthetase, one acting in lipid A synthesis and the other acting in the synthesis of group I1 capsules of E. coli. Cellular expression It was found with E. coli K1 (F.A. Troy, personal communication) and E. coli K5 (A. Finke et al. 1991) that the polymerization of the capsular polysaccharides occurs at the cytoplasmic side of the cytoplasmic membrane. The export of capsular polysaccharides thus consists of at least two stages, viz., the translocation across the cytoplasmic membrane and subsequent transport to the cell surface. We have analyzed both processes with expression mutants of E. coli K5. In immunogold electron microscopy of a region 3 mutant the polysaccharide was labelled in the cytoplasm. The extracted polysaccharide was more heterogenous and of a smaller average size than that isolated from the K5 capsule, and it had no phospholipid substitution (Kroncke et al. 1990a). The same experiment was performed with E. coli Kl , K5, and K 12 with deenergized membranes. Again, the polysaccharides were detected in the cytoplasm (Kroncke et al. 1990b). These results showed that (i) the translocation is effected by products of gene region 3, (ii) it is an energyrequiring process, and (iii) the polymerization of the K5 polysaccharide seems to be influenced by its translocation. The membranes of this mutant had a reduced polymerization activity (about 30% of those from the intact clone. Two different region 1 mutants were analyzed, and the results showed that the mutations had drastically different effects. In one mutant (Kroncke et al. 1990a) the polysaccharide was labelled with the immunogold electron microscopic method in the periplasm of the cells. The isolated polysaccharide had normal chain length and was substituted with the phospholipid. This indicated that substitution of the polysaccharide with the phosphatidic acid occurs during or immediately after its translocation through the cytoplasmic membrane. The membranes of this region 1 mutant had a normal polymerization activity. In the second region 1 mutant (D. Bronner and K. Jann, unpublished) .the polysaccharide was labelled with .the immunogold electron microscopic method in distinct areas of th cytoplasm. The isolated polysaccharide had normal chain length and did not contain detectable amounts of the phospholipid substituent. The membranes of this mutant had a very low polymerizing activity (about 1% of those from the intact clone). We assume that in this region 1 mutant a protein is lacking that is essential for the polymerization of the K5 polysaccharide and its translocation across the cytoplasmic membrane. Possibly, this protein anchors the enzymes of the polymerization complex to the (region 3 directed) translocation proteins. To gain insight into the cellular topography of capsular surface expression, we studied E. coli K l , K5, and K12, which had been grown at a capsule-restrictive temperature, shifted to 37"C, and plasmolyzed at the onset of de novo capsule expression. Immunoelectron microscopy (Kroncke et al. 1990b) showed that in these cells the capsular polysaccharides were exported at discrete sites. Such sites had been characterized as membrane adhesion sites (Bayer 1968). Membrane adhesion sites were also shown to be the export sites of the K29 polysaccharide (group I) (Bayer and Thurow 1977) and of lipopolysaccharides (Miihlradt et al. 1973). The observed similar export mode of these different surface
CAN. J. MICROBIOL. VOL. 38, 1992
TABLE2. Structures of the K1, K4-, and K5 capsular polysaccharides of E. coli Polysaccharide
Structure
Reference
K1 K5 K4
-- 8)-aNeuNAc-(2 --- 4)-PGlcA-(I-- 4)-aGlcNAc-(1 --- 3)-PGlcA-(I-- 4)-PGalNAc-(I--
Orskov et al. 1979 Vann et al. 1981 Rodriguez et al. 1988b
3
I
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by UNIVERSITY OF OREGON on 12/10/17 For personal use only.
