WO1989010755A1

WO1989010755A1 - Immunogenic complex and its use in vaccination

Info

Publication number: WO1989010755A1
Application number: PCT/GB1989/000475
Authority: WO
Inventors: Richard Edward Randall; Dan Frushard Young
Original assignee: National Research Development Corporation
Priority date: 1988-05-05
Filing date: 1989-05-04
Publication date: 1989-11-16
Also published as: GB8910262D0; GB2220140A; AU3565289A

Abstract

A support material-antibody-antigen (SMAA) complex for use in vaccination, comprising: 1) a support material capable of being taken up by antigen-presenting cells or attached to antigen-responding cells of the immune system, e.g. Staphylococcus aureus Cowan Strain A or inert beads, 2) antibodies of at least one kind bound to the support material, and 3) immunostimulating antigens of at least one kind directly or indirectly bound to the antibodies. Preferably different antigens of the same infective agent, such as surface and internal proteins of a paramyxovirus, are bound to their specific antibodies.

Description

IMMUNOGENIC COMPLEX AND ITS USE IN VACCINATION Background of the invention

1. Field of the Invention

This invention 1s in the fields of immunology and vaccination.

2. Description of the prior art Most vaccines currently available contain whole killed or live attenuated Infectious agents (herein termed "microbes" for brevity) as the immunizing antigens. However, for some diseases such an approach has not so far been successful or is not practical (e.g. for human immunodeficiency virus and malaria). There is a need for alternative types of vaccine against diseases and to improve on the efficacy and safety of existing vaccines. Alternatives proposed include the construction of microbial chimeras in which a single antigenic determinant from one microbe is Inserted into a non-critical position on the surface structure of another microbe (e.g. polio virus chimeras, K.L. Burke et al. , Nature, 322, 81-82 (1988)) and the attachment of viral proteins to a glycoside matrix to form immunostimulating complexes (ISCOMs). Immunogenlc peptides are also being manipulated to Improve their antigenicity, e.g. by coupling T helper cell determinants onto B cell determinants. However, when designing vaccines it is often important to induce both appropriate humoral and cell-mediated immune responses. For example, it may be necessary to include multiple antigens from the same microbe in some vaccines. Thus, the major target antigens for B cells are usually conformational epi topes which may be on different microbial proteins from those containing epitopes required for T cell responses. Further, different individuals of a heterozygous population may recognise different T cell target antigens depending on the individual's major histocompatibility complex (MHO status. Consequently, to ensure that all individuals respond to a particular vaccine, multiple T cell antigens may be required. A desirable property of vaccines will also be their ability to induce long term immunological memory. Thus vaccines that induce a broad immune response, including a T helper cell response, may be beneficial in situations when the level of antibodies and/or cytotoxic T cells in a patient at the time of infection may not be sufficient for protective immunity but where a rapid secondary immune response can prevent the development of disease symptoms. In this respect it has been shown in a number of viral systems that T helper cells which recognise target antigens on internal virus proteins may enhance B cell responses to surface antigens present on the same virus particle through the mechanism of B cell antigen presentation. Thus, vaccination schemes which prime a B cell response to only one antigenic determinant, while inducing an antibody response to that antigen, may not prime the immune system for a rapid secondary immune response. Immunization strategies need to be developed that are capable of presenting multiple microbial antigens (proteins, protein fragments or peptides) in an appropriate way to the immune system. Virus chimeras and conventional ISCOMs are not suitable for this purpose.

Additional prior art is mentioned in a separate section after the "Summary of the invention", without which its context would not be clear.

Summary of the invention

We have developed a method of antigen presentation which is applicable to multivalent vaccination and 1n certain cases is likely to modulate specific immune responses in relation to a particular infection. Our invention rests on the idea of linking the antigen which it is desired to present to a support material, for example solid particles, via an antibody. Conveniently, although not necessarily, the antibody is raised against either that antigen, or, for example if an antibody cannot be raised against it in a human or a particular animal, some "intermediate" antigen to which the final presented antigen is ultimately linked. Thus, the products with which this invention is concerned have as essential elements (1) a support material compatible with the use of the product in vaccination, (2) an antibody as a "bridge" and (3) the antigen which It is desired to present to the immune system.

While various forms of binding between the support material and the antibody and between the antibody and the antigen are possible, the invention requires at least one immunological linkage. It is therefore appropriate, for brevity, to call the products of the Invention "complexes".

The support material - antibody - antigen (SMAA) complexes of the invention have important advantages, for example: - they can be used to present two or more different antigens at the same time the SMAAs tested induce a stronger immune response than the purified antigen alone on an alum adjuvant the SMAAs tested induce humoral and cell-mediated immunity at least the simpler SMAAs are convenient to prepare, because the antigen can be purified in an affinity interaction, e.g. on a column, with supported antibody and the resultant complex used for vaccination as it is (without any subsequent step of dissociating the support from the antibody or the antibody from the antigen).

SMAA complexes of the invention are believed to be taken up in some way, e.g. by phagocytosisor pinocytosis, by the antigen-presenting cells. On this basis it is possible to define the SMAA complexes as comprising:

(1) a support material capable of being taken up by antigen-presenting cells or attached to antigen-responding cells of the immune system,

(2) antibodies of at least one kind bound to the support material, and

(3) immunostimulating antigens of at least one kind bound to the antibodies.

