COMPOSITION AND METHOD OF IMMUNIZATION WITH IMPROVED CARRIER AND HAPTEN-CA-RRIER VACCINES

Field of the Invention This invention relates to antigens or haptens conjugated or complexed to highly cationized protein carriers. The invention also relates to native antigens complexed with their highly cationized counterparts. These conjugates or complexes possess enhanced immunogenicity. This invention also relates to a method of enhancing the immune response to a native antigen by administering the conjugates or complexes to an animal.

Background of the Invention

The importance of antigens in the prevention of infectious disease through immunization is well known. The basis of immunization is the exposure of the animal to dead or weakened infectious agents (i.e. viruses, bacteria, toxins, etc) or extracts thereof, which are capable of evoking an immune response. These substances are generally referred to as antigens.

Most antigens are either wholly or partially composed of protein or saccharide. The action of antigens is thought to be dependent in part on the antigen's affinity for certain binding sites on a variety of cells of the immune system. These cells are present in blood and internal organs. Interaction of the antigens with the binding site stimulates the immune system cascades which, when actuated, defend the organism against infectious agents. But antigens can also have adverse side effects on the organism sought to be immunized. It is therefore desirable to achieve an effective immunogenic response while utilizing low levels of the antigen. If immunogenicity of a specific antigen can be increased, a smaller dosage of the antigen can be administered to achieve an appropriate protective level of immunity. This is important in the treatment of humans and animals.

Some substances do not evoke an immune response at all, or do so very poorly. It would be advantageous to be able to convert such substances from a non-immunogenic form to an immunogenic form. This would allow vaccines to be produced where it was not previously possible. By definition, a hapten is such a non-immunogenic or poorly immunogenic molecule whose immunogenicity can be enhanced when conjugated either covalently or ionically to a carrier molecule, usually a protein.

The cationization of protein-containing substances has been known for some time. Several methods of protein modification which result in a more positively charged version of a native protein have been described. However, none of the prior art has disclosed, suggested or demonstrated elicitation of an enhanced immunogenicity of haptens or other protein or non-protein containing substances when conjugated (covalently or non-covalently) to highly (to a pi of above about 9.3) cationized "carrier" proteins.

An aspect of the present invention relates to the discovery that highly cationized protein-carriers conjugated or complexed with proteinaceous haptens or other weakly immunogenic molecules possess enhanced immunogenic properties. Through this discovery a method of enhancing the immune response to a variety of im unogens is provided.

BACKGROUND ART

Border, W.A., H.J. Ward, E.S. Hamil and A.H. Cohen, Introduction of Membranous Nephropathy in Rabbits by Administration of an Exogenous Cationic Antigen, J. Clin. Invest. 69:451 (1982); discloses a carbodiimide- directed cationization of a protein, specifically bovine serum albumin (BSA), and the effects of charge modification of BSA on its ability to bind directly to the kidney glo erulus. In the course of these studies, Border, et al also examined an effect of charge modification on the immunogenicity of the antigen and determined that there was no significant difference between native BSA and cationized BSA. Border, et al does not disclose nor suggest combining or complexing a highly cationized antigen with a different antigen or with the same unmodified antigen to elicit an enhanced immune response.

The publication of Chu, F.S., H.P. Lau, T.S. Fan and G.S. Zhang,

Ethylenediamine Modified Bovine Serum Albumin as Protein by Carrier in the

Productions of Antibody against Mycotoxins. J. of Imm. Meth. 55: 73-78

(1982); relates to the modification of BSA via carbodiimide-directed cationization with ethylenediamine. BSA is modified under this protocol in order to facilitate and optimize the subsequent combining reaction between the non-immunogenic hapten ( ycotoxin or aflatoxin) and the carrier BSA molecule.

The reference discloses an enhancement of the efficiency and extent of conjugation of hapten to carrier by virtue of the addition of more a ino groups which are the targets available for conjugation chemistry. The increase in amino groups also increases the number of haptens that can be conjugated to the carrier. Immunization with the more highly conjugated BSA-

EDA-mycotoxin immunogen produced a greater antibody response than immunization with unmodified BSA-mycotoxin conjugates delivered at the same protein concentrations. The authors attribute the increase antibody responses to the higher proportion of hapten in the former im unogen as well as postulate a role for the stearic orientation of hapten relative to its carrier. At no point do they demonstrate or suggest that a change in charge contributes to the results observed, nor do they quantitate a charge modification that occurred on the BSA carrier or on the hapten-carrier conjugate. Furthermore, they do not attempt to regulate or control the level of any such modification. The efficacy of the carbodiimide/ethylenediamine reaction is measured only by its effect on the resulting efficiency of hapten binding to carrier. This reference does not suggest or disclose that a protein may be highly cationized and non-covalently complexed or conjugated with an antigen or hapten or the native protein itself and result in significant improvements in the immune responses to the antigen, hapten or native protein.

