AU6049599A

AU6049599A - Streptococcal c beta protein compositions

Info

Publication number: AU6049599A
Application number: AU60495/99A
Authority: AU
Inventors: Milan S Blake; Karen O. Long-Rowe
Original assignee: Baxter Healthcare SA
Current assignee: Baxter Healthcare SA
Priority date: 1998-09-17
Filing date: 1999-09-17
Publication date: 2000-04-03
Also published as: EP1114143A4; NO20011381L; WO2000015760A1; JP2002525041A; PL347058A1; EP1114143A1; HUP0103502A2; SK3482001A3; CA2343051A1; CZ2001951A3; NO20011381D0

Description

WO 00/15760 PCT/US99/21643 STREPTOCOCCAL C BETA PROTEIN COMPOSITIONS Background of the Invention Field of the Invention 5 The invention concerns protein fragments or peptides that elicit antibodies which are bactericidal to Gram positive bacteria with complement alone, i.e., the activity is not dependent on an opsonophagocytotic mechanism. The present invention also concerns fragments of the Group B streptococcal beta antigen as well as mutants thereof having a reduced or eliminated ability to bind human IgA. 10 These fragments retain the immunological properties useful for formulating a conjugate vaccine against Group B streptococci. The invention also concerns the isolation and purification of the Group B streptococcal CP protein and mutants and/or fragments thereof. Related Art 15 Streptococci are a large and varied set of Gram positive bacteria which have been ordered into several groups based on the antigenicity and structure of their cell wall polysaccharide (Lancefield, R.C., J Exp. Med 57:571-595 (1933); Lancefield, R.C., Proc. Soc. Exp. Biol. and Med 38:473-478 (1938)). Two of these groups have been associated with serious human infections. Those that 20 have been classified into Group A streptococci are the bacteria that people are most familiar and are the organisms which cause "strep throat." Organisms of Group A streptococci also are associated with the more serious infections of rheumatic fever, streptococcal impetigo, and sepsis. Group B streptococci were not known as a human pathogen in standard 25 medical textbooks until the early 1970s. Since that time, studies have shown that Group B streptococci are an important perinatal pathogen in both the United States as well as the developing countries (Smith, A.L. and Haas, J., Infections ofthe Central Nervous System, Raven Press, Ltd., New York, p. 313-333 (1991)).

WO 00/15760 PCT/US99/21643 -2 Systemic Group B streptococcal infections during the first two months of life affect approximately three out of every 1000 births (Dillon, H.C., Jr., et al., J Pediat. 110:31-36 (1987)), resulting in 11,000 cases annually in the United States. These infections cause symptoms of congenital pneumonia, sepsis, and 5 meningitis. A substantial number of these infants die or have permanent neurological sequelae. Furthermore, these Group B streptococcal infections may be implicated in the high pregnancy-related morbidity which occurs in nearly 50,000 women annually. Others who are at risk from Group B streptococcal infections include those who either congenitally, chemotherapeutically, or by 10 other means, have an altered immune response. Group B streptococci can be further classified into at least nine different types, such as types Ia, Ib, II, III, IV, V, VI, VII, and VIII, based on the bacteria's capsular polysaccharide. The most pathogenically important of these different types are streptococci having types Ia, Ib, II, III, and V capsular polysaccharides. 15 Group B streptococci of these four types represent over 90% of all reported cases. The structure of each of these various polysaccharide types has been elucidated and characterized (Jennings, H.J., et al., Biochemistry 22:1258-1263 (1983); Jennings, H.J., et al., Can. J Biochem. 58:112-120 (1980); Jennings, H.J., et al., Proc. Nat. Acad Sci. USA. 77:2931-2935 (1980); Jennings, H.J., et al., J. Biol. 20 Chem. 258:1793-1798 (1983); Wessels, M.R., et al., J. Biol. Chem. 262:8262-8267 (1987)). As is found with many other human bacterial pathogens, it has been ascertained that the capsular polysaccharides of Group B streptococci, when used as vaccines, provide very effective, efficacious protection against infections with these bacteria. This was first noted by Lancefield (Lancefield, 25 R.C., et al., J Exp. Med 142:165-179 (1975)) and more recently in the numerous studies of Kasper and coworkers (Baker, C.J., et al., N Engl. J. Med 319:1180-1185 (1988); Baltimore, R.S., et al., J. Infect. Dis. 140:81-86 (1979); Kasper, D.L., et al., J. Exp. Med 149:327-339 (1979); Madoff, L.C., et al., J Clin. Invest. 94:286-292 (1994); Marques, M.B., et al., Infect. Immun. 30 62:1593-1599 (1994); Wessels, M.R., etal., J Clin. Invest. 86:1428-1433 (1990); Wessels, M.R., et al., Infect. Immun. 61:4760-4766 (1993); Wyle, S.A., et al., J WO 00/15760 PCT/US99/21643 -3 Infect. Dis. 126:514-522 (1972)). However, much like many other capsular polysaccharide vaccines (Anderson, P., et al., J. Clin. Invest. 51:39-44 (1972); Gold, R., et al., J. Clin. Invest. 56:1536-1547 (1975); Gold, R., et al., J. Infect. Dis. 136S:S31-S35 (1977); Gold, R.M., etal., J. Infect. Dis. 138:731-735 (1978); 5 Makeld, P.R.H., et al., J. Infect. Dis. 136:S43-50 (1977); Peltola, A., et al., Pediatrics 60:730-737 (1977); Peltola, H., et al. N. Engl. J. Med. 297:686-691 (1977)), vaccines formulated from pure type Ia, Ib, II, and III capsular carbohydrates are relatively poor immunogens and have very little efficacy in children under the age of 18 months (Baker, C.J. and Kasper, D.L., Rev. Inf Dis. 10 7:458-467 (1985); Baker, C.J., et al., N. Engl. J. Med. 319:1180-1185 (1988); Baker, C.J., et al., New Engl. J Med. 322:1857-1860 (1990)). These pure polysaccharides are classified as T cell independent antigens because they induce a similar immunological response in animals devoid of T lymphocytes (Howard, J.G., et al., Cell. Immunol. 2:614-626 (1971)). It is thought that these 15 polysaccharides do not evoke a secondary booster response because they do not interact with T cells, and therefore fail to provoke a subsequent "helper response" via the secretion of various cytokines. For this reason, each consecutive administration of the polysaccharide as a vaccine results in the release of a constant amount of antibodies, while a T cell dependent antigen would elicit an 20 ever increasing concentration of antibodies each time it was administered. Goebel and Avery found in 1931 that by covalently linking a pure polysaccharide to a protein, they could evoke an immune response to the polysaccharide which could not be accomplished using the polysaccharide alone (Avery, O.T. and Goebel, W.F., J. Exp. Med. 54:437-447 (1931); Goebel, W.F. 25 and Avery, O.T., J. Exp. Med. 54:431-436 (1931)). These observations initiated and formed the basis of the current conjugate vaccine technology. Numerous studies have followed which show that when polysaccharides are coupled to proteins prior to their administration as vaccines, the immune response to the polysaccharides changes from a T-independent response to a T-dependent 30 response. For reviews, see Dick, W.E., Jr. and Beurret, M., Glycoconjugates of bacterial carbohydrate antigens In: Contributions to Microbiology and WO 00/15760 PCT/US99/21643 -4 Immunology. Cruse et al., eds., p. 48-114 (1989); Jennings, H.J. and Sood, R.K., Neoglycoconjugates. Preparation and Applications. Lee, Y.C. and Lee, R.T., eds., Academic Press, New York, p. 325-371 (1994); and Robbins, J.B. and Schneerson, R., J. Infect. Dis. 161:821-832 (1990). Currently, most of these 5 polysaccharide-protein conjugate vaccines are formulated with well known proteins such as tetanus toxoid and diphtheria toxoid or mutants thereof. These proteins were originally used because they were already licensed for human use and were well characterized. However, as more and more polysaccharides were coupled to these proteins and used as vaccines, interference between the various 10 vaccines which used the same protein became apparent. For example, if several different polysaccharides were linked to tetanus toxoid and given sequentially, the immune response to the first administered polysaccharide conjugate would be much larger than the last. If, however, each of the polysaccharides were coupled to a different protein and administered sequentially, the immune response to each 15 of the polysaccharides would be the same. Carrier suppression is the term used to describe this observed phenomenon. One approach to overcome this problem is to match the protein and polysaccharide so that they are derived from the same organism. Among the various antigens used to classify and subgroup Group B 20 streptococci, one was a protein known as the Ibc antigen. This protein antigen was first described by Wilkinson and Eagon in 1971 (Wilkinson, H.W. and Eagon, R.G., Infect. Immun. 4:596-604 (1971)) and was known to be made up of two distinct proteins designated as alpha and beta. Later, the Ibc antigen was shown to be effective when used as a vaccine antigen in a mouse model of 25 infection by Lancefield and co-workers (Lancefield, R.C., et al., J. Exp. Med 142:165-179 (1975)). The isolation, purification and functional characterization of the beta antigen (Co) protein of Group B streptococci was accomplished by Russell-Jones, et al. (Russell-Jones, G.J. and Gotschlich, E.C., J. Exp. Med 160:1476-1484 30 (1984); Russell-Jones, G.J., et al., J Exp. Med 160:1467-1475 (1984)).

WO 00/15760 PCT/US99/21643 -5 In testing the IgA binding activity of surface proteins from various GBS strains, Russell-Jones et al. initially extracted the CP protein using a Triton® X 100 detergent method (Russell-Jones, G.J. et al., JExp.Med 160: 1467-1475 (1984)). Although they identified a 130 kD protein that consistently associated 5 with IgA binding activity, the protein was always contaminated with smaller proteins, only some of which bound IgA. They speculated that these lower molecular weight proteins were degradation products from an uncharacterized bacterial protease. Russell-Jones et al. reduced the proteolytic activity by extracting the C3 protein in hot SDS, and then purified the protein by SDS gel 10 chromatography. Although they obtained a substantially pure protein, this method did not provide a final product in sufficient quantity and purity for subsequent conjugation to bacterial polysaccharide. Madoff et al. obtained CP protein using preparative SDS-PAGE and electroelution of CP protein (Madoff et al., Infect. Immun. 60 (12):4989-4994 15 (1992)). This method also did not lend itself to scale-up. Jerlstrdm et al. (1996) obtained fragments of the CP protein from inclusion bodies produced in cells which overexpressed the C3 protein fragments (Jerlstr6m et al., Infect. Immun. 64 (7): 2787-2793 (1996)). The inclusion bodies were first solubilized in 8M urea. The resulting supernatant was then subjected to FPLC using a Mono Q® 20 ion-exchange column. Although yields and purity were not described in detail, it appears that some degree of purification was achieved simply by isolating the inclusion bodies. Russell-Jones et al. demonstrated that one of the properties of the CP protein was to bind specifically to human IgA immunoglobulin. The binding site 25 on the IgA molecule was localized to the Fc portion of the heavy chain of this immunoglobulin. They further showed that the CP protein consisted of a single polypeptide having an estimated molecular weight of 130 kD. The gene responsible for the expression of the CP protein was cloned (Cleat, P.H. and Timmis, K.N., Infect. Immun. 55:1151-1155 (1987)) and sequenced (JerlstrOm, 30 P.G., et al., Molec. Microbiol. 5:843-849 (1991)) by a group led by Timmis. His later study demonstrated that the IgA binding activity could be assigned to a 746 WO 00/15760 PCT/US99/21643 -6 bp DNA fragment of the gene defined by a leading BglII restriction endonuclease cleavage site and ending with a HpaI restriction endonuclease cleavage site. As stated previously, the 1975 Lancefield study showed that the Ibc antigen was an effective vaccine antigen in a mouse model of Group B 5 streptococcal infection (Lancefield, R.C., et al., J. Exp. Med. 142:165-179 (1975)). It was not clear at the time whether the alpha or beta protein component of the Ibc antigen was responsible for this protection. Madoffet al. began to shed light on this question and demonstrated that the purified CP protein used as a vaccine could protect infant mice from experimental infection with Group B 10 streptococci expressing this protein (Madoff, L.C. et al., Infect. Immun. 60:4989-4994 (1992)). Madoff et al. also showed that when they coupled a Type III streptococcal capsular polysaccharide to the CP3 protein, this vaccine would protect infant mice against infection with either a Type III Group B streptococci (expressing no CP) or a Type Ib Group B streptococci (expressing Cp, but lacking 15 a Type III capsular polysaccharide) (Madoff, L.C., et al., J Clin. Invest. 94:286-292 (1994)). Thus, this CP protein conjugate vaccine served several functions: the polysaccharide elicited protective antibodies to the polysaccharide capsule and the C3 protein evoked protective antibodies to the protein as well as modified the immune response to the polysaccharide from a T-independent 20 response to a T-dependent response. This polysaccharide-CP protein conjugate strategy works well in mice. But clearly, the goal is to protect humans against Group B streptococcal infections. The only caveat with using the same strategy in humans is that the CP protein binds human IgA immunoglobulins non-specifically (CP does not bind 25 specifically mouse IgA). This human IgA binding activity of CP may diminish the efficacy of a polysaccharide-CP protein conjugate vaccine for humans, as antigens bound to IgA may be cleared from the system so rapidly that an antigen specific antibody response is not produced. Furthermore, potentially protective epitopes on the C[ protein may be hidden when the human IgA binds to the C3 30 molecule. Thus, it would be advantageous to obtain a mutant CP3 protein which WO 00/15760 PCT/US99/21643 -7 lacks the IgA binding capacity, but retains as much of the native structure as possible. With this goal in mind, several groups have attempted to determine the IgA binding region of the CP protein. Brady and Boyle (Infect. Immun. 57:1573 5 1581 (1989)) identified certain forms of the Group B streptococcal 3 antigen which did not bind the IgA Fc region. Brady and Boyle also observed that certain strains examined in their study secreted low-molecular-weight forms of the antigen. HG806 secreted an approximately 38 kD form which did not exhibit human IgA-Fc binding. Two other strains, 2AR and DL471B, secreted 10 approximately 55 and 53 kD forms, respectively. These forms exhibited IgA-Fc binding, however. Furthermore, Jerlstr6m et al. (Molec. Microbiol. 5:843-849 (1991)) used experiments wherein subfragments of the CP protein were expressed as fusion proteins to identify two regions of the CP protein capable of binding IgA. These 15 experiments localized the IgA binding domains to a 747 bp BglII-HpaI fragment and a 1461 bp HpaI-HindIII fragment of the CP protein. U.S. PatentNos. 5,595,740 and 5,766,606 describe deletion mutants of CP protein which lack a larger region encompassing this domain and do not bind IgA. In addition, International Patent Application No. PCT/US97/15319 describes a 20 12 amino acid domain of the CP protein, the mutation of which results in decreased or eliminated IgA binding. Summary of the Invention The invention relates to protein fragments or peptides that elicit antibodies which are bactericidal to Gram positive bacteria with complement alone, i.e., the 25 activity is not dependent on an opsonophagocytotic mechanism. The invention also relates to a protein fragment or peptide related to the Group B streptococcal (GBS) beta antigen, which comprises the immunogenic IgG binding domain of the protein. Preferably, the IgA binding by the protein fragment or peptide is reduced or eliminated.

WO 00/15760 PCT/US99/21643 -8 In particular, the invention relates to a protein fragment or peptide comprising a proline-rich region, wherein at least every third residue is proline. In a preferred embodiment, the protein fragment or peptide comprises continuous (repeated) amino acid sequences having the formula: -[P-Y1-Y 2

-P-YIY

2 r- or 5 [Y,-Y2-P-Y,-Y2-P]r-, where Yl represents either an acidic or basic amino acid residue, Y 2 represents a neutral amino acid, and r is an integer which may range from one to five. More preferably, Y 1 is D or K and Y 2 is V or L. Also more preferably, r is four. Most preferably, Y 2 is a hydrophobic amino acid. The invention also relates to a peptide or protein fragment ofthe invention 10 having the formula Y-X-Z, wherein X represents at least eight contiguous amino acid residues between amino acids 828 and 1027, inclusive, of FIG. 1 (SEQ ID NO: 2) (wild-type CP protein), Y represents hydrogen or the N-terminal amino acid sequence of FIG. 1 (SEQ ID NO: 2) that is bound to X, or an N-terminal fragment and/or mutant thereof, and Z represents hydrogen or the C-terminal 15 amino acid sequence of FIG. 1 (SEQ ID NO: 2) that is bound to X, or a C terminal fragment and/or mutant thereof, with the proviso that at least one of amino acids 1-164 of FIG. 1 (SEQ ID NO: 2), if present in Y, is non-wild-type, and with the further proviso that the protein is at least one of an N-terminal or C terminal fragment of the amino acid sequence of FIG. 1 (SEQ ID NO: 2). 20 The invention also relates to a peptide or protein fragment of the invention as described above, wherein Y represents an amino acid sequence comprising A

X

1

X

2

X

3

X

4

X

5

X

6

X

7

X

8

X

9 Xo 1 0

X

1 I X 12 -B, wherein A comprises amino acids 1 164 of the sequence shown in FIG. 1 (SEQ ID NO: 2) or an N-terminal fragment thereof, B represents a sequence starting from amino acid 177 and terminating at 25 an amino acid bound to X, and X, - X12 are each selected independently from the group consisting of Ala, Val, Leu, Ile, Pro, Met, Phe, Trp, a bond, and the wild type amino acid found at the corresponding position of the sequence shown in FIG. 1 (SEQ ID NO: 2), wherein the amino acid positions are numbered from the first amino acid of the native amino acid sequence encoding the protein, with the 30 proviso that at least one of X, through X 12 , inclusive, is other than the wild type amino acid.