2 PFru
polysaccharides allows the conclusion that surface exposition may be similar even if the transported molecules are differently synthesized and their biochemical pathways are directed from differently organized genes. Biological significance The role of capsules in the unspecific host defense The interaction of bacterial surfaces with the complement system is an important mechanism with which invasive bacteria can evade the unspecific host defense. This leads to an impairment or abatement of bacteriolysis and (or) phagocytosis which are effected by the terminal membrane attack complex MAC and complement component C3b, respectively. Normally the complement system is activated either by the interaction with antibody (classical pathway) or by triggering via the properdin system (alternative pathway) (Loos 1985; Cross 1990). Although the mechanisms with which bacterial capsules interfere with the activities of the complement system are not really clear, several findings should be mentioned that are helpful in the attempt to understand the complicated interplay between parasite and host defense. Both classical and alternative complement pathways are subject to regulation and control. As far as we know today, it is the interference of capsular polysaccharides with regulatory proteins that upsets complement action. A pivotal compound, joining both pathways, is factor C3. Its conversion to the active form C3b is achieved by the classical convertase C2bC4b, which is under the control of a C4 binding protein, C4bp (Nussenzweig and Melton 1981). Some capsular polysaccharides seem to bind C4bp and thus interrupt the complement cascade (Cross 1990). Activation of the central component C3 to C3b by the alternative pathway is controlled, inter alia, by factors B and H. These regulate the affinity of C3b to bacterial cell surfaces. Some capsules decrease the affinity of factor B to C3b and others increase its affinity to factor H (Loos 1985; Cross 1990). In both cases, the surface deposition of complement (opsonic C3b or lytic MAC) is decreased and the encapsulated bacteria escape the bactericidal activity of the complement system. It appears as if capsules consisting of or containing sialic acid (the K1 and K92 polysaccharides of E. coli, the b- and c-type capsular polysaccharides of N. meningitidis, capsules of group B Streptococci) exert their anticomplementary action in this way (Cross 1990; Stevens et al. 1978; Wessels et al. 1989). A somewhat simpler interpretation, which may be given in addition to the above, is that encapsulated bacteria bind the complement components too far away from the cell surface to be harmful (Rowley 1954; Cross 1990). Such an inter-
pretation has also been given for the complement resistance of certain uncapsulated E. coli and Salmonella (Joiner 1985). It was shown recently (Cross 1990) that the role of E. coli capsules as virulence determinants cannot be generalized. Thus, whereas the K1 capsule interacts with complement as described, the K5 capsule does not seem to add to serum resistance of the corresponding E. coli; K5-negative mutants have the same serum resistance as their encapsulated parent strains with the same 0 antigen (Cross 1990). It is possible that complement sensitivity is dependent not only on the chemical nature of the capsular polysaccharide but also on that of the cell wall lipopolysaccharide. Thus, it may be the combination of capsular polysaccharides (K antigens) with lipopolysaccharides ( 0 antigens) that decides the outcome of host-parasite interaction in infection immunity. In this context, it may be of interest that only a limited number of OK combinations is encountered with invasive E. coli. In addition, the possible induction of cytokins by certain capsules (Cross 1990; Harvey et al. 1987) may exert the opposite effect of enhancing the bactericidal activity of phagocytic cells. Zmm unogenicity of bacterial capsules Although bacterial capsules generally induce the formation of anticapsular antibodies, which activate the host defense, their immunogenicity is often age related. Thus, many encapsulated E. coli, Neisseria, Haemophilus, or group B streptococci are not immunogenic in newborns or young children (Jennings 1990). This is amply documented in the age-related susceptibility to infection with encapsulated invasive bacteria and in the poor response of infants to many polysaccharide vaccines. Certain bacterial capsules are not or only poorly immunogenic also in adults. Structural elucidation of these capsules has revealed that this is due to identity of the capsules with certain substances found in