CA2153733A1 - Immunogenic oligosaccharide compositions - Google Patents

Abstract

The invention provides immunogenic oligosaccharide compositions and methods of making and using them. In particular, the compositions comprise oligosaccharides covalently coupled to carrier protein, wherein the resultant conjugate has been shown to contain specific immunogenic epitopes and elicits a protectively immunogenic response.

Description

IMMUNOGT~,NIC OLIGOSACCHARIDE COMPOSITIONS

Field of the Invention:
This application relates to immnnogenic oligosaccharide compositions and methods of making and using them. In particular, the compositions comprise oligosaccharides covalently coupled to carrier protein, wherein the resultant conjugate elicits an immlln~protective response.

Refe~ellces:
1 0 The following references are cited in the application at the relevant portion of the application.

Anderson, P., Infect. Tmmnn. 39:233, 1983 Anderson, P. W., Immunogenic Conjugates, U.S. Patent No. 4,673,574, 1 5 1987.
Anderson, P., Pichichero, M. E., and Insel, R. A., J. Clin. Invest.
76:52, 1985a.
Anderson, P., Pichichero, M. E., and Insel, R. A., J. Pediatr. 107:346, 1985b.
Anderson, P. W., Pichichero, M. E., Insel, R. A., Betts, R., Eby, R., and Smith, D. J., J. Tmmnnnl. 137:1181, 1986.
Anderson, P. W., Pichichero, M. E., Stein, E.C., Porcelli, S., Betts, R.
F., Connuch, D. M., Korones, D., Insel, R. A., Zabradnick, J. M., and Eby, R., J. Tmmllnol. 142: 2464, 1989.
Avery, O.T. and Goebel, W. F., J. Exp. Med. 50:533, 1929.
Barro, A., Dogan, R., Prued'homme, J. L., Bajart, A., Danue, B. and Fritzell, B., Vaccine, II:1003, 1993.
Bixler, G.S. and Pillai, S., The cellular basis of the immlln~ response to conjugate vaccines in "Conjugate Vaccines. Contributions to microbiology and 30 immlmology", J. M. Creese and R. E. Lewis, eds., Kargen, Basel, 1989.

Bolan, G., Broome, C. V., Fracklam, R.R." Pitkaytis, B.D., Fraser, D.
W., and Schlech, W. F, III, Ann. Internal Med. 104:1, 1986.
Borgano, J. M., McLean, A. A., Vella, P. P., Canepa, I., Davidson, W.
L., and Hilleman, M. R., Proc. Soc. Exp. Biol. Med. 157:148, 1978.
Broome, C. V., Facklam, R. R., and Fraser, D. W., N. Engl. J. Med.
303:549, 1980.
Bruyn, G.A.W., and van Furth, R., Eur. J. Clin. Microbiol. Infect. Dis., 10:897, 1991.
Chudwin, D. S., Artrip, S. C., Korenbilt, A., Scl irrll,~,~, G., and Rao, 1 0 S., Infect. Tmmlln. 50:213, 1985.
Connelly, K. K., and Starke, J. R., Sem. Resp. Inf. 6: 209, 1991.
Cryz, S. J., and Furer, E., Conjugate vaccine against infections by gram-negative bacteria, method for its preparation and use, U.S. Patent No.
4,771,127, 1988.
1 5 Eby, R., Koster, M., Hogerman, D. and Malinoski, F., Pneumococcal Conjugate Vaccines, in "Modern Approaches to New Vaccines Including Prevention of AIDS", E. Norrby, F. Brown, R. Chanock and H. Ginsberg, eds., Cold Spring Harbor Laboratory Press, New York, 1994.
Fattom, A., Lue, C., Szu, S. C., Mestecky, J., Scllirrlll~, G., Brylar, 20 D., Vann., W. F., Watson, D., Kimzey, L. M., Robbins, J.B. and Schneerson, R., Infect. and Tmmlln 58: 2309, 1990.
Fattom, A., Vann, W. F., Szu, S. C., Schneerson, R., Robbins, J. B., Chu, C., Sutton, A., Vickers, J. C., London, W. T., Curfman, B., Hardagree, M. C., and Shiloach, J. Infect. Tmmlm 56:2292, 1988.
Forester, H. L., Jahnigen, D. W., and LaForce, F. M., Am. J. Med.
83:425, 1987.
Gaur, A., Arunan, K., Singh, O. and Talwar, G. P., Int. Tmmllnol.

2:151, 1990.
Giebink, G. S., Koskela, M., Vella, P. P., Haris, M. and Chap, T. L., J.
Inf. Dis. 167: 347, 1993.

21~7~3 Goebel, W. F. and Avery, O. T., J. Exp. Med. 50: 521, 1929.
Gordon, L. K., Polysaccharide-exotoxoid conjugate vaccines, U. S.
Patent No. 4,619,828, 1986.
Gordon, L. K., Haemophilus influenzae b polysaccharide-diphtheria 5 toxoid conjugate vaccine, U.S. Patent No. 4,644,059, 1987.
Hazelwood, M., Nusrat, R., Kumararatne, D.S., Goodal, M., Rayklln~ , C., Wang, D. G., Joyce, H. J., Milford-Wards, A., Forte, M. and Pahor, A., Clin. Exp. Immunol. 93:157, 1993.
Hakamori, S. and T~nn~gi, R., Carbohydrate antigens in higher ~nim~
1 0 in "Handbook of experimental immllnnlogy - vol. 1", D. M. Weir, L. A.
Herzenberg, C. Blackwell and L. A. Herzenberg, eds., Blackwell, Oxford, 1986.
Heidelberger, M. and Avery, O. T., J. Exp. Med. 38:73,1923.
Hilleman, M. R., Carlson, A. J., Jr., McLean, A. A., Vella, P. P., 1 5 Weibel, R. E., and Woodhour, A. F., Rev. Infect. Dis. 3 (suppl):S31, 1981.Jennings, H. J., and Lugowski, C., Tmml~nogenic polysaccharide-protein conjugates, C~n~ n Patent No. 1,181,344, 1985.
Jennings, H. J., Roy, R., and G~mi~n, A. J., Modified meningococcal group b polysaccharide for conjugate vaccine, C~n~ n Patent No. 1,261,320, 1989.
Jones, J.K.N. and Perry, M.B., J. Am. Chem. Soc. 79:2787, 1957.
Kenne, L. and Lindberg, B., Bacterial polysaccharides in "The polysaccharides - Vol 2", G. O. Aspinall, Ed., ~c~lernic Press, New York, 1983.
Lee, C-J., Banks, S. D. and Li, J. P., Crit. Rev. Microbio. 18:89, 1991.
Lees, A., Finkelman, F., Inman, J.K., Witherspoon, K., Johnson, P., Kennedy, J. and Mond, J.J., Vaccine 12:1160, 1994.
Lock, R. A., T-T~n~m~n, D., and Paton, J. C., Microbial Pathogen.
12:137, 1992.

2I5373~

Madore, D. V., Jackson, C. L., Phipps, D. C., Penridge Pediatric Association, Popejoy, L.A., Eby, R., and Smith, D. H., Pediatric, 85: 331, 1990.
Malcolm, A. J., Messner, P., Sleytr, U. B., Smith, R. H., and Unger, F.
5 M., Crystalline bacterial cell surface layers (S-layers) as combined carrier/adju~ for conjugate vaccines, in "Immobilized Macromolecules:
Application Potentials," U. B. Sleytr, P. Messner, D. Pum and M. Sara, eds, Springer-Verlag, London, 1993a.
Malcolm, A. J., Best, M. W., Szarka, R. J., Mosleh, Z., Unger, F. M., 1 0 Messner, P. and Sleytr, U. B., Surface layers of Racill~/~ alvei as a carrier for a Streptococcus pneumoniae conjugate vaccine in "Adances in Bacterial Paracrystalline Surface Layers," T. J. Beveridge and S. F. Koval, eds., Plenum Press, New York, 1993b.
Mandell, G. L., "Principles and Practice of Infectious Diseases, "
1 5 Churchill Livingston, New York, 1990.
Marburg, S., Jorn, D., Tolman, R. L., Arison, B., McCauley, J., Kniskern, P. J., Hagopian, A., and Vella, P.O., J. Am. Chem. Soc. 108:5282, 1986.
Marburg, S., Tolman, R. L., and Kniskern, P. J., Covalently-modified 20 polyanionic bacterial polysaccharides, stable covalent conjugates of such polysaccharides and immllnt)genic proteins with bigeneric spacers, and methods of prel,a~ g such polysaccharides and conjugates and of co"r" ",i~-g covalency, U.S. Patent No. 4,695,624, 1987.
Marburg, S., Kniskern, P. J., and Tolman, R. L., Covalently-modified 25 bacterial polysaccharides, stable covalent conjugates of such polysaccharides and immlmogenic proteins with bigeneric spacers and methods of prepalillg such polysaccharides and conjugates and of collrll ~ g covalency, U. S. Patent No.
4,882,317, 1989.
Mufson, M. A., Hughey, D., and Lydick, E., J. Infect. Dis. 151:749, 30 1985.

Mufson, M. A., Krause, H. E., Schiffm~n, G., and Hughey, D. E., Am. J. Med. Sci. 293: 279, 1987.
Nielsen, S. V., and Henrichsen, J., Scand. J. Infect. Dis. 25:165, 1993.
Paton, J. C., Lock, R. A., Lees, C-J., Li, J. P., Berry, A. M., Mitchell, 5 T. J., Andrew, P. W., T-T~n~m~n, D., and Boulnois, G. J., Infect. Tmmlln.
59:2297, 1991.
Peeters, C.C.A.M., Tenbergen-Meekes, A-M., Poolman, J. T., Berutett, M., Zegers, B. J. M. and Rijkers, G. T., Infect. Tmmlm. 59: 3504, 1991.
Penney, C.L., Michon, F., and Jennings, H.J., Improved Vaccine 1 0 Compositions, WO 92/04951, 1992.
Pe~ lu~lel, R. M., Hansburg, D., Briles, D. E., Nicolotti, R. A., and Davie, J. M., J. Tmmllnt)l. 121:566, 1978.
Porro, M., Oligosaccharide Conjugate Vaccines, C~n~ n Patent No. 2 052 323, 1992.
1 5 Porro, M., and Costantino, P., Glycoploteilleic conjugates having trivalent immllnngenic activity, U. S. Patent No. 4,711,779, 1987.
Porro, M., Oligosaccharide conjugate vaccines, U.S. Patent Application No. 07/590,649, 1990.
Saunders, L.A.M., Rijkers, G. T., Kuis, W., Tenbergen-Meekes, A. J., 20 de Graff-Meeker, B. R., Hiemstra, I. and Zegers, B. J. M., J. Allergy Clin.
Immunol. 91: 110, 1993 Schidt, R. A., Boyd, J. F., McCracken, J. D., Scllirr",~n, G., and Giolma, J. P., Med. Pediatr. Oncol. 11:305, 1983.
Schneerson, R., Barrera, O., Sutton, A., and Robins, J. B., J. Exp. Med.
25 152:361,1980.
Schneerson, R., Robbins, J. B., Chu, C., Sutton, A., Vann, W., Vickers, J. C., London, W. T., Curfman, B., and Hardegree, M. C.,Infect.
Tmml-n. 45:582, 1984.

