IE921407A1 - Recombinant b oligomer of pertussis toxin - Google Patents

Abstract

A genetically-engineered B oligomer-like multimeric protein of pertussis toxin, useful as a non-reactogenic component in an anti-pertussis vaccine, is produced by expressing the individual cistronic elements for subunits S2, S3, S4, and S5 separately in a heterologous host, recovering the resultant subunit polypeptides, and assembling them by association in vitro to form a multimeric composite.

Description

RECOMBINANT B OLIGOMER OF PERTUSSIS TOXIN BACKGROUND OF THE INVENTION Field Of The Invention The present invention relates to the recombinant expression of the subunits of the B oligomer of pertussis toxin and association of the subunits in vitro to form a non-reactogenic multimeric composite capable of use in a vaccine for eliciting an immunoprotective response against infection, and the effects of infection, by Bordetella sp.

Description Of The Art The demand for whooping cough vaccines with decreased adverse reactions has stimulated research to better understand the immunogenic constituents of the etiologic agent, Bordetella pertussis. An important component of new acellular vaccines is pertussis toxin (PT), a heterohexameric ADP-ribosyltransferase that interferes with hormonal signal transduction in mammalian cells (1-3). It is this enzymatic activity, residing in the SI subunit of the toxin (4), that results in toxicity and is generally accepted to be the culprit in untoward reactions in the recipients of pertussis vaccines. Thus, a major concern in producing toxin-based pertussis vaccines is the elimination of the enzymatic activity of the SI subunit. Currently, deactivation of SI is accomplished by treatment of vaccine materials with heat or toxoiding agents such as formalin and glutaraldehyde. The disadvantages of these methods of physical inactivation are that, if too stringent, they can substantially reduce the protective immunogenicity of the toxin; and if not stringent - 2 enough, the toxicity-related enzyme activity of the molecule can reappear upon shelf storage (5).

Methods have been developed for producing each of the pertussis toxin subunits by recombinant means in Escherichia coli (6) and identifying regions of the SI subunit crucial for enzymatic activity and formation of a dominant protective antigenic determinant (7). By applying site-specific mutagenesis, the enzymatic activity of the SI subunit is eliminated without compromising the ability of the SI molecule to acquire a conformation consistent with the formation of the protective epitope (8) or with association of the subunit with the B oligomer to assemble into a holotoxin-like multimeric species (9).

B oligomer is believed to be the delivery platform for the SI subunit, possessing cell receptor recognition domains (1,10-13); it can be isolated as a pentameric macromolecule without the SI subunit, comprised of subunits S2, S3, S4, and S5 in an approximate molar ratio of 1:1:2:1, respectively (1). It has recently been shown that natural (native) B oligomer, which intrinsically possesses none of the ADPribosyltransferase activity associated with the toxicity of pertussis toxin, is by itself sufficient to elicit immunoprotective responses (14-16); further, in studies with recombinant proteins conducted in collaboration with Dr. Drusilla L. Burns and Dr. Juan L. Arciniega of the U.S. Food and Drug Administration, the SI subunit did not appear to contribute significantly to the protective response of the B oligomer. Thus, B oligomer clearly has the potential as a component of acellular pertussis vaccines. However, the natural B oligomer used in these studies was isolated from the holotoxin molecule, and while largely free of the SI subunit (>95%), B oligomer derived in such a manner has IE 92140? - 3 measurable amounts of Sl-related activity which may contribute significantly to its reactogenicity.

BRIEF DESCRIPTION OF THE DRAWINGS 5 Figure 1. This figure depicts the SDS-PAGE of recombinant B oligomer-like multimer in accordance with this invention (hereinafter referred to simply as recombinant B oligomer). B oligomer was assembled in vitro from recombinant E. coii-produced subunits S2, S3, S4, and S5, as described in the following text, and purified by affinity chromatography on fetuin-Sepharose (17) . The sample, eluting from the chromatographic column in 4 M MgCl2, was precipitated with trichloroacetic acid (TCA) and subjected to SDS-PAGE (18) . The gel was subsequently stained with Coomassie Blue. Lane 1) natural PT (commercial-grade); lane 2) natural B oligomer (commercial-grade); lane 3) recombinant B oligomer.