WO 00/15760 PCT/US99/21643 -9 The invention also relates to a polynucleotide molecule encoding a peptide or protein fragment of the invention, as well as vectors comprising such polynucleotide molecules, and host cells transformed therewith. The invention also relates to a conjugate comprising a peptide or protein 5 fragment of the invention covalently conjugated to a capsular polysaccharide. Alternatively, the peptide or protein fragment may be conjugated to or complexed with a porin protein, proteosome, or other carrier protein. The invention also relates to a vaccine comprising a peptide or protein fragment of the invention and a pharmaceutically acceptable carrier. 10 The invention also relates to a method of inducing an immune response in an animal, comprising administering the vaccine of the invention to an animal in an effective amount. The invention also relates to a process for obtaining a substantially pure C3 protein or fragment and/or mutant thereof, comprising: 15 (a) obtaining the CP protein in cell extracts; (b) subjecting the CP protein to ion exchange chromatography and collecting the C3 protein-containing fractions; (c) pooling and diluting the C3 protein-containing fractions; and (d) subjecting the diluted CP protein-containing fractions to ligand 20 affinity or a gel filtration chromatography and collecting the fractions; whereby substantially pure CP3 protein or fragment and/or mutant thereof is obtained. The invention further relates to a process for obtaining a CP protein or a fragment and/or mutant from bacterial cells which are transfected with nucleotide 25 sequences encoding the CP protein and/or a fragment or mutant thereof, wherein the cells overexpress the CP protein or a fragment or mutant thereof. The invention also relates to a process for obtaining a CP3 protein from bacterial cells which naturally produce the CP protein.

WO 00/15760 PCT/US99/21643 -10 Brief Description of the Figures FIG. 1 shows the DNA sequence and deduced amino acid sequence of wild type C[1 (Jerlstrdm, P.G. et al., Molec. Microbiol. 5:843-849 (1991)). The BglII and PstI sites shown in FIGS. 2, 3 and 4 are identified. 5 FIGS. 2A and 2B show protein elution profiles for Q® column (1.6 x 20 cm) and heparin column (1.6 x 20 cm) chromatography, respectively, under the following conditions: flow rate - 5 ml/minute; attenuation - 0.2; and gradient 0 - 0.1 M NaCl/0.5% Zwittergent®/25 mM tris(HCL), pH 7.6 - 7.8, 10 ml/fraction. 10 FIGS. 3A and 3B show of CP protein purification data, and SDS-PAGE of corresponding preparations. FIG. 3A illustrates purification of CP protein on Q® and heparin columns. FIG. 3B depicts purification of CP protein on Q® and S-300 gel-filtration columns. FIGS. 4A and 4B illustrate thermolysin digestion of CP protein. FIG. 4A 15 shows SDS-PAGE illustrating the digestion pattern with resulting fragments of apparent molecular weights of 35 kD and 25 kD on tricine gels. FIG. 4B is a western transfer probed with antibody 52.2, showing that major fragments react with antibody. FIG. 5 shows separation and recovery of CP protein fragments by western 20 transfer and subsequent dissolution of nitrocellulose membrane. Aliquots were run on SDS-gels to confirm protein recovery. FIG. 6 shows SDS-PAGE of thermolysin-digested CP protein separated by gel-filtration on Sephacryl'-100. FIG. 7 shows SDS-PAGE of Arg-C fragments of the TI peptide. 25 FIG. 8 shows the sequence of native C3 protein from S. agalactiae. N-terminal sequence analysis indicated where thermolysin generated fragments TI and T2 begin (N-terminal section sequenced is underscored). Theoretical T1 fragment digestion by Arg-C is shown by circles. FIGS. 9A and 9B are graphic representations of antibody binding to 8-mer 30 and 20-mer mimotopes, respectively. Duplicate experiments were performed WO 00/15760 PCT/US99/21643 -11 with each mimotope with washing between each binding experiment (two bars). Some variability is due to incomplete washing. FIG. 10 is a graphic representation of inhibition of antibody binding to 20-mer mimotopes by pre-exposing the antibodies to a synthetic peptide 5 corresponding to amino acids 880-906 of FIG. 8. One set of data for the 20-mers of the experiments in FIG. 9B (open bars) is included as a positive control for antibody not exposed to the synthetic peptide. FIG. 11 is a graphic representation of bactericidal inhibition by C[ protein and thermolysin-digested fragments thereof. 10 FIG. 12 is a graphic representation of the survival of Group B streptococcal Ib in a bactericidal assay using antibody to native C3 protein. FIG. 13 is a graphic representation of the survival of Group B streptococcal Ib in a bactericidal assay using Ser-peptide antibody to CP protein. FIG. 14 depicts Ser-peptide antibody binding on Group B streptococcal 15 Ib colony lifts using antiserum purified by tetanus column alone and tetanus then Protein G columns. FIG. 15 is a graphic representation of the correlation of Group B streptococcal Ib growth and Ser-peptide antibody binding. Ser-peptide antibody binding begins during the lag growth phase and peaks during the log growth 20 phase. As bacteria approach the stationary phase and division ceases, antibody binding decreases to zero. FIG. 16 is a graphic representation of rabbit antibody 52.2 binding and inhibition of binding to overlapping 8-mer mimotope pins which represent sequences of the Arg-C digested R2 peptide fragment. Specifically, the bar graph 25 illustrates the sequences with the highest antibody affinity (unabsorbed) and the ability of intact C3 protein to inhibit antibody 52.2 to mimotopes (absorbed).

WO 00/15760 PCT/US99/21643 -12 Detailed Description of the Preferred Embodiments The antigenic protein fragments and peptides of the invention were found to elicit antibodies capable of killing streptococci with complement alone. This bactericidal activity, as opposed to opsonophagocytosis wherein the bacteria are 5 killed inside a white blood cell, has never been observed with Gram positive bacteria using an antibody specific therefor. In fact, antibodies elicited by the capsular polysaccharide of streptococci expressing the CP protein do not kill streptococci unless white cells are present. Investigation of the antigenic character of the CP protein led to the 10 discovery of an antigenic region with characteristics shared among Gram positive bacteria. Antibodies raised against protein fragments and peptides based on the amino acid sequence of this region of the C3 protein possess bactericidal activity which is not dependent on an opsonophagocytotic mechanism. Thus, the invention relates to protein fragments or peptides that elicit antibodies which are 15 bactericidal to Gram positive bacteria with complement alone, i.e., activity that is not dependent on an opsonophagocytotic mechanism. The antigenic protein fragments and peptides of the invention are derived from Gram positive amino acid sequences at the C-terminal end of the native proteins and are believed to be embedded in the bacterial cell wall. In the native 20 bacterial proteins, the antigenic regions corresponding to the protein fragments and peptides of the invention are exposed to antibody binding during the events of cellular division. In the area where one bacterium starts to form two bacteria, the cell wall has to be broken down into small pieces and remodeled. This area is known as the septal plane. It would be precisely in this area during the time of 25 division that the antigenic regions would be available for antibody binding and activation of complement. Confocal microscopy results (Example 13) clearly demonstrate that antibodies generated to the protein fragments bind specifically in the area of the septal plane and more of the antibodies bind to actively dividing cells as opposed to non-dividing cells. The results of the FACScan analysis WO 00/15760 PCT/US99/21643 -13 confirmed these results and added further evidence to this observation. In addition, Coleman et al. previously made similar observations. Coleman et al. used colloidal gold immunolabeling to study immunoglobulin-binding sites and CP antigen in Group B streptococci. Coleman 5 et al., Infect. Immun., 58:332-340 (1990). Colloidal gold was conjugated to IgA to characterize IgA binding properties and to IgG to characterize IgG binding. The electron micrographs of Coleman et al. showed that the nonimmunological human IgA binding was distributed over the surface of the bacteria, while the anti-CP specific IgG binding was localized to the area defining the septal planes 10 of dividing bacterial cells. Therefore, one object ofthe invention was to determine whether a specific site on the CP protein was responsible for eliciting antibodies capable of this unique bactericidal activity. Identification of such a site allows one to ensure that it is included within a vaccine for Group B streptococci. 15 In particular, the invention relates to a protein fragment or peptide comprising a proline-rich region, wherein at least every third residue is proline, and wherein antibodies raised against said protein fragment or peptide are bactericidal to Gram positive bacteria with complement alone. In a preferred embodiment, the invention relates to a protein fragment or peptide comprising a 20 continuous (repeated) amino acid sequences having the formula -[P-Y-Y 2

-P-Y

1 Y2]r- or -[Y 1

-Y

2

-P-Y-Y

2 -P]-, where Yi represents either an acidic or basic amino acid residue, Y 2 represents a neutral amino acid, and r is an integer from one to five. More preferably, Yi is D or K and Y 2 is V or L. Also more preferably, r is four. Most preferably, Y 2 is a hydrophobic amino acid. 25 In another embodiment, the protein fragment or peptide comprises a continuous (repeated) amino acid sequence having the formula -[P-D-Y3-P-K-L]r or -[K-L-P-D-Y3-P],- , wherein Y 3 represents V or A. More preferably, Y 3 is V. Also more preferably, r is at least 3. In another embodiment, the protein fragment or peptide comprises a 30 continuous (repeated) amino acid sequence having the formula -[S-P-K-Y 4

-P-E-

WO 00/15760 PCT/US99/21643 -14 A-P-Y5-V-P-E],-, wherein Y 4 represents T or A, and Y 5 represents H or R. More preferably, Y 4 is T and Y 5 is H. Also more preferably, r is at least 3. The invention also relates to a peptide or protein fragment of the invention, which comprises an immunogenic IgG binding domain capable of 5 eliciting antibodies as described above. Preferably, the IgA binding by the protein fragment or peptide is reduced or eliminated. In preferred embodiment, this domain comprises a peptide sequence of at least 8 amino acids of the sequence between about positions 828 and about 1027 of the amino acid sequence shown in FIG. 1 (SEQ ID NO: 2). 10 In another preferred embodiment, the invention relates to a peptide or protein fragment of the invention comprising an amino acid sequence having the formula Y-X-Z, wherein X represents at least eight contiguous amino acid residues between amino acids 828 and 1027, inclusive, of FIG. 1 (SEQ ID NO: 2), Y represents hydrogen or the N-terminal amino acid sequence of FIG. 1 15 (SEQ ID NO: 2) that is bound to X, or an N-terminal fragment and/or mutant thereof, and Z represents hydrogen or the C-terminal amino acid sequence of FIG. 1 (SEQ ID NO: 2) that is bound to X, or a C-terminal fragment and/or mutant thereof, with the proviso that at least one of amino acids 1-164 of FIG. 1 (SEQ ID NO: 2), if present in Y, is non-wild-type, and with the further proviso that the 20 protein is at least one of an N-terminal or C-terminal fragment of the amino acid sequence of FIG. 1 (SEQ ID NO: 2). More preferably, Y does not include at least amino acids 1-176 of FIG. 1 (SEQ ID NO: 2). Also more preferably, Z comprises at least amino acid 901 of FIG. 1 (SEQ ID NO: 2). Also more preferably, the peptide or protein fragment of the invention is not the amino acid sequence of 25 SEQ ID NO. 31. Also more preferably, the peptide or protein fragment of the invention is not the approximately 55, 53, or 38 kD polypeptide secreted by the group B streptococcus strain 2AR, DL471B, or HG 806, respectively. Preferably, epitopic sequence X is selected from the group consisting of PPKTPDVP (SEQ ID NO: 32), PDVPKLPD (SEQ ID NO: 33), KLPDVPKL 30 (SEQ ID NO: 34), VPKLPDVP (SEQ ID NO: 35), KLPDAPKL (SEQ ID NO: 36), APKLPDGL (SEQ ID NO: 37), ETPDTPKI (SEQ ID NO: 38), RTVRLALG WO 00/15760 PCT/US99/21643 -15 (SEQ ID NO: 39), and GGGTVRVF (SEQ ID NO: 40). In a preferred embodiment, X represents any one of eight, nine, ten, eleven, twelve, thirteen, fourteen, fifteen, sixteen, seventeen, eighteen, nineteen, twenty or twenty one contiguous amino acid residues between amino acids 828 and 1027, inclusive, of 5 FIG. 1 (SEQ ID NO: 2). In another embodiment, the invention relates to a peptide or protein fragment comprising an amino acid sequence consisting of any eight contiguous residues of an amino acid sequence between amino acids 828 and 1027, inclusive, of the sequence shown in FIG. 1 (SEQ ID NO: 2). More preferably, the eight 10 contiguous residues are selected from the group consisting of PPKTPDVP (SEQ ID NO: 32), PDVPKLPD (SEQ ID NO: 33), KLPDVPKL (SEQ ID NO: 34), VPKLPDVP (SEQ ID NO: 35), KLPDAPKL (SEQ ID NO: 36), APKLPDGL (SEQ ID NO: 37), ETPDTPKI (SEQ ID NO: 38), RTVRLALG (SEQ ID NO: 39), and GGGTVRVF (SEQ ID NO: 40). 15 In another embodiment, the invention relates to a protein fragment or peptide having the formula Y-X-Z, wherein X represents any twenty contiguous amino acid residues between amino acids 828 and 1027, inclusive, of FIG. 1 (SEQ ID NO: 2), Y represents hydrogen or the N-terminal amino acid sequence of FIG. 1 (SEQ ID NO: 2) that is bound to X, or an N-terminal fragment and/or 20 mutant thereof, and Z represents hydrogen or the C-terminal amino acid sequence of FIG. 1 (SEQ ID NO: 2) that is bound to X, or a C-terminal fragment and/or mutant thereof, with the proviso that at least one of amino acids 1-164 of FIG. 1 (SEQ ID NO: 2), if present in Y, is non-wild-type, and with the further proviso that the protein is at least one of an N-terminal or C-terminal fragment of the 25 amino acid sequence of FIG. 1 (SEQ ID NO: 2). More preferably, X is selected from the group consisting of SPKTPEAPKIPEPPKTPDVP (SEQ ID NO: 41), PEAPKIPEPPKTPDVPKLPD (SEQ ID NO: 42), KIPEPPKTPDVPKLPDVPKL (SEQ ID NO: 43), PPKTPDVPKLPDVPKLPDVP (SEQ ID NO: 44), PDVPKLPDVPKLPDVPKLPD (SEQ ID NO: 45), and 30 KLPDVPKLPDVPKLPDAPKL (SEQ IDNO: 46). Also more preferably, Y does not include at least amino acids 1-176 of FIG. 1 (SEQ ID NO: 2).

WO 00/15760 PCT/US99/21643 -16 In another embodiment, the invention relates to a peptide or protein fragment comprising any twenty contiguous residues of the amino acid sequence between amino acids 828 and 1027, inclusive, of the sequence shown in FIG. 1 (SEQ ID NO: 2). More preferably, the twenty residues are selected from the 5 group consisting of SPKTPEAPKIPEPPKTPDVP (SEQ ID NO: 41), PEAPKIPEPPKTPDVPKLPD (SEQ ID NO: 42), KIPEPPKTPDVPKLPDVPKL (SEQ ID NO: 43), PPKTPDVPKLPDVPKLPDVP (SEQ ID NO: 44), PDVPKLPDVPKLPDVPKLPD (SEQ ID NO: 45), and KLPDVPKLPDVPKLPDAPKL (SEQ ID NO: 46). 10 In another embodiment, the invention relates to a peptide or protein fragment having the formula Y-X-Z, wherein X is at least 21 contiguous residues between amino acids 828 and 1027, inclusive, of FIG. 1 (SEQ ID NO: 2), Y represents hydrogen or the N-terminal amino acid sequence of FIG. 1 (SEQ ID NO: 2) that is bound to X, or an N-terminal fragment and/or mutant thereof, and 15 Z represents hydrogen or the C-terminal amino acid sequence of FIG. 1 (SEQ ID NO: 2) that is bound to X, or a C-terminal fragment and/or mutant thereof, with the proviso that at least one of amino acids 1-164 of FIG. 1 (SEQ ID NO: 2), if present in Y, is non-wild-type, and with the further proviso that the protein is at least one of an N-terminal or C-terminal fragment of the amino acid sequence of 20 FIG. 1 (SEQ ID NO: 2). More preferably, X consists of the amino acid sequence between amino acids 828 and 1027, inclusive, of the sequence shown in FIG. 1 (SEQ ID NO: 2). Also more preferably, Y does not include at least amino acids 1-176 of FIG. 1 (SEQ ID NO: 2). In another embodiment, the invention relates to a peptide or protein 25 fragment comprising an amino acid sequence consisting of at least 21 contiguous residues of the amino acid sequence between amino acids 828 and 1027, inclusive, of the sequence shown in FIG. 1 (SEQ ID NO: 2). More preferably, the peptide or protein fragment of the invention comprises an amino acid sequence consisting of the amino acid sequence PDVPKLPDVPKLPDVPKLPDAPKL 30 (SEQ ID NO: 47). Most preferably, the peptide or protein fragment of the invention comprises an amino acid sequence consisting of the amino acid WO 00/15760 PCT/US99/21643 -17 sequence between amino acids 828 and 1027, inclusive, of the sequence shown in FIG. 1 (SEQ ID NO: 2). Although the numbering system of Jerlstr6m, P.G., et al., Molec. Microbiol. 5:843-849 (1991) has been utilized in FIG. 1, and elsewhere in 5 references to the sequences of the invention, it should be noted that the peptides, protein fragments, and nucleic acids of the invention also relate to the wild-type C3 protein sequence as disclosed in Hed6n, L-O., et al., Eur. J Immunol., 21:1481-1490 (1991). Although the sequences disclosed by Jerlstr6m et al. and Hed6n et al. 10 show only minor variation relative to one another, two amino acid differences are specifically noted as examples of variation encompassed by the present invention. While the Jerlstrbm et al. sequence depicted in FIG. 1 shows Lys in position 878, the corresponding amino acid in the sequence of Hed6n et al. is Glu. Also, Jerlstr6m et al. show three full repeats of the sequence PVDPKL (SEQ ID NO: 15 51), while Hed6n et al. show only two full repeats of the six residue sequence (the "third" repeat in the Hed6n et al. sequence having Ala instead of Val, corresponding to position 897 in the Jerlstr6m et al. sequence). Such variation may be due to DNA polymerase "stuttering" or slip-stranded synthesis when reading repeated sequences such as those described herein. The proteins and 20 nucleic acids of the present invention, therefore, include such minor variations as exists between the sequences disclosed by Jerlstr6m et al. and Hed6n et al. Mutation of a region of the CP protein located between about amino acid residues 163 and 176 of the wild-type CP sequence shown in FIG. 1 (SEQ ID NO:2) results in a protein fragment or peptide which has reduced or eliminated 25 IgA binding properties, but which retains enough of its tertiary structure to maintain the majority of its antigenicity (Examples 4 and 5). Amino acid substitutions or deletions in this region reduce or eliminate IgA binding while maintaining antigenicity of the protein. Thus, one may alter the amino acid sequence of the CP polypeptide so as to achieve a protein which does not bind 30 IgA. Appropriate amino acid substitutions which eliminate IgA binding include replacement of one or more residues with an amino acid having different WO 00/15760 PCT/US99/21643 -18 properties. For example, a strongly hydrophilic amino acid can be replaced with a strongly hydrophobic amino acid. Amino acids which can be grouped together include the aliphatic amino acids Ala, Val, Leu and Ile, the hydroxyl residues Ser and Thr, the acidic residues Asp and Glu, the amide residues Asn and Gln, the 5 basic residues Lys and Arg, and the aromatic residues Phe and Tyr. Thus, those of ordinary skill in the art will understand how to determine suitable amino acid substitutions or deletions in the region between about residues 163 and 176 in the CP3 protein in order to reduce or eliminate IgA binding. Further guidance concerning which amino acid changes are likely to have 10 a significant deleterious effect on a function can be found in Bowie, J.U., et al., "Deciphering the Message in Protein Sequences: Tolerance to Amino Acid Substitutions," Science 247:1306-1310 (1990). Thus, in another aspect, the invention also relates to a peptide or protein fragment as described above and which lack IgA binding, in particular, wherein 15 Y represents an amino acid sequence comprising A-X, X 2