- 6 - 21~37~3 Schneerson, R., Robbins, J. B., Parke, J. C., Bell, C., Schlesselman, J.
J., Sutton, A., Wang, Z., Schiffm~n, G., Karpas, A., and Shiloach, J., Infect.
Tmmlln. 52:519, 1986.
Schneerson, R., Levi, L., Robbins, J. B., Bryla, D. M., Scllirr~ n, G, 5 and Lagergard, T., Infect. and Tl~llllllnily 60:3528, 1992.
Seid, R. C., Jr., Boykins, R.A., Liu, D. F., Kibrough, K. W., Hsieh, C.L., Eby, R., Glycoconj. J. 6: 489, 1989.
Sell, S. H., Wright, P. F., Vaughn, W. K., Thompson, J., and ScllirrlllAll~ G., Rev. Infect. Dis. 3 (suppl):S97, 1981.
1 0 Shapiro, E.D., and Clemens, J. D., Ann. Intern. Med. 101 :325, 1984.
Shapiro, E. D., N. Engl. J. Med. 316:1272, 1987.
Shapiro, E. D., Pneumococcal vaccine, In: "Vaccines and Tmmllnotherapy," S. J. Fryz Jr., ed., Pergamon Press, New York, 1991.
Siber, G. R., Weitzman, S. A., Aisenberg, A. C., Weinstein, H. J., and 1 5 scllirrlllAll~ G., N. Engl. J. Med. 299:442, 1978.
Simberkoff, M. S., Cross, A. P., Al-Ibrahim, M., Baltch, A. L., Geiseler, P. J., Nadler, J., Richmond, A. S., Smith, R. P., Scl-irrll~ , G., andShepard, D. S., N. Engl. J. Med. 315:1318, 1986.
Simberkoff, M. S., Pneumococcal vaccine in adults, In: "T~ lion,"
20 M. A. Sande, and R. K. Root, ed., Churchill Livingstone, New York, 1989.
Sims, R. V., Steinman, W. C., McConville, J. H., King, L. K., Zwick, W. C., and Schwartz, J. C., 1988, Ann. Intern. Med., 108:653., 1988.
Slack, J., Der-Balian, G. P., Nahm, M. and Davie, J. M., J. Exp. Med.
151:853-1980.
Sorensen, U. B.S., J. Clin. Micro. 31: 2097, 1993.
Sloyer, J. L., Jr., Ploussard, J. H., and Howie, V. M., 1981, Rev.
Infect. Dis. 3 (suppl):Sl, 1981.
Stein, K. E., J. Inf. Dis. 165: 549, 1992.
Stein, K. E., Int. J. Tech. Assess, Health Care, 10: 167, 1994.

Stein, K. E., Zopf, D. A., Johnson, B. M., Miller, C. B. and Paul, W.
E., J. Tmmllnol. 128: 1350, 1982.
Stein, K. E., Zopf, D. A. and Miller, C. B., J. Exp. Med., 157: 657, 1983.
Steinhoff, M.C., Edwards, K., Keyserling, H., Thoms, M. L., Johnson, C., Madore, D. and Hogerman, D., Pediatr. Infect. Dis. J. 13: 368, 1994.
Tsay, G. C., and Collins, M. S., Vaccines for gram-negative bacteria, U.
S. Patent No. 4,663,160, 1987.

1 0 The disclosure of the above publications, patents and patent applications are herein incorporated by l~r~lellce in their entirety to the same extent as if the language of each individual publication, patent and patent application were specifically and individually included herein.

1 5 B~l~k~E~round of thP InvPntion:

Tl..-..~..-~ Responses to Poly~ h~rides Heidelberger and Avery (1923) demonstrated that the type specific antigens of pneumococci are polysaccharides. Bacterial capsular polysaccharides 20 are cell surface antigens composed of identical repeat units which form extended saccharide chains. Polysaccharide structures are present on pathogenic bacteria and have been identified on Escherichia coli, Neisseria meningitidis, Haemophilus influenzae, Group A and Group B Streptococcus, Streptococcus pneumoniae and other species. (Kenne and Lindberg 1983).
Specific blood group d~te- lllin~ i and "tumor-associated" antigens are examples of ~ n cell surface carbohydrates. Oncogenically l~ ro~ ed cells often display surface carbohydrates distinctly dirr~lclll from those of non-transformed cells. These glycans consist of only a few monosaccharides 30 (Hakomori and K~nn~gi 1986). The glycan structures by themselves are usually `_ 2153733 not antigenic, but constitute haptens in conjunction with protein or glycoploteimatrices.

A general feature of saccharide antigens is their inability to elicit 5 significant levels of IgG antibody classes (IgG isotypes) or memory responses,they are considered thymus-independent (TI) antigens. Conjunction of polysaccharide antigens or of immlmnlogically inert carbohydrate haptens to thymus dependent (TD) antigens such as pfoleills enhAnres their immlmogenicity.
The protein stimlllAtes carrier-specific T-helper cells which play a role in the1 0 induction of anti-carbohydrate antibody synthesis (Bixler and Pillai 1989).

Much of our current knowledge of TI and TD responses comes from studies of pertinent mouse models (Stein et al., 1983; Stein, 1992; Stein, 1994).
TI antigens generally elicit low affinity antibodies of restricted class and do not 1 5 produce immlmnlogic memory. Adjuvants have little effect on response to TI
antigens. In contrast, TD antigens elicit heterogeneous and high affinity antibodies with i~ lni~;ltion and produce immunologic memory. Adjuvallls enhance response to TD antigens. Secondary responses to TD antigens shows an increase in the IgG to IgM ratio, while for TI antigens the secondary response 20 IgG to IgM ratio is one-to-one, similar to that of a plilllaly response (Stein et al., 1982; Stein, 1992 and 1994). In mice and hllmAn~, TD antigens elicit predo~ A ~Illy IgGl isotypes, with some amounts of IgG2 and IgG3 isotypes. TI
responses to polysaccharides are restricted to IgG3 of the IgG isotypes (Pelllllullel et al., 1978; Slack et al., 1980).
Current Pneumococcal Vaccine Pneumococci are ~;ull~lllly divided into 84 serotypes based on their capsular polysaccharides. Although there is some variability of commonly occurring serotypes with geographic location, generally serotypes 1, 3, 4, 7, 8 30 and 12 are more prevalent in the adult population. Serotypes 1, 3, 4, 6, 9, 14, 21S37~3 -18, 19 and 23 often cause pneumonia in children (Mandell, 1990; Connelly and Starke, 1991; Lee et al., 1991; Sorensen, 1993; Nielsen and Henricksen, 1993).

At present, the most widely used anti-pneumococcal vaccine is composed 5 of purified capsular polysaccharides from 23 strains of pneumococci (Pneumovax~23, Merck Sharp & Dohme). The pneumococcal capsular types included in Pneumovax~23 are 1, 2, 3, 4, 5, 6B, 7F, 8, 9N, 9V, lOA, 1 lA, 12F, 14, 15B, 17F, 18C, l9F, l9A, 20, 22F, 23F, 33F (Danish nomenclature).
These serotypes are said to be responsible for 90 percent of serious 1 0 pneumococcal disease in the world.

Some controversy exists in the literature over the efficacy of the Pneumovax~23 vaccine (Borgano et al.,1978; Broome et al., 1980; Sloyer et al., 1981,; Shapiro and Clemens, 1984; Bolan et al., 1986; Simberkoff et al., 1986;
1 5 Forester et al., 1987; Shapiro, 1987; Sims et al., 1988; Simberkoff, 1989;
Shapiro, 1991). The pneumococcal vaccine is effective for induction of an antibody response in healthy young adults (Hilleman et al., 1981; Mufson et al., 1985; Bruyan and van Furth, 1991). These antibodies have been shown to have in vitro opsonic activity (Chudwin et al., 1985). However, there is marked 20 variability in the hllellsily of the response and in the persistence of antibody titers to the different serotypes (Hilleman et al., 1981; Mufson et al., 1987).

Children under 2 years of age are the group at highest risk of systemic disease, otitis media and acute lower re~pilatoly infection caused by 25 pneumococci, but they do not respond to this vaccine (Sell et al., 1981;
Hazelwood et al., 1993; Saunders et al., 1993). Furthermore, elderly and immunosuppressed patients have impaired or varied responses to Pneumovax~23 (Siber et al., 1978; Schildt et al., 1983; Forester et al., 1987; Simberkoff, 1989;
Shapiro, 1991). These population groups do not respond well to the thymus 30 independent polysaccharide antigens of this vaccine. Typical of thymus _ 21S37~

independent antigens, antibody class switching from an IgM to IgG isotype is notusually observed nor is an ~n~mn~stic response to a booster i~ ion (Borgano et al., 1978).

Recent occurrences of antibiotic resistant strains of bacteria stresses the need to develop efficacious vaccines for the prevention of childhood infection.
Clearly, new vaccines against pneumococci are n~e~le~l especially for high risk groups and children.

10 Conjugate Vaccines Avery and Goebel were the first to prepare vaccines against bacterial infections (Avery and Goebel 1929; Goebel and Avery 1929). More recently, several protein carrier conjugates have been developed which elicit thymus dependent responses to a variety of bacterial polysaccharides. To date, the 1 5 development of conjugate vaccines to Hemophilus influenzae type b (Hib) has received the most attention. Schneerson et al. (1980) have covalently coupled Hib polysaccharides (polyribitol-phosphate) to diphtheria toxoid. This group hasalso developed a Hib vaccine by derivatizing the polysaccharide with an adipic acid dihydrazide spacer and coupling this material to tetanus toxoid with 20 carbodiimide (Schneerson et al., 1986). A similar procedure was used to produce conjugates cont~ining diphtheria toxoid as the carrier (Gordon, 1986 and1987). A bifunctional spacer was utilized to couple the outer membrane protein of group B Neisseria meningitidis to Hib polysaccharides (Marburg et al., 1986, 1987 and 1989). Finally, Anderson (1983 and 1987) has produced a conjugate 25 vaccine using Hib oligosaccharides coupled by reductive amination to a nontoxic, cross-reactive mutant diphtheria toxin CRM~97.

Reports in the literature differ on the efficacy of these vaccines, and many studies are still in progress. However, oligosaccharide conjugates (Anderson et 30 al., 1985a, 1985b, 1986, 1989; Seid et al., 1989; Madore et al., 1990; Eby et 21 $3 7~3 al., 1994) and polysaccharide conjugates (Barra et al., 1993) are reported to beimmlmogenic in infants and elicit a thymus dependent response. Hapten loading is a key factor for conjugate immunogenicity (Anderson et al., 1989; Eby et al.,1994).

Other conjugate vaccines have been developed by Jennings et al. (1985 and 1989), who utilized periodate activation to couple polysaccharides of Neisseria meningitidis to tetanus or diphtheria toxoid carriers. Porro (1987) defined methods to couple esterified N. meningitidis oligosaccharides to carrier1 0 proteins. Conjugate vaccines cont~ining polysaccharides of Pseudomonas aeruginosa coupled by the periodate procedure to detoxified protein from the same organism (Tsay and Collins, 1987) have been developed. Cryz and Furer (1988) used adipic acid dihydrazide as a spacer arm to produce conjugate vaccines against P. aeruginosa.
Polysaccharides of specific serotypes of S. pneumoniae have also been coupled to classical carrier proteins such as tetanus or diphtheria toxoids (Schneerson et al., 1984; Fattom et al., 1988 and 1990; Schneerson et al., 1992), to N. meningitidis membrane protein (Marburg et al., 1987; Giebink et 20 al., 1993) and to a pneumolysin mutant carrier (Paton et al., 1991; Lock et al., 1992; Lee et al., 1994). Technology for coupling S. pneumoniae oligosaccharides to CRMl97 protein has been developed (Porro, 1990). These conjugate vaccines have variable or as yet und~te~ ed immlmopotentiation properties. Reproducibility of these coupling technologies with the maintenance 25 of immllnogenic epitopes is currently the greatest problem in developing effective S. pneumoniae glyco-conjugate vaccines. The optimal immllnogenic oligosaccharide size appears to vary dependent on the serotype, in-lic~ting a conformational aspect of certain immunogenic epitopes (Eby et al., 1994;
Steinhoff et al., 1994).

Vaccines to DTP, tuberculosis, polio, measles, hepatitis, Hib and pneumonia which induce long lasting protection are n~e~e(l. In order to induce protection in infants to S. pneumoniae, a multi-hapten protein conjugate cont~ining a high level of oligosaccharides of optimal immllnogenic size for each 5 serotype is desired.