Figure 2. This figure is a graphical representation of the mitogenic activity of B oligomer preparations, including recombinant B oligomer according to this invention. Two-fold serial dilutions of the indicated B oligomer preparations were made in RPMI 1640 medium. Mouse splenic lymphocytes were then added to the B oligomer preparations to yield concentrations of B oligomer indicated on the abscissa. [3H]thymidine incorporation was measured by conventional means (19).

Since the recombinant B oligomer preparation was stored in high concentrations of urea (2 M), and because of its relatively low initial protein concentration, the lower dilutions of the recombinant B oligomer preparations contained significant amounts of residual urea that resulted in cell death. The preparations were: natural B oligomer (·); natural B oligomer which at the highest - 4 concentration shown contained 0.125 M urea () ; and recombinant B oligomer which at the highest concentration shown contained 0.125 M urea (o) .

Figure 3. This figure illustrates the effect of immunization of mice with B oligomer preparations on the leukocytosis-promoting activity of PT. Five mice were each injected intraperitoneally (i.p.) with preparations of recombinant B oligomer (rB) according to this invention, or natural B oligomer (B), at doses of 8 pg (rBH/BH), 1 pg (rBM/BM), or 0.1 pg (rBi/Bi,) in a total volume of 0.5 ml of PBS-gel with alum (see Table 1). One group of 5 mice (marked PT) were mock-immunized with PBS-gel. One month later, all groups were challenged i.p. with 500 ng of PT (with the exception of the group marked PBS*, which were challenged with PBS); the leukocyte count was determined four days later. The mean of the leukocyte counts plus one standard deviation are shown.

SUMMARY OF THE INVENTION The present invention provides a geneticallyengineered B oligomer-like multimeric protein (recombinant B oligomer) of pertussis toxin derived from recombinant materials, as well as a method of preparation. Briefly described, the cistronic elements (genes) for the S2, S3, S4, and S5 subunits of the toxin, and any analogs of such subunits produced by genetic manipulation (e.g., site-specific mutagenesis) and derivatives thereof, are individually expressed in separate transformed heterologous (foreign) hosts, harvested, and then associated in vitro to form a multimeric composite which, like natural B oligomer, has the ability to elicit a mitogenic response in - 5 lymphocytes and, more important, to induce an immunoprotective response against pertussis toxin.

Immunization of mice with the recombinant B oligomer of this invention, or with native B oligomer isolated from PT produced by B. pertussis, resulted in antibodies reactive with PT in ELISA and which were able to neutralize the cytopathic effect of PT on cultured Chinese hamster ovary (CHO) cells; antibody titers evoked by each preparation were similar. Moreover, mice immunized with either natural or recombinant B oligomer were protected against the leukocytosis-promoting activity of PT. These data demonstrate that the pertussis toxin B oligomer of this invention, assembled in vitro from individual recombinant subunits, is a suitable candidate for inclusion in an acellular pertussis vaccine.