X

3

X

4

X

5

X

6

X

7 X 8 X 9

X

10 X I X 12 -B, wherein A comprises amino acids 1-164 of the sequence shown in FIG. 1 (SEQ ID NO: 2) or an N-terminal fragment thereof, B represents a sequence starting from amino acid 177 and terminating at an amino acid bound to X, and X 1 - X12 are each selected independently from the group consisting of 20 Ala, Val, Leu, Ile, Pro, Met, Phe, Trp, a bond, and the wild type amino acid found at the corresponding position of the sequence shown in FIG. 1 (SEQ ID NO: 2), wherein the amino acid positions are numbered from the first amino acid of the native amino acid sequence encoding the protein, with the proviso that at least one of X, through X 12 , inclusive, is other than the wild type amino acid. 25 The invention also relates to a peptide or protein fragment as described above and which lack IgA binding, in particular, wherein Y represents an amino acid sequence comprising A-Xj X 2

X

3

X

4

X

5

X

6

X

7

X

8

X

9

X

10

X

1

X

12 -B. In a particularly preferred embodiment, amino acids X 7 and X 12 are Ala (SEQ ID NO: 3). In another preferred embodiment, amino acids X 4 and X,, are Pro (SEQ ID 30 NO: 4). In another preferred embodiment, amino acid X 7 is Thr and amino acid WO 00/15760 PCT/US99/21643 -19

X

1 2 is Leu (SEQ ID NO: 5). In a more preferred embodiment, amino acids Xs, X 7 ,

X

s , X 1 0 , X 1 and X 12 are each replaced with a bond (SEQ ID NO: 6). In preferred embodiments, the protein fragments and peptides of the invention are substantially pure. 5 IgA binding ability of CO may require dimerization of CP. Thus, even where the IgA binding region of CP is not mutated as described above, mutation of the region of CP which is believed to be required for dimerization can result in a form of CP that cannot bind IgA. Deletion of a portion of CP from residue 729 to the C-terminus of the sequence shown in FIG. 1 (SEQ ID NO: 2) 10 eliminates dimerization of CP. The results of experiments supporting this finding may be found in Table 1. Several fragments of CP were inserted into each of two different vectors. Where sequences shown in the table are preceded or followed by an outward facing bracket, this indicates that the CP sequence does not extend further on that end of the fragment, i.e., that the nucleotide sequence inserted into 15 the vector encodes only those amino acids shown and no more of the C[ sequence. Where sequences shown in the table are preceded or followed by ellipses, this indicates that the remainder of the CP sequence at that end of the fragment is also included in the vector. Nucleotide sequences encoding the peptides shown in the upper part of the table were inserted into either the vector 20 pTOPE or the vector pET1 7b. Both of these vectors allow expression of inserted fragments from the T7 promoter, and both produce fusion proteins containing a fragment of the 4)10 capsid protein N terminal to the amino acid sequence encoded by the insert. However, while pET17b encodes only 8 amino acids of the 10 protein, pTOPE encodes a 288 amino acid fragment of the 10 protein. 25 As shown in Table 1, certain fragments of CP produced from pET17b exhibit reduced IgA binding, while the same fragment produced by pTOPE is capable of binding IgA. The fragments tested lack the region of C[ predicted to be involved in dimerization, but do not contain any mutations in the putative IgA.

WO 00/15760 PCT/US99/21 643 ______-20 o 0 0 0 0 0 0 0 0 0 0 F- -- < < CC VC C C , V V .2 CCC CC VC CCC VC V V V < < < >- >. < < < z z - :o >. C CCC C C , 0 V C l V aY V Co Co CY CV COC) CC Co C VC C C CCC CY C Y C CC 8 CC CC Co Co a ) a l a a a) aY a .2~~ ~ 0 0 0 0 0 Co Co Co Co C ~> > o o C CoC Co C Co C W. w. 0 .2 .2 . z000 0 0 C C C . 2 . 2 . RHSTRR SHE (UE 6 WO 00/15760 PCT/US99/21643 -21 binding domain (note that the CP fragments inserted into vector pET24b, shown at the bottom of Table 1, contain the putative dimerization region but nonetheless exhibit reduced IgA binding due to mutations in the IgA binding domain, as described above). It is postulated that these CP fragments bind IgA when 5 produced from pTOPE because the 288 amino acid fragment of the 4 10 protein allows dimerization of the CP fragment. This may be due to the fact that the $10 capsid protein normally forms oligomers; the region responsible for oligomerization may thus allow dimerization of the inserted CP fragments, and thus IgA binding. Thus, the invention also relates to a peptide or protein fragment 10 having a mutation in the dimerization domain of Cp, wherein the mutant CP protein is incapable of binding IgA. Of course, in the interest of producing a non IgA binding a peptide or protein fragment retaining as much of the antigenicity of the wild type C3 protein as possible, dimerization of CP should not be interrupted. 15 In a preferred embodiment, at least one of amino acid residues 521-541 of the amino acid sequence shown in FIG. 1 (SEQ ID NO:2) is either (a) deleted or (b) altered, so that the protein is not cleaved in this region when CP is produced in E. coli. In a more preferred embodiment, at least one of amino acid residues 533-541 of the sequence shown in FIG. 1 (SEQ ID NO:2) is either (a) 20 deleted or (b) altered. In an even more preferred embodiment, at least one of amino acid residues 537 and 538 is either (a) deleted or (b) altered. Of course, one of ordinary skill will be able to determine other suitable amino acid substitutions by routine experimentation, and by reference to the article by von Heijne (Nucleic Acids Res. 14: 4683-4690 (1986)). 25 As the CP protein is, in its wild type state, membrane bound, it is possible to improve purification of the above-mentioned CP protein, mutants and/or fragments by eliminating the hydrophobic residues of the transmembrane domain of the CP protein (the transmembrane domain corresponds to residues 1095-1127 of the sequence shown in FIG. 1 (SEQ ID NO: 2)). This can be accomplished by 30 substitution of non-hydrophobic residues for the hydrophobic residues (residues 1108-1116 of the sequence shown in FIG. 1 (SEQ ID NO: 2)) or by deletion of WO 00/15760 PCT/US99/21643 -22 the hydrophobic residues. While purification of membrane-bound CP requires the use of detergent, a mutant CP which lacks the hydrophobic membrane spanning region can be purified without using detergent. Thus, the invention also relates to a peptide or protein fragment wherein the nine hydrophobic residues 5 making up the transmembrane domain are deleted or replaced by non hydrophobic amino acids. Production of CP protein from E. coli can be problematic because the protein is cleaved at a specific region, presumably by an E. coli signal peptidase. This cleavage results in a truncated protein, which may not be ideal for a vaccine, 10 as it may lack antigenic epitopes of the wild-type CP protein. The cleavage site has been predicted by sequence analysis and by matrix assisted laser desorption initiated time of flight (MALDI-TOF) mass spectrometry (von Heijne, Nucleic Acids Res. 14: 4683-4690 (1986)). The cleavage site is between amino acid residues 538 and 539 (after alanine and before glutamine) of the amino acid 15 sequence shown in FIG. 1 (SEQ ID NO:2). The signal peptidase recognition site is located within a 20 amino acid stretch located between residues 521 and 541 of the amino acid sequence shown in FIG. 1 (SEQ ID NO:2). Therefore, by deleting this region, the protein can successfully be produced in E. coli. Furthermore, as signal peptidases have very strict sequence specificity, alteration 20 of the signal peptidase recognition sequence, including even a single, conservative amino acid substitution in this region, may eliminate cleavage of the protein by E. coli. The recognition sequence required for cleavage by this signal peptidase is believed to be GluLeuIleLysSerAlaGlnGlnGlu (SEQ ID NO: 11), corresponding to amino acid residues 533-541 of the sequence shown in FIG. 1 25 (SEQ ID NO:2). Alteration of either the serine or the alanine residue of this sequence by either deletion or non-conservative substitution is expected to eliminate cleavage by the signal peptidase. Of course, ideally, the mutagenesis of CP3 sequence will be kept to a minimum so as to retain the tertiary structure of the wild-type antigen for the purposes of eliciting an immunogenic response.

WO 00/15760 PCT/US99/21643 -23 The invention also relates to polynucleotide molecules encoding the peptides and protein fragments of the invention, vectors comprising those polynucleotide molecules, and host cells transformed therewith. The invention also relates to the expression of protein fragments and 5 peptides of the invention in a cellular host. Prokaryotic hosts that may be used for cloning and expressing the proteins or peptides of the invention are well known in the art. Vectors which replicate in such host cells are also well known. Preferred prokaryotic hosts include, but are not limited to, bacteria of the genus Escherichia, Bacillus, Streptomyces, Pseudomonas, Salmonella, Serratia, 10 Xanthomonas, etc. Two such prokaryotic hosts are E. coli DH10B and DH5aF'IQ (available from LTI, Gaithersburg, MD). The most preferred host for cloning and expressing the protein fragments and polypeptides of the invention is E. coli BL21 (Novagen, Inc., Madison, WI), which is lysogenic for DE3 phage. The present invention also relates to vectors which include the isolated 15 DNA molecules coding for the peptides and protein fragments of the present invention, host cells which are genetically engineered with the recombinant vectors, and the production of the proteins or polypeptides of the invention by recombinant techniques. Host cells can be genetically engineered to incorporate the nucleic acid 20 molecules and express the protein fragments or peptides of the present invention. For instance, recombinant constructs may be introduced into host cells using well known techniques of infection, transduction, transfection, and transformation. The polynucleotides may be introduced alone or with other polynucleotides. Such other polynucleotides may be introduced independently, co-introduced, or 25 introduced joined to the polynucleotides of the invention. Thus, for instance, the polynucleotides may be joined to a vector containing a selectable marker for propagation in a host. The vector construct may be introduced into host cells by the aforementioned techniques. Generally, a plasmid vector is introduced as DNA in a precipitate, such as a calcium 30 phosphate precipitate, or in a complex with a charged lipid. Electroporation also may be used to introduce polynucleotides into a host. If the vector is a virus, it WO 00/15760 PCT/US99/21643 -24 may be packaged in vitro or introduced into a packaging cell and the packaged virus may be transduced into cells. A wide variety of techniques suitable for making polynucleotides and for introducing polynucleotides into cells in accordance with this aspect of the invention are well known and routine to those 5 of skill in the art. Such techniques are reviewed at length in Molecular Cloning: A Laboratory Manual, second edition, Sambrook et al., eds., Cold Spring Harbor Laboratory (1989), which is illustrative of the many laboratory manuals that detail these techniques. In accordance with this aspect of the invention the vector may be, for 10 example, a plasmid vector, a single or double-stranded phage vector, or a single or double-stranded RNA or DNA viral vector. Such vectors may be introduced into cells as polynucleotides, preferably DNA, by well known techniques for introducing DNA and RNA into cells. The vectors, in the case of phage and viral vectors, also may be and preferably are introduced into cells as packaged or 15 encapsulated virus by well known techniques for infection and transduction. Viral vectors may be replication competent or replication defective. In the latter case, viral propagation generally will occur only in complementing host cells. Preferred among vectors, in certain respects, are those for expression of protein fragments or peptides of the present invention. Generally, such vectors 20 comprise cis-acting control regions effective for expression in a host operatively linked to the polynucleotide to be expressed. Appropriate trans-acting factors either are supplied by the host, supplied by a complementing vector, or supplied by the vector itself upon introduction into the host. In certain preferred embodiments in this regard, the vectors provide for 25 specific expression. Such specific expression may be inducible expression or expression only in certain types of cells or both inducible and cell-specific. Particularly preferred among inducible vectors are vectors that can be induced for expression by environmental factors that are easy to manipulate, such as temperature and nutrient additives. A variety of vectors suitable to this aspect of 30 the invention, including constitutive and inducible expression vectors for use in WO 00/15760 PCT/US99/21643 -25 prokaryotic and eukaryotic hosts, are well known and employed routinely by those of skill in the art (U.S. Patent No. 5,464,758). The engineered host cells can be cultured in conventional nutrient media, which may be modified as appropriate for, inter alia, activating promoters, 5 selecting transformants or amplifying genes. Culture conditions, such as temperature, pH, and the like, previously used with the host cell selected for expression generally will be suitable for expression of protein fragments or peptides of the present invention as will be apparent to those of skill in the art. A great variety of expression vectors can be used to express a protein 10 fragment or peptide of the invention. Such vectors include chromosomal, episomal and virus-derived vectors, e.g., vectors derived from bacterial plasmids, from bacteriophage, from yeast episomes, from yeast chromosomal elements, from viruses such as baculoviruses, papova viruses, such as SV40, vaccinia viruses, adenoviruses, fowl pox viruses, pseudorabies viruses and retroviruses, 15 and from vectors derived from combinations thereof, such as those derived from plasmid and bacteriophage genetic elements, such as cosmids and phagemids, all may be used for expression in accordance with this aspect of the present invention. Generally, any vector suitable to maintain or propagate, polynucleotides, or to express a polypeptide or protein, in a host may be used for 20 expression in this regard. The appropriate DNA molecule may be inserted into the vector by any of a variety of well known and routine techniques. In general, a DNA molecule for expression is joined to an expression vector by cleaving the DNA sequence and the expression vector with one or more restriction endonucleases and thenjoining 25 the restriction fragments together using T4 DNA ligase. Procedures for restriction and ligation that can be used to this end are well known and routine to those of skill in the art. Suitable procedures in this regard, and for constructing expression vectors using alternative techniques, which also are well known and routine to those skill, are set forth in great detail in Sambrook et al. cited above. 30 The DNA molecule inserted in the expression vector is operatively linked to appropriate expression control sequence(s), including, for instance, a promoter WO 00/15760 PCT/US99/21643 -26 to direct mRNA transcription. Representatives of such promoters include the phage lambda PL promoter, the E. coli lac, trp and tac promoters, the SV40 early and late promoters, and promoters of retroviral LTRs, to name just a few of the well known promoters. It will be understood that numerous promoters not 5 mentioned are suitable for use in this aspect of the invention are well known and readily may be employed by those of skill in the art in the manner illustrated by the discussion and the examples herein. In general, expression constructs will contain sites for transcription initiation and termination, and, in the transcribed region, a ribosome binding site 10 for translation. The coding portion of the mature transcripts expressed by the constructs will include a translation initiating AUG at the beginning and a termination codon appropriately positioned at the end of the polypeptide to be translated. In addition, the constructs may contain control regions that regulate as 15 well as engender expression. Generally, in accordance with many commonly practiced procedures, such regions will operate by controlling transcription, such as repressor binding sites and enhancers, among others. Vectors for propagation and expression generally will include selectable markers. Such markers also may be suitable for amplification or the vectors may 20 contain additional markers for this purpose. In this regard, the expression vectors preferably contain one or more selectable marker genes to provide a phenotypic trait for selection of transformed host cells. Preferred markers include dihydrofolate reductase or neomycin resistance for eukaryotic cell culture, and tetracycline or ampicillin resistance genes for culturing E. coli and other bacteria. 25 The vector containing the appropriate DNA sequence as described elsewhere herein, as well as an appropriate promoter, and other appropriate control sequences, may be introduced into an appropriate host using a variety of well known techniques suitable to expression therein of a desired polypeptide or protein. Representative examples of appropriate hosts include bacterial cells, 30 such as E. coli, Streptomyces, and Salmonella typhimurium cells. Hosts for a great variety of expression constructs are well known, and those of skill will be WO 00/15760 PCT/US99/21643 -27 enabled by the present disclosure readily to select a host for expressing a protein fragment or peptide in accordance with this aspect of the present invention. More particularly, the present invention also includes recombinant constructs, such as expression constructs, comprising one or more of the 5 sequences described above. The constructs comprise a vector, such as a plasmid or viral vector, into which such a sequence of the invention has been inserted. The sequence may be inserted in a forward or reverse orientation. In certain preferred embodiments in this regard, the construct further comprises regulatory sequences, including, for example, a promoter, operably linked to the sequence. 10 Large numbers of suitable vectors and promoters are known to those of skill in the art, and there are many commercially available vectors suitable for use in the present invention. As the invention also concerns the construction of a polypeptide or protein having a reduced or eliminated ability to bind human IgA, the invention thus 15 relates to using in vitro mutagenesis methods to generate the protein fragments or peptides of the invention. A number of in vitro mutagenesis methods are well known to those of skill in the art; several are provided here as examples. One such method introduces deletions or insertions into a polynucleotide molecule inserted into a plasmid by either partially or completely digesting the 20 plasmid with an appropriate restriction enzyme, and then ligating the ends to again generate a plasmid. Very short deletions can be made by first cutting a plasmid at a restriction site, and then subjecting the linear DNA to controlled nuclease digestion to remove small groups of bases at each end. Precise insertions may also be made by ligating double stranded oligonucleotide linkers 25 to a plasmid cut at a single restriction site. Chemical methods can also be used to introduce mutations to a single stranded polynucleotide molecule. For example, single base pair changes at cytosine residues can be created using chemicals such as bisulfite, which deaminates cytosine to uracil, thus converting GC base pairs to AT base pairs. 30 Preferably, oligonucleotide directed mutagenesis will be used so that all possible classes of base pair changes at any determined site along a DNA molecule can be WO 00/15760 PCT/US99/21643 -28 made. In general, this technique involves annealing an oligonucleotide complementary (except for one or more mismatches) to a single stranded nucleotide sequence of interest. The mismatched oligonucleotide is then extended by DNA polymerase, generating a double stranded DNA molecule 5 which contains the desired change in sequence on one strand. The changes in sequence can of course result in the deletion, substitution, or insertion of an amino acid if the change is made in the coding region of a gene. The double stranded polynucleotide can then be inserted into an appropriate expression vector, and a mutant polypeptide can thus be produced. The above-described 10 oligonucleotide directed mutagenesis can of course be carried out via PCR. An example of such a system is the Ex-Site® PCR site-directed mutagenesis technique (Stratagene, CA) used in Example 4. Using the Ex-Site® PCR site-directed mutagenesis technique, several different oligonucleotides were made to induce different changes in the DNA 15 sequence in the region of interest. In one particular example, overlapping primers were obtained, wherein both primers contained the sequence required to change lysine to alanine at amino acids 170 and 175 in the sequence shown in FIG. 1 (SEQ ID NO: 2) (Table 1). The forward primer, designated CP 613, had the sequence (SEQ ID NO: 6) 5'-GTT GAA GCA ATG GCA GAG CAA GCG GGA 20 ATC ACA AAT GAA G-3' and the reverse primer, designated CP 642R had the sequence (SEQ ID NO: 7) 5'-GAT TCC CGC TTG CTC TGC CAT TGC TTC AAC TTG ACT TTT TTG-3' (the substitutions are noted in BOLD). These oligonucleotides were combined with pNV222 template, which consists of the CP gene inserted into the pSP76 vector. PCR was performed, and the products were 25 ligated and introduced into E. coli strain DH5a, thus generating clones containing the mutant CP gene. The following vectors, which are commercially available, may be used in the practice of the invention. Among vectors preferred for use in bacteria are pQE70, pQE60 and pQE-9, available from Qiagen; pBS vectors, Phagescript 30 vectors, Bluescript® vectors, pNH8A, pNH16a, pNH18A, pNH46A, available from Stratagene; ptrc99a, pKK223-3, pKK233-3, pDR540, pRIT5 available from WO 00/15760 PCT/US99/21643 -29 Pharmacia; pUC 18, pUC 19 and pPROEX-1, available from LTI, and pTOPE, pET17b, and pET24a (Novagen, Inc., Madison, WI). These vectors are listed solely by way of illustration of the many commercially available and well known vectors that are available to those of skill in the art for use in accordance with this 5 aspect of the present invention. It will be appreciated that any other plasmid or vector suitable for, for example, introduction, maintenance, propagation, or expression of a polynucleotide, protein fragment, or peptide of the invention in a host may be used in this aspect of the invention. Promoter regions can be selected from any desired gene using vectors that 10 contain a reporter transcription unit lacking a promoter region, such as a chloramphenicol acetyl transferase ("CAT") transcription unit, downstream of a restriction site or sites for introducing a candidate promoter fragment, i.e., a fragment that may contain a promoter. As is well known, introduction into the vector of a promoter-containing fragment at the restriction site upstream of the 15 CAT gene engenders production of CAT activity, which can be detected by standard CAT assays. Vectors suitable to this end are well known and readily available. Two such vectors are pKK232-8 and pCM7. Thus, promoters for expression of polynucleotides of the present invention include not only well known and readily available promoters, but also promoters that readily may be 20 obtained by the foregoing technique, using a reporter gene. Among known bacterial promoters suitable for expression of polynucleotides and polypeptides in accordance with the present invention are the E. coli lacI and lacZ and promoters, the T3 and T7 promoters, the gpt promoter, the lambda PR, PL promoters, and the trp promoter. 25 Selection of appropriate vectors and promoters for expression in a host cell is a well known procedure and the requisite techniques for expression vector construction, introduction of the vector into the host and expression in the host are routine skills in the art. The present invention also relates to host cells containing the constructs 30 discussed above. The host cell can be a prokaryotic cell, such as a bacterial cell.