Various researchers have proposed enhancement of the immllnogenicity of conjugate vaccines by adjuvant a(lmini.~tration. Al~ lill.l.ll salt, which isapproved for human use, is an example. Carbohydrate moieties, such as beta 10 glucan particles and low molecular weight dextran, have also been reported topossess adjuvant activity. Adjuvax (Alpha-Beta Technology) is an adjuvant composition cont~ining beta glucan particles. Lees et al. (1994) have reported the use of low molecular weight dextran constructs as adju~ . Penney et al.
(1992) have reported a long chain alkyl compound with immunological activity.
Brief Description of the D~awi.~:

Figure 1 illustrates the repeat unit structures of the polysaccharides used in the Examples of the invention.
Figure 2 shows the separation profile of Streptococcus pneumoniae serotype 8 capsular polysaccharides through a BioGel P-10 column after acid hydrolysis (0.5 M trifluoroacetic acid, 100C, 20 minutes) resulting in discernible oligosaccharides of one to eight repeat units.
Figure 3 shows the relative size of the repeat units in peaks 1, 2, 3 and 4 of hydrolyzed Streptococcus pneumoniae serotype 8 capsular polysaccharides, as measured by HPLC analysis.

Figure 4 shows the HPLC retention times of the glucose, M-3 maltotriose, M-7 maltoheptose, and M-10 malto-oligosaccharide standards used to determine the relative size of various oligosaccharide repeat units.

Figure 5 is an example of the retention times of ribitol, rhamnose, galactose, fucose and mannose monosaccharide standards used to d~Le~ e carbohydrate content of the hydrolysed repeat unit.

Figure 6 shows the separation profile of S. pneumoniae serotype 6B
polysaccharide hydrofluoric acid hydrolysates passed over a P-10 BioGel column.

Figure 7 shows the separation profile of S. pneumoniae serotype 6B
polysaccharide TFA hydrolysates passed over a P-60 BioGel column.

Figure 8 shows the separation profile of S. pneumoniae serotype 14 polysaccharide TFA hydrolysates passed over a P-30 BioGel column.

Figure 9 shows a separation profile of S. pneumoniae serotype l9F
polysaccharide acetic acid hydrolysates acetic acid passed over a P-10 BioGel column.

Figure 10 shows the separation profile of S. pneumoniae serotype 23F
polysaccharide TFA hydrolysates passed over a P-10 BioGel column.
Figure 11 shows the separation profile of S. pneumoniae serotype 8 polysaccharide cleaved by cellulase passed over a P-10 Bio Gel column.

21~3733 Figure 12 shows the separation profile of pneumococcal C-substance polysaccharide hydrofluoric acid hydrolysates passed over a P-10 Bio Gel column.

Figure 13 shows the inhibition ELISA results using a mouse antiserum to Streptococcus pneumoniae serotype 8 oligosaccharide protein carrier conjugate.

Figure 14 illustrates the acidification of oligosaccharides for carbodiimide coupling.
Figure 15 shows the separation of reduced and periodate fractions of a polysaccharide (23 valent polysaccharide vaccine-Pneumovax~ 23, Merck, Sharp and Dohme).

Figure 16 demon~lldt~s separation of reduced and periodate fractions of oligosaccharides of serotype 6B of Streptococcus pneumoniae.

Figure 17 demonstrates separation of reduced and periodate fractions of oligosaccharides of serotype l9F of Streptococcus pneumoniae.
Figure 18 depicts the periodate and EDC coupling chemistry reactions.

Figure 19 shows how a mono-hapten 8-oligosaccharide tetanus toxoid conjugate inhibited anti-8 serum binding to a 8 polysaccharide coated ELISA
25 plate.

Figure 20 depicts the IgG antibody isotypes elicited by S. pneumoniae serotype 8 polysaccharide following i""~.."~ ion with an 8:14 di-hapten-oligosaccharide-TT conjugate.

- 15- 21~3:733 Figure 21 shows an increased level of IgGl antibody isotype elicited by polysaccharide following i~ ion with an 8:14 di-hapten-oligosaccharide-conjugate, typical of a TD response.

Figures 22A and 22B show IgG isotypes elicited from groups of mice il,,,,,lll~i,~d with 14-polysaccharide and oligosaccharide conjugates with and without adjuvant.

Summary of the Invention:
In one aspect, the invention provides compositions comprising: a) a size-separated carbohydrate hapten comprising at least one immlmogenic epitope; and b) a carrier, wherein said hapten is covalently coupled to said carrier and wherein said hapten-carrier conjugate is protectively immlmogenic.

In another aspect, the invention provides methods of making conjugate compositions comprising: a) cleaving a bacterial polysaccharide into oligosaccharides so as to preserve immunogenic epitopes on the res--lting oligosaccharides; b) s~alalhlg the resulting oligosaccharides based on size; c) selecting those oligosaccharides which contain immlmogenic epitopes based on inhibition ELISA; d) activating the oligosaccharides selected in step c); and e)coupling the activated oligosaccharides to a purified carrier, wheleill the res lting composition contains immunogenic epitopes and is protectively immlln~lgenic .
In a further aspect, the invention provides methods of providing protective immllni7~tion against a bacterial pathogen comprising ~llmini~tering to a ll~lllll,~l in need of such treatment an effective amount of the vaccine composition described above.

In still a further aspect, the invention provides compositions useful for stim~ ting an immlln~ response to an antigen, said immlmostimlll~tQry composition comprising an oligosaccharide of S. pneumoniae serotype 8 and a suitable ph~rm~ceutic~l carrier, wherein said oligosaccharide provides an 5 immllnostimlll~tQry effect.

In a yet further aspect, the invention provides methods of providing protective immllni7~tion against a bacterial pathogen comprising a-lmini~tering to a m~mm~l in need of such treatment an effective amount of the composition of 10 the serotype 8 composition described above.

A still further yet aspect of the invention provides methods of aug~ g an immlmQgenic response to an antigen colllplisillg atlmini~tering an oligosaccharide of S. pneumoniae serotype 8 which contains an immllnogenic 15 epitope as del~ lilled by inhibition ELISA along with said antigen.

Detailed Des~ tion of the Invention:
This invention relates to improved methods for preparing oligosaccharide-protein carrier conjugates. The conjugate product may be composed of various 20 haptens or carriers. Mono, di, and multi-hapten conjugates may be plepaled.
Methods to detellllille the presence of immllnogenic epitopes on the hapten or carrier of the resultant conjugate are described. Such conjugates have utility as vaccines, therapeutic and prophylactic agents, imml-nomodulators diagnostic agents, development and research tools.
This invention is particularly suited for developing conjugates as vaccines to such bacterial pathogens including, but not limited to Streptococcus pneumoniae, Neisseria meningitidis, Haemophilus influenzae B, Group B
Streptococcus, Group A Streptococcus, Bordetella pertussis, Escherichia coli, 30 Streptococcus mutans, Staphylococcus aureus, Salmonella typhi, Cryptococcus neoformans, Pseudomonas aeruginosa and Klebsiella pneumoniae. Conjugates of this invention convert weakly or non-immlmogenic molecules to molecules which elicit specific immnnoprotective antibody or cellular responses.

Poor immllnP responses to polysaccharide vaccines (thymus independent antigens, TI) have been observed with high risk groups, such as the elderly and children under 2 years of age. Several investigators are ~ ing to elicit thymus dependent (TD) responses to a variety of bacterial polysaccharides using protein carriers. Integrity of critical immllnogenic epitopes and inconsistency of 10 covalent linkage between the carbohydrate and protein are major limitations with these conjugate vaccines. The present invention is drawn to the discovery of coupling technology which gives good reproducibility with respect to the carbohydrate to carrier ratio of conjugates. This invention also provides methods to verify the presence of immlmogenic epitopes on and oligosaccharide 15 haptens and hapten-carrier conjugates.

Polysaccharide conjugates elicit non-boostable IgM antibody responses, typical of TI antigens. The antiserum produced in response to these polysaccharide conjugates does not have opsonic activity. In the present 20 invention, oligosaccharides prepared by cleavage of polysaccharides from various bacterial strains are size separated and used to produce mono-hapten conjugates. These conjugates elicit IgG antibody isotypes with immllnoprotective, opsonization ability. This antibody response is elicited without the use of any adjuvant. Thus, the methods of the inventions are ideally25 suited for producing immllnogenic oligosaccharide hapten-carrier conjugates which utilize weakly or non-immunogenic polysaccharides of various strains.
The presence of immllnogenic epitopes on these oligosaccharides was found to be critical for eliciting an immnnoprotective response.

. 21S3733 -The number of bacterial antigens needed to develop efficacious anti-pathogen vaccines is expanding. However, repeated ~-lmini~tration of tetanus or diphtheria toxoid (often used as carrier proteins in vaccine compositions and as a prophylactic measure following trauma) may cause a phenomenon called carrier-5 inflllce~ epitope suppression. Epitope suppression has been described in thelileldLule with synthetic peptide and saccharide-toxoid conjugates (Gaur et al., 1990; Peeters et al., 1991). Tmmlln~ responses to a hapten coupled to a carrier protein can be reduced or absent when the recipient has been previously illllllllni~ecl with the carrier.
The goal of many researchers is to develop vaccines which elicit protection to the predo~ all~ bacterial serotypes which cause acute lower respiratory infection, otitis media and bacteremia in infants, without inducing carrier suppression. The methods of the invention can be utilized to produce 15 multi-hapten conjugates with optimal immllnogenic epitopes to each bacterial serotype. These conjugates, which contain lower carrier protein amounts than traditional conjugates, reduce the occurrence of the carrier suppression phenomenon. The reduced antigen load possible using these conjugates ",i~ es the antigenic competition observed with traditional conjugates.
Previously, we reported that crystalline bacterial cell surface layers (S-layers) were useful as carriers for the development of prototype conjugate vaccines (Malcolm et al., 1993a) and as a means to avoid the carrier suppression phenomenon (Malcolm et al., 1993b). In our laboratory, we 25 identified several S-layer glycoproteins which elicit non-cross reactive antibody and cellular responses. Vaccines to a variety of diseases can be developed usingS-layers isolated from various bacterial strains, thereby avoiding carrier suppression observed with tetanus and diphtheria toxoids. However, S-layers are difficult to isolate and purify, as well as costly to produce, making them 30 impractical for wide usage as vaccine carriers. The present invention describes 21 ~3 733 methods to prepare mono, di and multi-hapten oligosaccharide conjugates which reduce the amount of carrier n~cess~ry to elicit specific responses, thereby decreasing the risk of carrier in~ ced epitope suppression, even when tetanus ordiphtheria toxoid is used as the carrier.

One specific application of the technology of the invention is for the development of effective vaccines for the prevention of pediatric pneumoniae infections. Another application of the invention is to develop vaccines for protection to strains of Group B Streptococcus, Group A Streptococcus, 10 Haemophilus influenzae B, Streptococcus pneumoniae and N. meningitidis prevalent in infant disease, in the elderly or the immllnosuppressed. Other applications include development of conjugates for eliciting protection to various bacterial or virus pathogens.

We have found that the use of conditions which cleave specific linkages (i.e., l - 4 linkages) but leave sugar monosaccharides and other immllnologically important compounds such as phosphate intact results in improved immllnogenicity of the resulting conjugates.

We have found that oligosaccharide size and collro~,llalion is important to m~ximi~e immlm~genicity of conjugate plepdlalion. Dirrelc"l oligosaccharide sizes are separated from hydrolyzed polysaccharide mixtures and isolated by sizefraction. The monosaccharide content and the relative size of separated oligosaccharides is measured by, for example, HPLC analysis. Dirrere"l size 25 repeat units are tested using inhibition ELISA. We have found that ELISA
inhibition is directly proportional to the immllnogenicity of the oligosaccharide preparation and the resultant conjugate.

In particular, oligosaccharides prepared from cleavage of polysaccharides 30 of S. pneumococcus strains 3, 6B, 8, 14, l9F and 23; pneumococcal C-21 $3733 substance; and N. meningitidis C-polysaccharide have been used in our laboratory. Preferred repeat units (R.U.) for oligosaccharides are as follows for some S. pneumococcus serotypes and pneumococcal C-substance:
Serotype 3: 4-8 R.U.
6B: 4-10 R.U.
8: 2-8 R.U.
14: 4-6 R.U.
l9F: 4-10 R.U.
C-substance: 6-10 R.U.
10 Preferred repeat units for N. meningitidis C-polysaccharide is 6-10 R.U.