The term "antibody" used herein includes monoclonal antibody (MAb) (preferred), immunoglobulin molecules (natural or man-made chimeras), binding fractions thereof and polyclonal antisera.

In this invention the term "bound" (or any other part of the verb to bind) is not to be taken as implying direct linkage except where the context obviously so requires. The preferred support materials are biodegradable microcarriers, e.g. inactivated or killed bacteria Staphylococcus aureus Cowan Strain A or protein A, or a biodegradable polymer such as cross-linked polylysine or biologically inert beads.

The SMAA complexes of the invention are, 1n their broadest generality, known per se from prior art cited above. To the extent that they are known per se they do not form part of this invention. However, many embodiments of the SMAA complexes of the invention are not known per se, in particular those in which an "intermediate" antigen or hapten is incorporated and such complexes are per se part of the present invention.

Although some of the SMAA complexes are known per se, their use in vaccination is believed to be novel. Accordingly, the invention resides in a first medical use (Europe, Japan) or a method of vaccinating an animal (including a human) which comprises administering to the subject a protection-inducing amount of the antigen in the form of its SMAA complex of the invention (USA).

The invention also includes a sterile pharmacological composition, especially an aqueous injectable solution, comprising a SMAA complex of the invention and a physiologically acceptable diluent or carrier. An adjuvant is optional.

Further description of prior art

The capacity of Staphylococcus aureus Cowan Strain A ("St.A" for brevity) and the Protein A present in its cell walls, for binding to immunoglobulin G is well known and has been used to isolate or purify antigens, see e.g. R.D. Ivarie and P.P. Jones,

Analytical Biochemistry 97, 24-35 (1979).

It is known that St.A is a B-cell mitogen, see

R.J.M. Falkoff et al., Journal of Immunology 129, 97 (1983) and E. Celis et al., Hepatology 7, 563-568 (1987). It is also known that St. A binds to IgG (via its Fc protein) in antibody-antigen complexes preferentially to free IgG. By this means antibody-antigen complexes formed in vivo, which normally competitively inhibit antibody-dependent cellular cytotoxicity, are prevented from doing so, F.M. Cowan et al., Biomedidne 30, 23-27 (1979). W.M. Siag and J.M. Jones, Clinical Immunology and Immunopathology, 24, 186-193 (1982), sought to investigate the influence of protein A on circulating immune complexes and refer to six other reports on the effects of protein A on the Fc protein of IgG, which are self-conflicting. In their own experiments, an immune complex of radioiodine-labelled murine leukemia virus protein p30 and anti-p30, with and without the prior addition of protein A to the complex, was injected into rats, the animals were sacrificed and their radioactivity in various organs was measured. This paper therefore relates purely to an experimental study of immune complexes and does not propose vaccination of patients with SMAAs.

Experiments have been carried out using (unsupported) antibody-antigen complexes as immunogens. See, for example, D.H. Watson and P. Wlldy, Journal of General Virology, 4, 163 (1969) and J.A. Berzofshy et al., Nature 334, 706 (1988).

M.E. Johansson et al., Journal of Immunological Methods 26, 141-149 (1979) describe the preparation of specific antisera against adenoviruses in which polyclonal antiserum to adenovirus (containing mainly high avidity antibodies directed largely against adeno group antigen) was bound to CNBr-activated "Sepharose" beads and reacted with cellular extract of adenoviruses of various types and the bound product washed. The affinity bead-antibody-antigen complexes were then used to immunize rabbits and antiserum taken from the rabbits was examined by an Indirect immunofluorescent assay. The resultant antisera were found to be highly specific to the type of the adenovirus antigen. This immunisation procedure for raising specific antisera was compared favourably with the then conventional procedure of Immunising animals with immune precipitates produced in agarose. Careful study of this reference Indicates that it is solely concerned with producing antisera of high type specificity, at a time when monoclonal antibody technology was in Its infancy. Nowhere is there any suggestion that the bead-anti body-antigen complexes would induce a protective Immune response to the antigen, as is required 1n vaccination. PCT Application WO 88/08429 (Biogen NV), filing date April 20, 1988, published November 3, 1988, aims to treat immunodeficient patients, such as those infected with HIV, by administering an immunogen covalently coupled to a T-cell-independent carrier, exemplified by Ficoll, LPS, dextran sulphate or St. A. It is perhaps a noteworthy indication of the unobviousness of the present invention that this almost contemporary patent application nowhere suggests that the antigen be coupled to the carrier via an antibody. Brief description of the drawings Figure 1, referred to in Example 1, part (b), reproduces an SDS-PAGE gel of stained protein bands of antigens present in SMAA complexes of the invention;

Figure 2, referred to in Example 1, part (c) reproduces an autoradiograph of radiolabelled immune precipitates formed by immunizing mice with SMAA complexes of the invention and reacting the antisera of the immunized mice with extracts containing a mixture of antigens;

Figure 3(a) and (b), referred to in Example 1, part (d), reproduces autoradiographs of Western blots of SDS-PAGE gels of lung extracts of mice which have been immunized with SMAA complexes and then challenged by an aerosol route;

Figure 4, referred to in Example 2, reproduces an autoradiograph in the manner of Figure 2, but relating to immunization with multivalent SMAA complexes of the invention; Figure 5, referred to in Example 3, part (b), reproduces an autoradiograph in the manner of Figures 2 and 4, but relating to immunization with other multivalent SMAA complexes of the invention; Figure 6, referred to in Example 7, reproduces an SDS-PAGE gel in the manner of Figure 1, but relating to SMAA complexes in which the antibody is anti to an "intermediate" hapten and this hapten is linked to the immunostimulating antigen;

Figure 7, also referred to in Example 7, reproduces an autoradiograph of a microtitre plate RIA showing that mice immunized with the SMAA complexes to which Figure 6 relates produce antisera which react with the specific antigen;

Figure 8, referred to in Example 8, reproduces an SDS-PAGE gel in the manner of Figures 1 and 7, but relating to a SMAA complex in which the antibody is anti to an "intermediate" hapten and this hapten is linked via a spacer arm to a chimeric peptide.