WO 00/15760 PCT/US99/21643 -30 Constructs in host cells can be used in a conventional manner to produce the gene product encoded by the recombinant DNA sequence. Alternatively, the protein fragments or peptides of the invention can be synthetically produced by conventional peptide synthesizers. 5 Following transformation of a suitable host strain and growth of the host strain to an appropriate cell density, where the selected promoter is inducible, it is induced by appropriate means, e.g., temperature shift or exposure to chemical inducer, and cells are cultured for an additional period. Cells typically are then harvested by centrifugation, disrupted by physical 10 or chemical means, and the resulting crude extract retained for further purification. Microbial cells employed in expression of proteins can be disrupted by any convenient method, including freeze-thaw cycling, sonication, mechanical disruption, or use of cell lysing agents; such methods are well known to those 15 skilled in the art. Preferably, the peptides may be prepared using well-known methods of solid phase peptide synthesis. By the term "peptide" is intended those having no more that 100 amino acids. Proteins consist of more that 100 amino acids. Preferably, the peptides comprise 8-50 amino acids. 20 In another aspect, the invention relates to a method of isolating and purifying the proteins and polypeptides of the invention. A method of purifiying the proteins and polypeptides of the invention has been discovered that yields large quantities of substantially pure protein. By "substantially pure" it is intended that the protein or polypeptide obtained by the process of the invention 25 is at least 85% pure with negligible contamination by nucleic acids and polysaccharides. One advantage of the method of the invention is that it is scaleable. That is, the method is readily adaptable for purification of large quantities of the protein. The columns used in purification will accept a large volume/concentration of sample, use a dilution step rather than a concentration 30 step to apply sample from one column to the next, and both columns use the same buffer as a mobile phase. Moreover, this purification method is applicable to the WO 00/15760 PCT/US99/21643 -31 purification of Cp protein mutants and/or fragments obtained both from recombinant E. coli and native streptococci. Additionally, the use of the protease inhibitors and detergent in purification permits purification using ion-exchange and heparin columns with 5 minimized degradation of products. Finally, product yields are higher than those obtained by the previously employed gel-filtration column chromatography procedure (Russell-Jones et al., J. Exp. Med 160:1467-1475 (1984); Madoffet al., Infect. Immun. 59(1):204-210 (1991)), since larger quantities of sample can be processed at one time. This large scale process also minimizes handling time 10 involved in purification. The invention relates to a process for obtaining a substantially pure C3 protein, or fragment and/or mutant thereof, comprising: (a) obtaining the CP protein in cell extracts; (b) subjecting the CP protein to ion exchange chromatography and 15 collecting the CP protein-containing fractions; (c) pooling and diluting the CP protein-containing fractions; and (d) subjecting the diluted CP protein-containing fractions to ligand affinity or a gel filtration chromatography and collecting the fractions; whereby substantially pure C[ protein or fragment and/or mutant thereof is 20 obtained. In a preferred embodiment, an anion-exchange medium is employed. More preferably, the anion medium has a functional group selected from the group consisting oftertiary amine, quaternary ammonium, quaternary alkylamine, quaternary alkylalkanolamine, diethylaminoethyl, diethyl-(2 25 hydroxypropyl)aminoethyl, trimethylaminohydroxypropyl, trimethylbenzylammonium, dimethylethanolbenzylammonium, and dimethylethanolamine. Still more preferably, the functional group is a quaternary ammonium group. Most preferably, the quaternary ammonium group is a trimethylaminomethyl group. 30 In a preferred embodiment, the ion-exchange medium comprises a functional group described above immobilized on a solid support material, e.g., WO 00/15760 PCT/US99/21643 -32 selected from the group consisting ofagarose, dextran, cellulose, and polystyrene. More preferably, the solid support material is in bead form. Most preferably, the solid support material is cross-linked agarose. In a preferred embodiment, a ligand-affinity medium is employed. 5 Preferably, the ligand of the ligand-affinity medium is a heparin-like molecule. More preferably, the heparin-like molecule is a glycosaminoglycan. More preferably the heparin-like molecule is selected from the group consisting of chondroitin sulfate, dermatan sulfate, keratan sulfate, and hyaluronate. Most preferably, the heparin-like molecule is heparin. 10 In a preferred embodiment, the ligand-affinity medium comprises a ligand immobilized on a solid support material, e.g., selected from the group consisting of agarose, dextran, cellulose, and polystyrene. More preferably, the solid support material is in bead form. Most preferably, the solid support material is cross linked agarose. 15 In another embodiment, the fractions containing the CP protein, or fragment and/or mutant thereof, are subjected to gel filtration chromatography. More preferably, the gel filtration medium is a crosslinked allyl dextran/N,N' methylenebisacrylamide copolymer. In a preferred embodiment, the C[ protein, or fragment and/or mutant 20 thereof, is subjected to ion-exchange or ligand affinity chromatography in a buffer containing zwitterionic detergent. Preferably, the zwitterionic detergent is CHAPS (3-((3-cholamidopropyl)dimethyl-ammonio)-1-propane sulfonate). Also more preferably, the zwitterionic detergent is Empigen® BB (Alkyl dimethylamine betaines, available from Albright & Wilson Americas Inc., Glen 25 Allen, VA). Most preferably, the zwitterionic detergent is selected from the group consisting of n-octyl, n-decyl, n-dodeyl, n-tetradecyl, and n-hexadecyl derivatives ofN,N-dimethyl-3-ammonio-1 -propanesulfonate (Zwittergent® from Calbiochem, La Jolla, CA). More preferably, the detergent is at a concentration of about 1% to about 10%. Most preferably, the detergent concentration is 30 about 5%.

WO 00/15760 PCT/US99/21643 -33 In a preferred embodiment, the CO protein, fragment and/or mutant thereof, is subjected to chromatography on the first medium in a buffer containing a protease inhibitor. Preferably, the protease inhibitor is a serine protease inhibitor. More preferably, the protease inhibitor is selected from the group 5 consisting of DFP, PMSF, and APMSF. Most preferably, the protease inhibitor is 4-(2-aminoethyl)benzenesulfonylfluoride-HCl (PEFABLOC SC or PEFABLOC PLUS from Boehringer Mannheim Corporation, Indianapolis, IN, which may also contain tyramine). In a preferred embodiment, the eluate containing the CP protein, or a 10 fragment and/or mutant thereof, from the first chromatographic medium is diluted approximately three-fold with buffer containing about 10% to about 20% zwitterionic detergent before ligand-affinity or gel filtration chromatography on the second medium. In a preferred embodiment, the CP protein, or a fragment and/or mutant 15 thereof, is eluted from the first or second chromatography medium by applying an eluant comprising a salt gradient. More preferably, the salt gradient is a step gradient comprising steps of about 0.0 M, about 0.05 M, 0.075 M, and 0.09 M of a salt. Most preferably, the salt gradient is a linear gradient from about 0.0 M to about 0.1 M of a salt. More preferably, the salt is NaCl or KC1. Most preferably, 20 the salt is NaC1. In a preferred embodiment, the process for obtaining a substantially pure CP3 protein, or fragment or mutant thereof, is carried out at pH from about 7.0 to about 8.0. More preferably, the process is carried out at pH from about 7.6 to about 7.8. 25 In a preferred embodiment, the eluant further comprises from about 0.1% to about 1.0% of a zwitterionic detergent. More preferably, the eluant comprises about 0.5% of a zwitterionic detergent. In another preferred embodiment, CP protein, or a fragment and/or mutant thereof, is obtained from bacterial cells which are transfected with nucleotide 30 sequences encoding the CP protein or a fragment and/or mutant thereof, wherein the cells overexpress the CP protein or a fragment and/or mutant thereof. More WO 00/15760 PCT/US99/21643 -34 preferably, the bacterial cells are E. coli cells. Most preferably, the CP protein or a fragment or mutant thereof is obtained by: (a) disrupting the cells; (b) precipitating non-proteinaceous material from the cells by adding 5 ethanol/CaCl 2 to a concentration of about 20% (v/v) ethanol/0.1 M CaCl 2 ; (c) removing the precipitated non-proteinaceous material to give a solution; (d) precipitating protein from the solution by adding ethanol to a concentration of about 80% (v/v) and collecting the precipitated protein; and 10 (e) resuspending the precipitated protein in a buffer solution containing from about 3% to about 7% of a zwitterionic detergent. In another preferred embodiment, CP3 protein is obtained from bacterial cells which naturally produce the CP protein. More preferably, the bacterial cells are Streptococcus agalactiae cells. Most preferably, the CP protein is obtained 15 by: (a) boiling the bacterial cells in a buffer containing from about 3% to about 7% of a zwitterionic detergent to give a solution; (b) cooling the solution in an ice bath; (c) adding a cold solution of CaCl 2 /ethanol to give a concentration of 20 about 20% ethanol(v/v)/about 0.1M CaCl 2 ; (d) removing the precipitated non-proteinaceous material to give a solution; (e) precipitating protein from the solution by adding ethanol to a concentration of about 80% (v/v) and collecting the precipitated protein; and 25 (f) resuspending the precipitated protein in a buffer solution containing from about 3% to about 7% of a zwitterionic detergent. The invention also relates to a vaccine comprising a protein fragment or peptide of the invention, together with a pharmaceutically acceptable carrier. In one aspect, the invention relates to protein fragments or peptides that elicit 30 antibodies which are bactericidal to Gram positive bacteria with complement alone, i.e., the activity is not dependent on an opsonophagocytotic mechanism.