DETAILED DESCRIPTION OF THE INVENTION The individual subunits were produced in recombinant E. coli as previously described (6). In this recombinant production system, each of the subunit polypeptides is synthesized with an initiating methionine residue substituting for its native signal peptide sequence; the heterologous amino-terminal methionine residue is substantially cleaved from each of the recombinant proteins, with the exception of subunit S4, by endogenous methionyl aminopeptidase. The S4 subunit is thus a methionyl-mature polypeptide, i.e. it is an analog of the natural mature S4 subunit bearing a methionine residue at its amino terminus. Each of the recombinant subunit proteins were synthesized in individual E. coli transformants in individual fermentations. Cell pastes were recovered from the fermentation broths and were subsequently stored at -60°C to -80°C. Cells in the individual cell pastes - 6 were disrupted by two passages each through a French press at approximately 10^ psig in the presence of 1 mM dithiothreitol (DTT) at a temperature of approximately 4°C. Recombinant subunit proteins were synthesized, under these conditions of fermentation, as insoluble inclusion bodies in E. coli. These inclusion bodies were recovered from the cell lysates, and washed at least twice, by differential centrifugation; the first wash was in 1% deoxycholate (DOC) and 25 mM Tris-HCl (pH 7.9), and the second wash was in 4 M urea and 25 mM Tris-HCl (pH 7.5).

The individual inclusion body preparations were weighed and then mixed in a stoichiometric amount, based upon their relative percentage of subunit protein, approximating their molar ratio in natural B oligomer (S2:S3:S4:S5 at 1:1:2:1, respectively). The inclusionbody mixture was solubilized by adding a solution of 6 M guanidinium-HCl (GuHCl) and 10 mM DTT in a buffer of 25 mM Tris-HCl (pH 8.5); the mixture was then stirred gently overnight at room temperature. The mixture was centrifuged (18,000 rpm in a JA20 rotor for 30 minutes at 2°C-room temperature) to remove insoluble materials. The soluble supernatant was recovered and excess DTT removed by desalting in a chromatographic column of GH25 with the same GuHCl-Tris buffer free of DTT. The protein peak eluting at, and near, the void volume of the GH25 column was recovered. The sample could be diluted with 6 M GuHCl anytime after this step.

CU2SO4 was added to a final concentration of 50 μΜ; the sample was gently stirred overnight in an air atmosphere at room temperature. Following this reoxidation step, the sample was dialyzed against two changes (5 liters per change) of 100 mM potassium phosphate buffer (pH 7.4) containing 2 M urea; dialysis tubing should have the tolerance of 3,500-MW cut-off.

An insoluble precipitate forms during dialysis; this was - 7 removed by centrifugation (13,000 rpm, JA14 rotor, 20 min, room temperature). The supernatant was recovered and stored at 2°-8°C. The sample was then dialyzed against phosphate-buffered saline (PBS) overnight at room temperature; dialysis volume should be no less than 10-fold greater than the sample volume and should be changed at least once. Again, dialysis tubing should have a tolerance of 3,500-MW cut-off.

Fetuin-Sepharose affinity resins were prepared 10 from commercial fetuin Sepharose 4B, purified by the Spiro method (available from Gibco) and covalently linked under published conditions. The dialyzed sample was applied twice to a chromatographic column of fetuinSepharose at temperatures of 2°C-room temperature; bulk (non-column) methods of application, wash, and elution may also be utilized. The loaded affinity resin was washed sequentially with PBS and with buffer A (0.05 M Tris-HCl, pH 7.5) containing 1 M NaCl. The sample binding to the resin under these conditions was eluted with buffer A containing 4 M MgCl2.

