WO1993024000A1

WO1993024000A1 - Methods for obtaining and purifying cell-free pneumococcal surface proteins from s. pneumoniae and use thereof

Info

Publication number: WO1993024000A1
Application number: PCT/US1993/005191
Authority: WO
Inventors: Janet Yother
Original assignee: Janet Yother
Priority date: 1992-05-29
Filing date: 1993-05-28
Publication date: 1993-12-09
Also published as: AU4400293A

Abstract

Cultures of S. pneumoniae can be caused to express substantially cell-free surface proteins, in particular PspA, into a medium from intact cells by exposing the culture to a medium comprising from about 0.5 % to not more than about 3 % choline either during growth of a culture or thereafter, or by growing a culture containing about 0.005 to about 0.1 % ethanolamine in the absence of choline. The supernatant medium may be used directly for preparation of immunogens, undesired materials may be removed before use, or the protein in the supernatant may be separated and purified as needed.

Description

Methods for Obtaining and Purifying Cell-free Pneumococcal Surface Proteins from S. pneumoniae and Use Thereof

FIELD OF THE INVENTION

The present invention concerns methods for causing

10 Streptococcus pneumoniae strains to release substantial quantities of their surface proteins, in particular pneumo¬ coccal surface protein A (PspA) , into a medium and methods for purification and recovery of the surface protein so produced. The surface protein may be used with or without

_J purification to protect mammals from otherwise fatal pneumococcal infections.

BACKGROUND OF THE INVENTION Streptococcus pneumoniae is an important cause of otitis media, meningitis, bacteremia and pneumonia. De-

20 spite the use of antibiotics and vaccines, the prevalence of pneumococcal infections has declined only slightly in the last quarter century.

It is generally believed that immunity to S. pneumoniae may be mediated by antibodies specific against

?5 the polysaccharide capsule of the pneu ococcus. Neonatal and young children appear to respond poorly to polysaccharide antigens and can have repeated infections involving the same capsular serotype. An approach to immunizing against encapsulated bacteria is to conjugate

30 the capsular polysaccharide antigens to protein to make them immunogenic. This .approach has been successful with Haemophilus influenzae b. There are over eighty capsular serotypes of S. pneumoniae known, of which some twenty- three account for most infections. Therefore, for the use 35 of a pneumoccal polysaccharide-protein conjugate to be successful, it would be necessary to make the capsular types responsible for most infections adequately immunogenic. Indeed, not all of the twenty-three polysaccharides included in the vaccine presently available

40 are adequately immunogenic, even in adults. An alternative would be to identify protein antigens that would elicit protective immune responses. Such proteins might serve as vaccine in and of themselves, or might be used in conjunction with protein-polysaccharide conjugates or as carriers for polysaccharides. cDaniel et al. (I), J.Exo.Med. 160:386-397 (1984) describes production of hybridoma antibodies which recognize cell surface polypeptide(s) on S. pneumoniae and the protection of mice against certain strains of encapsulated pneumococci by those antibodies. This surface protein antigen has been termed "pneumococcal surface protein A" or "PspA." Studies to characterize PspA are described in McDaniel et al. (II) , Microbial Pathoσenesis 1:519-531 (1986). These studies found that there was considerable diversity in the PspA molecules found in different strains and differences in the epitopes recognized by different antibodies. Further efforts described in McDaniel et al. (Ill), J.Exp.Med. 165:381-398 (1987) showed that immunization of mice with non- encapsulated pneumococci which expressed PspA protected them from subsequent fatal infection with pneumococci, but that when isogenic pneumococci which lacked PspA were used, no protection was observed. McDaniel et al. (IV), Infect. Iirunun. .59.:222-228 (1991) described immunization of mice with a recombinant full length fragment of PspA which elicited protection against strains of pneumococcal capsular types 6A and 3. Crain et al. , Infect. Immun. 58.:3293-3299 (1990) describe a rabbit antiseru that detects PspA in 100% of clinical and laboratory isolates of S. pneumoniae strains. Fifty-seven isolates reacted with seven monoclonal antibodies to PspA exhibited thirty-one different patterns of reactivity.

PspA protein type is independent of capsular type. It appears that genetic mutation or exchange in the environment has allowed development of strains which are highly diverse with respect to capsule, PspA, and possibly other molecules as well. PspAs from various strains vary in their molecular weights, ranging from 67 to 99 kD. These differences are inherited stably and do not result from protein degradation. While PspA varies in structure between different pneumoccal strains, all PspAs exhibit 5 numerous cross-reactions, which suggests that sufficient common epitopes may exist to allow a single PspA or a small number to elicit protection against a large number of S. pneumoniae strains.