Creating charged groups on saccharide haptens has been discovered to facilitate the coupling of the haptens to the carrier. Use of cation or anion exchange columns is effective in allowing coupling of oligosaccharide to carrier1 5 at a higher sugar to carrier ratio. This provides more hapten per carrier, and reduces the carrier suppression phenomenon. R~duced fractions of carbohydrate are used for coupling to carrier.

Another important aspect to produce effective conjugate vaccines is the 20 use of purified carrier. In~u~iLies found in a carrier prepalation may hlle,r~le with coupling procedures. Aggregates of carrier proteins found in a carrier p,epa,alion can affect opli~lulll hapten to carrier ratios n~cess~ry to elicit the desired response. Carriers are generally purified using size exclusion column chromatography, although any standard method which removes in~ulilies and 25 aggregate may be used.

The coupling reaction time and the amount of oligosaccharide, coupling reagent and carrier are critical for obtaining an ideal carbohydrate to carrier conjugate ratio. We have developed methods which quantify carbohydrate to 30 carrier ratios by reproducible assays. Maillle~ ce of pH and temperature 215373~

conditions determined to be optimal during the coupling reaction is also important to produce an effective conjugate. Likewise, the use of effective blocking reagents which stop the coupling reaction but do not mask the immllnogenic groups is important to create effective conjugate compositions.

Use of coupling chemistry which m~int~in~ immllnogenic epitopes on oligosaccharides/polysaccharides is essential. We have found that EDC and periodate coupling, as described below may be used for coupling oligosaccharides to carriers. In addition to direct coupling of sugar to carrier, 10 various linkers may be used to space the saccharide from the surface of the protein. Appropliate linkers may also provide charged or uncharged moieties as desired. The immllnogenicity of coupled sugar-carrier compositions is determined by inhibition ELISA.

Using the methods of the present invention, we have discovered means to produce di-hapten and multi-hapten conjugates which still m~int~in their immunogenic epitopes. Conjugates with various oligosaccharide sequences and/or sizes can be produced. Similarly, conjugates comprising oligosaccharide and polysaccharide combinations may be synthesized. Such conjugates are able 20 to reduce or elimin~te antigenic competition.

Thus, appropliate conjugate design provides the ability to reduce carrier in~ ce~1 epitope suppression. Keys in this regard are the identification and use of immllnogenic oligosaccharide epitopes and more effective coupling of sugar to 25 protein. Binding a larger number of immllnogenic epitopes per protein molecule means that less carrier is needed to provide protective i~""~ ion.

We have developed methods to quantify immllnnprotective antibody response to conjugate compositions by isotyping ELISA, bactericidal and 30 opsonization assays. This allows d~telmillation of which conjugates will elicit 21~3733 the appropliate immlmoglobulin isotype response, i.e., IgG isotypes, when used to protectively immllni7e m~mm~

D~fin;tiQn.c:
The following terms have the following m~ningc when refelellced herein:
Oligosaccharide means a carbohydrate compound made up of a small number of monosaccharide units. In particular, oligosaccharides may be formed by cleaving polysaccharides.
Polysaccharide means a carbohydrate compound cont~ining a large number of saccharide groups. Polysaccharides found on the outer surface of bacteria or viruses are particularly useful in the present invention.

Carrier means a substance which elicits a thymus dependent immlln~
response which can be coupled to a hapten or antigen to form a conjugate. In particular, various protein, glycoproteill, carbohydrate or sub-unit carriers can be used, including but not limited to, tetanus toxoid/toxin, diphtheria toxoid/toxin, bacteria outer membrane proteins, crystalline bacterial cell surface layers, serum albumin, gamma globulin or keyhole limpet hemocyanin.

Immunogenic means causing an immlln~ response. An immllnogenic epitope means that portion of a molecule which is recognized by the immlm~
system to cause an immllnogenic response.
Hapten means an antigen, including an incomplete or partial antigen which may not be capable, alone, of causing the production of antibodies. Di-and multi-hapten, for purposes of this application, refer to compositions including two (di) or more (multi) oligosaccharide haptens conjugated to carrier.

21~3733 Protectively immllnogenic or immllnoprotective means stim~ tin~ an immllnP response which prevents infection by pathogen.

Immunostimnl~tory means stimlll~ting or enhancing an immlln~ response 5 to weakly immllnogenic haptens or antigens.

Neonate means a newborn animal, including an infant.

Methodology:
Prep~ration and Separation of Cleaved Polysaccharides:
Polysaccharides, available through American Type Culture Collection, Rockville, Maryland or by isolation procedures known in the art, were cleaved into oligosaccharide units using d~ropliate concentrations of chemicals. These 15 chemicals include, but are not limited to trifluoroacetic acid, acetic acid, hydrofluoric acid, hydrochloric acid, sodium hydroxide and sodium acetate.
Dirre~ time periods and tenlpeldlules may be used depending on the particular chemistry and concentration and on the resulting oligosaccharide desired.
Commercially available enzymes (e.g., cellulase and ,~-galactosidase) or isolated 20 bacteriophage-associated endoglycans known in the art can also be used to prepare oligosaccharides from polysaccharides.

Figure 1 shows the repeat unit structures of the polysaccharides used in the Examples of the invention. Other bacterial and viral polysaccharide are 25 known to those of skill in the art, and may be used in the methods and compositions of the present invention. Various polysaccharides can be cleaved including, but not limited to, pneumococcal group antigen (C-substance) and capsular polysaccharides of serotypes of Streptococcus pneumoniae, Neisseria meningitidis, Haemophilus inJ?uenzae, Group A Streptococcus and Group B
30 Streptococcus.

21~3733 After cleavage, the resulting oligosaccharide nli~lules are separated by size using P-10 (fractionation range 1,500 - 20,000 molecular weight), P-30 (2,500 - 40,000 molecular weight) and P-60 (3,000 - 60,000 molecular weight) BioGel columns. The presence of carbohydrates in the various column fractions 5 is d~lelmilled using phenol-sulphuric or sialic acid assays and thin layer chromatography (TLC). Carbohydrate-cont~ining column fractions are then analyzed by HPLC.

The presence of immunngenic epitopes on size-separated fractions of 10 cleaved polysaccharides is d~le.lllil1ed by inhibition ELISA, as described below.
If a preparation does not result in oligosaccharide fractions which inhibit in the ELISA test, cleavage procedures may be modified by ch~nging enzymes or chemicals, molarity, reaction time or l~ el~lule in order to produce immlmogenic epitopes.

D~le,lllhlation of Tmmuno~enic Epitopes in Oligosaccharide Preparations:
The presence of immlmogenic epitopes in column fractions is confirm~cl by inhibition ELISA and phosphorous assay as set forth in the Examples section.
Oligosaccharide fractions cont~ining immllnogenic epitopes (defined as those 20 which produce at least about a 50% reduction in O.D.40s at 12.5 llg concentration) are selected for coupling to carrier.

Couplir~ to C~rrier:
The oligosaccharide or polysaccharide to be used for coupling to carrier 25 is acidified or reduced in p-epalalion for EDC or periodate oxidation coupling.
For example, the oligosaccharide plepal~lion may be reduced using a RexynTM
101 (H) organic acid cation exchange column to acidify the sugar for EDC
coupling. Similarly, sugars may be reduced using standard methods for periodate oxidation coupling. When pl~alillg di-hapten or multi-hapten 21 ~3 73~
-conjugates, each oligosaccharide is activated individually for EDC or periodate conjugation.

Preferred di-hapten oligosaccharide conjugates include: 3:8-TT, 6:8-TT, 5 6:14-TT, 8:14-TT, 8:19-TT, 8:23-TT and 14:19-TT.

Carrier:
Various protein, glycoprotein, carbohydrate or sub-unit carriers can be used, including but not limited to, tetanus toxoid/toxin, diphtheria toxoid/toxin, 10 bacteria outer membrane proteins, crystalline bacterial cell surface layers, serum albumin, gamma globulin or keyhole limpet hemocyanin. In the specific examples of this invention, tetanus toxoid was used as the carrier. Tetanus toxoid l)r~al~lions routinely contain aggregates and low molecular weight hll~ulilies. Purity of carrier is essential for obtaining consistency with coupling 15 reactions, size exclusion chromatography is used to obtain a purified carrier preparation.

Size separated, immllnogenic epitope-cont~ining oligosaccharides are coupled to purified carriers by carbodiimide (EDC) or periodate activation, using 20 the procedures described in the Examples section. Any free hapten oligosaccharides are separated from hapten-carrier conjugates by column chromatography. The carbohydrate to protein ratio of conjugates is delellllilledby phenol sulfuric or sialic acid and Lowry protein assays. Typically, conjugates prepared by EDC coupling have a carbohydrate to carrier ratio of 1:2, while 25 conjugates prepared using periodate oxidation coupling have carbohydrate to carrier ratios ranging from 1:5 to 1:10.

Delellllillalion of Tmmllno~enic Epitopes on Conju~ates:
As stated previously, integrity of critical immunogenic epitopes is a 30 problem with previously known conjugation technologies. In the present 21~373~

invention, the ELISA inhibition assay is used to determin~ the potential immunogenicity of various conjugates produced by our conjugation procedures.
We have found that conjugates which demonstrate inhibition in this assay (at least about 50% reduction in O.D.4~,5 at 6.25 ~lg concentration) using the 5 methods set forth in the Examples, provide protective immlmc)genicity when used as a vaccine in m~mm~ . Thus, this assay is used to screen for useful conjugate compositions.

Tmml1ni~tion to Flicit Tmmunoprotective Antibody Respon.~es:
Typically, mice are immuni7ed on day 0 (1-primary immuni7~tion) day 7 (2-secondary immuni7~tion) and day 28 (3-tertiary i"",~."~ ion) by subcutaneous injection (100~11 into 2 flank sites) with antigens (polysaccharide-conjugates, oligosaccharide-conjugates, uncoupled polysaccharide or oligosaccharide, or uncoupled tetanus toxoid) at doses of 0.1, 0.5, 1, 2.5 and 515 ,ug, based on carbohydrate content for EDC conjugates and protein content for periodate conjugates.

Antigens were diluted to various doses in 0.9% NaCl and mice injected with 0.9% NaCl were used as negative controls. Mice were bled 7-10 days post 20 -2 and 3 immuni7~tion to collect serum to assay immunoprotective antibody responses. A typical i~ ni~;.lion schedule is shown in Table 1 for S.
Pneumoniae serotype 3 polysaccharide and oligosaccharide-tetanus toxoid conjugates prepared using EDC coupling.

Various other i~ lni7i.tion schedules are effective, including: day 0 (1), day 14 (2) and day 44 (3); and day 0 (1), day 30 (2) and day 60 (3).

The conjugates of this invention may be used as classical vaccines, as immunogens which elicit specific antibody production or stim~ te specific cell 30 mediated immlmity responses. They may also be utilized as therapeutic modalities, for example, to stim~ te the immlln~ system to recognize tumor-associated antigens; as immlmomodulators, for example, to stimlll~te lymphokine/cytokine production by activating specific cell receptors; as prophylactic agents, for example, to block receptors on cell membrane 5 preventing cell adhesion; as diagnostic agents, for example, to identify specific cells; and as development and/or research tools, for example, to stimlll~te cells for monoclonal antibody production.

Detellllhla~ion of Response:
As previously discussed, antibody responses to TI and TD antigens differ.
In the mouse, the response to a polysaccharide (TI) antigen is usually composed of a one-to-one ratio of IgM and IgG. In general, IgG isotypes are restricted, with IgG3 being over-expressed in anti-polysaccharide serum. IgA isotypes may also be present. TI antigens elicit antibodies with low affinity and immllnologic 15 memory is not produced.

