CROSSREACTIONS OF ESCHERICHIA COLI K AND 0 POLYSACCHARIDES IN ANTIPNEUMOCOCCAL AND ANT I -SALMONELLA SERA BY MICHAEL HEIDELBERGER,* KLAUS JANN,* AND BARBARA JANNf From the *New York University Medical Center, Department of Pathology, New Yorh 10016; and the $Max Planck Institut fur Immunobiologie, 4800 Freiburg-Zahringen, Federal Republic of Germany The study of crossreactivity in relation to chemical structure of the microbial polysaccharides has shown both theoretical and practical utility. In this paper, we make use of recently determined structures of E. coli K and 0 polysaccharides to derive new relationships of structure and specificity, and to update some of the data in earlier work. Materials and Methods These have been described in earlier papers (I -3). Pneumococcal (Pn)' type-numbers are given in Roman type to avoid confusion with those of E. coli and Klebsiella. Results and Discussion Data obtained are summarized in Table I. E. coli K Polysaccharides. K2 has 1,4-linked D-galactose (D-gal) and 1,5-1inked D-galf (furan form of D-galactose; all sugars are pyranose form unless otherwise stated) in addition to glycerophosphate residues in its repeating unit (4). Precip- itation (++-e) in anti-Pn XI1 (see Table I) may be caused by a loose fit of glycerophosphate into antibody sites designed for the DN-acetylmannosaminic acid (DmanNAcA) of Pn soluble specific capsular polysaccharide (PnS) XI1 (5, 6). PnS XXIX has two 1,6-linked ~-galf residues in its repeating unit ('7). Since carbons 5 and 6 of galf are outside the furanose ring, it is possible that the 1,5- D-galf of E. coli K2 would fit partially into antibody sites in anti-Pn XXIX designed for 1,6-linked Dgatf residues, to give the ++ precipitation found. There was also ++ in anti-Pn XVI, but the structure of PnS XVI is not known. K4 is omitted from Table I, as it gave only a single + reaction (on a scale of - to ++++) in anti-Pn XXII serum. K5 has the repeating unit $. 4)glcNAc-a-(l + 4)glcA-B-(1j;; (glcNAc, N- acetylglucosamine; glcA, glucuronic acid) (8). Its faint (+) crossprecipitations were in anti-Pn 111, IX, XII, XVI, and XXV. K7 has the repeating unit -f+ 3)-amanNAcA-P-(1 + 4)-~-glc-@-(l+, (glc, glucose) (9). It reacts with so many equine anti-Pn sera that at least some of the ' Ab6rcviattons used in lhrr paper: f, furan form; Pn, pneumococcal; PnS, Pn soluble specific c.~psular polysacchdride. 1350 J. EXP. MED. 0 The Rockefeller University Press . 0022-l007/85/lO/I350/09 $1.00 Volume 162 October 1985 1350-1358 HEIDELBERGER El' AL. 1351 reactions might be due to previous or inapparent infections of the horses by strains of E. coli containing K7 or a crossreacting antigen such as the enterobac- terial common antigen (ECA), which also contains DmanNAcA in its repeating unit (lo), and has shown some unexpected crossreactions. Nevertheless, some of the precipitations appear to be correlated with known structural features of Pn polysaccharides. The 1,3-linked DmanNAcA of K7 probably reinforces a cross- reaction in anti-Pn 111 due to 1,4-linked D-gk, since it would be expected to have some of the serological properties of the DglcA of PnS I11 $, 3)-0-~-GlcA- (1 + 4)-b-~-glc(l+ (sugars are capitalized only when known to be immunodom- inant). For examples of the partial serological equivalence of glc, glcNAc, and manNAcA, cf. previously published data (1 1-1 3). K7 precipitates 242 pg/ml of antibody nitrogen from anti-Pn I11 792C, far more than the +++ in anti-Pn