The classic publication by Mitchinson, N.A. The Carrier Effect in the Secondary Response to Hapten Carrier Conjugates: I, II in Eur. J. Immunol. 1:10 (1971) discusses in detail the effects of carrier conjugation on hapten immunogenity. These references do not disclose or suggest, however, that the phenomenon can be modulated by charge modification of carriers.

Summary of the Invention

There is disclosed an immunological composition which consists of a protein-containing antigenic substance which has been chemically cationized such that it exhibits a pi of at least 9.3 and is coupled covalently or ionically to a native antigen or hapten. This composition displays enhanced immunogenicity as compared to the native antigen or hapten alone or the native antigen or hapten coupled to a native protein-containing carrier molecule.

There is also disclosed a method in which an antigen which has been so highly cationized as to diminish its effectiveness as an antibody producing immunogen for the native protein, can be combined with a native antigen of the same type and, when the combination is used as an immunogen, elicits a synergistic enhancement of the antibody immune response to the native antigen.

There is further disclosed a method of immunization of an animal against an antigen or hapten comprising administration a complex of said antigen and a protein carrier to the animal in an amount effective to induce an immune response to said antigen, the improvement comprising: modifying the protein carrier through the covalent attachment of at least one agent selected from the group consisting of primary amines, secondary amines, tertiary amines, and ammonium groups, where a) said modified protein carrier has an isoelectric point, as measured by isoelectric focusing in a polyacrylamide gel, of at least a value of about 9.3; b) the immunological response of the animal to the complex of antigen and modified carrier is greater than to the antigen complexed to an unmodified carrier. It is preferred that the modified protein carrier have an isoelectric point value within the range of from about 9.3 to about 11.0 or higher. The complex of antigen or hapten and modified carrier maybe administered with or without an adjuvant.

The invention is more specifically directed to the use of a vaccine which comprises a mixture of highly cationized antigen with a pi of greater than about 9.3 and the native antigen at ratios of 10:1 to 1:10.

As used herein, "protein-containing substance" (PCS) includes all proteins, as well as substances whose molecular composition is in some part proteinaceous, such as lipoproteins. They may or may not be highly antigenic in their native form. More specific examples of such substances include, without limitation, bovine serum albumin (BSA), hen egg albumin (OVA), diphtheria toxoid (DT), tetanus toxoid (TT), the outer membrane protein of Neisseria meningitidis (OMP), other components of bacterial, fungal and viral microorganisms and genetic variations thereof, peptides such as hormones and cytokines, and non-mammalian, non-homologous "carrier" proteins such as keyhole limpet hemocyanin (KLH).

As used herein, "cationization" means the conversion, substitution or addition of functional groups of the native protein-containing substance whereby the substance is rendered more cationic (positively charged). Such native protein-containing substances are generally anionic within a physiologic pH range. The native PCS is converted to a cationic derivative by addition of functional groups. An example of such a cationization is the reaction whereby anionic side chain carboxyl groups of the PCS are substituted with polycationic aminoethylamide groups.

As used herein, the unreacted or "native" form of the protein- containing substance is indicated by a prefix "n" and the cationized forms are indicated by a prefix "c". For example, diphtheria toxoid (DT) may be expressed as "nDT" for the native form and "cDT" for the cationized form.

As used herein, the terms "antigen" and "immunogen" mean any substance which is capable of eliciting an immune response. The term "immunogenicity" denotes the immune response to an antigen as determined by assessment of the humoral and/or cellular response to the native molecule.

The term "carrier molecule" as used herein refers to a protein- containing substances, such as these described above. The carrier molecule may be in its native or a cationized state. A carrier molecule participates in eliciting the immune response to the desired antigen by virtue of the fact that it is covalently or ionically associated with the hapten or antigen to which immunity is sought.

The term "hapten" means a separable part of an antigen that can react specifically with an antibody but is incapable of stimulating significant immune response except in combination with a carrier molecule. Usually a hapten is not immunogenic or is poorly immunogenic, but when conjugated either covalently or ionically to a carrier protein, exhibits strong immunogenicity. For example, it is know that when poorly immunogenic polysaccharides are conjugated to a protein carrier they display greater humoral immunogenicity as well as the capacity to stimulate a T-cell dependent cellular immune response. Representative haptens useful in the instant invention include oligo and polysaccharides, glycoproteins, lipoproteins; peptides derived from viruses, bacteria, higher organisms, hormones, cytokines; nucleic acids; plant derived protein containing substances; synthetic chemical compounds and the like.