With TD antigens, increased secondary IgG antibody responses (an ~n~mn~stic response) are found, with a higher IgG to IgM ratio. Marked levels of IgA are usually not present. The TD antigen elicits a heterogeneous IgG
20 isotype response, the predominant isotype being IgGl. IgGza and 2b isotypes can be expressed, while the IgG3 isotype level is usually relatively low. TD antigens elicit immllnologic memory and antibody affinity increases with i~ i7~ions.
Thus, analysis of the immlmnglobulin isotypes produced in response to conjugate ~tlrnini~tration enables one to determine whether or not a conjugate will be 25 protectively immllnogenic.

We have found that the conjugates of the present invention induce a response typical of TD, rather than TI antigens, as measured by direct and isotyping ELISA and opsonization assay.

Conjugates prepared using our EDC coupling methods elicited better antibody responses than conjugates prepared by periodate activation. Doses of l ~lg were most immunogenic. Oligosaccharide-conjugates plcpalcd with diphtheria toxoid carriers elicited antibody responses similar to the responses 5 elicited with the oligosaccharide-tetanus toxoid conjugate.

As described previously, several investigators have ~llellll)t~d to increase immunogenicity and elicit thymus-dependent antibody protection by coupling polysaccharide material to tetanus and diphtheria toxoids. Results indicate that10 these conjugates are only slightly more imm-lnogenic than uncoupled capsular polysaccharide (CPS). One possible explanation for this may be that pertussis, diphtheria and tetanus toxoids (in allllllilllllll salt adjuvant) are often ~lmini~tered as a prophylactic four dose i~ lni~tion regime to infants. This regime may tolerize the infant, making the infant incapable of mounting a protective antibody 15 response to a hapten/antigen coupled to these toxoid carriers (carrier suppression). Another possible reason for failure to induce protection may be structural. Protein carriers elicit and augment the immlmP response to haptens, but in the case of CPS-protein conjugates, the CPS portion is a relatively largeTI antigen. The immlm~ system may not recognize the CPS-protein as a 20 conjugate, but simply as two distinct entities, resulting in a thymus-independent response to the CPS and a thymus-dependent response to the carrier.

This appears to be the case in our studies, as shown in Table 2. The immlmP. system recognizes the polysaccharide of our polysaccharide-tetanus 25 toxoid (TT) conjugate as a TI antigen. The potential TD inducing capability of the carrier with respect to antibodies to the polysaccharide is not observed. Wepostulate that the immllnngenic epitopes of the carbohydrate haptens (oligosaccharides) must be in close proximity to the TD inducing epitopes of thecarrier in order to convert a TI response to a TD response.

We have also used linker arm technology to prepare conjugates. We have used, for example, 6-amino-n-hexanoic acid as a linker. The resulting conjugates were found to be less effective in eliciting antibody responses than conjugates prepared by directly coupling EDC activated oligosaccharide haptens 5 to carriers. This finding supports our hypothesis that close hapten to carrier proximity is needed to elicit TD responses.

We have also developed methods to dele~ e the level of immllnoprotective antibody elicited by the conjugates of the present invention 10 using bactericidal or opsonization assays. These tests have shown that the conjugates of the present invention are effective in eliciting protective antibodies, as measured by these assays.

As discussed previously, the epitope-carrier suppression phenomenon has 15 been observed by other researchers and in our laboratory with the S-layer carrier studies (Malcolm et al., 1993b). Our multi-hapten conjugates will reduce or circumvent this suppression, because these conjugates will contain greater mass of immllm)genic epitope per molecule of carrier than conventional conjugate vaccines. With our conjugates, the immlmP system will not be "overchallenged"
20 by the carrier. For example, a tri-hapten conjugate prepared by methods of this invention will require only three injections to elicit specific immllnP responses to three different target pathogens. In contrast, using conventional monohapten conjugates, one would need to ~lmini.cter nine injections to elicit similar responses. This means three times the amount of protein would be required.
Further, i~ tion regimes convert an anti-polysaccharide TI
response to a TD response can be designed using the conjugates of the present invention. Economical initial exposure to polysaccharide (e.g., using Pneumovax 23) followed by a single ~(1mini~tration of a conjugate of the present30 invention would induce IgG antibody levels (an ~n~mn~stic response). Such an i""",."i~lion regime would not induce carrier suppression. In such cases, the imml-n-~ system initially e~luc~ted to various carbohydrate epitopes and antigens (a TI response) would be in(l~1ce~1 by multi-hapten conjugates to elicit stronger immllnngenic responses to pathogens frequently causing disease in specific 5 population groups (e.g., serotypes 1, 3, 4, 6, 9, 14, 18, 19 and 23 in infants).

Pharm~^e~lt~ C~ o~ ns:
To elicit antibodies to specific pathogens and/or various carbohydrate moieties the conjugates of the invention may be ~flmini~tered by various delivery 10 methods including intraperitoneally, in~ldu"lsc~ rly, intradermally, subcutaneously, orally or nasally.

The formulation of the compositions of the present invention may include suitable ph~rm~ceutical carriers. The conjugates of the invention are 15 immllnngenic without adjuvant, however adjuvants may increase immunnprotective antibody titers or cell mtq/li~tecl i~l~u~u~iLy response. Such adjuvants could include, but are not limited to, Freunds complete adjuvant, Freunds incomplete adjuvant, alumilliulll hydroxide, dimethyldioctadecyl-ammonium bromide, Adjuvax (Alpha-Beta Technology), Inject Alum (Pierce), 20 Monophosphoryl Lipid A (Ribi Tmmnnnchem Research), MPL+TDM (Ribi Immunochem Research), Titermax (CytRx), toxins, toxoids, glycoproteins, lipids, glycolipids, bacterial cell walls, subunits (bacterial or viral), carbohydrate moieties (mono-, di-, tri- tetra-, oligo- and polysaccharide), various liposome formulations or saponins. Combinations of various adjuvants may be used with 25 the conjugate to prepare the immnnogen formulation.

Exact formulation of the compositions will depend on the particular conjugate, the species to be i~ e~l and the route of ~lmini~tration.

21~3733 Such compositions are useful for i~ any animal susceptible to bacterial or viral infection, such as bovine, ovine, caprine, equine, leporine, porcine, canine, feline and avian species. Both domestic and wild anim~lc may be immnni7~1. Humans may also be il~ kd with these conjugate 5 compositions.

The route of a(lmini.ctration may be any convenient route, and may vary depending on the bacteria or virus, the animal to be i~ e~l and other factors. P~ellLelal ~flminictration, such as subcutaneous, h~ .lcc~ r, or 10 intravenous a(lminictration, is prefelled. Subcutaneous a~lminictration is most pler.,ll~d. Oral atlminictration may also be used, including oral dosage forms which are enteric coated.

The schedule of a~lminictration may vary depending on the bacteria or 15 virus pathogen and the animal to be i~"",l~i7e~1. Animals may receive a single dose, or may receive a booster dose or doses. Annual boosters may be used for continued protection. In particular, three doses at days 0, 7 and 28 are pleÇ~.red to initially elicit antibody response.

The following examples are not intended to limit the scope of the invention in any manner.

Examples F7~ml)le l:
Pl~paldLion and Separation of Polysaccharide Hydrolysates Figure 2 shows the separation profile of Streptococcus pneumoniae serotype 8 capsular polysaccharides through a BioGel P-l0 column after acid hydrolysis (0.5 M trifluoroacetic acid, 100C, 20 Illill~ s) resulting in discernible oligosaccharides of one to eight repeat units. Numbers one to eight - 32 - 21 ~3 733 correspond to the number of repeat units found in each peak, peak nine contains oligosaccharides of greater than eight repeat units. Oligosaccharides derived from hyaluronic acid were used to standardize the chromatographic system.

The relative size of the repeat units in peaks 1, 2, 3 and 4 were measured by HPLC analysis (Figure 3). The HPLC retention times of glucose, M-3 maltotriose, M-7 maltoheptose, and M-10 malto-oligosaccharide (Sigma Chemical Co.) used as standards to delelnlille relative size of various oligosaccharide repeat units is shown in Figure 4. Monosaccharide content of 10 the repeat structure was established by further hydrolysis of the oligosaccharide repeats with 2.0 M trifluoroacetic acid (TFA) at 100C for 2 hours. An example of the retention times of ribitol, rhamnose, galactose, fucose and mannose monosaccharide standards used to determine carbohydrate content of the hydrolysed repeat unit is shown in Figure 5. The chemical structure of one 15 serotype 8 repeat unit was determine to be ,B-glucose (1 ~ 4) ,~-glucose (1 ~ 4) a-galactose (1 ~ 4) aglucuronic acid (1 ~ 4) by GC-MS and NMR analysis.
This corresponds to the repeating unit structure cited in the lileralule (Jones and Perry 1957).

Figures 6 - 10 are examples of separation profiles of S. pneumoniae serotypes 6B, 14, l9F and 23F polysaccharide hydrolysates (TFA, acetic acid or hydrofluoric acid) passed over P-10, P-30 or P-60 BioGel columns.

Figure 11 shows the separation of an enzyme cleaved polysaccharide (serotype 8 cleaved by cellulase). The separation of C-substance oligosaccharides is shown in Figure 12.

Example 2:
Inhibition FT T~SA to Determine Immunogenic F,r~itopes of Oli~osacch~ride Preparations The basic procedure utilized for inhibition ELISA to test for the presence 5 of immllnogenic epitopes on oligosaccharide ple~al~lions and oligosaccharide or polysaccharide-conjugates was as follows:
1. Coat 96 well EIA plates (NUNC) with 1 ~lg well of the antigen (Ag) using 0.05 M NaCO3 coating butter (100 ~ll/well), incubate at 4C overnight.
2. On the same day, prepare inhibiting Ag tubes (e.g., various 1 0 oligosaccharide hydrolysates) using 1 x PBS - 0.01 % Tween 20 as diluent.
- Make a 7 fold serial dilution in the tubes (starting from 25 ~lg/well to 0.391 ,ug/well in triplicate), the total volume in each tube should be 175 ~11 after serial dilution.
- Prepare 1: 1000 dilution of anti-serum of a specific type (e.g., Diagnostic 1 5 anti-serum 14 that has been raised in rabbits, Statum Seruminstitut), in 1 x PBS
+ Tween.
- Add 175 ~11 of this solution to each tube. Total volume in each tube is now 350 ~1. Incubate the tubes at 4C overnight.

3. Next day, block the EIA plates with 100 ~ll/well of blocking buffer (1 x 20 PBS + 1 % BSA), incubate at room temperature for 1 hour.

4. Flick off the plates and Llall~re~ content of each tube to the wells (100 ~ll/well, incubate at room temperature for 2 hours.

5. Wash the plates 3 times with wash solution (0.01 % Tween+ 1 x PBS).

6. Prepare 1: 1500 dilution of Goat-anti-rabbit (or anti-species of serotype 25 specific serum used in Step 2) IgG Alkaline Phosphatase conjugate (TAGO) in 1x PBS + 1 % Tween buffer (100 ~ll/well). Incubate at room temperature for 2 hours.

7. Wash the plates 4 times with wash solution, flick off excess liquid.

8. Dissolve Alkaline Phosphatase substrate tablets (# 104 - Sigma) in the 30 DEA (diethylen~min~-) buffer pH=9.98, 5 ml/tablet, 100 ~l/well.

21~3733

9. Incubate the plates in the dark and read the Absorbance at 405 nm wavelength every 15 minutes.

Various commercial and laboratory prepared antiserum can be used in this 5 assay, including, but not limited to, serum produced in mice, rat, rabbit, goat, pig, monkey, baboon and human.

Figure 13 shows the inhibition ELISA results using a mouse antiserum to Streptococcus pneumoniae serotype 8 oligosaccharide protein carrier conjugate

10 (2-4 repeat units coupled using EDC to TT). Inhibition was tested with type 8oligosaccharides (0.5 M TFA, 100C, 20 minute preparation) of 1, 2, 3, 4, 6, &
8 + repeat units, and with type 8 polysaccharides. From these results, it can beseen that the 1 repeat unit (a 4 monosaccharide chain) does not contain an immlmogenic epitope. The 2 repeat unit (8 monosaccharide chain) was capable 15 of inhibiting antibody binding to the ELISA plate, indicating that it contains an immllnogenic epitope. The molecular weight of repeat unit 2 was d~Le~ ed to be 1365 by FAB-MS analysis. This correlates well with the theoretical molecular weight of 8 saccharides. Repeat units of 3, 4, 6, 8 + and the whole polysaccharide also inhibited antibody binding to the ELISA plate, again 20 indicating that immllnogenic epitopes were present in these oligo/polysaccharides .