WO 00/15760 PCT/US99/21643 -35 These protein fragments and peptides of the invention are particularly well-suited to the production of a vaccine which is useful in patients with impaired cellular immunity, e.g., chemotherapy patients and leukemia patients. As the proline-rich region of streptococci described herein is characteristic 5 of Gram positive bacteria, it is expected that the vaccines of the invention will be useful for raising an immune response against other Gram positive bacteria, including, but not limited to Bacillus acidocaldarius, Bacillus acidoterrestris, Bacillus alcalophilus, Bacillus alvei, Bacillus amyloliquefaciens, Bacillus amylolyticus, Bacillus anthracis, Bacillus azotofixans, Bacillus azotoformans, 10 Bacillus badius, Bacillus benzoevorans, Bacillus brevis, Bacillus cereus, Bacillus circulans, Bacillus coagulans, Bacillus cycloheptanicus, Bacillus fastidiosus, Bacillus firmus, Bacillus flexus, Bacillus fusiformis, Bacillus globisporus, Bacillus glucanolyticus, Bacillus gordonae, Bacillus halodenitrificans, Bacillus insolitus, Bacillus kaustophilus, Bacillus larvae, Bacillus laterosporus, Bacillus 15 lautus, Bacillus lentimorbus, Bacillus lentus, Bacillus licheniformis, Bacillus macerans, Bacillus macquariensis, Bacillus marinus, Bacillus megaterium, Bacillus mycoides, Bacillus pabuli, Bacillus pantothenticus, Bacillus pasteurii, Bacillus polymyxa, Bacillus popilliae, Bacillus psychrophilus, Bacillus psychrosaccharolyticus, Bacillus pumilus, Bacillus schlegelii, Bacillus simplex, 20 Bacillus sphaericus, Bacillus stearothermophilus, Bacillus subtilis, Bacillus thermoglucosidasius, Bacillus thermoleovorans, Bacillus thuringiensis, Bacillus validus, Clostridium absonum, Clostridium aceticum, Clostridium acetobutylicum, Clostridium acidiurici, Clostridium aerotolerans, Clostridium aminovalericum, Clostridium argentinense, Clostridium aurantibutyricum, 25 Clostridium barati, Clostridium barkeri, Clostridium beijerinckii, Clostridium bifermentans, Clostridium botulinum, Clostridium butyricum, Clostridium cadaveris, Clostridium carnis, Clostridium celatum, Clostridium cellobioparum, Clostridium cellulolyticum, Clostridium cellulovorans, Clostridium chauvoei, Clostridium clostridiiforme, Clostridium coccoides, Clostridium cochlearium, 30 Clostridium cocleatum, Clostridium colinum, Clostridium collagenovorans, Clostridium cylindrosporum, Clostridium difficile, Clostridium disporicum, WO 00/15760 PCT/US99/21643 -36 Clostridium durum, Clostridium fallax, Clostridium felsineum, Clostridium fervidus, Clostridium formicoaceticum, Clostridium ghoni, Clostridium glycolicum, Clostridium haemolyticum, Clostridium hastiforme, Clostridium histolyticum, Clostridium indolis, Clostridium innocuum, Clostridium intestinalis, 5 Clostridium irregularis, Clostridium kluyveri, Clostridium lentoputrescens, Clostridium leptum, Clostridium limosum, Clostridium lituseburense, Clostridium magnum, Clostridium malenominatum, Clostridium mangenotii, Clostridium methylpentosum, Clostridium novyi, Clostridium oceanicum, Clostridium orbiscindens, Clostridium oroticum, Clostridium papyrosolvens, Clostridium 10 paraputrificum, Clostridiumpasteurianum, Clostridiumperfringens, Clostridium polysaccharolyticum, Clostridium populeti, Clostridium propionicum, Clostridiumproteolyticum, Clostridiumpurinolyticum, Clostridiumputrefaciens, Clostridium putrificum, Clostridium quericolum, Clostridium ramosum, Clostridium rectum, Clostridium roseum, Clostridium saccharolyticum, 15 Clostridium sardiniensis, Clostridium sartagoformum, Clostridium scatologenes, Clostridium scindens, Clostridium septicum, Clostridium sordellii, Clostridium sphenoides, Clostridium spiroforme, Clostridium sporogenes, Clostridium sporosphaeroides, Clostridium stercorarium, Clostridium sticklandii, Clostridium subterminale, Clostridium symbiosum, Clostridium tertium, 20 Clostridium tetani, Clostridium tetanomorphum, Clostridium thermoaceticum, Clostridium thermoautotrophicum, Clostridium thermocellum, Clostridium thermohydrosulfuricum, Clostridium thermolacticum, Clostridium thermosaccharolyticum, Clostridium thermosulfurogenes, Clostridium tyrobutyricum, Clostridium villosum, Clostridium xylanolyticum, 25 Corynebacterium accolens, Corynebacterium ammoniagenes, Corynebacterium amycolatum, Corynebacterium bovis, Corynebacterium callunae, Corynebacterium cystitidis, Corynebacterium diphtheriae, Corynebacterium flavescens, Corynebacterium glutamicum, Corynebacterium hoagii, Corynebacterium jeikeium, Corynebacterium kutscheri, Corynebacterium 30 liquefaciens, Corynebacterium matruchotii, Corynebacterium minutissimum, Corynebacterium mycetoides, Corynebacterium pilosum, Corynebacterium WO 00/15760 PCT/US99/21643 -37 pseudodiphtheriticum, Corynebacterium pseudotuberculosis, Corynebacterium renale, Corynebacterium striatum, Corynebacterium variabilis, Corynebacterium vitarumen, Corynebacterium xerosis, Erysipelothrix rhusiopathiae, Erysipelothrix tonsillarum, Lactobacillus acetotolerans, Lactobacillus 5 acidophilus, Lactobacillus alimentarius, Lactobacillus amylovorus, Lactobacillus animalis, Lactobacillus aviarius, Lactobacillus bifermentans, Lactobacillus brevis, Lactobacillus buchneri, Lactobacillus casei, Lactobacillus catenaformis, Lactobacillus collinoides, Lactobacillus confusus, Lactobacillus coryniformis, Lactobacillus crispatus, Lactobacillus curvatus, Lactobacillus delbrueckii, 10 Lactobacillus farciminis, Lactobacillusfermentum, Lactobacillusfructivorans, Lactobacillus fructosus, Lactobacillus gasseri, Lactobacillus halotolerans, Lactobacillus hamsteri, Lactobacillus helveticus, Lactobacillus hilgardii, Lactobacillus homohiochii, Lactobacillus intestinalis, Lactobacillus jensenii, Lactobacillus kefir, Lactobacillus kefiranofaciens, Lactobacillus malefermentans, 15 Lactobacillus mali, Lactobacillus maltaromicus, Lactobacillus minor, Lactobacillus minutus, Lactobacillus murinus, Lactobacillus oris, Lactobacillus parabuchneri, Lactobacillus paracasei , Lactobacillus pentosus, Lactobacillus plantarum, Lactobacillus reuteri, Lactobacillus ruminis, Lactobacillus sake, Lactobacillus salivarius , Lactobacillus sanfrancisco, Lactobacillus suebicus, 20 Lactobacillus vaccinostercus, Lactobacillus vaginalis, Lactobacillus viridescens, Lactobacillus vitulinus, Leuconostoc amelibiosum, Leuconostoc carnosum, Leuconostoc citreum, Leuconostoc gelidum, Leuconostoc lactis, Leuconostoc mesenteroides, Leuconostoc oenos, Leuconostoc paramesenteroides, Leuconostoc pseudomesenteroides, Listeria grayi, Listeria innocua, Listeria ivanovii, Listeria 25 monocytogenes, Listeria murrayi, Listeria seeligeri, Listeria welshimeri, Micrococcus agilis, Micrococcus halobius, Micrococcus kristinae, Micrococcus luteus, Micrococcus lylae, Micrococcus nishinomiyaensis, Micrococcus roseus, Micrococcus sedentarius, Micrococcus varians, Peptostreptococcus anaerobius, Peptostreptococcus asaccharolyticus, Peptostreptococcus heliotrinreducens, 30 Peptostreptococcus hydrogenalis, Peptostreptococcus indolicus, Peptostreptococcus magnus, Peptostreptococcus micros, Peptostreptococcus WO 00/15760 PCT/US99/21643 -38 prevotii, Peptostreptococcus productus, Peptostreptococcus tetradius, Propionibacterium acidipropionici, Propionibacterium acnes, Propionibacterium avidum, Propionibacterium freudenreichii, Propionibacterium granulosum, Propionibacterium jensenii, Propionibacterium lymphophilum, 5 Propionibacterium propionicum, Propionibacterium thoenii, Sebaldella termitidus, Staphylococcus arlettae, Staphylococcus aureus, Staphylococcus auricularis, Staphylococcus capitis, Staphylococcus caprae, Staphylococcus caseolyticus, Staphylococcus chromogenes, Staphylococcus cohnii, Staphylococcus delphini, Staphylococcus epidermidis, Staphylococcus equorum, 10 Staphylococcus felis, Staphylococcus gallinarum, Staphylococcus haemolyticus, Staphylococcus hominis, Staphylococcus hyicus, Staphylococcus intermedius, Staphylococcus kloosii, Staphylococcus lentus, Staphylococcus lugdunensis, Staphylococcus saccharolyticus, Staphylococcus saprophyticus, Staphylococcus schleiferi, Staphylococcus sciuri, Staphylococcus simulans, Staphylococcus 15 warneri, Staphylococcus xylosus, Streptococcus adjacens, Streptococcus agalactiae, Streptococcus alactolyticus, Streptococcus anginosus, Streptococcus bovis, Streptococcus canis, Streptococcus cricetus, Streptococcus defectivus, Streptococcus downei, Streptococcus dysgalactiae, Streptococcus equi, Streptococcus equinus, Streptococcus ferus, Streptococcus gordonii, 20 Streptococcus hansenii, Streptococcus hyointestinalis, Streptococcus iniae, Streptococcus intestinalis, Streptococcus macacae, Streptococcus mitis, Streptococcus mutans, Streptococcus oralis, Streptococcus parasanguis, Streptococcus parvulus, Streptococcus pleomorphus, Streptococcus pneumoniae, Streptococcus porcinus, Streptococcus pyogenes, Streptococcus rattus, 25 Streptococcus sanguis, Streptococcus sobrinus, Streptococcus suis, Streptococcus thermophilus, Streptococcus uberis, and Streptococcus vestibularis. In a preferred embodiment, the protein fragment or peptide is conjugated to a polysaccharide. The conjugates of the invention may be formed by reacting the reducing end groups of the polysaccharide to primary amino groups (for 30 example, of lysine residues) of the protein by reductive amination. The polysaccharide may be conjugated to any or all of the primary amino groups of WO 00/15760 PCT/US99/21643 -39 the protein. The reducing groups may be formed by selective hydrolysis or specific oxidative cleavage, or a combination of both. Preferably, the protein fragment or peptide is conjugated to the polysaccharide by the method of Jennings et al., U.S. Patent No. 4,356,170, which involves controlled oxidation 5 of the polysaccharide with periodate followed by reductive amination with the protein of the invention. In a preferred embodiment, the polysaccharide is one of the Group B streptococcal capsular polysaccharides selected from types Ia, II, III, and V. See Baker, C.J. and Kasper, D.L., Rev. Inf Dis. 7:458-467 (1985); Baker, C.J., etal., 10 N. Engl. J. Med 319:1180-1185 (1988); Baker, C.J., et al., New Engl. J. Med 322:1857-1860 (1990). The vaccine may also be a combination vaccine comprising one or more of the protein fragment or peptide-polysaccharide conjugates selected from the group consisting of the protein fragment or peptide of the invention conjugated to Group B capsular polysaccharide type Ia (protein 15 Ia); conjugated to Group B capsular polysaccharide type II (protein-II); conjugated to Group B capsular polysaccharide type III (protein-III); and conjugated to Group B capsular polysaccharide type V (protein-V). Most preferably, the vaccine is a combination vaccine comprising protein-Ia, protein-II, protein-III, and protein-V. Such a combination vaccine will elicit antibodies to 20 Group B streptoccoci of Types Ia, II, III, V, and Ib (as Type Ib Group B streptococci also express Cp). Furthermore, the immune response to the polysaccharides of the combination vaccine will be a T-dependent response. The vaccine of the present invention may comprise a conjugate comprising a peptide or protein fragment of the invention covalently conjugated 25 to a capsular polysaccharide. Alternatively, the peptide or protein fragment may be conjugated to or complexed with a porin protein, proteosome, or other carrier protein. See U.S. Patent No. 5,439,808, which describes methods for the preparation of outer membrane Group B porin from N. meningitidis, and WO95/03069, which describes the preparation ofthe P2 porin from H influenzae. 30 The vaccine of the present invention comprises one or more of the protein fragment or peptide vaccines or conjugate vaccines in amounts effective WO 00/15760 PCT/US99/21643 -40 depending on the route of administration. Although subcutaneous or intramuscular routes of administration are preferred, the vaccine of the present invention can also be administered by an intraperitoneal or intravenous route. One skilled in the art will appreciate that the amounts to be administered for any 5 particular treatment protocol can be readily determined without undue experimentation. With respect to each conjugate, suitable amounts are expected to fall within the range of 2 micrograms of the protein fragment or peptide per kg body weight to 100 micrograms per kg body weight. In a preferred embodiment, the vaccine comprises about 2 Vg of the protein fragment or peptide or an 10 equivalent amount of the protein fragment or peptide-polysaccharide conjugate. In another preferred embodiment, the vaccine comprises about 5 pg of the protein fragment or peptide or an equivalent amount of the protein fragment or peptide polysaccharide conjugate. The vaccine of the present invention may be employed in such forms as 15 capsules, liquid solutions, suspensions or elixirs for oral administration, or sterile liquid forms such as solutions or suspensions. Any inert carrier is preferably used, such as saline, phosphate-buffered saline, or any such carrier in which the protein fragment or peptide or conjugate vaccine has suitable solubility properties. The vaccines may be in the form of single dose preparations or in 20 multi-dose flasks which can be used for mass vaccination programs. Reference is made to Remington's Pharmaceutical Sciences, Mack Publishing Co., Easton, PA, Osol (ed.) (1980); and New Trends and Developments in Vaccines, Voller et al. (eds.), University Park Press, Baltimore, MD (1978), for methods of preparing and using vaccines. 25 The vaccines of the present invention may further comprise adjuvants which enhance production ofprotein-specific antibodies. Such adjuvants include, but are not limited to, various oil formulations such as Freund's complete adjuvant (CFA), stearyl tyrosine (ST, see U.S. Patent No. 4,258,029), the dipeptide known as MDP, saponin (see U.S. Patent No. 5,057,540), aluminum 30 hydroxide, and lymphatic cytokine.

WO 00/15760 PCT/US99/21643 -41 Freund's adjuvant is an emulsion of mineral oil and water which is mixed with the immunogenic substance. Although Freund's adjuvant is powerful, it is usually not administered to humans. Instead, the adjuvant alum (aluminum hydroxide) or ST may be used for administration to a human. The protein 5 fragment or peptide vaccine or a conjugate vaccine thereof may be absorbed onto the aluminum hydroxide from which it is slowly released after injection. The vaccine may also be encapsulated within liposomes according to Fullerton, U.S. Patent No. 4,235,877. In another preferred embodiment, the present invention relates to a 10 method of inducing an immune response in an animal comprising administering to the animal the vaccine of the invention, produced according to methods described, in an amount effective to induce an immune response. Having now generally described the invention, the same will be more readily understood through reference to the following Examples which are 15 provided by way of illustration, and are not intended to be limiting of the present invention, unless specified. Examples Example 1 Cloning and Expression of the Gene Encoding C/ 20 To locate the IgA binding site on the CP protein, two oligonucleotides were synthesized. The first oligonucleotide, oligo 1, corresponds to the 5' end of the mature protein, and has the sequence (SEQ ID NO: 8) 5' AAGGATCCAAGTGAGCTTGTAAAGGACGAT-3', which includes a BamHI site. The second oligonucleotide falls just short of the 3' end of the gene, and has 25 the sequence (SEQ ID NO: 9) 5'-AAAACTCGAGTTTCTTTTCCGTTGTTG ATGTA-3', and includes a XhoI site. The oligonucleotide for the 3' end of the gene was chosen to eliminate the LPXTG motif found in most Gram positive cell wall proteins. This sequence motif has been shown to be involved in the WO 00/15760 PCT/US99/21643 -42 processing of these cell wall proteins and is the part of these proteins which eventually becomes covalently bound to peptidoglycan (Navarre, W.W. and Schneewind, O., Molec. Microbiol. 14:115-121 (1994); Schneewind, O., et al., Science 268:103-106 (1995)). Using chromosomal DNA from Strain A909 5 Group B streptococci containing the gene for the C3 protein as a template, and standard PCR procedures, a product of approximately 3.2 kb was produced as observed when electrophoresed on a 1% agarose gel. The PCR product containing the CP protein gene was cleaved with the endonuclease restriction enzymes BamHI and Xhol. This BamHI-XhoI DNA fragment contained the 10 sequence for the entire CP protein except for the last 33 amino acids at the carboxyl terminus, including the putative IgA binding site. The DNA fragment was then ligated into the appropriately restricted T7 expression plasmid pET1 7b (Novagen, Inc., Madison WI) using a standard T4 ligase procedure. The plasmid was then transformed into the E. coli strain BL21 (DE3) using the manufacturer's 15 suggested protocols (Novagen, Inc.). E. coli cells containing the plasmid were selected on LB plates containing 50 pg/ml carbenicillin. These plates were incubated overnight at 37oC. The transformant colonies were carefully lifted onto nitrocellulose filters saturated with IPTG. After 30 minutes, the bacteria were lysed by placing the filters into a chloroform vapor chamber for 15 minutes 20 at room temperature. After the filters were removed from the chamber, they were placed, colony-side up, onto a Whatman® 3MM filter which had been previously saturated with 20 mM Tris-HC1, pH 7.9, 6 M urea, and 0.5 M NaC1. After 15 minutes, the filters were washed three times in PBS and incubated for 1 hour with 25 purified human IgA in PBS-Tween®. The filters were then rewashed in the PBS Tween® and developed by standard procedures (Blake, M.S., et al., Analyt. Biochem. 136:175-17 (1984)) using a goat antihuman IgA-alkaline phosphatase conjugate (Cappel Research Products, West Chester, PA). Several colonies demonstrating high IgA binding activity were selected and grown overnight in 1 30 ml LB broth containing carbenicillin at 30'C. These cultures were then diluted 1 to 100 with fresh LB-carbenicillin broth and incubated at 30 oC for an additional WO 00/15760 PCT/US99/21643 -43 6 hours. Expression was then induced by the addition of IPTG and the culture continued for an additional 2 hours at 30 0 C. The cells were collected by centrifugation, resuspended in water and subjected to several freeze-thaw cycles. The cells were once again collected by centrifugation and the supernatants saved 5 for examination of their IgA binding activity. Example 2 Identification of the IgA Binding Domain of C/i Once certain a stable plasmid producing a recombinant CP protein had been achieved and that the expressed protein bound human IgA, a strategy similar 10 to that of the Novatope® System (Novagen, Inc.) was utilized to locate the IgA binding region of Cp. This procedure was performed according to the manufacturer's instructions. Briefly, the purified plasmid containing the C[ gene was randomly digested with DNase I and electrophoresed in a 2% low melting point agarose gel. Fragments of the DNA corresponding to sizes between 100 to 15 300 base pairs were excised from the gel, purified, and resuspended in TE buffer. A single dA was added to the fragments using the recommended reaction mixture and the fragments ligated into the pETOPET vector which contained single dT ends. After the standard ligation procedure, the plasmids were transformed into competent NovaBlue (DE3) cells (Novagen, Inc.) and plated on LB plates 20 containing 50 pg/ml carbenicillin. These plates were incubated overnight at 37oC. The transformant colonies were tested for IgA binding activity as described in Example 1. Several clones were selected on the bases of their binding to the IgA. The bacteria from each of these clones were inoculated separately onto fresh LB plates and retested for their IgA binding ability as before. 25 Plasmid preparations were made from each by standard means and sequenced. The nucleotide sequences of the cloned CP3 protein gene fragments were determined by the dideoxy method using denatured double-stranded plasmid DNA template as described (Current Protocols in Molecular Biology, John Wiley & Sons, New York, N.Y. (1993)). Sequenase II kits (United States Biochemical WO 00/15760 PCT/US99/21643 -44 Corp., Cleveland, OH) were used in accordance with the manufacturer's instructions. The smallest fragment of DNA obtained that included part of the CP gene is shown in FIG. 1. The translation of this sequence corresponds to amino acids 101 to 230 of the mature CP protein shown in FIG. 1 (SEQ ID NO: 2). 5 Attempts to further shorten this DNA fragment failed to give any IgA binding activity. Example 3 ELISA Inhibition Assays: Peptide Binding Studies Several synthetic peptides were made corresponding to the amino acid 10 sequence contained within this region of the CP protein. Peptides were synthesized using NMP t-butoxycarbonyl chemistry on an ABI 430A peptide synthesizer (Applied Biosystems, Foster City, CA) and were deprotected. Peptides from a sample of the resin were removed from the resin by treatment with HF in the presence ofanisole (0 oC/1 hour). Preparative purification of these 15 peptides were performed using a C18 column (2.14 ID x 30 cm) (Dynamax Rainin, Woburn, MA). The peptides were quantitated by PTC amino acid analysis using Waters Picotag system (Waters, Milford, MA). The synthesized peptides eluted from the C 18 column as a major peak consisting of usually 75 85% of the total elution profile. The amino acid composition of the purified 20 peptides were in good agreement with the sequence which was used to synthesize the peptides. These peptides were used in ELISA inhibition assays to block the binding of human IgA to the purified CP protein as follows. Microtiter plates (Nunc-Immuno Plate IIF, Vangard International, Neptune, NJ) were sensitized by adding 0.1 ml per well of purified CP at a concentration of 2.0 ig/ml in 0.1 M 25 Carbonate buffer, pH 9.6 with 0.02% azide. The plates were incubated overnight at room temperature. The plates were washed five times with 0.9% NaC1, 0.05% Brij 35, 10 mM sodium acetate pH 7.0, 0.02% azide. A purified human IgA myeloma protein was purchased from Cappel Laboratories, diluted in PBS with 0.5% Brij 35, added to the plates, and incubated for 1 hour at room temperature.