VI11 1008 of higher antibody content. The repeating unit of PnS VI11 is half cellobiouronic acid (14), and the remainder is Dglc + gal (galactose), but all linkages are 1,4, instead of the 1,3-linked D-glcA of PnS 111 $, 3)-~-GlcA-/3-(1 -+ 4)-D-gk-P-(1-+, (15), so that the greater reactivity in anti-Pn 111 is in accord with the three structures. PnS 11, XIV, XV, XIX, XXII, and XXIII contain 1,4-linked D-glc in their repeating units: the crossprecipitations in antisera to these types appear due (1 6) to multiples of these residues, reinforced in the large reaction (99 rg/ml of antibody nitrogen) in anti-Pn XIX by 1,4-1inked D-manNAc (D-N-acetylmannos- amine), a determinant of PnS XlX (17). Precipitation in anti-Pn 1, VI, IX, and X may be caused by inapparent crossreacting E. coli infections of the immunized horses, as indicated above. K8 is made up of gal, glcA, galN (galactosamine), and glcN (glucosamine) (1 8). The reaction was only ++ in anti-Pn XXII, + in anti-Salmonella typhi and anti- S. paratyphi A, and ++ in anti4 paratyphi B. K9 (1 8, 19) contains gal, N-acetylneuraminic acid, and N-acetylgalactosamine (galNAc). It gave a + reaction in anti-Pn 1, 111, VIII, IX, XV, and XXV and ++ in anti-Pn VI, VII, and X. In the last three, common linkages of D-gal and/or galN Ac in the relevant polysaccharides probably give rise to antibodies producing the slight crossprecipitations. K12 (20) showed a + reaction in anti-Pn V and XXVII. K14 (21) and 17 were uniformly - to f and, with K12, are not included in Table I. Immunodominance of partially 0-acetylated 2-keto-3-deoxymannosoc- tonic acid (KDO), which does not occur in PnS, may explain the lack of reactivity of K12 and 14 in anti-Pn sera. K25 is made up of glcA, galNAc, and fucose (fuc) (18, 19). It crossreacts weakly in anti-Pn V and MXXV. Possibly, 1,2-linked D-glcA will be found in K25, as it occurs in PnS V (22); the structure of PnS XXV is not known. K26 contains glcA, gal, and rhamnose (rham) (1 8, 19). Strong crossprecipita- tion in anti-Pn I1 and XXIlI indicates that its repeating units contain nonreduc- ing lateral end groups of D-~~cA, as in PnS 11 (23), and of L-rham, as in PnS XXIII (24-26). With these structural similarities to K85 (see below), there should be reciprocal crossreactivity between K26 and K85 and antisera to these, but this does not seem to have been looked for. The reaction in anti-Pn VI is 1352 CROSSREACTIONS OF E. COLI K AND 0 POLYSACCHARIDES TAB Crossreactions of E. coli Polysaccharides in Antipneumo K2 - K7 ++* 25 26 27 if 28 90 31 (-) 42 8 54 51 +++ 85 * (Of147Ki) 100 ++ OEM + 08LPS (++) 08PS ++* 9 (+)* - - - - 87 (-1 09 (ftf 01s" (+*) 11 4000 V 1060 864 I 792 1 1940 1 1010 * Readings in parentheses: tests made in two or more rows of antisera that were - to ++ or ++e with * E. coli K26 precipitated Ranti-K47 Klebsiella. K1 47, like PnS XXIII, has lateral nonreduciff~end groups * From Heidelberger et at. (37). other polysaccharides. of L-rham. +++ in RXVII. probably due to 1,3-~-rham and/or to 1,2-~-gai. The explanations of these crossreactivities in @rskov et al. (18) are thus extended and modified. K27 has the probable (27) structure: -0Ac E 4 [.GafP 43;; DglC( I -+ 3)-~-glcA-P-( 1 4 ~)-L-~uc( 1 Weak precipj~t~on in anti-Pn I is difficult to expIain. Failure to react in anti- Pn X1V may indicate that -0Ac is on the gal. PnS X (28) and XX (29) have lateral nonreducing end groups of D-galf. Crossreactions in anti-Pn X and XX might indicate that