Table 3 demonstrates similar results found using a rabbit anti-S.
pneumoniae serotype 8 specific serum (Statems S~lulllin~LiLuL). Repeat unit 1 did 25 not markedly inhibit binding; repeat units 2, 3, 4, 5, 6, 7, 8+ and whole polysaccharide inhibited binding.

Inhibition ELISA was also used to de~ellllille the presence of immunogenic epitopes on oligosaccharides prepared using dirrelellL hydrolysis 30 procedures on various polysaccharides. Table 4 shows results with methods used `_ 21 53733 by the prior art, for example, Porro C~n~ n Patent No. 2 052 323 to hydrolyse S. pneumoniae serotype 6 polysaccharide (0.01 M acetic acid, 100C, 30 hours).
Whole polysaccharide blocked binding at low antigen concentration (effective at 0.39 ~g concentration) while the acetic acid hydrolysate did not. Note that we 5 could not size separate the hydrolyzed preparation because it was "caramelized."

We discovered that dirrelell~ hydrolysing agents (e.g., TFA) and reduced time and tempeMture produced oligosaccharides with more immllnogenic epitopes, as shown in Table 5. A 0.5 M TFA, 70 C 1 or 2 hour hydrolysate 1 0 effectively inhibited antibody binding at a 3.13 ~lg concentration, a 4 hourpreparation did not. Tables 6 and 7 also illustrate the effect of time for plepalillg 6B oligosaccharides with or without immllnogenic epitopes. A 2 hour acetic acid preparation blocked antibody binding (at 3.13 llg concentration), the 24 and 48 hour prepalations did not. Similarly, a 1.5 hour TFA preparation 1 5 more effectively blocked binding than a 3 hour preparation.

As shown in Table 8, 0.5 M TFA hydrolysis of S. pneumoniae serotype 14 at 70C for 7 hours, as disclosed in the prior art (Porro, C~n~ n Patent 2 052 323), is not ~refelled. l~ed~1cecl molar concentrations of TFA (e.g., 0.1 M)20 is better for prepalillg immunogenic 14 oligosaccharides.

Table 9 illustrates the importance of selecting oligosaccharides which contain immllnogenic epitopes for coupling to carrier. The 3 repeat unit structure of serotype 14 oligosaccharide could not inhibit antibody binding, the 4 25 and 8 repeats, however, contain the immlln()genic epitopes and effectively blocked antibody binding.

Table 10 shows the effect of hydrolysate concentration and reaction time for pl~alillg 14 oligosaccharides cont~ining immllnogenic epitopes.

21~3733 Tmmllnogenic epitopes were conserved by a TFA 7 hour hydrolysis, but destroyed when hydrolysed for 24 hours.

Table 11 illustrates the importance of using optimal heat conditions for 5 producing 19F oligosaccharides cont~ining immlmogenic epitopes. Tmmlmogenic epitopes were destroyed by HCl hydrolysis at room temperature, but m~int~in~d when hydrolysis was performed at 70C.

As shown in Table 12, poor inhibition of antibody binding was observed 1 0 with 0.25 M TFA, 70C, 3 hr hydrolysates of 23F polysaccharides, (Porro, C~n~ n Patent 2 052 323). Table 13 demonstrates the effect of time on the generation of immllnogenic 23- oligosaccharides. Oligosaccharides produced by 0.1 M TFA hydrolysis, 70C for 3 hours inhibited antibody binding, oligosaccharides prepared by hydrolysis for 5 hours did not inhibit. Table 14 1 5 demonstrates the presence of immlmogenic oligosaccharides after 0.5 M TFA
hydrolysis at 70C for 15 minutes or with 5 M acetic acid at 70C for 5 hours.
These hydrolysates effectively inhibited to 0.78 ~g concentration.

Table 15 demonstrates the utility of the inhibition ELISA to recognize 20 immunogenic oligosaccharides of Neisseria meningitidis serotype C.
Hydrolysates prepared with NaOAc, blocked antibody binding as effectively as the whole polysaccharide.

Example 3:
25 Acidification of Carbohydrate Moieties for Carbodiimide Couplin~

A Rexyn 101 (H) organic acid cation exchange column (Fisher Scientific) was prepared and washed with dH20. Polysaccharide or oligosaccharide samples dissolved in dH20 (pH neutral) were run over this column and collected at a rate30 of one drop per six seconds. Acidification was confirm~d by pH colour-fixed intli~ator sticks. Excess dH20 was used to wash the column. Acidified fractions were pooled and lyophilized for use in coupling reactions.

Figure 14 depicts a TFA cleavage between ,~-D-Glcp (1 ~ 4),B-D-Gal of 5 an oligosaccharide structure resulting in the formation of an aldehyde and hydroxyl group. Further oxidation of the aldehyde results in a carboxyl group.
When this material is passed through a cation exchange column, a COO~ group results.

10 Fx~mple 4:
Couplin~g Procedures and Quantification Assays Carbodiimide (EDC) Couplin~ Procedure A 1:1 weight ratio of ion charged carbohydrate (polysaccharide or 15 oligosaccharide) sample (e.g., 3 mg) and EDC (3 mg) was dissolved in 2 mls of 0.1 M KH2PO4, a pH of 4.5 was m~int~in~d with lN NaOH or HCl. This mixture was stirred for 1 hour at room temperature. Carrier (3 mg) was added to the EDC activated carbohydrates and then stirred for 4 hours at room temperature. This reaction was stopped by the addition of 200 ~11 of 10%
20 ammonium bicarbonate, the mixture was then further stirred for 1 hour at room temperature.

The resultant conjugate was dialysed against dH20 overnight using 50,000 molecular weight cut off (MWCO) dialysis tubing.
Conjugates were lyophilized and then assayed by Lowry protein, phenol-sulfuric acid, sialic acid and phosphorous assays for composition (methods described below). Typically, conjugates prepared with this coupling methods have a carbohydrate to carrier ratio of 1:2.

Phenol-Sulfuric Acid Assay for Quantification of Carbohydrates Reagent: 5% phenol solution (5.5 mL liquid phenol (90%) added to 94.5 mL
distilled water).

5 Standard: Glucose 1 mg/ml stock solution. Prepare 2 to 60 ,ug/200 ,ul sample buffer for standard curve.
Procedure: (Adapted from: Handbook of Micromethods for the Biological Sciences. Keleti, G. & W.H. Lederer (eds). 1974. Van Nostrand Reinhold Co., New York.) 1 0 1. Place 200 ~11 samples into very clean tubes.
2. Add 200 111 phenol reagent.
3. Rapidly add 1 mL concentrated sulfuric acid.
4. Vortex well.
5. Let stand at room temperature for 30 minutes.
1 5 6. Color is stable at room temperature for 2 to 30 hours.
7. Read absorbance at 490 nm: blank with tube cont~ining H20 only as sample in # 1.

Q~1~ntit~tive Estim~tion of Si~lic Acid 20 Reagents:
a. 6 gram of A12(SO4)3 . 18 H20 dissolved up to 20 ml with distilled water.
b. 1 gram para-dimethylaminobenzaldehyde dissolved up to 20 ml with 6 N
HCl. (Store in a dark bottle in the refrigerator).
Standard: N-acetyln~ul~llh~ic acid at 0, 2.5, 5, 10, 15, 20, 25, 30, 35, 40, 45 25 and 50 ~lg/~ll total volume is 350 ~11. Use distilled water to make up to 350 Method:
1. 200 ~11 of sample in duplicates. Make up to 350 ~1 with distilled water.
2. Add 700 ,ul of reagent A to each tube. Shake.
3. Add 350 ~11 of Ehrlich reagent B.
30 4. Cover all tubes with marbles.

. ~ 2153733 5. Heat the tubes at 100 C for 30 minutes using Pierce Heating modules.
6. Cool the tubes rapidly to room temperature in an ice bath.
7. Read optical density at 530 nm wavelength.

5 Phosphorous Assay Reagents: a. 2.5 % ammonium molybdate; b. 10% ascorbic acid; c. 70%
perchloric acid; and d. 1 mM sodium phosphate standard.
Procedure: (Adapted from: Rouser, G., Siakotos, A. N. and Fleischer, S. 1966.
Lipids 1:85-86) 1 0 1. Place samples and standards (0, 25, 50, 100 and 200 ml; 25 to 200 nmoles) into clean tubes.
2. Dry samples in a heater block at 180C for 5 minutes in the fume hood.
3. Add 450 ml perchloric acid to each tube, cover each tube with a marble and heat at 180c for 30 - 60 minutes.
1 5 4. Add 2.5 mL d.H20. after tubes have cooled.
5. Add 0.5 ml ammonium molybdate and vortex imm~ tely.
6. Add 0.5 ml ascorbic acid and vortex immediately.
7. Place tubes in 95c water for 15 minutes.
8. Read absorbance at 820 nm after tubes have cooled.
20 9. Samples can be left for several hours before being read.

T owry Protein Assay Reagents:
a. 2% (w/v) Na2CO3 in 0.1 M NaOH (1 L) 25 b. 0.5% CuS04 in 1% sodium citrate (100 mL) c. Folin-Ciocalteu phenol reagent (2X) d. Bovine serum albumin (1 mg/mL) Procedure:
1. Prepare standard curve which consists of: 0, 12.5, 25, 50, 100 and 200 30 mg of BSA in a final volume of 200 ~lL.

_ 21~373~

2. Bring unknown protein samples to 200 mL with d.H20.
3. Mix reagents A and B 50: 1 (v/v) and add 2 mL to each sample.
4. Vortex and let stand at room t~lllpel~Lule for 10 minutes.
5. Dilute Folin-Ciocalteu phenol reagent 1: 1 with d.H20 and add 200 mL to5 each sample.
6. Vortex and let stand at room temperature for 30 minutes.
7. Read absorbance at 660 nm.

Periodate Oxidation Couplin~ Procedure 1 0 Samples of polysaccharide or oligosaccharide (e.g., 3 mg) were dissolved in 3 ml of freshly prepared 60 mM sodium meta-periodate in 50 mM sodium acetate. This preparation was then stirred overnight at 4C. Ethylene glycol (300 ~l) was then added to stop the reaction, this mixture was subsequently stirred at room temperature for 1 hour and then lyophilized. Samples dissolved in 1.5 ml 1 5 of 0.03 M ammonium bicarbonate (pH = 8.0) were run over a P-2 Bio-Gel column. The phenol-sulfuric acid or sialic acid assays were used to del~ line fractions cont~ining the periodate reduced form of the samples, which were subsequently lyophilized.

Figure 15 shows the separation of a reduced polysaccharide (23 valent polysaccharide vaccine-Pneumovax~ 23, Merck, Sharp and Dohme) fraction.
Figures 16 and 17 demonstrate separation of reduced oligosaccharides of serotypes 6B and l9F of Streptococcus pneumoniae, respectively.

Three mg of reduced saccharide and 3 mg of carrier were dissolved in 3 mls of 0.1 M sodium tetraborate decahydrate, pH 8.9. Sodium cyanoborohydride (H+ source) was then added to this mixture and stirred for 48 hours at 50C. This reaction was stopped by adjusting the pH to 3 - 4 with 80%
acetic acid. This conjugate was then dialysed for 48 hours against dH20 (2 - 3 30 dH20 changes) using 50,000 MWCO dialysis tubing.

The conjugate was lyophilized, and the composition of the conjugate determined by Lowry protein assay, phenol-sulfuric, sialic acid and phosphorous assays. Typically, conjugate prepared using this coupling method have carbohydrate to carrier ratios of 1:5 to 1:10.

Figure 18 depicts the periodate and EDC coupling chemistry reactions.

Example 5:
Conjugate Carriers Example 4 describes methods used to produce immlln~genic oligosaccharide/polysaccharide conjugates from weakly or non-immunogenic polysaccharides .