Crain et al., op. cit. , have isolated full length PspA

"■o from S. pneumoniae strain Rxl from a cell-wall extract.

McDaniel et al., McDaniel IV, described a recombinant full- length PspA and its expression in E. coli. The supernatants of lysates from the modified E. coli were used for immunizations without further purification.

15 More recently the inventor and her coworkers, Yother et al., J.Bacteriol. 174:610-618 (1992), have described the use of insertion-duplication mutagenesis to prepare bacteria which express soluble truncated mutants of PspA which constitute from 20 to 80% of the complete molecule.

20 The PspA expressed in E. coli were localized to the periplasmic space from which they could be released by standard osmotic shock procedures. The isolated truncated segments contain immunoprotective epitopes of the protein, but lack the region which anchors the protein to the cell

25 membrane. A purified 43 kD segment was shown to protect mice from type 3 S. pneumoniae by Talkincrton et al.. Infect.Immun. 59:1285-1289 (1991). Although this complex procedure may be capable of supplying material suitable for preparing immunogenic material(s), a simpler method of

30 preparing PspA would be desirable, whether the PspA from only one, several, or even most strains is required to obtain the protection sought.

S. pneumoniae appears to be unique among gram positive bacteria in having surface proteins, in particular PspA,

35 which do not contain a common motif that includes a 16 to 20 amino acid hydrophobic C-terminal membrane-spanning region that, together with a short (5 to 6 amino acid) charged tail, anchors the protein to the cell. One such protein is PspA.

In previous studies: a) Rane and SubbaRow, J. Bacteriol. 40: 695-704

5 (1940) , have shown that S. pneumoniae requires choline for growth; b) the level of choline required is 0.00005%; c) Briese et al., op. cit. , and Giudicelli and Tomasz, in Nombela, C. ed. , Microbial Cell Wall Synthesis and

^■• Autolysis, pp. 207-212, Elsevier (1984), have shown that growth in high levels of choline (1-2%) prevents the autolysis of cells which normally occurs at the stationary phase of growth (the high level of choline apparently inhibits autolysin by competing with cell wall-associated

15 choline as a binding site) ; d) Badger, J.Biol.Chem. 153: 183-191 (1944) , has shown that ethanolamine can be substituted for choline as a growth requirement, but autolysis does not occur at the stationary phase because there is no choline present for

20 the autolysin to bind (Tomasz, Proc. Natl. Acad. Sci. USA .59:86-93 (1968) and Nature 227: 138-140 (1970). e) Briese et al., in Hakenbeck, R. , J.V. Holtje and H. Labischinski ed. , The Target of Penicillin, pp. 173-178, Walter de Gruyter (1983) and Diaz et al., J.Biol.Chem. 264:

25 1238-1244 (1989) , have shown that autolysin remains cell associated during growth in high or low levels of choline or in ethanolamine.

Sanz et aL.FEBS Lett. 232: 308-312 (1988), have shown that a number of tertiary amines and quaternary ammonium

30 compounds elicit choline-like responses from a surface- bound pneumococcal amidase and cause in vitro activation of autolysin. These choline analogs comprise compounds which in the ammonium form have the general structure:

R

35 \Φ

R^,~ N - R

/

R wherein each R is individually selected from the group com¬ prising methyl, ethyl, hydroxyethyl and esterified hydroxy- ethyl radicals; and R' is methyl or hydrogen. Tomasz et al., J.Supramol. Struct. 3: 1-16 (1975), have shown that 5 diethylaminoethanol (DEAE) is incorporated into cells of S. pneumoniae grown in a choline-free culture to which DEAE has been added. Therefore it is believed that the choline analogs of Sanz et al. will be incorporated in place of choline in cells of S. pneumoniae grown when added to

""i culture media which are substantially free of choline. Briese and Hakenbeck. Eur.J.Biochem. 146:417-427 (1985) , have shown that at least four cell-associated proteins c' S. pneumoniae bind choline. One of the four was identified as autolysin. Subsequently Garcia et al.,

15 Biochem. Biophvs. Res. Comm. 158:251-256 (1989) have identified another choline-binding molecule as an autolytic glycosidase. The inventor and a coworker, J. Bacteriol. 174:601-609 (1992), have found that the C-terminal region of PspA is similar to that region of the pneumoccal lysins

20 and it, like the autolysin, binds to choline residues found in the membrane-attached lipoteichoic acid (LTA) . However, in distinct contrast to the present invention, autolysin has never been observed to be released by intact cells under any conditions. Indeed, recent studies have shown

25 that autolysin is localized intracellularly and binds LTA only upon cell lysis.