Figure 9, also referred to in Example 8, reproduces an autoradiograph of a microtitre plate RIA similar to that of Figure 7, but relating to the SMAA complex of Figure 8; Figure 10, also referred to in Example 8, reproduces an autoradiograph of an RIA of a Western blot of SDS-PAGE-separated polypeptides from cells which were infected with a virus and then reacted with various antibody preparations, including sera from mice immunized with the SMAA complex of Figure 8; and Figure 11 shows schematically the incorporation of a fusion polypeptide in SMAA complexes of the invention.

Figures 1-3 are Figures 3-5 of R.E. Randall, D.F. Young and J.A. Southern, J. Gen. Virology 69, 2517-2526 (1988), Figure 4 is Figure 8 of R.E. Randall and D.F. Young, J. Gen. Virology, 69, 2505-2516 (1988) and Figure 5 is Figure 1 of R.E. Randall and D.F. Young, J. Virology 63 (No.4) 1808-1810 (1989). All these papers were published after the first priority date of this application. Description of the preferred embodiments

The support material can be any which is normally insoluble in aqueous immunological media and to which antibodies may be bound. These support means can be in the form of particles, for example beads, or in sheet form or in the form of rods, fibres or the like. Suitable support means include, for example, spherical particles such as biologically inert, especially polymeric, beads such as "Sepharose" or "Dynabead" magnetic beads or particles of biodegradable microcarriers, e.g. of killed bacteria such as Staphylococcus aureus Cowan Strain A.

The antibodies may be bound to the support material by any conventional method. Such methods include, for example, chemical bonding whereby the antibody is chemically bonded to the support means, or non-chemical absorptive bonding. It is however essential that some of the antibody should be able to "dangle" clear of the support material and not become sterically hindered by it, in order that the antibody shall bind to its partner antigen.

The size of the support material is not critical within reasonable commonsense limits. The support material must be capable of being taken up by antigen-presenting or attached to antigen-responding cells, e.g. macrophage-like cells and lymphocytes. There is virtually no lower limit on its size, which can be sub-micron. Thus, for most practical purposes a range of 0.1 to 5 microns will ordinarily be the most suitable. Larger particles are usable but are generally less preferred and are probably less well taken up by many of the cells.

The antibodies used in the present invention are preferably monoclonal antibodies; however other types of antibodies, for example, polyclonal IgG or antisera, may also be used. The antibodies, used may come from any animal species, for example, mice, rabbit, human or be chimaeric antibodies (e.g. human/mouse, Luna, rat) and may even be mixtures of antibodies.

A convenient feature of the invention is that the supported antibodies can be used to purify, from a mixture, an antigen which will bind to the antibodies. This is, of course, not novel per se as a generality, but the idea of going on to use the resultant supported antigen-antibody complexes, with the support means still attached, for the purpose of vaccination, is novel. For example, a mouse monoclonal antibody can be raised to a specific antigen, for example, the matrix (M) protein of the simian virus 5 (SV5). This mouse monoclonal antibody is then bound to Staphylococcus aureus which acts as the support. The mouse monoclonal antibody bound to Staphylococcus aureus can then be used to purify M protein from a homogenate of SV5 by passing the homogenate over the Staphylococcus aureus with the antibody bound to it. The only material that should bind to the antibody on the Staphylococcus aureus should be the M protein. Suitably the antigen is obtained from the microbe, e.g. virus or bacterium, against which protection is to be induced. However, synthetic peptides produced e.g. by recombinant DNA technology or by chemical synthesis can also serve as antigens for the purposes of this invention. Conveniently, the immunostimulating antigen is bound directly to the antibody by an ordinary immunological linkage.

The SMAA complexes can contain a single kind of antibody or a variety of different antibodies. In the latter case, there are various possible configurations. For clarity of discussion it is convenient to refer to "particles" of the support material, without prejudice to analogous use of other finely divided forms. Each particle can have one, two or more molecules of antibody attached to it. Thus, a single particle can have many different kinds of antibody attached, each being anti to a different biological "factor" or epitope. Alternatively or additionally, the SMAA complexes can be mixtures of support particles, in which all the antibodies present on any individual particle are of the same kind, but in which different kinds of antibodies are attached to different individual particles. In a preferred embodiment the complexes contain two or more different kinds of antibodies, in order easily to provide two or more different kinds of immunostimulating antigen. It is believed that a more appropriate immune response may be elicited when more than one antigen of a particular infectious agent, for example a virus, is presented in vaccination. Thus, in some cases an enhanced immune response can be obtained when the SMAA complex contains antigens derived from both surface and internal viral proteins, than from merely surface antigens. For example a F or HN surface antigen and M or NP internal antigen of a paramyxovirus such as Simian Virus 5 or Newcastle Disease Virus could be embodied in the SMAAs.