Tetanus toxoid was purified for use as a carrier by column chromatography. This purified toxoid elicited high levels of IgM (e.g., SO~lg/mlmouse serum) and IgG isotypes (e.g., IgG, 100 llg/ml of serum; IgG2a, 38 ~g/ml of serum; IgG2b, 68 llg/ml of serum; and IgG3, 105 ~g/ml of serum).

20 Example 6:
Delel",i~-~tion of TmmllnQgenic F.pitopes on Oligosaccharide/Polysacch~ride Conju~ates:
The inhibition ELISA as described in Example 2 was used. The presence of immllnogenic epitopes on a mono-hapten 8-oligosaccharide tetanus 25 toxoid conjugate was confirmed by inhibition ELISA. This conjugate inhibited anti-8 serum binding to a 8 polysaccharide coated ELISA plate (Figure 19). Free tetanus toxoid did not inhibit binding. The presence of immllnogenic 8 oligosaccharide on di-hapten 6:8; 14:8 and 19:8 conjugates was also shown.
This figure illustrates the reproducibility of our coupling procedures, as the 8-21~3733 mono-hapten and di-hapten conjugates equally blocked antibody binding, indicating that each conjugate contained equivalent amounts of 8 oligosaccharide.

Table 16 shows results of inhibition ELISA when 6B polysaccharide, 6B
5 oligosaccharides, a 6B:8 di-hapten-oligosaccharide tetanus toxoid conjugate ortetanus toxoid alone was used as inhibiting antigens. Tetanus toxoid did not inhibit binding of anti-6B serum to a 6B-polysaccharide coated ELISA plate.
Free 6B-oligosaccharide or polysaccharide did inhibit binding. The 6B:8 di-hapten-oligosaccharide-TT conjugate also inhibited binding. This confirms the 10 presence of immllnogenic 6B epitopes on the 6B:8 di-hapten-TT conjugate.

Similarly, a 14:8-di-hapten-TT conjugate inhibited anti-14 serum binding, demonstrating the presence of serotype 14 immllnogenic epitopes (Table 17).
Note that at high concentrations, there was non-specific inhibition by TT alone.15 We have found that this is an artifact of anti-14 in this assay.

Various oligosaccharide fractions of a 23F hydrolysate were coupled to TT. All contained immnnngenic epitopes of the 23F serotype as shown in Table 18.
The immnnogenic epitopes of N. meningititli.~ oligosaccharides (NaOAc preparation) were similarly m~int~in~d when coupled to tetanus toxoid (see Table19).

25 Example 7:
Detellnillillg Antibody Isotype Levels Elicited by Thymus Independent (TV and Thymus Dependent (TD) Antigens The basic procedure to measure antibody isotype levels is as follows to quantify IgM, IgG and IgA isotypes elicited by various conjugates:

1. Coat EIA plates (NUNC, IMMUNOSORB) with 1 mg/well of Ag in 0.05 M sodium carbonate/sodium bicarbonate buffer pH-9.5, 100 ,ul/well.
2. Incubate at 4C overnight.
3. Next day, block plates with 100 ml well of blocking buffer (1 x PBS +
5 1% BSA). Tnrl1bate at room temperature for approximately 1 hour.
4. Prepare 1:25 dilution mouse serum in working-buffer (1 x PBS + 0.1%
Tween). Add 100 Ill/well into the approl)liate well, incubate at room temperature for 2 hours.
5 . Wash plates 3 x with washing buffer ( 1 x PBS + 0.05 % Tween). Flick 10 off excess liquid by tapping the plates on the bench top.
6. Prepare 1:2 dilution of EIA Grade Mouse Type (Rabbit Anti-Mouse, IgM, IgG~, IgG2a, IgG2b, IgG3 and IgA, Bio-Rad) in working buffer at 100 ~l/well. Tnrllbate at room temperature for 2 hours.
7. Wash plates 3 x with washing buffer.
1 5 8. Prepare 1:1500 dilution of Goat-anti-Rabbit IgA Alkaline Phosphatase conjugate (TAGO) in working buffer at 100 ml/well. Incubate at room l~lll~elaLu~ for 2 hours.
9. Wash plates 4 x with washing buffer.
10. Prepare enzyme substrate using Sigma # 104 ALkaline Phosphatase 20 Substrate tablets (one tablet/5 mls of 10% diethanolamine substrate buffer), 100 Ill/well. Tn~llbate at room temperature in the dark and read every 30 minl-tes at 405 nm wavelength.

11. Convert absorbance readings to mg antibody/ml serum using dose-response curves geneldL~d from ELISA responses, of the rabbit anti-mouse 25 isotype antibodies to various concentrations of mouse class and subclass specific immunoglobulin (Zymed Labs. Inc.).

Table 2 shows the antibody elicited in mice when immllni7Pcl with S.
Pneumoniae serotype 8 oligosaccharide and polysaccharide conjugates. Only the 30 8 oligosaccharide-conjugate elicited IgG antibodies of all isotypes, the ~1~373~

unconjugated oligosaccharide was not immlln~genic, the polysaccharide and the polysaccharide-conjugate elicited antibody isotypes typical of TI responses (mainly IgM, IgA, and IgG3 isotypes). Adjuvant was not n~cess~,ry to elicit the IgG isotypes with our oligosaccharide-tetanus toxoid conjugate. Conjugates 5 comprising relatively small oligosaccharides, haptens of 2 - 4 repeat units (8 - 16 saccharides), elicited the best antibody responses as measured by direct ELISA.

Direct FT T~SA Protocol 1. Use NUNC Maxisorp Immunoplate.
10 2. Dilute coating antigen to 1.0 mg/100 ml in carbonate-bicarbonate buffer.
Use glass tubes as antigen will stick to plastic.
3. Add 100 ml to each well of plate. Store overnight at 4C.
4. Wash 3 x in PBS-.05 % Tween. Shake out excess PBS by tapping on Kimwipes/paper towels.
15 5. Add 100 ml/well of blocking agent (1 x PBS - 1% BSA). Incubate for 60 minutes at room temperature.
6. Wash 3 x as in Step 4.
7. Add 100 ml/well of test antibody appropliately diluted in PBS - .01%
Tween. Incubate for 90 minlltes at room temperature.
20 8. Wash 3 x as in Step 4.
9. Dilute ~lk~lin~ phosphatase conjugated anti-mouse Ig (TAG0 Cat # 4653) in PBS-Tween 1/1500. Add 100 Ill/well and incubate for 90 minutes in the dark.
10. Wash 3 x as in Step 4.
11. Add 100 ml/well Sigma 104 Phosphatase Substrate (disodium-p-25 nitrophenyl phosphate tables). Add 2 tablets (5 mg/tablet) of substrate to 10 mLdiethanolamine buffer. Keep in dark as substrate is inactivated by light.

12. Incubate in dark at room temperature. The development of the reaction varies depending on the antibody. Absorbance can be read on the Microelisa Auto Reader (405 nm) at approximate 30 minlltes intervals.

Results in Table 20 show a comparison of IgG1 and IgG3 levels in mice immuni7ed with 8-conjugate at 3 weeks of age or at 8 weeks of age. Significant IgGI levels were elicited by the 8-oligosaccharide-TT-conjugate in mice ni~ed at 3 weeks old (0.273 ~lg/ml) and at 8 weeks old (0.700 ~g/ml).
5 Tnrlic~tive of a TD response, an adjuvant (e.g., FCA) increased specific IgGI
(1.22 llg/ml). The 8-polysaccharide imillced over-expression of IgG3 and low IgGl, typical of a polysaccharide TI response. The 8-polysaccharide-TT
conjugate, considered a "TD antigen", in~ ced only low levels of IgGI, with overexpression of IgG3, characteristic of TI polysaccharide antigens. Also, 1 0 adjuvant in combination with the 8-polysaccharide-TT conjugate did not enhance IgGI levels, but did increase IgG3 antibody (TI-like response). Some polysaccharide-conjugates are known to elicit combinations of TI and TD
antibody response profiles (Stein, 1992; Stein, 1994).

1 5 Figure 20 depicts the IgG antibody isotypes elicited by a 8:14 di-hapten-oligosaccharide-TT conjugate to 8 polysaccharide. Like the 8-mono-hapten conjugate, this di-hapten conjugate could induce much higher levels of specific IgGI antibody (a TD response) than a 8-polysaccharide-conjugate or 8-polysaccharide alone. Overexpression of the IgG3 isotype to polysaccharide 20 immllnngen is shown. Control mice were injected with tetanus toxoid alone.

Results obtained with serotype 14-oligosaccharide conjugates are shown in Table 21. A 14-oligosaccharide-TT-conjugate prepared by 0.1 M TFA
hydrolysis elicited IgGI, G2a, G2b, and G3 isotypes, the 1 ~g dose was the most 25 immunogenic. Oligosaccharide-TT conjugates prepared using carbohydrate fractions of separation peaks 7 and 8 of a 0.5 M TFA hydrolysate elicited lower levels of IgG isotypes. Smaller oligosaccharides (peaks 4 and 5 of the 0.5 M
TFA plepdldlion) in conjugate form elicited low levels of IgG isotypes. The 14-polysaccharide-TT conjugate elicited relatively high levels of IgGI isotypes.
30 However, serum from mice injected with this polysaccharide conjugate was not - 21 ~3 733 immllnt)protective (as will be shown in Example 8, Table 24). There appears to be a required threshold level of IgG antibody isotypes to provide immllnoprotection to the serotype 14 pathogen. The uncoupled 14 polysaccharide, tetanus toxoid alone, or 0.9 % NaCl negative control serum all 5 displayed low levels of all isotypes, equivalent to normal mouse serum (NMS) levels.

Figure 21 shows an increased level of IgGl antibody isotype to 14-polysaccharide elicited by a 8:14 di-hapten-oligosaccharide-conjugate, typical of 1 0 a TD response. The 14-polysaccharide in-luce~l overexpression of IgG3 (TI
response), the 14 oligosaccharide alone was not immllnogenic. Uncoupled tetanus toxoid was a negative control.

As with individual hllm~n.~, dirrelcllL groups of mice displayed variable 1 5 responsiveness to oligosaccharide- and polysaccharide-conjugates. In certaingroups of mice, variations in the dirrelellL IgG antibody isotype levels were observed. Figure 22A shows results from a group of "good responser" mice which produced IgGl to a 14-polysaccharide conjugate (a TD-like response).
Nevertheless, a 14-oligosaccharide-conjugate elicited higher IgGl levels. This 20 conjugate also elicited substantial levels of IgG2b (0.955 ~g/ml =
oligosaccharide-conjugate; 0.139 ~lg/ml = polysaccharide-conjugate). This response was TD driven as FCA enhanced these IgG2b antibodies, Figure 22B.
(1.509 ~g/ml = oligosaccharide-conjugate; 0.474 ~lg/ml = polysaccharide-conjugate).
The ability of oligosaccharide-conjugates of the invention, to elicit greater TD antibody responses than polysaccharide-conjugates was not limited to S.
pneumoniae immllnogens. Oligosaccharide-conjugates of Neisseria meningitidis Group C elicited greater levels of IgGl isotype antibody (7.01 ~g/ml) than the 30 polysaccharide-conjugate (3.60 llg/ml) or polysaccharide alone (0.162 ,ug/ml).

_ 21 ~3 733 Interestingly, the IgG3 isotype amounts in~ ce~l by the oligosaccharide conjugates was also more (13.11 llg/ml = oligosaccharide-conjugate; 9.84 ~lg/ml = polysaccharide-conjugate; 3.81 ~g/ml = polysaccharide alone).