the lateral gal of K27 is also galf, although gal (pyran) and galf are not necessarily noncrossreactive. K28 has (30) the structure: wgal-@-(l ---f 4)A D-gk-ff-( 1 -+ 4>D+gkA-@-( 1 4 $)-L-fuc-a-( f -~.3~~ OAc L-fuc is acetylated at positions 2 or 3 in 70% of the repeating units. It is difficult to see why K28 should give even +f in anti-F'n I1 or why, if reactivity in anti- Pn VI1 is due to the lateral %gal, there is no definite precipitation in anti-Pn HEIDELBERGER ET AL. LE I Antipneumococcal and Anti-Salmonella Sera I353 X \' 770 XVlll XIX 2200 2250 +- - 99 (3 R-) - - -- +- -f (*) (-1 - - f- - - (-) (-1 -to* -to* -to* -to* - - - ++ + ++ (++) (++) (++) (+) (+++) (++) xx XXll 355 878 + f f 11 f ++ (*) - +* f i (-) - (f) - f - - (-) -lo* -to* - to * ++* ++f +* ++ (+++) (+++) +++ +++ I06 (++*) XXlll MXXV 420 640 ** ++* ++' (-) (+) 1Y7* - - +* - +* *+ -* (+++*) -or* - 1523.'. (f) (*) 20' -- -+ - + (++e) -- -* ++* * XXVll 277 XXVlll 785 XXlX 389 I 0Phi ++ + (*) - - - + - - -or* -or* f - (+I - - ++* ++ + ++* (++*) (++*) - - parotppht parolpphi A B + - -or+ -or* - + - +* ++ ++* ++ (++*) (+++) ` Tests with equine anti-Pn XXV 51 3C (New York City Department of Health). ** From Heidelberger et al. (37) table I. footnote d. (278 - 126 = 152). 27 pg after treatment with alkali. * From C. Springer, Northwestern University School of Medicine, Evanston, IL. I' From A. Zweibauni, H6spital Broussais, Paris. XI1 or XIV; the +& reaction in anti-Pn X is probably for the reason given for K27. K29 contains 1,3-linked glucosyl residues in its repeating unit (3 1). These are probably responsible for the ++ in anti-Pn VI. There should also be a heavy crossreaction with Klebsiella (KI) K31, which has the same pyruvyl (Py) 4,6-glc- (1 + 2)-~-man sidechain (32), and perhaps also with KI 36 (33) and KI 64 (34), which also have 4,6-Py-glc nonreducing lateral end groups. Crossreactivity of K30 (35) and K85 (36) in anti-Pn I1 and V, and of K85 in anti-Pn XXIIl was expected, found, measured, and discussed previously (37). It may now be added that, since K85 precipitates more antibody from anti-Pn V (PnS V has 1,2-linked aglcA [22]) than from anti-Pn 11, this would favor assignment of the 1,2-linkage to the nonterminal glcA of K85, rather than the 1,4-linked alternative also proposed (18, 37). Recent work (28) has shown that pure PnS X contains no glcNAc, so that the crossreaction of K85 in anti-Pn X can not be attributed to this sugar, as was done previously (18, 37). Alternatively, the glcA residues of K85 might fit partially into combining sites of anti-Pn X designed for reception of the ribitolphosphate residues of PnS X. Recently (38), Klebsiella K63 was found to contain the same sugars in the same linkages as E. coli K42 (39). Quantitative analyses (I), however, show that K1 63 precipitates 148 gg/ml of antibody nitrogen from anti-Pn I 1057C whereas E. coli K42 gives only 8 gg/ml (37). Although the primary structures of the two polysaccharides are the same, KI 63 is said to contain (0.2 residues of -0Ac per 1354 CROSSREACTIONS OF E. COLI K AND 0 POLYSACCHARIDES repeating unit, while E. coli K42 has 0.5. Assuming that K42 has not been partially depolymerized or degraded, one might explain the analytical data as follows. If the -0Ac of K42 were on galA, and antibodies in anti-Pn I were partly directed against unacetylated galA, the discrepancy would be accounted for. Alternatively, the three-dimensional form of the more highly acetylated K42 might