SUMMARY OF THE INVENTION

It has been found that cultures of S. pneumoniae can be caused to express substantially cell-free surface proteins, in particular PspA, into a medium from intact 5 cells either during growth of a culture or thereafter. The supernatant medium may be used directly as an immunogen or may be used for preparation of more purified immunogens, undesired materials may be removed before use, or the PspA in the supernatant may be separated and purified as needed.

^■"^■ DETAILED DESCRIPTION OF THE INVENTION

It has been found that pneumococcal surface protein A (PspA) is bound to the S. pneumoniae cell by interaction with choline residues of the cell membrane and cell wall teichoic acids and, along with any other surface proteins

15 bound to choline residues (hereinafter referred to as

"choline-bindable proteins") , may be freed from the walls of intact cells of S. pneumoniae by exposing the cells during growth or thereafter to a medium comprising from about 0.5% to about 3% or more of choline or a choline ana-

20 log. The surface protein-containing medium is then sepa¬ rated from the cells by any convenient means such as decantation, etc. and may then be freed of cells and cellular debris by means such as filtration or centrifugation in the usual manner. The resulting solution

25 of choline-bound proteins (i. e. proteins which had been bound to choline residues in the cells) may then be used as such as a vaccine or to induce the formation of antibodies in animals to provide passive protection against pneumococcal infection, or they may be further treated as

30 described beyond before use.

A particular method for practice of the invention is to cause S. pneumoniae to express surface proteins which are ordinarily bound by interaction with choline residues of the S. pneumoniae cell membrane or wall by growing the

35 S. pneumoniae in a medium comprising from about 0.5% to not more than about 3% choline or a choline analog (most strains begin to exhibit poor growth at levels of 3% or more) , preferably from about 1% to about 2% choline. Under such conditions substantially cell-free surface choline- bindable protein, in particular PspA, are released into the supernatant culture medium. The amount of PspA present in cells and the supernatant over a range of choline concentrations was studied using S. pneumoniae strain Rxl with the following results;

% choline cell-associated PspA cell-free PSPA about 0.05% about 98% about 2% 0.1% about 50% about 50%

0.5% about 5% about 95% about 0.87% not more than 2% not less than 98%

Western blot testing showed that autolysin remained in the cells under each of these conditions. The release of surface bound proteins by high concentrations of choline (2%) in the culture medium as shown by the presence of PspA was tested for eight strains of S. pneumoniae which repre¬ sented different capsular and PspA serotypes. strain capsular PspA PSPA MW % cell-free PspA serotvpe serotype *

25 84kD 98

1 92kD 90

13 92kD 70

84kD 50

20 94kD 80

0 86kD 98

33 86kD 70 * based on PspA typing in Crain et al. , op. cit. ** Rxl is a non-encapsulated derivative of D39

As can be seen, at least 50% of the contained PspA was found in the supernatant in every instance, and for five of the strains from 80-98% was found in the supernatant.

The choice of growth medium is not critical, but the use of a chemically defined medium (CDM) such as that de¬ scribed by van de Rijn and Kessler, Infect. Immun. 27:444- 448 (1980) , is preferred in order to reduce the amount of components which may be co-isolated with the surface proteins or which may have proteolytic activity which would degrade the surface proteins. Other culture conditions may be chosen from those well known to the art. When grown under these conditions S. pneumoniae releases choline-bound surface proteins into the supernatant. The time to reach saturation is dependent upon the strain of S. pneumoniae and the culture conditions selected. At 37^βC the doubling 5 time in CDM with 2% choline ranges from about 30 minutes to 1 hour.

Because the cells usually settle to the bottom of the culture vessel, the supernatant may be separated from the bulk of the cells by any convenient method such as

" -^■ decantation, etc. and may then be freed of cells and cellular debris by means such as filtration or centrifu- gation in the usual manner to yield a substantially cell- free PspA. The growth may also be carried out by continuous or batch-wise addition of new culture medium and

15 withdrawal of supernatant containing the expressed proteins in a chemostat or similar means.

It may be desirable to remove the choline from the supernatant after it has been separated from the cells. Removal may be accomplished for example by dialysis against

20 a buffer as is well-known in the art. A suitable buffer is a phosphate buffered saline with a pH of about 7.2.