Detailed Description of the Invention

The protein-carrier can be cationized by several methods known in the art. See for example Border, et al (supra). A preferred method is the reaction of ethylenediamine (EDA) and then the reaction with l-ethyl-3-(3- dimethylaminopropyl)-carbodiimide (EDC) with the protein. This reaction involves the activation of the carboxyl groups of a protein with carbodiimide and the subsequent reaction of the activated carboxyl group with a nucleophile of the general type +R-NH₂ to obtain the primary amine derivatives. Considerable versatility can be achieved since both the chemical nature of the modification, i.e. introduction of primary, secondary or tertiary amine groups and the degree of modification of the protein carboxyl groups can be varied by proper choice of reagents, reaction time and pH of the coupling reaction.

The pH of the EDC reaction is generally maintained in the range from about 4.75 to about 6.25. More rapid substitution of carboxyl groups in the EDC reaction occurs at the lower pH levels within this general range. .Reaction time is determined by the concentration of the reactants and by the degree of cationization desired. Representative reaction times are about 15- 180 minutes. The reaction is maintained within the general range from about 4°C to about 37°C and is generally maintained at about 25°C.

Each known method of cationization may be halted or quenched according to several methods known in the art. Such quenching prevents further addition of cationic groups to the PCS thereby facilitating the production of molecules with the desired level of cationization. The EDC reaction is quenched with a buffer, preferably an acetate buffer, which terminates the reaction. Preferably the concentration of the acetate buffer is about 4M.

Cationization can be verified and quantified using isoelectric focusing techniques known in the art and other known chemical methodologies see for example; (Hoare, D.G. and D.E. Koshland, A Method for the Quantitative Modification and Estimation of Carboxylic Acid Groups in Proteins. J. Biol. Chem. 242:2447 (1967)).

The protein-containing substance is cationized to an extent whereby it exhibits enhanced utility as a carrier molecule. It is within the skill of the art to determine and adjust the degree of cationization which increases efficacy as a carrier molecule for each PCS type. The following examples are intended to illustrate the present invention as practiced on several antigens. Variations of the parameters and methodology for optimization of the present invention for any specific PCS and/or hapten is within the skill of the artisan.

Preparation of Cationized Diphtheria Toxoid

Example 1

Diphtheria toxoid (DT) (purchased from Connaught Laboratories) contained 11.7 mg/ml protein with a specific activity of about 250 Lf/mg (Lf is a standard flocculating unit.) A solution of 1M ethylenediamine (EDA) in

10 mM 2-(N-morphol ino) ethanesulfonic acid (MES) was prepared by dissolving

EDA dihydrochloride (Sigma) and MES (Sigma) in deionized water (Milli-Q system by Millipore) in a polyethylene or polypropylene container, and the pH adjusted to 5.5. To this was added the DT. The solution was stirred continuously in a constant temperature water bath at 25°C until the DT was dissolved. At this point, a sample was removed, mixed with 4M acetate buffer adjusted to match the pH of the reaction. This sample was used as a time zero sample for later comparisons. The reaction was then initiated with the additionof1-ethyl -3-[(dimethylaminopropyl)-carbodiimidehydrochloride] (EDC,

Pierce) dissolved in deionized water. The reaction solution was stirred continuously, maintaining temperature and pH. At selected time intervals, up to 180 minutes, samples were removed and quenched with pH adjusted 4M acetate.

The samples were then placed in dialysis bags (Spectra-Por, MW cut-off 14,000) and dialyzed twice against 10 M sodium phosphate, pH 7.2; four times against a solution of 10 mM sodium phosphate and 150 mM sodium chloride, pH 7.2; and twice against deionized water. All dialyses were performed at 4°C for a minimum of 8 hours each. The samples were stored in polypropylene containers at 4°C.

Preparation of Covalent Conjugates Example 2

Covalent conjugates of nDT and clδODT coupled to H. influenzae b capsular polysaccharide were prepared by reductive amination, using a modification of the procedure of Anderson et al . (Anderson, P.W., Pichichero, M.E., Insel , R.A., Betts, R., Eby, R. and Smith, D.H.: Vaccinnes Consisting of Periodate-Cleaved Oligosaccharides from the Capsule of Haemoohilus Influenza type b Coupled to a Protein Carrier: Structural and Temporal Requirements for Priming in the Human Infant. J. Immunol. 137: 1181-1186 (1986)). This approach involves generation of oligosaccharides from the polysaccharide substance, polyribosyl-ribitol -phosphate (PRP), by periodate oxidation, followed by coupling of sized oligosaccharide fragments to carrier protein via reductive amination.