differ from that of KI K63. Strictly, then, E. coli K42 and KI 63 are not identical. K31 (described by K. Jann and B. Jann, unpublished results), with its only heavy crossreaction in anti-Pn XXIII, provides another instance of a 1,2-linked sugar acting serologically much like a nonreducing lateral end group. In the repeating unit of K31, it is the 1,Z-linked L-rham that reacts like the L-rham in K26 (above). K51 is omitted from Table I because the only significant reaction, (++) in anti-Pn XVIII, is not interpretable with the information at hand. K52 is also omitted, as all tests showed little (+) or no (-) crossreactivity. K54 has a unique repeating unit: $, 3)-~-glcA-~-threonylamide-~-( 1 .--, 3)-~- rham-cy-( Itn (40) (occasional units have serine instead of threonine). From its crossprecipitation in anti-Pn I11 and not in anti-Pn VIII, predicted earlier that the glcA would be D-, and 1,3-linked, as in PnS 111, and not 1,4-linked as in PnS VIII. Amidation of glcA seems not to have affected its specificity. 1,3-linked L- rham explains the crossreactivity in anti-Pn VI (41). K57 contains gal, ribose, galA, and galNAc (18). Its only significant (+++) crossprecipitation, in anti-Pn I, appears due to DgalA in a 1,3- or 1,4-linkage, as in PnS I (42). K87 has (43) the repeating unit, f. 4)-D-gkA-B-( 1 + ~)-L-~ucNAc( 1 + 3)t o-glcNac (1 + 6)-~-ga\( 1 D-gk-/3-(1 + 4)f I -0Ac The strongest crossprecipitations (+&) were in anti-Pn VI1 and 18 pg/ml antibody nitrogen in anti-Pn VIII, the latter value increasing to 27 pg after treatment of K87 with alkali. Precipitation of anti-Pn VI11 is undoubtedly caused by the existence of D-glcA in K87 and PnS VI11 (1 4) in 1,4-linkage. KlOO gave crossreactions not exceeding +++ in anti-Pn 1, IV, VI, IX, X, XX, XXII, and anti-S. typhi. Since KlOO is said (44) to be identical to the polysaccha- ride of H. inJuenzae b, $. 3)-D-ribosyl-&(l --., 1)ribitol-5-phosphatefn (45), it is possible that the precipitation in anti-Pn VI and X might be due to the known occurrence of ribitolphosphate as part of the repeating units of PnS VI and X. Or, as in other instances recorded herein, the horses used for production of antisera might have had inapparent infections with a K1 00-containing or cross- reacting microorganism. Three different preparations of 08 were tested, as well as one of 09. Since 60% of the repeating unit of 09 consists of the repeating unit of 08 (46, 47) it is surprising that there is no crossreactivity between these types (48). However, one of the 1,8mannosyl residues on 08 is 0-linked Uann, Jann, and Himmelspach, unpublished observations), which might 0 Polysaccharides and Lipopolysaccharides. HEIDELBERGER ET AL. 1355 bend the molecule into a shape different from that of 09, in which all linkages are a. The reaction of 09 in anti-Pn XI1 may be caused by its multiples of a- 1,2 niannobiosyl residues. Since mannose is partially equivalent serologically to glucose (12), 09 could fit loosely into antibody sites designed to bind the kojibiosyl residues of PnS XII. The number of relatively weak crossprecipitations in anti-Pn sera may be due, as in other instances, to inapparent infection of the immunized horses with 08, 09, or to crossreacting strains. Cells of E. coli 013 yield antisera in rabbits that precipitate heavily with glycogen (49). A glc-containing 0 13 polysaccharide also sent by Dr. A. Zweibaum (Hbpital Broussais, Paris, France) reacts to various extents with anti-Pn I-XX, so that it is questionable whether all of the precipitates involve type-specific antibodies. 