the host. The K1 capsular polysaccharide of E. coli, a poly-a-2,8-sialic acid, has .the same structure as the carbohydrate terminus of the embryonic neural cell adhesion molecule n-CAM (Finne et al. 1983). By the same token, the E. coli K5 capsular polysaccharide, a heteropolysaccharide consisting of 4 4 glucuronic acid and 4-a-N-acetylglucosamine (Vann et al. 1981), is identical with the first polymeric intermediate of heparin (Navia et al. 1983). The K4 capsular polysaccharide of E. coli (structure in Table 2) can be described as a fructosylated chondroitin (Rodriguez et al. 1988b). Its relatively poor immunogenicity may be due to the lability of the fructosyl linkage. Thus, growth of the bacteria in body compartments of low pH or their processing in phagolysosomes may convert the bacteria from a form in which they induce and react with anti-K4 antibodies to one in which
JANN AND JANN
Kroncke, K.-D., Golecki, J.R., and Jann, K. 1990b. Further electron microscopic studies on the expression of Escherichia coli group I1 capsules. J. Bacteriol. 172: 3469-3472. Laakso, D.H., Homonylo, M.K., Wilmot, S.J., and Whitfield, C. 1988. Transfer and expression of the genetic determinants for 0 and K antigen synthesis in Escherichia coli 09:K(A)30 and Klebsiella sp. 0 1 :K20, in Escherichia coli K12. Can. J. Microbiol. Bayer, M.E. 1968. Areas of adhesion between wall and membrane 34: 987-992. of Escherichia coli. J. Gen. Microbiol. 53: 395-404. Bayer, M.E., and Thurow, H. 1977. Polysaccharide capsule of Loos, M. 1985. The complement system: activation and control. Curr. Top. Microbiol. Immunol. 121: 7- 18. Escherichia coli: microscope study of its size, structure, and sites of synthesis. J . Bacteriol. 130: 91 1-936. Masson, L., and Holbein, B.E. 1985. Role of lipid intermediate(s) Boulnois, G. J., and Jann, K. 1989. Bacterial polysaccharide in the synthesis of serogroup B Neisseria meningitidis capsular synthesis, export and evolution of structural diversity. Mol. polysaccharide. J. Bacteriol. 161: 861-867. Microbiol. 3: 1819- 1823. Miihlradt, P.F., Menzel, J., Golecki, J.R., and Speth, V. 1973. Boulnois, G.J., and Roberts, I.S. 1990. Genetics of capsular Outer membrane of Salmonella. Sites of export of newly synthesized lipopolysaccharide on the bacterial surface. Eur. J . polysaccharide production in bacteria. Curr. Top. Microbiol. Biochem. 35: 47 1-481. Immunol. 150: 1-18. Navia, J.L., Riesenfeld, J., Vann, W.F., et al. 1983. Assay of NCross, A.S. 1990. The biological significance of bacterial encapacetylheparosan deacetylase with a capsular polysaccharide from sulation. Curr. Top. Microbiol. Immunol. 150: 87-95. Escherichia coli K5 as a substrate. Anal. Biochem. 135: 134-140. Finke, A., Roberts, I., Pazzani, C., et al. 1989. Activity of CMPNussenzweig, V., and Melton, R. 1981. Human C4 binding proKDO synthetase in Escherichia coli expressing the capsular K5 tein (C4bp). Methods Enzymol. 80: 124- 133. polysaccharide-implication for biosynthesis of the K5 polysacOrskov, I., Sharma, V., and Orskov, F. 1976. Genetic mapping charide. J. Bacteriol. 171: 374-379. of the K1 and K4 antigens (L) of Escherichia coli. Non-allelism Finke, A., Jann, B., and Jann, K. 1990. CMP-KDO-synthetase of K(L) antigens with K antigens of 08:K27(A), 08:K8(L) and activity in Escherichia coli expressing capsular polysaccharides. 09:K57(B). Acta Pathol. Microbiol. Scand. Sect. B, Microbiol. FEMS Microbiol. Lett. 69: 129- 134. 84: 125-131. Finke, A., Bronner, D., Nikolaey, A.V., Jann, B., and Jann, K. Orskov, F., Orskov, I., Sutton, A., et al. 1979. Form variation 1991. Biosynthesis of the E. coli K5 polysaccharide, a represenin Escherichia coli K1: determined by 0-acetylation of the tative of group I1 capsular polysaccharides: polymerization in capsular polysaccharide. J. Exp. Med. 149: 669-685. vitro and characterization of the product. J. Bacteriol. 173: Rodriguez, M.-L., Jann, B., and Jann, K. 1988a. Comparative 4088-4094. structural elucidation of the K18, K22, and KlOO antigens of Finne, J., Leinonen, M., and Makela, P.H. 1983. Antigenic Escherichia coli as related ribosyl-ribitol phosphates. Carbohydr . similarities between brain components and bacteria causing Res. 173: 243-253. meningitis. Lancet, 2(8346): 355-357. Rodriguez, M.-L., Jann, B., and Jann, K. 1988b. Structure and Goldmann, R.C., Bolling, T.J., Kohlbrenner, W.E., et al. 1986. Primary structure of CTP:CMP-3-deoxy-D-manno-octulosonate serological characteristic of the capsular K4 antigen of Escherichia coli 05:K4:H4, a fructose containing polysaccharide cytidyltransferase (CMP-KDO-synthetase) from Escherichia coli. with a chondroitin backbone. Eur. J. Biochem. 