WO 00/15760 PCT/US99/21643 -45 The plates were again washed as before and the secondary antibody, alkaline phosphatase conjugated goat anti-human IgA (Tago Inc., Burlingame, CA), was diluted in PBS-Brij, added to the plates, and incubated for 1 hour at room temperature. The plates were washed as before and p-nitrophenyl phosphate 5 (Sigma 104® phosphatase substrate) (1 mg/ml) in 0.1 M diethanolamine, 1 mM MgCl 2 , 0.1 mM ZnCl 2 , 0.02% azide, pH 9.8 added. The plates were incubated at 37'C for 1 hour and the absorbance at 405 nm determined using an Elida-5 microtiter plate reader (Physica, New York, NY). Control wells lacked either the primary and/or secondary antibody. This was done to obtain a titer of the human 10 IgA myeloma protein which would give a half-maximal reading in the ELISA assay. This titer would be used in the inhibition ELISA. The microtiter plates were sensitized and washed as before. Purified synthetic peptides were added and diluted in PBS-Brij. The dilution of the human IgA myeloma protein which gave the half-maximal reading was then added. The mixture was then incubated for 15 1 hour at room temperature. The plates were rewashed and the conjugated second antibody added as stated. The plates were then processed and read as described. The percentage of inhibition would be calculated as follows: 1 - (ELISA value with the peptide added)/(ELISA value without the peptide added). The peptide which inhibited in this ELISA assay contained the sequence Asn-His 20 Gln-Lys-Ser-Gln-Val-Glu-Lys-Met-Ala-Glu-Gln-Lys-Gly (SEQ ID NO: 10). This suggested that at least part of the IgA binding domain of the C3 was comprised within the region of the protein containing this sequence. Example 4 Oligonucleotide Directed Mutagenesis of the Gene Encoding C/i 25 In order to confirm the importance of this region in the CP protein for IgA binding activity and to begin to generate the mutant proteins that in the end would be used in the vaccine formulation, a modification of the Ex-Site® PCR site directed mutagenesis protocol was employed as developed by Stratagene WO 00/15760 PCT/US99/21643 -46 (Stratagene, CA). The template used was a plasmid called pNV222 which consisted of the CP gene inserted into the pSP76 vector (Promega, Madison, WI). DNA oligonucleotides were synthesized on an Applied Biosystems model 292 DNA Synthesizer (Foster City, CA). The oligonucleotides were manually cleaved 5 from the column by treatment with 1.5 ml of ammonium hydroxide for 2 hours with gentle mixing every 15 minutes. They were deprotected at 55 0 C for 16-18 hours. After deprotection they were dried down and used directly or purified using oligonucleotide purification columns (Applied Biosystems, Foster City CA). Several different oligonucleotides were made to induce different changes 10 in the DNA sequence in the region of interest. An example of which is the following. The primers, in this particular example, were overlapping primers, both containing the sequence required to change lysine to alanine at amino acids 170 and 175 in the sequence shown in FIG. 1 (SEQ ID NO: 2). The forward primer, designated CP 613, had the sequence (SEQ ID NO: 6) 5'-GTT GAA GCA 15 ATG GCA GAG CAA GCG GGA ATC ACA AAT GAA G-3' and the reverse primer, designated CP 642R, had the sequence (SEQ ID NO: 7) 5'-GAT TCC CGC TTG CTC TGC CAT TGC TTC AAC TTG ACT TTT TTG-3' (the substitutions are noted in BOLD). The reaction conditions were as follows: 10 ng pNV222 template, 15 pmol of each primer, 1 mM of each dNTP, lX Vent 20 polymerase buffer (20 mM Tris-HC1, pH 7.5; 10 mM KCl; 10 mM (NH 4

)

2

SO

4 ; 2 mM MgSO 4 0.1% (v/v) Triton® X-100; 0.1 mg/ml bovine serum albumin (BSA)), 10 units of Vent polymerase, and H 2 0 to 100 pl. The reactions were prepared with PCR Gem 10 wax beads as per the Hot Start Protocol (Perkin Elmer, Foster City, CA). The reactions were run in a Perkin Elmer® 25 Thermocycler (Perkin Elmer, Foster City, CA) under the following conditions: 1 cycle of 94oC for 5 minutes; 10 cycles of 94'C for 30 seconds, 37 0 C for 2 minutes, 72oC for 10 minutes; 30 cycles of 94'C for 30 seconds, 55'C for 2 minutes, 72oC for 10 minutes; and 1 cycle of 72oC for 12 minutes. The reaction was treated with 10 units of DpnI at 37'C for 30 minutes to destroy the template 30 DNA, followed by a 60 minute treatment at 72oC with PfuI polymerase to fill in any remaining overhangs. The reaction was diluted 1:4.6 in 1 X Vent polymerase WO 00/15760 PCT/US99/21643 -47 buffer plus 0.38 mM dATP. The diluted reaction was ligated for 24 hours at 25 0 C and transformed into competent DH5a. cells (Gibco/BRL, Gaithersburg, MD). Selected colonies were grown in 3 ml of LB plus kanamycin (50 mg/ml) at 37 oC for 16-18 hours. DNA was prepared using QIAspin® columns (Qiagen, 5 Chatsworth, CA). The clones were analyzed for insert size on 0.8 % agarose gels and then sequenced. Selected clones were then grown in 100 ml LB plus kanamycin (50 mg/ml) at 37 0 C for 16-18 hours. DNA was prepared using the Qiagen® tip 100 (Qiagen, Chatsworth, CA). They were then digested with NdeI and PstI and run on 0.8% agarose gels to separate the mutated region. The 2300 10 bp fragment was isolated and purified from the gel using the Gene-Clean Spin Kit® (Bio 101, Vista, CA). A clone named pNV34 which consisted of the expression vector pET 24a (Novagen, Inc.) and the native CP gene, was also digested with NdeI and PstI and run on a 0.8% agarose gel. The large band (6300 bp) containing the pET vector and the remainder of the CP gene was isolated and 15 purified from the gel using the Gene-Clean Spin Kit® (Bio 101). These two fragments were ligated at 4oC for 24 hours and transformed into competant BL21(DE3) cells. Selected colonies were grown in 3 ml of LB plus kanamycin (50 mg/ml) at 37'C for 16-18 hours. DNA was prepared using QIAspin® columns (Qiagen) and the clones were analyzed for insert size on 0.8% agarose 20 gels. Also constructed were clones encoding mutant CP proteins wherein two glutaminyl residues were replaced by prolinyl residues, and wherein a deletion in the CP gene had occurred resulting in a 6 amino acid deletion in the region of interest (Table 1). 25 Clones expressing a CP protein which lacked or had reduced IgA binding activity, but still reacted with the anti-Pag antiserum were selected (Example 5) and grown in 100 ml LB plus kanamycin (50 mg/ml) at 37 0 C for 16-18 hours. Plasmid DNA from these clones was prepared using Qiagen® tip 100 (Qiagen) and the mutated CP gene entirely sequenced.

WO 00/15760 PCT/US99/21643 -48 Example 5 Western Blot and ELISA Analysis of IgA Binding by C/ Mutants The proteins encoded by the mutated genes were expressed and subjected to SDS-PAGE and western blot analysis in order to determine if mutations in the 5 gene encoding the CP protein reduced or eliminated IgA binding, while retaining C3 antigenicity. Two western blots were made for each sample and reacted with either the purified human IgA myeloma protein or hyperimmune rabbit anti-Pag protein antiserum. The clone expressing a C3 protein wherein lysine was changed to alanine at amino acids 170 and 175 in the sequence shown in FIG. 1 10 (SEQ ID NO: 2) demonstrated almost no IgA binding activity, but the ability of the protein to react with anti-Cp antiserum remained high. IgA binding activity was also substantially eliminated in the clone expressing a C3 protein wherein two glutaminyl residues are replaced by prolinyl residues and in the clone encoding a CP protein having a six amino acid deletion (Table 1), while reactivity 15 with the anti-Cp antiserum was maintained for both. The data for the clone having a six amino acid deletion suggested that the residues responsible for the IgA binding activity of the CP protein were located within this region of the protein, and that other possible mutations within this area would affect the IgA binding activity. 20 A competitive inhibition ELISA was used to more precisely determine the amount of antigenic and/or structural changes the sequence modifications had on the CP protein. Microtiter plates (Nunc-Immuno Plate IIF, Vangard International, Neptune, NJ) were sensitized by adding 0.1 ml per well of purified CP at a concentration of 2.0 pg/ml in 0.1 M carbonate buffer, pH 9.6 with 0.02% azide. 25 The plates were incubated overnight at room temperature. The plates were washed five times with 0.9% NaC1, 0.05% Brij 35, 10 mM sodium acetate pH 7.0, 0.02% azide. Hyperimmune rabbit antiserum to the C[ protein was diluted in PBS with 0.5% Brij 35 and added to the plate and incubated for 1 hour at room temperature. The plates were again washed as before and the secondary antibody, 30 alkaline phosphatase conjugated goat anti-rabbit IgG (Tago Inc., Burlingame, WO 00/15760 PCT/US99/21643 -49 CA), was diluted in PBS-Brij, added to the plates and incubated for 1 hour at room temperature. The plates were washed as before and p-nitrophenyl phosphate (Sigma 104® Phosphatase Substrate) (1 mg/ml) in 0.1 M diethanolamine, 1 mM MgCl 2 , 0.1 mM ZnCl 2 , 0.02% azide, pH 9.8, was added. 5 The plates were incubated at 37oC for 1 hour and the absorbance at 405 nm determined using an Elida-5 microtiter plate reader (Physica, New York, NY). Control wells lacked either the primary and/or secondary antibody. This was done to obtain a titer of the rabbit anti-Cp protein which would give a half maximal reading in the ELISA assay. This titer would be used in the inhibition 10 ELISA. The microtiter plates were sensitized and washed as before. Purified C[ protein or mutations of the CP protein were added and diluted in PBS-Brij. The dilution of the rabbit anti-Cp protein which gave the half-maximal reading was then added. The mixture was then incubated for 1 hour at room temperature. The plates were rewashed and the conjugated second antibody added as stated. The 15 plates were then processed and read as described. The percentage of inhibition would be calculated as follows: 1 - (ELISA value with the protein added)/(ELISA value without the proteins added). In this assay, the inhibition of the wild-type CP protein from streptococci is compared with the recombinant CP protein and the glutaminyl to prolinyl 20 mutants, both expressed in E. coli (data not shown). This assay is sensitive enough to detect the absence of the membrane spanning region in the recombinants of the CP proteins. However, when the recombinant C[ protein containing the wild-type sequence is compared to the substitution mutant lacking IgA binding activity, the antigenic differences are minimal. This would suggest 25 that such substitution mutants maintain most of the antigenic character of the CP protein but lack the unwanted the IgA binding activity.

WO 00/15760 PCT/US99/21643 -50 Example 6 Initial Separation of Recombinant C/ Protein from E. coli Cell pastes were resuspended in ice cold cell disruption buffer (5 mM Tris(HC1), pH 7.6-7.8, containing 5 mM DTT, 0.2 mg/ml EDTA and PEFABLOC 5 PLUS® Protease Inhibitor (Boehringer Mannheim, Indianapolis, IN)). The cell suspension was passed through a Stansted® cell disruptor twice, then centrifuged at 12,000 x g for 30 minutes. The supernatant was decanted and non proteinaceous material was precipitated by addition of ethanol/CaCl 2 to a final concentration of 20% ethanol/0.1 M CaCl 2 . 10 Proteins were precipitated with final 80% ethanol after clarification by centrifugation at 10,000 x g for 30 minutes, then solubilized in buffer containing 5% Zwittergent® 3,14, 25 mM Tris(HC1) and 5 mM DTT, pH 7.6-7.8. Example 7 Initial Separation of C/ Protein from S. agalactiae 15 Native C3 protein was recovered from the membrane of Streptococcus agalactiae by boiling the bacteria in buffer containing 5% Zwittergent® 3,14 and 25mM Tris(HC1), pH 7.6-7.8. After boiling, the suspension was cooled to between O'C and 10 0 C in an ice-bath, then a cold (0OC - 4oC) solution of ethanol/CaCl 2 was added to a final concentration of 20% ethanol/0.1 M CaCl 2 . 20 The solution was clarified by centrifugation, after which proteins were precipitated from the supernatant with 80% ethanol. Protein was resolubilized in a buffer containing 5% Zwittergent®, 25mM Tris(HC1), 5mM DTT, and "PEFABLOC PLUS" Protease Inhibitor (Boehringer Mannheim Corporation, Indianapolis, IN), pH 7.6-7.8.