01 11 has the structure. (50,51) i. Colitose (1 3)= +--. Colitose (1 --* 6)9 This polysaccharide, from Dr. 0. Westphal (University of Freiburg, Federal Republic of Germany) contains colitose, glc, gal, and glcNAc. It was tested in all listed antisera, giving only very weak (+) to negative (-) reactions, and is therefore omitted from Table I. After degradation with 1% acetic acid, at 100°C for 90 min, which removes both colitose residues from the glc, it precipitated 60 rg/ml of antibody nitrogen from anti-Pn VI11 1008, confirming the presence of 1,4-linked D-glc in 0 1 1 1. 4-~-glc-a-( 1 + 4)-~-gal-a-( 1 + 3)-~-glcNAc-B-( 1 Summary Crossreactions of 24 K polysaccharides and 4 0 polysaccharides of E. coli in antisera to 27 pneumococcal types, 3 anti-Salmonella sera, and anti-Klebsiella K1 serum are discussed in relation to structural features of the polysaccharides insofar as these are known. Predictions based on the crossprecipitations are also ventured for several instances in which structures are as yet undetermined. Received for publication 24 June 1985. References 1. Heidelberger, M., and W. Nimmich. 1976. Immunochemical relationships between bacteria belonging to two separate families: pneumococci and Klebsiella. Immunochem- isty 13:67. 2. Heidelberger, M., and P. A. Rebers. 1958. Cross-reactions of polyglucoses in anti- pneumococcal sera. V1.J. Am. Chem. SOL. 80:116. 3. Heidelberger, M., and J. M. Tyler. 1964. Cross-reactions of pneumococcal types. Quantitative studies with the capsular po1ysaccharides.J. Exp. Med. 120:711. 4. Jann, K., B. Jann, M. A. Schmidt, and W. Vann. 1980. Structure of the Escherichia coli K2 capsular antigen, a teichoic acid-like polymer.]. Bacteriol. 143: 1108. 5. Duke, J. L., I. J. Goldstein, and J. A. Cifonelli. 1974. Structural studies of capsular polysaccharide of type XI1 Diplococcus pneumoniae. Carbohydr. Res. 37:8 1. 1356 CROSSREACTIONS OF E. COLI K AND 0 POLYSACCHARIDES 6. Leontine, K., B. Lindberg, and J. Lonngren. 1981. Structural studies of the capsular polysaccharide from Streptococcus pneumoniae type 12 F. Can. J. Chem. 59:208 1. 7. Rao, E. V., M. J. Watson, J. G. Buchanan, and J. Baddiley. 1969. Type-specific substance from Pneumococcus type 29. Biochem. J. 11 1:547. 8. Vann, W. F., M. A. Schmidt, B. Jann, and K. Jann. 1981. Structure of the capsular polysaccharide (K5 antigen) of urinary tract-infective Escherichia coli 010:K5: 144. A polymer similar to desulfo-heparin. Eur. J. Biochem. 1 16:359. 9. Tsui, F. P., R. A. Boykins, and W. Egan. 1982. Studies of the Escherichia coli K7 (K56) capsular polysaccharide. Carbohydr. Res. 102:263. 10. Dell, A., J. Oates, C. Lugowski, E. Romanowska, L. Kenne, and B. Lindberg. 1984. Enterobacterial common antigen, a cyclic polysaccharide. Carbohydr. Res. 133:95. 11. Heidelberger, M., G. G. S. Dutton, J. Eriksen, W. Nimmich, and S. Stirm. 1982. Additional correlations of chemical structure and immunological specificity among cross-reactions of pneumococci and Klebsiella. Acta Pathol. Microbiol. Immunol. Scand. Sect. C. Immunol. 90237. 12. Heidelberger, M., and M. E. Slodki. 1972. Cross-reactions of Lipomyces in antipneu- mococcal and other antisera. Carbohydr. Res. 24:401. 13. Robbins, J. B., R. L. Myerowitz, J. K. Whisnant, M. Argaman, R. Schneerson, Z. T. Handzell, and E. C. Gotschlich. 1972. Enteric bacteria cross-reactive with Neisseria meningitidis groups A and C and Diplococcuspneumoniae types I and 111. Infect. Immun. 