When an antibody can be easily raised against a particular desired immunostimulating antigen, it will usually be convenient to prepare SMAA complexes by directly coupling the antigen to its specific binding antibody. However, this procedure might not always be convenient or possible. This invention has the great merit that one or more molecules of immunostimulating antigen can be bound indirectly to a single antibody molecule, thereby providing a multiple copy (same kind of antigen) or multivalent (two or more different kinds of antigen) immunogen. Thus, according to a feature of the invention, at least one immunostimulating antigen is bound to an "intermediate" antigen or hapten which is in turn bound to the antibody in the SMAA complex. Virtually any molecule can serve as a hapten for this purpose, the only requirements being that an antibody must be able to be raised against it and that it can be bonded to the immunostimulating antigen(s). Where two or more molecules of antigen are to be covalently bonded to one molecule of the intermediate hapten, either the hapten should itself be polyvalent (e.g. DNP or a reactive derivative thereof) or it should be bonded to another molecule which is polyvalent. In other words, the linkage between the anti-hapten antibody and the immunostimulating antigens will at some point be branched. In another sub-embodiment, the "intermediate" antigen or hapten is fused directly to the immunostimulating antigen. For example a fusion polypeptide can be made, by recombinant DNA means or chemical synthesis, comprising a carrier polypeptide, serving as an "intermediate" antigen or hapten element, directly bonded to an immunostimulating or target polypeptide or protein. An antibody raised against the intermediate element is bonded to the support material. Figure 11 of the drawings shows schematically the incorporation of up to four different fusion polypeptides in a SMAA complex of the invention. Each polypeptide is produced in the conventional manner from an expression vector comprising a DNA sequence coding for a carrier polypeptide joined to DNA sequence coding for the desired polypeptide. The above sub-embodiment must be approached with care. In some cases, the antibody might be so spatially near to the immunostimulating antigen that it sterically hinders its functions. According to another very valuable and advantageous feature of the invention, therefore, the immunostimulating antigen is linked to the intermediate antigen or hapten via a spacer arm, that is to say any innocuous molecule or combination of molecules which provides the desired distance between the antibody and the immunostimulating antigen. It is Impossible to lay down an all-embracing definition of what the length of the spacer arm should be. It will depend on very many factors such as rigidity of the spacer arm, preferred conformations of the antibody and immunostimulating antigen, their molecular sizes, their precise configurations, the distance between the point of attachment of the spacer arm and the binding epi topes of antibody or antigen and so on. It is a matter of experimentation in any particular case, by shortening or lengthening a trial arm. Spacer arms are not a new concept in biological or pharmacological chemistry and some guidance is therefore obtainable from the literature, particularly in diagnostics. Although not a new concept per se, the realisation of their particular "synergy" with the SMAA complex concept is believed to confer additional inventivity on this embodiment of the present invention. In Example 8, a chimeric peptide made of a herpes simplex virus polypeptide fused to a poliovlrus polypeptide is linked to the hapten DNP and thence via an anti-DNP antibody to the support material. In this embodiment, there is a spacer arm provided by the amino acid lysine linked to the poliovlrus polypeptide sequence between the DNP and the HSV polypeptide. (The poliovirus polypeptide was chosen as a matter of experimental convenience, not as an ideal form of spacer arm). Simply bonding the HSV polypeptide to the DNP will not work : the MAb to the DNP seems to block an epitope on the HSV polypeptide. In the light of this result, a spacer arm at least 5 or 6 amino acids long between the hapten and the peptide of interest seems advisable, but, as explained above, different lengths will be appropriate in different circumstances.

The spacer arm need not be biologically inert. It can for example contain a protease cleavage site, whereby the immunostimulating antigen could be cleaved in vivo from the remainder of the SMAA complex, or it could contain an antigenic determinant which influences the desired immune response.

Reverting to the generality of the invention, the antigens selected for incorporation in SMAA complexes are not critical. As will have been appreciated from the above account, just about any immunogenic antigen can be linked by some means, whether wholly by an immunological linkage or partly by covalent bonding and partly immunologically (in any combination thereof), to the support material. The invention is particularly applicable to antigens of infectious agents such as viruses, bacteria, parasites and fungi, but it may be desirable, alternatively or additionally, to introduce other antigens which are not present in infectious agents. It may be possible to enhance the immunogenicity of SMAA complexes of the invention. Thus, SMAA complexes can be constructed which additionally have free antibodies attached. (By "free antibodies" is meant antibodies not bound to any antigen or hapten). Thus, the SMAA complexes may also incorporate antibodies to antigens on the surface of some of the cells of the animal (host) being immunized. These antibodies will bind the SMAA complexes to specific cells in the host, thus targetting the SMAA complex onto these cells, for example leucocytes, thus enhancing the immune response. These particular SMAA complexes may be useful in the development of oral vaccines for the treatment of certain respiratory diseases.

Also, it should be possible to modulate one arm of the immune response, that is the T or B cell response, by incorporating antibodies into the SMAA complexes which may lead to increased activation or proliferation of T or B cells. For example anti-CD3 (CD3 is an antigen found on the surface of human lymphocytes) Incorporated into the SMAA complexes may induce local non-specific activation of T cells and the release of lymphoklnes, which may act upon any B cell clones specific for the immunostimulating antigen of interest.

Also, the SMAA complexes can have a protein such as a cholera toxin bound to the support means, with or without any intermediate antibody in the binding linkage, for enhancing a particular immune response, e.g. immunity in the gut, see e.g.C.O. Elson and W. Ealding, J. Immunology, 133, 2892-2897 (1984).