WO 00/15760 PCT/US99/21643 -51 Example 8 Purification of C/3 Protein Optimally, the protein concentration in the 5% Zwittergent®/Tris buffer should be between 0.5-0.8 mg/ml before applying to a Pharmacia High 5 Performance Q® column (1.6 x 20 cm). When a linear salt gradient was used to elute CP protein, the gradient used was from 0 - 0.1 M NaCl in 25 mM Tris(HC1)/0.5% Zwittergent® 3,14, pH 7.6-7.8. A step gradient may also be applied at 0, 50, 75, and 90 mM NaCl in the Tris/Zwittergent® buffer and the CP protein eluted in 75mM NaC1. Alternatively, CP protein may be eluted using 60 10 mM NaCl/0.6% CHAPS in Tris(HC1), pH 7.6-7.8, after washing the column in 50 mM NaCl/Tris(HC1)/0.5% Zwittergent® buffer. CP protein will elute at a lower salt concentration with increased total protein applied to the column. SDS-PAGE was used to identify CP fractions from the Q® column. These fractions were pooled and protease inhibitor PEFABLOC PLUS® (Boehringer 15 Mannheim, Indianapolis, IN) was added. The protein pool was then diluted 3 fold, i.e., one part protein pool to two parts diluent (v/v) containing 15% Zwittergent®, 25 mM Tris(HC1) and 5 mM DTT, pH 7.6-7.8. The protein was then applied to a Pharmacia High Performance Heparin Sepharose® column (1.6 x 20 cm). CP protein was eluted using a linear salt gradient (0 - 0.1 M NaCl in 20 25 mM Tris(HC1)/0.5% Zwittergent®). Generally, the CP protein may be eluted from the heparin column utilizing the same buffer and gradient conditions as with the Q® column. FIG. 10 illustrates elution profiles of Q® and heparin columns. FIGS. 3A and 3B show purification data and corresponding SDS-PAGE gel analysis. 25 When a gel filtration column was used in the second chromatographic separation, fractions from the Q® column containing the CP protein were precipitated by the addition of absolute ethanol to a final concentration of 80% ethanol. The precipitated proteins were collected by centrifugation and resolubilized in a minimal volume of 25 mM HEPES, 1 M NaC1, 5 mM DTT, pH 30 8.0 with 10% Zwittergent®. The redissolved sample was applied to a gel filtration WO 00/15760 PCT/US99/21643 -52 column previously equilibrated in the same buffer, but lacking the Zwittergent® and DTT for the recombinant proteins, and containing 0.5% Zwittergent® for purification of the native protein. Example 9 5 Protease Digestion of C/ Protein and Isolation of Fragments Purified CP protein (6.0 - 8.0 mg/ml) in Dulbecco's PBS (Life Technologies, Gaithersburg, MD) was digested by adding 0.025 ml of a 10 mg/ml stock solution ofthermolysin (Boehringer Mannheim, Indianapolis, IN) in 20 mM CaCl 2 . The reaction mixture was incubated at 60'C overnight with constant 10 agitation. Protease activity was then terminated by the addition of 10 mM EDTA (FIGS. 4A and 4B). Initially, thermolysin generated peptide fragments TI and T2 (apparent molecular weights of 35 kD and 25 kD, respectively) were separated by electrophoresis, transferred to nitrocellulose, stained in Ponceau S, then recovered 15 by dissolving the membrane in acetonitrile (FIG. 5). After dissolution of the membrane, peptide fragments were precipitated with ice-cold acetone and collected by centrifugation. After centrifugation, the peptide precipitates were dissolved in PBS and assayed for inhibition of bactericidal activity by a method of Fusco et al., Adv. Exp. Med. Biol 418: 841-845 (1997). Aliquots of acetone 20 precipitated protein were also run on SDS-gels to confirm protein recovery. In order to increase the amount of protein for subsequent digestion, thermolysin fragments were precipitated with 80% final ethanol and collected by centrifugation. The pellet was resolubilized in 1-3 ml of 10 mM Tris(HC1)/10 mM CaCl 2 , 10 % glycerol, pH 7.2 and applied to a Sephacryl®-100 column (1.6 25 X 40 cm) equilibrated in 10 mM Tris(HC1)/10 mM CaCl 2 , pH 7.2 (FIG. 6). The isolated peptide fragment T 1 was further digested using endoprotease Arg-C from Clostridium histolyticum (Boehringer Mannheim, Indianapolis, IN). The Arg-C digest was assayed for inhibition of bactericidal activity. Fragments WO 00/15760 PCT/US99/21643 -53 designated R1 and R2 were later separated by chromatofocusing on a Mono P® column using Polybuffer® 74 (Pharmacia, Piscataway, NJ) (FIG. 7). Inhibition of bactericidal activity using unseparated thermolysin fragments T1 and T2 indicated 50% inhibition at a protein concentration of 36 tg/ml. When 5 thermolysin fragments TI and T2 were separated and recovered by electrophoresis and western transfer, 50% inhibition of bactericidal activity occurred at 15 tg/ml and 20 pg/ml for Tl and T2 fragments, respectively. Tl and T2 peptides at 18 pg/ml and 40 pg/ml, respectively, resulted in 100% inhibition. The yield of T1 peptide after Arg-C digestion was minimal. Protein 10 concentration for the Arg-C digested fragments was not determined. However, 50% inhibition of bactericidal activity was obtained with a 1:50 dilution of the sample. This activity was compared with a T1 peptide fraction separated by gel filtration. This T1 fraction demonstrated 50% inhibition at 2.5 Ig/ml. N-terminal sequence data for two major peptide fragments from 15 thermolysin digestion T I and T2 of apparent molecular weights near 35 kD and 25 kD was obtained by SDS-PAGE on tricine gels, transferred to PDVF membranes in 10 mM CAPS/1 0% methanol, stained in 0.1% Ponceau S in 1% acetic acid, destained in 5% acetic acid, and washed extensively in water with a vortex. Well-dried PDVF pieces were excised and sent to MA BioServices® 20 (Rockville, MD) for sequencing. Sequence analysis results ofthermolysin digested products showed the N terminus for T I was VEQDQPAPIPENSE (SEQ IDNO: 52). For T2 peptide, the N-terminus was LAANENNQQKIELTV (SEQ ID NO: 53). The TI peptide fragment is closer to the C-terminal end of the CP protein than the T2 peptide 25 fragment (FIG. 8). Example 10 Peptide Synthesis The Chiron Mimotopes Multipin Non-Cleavable Peptide Kit (Chiron Mimotopes, Raleigh, NC) was used to synthesize peptides of the C[ mutant WO 00/15760 PCT/US99/21643 -54 holoprotein. A computer program supplied with the kit assisted in the generation of continuous 8-mer peptides with overlapping sequences of 4 amino acid residues to determine general regions of antibody binding. Peptide synthesis utilized FMOC protected amino acids and DIC/HOBt for amino acid activation 5 as recommended by the kit manual. Side-chain protecting groups were removed and the N-termini of peptides were acetylated. Preliminary ELISA results obtained utilizing 8-mer peptides indicated an antibody binding region. Twenty mer peptides with overlapping sequences were synthesized and utilized to confirm initial results. 10 The 24-mer peptide sequence, PDVPKLPDVPKLPDVPKLPDAPKL (SEQ ID NO: 47), was also synthesized by solid phase peptide synthesis, cleaved, and purified to obtain a peptide at a multimilligram scale (Protein/DNA Technology Center, Rockefeller University, NY). This peptide was used in ELISA inhibition assays. 15 For ELISA, mimotope pins were blocked in 2% BSA/PBS-0.1% Tween® (polyoxyethylenesorbitan) (PBS-T) for 30 minutes. After washing pins in at least 5 changes of PBS-T, pins were gently agitated in 1:20,000 dilution of polyclonal rabbit antibody 52.2 in 0.05% BSA/PBS-T for 1.5 hours. Pins were again washed in at least 5 changes of PBS-T before agitation in a 1:2000 dilution of alkaline 20 phosphatase conjugated anti-rabbit antibody in 0.2% fetal bovine serum/PBS-T. ELISA plates were developed in SIGMA 104® Phosphatase substrate (Sigma Chemical Company, St. Louis, MO) in a diethanolamine buffer. Plates were developed with constant shaking at room temperature for 1 hour and were then read at 405 nm on a plate reader. 25 Results of the ELISA indicated that most antibody binding was directed towards a repetitive sequence located near the C-terminal end of the protein. The repetitive sequence exists on the CP protein as PDVPKLPDVPKLPDVPKLPDAPKL (SEQ ID NO: 47). The 8-mer sequences PDVPKLPD (SEQ ID NO: 33), KLPDVPKL (SEQ ID NO: 34), VPKLPDVP 30 (SEQ ID NO: 35), KLPDAPKL (SEQ ID NO: 36) elicited most antibody binding (FIG. 9A). To a lesser extent, some antibody binding activity was directed WO 00/15760 PCT/US99/21643 -55 towards the 8-mer sequences ETPDTPKI (SEQ ID NO: 38), RTVRLALG (SEQ ID NO: 39), and sometimes GGGTVRVF (SEQ ID NO: 40). The 20-mer peptides that were synthesized (in duplicate) to verify antibody binding were: SPKTPEAPKIPEPPKTPDVP (SEQ ID NO: 41), PEAPKIPEPPKTPDVPKLPD 5 (SEQ ID NO: 42), KIPEPPKTPDVPKLPDVPKL (SEQ ID NO: 43), PPKTPDVPKLPDVPKLPDVP (SEQ ID NO: 44), PDVPKLPDVPKLPDVPKLPD (SEQ ID NO: 45), and KLPDVPKLPDVPKLPDAPKL (SEQ ID NO: 46) (FIG. 9B). Example 11 10 Inhibition ELISA by C/6 Protein and Synthetic Peptide Twenty ml of a 1:20,000 dilution of polyclonal rabbit antibody 52.2 in 0.05% BSA/PBS-T containing 0.1 mg/ml recombinant CP protein was incubated for 30 minutes at room temperature before assaying pins for antibody binding as described above. ELISA results indicated inhibition of antibody binding to pins 15 of the sequences PDVPKLPD (SEQ ID NO: 33), KLPDVPKL (SEQ ID NO: 34), VPKLPDVP (SEQ ID NO: 35), and KLPDAPKL (SEQ ID NO: 36). However, some antibody binding activity occurred with the sequence RTVRLALG (SEQ ID NO: 39) (data not shown). In inhibition experiments using the synthetic peptide 20 PDVPKLPDVPKLPDVPKLPDAPKL (SEQ ID NO: 47), the 8-mer and 20-mer pins were blocked in BSA. At the same time, 20 ml of a 1:20,000 dilution of 52.2 anti-bag rabbit antibody in 0.05% BSA / 0.1% Tween®-PBS was incubated 30 minutes with (0.1 mg/ml) synthetic peptide (Protein/DNA Technology Center, Rockefeller University, NY) before assaying pins for antibody binding as 25 described above. One set of the duplicate 20-mer pins served as a positive control for antibody not exposed to the synthetic peptide. ELISA results indicated inhibition of antibody binding to pins of the sequences PDVPKLPD (SEQ ID NO: 33), KLPDVPKL (SEQ ID NO: 34), VPKLPDVP (SEQ ID NO: 35), and KLPDAPKL (SEQ ID NO: 36). Again, WO 00/15760 PCT/US99/21643 -56 antibody binding was observed with sequences ETPDTPKI (SEQ ID NO: 38), and somewhat less, RTVRLALG (SEQ ID NO: 39). Antibody binding to 20-mer pins was inhibited similarly. The positive control pins indicated antibody binding for comparison (FIG. 10). 5 Example 12 Bactericidal Inhibition Assay to Establish that a Particular Repetitive Amino Acid Sequence on C/ Elicits Antibody Binding Rabbits were vaccinated with a synthetic peptide coupled to tetanus toxoid to generate specific antibody to a region on Cp. A specific antibody directed 10 toward the sequence PDVPKLPDVPKLPDVPKLPDAPKL (SEQ ID NO: 48) was obtained and purified and able to demonstrate killing GBS (Group B streptococci) type Ib. This killing was compared to the bactericidal activity of rabbit antiserum to the native C3 holoprotein. Materials and Methods 15 Coupling a Peptide to a Carrier Protein Peptides were synthesized by The Rockefeller University Protein Sequencing Facility using NMP t-butoxycarbonyl chemistry on an ABI 430A peptide synthesizer (Applied Biosystems, Foster City, CA). The peptides were deprotected and removed from the resin by treatment with HF in the presence of 20 anisole (0 0 C/1 hour). Preparative purification of each peptide was performed using a C18 column (2.14 ID x 30 cm) (Dynamax-Rainin, Woburn, MA). The peptides were quantitated by PTC amino acid analysis using Waters Picotag system (Waters, Milford, MA). The synthesized peptides eluted from the C 18 column as a major peak consisting of greater than 95% of the total elution profile. 25 The amino acid composition of the purified peptides were in good agreement with the sequence. Mass spectral analysis (MALDI-TOF) conducted using a WO 00/15760 PCT/US99/21643 -57 Perseptive Biosystems® DERP Mass Spectrometer run in a linear mode using alpha-hydroxy-cynnamic acid as a matrix, indicated a peptide mass of 2677 kD, which is in good agreement with the calculated molecular weight of approximately 2658 kD. 5 Approximately 26 mg of synthetic peptide, SPDVPKLPDVPKLPDVPKLPDAPKL (SEQ ID NO: 49) (designated Ser peptide), dissolved in 1.5 ml deionized water, was citraconylated (Atassi, M.Z. and Habeeb, A.F.S.A., "Reaction of proteins with citraconic anhydride," in: Methods in Enzymology, Vol. XXV Enzyme Structure Part B, edited by Hirs, 10 C.H.W. and Timasheff, S.N., New York, Academic Press, p. 546-533 (1972)) by adding 60 pl of citraconic anhydride in 15 pl aliquots at 30 minute intervals at room temperature. The reaction mixture was maintained above pH 8 by titration with 5 M NaOH. The reaction proceeded for 1 hour after the final addition of citraconic anhydride, then the solution was exhaustively dialyzed against 0.2 M 15 Na 2

HPO

4 , pH 8.5, using 100 MWCO dialysis tubing. The serine residue on the peptide was then oxidized in the dark at room temperature for 10 minutes using 2.2 mg of sodium periodate (Geoghegan, K.F. and Stroh, J.G., "Site-directed conjugation of non-peptide groups to peptides and proteins via periodate oxidation of a 2 amino-alcohol - Application to 20 modification atN-terminal serine," Bioconjugate Chem., 3: 138-146 (1992)). The reaction was quenched by adding 20 pl of ethylene glycol with constant stirring for 30 minutes, then again exhaustively dialyzed against 0.2 M Na 2

HPO

4 , pH 8.5. Purified tetanus toxoid monomer (Statens Seruminstitut, Copenhagen S., Denmark) was rechromatographed on Superdex® 200. The serine peptide was 25 coupled to 15 mg tetanus toxoid by reductive amination using 13 mg sodium cyanoborohydride. The reaction proceeded overnight at 37 0 C, then was exhaustively dialyzed against citric acid water (pH 4.2) to remove the protecting groups on the lysine residues of the peptide. Approximately 8 mg product was recovered (gravimetric determination which included uncoupled peptide). 30 Coupling was monitored by HPLC using Superdex® peptide column, by injecting WO 00/15760 PCT/US99/21643 -58 aliquots of peptide, tetanus toxoid, and a mixture of peptide and tetanus toxoid after coupling. Furthermore, conjugate dialyzed with 12,000 - 14,000 MWCO dialysis tubing to remove uncoupled peptide was assayed by ELISA. A microtiter plate was coated with 1 pg Ser-peptide conjugate overnight at 4oC. 5 The polyclonal rabbit 52.2 primary antibody was used at 1:20,000 dilution in PBS-T, followed by secondary antibody alkaline phosphatase goat anti-rabbit conjugate (ICN, Costa Mesa, CA) at 1:2000 dilution in PBS-T. The ELISA plate was developed in SIGMA 104® Phosphatase substrate in 1 M diethanolamine buffer/0.5 mM MgCl 2 , pH 9.8. The plate was read at 405 nm on a plate reader 10 (Dynex Technologies, Inc., Chantilly, VA). Ser-Peptide Antibody New Zealand white rabbits were immunized on day 0 using a 100 pg dose of mixture of both coupled and uncoupled peptide in Freund's complete adjuvant. On days 21 and 42, rabbits were again immunized with a 100 tg dose in Freund's 15 incomplete adjuvant. Rabbits were bled on day 51. Upon obtaining rabbit anti serum, ELISA was conducted by coating a 96-well ELISA plate with 2 [tg/ml C[ protein, then diluting the Ser-peptide antibody starting at 1:1000 with duplicate 2-fold dilutions to obtain titers. Additionally, to evaluate the specificity of the antibody binding, 20 mimotope pins representing 8-mer sequences of the entire CP molecule were blocked in 2% BSA/PBS-T for 30 minutes. After washing, pins were incubated with 1:20,000 dilution of ser-peptide antibody for 1.5 hours. Pins were again washed, then incubated in 1:2000 alkaline phosphatase anti-rabbit antibody for 1.5 hours. ELISA plates were developed in SIGMA 104® Phosphatase substrate 25 in 1 M diethanolamine buffer/0.5 mM MgCl 2 , pH 9.8. The plates were read at 405 nm on a plate reader.

WO 00/15760 PCT/US99/21643 -59 Purification of Ser-Peptide Rabbit Antibody 30 mg tetanus toxoid were dissolved in 7 ml of 0.2 M NaHCO 3 /0.5 M NaC1, pH 8.3 and were coupled onto a 5 ml HiTrap® NHS-activated affinity column (Pharmacia, Piscataway, NJ) by recycling for 4 hours at room 5 temperature using a peristaltic pump. After washing and deactivating, as recommended by the manufacturer, the column was equilibrated in PBS, then 20 ml of ser-peptide rabbit serum was passed through the column. Void fractions were pooled and assayed using mimotope pins representing the N- and C terminal sequences of C3. Immunoglobulin Gs were further purified using an 10 Immunopure® G column (Pierce, Rockford, IL). Fractions eluting from the column in a low pH buffer were neutralized and desalted on 5 ml HiTrap® desalting column (Pharmacia, Piscataway, NJ) before again testing antibody binding specificity on mimotope pins which represented N and C-terminal ends of CP. 15 Bactericidal Activity of Ser-Peptide Antibody Antibody and complement mediated bactericidal activity was conducted according to the assay of Fusco et al., "Bactericidal activity elicited by the beta C protein of Group B streptococci contrasted with capsular polysaccharides," Adv. Exp. Med. And Biol., 418: 841-845 (1997). A comparison was made with 20 killing by: (1) antibody to native CP3 protein, (2) Ser-peptide antibody purified over tetanus column, and (3) Ser-peptide antibody purified over tetanus then Protein G columns. The peptide PDVPKLPDVPKLPDVPKLPDAPKL (SEQ ID NO: 48) was used to inhibit bactericidal activity of these Ser-peptide antibodies. GBS Colony Lifts to Probe Ser-Peptide Antibody Binding 25 GBS colonies were lifted from chocolate agar plates onto round nitrocellulose sheets, washed in PBS-Tween®, then incubated for 1.5 hours with WO 00/15760 PCT/US99/21643 -60 either tetanus purified Ser-peptide antibody at 1:10,000 and 1:20,000 dilutions or tetanus and Protein G purified Ser-peptide antibody at 1:3000 and 1:6000 dilutions. After several washes in PBS-Tween®, the nitrocellulose sheets were incubated for 1.5 hours with goat anti-rabbit alkaline phosphatase conjugate, then 5 developed using NBT/BCIP substrate (Blake, M.S. et al., "A rapid sensitive method for detection of alkaline phosphatase conjugated anti-antibodies on western blots," Anal. Biochem., 136: 175-179 (1984)) after washing in substrate buffer (0.1 M Tris/0.1M NaCl/1 mM MgCl 2 ). Results and Discussion 10 Coupling a Peptide to a Carrier Protein It was necessary to citraconylate the lysine residues on the Ser-peptide to prevent coupling of these peptides to each other rather than to the tetanus toxoid. After dialysis, ELISA indicated that the 52.2 antibody recognized the Tetanus Ser-peptide conjugate. 15 Ser-Peptide Antibody The ELISA results of the Ser-Peptide antibody to the 8-mer mimotope sequences of CP showed that the antibody recognized the sequence PDVPKLPD (SEQ ID NO: 50) solely on the block of pins representing the carboxyterminal 20 portion of Cp. Several sequences located towards the N-terminal portion of C[ also indicated positive signals in the ELISA. It was suspected that antibodies raised to the tetanus-conjugate cross-reacted with the N-terminal sequence of CP and this reactivity was eliminated by passing the anti-serum over the tetanus affinity column.