Fx~ml?le 8:
Bactericidal and Opsonization Assays to Measure Tmmllnoprot~tive Antibodies Elicited by Conjugates The basic bactericidal and opsonization assays used are as follows:
Bactericidal Assay 1 0 1. Streak a blood agar plate with desired gram negative bacteria procured from the American Type Culture Collection. Incubate at 37C, overnight.
2. Next day, pick an isolated colony and inoculate it in 1.0 ml of Todd-Hewitt Broth (THB) + Yeast Extraction (YE) media in a sterile test tube.
Tn~ bate at 37C overnight.
1 5 3. On the following day, measure O.D. of inoclll~te~l bacteria at 420 nm wavelength. Use THB +YE media as blank.
4. To a sterile flat bottom 96-well plate, add a sterile 2.5 mm glass bead in each well.
5. To each well, add:
a. 5 ml of bacteria.
b. 10 ml of mouse serum to be tested.
incubate at 37C for 1 hour.
Note: Step # 5 and # 6 are done in triplicate 6. After 1 hour incubation, prepare 1 :20 dilution of exogenous complement e.g. (Low Tox Rabbit Complement, Cedarlane) sterilely in THB+YE. Add 50 ~ll/well. Incubate at 37C for 1 hour.
7. After complement incubation, 50 ml aliquot is plated out on blood agar plates using a glass spreader.
8. Wrap all agar plates in plastic bags and incubate at 37C for 12 hours.
30 9. Next day, count plaque forming colonies.

Opsonization Assay 1. Streak a blood agar plate with desired gram negative or positive bacteria (procured from the American Type Culture Collection). Incubate at 37 5 overnight.
2. Next day, pick an isolated colony and mix it with 1.0 ml of THB+YE
media in sterile test tube. Incubate at 37C overnight.
3. The next day, prepare 100 U/ml of sterile heparin.
4. I.V. inject 100 ml of sterile heparin into tail of each mouse (5 - 10 mice).
10 After 10 minutes, bleed mice retro-orbitally into a sterile tube.
5. Measure O.D. of bacteria at 420 nm wavelength. Use THB+YE media as blank. (Use spectrophotometer 4040 to measure O.D.) 6. To a sterile flat bottom 96 well plate with sterile 2.5 mm glass bead in each well, add:
a. 50 ml of hepalil~ized blood.
b. 10 ml of serum c. 5 ml of bacteria Do this step in triplicate 7. Wrap plate in tinfoil and incubate at 37C incubator for one hour on a 20 shaker (slow motion.
8. After one hour, a 50 ml aliquot is plated out on blood agar plates using a glass spreader.
9. Wrap all plates in plastic wrapper and incubate at 37C for 12 hours.
10. Next day, count plaque forming colonies.
Serum from mice immlmi~P~l with a S. pneumoniae type 8 oligosaccharide conjugate was found the be immllnnprotective as measured by the opsonization assay. Opsonization of S. pneumoniae bacteria mPfli~te~l by specific anti-capsular antibodies is essential for host defense (Saunders, et al., 30 1993). This assay is generally considered a reliable indication of immllnnprotective capability in vivo. Results from assays show that antibodies to the 8 oligo-conjugate greatly reduce growth of colony forming units of S.
pneumoniae serotype 8 on blood agar plates (Table 22). This reduction was specific, as colony growth of serotypes 3 and 6B (used as specificity controls) 5 were not inhibited. ~ on with the unconjugated oligosaccharide or polysaccharide (which is used in the commercially available pneumoniae vaccine) elicited no protection. Protection elicited with the polysaccharide-conjugate was much less (39% reduction) than the protection elicited with the oligosaccharide conjugate (98% reduction). These results demon~tldt~ that our 8 oligo-tetanus 10 toxoid conjugate elicits high levels of immunoprotective antibodies against the serotype 8 S. pneumoniae pathogen. The level of imml~nnprotective antibody elicited by poly-conjugates was marginal.

As well, the 8-oligo conjugate could elicit an immllnoprotective antibody 15 response in mice previously ~(lmini~tered the whole polysaccharide alone. Mice injected with 2 doses of 8-polysaccharide followed by a tertiary oligo-conjugateinjection had immunoprotective antibodies in their serum (70% colony reduction in opsonization assay). As in previous experiments, mice receiving 3 injections of polysaccharide elicited no significant amount of protective antibody. Specific 20 oligosaccharide serotypes coupled to a carrier protein may be beneficial as abooster to augment the immunoprotection of high risk groups, non-responsive or only lllalghlally responsive to the current 23-valent polysaccharide vaccine.

We have performed an immunogenicity study with di-hapten 3 oligo/8 25 oligo-tetanus toxoid conjugates. Oligosaccharides of both serotypes were prepared by TFA hydrolysis. Mice injected with this multi-hapten conjugate elicited immunoprotective antibodies to the 3 and 8 serotypes (96 - 99% colony reduction - Table 23). A 3/8-polysaccharide conjugate elicited little immunoprotective antibody (10 - 12%). The mono-hapten 3 oligo-tetanus toxoid 30 conjugate used in this study was not prepared with oligosaccharides that had been 21S37~

- so -determined to have immlmogenic epitopes by inhibition ELISA and was not capable of eliciting an irnmunoprotective response. The mechanism which allows the immlm~ system to response to epitopes on the 3 oligosaccharide in thedi-hapten form is, of course, speculative. However, we suggest that the 8 5 oligosaccharides stim~ te clones of cell (i.e. accessory or helper cells) which can augment the response to the epitopes on the serotype 3 oligosaccharide.

We have discovered that the 8 oligosaccharide structure has adjuvant or adjuvant "like" activity. The relatively simple repeating unit structure of the 8-1 0 oligosaccharide (~-glucose (1 ~ 4) ,~-Glucose (1 ~ 4)a-galactose (1 ~ 4) a gluconic acid) may specifically or non-specifically ctim~ te/activate i~
cells or induce receptors or factors to enhance a humoral/cellular response to non-immlmogenic or weakly immlm~genic polysaccharides/oligosaccharides.
Serotype 8 oligosaccharides has adjuvant activity in conjugate form or as an 1 5 ~(lmixtllre to the vaccine formulation.

Opsonization results of a 14-oligosaccharide-TT conjugate (0.1 M TFA
preparation - Table 24) show good bacterial colony reduction of the 14 serotype (76%). The 14-oligo-TT 0.5 M TFA plepal~tion elicited less immllnoprotective 20 antibody (54% reduction). The serums from the polysaccharide-TT conjugate, the polysaccharide alone and the tetanus toxoid injected mice showed greatly reduced inhibition capacity (18, 2 and 15% respectively). Serum from control mice (0.9 NaCL injected and NMS) showed no reductive capacity.

Di-hapten-oligosaccharide conjugates also elicited antibody with opsonic activity. A serum to a 8:14-oligo-TT conjugate reduced serotype 14 colony forming units by 65% (Table 25). This di-hapten conjugate was as immlmogenic as the mono-hapten 14-conjugate (reduction of CFU = 68%). Serum from mice immllni~ed with the polysaccharide-conjugate marginally reduced CFU's by 30 37%.

- Sl - 21 $3 733 Example 9:
Circumvention of Carrier Suppression and Reduction of Antigenic Competition ~educed responses due to antigenic competition when multiple antigens are injected has been reported in the literature under some conditions. Results obtained from i.,~"~ni~ ion schedules A and D (Table 26) will be used to determine if the response to each component of our multi-hapten conjugate is equal to the response elicited by the single mono-hapten conjugates.
The unit mass of carbohydrate antigen of our mono- and multi-hapten conjugates will be equivalent (i.e., l :2 CHO:protein ratio for EDC conjugates).The design of our multi-hapten conjugates using reduced antigen load will ",il~i",i,e the potential for developing antigenic colllpc;Lilion.

Schedules B and E will determine if a plilllaly injection with the conjugate is sufficient to educate the immlm~ system to elicit a T dependent response when boosted with uncoupled polysaccharide(s).

Schedules C and F will establish the capability of our conjugates to enhance immllnoprotective antibody responses in mice previously primed with polysaccharide(s) alone. If so, a multi-hapten pneumoniae vaccine coll~ g oligosaccharides of 3 to 4 serotypes may be very useful to augment the response to Pneumovax~ 23 in high risk patients.
Groups of mice will be injected by 3 doses (1, 2, 3) of tetanus toxoid (titers to tetanus toxoid to be confirmed by ELISA) followed by 3 injections of various S. pneumoniae oligo or poly-TT conjugates as in G (Table 26).
In all studies, conjugates will be ~tlmini.ctered orally and by subcutaneous injection.

-- 52 - 21 ~3 73~

The conjugates of the present invention will stimnlAte immnn~ responses in infants, in children with i~ llAIIlle imml-n~ ~y~Lell~s and in the immlmosuppressed. As models for these situations, we will d~le.ll-il~e the 5 immnn~po~ g efficacy of our conjugates in young mice, in SCID and nude mice. As described above, these mice will also be pre-sensitized with tetanus toxoid prior to multi-conjugate inoculation to study the carrier ~upplession phenomenon.

Modification of the above-described modes of carrying out the various embodiments of this invention will be ~al~llL to those skilled in the art following the te~c-hing~ of this invention as set forth herein. The examples described above are not limiting, but are merely exemplary of this invention, the scope of which is defined by the following claims.

Claims (22)

THE EMBODIMENTS OF THE INVENTION IN WHICH AN EXCLUSIVE
PROPERTY OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:

1. A composition comprising:
a) at least one size-separated carbohydrate hapten comprising at least one immunogenic epitope; and b) a carrier, wherein said hapten is covalently coupled to said carrier and wherein said hapten-carrier conjugate is protectively immunogenic.

2. The composition of Claim 1 wherein said hapten is an oligosaccharide of a bacterial or viral polysaccharide.

3. The composition of Claim 1 wherein the presence of said immunogenic epitope is determined using inhibition ELISA.

4. The composition of Claim 2 wherein said oligosaccharide is produced by acid hydrolysis of said polysaccharide.

5. The composition of Claim 1 wherein said protective immunogenicity is determined by isotype ELISA.

6. The composition of Claim 1 wherein said protective immunogenicity is determined by bactericidal or opsonization assay.

7. The composition of Claim 2 wherein said polysaccharide is selected from the group consisting of capsular polysaccharides of S. pneumococcus serotypes 1, 2, 3, 4, 5, 6B, 7, 7F, 8, 9N, 9V, 10A, 11A, 12, 12F, 14, 15B, 17F, 18C, 19F, 19A, 20, 22, 23F and 33F.

8. The composition of Claim 1 which comprises two or more haptens.

9. The composition of Claim 1 which does not induce carrier suppression.

10. The composition of Claim 1 which does not induce antigenic competition.

11. The composition of Claim 1 further comprising an adjuvant.

12. A method of making a conjugate composition comprising:
a) cleaving a bacterial or viral polysaccharide into oligosaccharides so as to preserve immunogenic epitopes on the resulting oligosaccharides;
b) separating the resulting oligosaccharides based on size;

c) selecting those oligosaccharides which contain immunogenic epitopes based on inhibition ELISA;
d) activating the oligosaccharides selected in step c); and e) coupling the activated oligosaccharides to a purified carrier, wherein the resulting composition is protectively immunogenic.

13. The method of Claim 12 wherein said cleavage is performed using acid hydrolysis.

14. The method of Claim 12 wherein said activation is acidification on a cation column.

15. The method of Claim 12 wherein said coupling is performed using EDC
or periodate.

16. The method of Claim 12 wherein said coupling provides a predictable ratio of hapten to carrier.

17. The method of Claim 12 wherein steps a) through d) are repeated using at least one additional bacterial or viral polysaccharide to result in a di- or multi-valent conjugate.

18. The method of Claim 17 wherein said bacterial polysaccharides are selected from the group consisting of capsular polysaccharides of S.
pneumococcus serotypes 1, 2, 3, 4, 5, 6B, 7, 7F, 8, 9N, 9V, 10A, 11A, 12, 12F, 14, 15B, 17F, 18C, 19F, 19A, 20, 22, 23F and 33F.

19. A method of providing protective immunization against a bacterial pathogen comprising administering to a mammal in need of such treatment an effective amount of the composition of Claim 1.

20. The method of Claim 19 wherein said administration is selected from the group consisting of oral and parenteral.

21. The method of Claim 19 wherein said mammal is a neonate.

22. The method of Claim 19 wherein said mammal is selected from the group consisting of immune suppressed mammals and elderly mammals.