The invention is useful for the vaccination of animals including humans. Normally the antibody incorporated in the SMAA complexes will be one which is raised in or derived from an animal of the same species as is to be vaccinated by the present invention. For example, a human antibody is preferably used in a SMAA complex intended for the vaccination of humans. It is possible to use antibodies from a species other than that of the subject to be vaccinated. For example, mouse antibodies can be used in SMAA complexes to treat rabbits. SMAA complexes incorporating mouse antibodies have been shown to produce a vigorous immune response in rabbits to the antigen contained in the SMAA complexes. It might not be acceptable, however, to use in humans SMAA complexes which contain antibodies derived from animals other than humans. There are many solutions to this problem. One is to "humanise" the desired antibodies by the recently invented methodology. Another is to attach a known "intermediate" antigen ("antigen K"), against which a human antibody has been raised and therefore exists or will be raised, to the immunostimulating antigen ("antigen N") against which It is desired to create an immune response, for which no human antibody exists. Methods of attachment by use of haptens, fusion polypeptides etc., have been described above.

To prepare SMAA complexes of the invention, the antibody is bound to the support material by any conventional method familiar in the immunoaffinity column or diagnostics fields. In the case of St.A, one incubation will suffice, while in other instances the support material has to be reacted with a linker to covalently bond it to the antibody. The antigen is bound to this supported or immobilised antibody, using conventional techniques, for example, mixing the supported antibody with a homogenate of the organism being immunized against. The supported antibody will then selectively bind specific antigens from the homogenate. The rest of the homogenate can be removed from the mixture by conventional means such as filtration and washing, leaving a preparation comprising the antigen bound to the antibody which is in turn bound to the support material.

Where the SMAA complex contains an "intermediate" hapten or antigen it will frequently be necessary first to prepare a compound of the general formula:

(intermediate antigen or hapten) - spacer arm* - (immunostimulating antigen)_n,

where n is 1 or more and the asterisk denotes that the spacer arm is optional according to circumstances. An antibody to the intermediate antigen or hapten, usually raised merely against the unbound intermediate, is then bonded to the support material and then to the above-prepared compound.

Typically the molar ratio of the antigen which is directly bound to the antibody is approximately 1:1. This can be achieved merely by use of excess of the antigen in the immunological reaction. However, effective immune responses have been elicited when the ratio has been 1:10 or lower.

The amount of SMAA complex administered as a vaccine may be any which will provide a conventional amount of antigen, according to the protocol of vaccination, e.g. whether it is a priming dose or main dose, and route of administration, e.g. i.v., i.p., i.m., oral etc. chosen, and according to the size of the animal being treated. The SMAA complexes of the present invention do not normally need to be administered along with a conventional adjuvant.

Often the support material enhances the immune response that would otherwise be obtained with the antibody-antigen complex on its own. It is thought that this may be because the support means triggers the complement system, in the animal being immunized, which is thought to enhance the humoral Immune response and may also play a role in the generation of memory B cells. At the same time the support material appears to be taken up by antigen-presenting cells and this may aid in the presentation of antigens to lymphocytes involved in generation of both humoral and cell mediated immunity.

The present invention will now be illustrated by way of

Examples.

EXAMPLE 1 Preparation and use of SMAA complexes of Simian Virus 5 antigens and antibodies

Introduction

Simian virus 5 (SV5) has six major structural proteins, namely the haemagglutinin-neuraminidase (HN), fusion (F), matrix (M), nucleo-(NP), phospho-(P) and large (L) proteins. The HN and F proteins are glycosylated surface proteins that protrude from the virus envelope and it is against these proteins that neutralizing antibody is directed. The other four structural proteins are internal. The NP, P and L proteins, together with the single stranded genomic RNA, make up the nucleocapsid complex, while the M protein is located between the nucleocapsid and the virus envelope.

(a) Preparation of Staphylococcus aureus Cowan Strain A - SV5 protein monoclonal antibody - SV5 protein SMAA complexes The immunoglobulin binding sites on St.A were saturated separately with MAbs to each of five SV5 proteins in turn by mixing equal volumes of a suspension of St.A (10% w/v) with ascitic fluid containing the SV5 protein-specific MAb (anti to HN, F, NP, M or P protein) for 2h at 4°C. The five St.A-bound antibodies were washed three times by sedimentation from (3,300 g for 3 min) and resuspended in phosphate buffered saline (PBS). Soluble antigen extracts were made from SV5-infected BHK cells as described by R.E. Randall, D.F. Young, K.K.A. Goswami and W.C. Russell, Journal of General Virology 68, 2769-2780 (1987) and these extracts were then mixed continuously with the St.A-bound antibody (20 μl of a 10% w/v St.A-bound antibody per 2 × 10⁸ cell equivalents of antigen extract) for 4-6h at 4°C. The excess antigen used normally saturate the MAbs, thus ensuring, after washing, an approximately equimolar ratio of antibody to antigen in the complexes. The resulting St.A-antibody-antigen complexes were washed three times by sedimentation (3,300g for 3-10 mins) and resuspended in immune precipitation buffer (20 mM Tris. HCl , pH 7.2, 5 mM EDTA, 0.5% Nonidet P40, 0.1% sodium dodecyl sulphate and 0.65 M NaCl) and then washed twice in PBS. (b) Dissociation of the antigens from the complexes