PRP oxidation. 33.3 mg PRP (Connaught Labs) was dissolved in 0.2 ml 0.2M sodium phosphate buffer, pH 7.0, and combined with 4.8 mg Na etaperiodate. The preparation was incubated for 30 minutes at room temperature, and cleavage of the polysaccharide chain was verified by monitoring reduction in apparent molecular weight by size exclusion HPLC (BioRad BioSil SEC-250 column). PRP oxidation products were subsequently sized on a Sephadex G75 column, using a 0.02M triammonium acetate buffer. Fractions were collected and monitored for protein (A280) and carbohydrate content (phenol sulfuric acid assay, D-ribose standard). PRP oligosaccharides corresponding to approximate molecular weights ranging from that of glucose to about 60,000, were pooled, lyophilized and rehydrated in 1 ml distilled water. The oxidized PRP pool was then assayed for carbohydrate content and reducing sugar activity (tetrazolium blue reduction assay, D-ribose standard).

Conjugation. For coupling to either nDT (Connaught) or cDT by reductive amination, 2 mg either carrier protein was mixed with 10.25 reducing equivalents of oxidized PRP in 0.2M sodium phosphate buffer, pH 8.0 (reducing equivalents calculated to represent 1:1 molar ratio with primary amines in nDT). Sodium cyanoborohydride was added at 1.2 mg/sample, and the sample incubated at room temperature for 3 days followed by 8 days at 37°C (nDT conjugate), or 3 days at room temperature followed by 5 days at 37°C (cDT conjugate). Conjugation was monitored by size exclusion HPLC by evaluating the shift in apparent PRP molecular weight. The reaction was terminated by addition of 0.25 mg/sa ple sodium borohydride. Conjugates were then separated on a G75 column, and fractions in which carbohydrate and protein co-eluted were pooled as carrier-PRP conjugate.

Formation of Homologous Complexes of I munogens

Example 3

DT (Connaught) was cationized by the method of Example 1. nDT plus clδODT (22EDA/DT) mixtures were prepared at various ratios, and allowed to incubate for 72 hours at 4°C to facilitate ionic interactions. Immunogen preparations of mixtures or nDT alone were then adsorbed to aluminum hydroxide (alum, Banco: 15.9 mg/mg immunogen) by end over end rotation at room temperature for 4.5 hours and adjusted to deliver the desired amount of immunogen in 0.2 ml volumes.

Preparation of Cationized OVA Example 4

Ovalbumin (OVA) from Sigma Chemical, was cationized according to the general procedure described by Border (1). Specifically, five grams of OVA was dissolved in distilled water to a volume of 25 ml and admixed with a solution of EDA in 500 ml distilled water for a final EDA concentration of 1M. The pH of this solution was adjusted to about 5.0 with 6 HC1. To this was added 1.8 grams of EDC.

The reaction was permitted to proceed for 60 minutes, with constant stirring, while the temperature was maintained at about 25°C and the pH was held constant. After quenching with 4M acetate buffer, the reaction mixture was subjected to multiple dialysis treatments against distilled water and

lyophilized. It was then chromatographed through Sephadex G-25 and lyophilized again before use. The resulting cOVA had a pi of greater than 9.3.

Preparation of Cationized Serum Associated Amyloid

Example 5

Serum Associated amyloid (SAA) is a low molecular weight polypeptide (MW 11,685), found associated with high density lipoprotein in acute phase serum. SAA is an anionic molecule, with a reported pi of 5.7-6.1.

In this present example, the ability of cationic SAA and a mixture of cationic and native SAA to induce polyclonal anti-nSAA antibodies in rabbits, as compared to nSAA immunogen, was evaluated.

Highly purified nSAA, derived from acute phase human serum, was obtained and the starting pi of the material was verified to be 5.7, based on the Phast system. SAA was cationized essentially as per Example 1, with the following modifications: dialysis against pH 5.5 MES buffer was done with a 2,000 M cutoff dialysis membrane, and the SAA concentration during cationization was 1.6mg/ml. Time points were taken after 5, 15 and 30 minutes. The three cSAA time points possessed a pi greater than 9.3 in all cases, so the three time points were pooled for immunization studies, and are subsequently referred to as cSAA.

Analytical Methods Isoelectric Focusing

Samples of native and cationized protein carriers were analyzed by isoelectric focusing in the PHAST system (Pharmacia) using conditions specified by the manufacturer. The standard curve obtained for isoelectric point (pi) determinations used Sigma standards ranging from a pi of 3.5-9.3. The gels show that a number of cationized species are present in each sample, generally appearing as a broad smear covering a range of pi's. Several bands within this broad range can be distinguished. The change or increase in isoelectric point is used as one of the measures to determine the degree to which a given PCS has been cationized. Where a specific pi value is reported, it is the median of the range. EDA Determination

The cationization of protein carriers covalently couples ethylenediamine (EDA) to the anionic carboxyl groups of the protein. The number of EDA additions was determined in hydrolyzed samples. Concentrated hydrochloric acid, Baker Ultrex, was used for preparation of 6 N HC1 for hydrolysis. Lithium diluent, pH 2.2 (Pickering Laboratories) was used as a sample and standard diluent.