In summary, then, the individual recombinant subunits were isolated as insoluble inclusion bodies following lysis of the bacteria. Individual inclusionbody preparations were solubilized in 6 M guanidinium hydrochloride (GuHCl) in the presence of 1 mM dithiothreitol (DTT) and combined to give a molar ratio approximately equivalent to that estimated to be found in natural B oligomer (i.e., S2:S3:S4:S5 at 1:1:2:1, respectively). Reductant is removed by size-exclusion chromatography, reoxidation enabled by stirring in an air atmosphere in the presence of 50 μΜ CU2SO4, and spontaneous association of the subunits facilitated by equilibrium dialysis against 2 M urea. By virtue of its ability to bind complex carbohydrates, a cell targeting mechanism of the toxin (20), the recombinant B oligomer was purified to near homogeneity (as assessed by the - β purity of the individual subunits in the multimer) by affinity chromatography on fetuin-Sepharose (Fig. 1); in our experience, very little if any of the individual subunits bind specifically to the fetuin resin, indicating that the bulk of material eluting in MgCl2 is in a multimeric form. The eluant was tested, and found to be positive, for agglutination of goose erythrocytes. Western blotting with polyclonal anti-toxin serum and a modified ELISA (using fetuin-coated plates and monoclonal antibodies against S2, S3, S4, and S5, and an anti-Sl monoclonal antibody as a control) both confirmed the presence of each of the immunoreactive B subunit polypeptides. Although the ratio of subunits S2 and S3 (as measured by integrative scanning densitometry of stained SDS-polyacrylamide gels) differed slightly from that of natural B oligomer, this may be accounted for by excess S3 subunit (but see above) or S3-S4 dimers coeluting from the fetuin-Sepharose affinity resin. That the isolated multimer contained a significant amount of the pentameric B oligomer species, however, is clearly demonstrated by the mitogenicity experiments (Fig. 2): whereas B oligomer and certain of the individual subunits and dimers comprising it are capable of erythrocyte agglutination, only intact B oligomer is mitogenic (10,21). Notably, and regardless of the relative proportions (ratio) of the individual protein subunits (S2, S3, S4, and S5), the recombinant B oligomer of this invention differs from the natural form in that it always possesses a methionyl-mature S4 subunit.

The recombinant B oligomer preparations were then evaluated for their ability to stimulate toxinneutralizing immune responses in experimental animals, now a standard procedure for assessing the potency of pertussis vaccines. This testing was performed by Dr. D. L. Burns and Dr. J. L. Arciniega of the FDA. - 9 Mice were inoculated intraperitoneally (i.p.) with graded doses of either natural or recombinant B oligomer. Two weeks after immunization, mice were bled to measure titers of antitoxin antibodies by ELISA and then challenged with lethal doses of PT; animals were observed for toxin-induced death and all were evaluated four days after challenge for toxin-mediated leukocytosis. As demonstrated by the results in Table 1, mice responded to each immunizing agent with significant antitoxin and toxin-neutralizing titers, the latter measured by reduction of PT-mediated CHO cell clustering; moreover, the neutralizing antibody titers induced by the recombinant B oligomer were similar to those observed when natural B oligomer was used as the immunogen. -10υ Μ 0) Ρ> -Η Ρ Cn c N P ι—I (0 P P 3 tt) 2 Ο VO rd ο VO 00 cn in CM rd w CM rd rd CM Ρ rtJ Ό V δ Μ Ο η δ η Ό ρ (fl Η Η μ η Ή Ο φ Md Ό θ' β β •Η fl . θ' Λ fl · θ' „ ό ι Φ φ d Μ VO φ nJ β Φ Φ υ θ' ·Η :d 6 •U β Φ >ι ιη ια ι ~ m £ Ό ό Ρ C (0 Ή θ' <ο 3 C •Η -0 — 3 φ • Ο ρ O' tt) CL Ρ Λ > · Ρ Ο ·Η Ο c — ρ C Ή Ρ φ (0 Μ Ρ 3 tt) Ρ nJ η Ρ 3 α> ρ θ' ο.