(for the purposes of verifying their unique attachment to the MAbs in the complexes)

The antigen-antibody complexes were dissociated from the St.A by heating at 80°C for 5 min in disruption buffer (0.05 M Tris. HCl , pH 7.0, 2% sodium dodecyl sulphate, 5% 2-mercaptoethanol and 5% glycerol), the St.A was removed by sedimentation (6,000g for 3 min) and the dissociated polypeptides separated by electrophoresis through a 15% SDS polyacrylamde mini-slab gel. The protein bands were stained with Coomassie Brilliant Blue. Fig. 1 is a photograph of the stained gel showing that the HN (slot 1), F (slot 2), NP (slot 3), M (slot 4) and P (slot 5) proteins of SV5 were isolated from the SV5 extract by the MAbs, the proteins being attached uniquely to their respective MAbs. The positions of the SV5 polypeptides are shown in the left-hand column and the positions of the antibody heavy (IgH) and light (IgL) chains in the right-hand column, (c) Immunization of mice with SMAA complexes and testing of the resultant mouse sera

Groups of three mice were immunized twice intraperitoneally with SMAA complexes prepared as in part (a) above containing 2-5 μg (NP, M, HN, P) or 0.1-0.2 μg (F) of virus protein attached to the antibody. A control group was given SV5 purified as described by R.E. Randall et al., loc. cit. (1987). The mice were injected with 200 μl aliquots of 0.5% w/v suspensions of the SMAA complexes, with a gap of 3-4 weeks between the first and second immunizations. Ten days after the second immunization the mice were bled from the tail vein and the sera tested for their ability to immune-precipitate SV5 proteins. Immune precipitates were prepared by reaction of these sera with soluble antigen extracts of SV5-infected BHK cells, these cells having been [35S] methionine-radiolabelled by the method of R.E. Randall et al., loc. cit.. (1987).

[³⁵S] methionine-labelled polypeptides were separated from the immunoprecipitates by electrophoresis through a 15% SDS polyacrylamide slab gel. Referring to the autoradiograph of Figure 2, the lanes corresponding to the antisera to purified SV5 are labelled a-SV5 (in duplicate), while the antisera to SMAAs containing the individual SV5 proteins are labelled a-HN, A-F etc. Sera from mice immunized with the HN, F and M proteins precipitated only these respective proteins. However, mice immunized with the NP protein also precipitated the P protein, the mice immunized with the P protein precipitated small amounts of the NP protein. (This is consistent with observation and knowledge of the NP and P protein which may be found complexed together).

A neutralization test of virus infectivity was carried out as follows. Sera were diluted initially 1 in 20 and thereafter by doubling dilution. The dilutions of sera (100μl), diluted in tissue culture medium containing 2% calf serum, were incubated at 37ºC for 2h with 100μl of SV5 (5 × 10⁵ pfu/ml). The antibody-virus mixtures were then used to infect Vero cells growing as monolayers in 96-well microtitre plates. After a 2 to 4h adsorption period the antibody-virus inoculum was replaced with tissue culture medium containing 2% calf serum and the cells were reincubated at 37°C for a further 24 to 30h. The cells were then fixed with 5% formol saline, 2% sucrose for 10 min. permeabilized with 0.5% NP40, 10% sucrose for 5 min and washed three times with PBS. The presence of virus protein was detected by incubating the cells with a mixture of MAbs specific for SV5 (as 1/500 dilutions of ascitic fluids), and bound antibody was detected with ¹²⁵I-labelled Protein A (Amersham) as described by Randall et al., loc. cit. (1987).

All the sera were tested for their ability to neutralize virus infectivity. The anti-HN sera had high titres of neutralizing antibody (1/800 - 1/1600) while the anti-F sera had lower titres of neutralizing antibody (1/40 - 1/100). The latter result may have been in part due to the smaller amount of F antigen bound to the St.A-antibody complexes used (0.1 - 0.2 μg of F compared to 2-5 μg of HN, NP, M and P; Fig. 2). No neutralizing activity was measured in any of the anti-NP, P or M sera even at 1/20 dilutions of sera. (d) Detection of virus in the lungs of mice immunized with SMAA complexes

Two independent experiments were performed. Mice were immunized as described in part (c) above and the mice challenged In the first experiment (Fig. 3a) were the same mice whose sera were used in the immune precipitation test of part (c). The immune precipitation results of sera taken from the mice challenged in the second experiment (Fig. 3b)( were similar to those of part (c). Mice, while anaesthetized with ether, were challenged by inhalation of approximately 1 × 10⁷ pfu of the LN strain of SV5 in 100 μl of growth medium. Five days post infection the mice were sacrificed and their lungs removed into PBS (20% w/v). The lungs were then homogenised, using an MSE overhead homogeniser, sonicated, added to disruption buffer [see part (b) above] and heated for 5 min at 80°C. Particulate material was removed by sedimentation (6000 g for 3 min) and the dissociated polypeptides separated by electrophoresis through a 15% SDS polyacryl amide mini slab gel. The separated polypeptides were transferred to nitrocellulose using a semi-dry multi gel electroblotter, the nitrocellulose was then reacted with a pool of MAbs to the P protein and bound antibody was detected by ¹²⁵I labelled protein A and autoradiography in a manner previously described. (The amount of the MAb to the P protein bound to the Western blot is proportional to the total amount of virus present within the injected lungs).