If PspA of greater purity is desired, purification may be accomplished by any of the methods known to the art, more particularly those in the references cited above.

25 A preferred method for purification of solutions of surface proteins such as PspA which bind to choline is to adsorb the proteins on an ion exchange column modified with a tertiary amine such as diethylaminoethanol (DEAE) . This method is useful regardless of the source of the choline-

30 bindable surface protein. Other tertiary amine choline analogs included in the previous listing derived from Sanz et al., OP. cit.. bound to a suitable support may also be used. If the choline-bindable surface protein solution contains choline, the choline should be removed by a method

35 such as dialysis before passing the solution through the column to adsorb the protein. The column support does not appear to be critical. A preferred method is the use of an equilibrated DEAE-cellulose column such as that described by Sanz et al. , OP. cit. A suitable method for equilibration is the use of a solution of about 50mM Tris and about 0.15M sodium chloride, having a pH of about 7.4. 5 The presence of the anchor portion of a choline-bindable surface protein such as PspA causes it to be strongly adsorbed on the amine-modified column, which allows other materials in the PspA solution to be washed out of the column readily with a saline buffer such as that used to

"Ϊ equilibrate the column. The salt content of the buffer is not critical. Washes with low to high salt concentrations of from about 0.15M to about 1.5M are beneficial because other proteins will be eluted more readily, thus yielding a higher-purity choline-bindable surface protein. The

15 choline-bindable surface protein such as PspA may then be easily and substantially quantitatively washed from the column using an aqueous wash solution containing from about 0.5% to about 4%, preferably from about 1.5% to about 2.5%, choline. The minimum amount of choline solution needed to

20 substantially completely elute the protein may be as little as 1/lOOth the original culture volume or less, but ultimately will depend on the strain used to grow the protein and the concentration of choline used in the wash solution. The minimum amount is not critical, and can

25 easily be determined by routine experimentation. If choline will not interfere, the resulting protein solution may then be used as such, or if desired or necessary, it may be freed of choline by means such as dialysis against a saline buffer, as described above.

30 In another embodiment of the invention a culture of S. pneumoniae is grown in a medium containing only a small amount of choline, preferably less than about 0.05%. The choice of medium is not critical. Suitable media include CDM with added choline or Todd-Hewitt plus yeast extract

35 (Difco Laboratories, Detroit, Michigan) . The cells are harvested by any suitable means known to the art such as centrifugation at 10,000 RPM for 10 to 20 minutes and decantation of the supernate. The harvested cells may be washed first with small amounts of water such as twice with 1/lOth the amount of the original culture volume, and then are washed with from about 1/20th to about one culture

5 volume or more of aqueous solution of from about 1% to about 10%, preferably about 1.5% to 2.5% choline or choline analog to extract choline-bindable proteins including PspA from the cells. The use of choline is a preferred method. The wash solution contains most of the choline-bindable

~ -> protein contained in the original culture. When Rxl and

D39 strains were used a one volume wash of 2% choline con¬ tained about 98% of the contained PspA, and when the A66 strain was used the recovery with such a wash was about 70%. Western blot tests showed that no autolysin was

15 released from the cells by this procedure. As before, the extract may be used without further purification or may be further purified as set out above, keeping in mind that the choline or choline analog should be removed from the solution before utilizing ion-exchange to purify the PspA.

20 In a preferred form of the invention a culture of S. pneumoniae is grown in a CDM in the substantially complete absence of choline or choline analogs in the presence of from about 0.005 to about 0.1%, preferably from about 0.02% to about 0.04% of ethanolamine. The culture utilizes

25 ethanolamine in the absence of choline, but apparently the choline-bindable proteins cannot be substantially bound by the cells in the absence of cellular choline residues and are expressed into the supernatant. The supernatant may be harvested as set out above to yield a substantially cell-

30 free PspA. Tests showed once again that autolysin was not released from the cells by this procedure. If ethanolamine will not interfere, the supernatant may then be used directly, or if desired or necessary, ethanolamine may be removed by any suitable means such as dialysis. The

35 supernatant may also be purified for further use by adsorbing the choline-bindable proteins on an amine- odified ion exchange column, eluting them with a choline wash and purifying the choline wash as set out above. Application of ethanolamine supernatants to such columns does not require pretreatment such as dialysis.