Md ο β ο Ρ » ο. 2 σι Μ Φ Ρ ο G •σ Ή Ο oo in cn *»· 00 cn CO O oo ·» CM rd rd σι sj* r* 00 ’T 00 cn cn io VO rd oo vo O 00 K *M K K cn CM CM rd «r CM rd <*> C Ο <\Ι φ ρ ο θ' • Ρ Ο Ο ρ ο C ·3· O' nJ >, β) Ρ Λ Ρ Ε-» Φ 0) C · Ο Ό ρ φ Μ « C C _ ρ ρ tn nJ nJ φ (Ο Μ ρ ρ CM ρ CO C Ο Μ Ρ Ρ ο α « Ί φ Οι φ Ο, φ Μ Ο. σι nJ φ1 Ρ Ό Φ φ Ρ C <0 Ρ Ρ Ρ ο ιθ Ρ η) ρ Ό φ θ' □ C η) φ nJ Ρ C Ό ο Ο £ Ρ Φ [ρ Ό ρ O' Ρ Φ Ρ Φ Λ Μι ρ C Φ C Ρ Ή Φ Φ Ρ Ρ > 3 « Δ > ΜΙ φ Φ 3 ΜΙ β) Ο—3 θ' ρ Ό σι 3 φ ΜΟ Ρ σ ν ρ φ >< ι ε rt) C 4) O' o G p Is o O' Ή O σ» α, φ φ ρ c « ό ρ — φ ΟΡΌ C X ΛΙ Ο. Ο >, Μι Ρ Δ Ό Ρ Ρ >1 C Ρ Δ ‘ c Μι ΛΙ Φ σ' Ρ Ρ nJ ρ O’ ffl « Φ >J >1 Οι c c o o •d Ή MJ mJ fl «0 N N o CQ MJ c <0 c c c o o Ρ P P p nJ Φ c η < φ =l ρ ρ Λ β) Ό 3 ε 3 ρ 3 η φ φ ρ ρ PC Ό ιο Ό φ ρ Ό C ε C Φ Ν (Ν CD β β MJ fl ο o Fd CM o o φ at ο ο Ρ fM C £ Ρ G ρ 2 φ φ δ ρ ϊ (X Φ ε ρ 2 -ο QJ C α Ό φ Μι ρ ρ X rt) φ φ Ρ υ ο. 2 φ rtJ ο -Q ρ ϋ φ ο ΰ η c Λ ρ «» 3? J.2 J -Ρ Ρ 3 C Ρ <0 ·ζ) ρ ρ ο 3 ” Φ ο > ρ C θ' Ρ ο Ρ η <0 θ' C •ο Ρ Φ Ρ Ρ (0 φ Ρ Ρ φ 3 Μι Ο m CO d 00 d fl Φ 00 Φ X Φ oo < to o β *«*» Md rd oo o • • <0 β Ό 3 β •d Φ + o X mJ 00 •d o Id rd P 4J Li O Φ P □ o CU O' Φ *d rd Φ 1 υ •d 0 M CQ Ό x: X o 00 X X rd Md •d υ A 0 x: g oo 4-> ’d >1 Φ •d 0 X MJ 3 3 Λ •d rd O' g > 0 Ό c ♦d > Φ P φ MJ MJ Ml x: U o Q Φ 4J <0 3 Φ P MJ •m in Md O' β 3 O β Φ ♦d P •d x: υ rd d 44 0) <0 Φ υ Ό Q •U Md •d φ O co O ε u d 3 3 0. P β a Ό •d υ m o c υ φ d • P φ rd g Md I d rd CU C Φ o Φ Μι Φ W φ Ο Δ POP X Φ η υ ε nj - ·° η Ό Φ Ο Λ -Η Ρ C ηι θ' c •Η Ν Ρ Ρ Φ Μι Ρ Φ C Δ Φ ρ υ φ Ρ ω Ό Ρ Ρ Δ ρ £ X φ Ρ ο C φ Ο = Ό Ο Ρ Φ Ό θ' υ O' 00 4) N fl •d •d 3 n •d Φ β rd MJ in rd x: X O d Φ fl MJ O • m Φ Ml d o X CU Q. mJ Ό rd CM β X 3 β 4 • · o o Φ Φ Φ fl rd d MJ N β + Md •d Φ Φ 00 d rd Ml x: mJ rd fl <0 Φ fl nj υ fl Φ 3 d MJ d •d Q O' Φ •d MJ n x: •d I Ό oo MJ 3 Φ 3 rd to Φ Φ 3 0. X MJ Md Φ β P 3 CU oo O rd nJ Φ Ό Φ 43 O > O x: MJ ε ρ ίλ Ρ tt) C Ό ρ — m Ό Φ ιη Φ Ό η ηι ε ε tt) — ο Ε ρ rd Φ O Md Ml o O fl > fl O Ό IP Φ O Ml Ό Φ Ml •n 3 Φ mJ > ja Ό 2 C 00 υ C •d mJ •d Φ o. P Φ P JQ d d P 3 Δ nJ •d rd υ o φ Φ rd υ P Λ •d CO o Ml o fl nj Δ X Λ φ (X Φ •d > Φ O Φ 2 3 ε - 11 More important, there was no significant difference in the level of protection afforded against the toxic effects of PT in animals immunized with natural B oligomer or with recombinant B oligomer (Fig. 3). Neither individual subunits (6,22) nor individual S2-S4 or S3-S4 dimers of the B oligomer (23) are capable of eliciting immunoprotection against the leukocytosispromoting effects of PT. Although the potency of the recombinant material at the lowest dose (0.1 μg) may be 10 slightly lower than an equivalent amount of native B oligomer, this difference may more likely be ascribed to accuracy in determining the protein concentration of the recombinant B oligomer by integrative densitometry of stained SDS-polyacrylamide gels .