Fig. 3 shows autoradiograph of those Western blots for experiments 3(a) and 3(6). Those for 3(b) are shown with a short (1 day) and long (4 day) radlographic exposure.

These experiments show that immunization with both external virus glycoproteins and internal structural proteins reduced the amount of virus present within the lungs compared to unimmunized mice or mice immunized with St.A-antibody complexes alone. There was no correlation between the degree of protection observed and the level of neutralizing activity in the sera. Immunization with the NP and M proteins appeared to be particularly effective in inducing a protective immune response but neutralizing antibody was not detected in these animals even at the time of sacrifice. While the mechanism of protection by immunization with the internal virus proteins is not known, we believe that it is mediated via a cytotoxic T cell response. This would be consistent with the progress of the disease in unimmunized animals in which there is an increase in the amount of virus present within the lungs until 4-6 days post infection followed by a rapid decrease. In these mice, cytotoxic T cell activity can be detected at 5 days post infection, whereas neutralizing antibodies cannot.

Further, we have shown that immunization with SMAA complexes induces a strong cytotoxic T cell response. In the interests of brevity, the data are not given here, but appear in R.E. Randall and D.F. Young, J. General Virology 69, 2505-2516 (1988), with reference to Fig. 7 thereof. EXAMPLE 2 Preparation and testing of St.A-(anti-SV5 HN anti-SV5 NP/anti-SV5 P/anti-SV5 M mixed antibodies)-SV5 HN/NP/P/M mixed antigen SMAA complexes

This Example illustrates a multivalent vaccine based on SMAA complexes in which the St.A support has several different antigens of the same virus attached to their respective antibodies.

(a) Preparation of the SMAA complexes

The St.A-antibody-antigen complexes were prepared as described in Example 1, part (a) above except that four ascitic fluids containing MAbs to the HN, NP, P and M were mixed together before being reacted with the fixed and killed suspension of St.A. The use of the mixture of MAbs implies that two or more different MAbs will become bound to the same particle of St.A support. (b) Immunization of mice with the SMAA complexes and testing of the resultant mouse sera

Mice were immunized as described in Example 1, part (c) above either with purified virus or with the above-mentioned mixture of St.A-antibody-antigen complexes, containing the HN, NP, M and P proteins.

Immune precipitates were formed as described in Example 1, part (c) above by the reaction of labelled soluble antigen extracts of SV5-infected BHK cells with polyclonal sera. The [³⁵S] methionine-labelled polypeptides were separated from the immune precipitates by electrophoresis through a 12% SDS polyacrylamide slab gel.

Referring to Figure 4 which shows an autoradiograph of this gel, slot 1 refers to the mice immunized with the purified virus, slots 2 to 4 to the mice immunized with the multivalent SMAA complexes. The SMAA complex-immunized mice produced antibodies against all four proteins. (Note: the original autoradiograph showed a faint band due to the F-* polypeptide of the F protien 1n slot 1. This is not reproduced in Figure 4). Example 3

Preparation and testing of St.A-anti (viral proteins of five different viruses) - viral antigen SMAA complexes

This Example illustrates a multivalent vaccine based on SMAA complexes, in which the St.A support has five different antigens, from different viruses, attached to their respective antibodies, (a) Preparation of SMAA complexes

St.A bound antibody was made by incubating equal volumes of fixed and killed suspensions of the Cowan A strain of Staphylococcus aureus (10% w/v) for 2h. at 4°C with a mixture of protein A-binding MAbs (as ascitic fluids) with respective specificities for (1) the glycoprotein D of HSV CMAb LP14 kindly provided by Dr. A.C. Minson, Dept. of Pathology, Cambridge University UK; A.C. Minson et al., J. Gen. Virology 67, 1001 (1986)], (ii) the HA protein of influenza virus strain PR8 (MAb HC245 kindly provided by A. Douglas, NIMR, Mill Hill, London, UK); (iii) the HA protein of measles viruses LMAb B10; from W.C. Russell, Dept. of Biochemistry and Microbiology, University of St. Andrews, Scotland; W.C. Russell & K.K. Goswani 1n C. Mims et al., (ed.) "Viruses and Demyelating diseases", Academic Press, London, (1984)]; (iv) the HN proteins of SV5 [MAb SV5-HN-4a; R.E. Randall et al., loc . cit., (1987)]; and (v) the HN protein of para-influenza virus type 2 (PIV-2) [MAb PIV-2-HN-3a; R.E. Randall & D.F. Young, J. Gen. Virology 69, 2051 (1988)]; Unbound antibody was removed by suspension and sedimentation (2500g for 3 min) of the solid material three times in PBS.

Soluble antigen extracts were made from SV5, PIV-2, or HSV infected BHK cells and from measles virus-infected vero cells as described by R.E. Randall et al. , loc. cit., 1987. A detergent extract was also made from purified influenza virus strain PR8 (also kindly provided by A. Douglas, see above) by incubating the virus with extraction buffer (20 mM Tris-HCl pH 7.2, 5mM-EDTA, 0.5% Nonidet P40, 0.1% SDS and 0.65M NaCl) for 30 min at 4°C, the parti culate material being then removed by centrifugation at 100,000g for 30 min. These five extracts were mixed together and the mixture incubated with the St.A-bound antibody as described in Example 1 , part (a).