The presence of choline-bindable protein(s) at all 5 stages of the preparation of solutions of surface-bound proteins is readily determined by methods well-known to the art such as the Western blot or silver staining of protein gels. The Western blot is particularly useful for detecting PspA. Silver stain of protein gels may also be """■ used to determine PspA, but the PspA of some strains such as Rxl give only faint yellowish stains.

The amount of PspA in supernatants of cultures grown in high choline appears to be dependent on the strain of S. pneumoniae used, but in most cases appears to be about 2 to 15 5% of the total protein. In the supernatant from one etha¬ nolamine culture which was studied the PspA constituted about 25% of the total protein. When a PspA solution was prepared by washing a low-choline culture of each of two strains, PspA was about 50% of the total protein. 20 The presence of PspA was determined by using silver staining and antibodies specific to PspA.

If in vivo activity is considered to be an ability to elicit a protective response, both full length and truncated molecules of PspA are known to elicit responses, 25 but specific differences have not been examined in detail because prior to the present invention, full-length PspA has been difficult to obtain. The PspA prepared according to the present invention from non-encapsulated and encapsulated pneumococci was able to elicit a high level of 30 protection in mice against otherwise fatal pneumococcal infections. Unlike fragments comprising the N-terminal half of PspA which must be administered with adjuvants, immunogenicity to elicit effective antibody response is demonstrated with the PspA preparations of the invention 35 whether or not an adjuvant is used. Whole PspA prepared according to the invention is immunogenic when injected as the PspA-containing isolate which may have undergone further purification or may be administered as the substantially cell-free supernatant. In mice, two injections of PspA are required to elicit a significant antibody response. The isolated PspA-containing material

5 remains immunogenic after lyophilization. If whole supernatant is used for immunization, the cost of purification may be avoided. Since the pneumococci grown in ethanolamine do not autolyze, the supernatant should not contain significant amounts of inflammatory cell-wall

" -- products. When rough (unencapsulated) pneumococci are used to produce the PspA there should be little if any free capsular polysaccharide in the supernatant. The full- length PspA may be expected to elicit a broader response than its fragments because of the presence of multiple

15 conserved epitopes. The present invention makes available PspA from strains which have not been tested previously because no truncated or full PspAs were available. Such full PspAs should elicit more protection than the two truncated PspAs and one full-length clone which were

20 available until now. The lyophilized PspA-containing materials present an advantage in that they are more easily stored and shipped in closed vials. Lyophilized product may then be prepared for administration by adding a suitable carrier.

25 The immunogenically effective amount of a cell-free protein of the invention may be determined by routine experimentation. It may be administered by any of the usual means by which vaccinations are accomplished, as for example, by intradermal or subcutaneous injection, or by

30 application to an abraded epithelium and the like. The substantially cell-free preparations of the invention may also be used to induce the production of antibodies in a suitable host without substantial modification. Lyophilized protein may be used with any suitable carrier

35 such as saline and the like. Although adjuvants are not necessary, they may be used in administering the protein. The foregoing and the Examples below provided herein are not exhaustive but are intended merely to illustrate the principles of the invention. Further, since numerous changes and variations will be readily apparent to those 5 skilled in the art, it is not intended that the invention be limited to the exact construction and operation described herein. Thus all suitable modifications and equivalents that may be resorted to fall within the scope of the invention. ^~ - EXAMPLES

Example 1

This Example illustrates the method used to determine the amount of choline-bindable protein (as PspA) which was expressed by the Rxl and A66 strains of S. pneumoniae. 15 Approximately 10⁵ S. pneumoniae cells were inoculated into 3ml of CDM (Hazelton Research Products, Denver, PA) to which 2% choline had been added. The culture was grown standing at 37^βC overnight (about 15 hours). One ml of res spended culture was centrifuged (about 15,000g, 2 20 minutes for Rxl, an unencapsulated strain, and 5 minutes for A66, an encapsulated strain) . The supernatant fluids were saved and the cells were resuspended in 1 ml of a sodium deoxycholate buffer made up to volume with sodium chloride-citrate (SSC) . The cells and supernatants were 25 stored at -20"C.

For protein analysis, a 20 microliter aliquot of supernatant or cell suspension was mixed with 4 microliters 5XSDS-PAGE loading buffer and boiled for 5 to 10 minutes. The sample was loaded and electrophoresed through 2 30 identical 12% SDS-PAGE gels. One gel was developed with silver stain, and the other was transferred to nitrocellulose and developed in a Western blot with an antibody specific for PspA.