The results presented herein demonstrate the feasibility of creating complex heteromeric proteins from recombinant DNA-derived subunits. In the case of PT, it is now possible to produce a highly immunogenic subspecies of the toxin molecule. Although B oligomer can be produced in B. pertussis by inactivating the ability of the SI subunit to associate with B oligomer by mutations in its encoding cistronic element, little if any of the B oligomer is subsequently secreted by this organism (24). B oligomer can be purified from natural toxin, but this is an intensive process which still results in measurable contamination by Slcontaining holotoxin (9,19,25). Not only does the recombinant B oligomer of this invention lack intrinsic toxicity by virtue of deletion of the SI subunit, but removal of the pathogenic organism from the manufacturing process eliminates the potential for contamination by other B. pertussis toxic components (26,27) which may also contribute to the adverse reactions of pertussis vaccines.

The genetically-engineered B oligomer-like multimer of this invention can be formulated in a - 12 conventional manner into a vaccine. In the case of a toxin that has been genetically inactivated, such as pertussis toxin in the present invention, further inactivating steps (such as chemical treatment or heat treatment) should not normally be required since these materials are produced in non-pathogenic organisms and are inherently free of the Sl-related enzyme activity that is generally accepted to elicit the adverse reactions to whole-cell pertussis vaccines and to untreated (or not stringently inactivated) pertussis holotoxin vaccines. Nevertheless, it is necessary to control purity of the recombinant products, particularly with regard to the endotoxin content. In general, recombinant B subunit macromolecules described in the present disclosure as potential vaccinating antigens would be purified to >90% homogeneity. The nature and estimated quantity of contaminants, if any, would be evaluated to ensure that the extent of endotoxin contamination meets the standards of the individual regulatory agencies.

For purposes of parenteral delivery, the vaccine materials would normally be adsorbed onto aluminum adjuvants. This can be accomplished by at least two means: precipitation with preformed alum and precipitation with aluminum salts. The adsorbed precipitates are then resuspended in an excipient to yield a dosage concentration of vaccine antigen generally in the range of 5-100 μg per dose and an alum amount usually not exceeding 1.5 mg/dose; volume per dose is generally in the range of 0.1-1.0 ml. The suspending excipient is commonly a buffered solution (e.g., phosphate-buffered saline, pH 7.0), may have added stabilizers (e.g., glycerol), and will likely contain a preservative (e.g., 0.01% Thimerosal) to prevent microbial contamination and to extend shelf life. For stimulation of secretory mucosal immune - 13 responses, delivery of antigen to mucosal lymphoid tissues would be indicated. In this regard, the vaccine materials could be formulated in enteric-coated capsules or other delivery vehicle to prevent degradation of the vaccine product until it reaches lymphoid tissues of the gut (e.g., Peyer's patches). Elicitation of a secretory immune response in the gut can be transmitted globally (via slgA and/or slgA-secreting lymphocytes) to the lamina propria of other mucosal-lined organs, and particularly to the respiratory tract for protection against pertussis. Alternatively, the vaccine product could be formulated as an aerosol or liquid for respiratory, buccal, or nasal delivery in an appropriate excipient and/or delivery vehicle (e.g., liposomes), such that it stimulates a secretory immune response directly in the respiratory tree or is transmitted (as above) from the buccal or nasal lymphoid tissue to the lamina propria of the respiratory tract.