WO 00/15760 PCT/US99/21643 -61 Purification of Ser-Peptide Antibody Antibody binding to the N-terminal sequences was eliminated once the antiserum passed through the tetanus column. Antibody bound specifically to the sequence PDVPKLPD (SEQ ID NO: 50). Additional purification of Ser-peptide 5 antibody on a Protein G column further purified IgGs and maintained affinity to the sequence PDVPKLPD (SEQ ID NO: 50). Bactericidal Activity of Ser-Peptide Antibody Fifty percent survival of bacteria occurred at antiserum dilutions greater than 1:5500 for 52.2 antibody to native CP3 (FIG. 12). Similarly, 50% survival 10 occurred at dilutions of approximately 1:8000 for Ser-peptide antiserum, regardless of passage over the tetanus column or not (FIG. 13). Interestingly, the antiserum purified over tetanus and protein G columns failed to elicit bactericidal activity. This phenomenon may have been attributed to structural changes in antibody due to the acidic conditions used to elute these proteins during 15 purification on Protein G column, thus preventing complement activation. GBS Colony Lifts to Probe Ser-Peptide Antibody Binding FIG. 14 depicts Ser-peptide antibody binding on GBS Ib cells using antiserum purified by tetanus column alone and tetanus then Protein G columns. 20 In either case, antibody binding occurred for both preparations notwithstanding results indicating the inability of Protein G purified antibody to kill GBS. Summary A synthetic peptide sequence from CP protein was coupled to tetanus toxoid and elicited an antibody response in rabbits vaccinated with the conjugate. 25 After purifying the anti-serum over a tetanus column or tetanus then Protein G WO 00/15760 PCT/US99/21643 -62 affinity columns, the antibody bound specifically to the sequence PDVPKLPD (SEQ ID NO: 50) as measured by ELISA on mimotope pins. Antibody and complement mediated bactericidal activity of this anti serum was measured and compared to anti-serum directed towards C[ 5 holoprotein. Anti-serum to the CP peptide purified over the tetanus column demonstrated killing in bactericidal assays similar to killing by anti-serum to CP holoprotein. It is believed that antibodies infiltrate the peptidoglycan layer of this Gram positive organism during cell division to allow complement activation resulting in cell lysis. 10 Example 13 Ser-Peptide Antibody Binding Introduction Many cell wall bound proteins of Gram positive bacteria have some common characteristics. The most notable one is the presence of a LPXTG 15 sequence close to the carboxy terminal portion of these proteins. The motif has been well characterized and it is through this sequence that the cell wall proteins become covalently bound to the cell wall, i.e., an enzyme cleaves the sequence between the T-G and then proceeds to couple the protein to the peptide cross bridges of the peptidoglycan through the carboxyl group on the threoninyl 20 residue. Another common feature of these cell wall proteins of Gram positive organisms is that the area just prior to the LPXTG sequence toward the amino terminus, contains numerous prolyl residues. Prolyl residues, because of their unique structure, cause kinks in proteins. Thus, this high prolyl region on the cell wall proteins has been hypothesized as an area that weaves in and out of the 25 peptidoglycan layer. However, no direct evidence for this theory has been presented. It is exactly this region on the CP protein where the anti-Ser-peptide antibodies bind and cause cell death in the presence of complement. As stated previously, a theory of how this unusual activity could occur was that the antibody bound to this region of CP during cell division at the septal planes where WO 00/15760 PCT/US99/21643 -63 the peptidoglycan is being broken down and remodeled. To obtain further evidence for this theory, confocal microscopy and FACScan analysis were used to determine where and when the antibodies to the Ser-peptide bound to the streptococcal organisms expressing the CD protein. 5 Materials and Methods Confocal Microscope (CLSM) Analysis: CLSM was used to study the binding of the rabbit antibody, anti-Ser-peptide to Group B streptococcus purified over a tetanus column. Bacterial cells of Group B streptococcus, strain H36B, were grown in Todd-Hewitt broth and samples taken at different stages in the growth. 10 The stage of growth was determined by measuring the optical density of the sample at 600 nm. The cells were then fixed in 2% paraformaldehyde and aliquots were dried onto microscope slides. The slides were blocked with 5% normal goat serum. Samples were incubated with a 1:500 dilution of rabbit antibody, anti-Ser-peptide for two hours, washed and incubated with a FITC 15 conjugated goat antiserum to rabbit IgG (Molecular Probes, Eugene, OR). The sample was counterstained with ethidium bromide. The samples were covered with coverslips with Vectashield® mounting medium (Vector Laboratories, Inc., Burlingame, CA) and examined by dual wavelength laser in a BioRad 1024 CLSM. 20 FACScan Analysis: FACScan analysis was performed on samples of bacterial cells obtained at different stages of growth using a Becton-Dickenson FACScan (Research Triangle Park, NC). Bacterial cells of Group B streptococcus, strain H36B, were grown in Todd-Hewitt broth and samples taken at different stages in the growth. The stage of growth was determined by measuring the optical density 25 of the sample at 600 nm. The bacterial cells were fixed in 2% paraformaldehyde, washed and incubated either with the anti-Ser-peptide antiserum or with the tetanus-absorbed pre-bleed serum for one hour. Cells were washed three times and incubated with a FITC-conjugated goat antiserum to rabbit IgG (Molecular WO 00/15760 PCT/US99/21643 -64 Probes, Eugene, OR) for one hour. Bacterial cells were washed three times in PBS, counterstained with ethidium bromide, and raised to a volume of 1 ml in diluent 2 buffer (J & S Medical Associates, Farmingham, MA). The data was analyzed using the Flo-Jo program (Tree Star Corp., San Carlos, CA). 5 Results Confocal Microscope (CLSM) Analysis: The CLSM analysis demonstrated that the anti-Ser-peptide bound to small areas between streptococcal cells. The antibody bound more to those cells actively dividing, i.e., during log-phase growth samples, and less when the culture entered into stationary phase as will 10 be further shown vide infra. These data further support the idea that the Ser peptide antibody only gains access to this area on the CO protein in areas where the peptidoglycan is in a state of flux, i.e., the divisional septal plane. In no field were bacteria observed completely covered with the Ser-peptide antibody. FACScan Analysis: The data for the FACScan analysis can be seen in FIG. 15. 15 The x-axis is the time elapsed since the beginning of the culture when the bacterial samples were taken. The y'-axis is the OD600 of the samples representing the number of bacteria within the sample. The y 2 -axis is the amount of green fluorescence, i.e., the Ser-peptide antibody binding, relative to the red fluorescence, the ethidium bromide staining all the bacteria. As can be seen, at 20 the beginning of the culture, very little of the Ser-peptide antibody binds to the cells. This is a time when stationary phase bacteria from an overnight culture are diluted into fresh culture media. By 2 hours, the streptococci begin to divide and cultivate as demonstrated by an increase in OD600 and, at the same time, an increase in the amount of bound ser-peptide antibody is seen. The peak of 25 antibody binding coincides with the most active part of the culture during log phase, between 4 to 6 hours. After 8 hours, the culture enters into stationary phase where bacterial division slows down and stops. The bacteria sampled in stationary phase did not bind the Ser-peptide antibody and resembled those WO 00/15760 PCT/US99/21643 -65 organisms found in the beginning of the culture. These data further confirm that the original theory was correct; that the epitope to which the Ser-peptide antibody binds is only exposed during active cell division and only in relatively small areas, i.e., the septal plane of these dividing cells. Such septal division areas 5 would also bare the cell membrane of the Gram positive organism, allowing the complement components, activated by the Ser-peptide antibodies, unhindered access to the bacterial cell membrane to complete the lysis of the cell. Although the foregoing refers to particular preferred embodiments, it will be understood that the present invention is not so limited. It will occur to those 10 of ordinary skill in the art that various modifications may be made to the disclosed embodiments and that such modifications are intended to be within the scope of the present invention, which is defined by the following claims. All patents, patent applications, and publications cited herein are incorporated by reference herein in their entirety.

Claims

1. A process for obtaining a substantially pure CP protein or fragment and/or mutant thereof, comprising: (a) obtaining the CP protein in cell extracts; 5 (b) subjecting the CP3 protein to ion-exchange chromatography and collecting the CP protein-containing fractions; (c) pooling and diluting the C3 protein-containing fractions; and (d) subjecting the diluted CP protein-containing fractions to 10 ligand-affinity chromatography and collecting the fractions; whereby substantially pure CP protein or a fragment and/or mutant thereof is obtained.

2. The process of claim 1, wherein the ion-exchange chromatography is performed utilizing an anion-exchange medium comprising a 15 trimethylaminomethyl group.

3. The process of claim 1, wherein the ligand-affinity chromatography is performed utilizing an ligand-affinity medium comprising a heparin ligand.

4. The process of claim 1, wherein the CP protein or fragment and/or 20 mutant thereof is subjected to chromatography in a buffer containing about 5% of a zwitterionic detergent.

5. The process of claim 1, wherein the pooled fractions from the ion exchange chromatography containing the CP protein or a fragment and/or mutant thereof are diluted approximately three-fold with buffer containing about 10% to 25 about 20% ofa zwitterionic detergent before the ligand-affinity chromatography. WO 00/15760 PCT/US99/21643 -67

6. The process of claim 1, wherein the CP protein or fragment and/or mutant thereof is eluted during the ion-exchange or ligand-affinity chromatography medium by applying an eluant comprising a salt gradient.

7. The process of claim 6, wherein the eluant comprises about 0.5% 5 of a zwitterionic detergent.

8. The process of claim 1, wherein the C[ protein or fragment and/or mutant thereof is obtained from bacterial cells which are transfected with nucleotide sequences encoding the CP protein or a fragment and/or mutant thereof, wherein said cells overexpress the CP protein or a fragment and/or 10 mutant thereof.

9. The process of claim 8, wherein the CP protein and/or a fragment or mutant thereof is obtained by: (a) disrupting the cells; (b) precipitating non-proteinaceous material from said cells by 15 adding ethanol/CaCl 2 to a concentration of about 20% (v/v) ethanol/about 0.1 M CaCl 2 ; (c) removing the precipitated non-proteinaceous material to give a solution; (d) precipitating protein from the solution by adding ethanol 20 to a concentration of about 80% (v/v) and collecting the precipitated protein; and (e) resuspending the precipitated protein in a buffer solution containing from about 1% to about 10% of a zwitterionic detergent.

10. The process of claim 1, wherein CP protein is obtained from bacterial cells which naturally produce the CP protein. 25

11. The process of claim 10, wherein the CP protein is obtained by: (a) boiling the bacterial cells in a buffer containing from about 1% to about 10% of a zwitterionic detergent to give a solution; WO 00/15760 PCT/US99/21643 -68 (b) cooling the solution in an ice bath; (c) precipitating non-proteinaceous material from said cells by adding a cold solution of ethanol/CaCl 2 to give a concentration of about 20% ethanol/about 0.1M CaCl 2 ; 5 (d) removing the precipitated non-proteinaceous material to give a solution; (e) precipitating protein from the solution by adding ethanol to a concentration of about 80% (v/v) and collecting the precipitated protein; and (f) resuspending the precipitated protein in a buffer solution 10 containing from about 1% to about 10% of a zwitterionic detergent.

12. An isolated nucleic acid molecule coding for a protein fragment or peptide comprising a proline-rich region, wherein at least every third residue is proline, and wherein antibodies raised against said protein fragment or peptide are bactericidal to Gram positive bacteria with complement alone. 15

13. The isolated nucleic acid molecule of claim 12, wherein said protein fragment or peptide comprises a continuous (repeated) amino acid sequences having the formula -[P-Yi-Y2-P-Y,-Y2]r- or -[Y1Y2-P-Y1-Y2-P]r-, wherein Y 1 represents either an acidic or basic residue, Y 2 represents a neutral amino acid, and r is an integer from one to five. 20

14. The isolated nucleic acid molecule of claim 12, wherein said protein fragment or peptide comprises a continuous (repeated) amino acid sequence having the formula -[P-D-Y3-P-K-L],- or -[K-L-P-D-Y3-P],-, wherein Y 3 represents V or A, and r is an integer from one to five.

15. The isolated nucleic acid molecule of claim 12, wherein said 25 protein fragment or peptide comprises a continuous (repeated) amino acid sequence having the formula -[S-P-K-Y 4 -P-E-A-P-Y 5 -V-P-E]r-, wherein Y 4 represents T or A, Y 5 represents H or R, and r is an integer from one to five. WO 00/15760 PCT/US99/21643 -69

16. An isolated nucleic acid molecule coding for a protein fragment or peptide having the formula Y-X-Z, wherein X represents at least eight contiguous amino acid residues between amino acids 827 and 1028, inclusive, of FIG. 1 (SEQ ID NO: 2), Y represents hydrogen or the N-terminal amino acid 5 sequence of FIG. 1 (SEQ ID NO: 2) that is bound to X, or an N-terminal fragment and/or mutant thereof, and Z represents hydrogen or the C-terminal amino acid sequence of FIG. 1 (SEQ ID NO: 2) that is bound to X, or a C-terminal fragment and/or mutant thereof, and wherein said protein fragment or peptide does not bind to the Fc region of human IgA immunoglobulin, with the proviso that the protein 10 is at least one of an N-terminal or C-terminal fragment of the amino acid sequence of FIG. 1 (SEQ ID NO: 2), with the proviso that said protein fragment or peptide is not the approximately 38 kD polypeptide secreted by the Group B streptococcus strain HG 806, and with the further proviso that at least one of amino acids 1-164 of FIG. 1 (SEQ ID NO: 2), if present in Y, is non-wild-type. 15

17. The nucleic acid molecule of claim 16, wherein Y does not comprise at least amino acids 1-176 of SEQ ID NO: 2.

18. The nucleic acid molecule of claim 16, wherein Z comprises at least amino acid 901 of SEQ ID NO: 2.

19. The nucleic acid molecule of claim 16, wherein X is selected from 20 the group consisting of PPKTPDVP (SEQ ID NO: 32), PDVPKLPD (SEQ ID NO: 33), KLPDVPKL (SEQ ID NO: 34), VPKLPDVP (SEQ ID NO: 35), KLPDAPKL (SEQ ID NO: 36), APKLPDGL (SEQ ID NO: 37), ETPDTPKI (SEQ ID NO: 38), RTVRLALG (SEQ ID NO: 39), GGGTVRVF (SEQ ID NO: 40), SPKTPEAPKIPEPPKTPDVP (SEQ ID NO: 41), 25 PEAPKIPEPPKTPDVPKLPD (SEQ ID NO: 42), KIPEPPKTPDVPKLPDVPKL (SEQ ID NO: 43), PPKTPDVPKLPDVPKLPDVP (SEQ ID NO: 44), PDVPKLPDVPKLPDVPKLPD (SEQ ID NO: 45), WO 00/15760 PCT/US99/21643 -70 KLPDVPKLPDVPKLPDAPKL (SEQ ID NO: 46), and PDVPKLPDVPKLPDVPKLPDAPKL (SEQ ID NO: 47).

20. The nucleic acid molecule of claim 16, wherein X represents any eight to twenty-one contiguous amino acid residues between amino acids 828 and 5 1027, inclusive, of SEQ ID NO: 2.

21. The nucleic acid molecule of claim 16, wherein X represents the amino acid sequence between amino acids 828 and 1027, inclusive, of the sequence shown in SEQ ID NO: 2.

22. A vector comprising the polynucleotide molecule of claim 12 or 10 claim 16.

23. A host cell transformed with the vector of claim 22.

24. A protein fragment or peptide encoded by the nucleic acid molecule of claim 12 or claim 16.

25. A protein fragment or peptide-polysaccharide conjugate 15 comprising the protein fragment or peptide of claim 24.

26. A vaccine comprising at least one protein fragment or peptide of claim 24, together with a pharmaceutically acceptable carrier.

27. The vaccine of claim 26, wherein said protein fragment or peptide is conjugated to a polysaccharide selected from the group consisting of Group B 20 streptococcal capsular polysaccharide types Ia, II, III, and V.

28. A combination vaccine comprising at least two protein fragment or peptide-polysaccharide conjugates of claim 27. WO 00/15760 PCT/US99/21643 -71

29. A method of inducing an immune response in an mammal, comprising administering a vaccine comprising at least one protein fragment or peptide according to claim 24, together with an pharmaceutically acceptable carrier, in an amount sufficient to induce an immune response in a mammal.