For Rxl, the Western analysis showed that 98% of PspA 35 was in the supernatant fluid. The silver stain indicated that about 1% of the total protein in the supernatant was PspA. For A66, the Western analysis showed that 70% of PspA was in the supernatant fluid. Silver stain indicated that about 5% of the total protein in the supernatant was PspA.

It is important to note that silver stain does not 5 show PspA from Rxl as well as PspAs from other strains.

In subsequent work with these and other strains when only PspA in supernatant was sought, the settled cells were left on the bottom, and 1 ml of supernatant was withdrawn from the top and filtered through a 0.22 micrometer low - "^■ protein binding syringe filter. The supernatant was then ready for analysis. For larger quantities as much supernatant as possible without removing cells was withdrawn and a larger volume filter unit was used.

15 Example 2

This Example is illustrative of growth in a choline-free medium.

The procedures of example 1 were followed except that the medium was CDM plus 0.02% ethanolamine. Cultures were

20 grown of Rxl, D39, WU2, A66, BG9163, EF3296, DBLl and DBL5 strains. The Rxl culture was harvested after overnight growth. The other strains required up to 36 or 48 hours for saturation. Western analysis showed that PspA was about 25% of the supernatant from Rxl. This determination

25 was not made for the other strains. The results were as follows:

30

35

** Rxl s a non-encapsulated der vative of D39

40

Example 3 This Example illustrates growth in a low level of choline.

The procedures of Example 1 were followed using the Rxl strain, except that the choline concentration was 0.0005%, and growth was not carried beyond mid-log phase to avoid the onset of lysis. The cells were harvested by centrifugation, washed twice with a culture volume of water, and then once with a culture volume containing 2% choline to cause release of cellular PspA and any other choline-bound surface proteins. The choline wash was filtered as above to remove any cell carry-over. Analysis of the choline wash showed that PspA represented about 50% of the total protein.

Example 4

This Example illustrates purification by passage over a

DEAE column.

The Rxl strain was grown overnight using the conditions of Example 2, but using 10 ml medium. The supernatant fluid was withdrawn and applied to an ion- exchange column prepared by loading a P1000 pipet tip with 200 microliters of DEAE-Affi-Blue Gel (Biorad) . The inside diameter at the top of the packing was 0.5 cm, and at the bottom 0.25 cm. The length was 1.7 cm. The column was equilibrated with 50mM Tris, pH 7.4, plus 0.15M sodium chloride (low salt buffer) . The supernatant (about 10 ml) was loaded, and the column was washed once with 0.5 ml low salt buffer, once with buffer plus 1.5M sodium chloride (high salt buffer) , and then once with high salt buffer plus 2% choline. Fractions were collected as follows: fraction 1 = 10 ml void volume fraction 2 = 0.5 ml low salt wash fraction 3 = 1.0 ml high salt wash fraction 4 = 0.15 ml choline wash (about 1 bed volume) fraction 5 = 0.3 ml choline wash fraction 6 = 0.3 ml choline wash fraction 7 = 0.3 ml choline wash

Approximately 60 to 70% of PspA bound to the column. The remainder was mostly in fractions 2 and 3 (about equally) .

The PspA which had been bound to the column was found in fractions 4 and 5, mostly fraction 5. Recovery was about

90%, and the PspA was concentrated 30-fold. PspA appeared to represent about 70% of the total protein. The column employed in this Example was one which was on hand; a less costly DEAE-modified cellulose column should be equally suitable.

Example 5 This Example illustrates dialysis to remove choline as when it is necessary to remove choline before applying super¬ natant to a column according to Example 4, or if removal of choline after elution from a column is desired. The sample is placed in a dialysis bag with molecular weight cut-off of more than 50,000. (If lower molecular weight molecules are to be retained, the cut-off may be lowered.) Dialysis is carried out by a standard method, for example, at 4^βC in phosphate buffered saline, pH 7.2. The volume of buffer to sample is about 100:1, and the buffer is changed two or three times.

"^*ι Example 6

This Example illustrates the immunogenic activity of the substantially cell-free PspA supernatant preparations according to the invention.

Supernatants of various strains S__±. pneumoniae were

15 prepared as in Example 1, but adding 1.2% choline chloride to the CDM. CBA/N (XID) mice were given two interperitoneal injections, two weeks apart (preliminary experimentation indicated that two injections of about 0.1 ml were required to generate an immunogenic response) , of

20 0.1 ml (without adjuvant) of the supernatants diluted with saline or concentrated (for example a 10-fold concentration is "10x") to the concentrations shown in Table 1 below, and then were challenged 7 days after the second injection with an intravenous injection of 100 times the LD₅₀ of WU2 strain

25 pneumococci.