* ★ * While this invention has been illustrated with 25 respect to specific embodiments, it should be understood that other modifications and variations in the manner in which the invention may be practiced are possible. It is intended that the present invention include all such modifications and variations as come within the scope of the invention as defined in the appended claims. - 14 BIBLIOGRAPHY 1. Tamura, M., et al. (1982) Biochemistry 21:5516-5522 . 5 2 . Katada, T., et al. (1982) Proc. Natl. Acad. Sci. USA 79:3129-3133. 3. Bokoch, G. M., et al. (1983) J. Biol. Chem. 10 258:2072-2075. 4 . Katada, T., et al. (1983) Arch. Biochem. Biophys 224:290-298. 15 5. Ashworth, L. A. E., et al. (1983) Lancet ii:878-881. 6. Burnette, W. N., et al. (1988) Bio/Technology 6:699-706. 20 7 . Cieplak, W., et al. (1988) Proc. Natl. Acad. Sci USA 85:4667-4671. 8. Burnette, W. N., et al. (1988) Science 242:72-74 25 9. Bartley, T. D., et al. (1989) Proc. Natl. Acad. Sci. USA 86:8353-8357. 10. Tamura, M., et al. (1983) J. Biol. Chem. 30 258:6756-6761. 11. Sekura, R. D., et al. (1985) in Pertussis toxin (Sekura, R. D., et al., eds.) Academic Press, Inc., New York, pp. 45-64. 12 . Schmidt, M. A., et al. (1989) Infect. Immun. 57:3828-3833 . 13 . Witvliet, Μ. H., et al. (1989) Infect. Immun. 5 57:3324-3330. 14 . Shahin, R. D., et al. (1989) in Vaccines 89: modern approaches to new vaccines including prevention of AIDS (Lerner, R. A., et al., eds.) Cold Spring 10 Harbor Laboratory, Cold Spring Harbor, NY. pp. 249-252. 15 15. Shahin, R. D., et al. (1990) J. Exp. Med. 171:63-73. 16. Shahin, R. D., et al. (1990) Infect. Immun. 58:4063-4068. 17 . Sekura, R. D., et al. (1983) J. Biol. Chem. 20 258:14647-14651. 18. Laemmli, U. K. (1970) Nature 227:680-685. 19. Burns, D. L., et al. (1987) Infect. Immun. 25 55:24-28 . 20. Irons, L. I., et al. (1979) Biochim. Biophys. Acta 580:175-185. 30 21. Nogimori, K., et al. (1984) Biochim. Biophys. Acta 801:232-243. 22 . Nicosia, A., et al. (1987) Infect. Immun. 55:963-967. - 16 23. Hausman, S. Z., et al. (1989) Infect. Immun. 57:1760-1764 . 24. Pizza, M., et al. (1990) J. Biol. Chem. 265:17759-17763 .

. Arciniega, J. L., et al. (1987) Infect. Immun. 55:1132-1136. 26. Weiss, A. A., et al. (1986) Ann. Rev. Microbiol 40:661-686. 27. Burnette, W. N., et al. (1990) Bio/Technology 8:1002-1005 .