177: 117-124. J . Biol. Chem. 261: 15 831 - 15 835. Rohr, T.E., and Troy, F.A. 1980. Structure and biosynthesis of Gotschlich, E.C., Fraser, B.A., Nishimura, O., et al. 1981. Lipid surface polymers containing polysialic acid in Escherichia coli. on capsular polysaccharides of gram-negative bacteria. J. Biol. J . Biol. Chem. 255: 2332-2342. Chem. 256: 8915-8921. Rowley, D. 1954. The virulence of strains of Bacterium coli for Harvey, W., Kamin, S., Meghji, S., and Wilson, M. 1987. mice. Br. J. Exp. Pathol. 35: 528-538. Interleukin-1-like activity in capsular material from Haemophilus Schmidt, G., Jann, K., Orskov, I., and Orskov, F. 1977. Genetic actinomyces comitans. Immunology, 60: 4 15-4 18. determinants of the synthesis of the polysaccharide capsular Jann, K., and Jann, B. 1984. Structure and biosynthesis of 0 antigen K27 (A) of Escherichia coli. J. Gen. Microbiol. 100: antigens. In Chemistry of endotoxin. Edited by E.T. Rietschel. 355-361. (Handbook of endotoxin. Vol. 1). Elsevier, Amsterdam. Schmidt, M.A., and Jann, K. 1982. Phospholipid substitution of pp. 138-186. capsular (K) polysaccharide antigens from Escherichia coli Jann, B., and Jann, K. 1990. Structure and biosynthesis of the causing extraintestinal infections. FEMS Microbiol. Lett. 14: capsular antigens of Escherichia coli. Curr . Top. Microbiol. 69-74. Immunol. 150: 19-42. Silver, R.P., Vann, W.F., and Aaronson, W. 1984. Genetic and Jennings, H. J . 1990. Capsular polysaccharides as vaccine canmolecular analyses of Escherichia coli K 1 antigen genes. didates. Curr . Top. Microbiol. Immunol. 150: 97-127. J . Bacteriol. 157: 568-575. Joiner, K.A. 1985. Studies on the mechanism of bacterial resistance Steenbergen, S.M., and Vimr, E.R. 1990. Mechanism of polysialic to complement-mediated killing and on the mechanism of action acid chain elongation in Escherichia coli. Mol. Microbiol. 4: of bactericidal antibody. Curr. Top. Microbiol. Immunol. 121: 603-61 1. 99- 133. Stevens, P., Huang, S.N.H., Welch, W.D., and Young, L.S. 1978. Kroll, J.S., Zamze, S., Loynds, B., and Moxon, E.R. 1989. Restricted complement activation by Escherichia coli with the Common organization of chromosomal loci for the production K1-capsular serotype: a possible role in pathogenicity. J. of different capsular polysaccharides in Haemophilus influenzae. Immunol. 121: 2174-2180. J . Bacteriol. 171: 3343-3347. Troy, F.A., Frerman, F.E., and Heath, E.C. 1971. The biosynKroll, J.S., Loynds, B., Brophy, L.N., and Moxon, E.R. 1990. thesis of capsular polysaccharide in Aerobacter aerogenes. The bex locus in encapsulated Haemophilus influenzae: a J . Biol. Chem. 246: 118-133. chromosomal region involved in capsule polysaccharide export. Troy, F.A., Vijai, I.K., and Tesche, N. 1975. Role of undecaprenyl Mol. Microbiol. 4: 1853-1862. phosphate in synthesis of polymers containing sialic acid in Kroncke, K.-D., Boulnois, G., Roberts, I., et al. 1990a. Expression Escherichia coli. J . Biol. Chem. 250: 156-163. of the Escherichia coli K5 capsular antigen: immunoelectron Vann, W .F., Liu, T.-Y., and Robbins, J.B. 1978. Cell free biosynmicroscopic and biochemical studies with recombinant E. coli. thesis of the 0-acetylated N-acetylneuraminic acid capsular J. Bacteriol. 172: 1085-1091.
they no longer do so. These findings emphasize that capsules counteract the host defense not only by interaction with complement components but also by evasion of immune recognition owing to molecular mimicry.
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by UNIVERSITY OF OREGON on 12/10/17 For personal use only.
709
7 10
CAN. J . MICROBIOL. VOL. 38, 1992
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by UNIVERSITY OF OREGON on 12/10/17 For personal use only.
polysaccharide of group C meningococci. J . Bacteriol. 133: 1300-1306. Vann, W.F., Schmidt, M.A., Jann, B., and Jann, K. 1981. The structure of the capsular polysaccharide (K5 antigen) of urinary tract infective Escherichia coli 010:K5:H4. A polymer similar to desulfo heparin. Eur. J. Biochem. 116: 359-364. Vimr, E.R. 1990. Map position and genomic organization of the kps cluster for polysialic acid synthesis in Escherichia coli K1. J . Bacteriol. 173: 1335-1338.
Vimr, E.R., Aaronson, W., and Silver, R.P. 1989. Genetic analysis of chromosomal mutations in the polysialic acid gene cluster of Escherichia coli K1. J. Bacteriol. 171: 1106-1 117. Wessels, M.R., Craig, E.R., Vicente-Javier, B., and Kasper, D.L. 1989. Definition of a bacterial virulence factor: sialylation of the group B streptococcal capsule. Proc. Natl. Acad. Sci. U.S.A. 86: 8983-8987.