TABLE 1

Medium alone 2, 2, 2, 2

It can be seen that with a 1/3 dilution or undiluted supernatant, the presence of PspA was generally required to elicit protection. When supernatant from PspA-free cultures concentrated 10 fold or more was used, the results indicate evidence for other protection eliciting factor(s) , since extension of life and/or protection from death was observed.

Example 7

This Example illustrates the use of substantially cell-free

PspA of the invention with an adjuvant.

Supernatants of various strains S__. pneumoniae were prepared as in Example 1, but adding 1.2% choline chloride to the CDM. CBA/N (XID) mice were given two interperitoneal injections, two weeks apart, of 0.1 ml of the supernatants to which about 40 μg alum had been added as an adjuvant, and then were challenged 7 days after the second injection with an intravenous injection of 100 times the LD₅₀ of WU2 strain pneumococci. The results are shown in Table 2. TABLE 2

Strain Properties Concentration Day of providing of strain of supernatant death supernatant

>21, >21, >21

>21, >21, >21 0

2, 2, >21

Medium alone 2, 2 ^a Strains were prepared by transforming different PspA genes 15 into WG44.1. Thus each strain has a different PspA on the Rxl background. ^b Strain was derived from Rxl and has an insertion in PspA which results in production of only the N-terminal half of PspA.

20 It can be seen that all four different full-length PspAs were able to elicit protection from fatal infection with strain WU2 (type 1 PspA) . Immunization with the truncated N-terminal half of PspA was significantly less able to elicit protection.

___> Example 8

This Example illustrates the retention of immunogenic activity by isolated and lyophilized PspA of the invention.

The R36A strain of S♦ pneumoniae was cultured to produce substantially cell-free PspA as in Example 2 , but

30 using 0.03% ethanolamine. A portion of the supernatant was isolated on a choline-Sepharose column, eluted with 2% choline, dialyzed in 1/10 XID normal mouse serum to remove choline and lyophilized. Another portion was used without further treatment. Portions were diluted with saline to

35 the concentrations shown in Table 3 and 0.1 ml was injected interperitoneally into XID mice at day 0 and day 14. The mice were then challenged with an intravenous injection of 100 times the LD₅₀ of WU2 strain pneumococci at day 21.

none 18, 18, 25, 26, 29, 29, 68

It can be seen that little or no loss in PspA immunogenic activity occurred after isolation and lyophilzation when compared to untreated supernatant. The optimal dose appears to be between 1 and 1/25 of the concentration in the 0.03% ethanolamine supernatant.

Example 9

This example shows that factors in the supernatants other than PspA elicit protection of XID mice.

Supernatants of various strains S_j_ pneumoniae were prepared as in Example 1, but adding 1.2% choline chloride to the CDM. CBA/N (XID) mice were given two interperitoneal injections, two weeks apart, of 0.2 ml 10- fold concentrations of the supernatant in complete Freund's adjuvant, and then were challenged 7 days after the second injection with 100 times the LD₅₀ of strain WU2 pneumococci. The results are shown in Table 4.

Medium alone 2 (13 mice)

On comparison of the results in Table 4 and with those in earlier Examples it can readily be seen that the protective action of the proteins or protein-like substances expressed by the non-PspA-expressing WG44.1 are significantly less than that of PspA.

Example 10

This Example demonstrates that the non-PspA substance expressed by WG44.1 which elicited a lesser degree of protection than PspA in Example 9 appears to be protein or protein-like.

A substantially cell-free supernatant of a culture of strain WG44.1 (a modification of Rxl which lacks PspA), prepared according to the method of Example 1, but using 1.2% choline chloride was concentrated by ultrafiltration followed by 60% (NH₄)₂S0₄ precipitation. CBA/N (XID) mice were given two interperitoneal injections, two weeks apart, of 0.2 ml 10-fold concentrations of the supernatant in complete Freund's adjuvant, and then were challenged 7 days after the second injection with 100 times the LD₅₀ of strain WU2 pneumococci. A second group of mice was treated in the same fashion with the same supernatant preparation in adjuvant to which pronase had been added. The results are shown in Table 5.