6:65 1. 14. Jones, J. K. N., and M. B. Perry. 1957. Structure of the type VI11 pneumococcus specific polysaccharide. J. Am. Chem. SOC. 79:2787. 15. Reeves, R. E., and W. F. Goebel. 1941. Chemoimmunological studies on the soluble specific substance of pneumococcus: structure of type I11 polysaccharide. J. Bioi. Chem. 139:511. 16. Heidelberger, M., and F. E. Kendall. 1934. Quantitative studies on the precipitin reaction. The role of multiple reactive groups in antigen-antibody union as illustrated by an instance of cross-precipitation. J. Exp. Med. 59:519. 17. Ohno, N., T. Yadomae, and T. Miyazaki. 1980. Structure of the type-specific polysaccharide of Pneumococcus type XIX. Carbohydr. Res. 80:297. 18. Brskov, I., F. Brskov, B. Jann, and K. Jann. 1977. Serology, chemistry, and genetics of 0 and K antigens of Escherichia coli. Bacteriot. Rev. 41:667. 19. Jann, K., and B. Jann. 1983. K antigens of Escherichia coli. Prog. Allergy. 33:53-79. 20. Schmidt, M. A., and K. Jann. 1983. Structure of the 2-keto-3-deoxy-~1-mannooctonic acid-containing capsular polysaccharide (K 12 antigen) of the urinary tract-infective E. coli 04:K 12+:H-: Eur. J. Biochem. 13 1 :509. 2 1. Jann, B., P. Hofmann, and K. Jann. 1983. Structure of 3-deoxy-D-mannooctu~osonic acid (KD0)-containing capsular polysaccharide from Escherichia coli 06:K 14:H3 1. Carbohydr. Res. 120: 13 1. 22. Barker, S. A., S. M. Bick, J. S. Brimacombe, M. J. How, and M. Stacey. 1966. Structural studies on the capsular polysaccharide of pneumococcus type V. Carbohydr. Res. 2: 224. 23. Kenne, L., B. Lindberg, and S. Svensson. 1975. Structure of the capsular polysaccha- ride of pneumococcus Type 11. Carbhydr. Res. 4059. 24. Heidelberger, M., M. Davie, and R. M. Krause. 1967. Cross-reactions of the group- specific polysaccharides of streptococcal groups B and G in antipneumococcal sera with special reference to type XXIII and its determinants.]. Immunol. 99:794. 25. Jones, C. 1985. Identification of the tetrasaccharide repeating unit of the Streptococcus pneumoniae type 23 polysaccharide by high-field proton NMR spectrography. Carbo- hydr. Res. 139:75. HEIDELBERGER ET AL. 1357 26. Heidelberger, M. 1983. Precipitating cross-reactions among pneumococcal types. Infect. Immun. 4 1: 1234. 27. Jann, K., B. Jann, K. F. Schneider, F. Brskov, and I. arskov. 1968. Immunochemistry of K antigens of Escherichia coli. 5. K antigen of E. coli OS:K27(A):H-. Eur.J. Biochem. 5:456. 28. Perry, M. B., V. Daoust, and R. Lowe. 1980. Immunity, immunization, cerebrospinal meningitis. World Health Organization International Conference, Marburg. (Abstr.). 29. Richards, J. C., M. B. Perry, and D. J. Carlo. 1983. The specific capsular polysaccha- ride of Streptococcus pneumoniae type 20. Can. J. Bwchem. Cell Biol. 6 1 : 178. 30. Altman, E., and G. G. S. Dutton. 1985. Capsular polysaccharide from Escherichia coli 09:K28(A)H (K28 antigen). Carbohydr. Res. 138:293. 31. Choy, Y. M., F. Fenmel, N. Frank, and S. Stirm. 1975. Escherichia coli capsule bacteriophages. VI. Primary structure of the bacteriophage 29-receptor, E. coli serotype 29 capsular po1ysaccharide.J. Vtrol. 16:587. 32. Cheng, C.-C., S.-L. Wong, and Y.-M. Choy. 1979. Structure of the capsular polysac- charide of Klebsiella K type 3 1. Carbohydr. Res. 73: 169. 33. Dutton, G. G. S., and K. L. Mackie. 1977. Structural investigation of Klebsiella serotype K36 polysaccharide. Carbohydr. Res. 55:49. 