A standard of ethylenediamine dihydrochloride (Aldrich) was prepared by dissolving approximately 26 mg in 50 ml of pH 2.2 lithium diluent yielding a solution of approximately 3.9 mM. A standard of arginine was prepared by dissolving 17.4 mg of arginine in 100 mL of pH 2.2 lithium diluent, yielding a solution of approximately 1 M. A dilute standard was prepared from these solutions by diluting 3.0 mL of the ethylenediamine (EDA) standard and 10.0 mL of the arginine standard to 50 mL with pH 2.2 lithium diluent.

The following reagents were used for the post-column derivatization and chromatographic analysis and all reagents were supplied by Pickering Laboratories unless noted otherwise: lithium eluents, pH 2.75 (Li275), lithium column regenerant (RG003), and trione, ninhydrin reagent. A modified lithium regenerant was prepared by adding lithium chloride to the Pickering regenerant. (This solution was used when necessary to sharpen the ethylenediamine peak). A solution of approximately 5 M lithium chloride (Pierce) was prepared by dissolving 21 mg of lithium chloride in 100 mL of reagent grade water. 25 L of this solution was combined with 175 mL of the regenerant yielding a solution approximately 0.89 N lithium.

Sample hydrolysis was performed using a Pierce Reacti-Therm Heating Module. Chromatograms were obtained using a Dionex BioLC Amino Acid Analyzer equipped with an Autolon Reagent Controller, ninhydrin post-column derivatization unit, an IonChrom UV/Vis detector with a NIN filter, and an Automated Sampler Module. The analytical column used consists of a Pickering Lithium Guard Column, 3x20 mm P/N 0373020.

An accurately known weight (or volume) of sample equivalent to 1-4 mg was transferred to a screw-capped glass culture tube. If the sample was dry, 1 mL of reagent grade water was added. An equivalent volume of concentrated HC1 was added, and the tube vortexed briefly to mix the contents. The tube was placed in a heating module and hydrolyzed for 24 hours at approximately 100°C. The contents were transferred to a 50 ml round bottom flask, the tube rinsed several times with 1 ml of reagent grade water, and the rinsings added to the flask. The hydrochloric acid and water were removed using a rotary-evaporator, high vacuum and a water bath at approximately 60- 80°C. Approximately 5 ml of reagent grade water was added and the evaporation repeated twice. After the final evaporation, the sample was reconstituted with an accurately known volume of pH 2.2 lithium diluent using a volume five times the sample starting volume. After filtering the sample through a 0.45 micron filter to remove any solids, it was placed in an autosa pler vial.

The Amino Acid Analyzer is configured to inject a fixed volume (20 microliters) of standard and sample. The following eluent profile is used to chro atograph the EDA.

Calculation of results was performed against the dilute standard

EDA and arginine solution. Both percent EDA (by weight) and EDA/Arg molar ratio were determined, the latter being used as an internal standard for the sample analyzed. The calculations are shown below:

% EDA = Area of EDA (sample) x mM EDA (std) x 5 ml x 60.10 mo/mMole x IL x 100 Area of EDA (std) 1000 mL x mg sample

EDA/Arg Molar Ratio = (Area of EDA (sample)ZArea of EDA (std) x mM EDA (std)

(Area of Arg (sample)/Area of Arg (std) x mM Arg (std)

Immunologic Methods Experimental Animals: Mice: BDF₁ or Balb/c mice were used for all studies. Mice 6-10 weeks of age were purchased from the Jackson Laboratory, Bar Harbor, ME. Each experimental group consisted of 5 to 10 mice. Rabbits: New Zealand white, females were also used at 5 animals/group.