TABLE 5 Supernatant Pay of death

WG44.1 2, 2, 5, >21 (7 mice)

WG44.1 + pronase 2, 2, 2 , 2

CDM medium only 2 (10 mice)

Claims

1. A method for producing a solution which comprises substantially cell-free choline-bindable proteins of one or more Streptococcus pneumoniae strains which comprises: exposing a culture of the Streptococcus pneumoniae to a solution which comprises at least about 0.5% choline or a choline analog during its growth stage or subsequent thereto, and separating the resulting supernatant from the Streptococcus pneumoniae cells to form said solution.

2. The method of Claim 1 wherein the Streptococcus pneumoniae is grown in a growth medium which comprises at least about 0.5% choline or a choline analog.

3. The method of Claim 2 wherein the growth medium comprises from about 1% to about 2% choline.

4. The method of Claim 1 wherein the Streptococcus pneumoniae is grown in a medium which contains not more than about 0.05% choline or choline analogs, the cells are harvested, and the harvested cells are washed with an aqueous solution comprising from about 1 to about 10% choline to form the solution of proteins.

5. The method of Claim 1 wherein the solution is further purified after separation from the cells by removing the choline or choline analog.

6. The method of Claim 1 wherein the choline-bindable protein in the solution is further purified after separation from the cells by removing the choline or choline analog, adsorbing the contained protein on an ion- exchange column, and then eluting the protein with an aqueous solution comprising from about 1% to about 3% choline to form a solution of proteins. 7. The method of Claim 6 wherein the aqueous solution of protein eluted from the ion-exchange column is further purified by removing the choline therefrom.

5 8. A method for producing a solution which comprises substantially cell-free choline-bindable protein of one or more Streptococcus pneumoniae strains which comprises: growing the Streptococcus pneumoniae in a growth medium which is substantially free of choline and choline " analogs and contains from about 0.005 to about 0.1% ethano1amine, and separating the resulting supernatant from the Streptococcus pneumoniae cells to form said solution.

15 9. The method of Claim 8 wherein the growth medium comprises from about 0.02% to about 0.04% ethanolamine.

10. The method of Claim 8 wherein the protein in the solution is further purified after removal of the cells by

20 adsorbing the protein on an ion-exchange column and then eluting the protein with an aqueous solution comprising from about 1% to about 3% choline.

11. The method of Claim 10 wherein the aqueous solution

25 eluted from the ion-exchange column is further purified by removing the choline therefrom.

12. Choline-bindable surface protein of one or more strains of S. pneumoniae bound to an ion-exchange column which

30 comprises a tertiary amine analog of choline.

13. The composition of Claim 12 wherein the choline analog is diethylaminoethanol.

35 14. A process for purifying choline-bindable surface protein of one or more strains of S. pneumoniae by adsorbing the protein from a solution which is substantially free of choline onto an ion-exchange column which comprises a tertiary amine analog of choline and then removing the protein bound to the column by washing the column with an aqueous solution which comprises from about 0.5% to about 4% choline to form a solution containing purified choline-bindable surface protein.

16. Lyophilized choline-bindable protein of one or more Streptococcus pneumoniae strains.

17. The lyophilized protein of claim 16 wherein the protein has been purified by adsorbing the protein on an ion- exchange column and then eluting the protein with an aqueous solution comprising from about 1% to 3% choline.

18. A composition of matter comprising choline-bindable protein of one or more Streptococcus pneumoniae strains in a cell growth supernatant which is substantially free of cells and cell debris.

19. A composition of claim 18 containing at least 0.02% ethanolamine.

20. A composition of matter comprising a cell growth supernatant substantially free of cells and cell debris containing as an active immunogenic agent at least one choline bindable surface protein of one or more Streptococcus pneumoniae strains.

21. A composition of claim 20 containing ethanolamine.

22. A composition of claim 20 containing choline.

23. A composition of matter comprising a cell growth supernatant containing a full-length PspA in a dialysis bag. 24. A composition of matter comprising lyophilized choline bindable surface protein of one or more Streptococcus pneumoniae strains in a sealed container.

5 25. A method of immunizing a susceptible host against infection with Streptococcus pneumoniae comprising administration of the composition of claim 20 to the host.

26. A method of immunizing a susceptible host against

^*~» infection with Streptococcus pneumoniae comprising the steps of: a) dissolving a lyophilized choline bindable surface protein of one or more Streptococcus pneumoniae strains in an appropriate pharmaceutical carrier and

15 b) administering an immunogenically effective amount of the dissolved protein to the susceptible host.

27. The method of claim 27 wherein the composition administered contains an adjuvant.