34. Merrifield, E. H., and A. M. Stephen. 1979. Structural studies on the capsular polysaccharide from Klebsiella serotype K 64. Carbohydr. Res. 74:241. 35. Chakraborty, A. K., H. Friebolin, and S. Stirm. 1980. Primary structure of the Escherichia coli serotype K30 capsular polysaccharide. J. Bacteriol. 14 1 :97 1. 36. Jann, K., B. Jam, F. Orskov, and 1. grskov. 1966. Immunchemische Untersuchungen an K-Antigenen von Escherichia coli. HI. lsolierung und Untersuchung der chem- ischen Struktur des sauren Polysaccharides aus E. coli 0141 :K85(B):H4(K85-Anti- gen). Biochem. Z. 346:368. 37. Heidelberger, M., K. Jam, B. Jann, F. arskov, 1. Brskov, and 0. Westphal. 1968. Relations between structures of three K polysaccharides of Escherichia coli and cross- reactivity in antipneumococcal sera. J. Backriol. 95:24 15. 38. Joseleau, J.-F., and M.-E. Marais. 1979. Structure of the capsular polysaccharide of Klebsiella K-type 63. Carbohydr. Res. 7'7: 183. 39. Jann, K., and B., F. and I. Brskov, and 0. Westphal. 1966. Immunchemische Untersuchungen an K-Antigen von Escherichia coli. 11. K-Antigen von E. coli 08:K42(A):H-. Biochem. Z. 3421. 40. Hofmann. P., B. Jann, and K. Jann. 1985. Structure of the amino acid-containing capsular polysaccharide (K54 antigen) from Escherichia coli 06:K54:H 10. Carbohydr. Res. 13926 1. 41. Rebers. P. A., and M. Heidelberger. 1961. The specific polysaccharide of type VI pneumococcus 11. The repeating unit. J. Am. Chem. SOC. 83:3056. 42. Lindberg, B., B. Lindqvist, J. Lonngren, and D. A. Powell. 1980. Structural studies of the capsular polysaccharide from Streptococcus pneumoniue type 1. Carbohydr. Res. 78:11 1. 43. Tarcsay, L., Jann, B., and K. Jann. 1971. Immunochemistry of K antigens of Escherichia coli K 87 antigen from E. coli 08:K87(B)H 19. EUY. J. Biochem. 23:505. 44. Robbins, J. B., R. Schneerson, J. C. Parke, T. Y. Lin, Z. T. Handzel, and F. arskov. 1976. E. coli 075:Kfl47:H-5 cross-reactive with capsular polysaccharide of Haemo- philus injluenrae type b immunochemical relationship and prospects for heteroim- rnunization on humans to H. infuenzae b. In Role of Immunological Factors in Infections. R. F. Beers and E. G. Bassett, editors. Raven Press, New York. 103-120. 45. Branefors-Helander, P., C. Erbing, L. Kenne, and B. Lindberg, 1976. Structural 1358 CROSSREACTIONS OF E. COLI K AND 0 POLYSACCHARIDES studies of the capsular antigen from Haemophilus injluenzae type b. Acta Chem. Scand. Ser. B Org. Chem. Biochem. 30:276. 46. Reske, K., and K. Jann. 1972. 08 antigen of Escherichia coli. Structure of the polysaccharide chain. Eur. J. Biochem. 3 1 :320. 47. Prehrn, P., B. Jann, and K. Jann. 1976. 09 antigen of Escherichia coli. Structure of the polysaccharide chain. Eur. J. Biochem. 67:53. 48. Jann, K., B. Jann, V. Winter, and C. Wolf-Ullisch. 1979. Serological specificity of Escherichia coli 08 and 09 antigens. J. Gen. Microbiol. 110:203. 49. Zweibaurn, A,, M. Rousset, and S. Leon. 1976. Activite antiglycogene de serums de lapins imrnunisis avec une souche d'Escherichia coli 013. Compt. Rend. Acad. Sci. (France) D 283:1815. 50. Edstrom, R. D., and Heath, E. C. 1967. Biosynthesis of cell wall lipopolysaccharides in Escherichia coli. VII. Studies on the 0-antigenic polysaccharides. J. Eiol. Chem. 242:4125. 51. Eklind, K., P. J. Caregg, L. Kenne, A. A. Lindberg, and B. Lindberg. 1978. Structural studies on the Escherichia coli 01 1 1 lipopolysaccharide. Internal. Symp. Carbohydr. Chem. 9:493a (Abstr.).