Adjuvants: Incomplete Freund's. adjuvant (IFA) was purchased from Difco Laboratories, Detroit Ml. Aluminum hydroxide gel was prepared according to the method of Levine and Vaz (Levine, B.B. and N.M. Vaz, Effect of Combinations of Inbred Strain Antigen and Antigen Dose on Immune Responsiveness and Reagin Production in the Mouse. Int. Arch. Allergy Appl . Immunol. 39:156 (1970)) or was in the form of commercial Maalox (Rorer Inc., Fort Washington, PA), or purchased as alum (from Banco). Antibody Measurement: Quantitative ELISA techniques were used to assay the response to various antigens. Standard curves were established each time an assay was performed, using known amounts of antibody raised in response to appropriate native antigens. Variations are described where applicable. T Cell Proliferation Assays: Balb/c mice were injected i.p. with antigen absorbed to alum. The spleens were removed 10 days later and the subsequent cell suspension was passed over a nylon wool column as described (Julius, M.H., E. Simpson and L.A. Herzenberg, A Rapid Method for the Isolation of Functional Thv us-derived Murine Lymphocytes. Eur. J. Immunol. 3:645 (1973)). The nylon wool non-adherent cells were then resuspended in complete RPMI 1640 medium (Gibco Labs) containing 10% fetal bovine serum, ImM glutamine, 5 x 10^"5 M 2-mercaptoethanol , 25 mM HEPES and plated in 96 well flat bottom plates (Falcon Microtest III) at 5 x 10⁵ cells/well. Native or cationized immunogens were added at various concentrations in 20 ul serum-free complete RPMI 1640 to triplicate wells. Serum-free medium served as a control. Cells were incubated in a final volume of 220 μl at 37°C with 5% C0₂ for 72 hours at which time 1 μCi ³H-thymidine was added to each well. Cells were harvested 20 hours later using a cell harvester and radioactivity was determined by liquid scintillation spectrophotometry.

The Effect of Cationization on the Immune Response to Diphtheria Toxoid Alone DT was cationized as described in Example 1. The analytical data for the cationized DT is shown in Table II. There is an increase in EDA additions with increasing time of cationization (and a concomitant increase in surface charge/pl range.

BDF-, mice were immunized intraperitoneally (i.p.) with either 1 Lf or 10 Lf doses of nDT or cDT with alum. The immunizing dose was given on day 0 and the same dose was given as a booster on day 14. All animals were bled prior to immunization and at 14 days after the booster dose, day 28. The antibody at 28 days was measured in an ELISA assay with DT as the antigen on the icrotiter plate. The results are set forth in Table III.

The humoral immune response to the cDT and nDT is shown in Table III. cDT given at the 1 Lf dose showed significantly enhanced immunogenicity with the 20 minute cationization sample, while the 10 Lf dose of cationic DT showed significantly enhanced immunogenicity with the 2 minute cationization sample. Note that at levels of cationization greater than 45 minutes, the measured immune response is significantly less than that to native DT. From this data it would appear that highly cationized antigens (e.g. cl80DT) would offer no immunological benefit either as a primary antigen or carrier.

The effect of cationization of DT on the cellular immune response was also studied. Balb/c mice were immunized i.p. with 10 Lf each of either native DT or DT cationized for 45 minutes (c45DT), adsorbed to alum. After 21 days, T cells were prepared as described above and their ability to respond to either nDT or a variety of cDT forms was examined in a standard T cell proliferation assay. The results are shown in Table IVa.

The above data show that c45DT consistently functions more effectively than nDT as an immunizing antigen. It is also clear that when used as a challenging antigen, all forms of cationized DT produce a greater proliferative response than does nDT regardless of the immunizing antigen. Unlike the results observed with humoral responses to cationization, (eg Table III) there appears to be no plateau of response dependent on the level of cationization (at least within the level of modification tested in this experiment). Each incremental increase in cationization leads to an increase in the proliferative response.

Table IVb examines this relationship between cationization levels and T cell proliferation in a slightly different way. Since the cl80DT seems to elicit the highest response, it has been compared with nDT in both immunizing and challenge antigen roles, and in a dose response study. Animals were immunized and T-cells prepared for a standard proliferation assay as above except that the challenge antigens (either nDT or clδODT) were used in 2, 5, or 20 Lf/ml concentrations.

The data in Table IVb demonstrate the powerful effect that cationization of DT can have on the T cell (or cellular immune system) response. clδODT is more effective for stimulating the T cell population than nDT for a challenge from the native DT as well as stimulating a greater proliferative response of cells which have been immunized with either nDT or clδODT. The full implications of this observation and the role it may play in the elicitation of a maximal immune response in an animal is discussed in a later section.

Enhanced Immunogenicity of Haptens Covalently Conjugated to Cationic Carrier Proteins

Cationized protein containing substances can also be shown to confer enhanced immunogenicity on poorly immunogenic or nonimmunogenic molecules via its role as a carrier. This enhancement can be observed in humoral as well as cellular responses. Immunogenicity of Haemophilus Influenzae PRP Conjugated to Native and Cationic PJ

PRP-carrier conjugates prepared as in Example 3 were adsorbed to alum by rotation at room temperature for 5 hours (15.9 mg alum/mg immunogen). The concentration of the PRP-conjugates was then adjusted to 25 ug/ml carrier protein with PBS. Rabbits (New Zealand white females, 5 per group) were prebled, then immunized with 1 ml immunogen per animal, given as two separate 0.5 ml intramuscular injections. Samples were taken on days 14 and 28 following immunization, followed by boosting with the same immunogens on day 28 (week 4). Additional samples were taken 7 and 21 days later (weeks 5 and

7).