Claims (24)

WHAT IS CLAIMED IS:

1. A pertussis toxin B oligomer-like multimer protein comprised of subunits S2, S3, S4, and S5, in 5 which the S4 subunit is a methionyl-mature polypeptide, or an analog or derivative of the multimer.

2. The pertussis toxin B oligomer-like multimer protein according to claim 1 in which subunits 10 S2, S3, S4, and S5 are in a stoichiometric molar ratio approximatinc 1:1:2:1, respectively.

3. An anti-pertussis vaccine, comprised of an effective amount of the pertussis toxin B oligomer-like 15 multimeric protein of claims 1 or 2.

4. A method of producing a pertussis toxin B oligomer-like multimeric protein from recombinant subunits by expressing the cistronic elements for 20 subunits S2, S3, S4, and S5, or any site-specific codon substitutions or additions thereof, separately in heterologous expression hosts, and permitting an admixture of the recombinant subunits to associate in vitro to form a multimeric composite of the subunits 25 having the ability to induce an immunoprotective response against pertussis toxin.

5. The method of claim 4 wherein the multimeric composite possesses the ability to induce an 30 immunoprotective response against pertussis disease.

6. The method of claim 4 wherein the individual S2, S3, S4, and S5 subunits are admixed in any stoichiometric molar ratio. - 18 7. The method of claim 4 wherein the individual S2, S3, S4, and S5 subunits are admixed in a stoichiometric molar ratio approximating 1:1:2:1. 5 8. The method of claim 4 wherein the admixture is formed by solubilizing the recombinant subunits S2, S3, S4, and S5 under reducing conditions and thereafter subjecting them to oxidative conditions.

7. 10 9. The method of claim 8 wherein the reducing agent is dithiothreitol (DTT). 10. The method of claim 8 wherein the oxidants are atmospheric air and CU2SO4.

8. 11. The method of claim 8 wherein the admixture is formed in the presence of a chaotrope, detergent, or denaturant. 20

9. 12. The method of claim 11 wherein the chaotrope/denaturant is guanidinium hydrochloride.

10. 13. The method of claim 11 wherein the chaotrope/denaturant is urea.

11. 14. The method of claim 11 wherein, subsequent to oxidation, the concentration of the chaotrope, detergent, or denaturant is lowered or exchanged for another chaotrope, detergent, or 30 denaturant at a lower concentration, to facilitate refolding of the subunits and their association into a multimeric protein.

12. 15. The method of claim 14 wherein the 35 exchange chaotrope/denaturant is guanidinium hydrochloride . - 19

13. 16. The method of claim 14 wherein the exchange chaotrope/denaturant is urea.

14. 17. The method of claim 4 wherein the 5 heterologous host is a prokaryotic organism.

15. 18. The method of claim 17 wherein the prokaryotic host is Escherichia coli. 10

16. 19. The method of claim 18 wherein the recombinant S2, S3, S4, and S5 subunits are recovered from E. coli in the form of inclusion bodies.

17. 20. The method of claim 19 wherein the 15 multimeric composite of the recombinant subunits is purified by chromatography.

18. 21. The method of claim 20 wherein the multimeric composite of the recombinant subunits is 20 purified by affinity chromatography.

19. 22. The method of claim 21 wherein the multimeric composite of the recombinant subunits is purified by fetuin affinity chromatography.

20. 23. The method of claim 22 wherein the multimeric composite of the recombinant subunits is purified by haptoglobin affinity chromatography. - 20

21. 24. A pertussis toxin B oligomer-like multimer protein according to claim 1, substantially as hereinbefore described.

22. 25. substantially as An anti-pertussis vaccine according to claim 3, hereinbefore described.

23. 26. A method of producing a pertussis toxin B oligomer-like multimer protein according to claim 1, substantially as hereinbefore described.

24. 27. A pertussis toxin B oligomer-like multimer protein according to claim 1, whenever produced by a method claimed in a preceding claim.