A quantitative ELISA was used to determine the anti-PRP antibody levels for individual rabbit samples. PRP containing antigen (Merck PedVax Hib vaccine) was coated at 1 ug PRP/ml onto microtiter plates. Wells were blocked by incubation with assay diluent for 1 hr at 37°C and washed prior to sample addition.

As shown in Table V, both nDT-PRP and cl80DT-PRP conjugates induced similar, low titer anti-PRP antibody in the primary response (weeks 2 and 4). Following a booster inoculation at week 4, the clδODT-PRP conjugate demonstrated an enhanced anti-PRP memory response at weeks 5 and 7, whereas the titer generated by the nDT-PRP conjugate booster inoculation at week 4 had a modest response at weeks 5 and 7. Enhanced Immunogenicity of Homologous Ionic Complexes-Unmodified Plus Modified Antigen (nDT/cDT)

As noted above, there are differential effects of different levels of cationization on humoral and cellular immune responses (See tables III & IV). One can take advantage of those differences to produce optimal carrier- induced responses for a variety of antigens. The relationship between degree of cationization of a protein and its ability to induce augmented levels of anti-native antibody in the in vivo murine model system can be represented as a bell shaped curve. Initially, increasing EDA incorporation is associated with an increased ability to induce anti-native antibody, which eventually plateaus. Further increases in the pi of the protein by continued increase in EDA incorporation are associated with a decrease in antibody immunogenicity, which may actually fall below that induced by the corresponding native protein. Specifically, for nDT, enhancement is observed for low EDA incorporation, while high incorporation, e.g. clδODT, is associated with significant suppression of anti-nDT antibody production (See Table III).

These results contrast surprisingly with those derived from studies of the murine T cell response to cDT. T cells isolated from spleens of nDT-im une mice exhibit enhanced proliferation rn vitro to cDT, which is directly proportional to the extent of cationization (See Table IV). Thus, for the T cell component of the immune response, clδODT functions as the most efficient activator of cellular function tested, unlike the suppression noted for the in vivo antibody response. These results clearly demonstrate differential effects of cationization on cell -mediated and humoral immune responses to cDT.

These findings were interpreted in light of current information on the nature of T cell and B cell epitopes of macromolecules. Protein sequences recognized by cell surface Ig receptors on antibody-forming cells

(B cell epitopes), tend to be relatively complex, and are often dependent on both amino acid sequence and three-dimensional protein conformation. Chemical modification of proteins, which at high levels of substitution may alter conformation, may also have deleterious effects on the recognition of B cell epitopes characteristic of the native protein. In contrast, T cell epitopes tend to be linear sequences of amino acids and are, therefore, less sensitive to changes in protein conformation. Hence, T cell epitopes are preserved with high levels of excessive cationization, while B cell epitopes are not effective im unogens when highly cationized. One aspect of the present invention resides in the discovery that high levels of cationization that selectively inhibit B cell epitopes are potent effectors of T cell immunity.

Another aspect of this invention is the discovery that highly cationized antigen (above a pi of 9.3), which by itself fails to induce antibody (presumably due to effectively absent B cell epitopes) can, by virtue of its ability to induce T cells, enhance the production of antibody to native antigen when animals are immunized with the combination of native and highly cationic antigen. The prior art does not disclose or suggest that an immunogen consisting of a highly cationized antigen complexed with its native counterpart would lead to an enhancement of the production of anti-native antigen antibody. The following experiment using a highly cationized DT-nDT homologous carrier-antigen system tests this discovery.

Specifically, an admixture of nDT and cl80DT (pi greater than 9.3) was prepared as described in Example 4. Ten female Balb/c mice per group were immunized interperitoneally on day 0, and boosted with the same immunogen preparation on day 14. Serum samples were obtained by retroorbital bleeding of individual animals 1 and 3 weeks following the booster, and anti-nDT antibody titers determined by quantitative ELISA.

For determination of anti-nDT antibody titers in individual murine serum samples, a quantitative solid phase ELISA was employed. Microtiter plates (Corning) were coated with lOOμl/well DT (Connaught) diluted to 10 μg/ml with sodium carbonate-bicarbonate buffer (pH 9.5-9.7). Background wells were coated with lOOμl/well buffer only. Serum samples, 100 μl/well, diluted in assay diluent (PBS-1%, BSA-0.1%, Tween 20, 0.011% thimerosal) were added and incubated for 1 hour at 30°C. A control murine anti-DT serum was assayed in parallel on each plate, to control for plate-to-plate and day-to-day assay variation. Following the incubation, plates were washed with water, and 100 μl/well of goat-anti-mouse IgG-horseradish peroxidane (gamma chain specific, Kirkegaard & Perry), diluted to 133ng/ml in assay diluent, was added, followed by a 1 hour incubation at 30°C. The plates were then water washed, and lOOμl/well of OPD substrate (Abbott) dispensed to all wells. The enzyme- substrate reaction was stopped by addition of 100 μl/well of IN H₂S0_A (Abbott). Optical densities (0D) were read at 490nm on a BioTek EIA reader. All plating and serial dilutions were performed by a Biomek 1000 workstation.

Calculation of anti-nDT antibody titers was done by importing raw 490 values into a data reduction Lotus 1-2-3 spreadsheet, which calculated average 0D minus background, standard deviation and percent coefficient of variation within a duplicate set of wells. A regression line was generated for each sample, and the anti-nDT titer, defined as the inverse of the serum dilution yielding a net 0D of 1.5, calculated. The regression generated slope was also compared to the slope of the control serum, to verify curve parallelism.

A nonparametric test, described by Hochberg and Tamhane (Hochberg, Y. and Tamhane, A.C. (1987), Multiple Comparison Procedures. Wiley, New York, p. 235), was used to compare the anti-nDT response of the 1 Lf nDT group to that of animals receiving nDT complexed with clδODT.

As observed in the data in Table IX, 1 Lf cl80 DT (group 2) fails to induce a significant, lasting anti-nDT response, as compared to the same amount of nDT (group 1) as an immunogen. However, addition of various concentrations of clδODT to suboptimal concentrations of nDT results in production of anti-nDT antibody responses which are significantly greater than those with nDT alone. These results demonstrate that immunization with homologous complexes such as nDT-clδODT results in a synergistic enhancement of the anti-nDT antibody response, despite the inability of clδODT to induce significant anti-nDT antibody on its own. This is consistent with clδODT providing an effective T cell potentiation and facilitating the B cell response to epitopes of to the native protein. Enhanced Humoral Immune Response of nOVA-cOVA Complexes

Table X shows the antibody titer observed following inoculation of O.lμg nOVA; O.lμg c300VA; or a combination of O.lμg each. nOVA or cOVA was inoculated i.p. in the presence of alum, or individual mice received two sequential inoculations (one dose each, nOVA and cOVA) . Mean antibody titers (as determined by standard ELISA) were obtained at specified days following the primary inoculation.

As the data evidence, the instant discovery provides a highly unexpected enhanced immune response when an animal is immunized with a highly cationized antigen mixed with its native counterpart.

Enhanced anti-SAA Production by nSAA - cSAA Complexes

The cSAA prepared in Example 5 was used to demonstrate that homologous complexes according to this invention yield unexpected antibody production. New Zealand white female rabbits, 5/group, were used for these studies. Animals were prebled, and immunized on day 0 with 100 ug each nSAA, cSAA or lOOug nSAA+lOOug cSAA, given in a 1 ml/animal volume i.p., with alum.

Animals were boosted on day 28, and invididual bleeds taken 2 weeks later

(IM2+2). Two weeks later, the rabbits were boosted again, and an additional bleed taken 2 weeks later (IM3+2). Individual rabbit bleeds were titered for anti-SAA activity using a solid phase microtiter EIA, with highly purified native SAA as the plate coat antigen. All samples were diluted in a normal human serum-containing sample diluent, to eliminate potential reactivity with non-SAA serum components. Titers were expressed as the reciprocal dilution of serum which gave a 0.6 OD in the ELISA. The results are set forth in Table XI.

From the data contained in Table XI it is quite apparent that both cSAA and the mixture show increased antibody over native at IM2+2L. More importantly there is a clear increase in anti-nSAA with the mixture, over both nSAA and cSAA immunogens at the IM3+2 time point.

Industrial Applicability This invention provides a means of greatly improving the efficacy of hapten-carrier complexes as vaccines. Although hapten-carrier conjugate vaccines are known to produce long lasting immunity for certain antigens, the current approaches are not adequate for all antigens desired. Cationization of the "carrier" molecule (especially to a pi greater than 9.3) as described herein has been shown to enhance both the B and T cell responses of an immunized animal to a specific antigen or "hapten." Application of this technique should not only improve the efficacy of currently available hapten- carrier conjugate vaccines, but should also permit the development of effective vaccines for otherwise intractable antigens. There is a need in the industry for methods to produce novel vaccines to pathogens which have not previously produced an effective antibody response. There is also a need for new techniques which allow for effective immunizations with lower quantities of rare or potentially hazardous antigens. The present invention opens the door for production of vaccines to protect humans and animals from previously unreached infectious diseases. It also provides a means of generating high levels of antigen specific antibodies and other im unologically significant molecules for commercial and research purposes.

This invention has been described in detail with particular reference to preferred embodiments thereof, but it will be understood that variations and modifications can be effected within the spirit and scope of the invention.