(b) Immunization of mice with SMAA complexes and testing of the resultant mouse sera Groups of three mice were immunized twice intraperitoneally with SMAA complexes prepared as in part (a) above. The protocol for immunization immunoprecipitation and analysis of polypeptides released from the immunoprecipitates was as described in Example 1, part (c), except as follows. The amounts of the viral proteins present in the SMAA complexes ranged from 1 to 5 μg of each individual viral protein per molecule of the antibody. A 12% polyacryl amide gel was used in the SDS-PAGE.

Fig. 5 demonstrates the ability of the sera to immune-precipitate virus proteins from soluble antigen extracts made from cells infected with SV5 (track 6), PIV-2 (track 7), measles virus (track 8) and HSV (track 9). Track 1 shows the ability of the sera to precipitate all the appropriate virus proteins from a soluble antigen extract made by mixing extracts of SV5, PIV-2, measles and HSV infected cells together before the Immunoprecipitation was carried out. Tracks 2-5 are the respective controls for tracks 6-9, being immunoprecipitates between the MAb and its specific antigen (from the same soluble extract). The position of the virus proteins is Indicated in the left-hand column. The results show that the injection of the SMAAs induced antisera which reacted solely with the protein present in the SMAAs .

The sera from the three mice immunised with the SMAA prepared from the mixture of all five antigens (sera 1-3) and from unimmunized mice (serum 4) were tested for their ability to neutralize HSV, measles virus, PIV-2 and SV5 by the method described in Example 1, part (c). The sera were also tested for their ability to react with purified influenza virus in a radioimmunoassay (RIA). The RIA was based on a method described by R.E. Randall etal., Journal of Virology 52, 872-883 (1984), with the modification that the nitrocellulose sheets were sandwiched between 84-well Terasaki plates (10μl of a diluted sera per well). The purified influenza virus was bound to the nitrocellulose (approx 100μg per 7 × 5 cm sheet) and the protein-binding sites on the nitrocellulose were blocked with bovine serum albumin (BSA). Controls were sheets of nitrocellulose that had been reacted with BSA alone. After reaction of sera with the nitrocellulose, bound antibody was detected with ¹²⁵I-label led Protein A and autoradiography.

The following Table summarises the neutralization and RIA titre results, from which it will be seen that the sera of mice which had received the 5-valent SMAA complex vaccine, reacted with the HSV measles, PIV-2, SV5 and influenza viruses whereas the unimmunized mice reacted poorly. TABLE Neutralization titre RIA

Serum HSV measles PIV-2 SV5 Influenza virus

1 >1/1280 >1/1280 1/2560 <1/1280 1/640

2 1/1280 1/2560 1 /2560 1/1280 1/640

3 1/1280 1/2560 1/2560 1/1280 1/640

4 <1/20 <1/20 <l/20 <1/20 <1/20

EXAMPLE 4

Preparation and testing of a Dynabead-antibody-antigen complex

This Example shows that the invention does not arise from some peculiar effect of St.A or protein A, since the support material can equally be of biologically inert beads exemplified here by "Dynabeads".

(a) Preparation of the SMAA complex

"Dynabeads" are biologically inert magnetic polystyrene beads primarily used in cell separation techniques (DYNAL, Wirral, Merseyside, UK). They have a diameter of 4.5μm, specific gravity 1.5, magnetic susceptibility 10^-2cgs units and surface area 3-5 m.²/g. Purified monoclonal antibody CSV5-HN-Ib; R.E. Randall et al., loc. cit. , (1987)] to the HN protein of SV5 was directly coupled to tosyl-activated Dynabeads by standard methods recommended by the manufacturer. The Dynabead-bound antibody product was then used to purify the HN protein from a soluble antigen extract of SV5 infected cells as described in Example 1.

(b) Immunization of mice with the SMAA complex and testing of the resultant mouse sera After a single immunization of the resultant Dynabeads SMAA complex containing 2μg of the HN protein per molecule of antibody, HN antibody could be detected in the sera of the immunized mice, e.g. the titre of SV5-neutralizing activity [see Example 1, part (c)] was between 1/50 and 1/200. EXAMPLE 5

Preparation and testing of a "Sepharose"-Protein

A-antibody-antigen complex

"Sepharose"-Protein A beads are large, 20-40μm in diameter. A suspension of these beads (10% w/v) were reacted for 4h at 4°C with the ascitic fluid of a Simian virus 5 anti-HN antibody, and the resultant bound antibodies washed and incubated with SV5 soluble antigen extract as described in Example 1, part (c).

Mice were immunized twice with these SMAA complexes, the sera immunoprecipitated with labelled SV5 extract and the polypeptides analysed, as described in Example 1, part (c), except that a 10% polyacrylamide gel was used in the SDS-PAGE. Only the HN protein was detected in the autoradiograph (not reproduced here). The level of neutralizing activity of the mouse sera, determined as 1n Example 1, part (c) was less (1/200 to 1/800) than in the sera of mice immunized with equivalent amounts of St.A-antibody to SV5 HN-SV5 HN complex (1/800 to 1/3200). EXAMPLE 6 Preparation of a St.A-human anti-tetanus toxin monoclonal antibody - tetanus toxin SMAA complex

This Example relates to the binding of a human monoclonal antibody (A46; Watts and Davidson, EMB0 Journal 7, 1937-1945. 1988) to St.A. The monoclonal antibody has specificity for tetanus toxin. Cells secreting the antibody were grown in culture medium containing 10% foetal calf serum. The monoclonal antibody was purified from this medium on a Protein A-"Sepharose" column before being complexed to St.A as described 1n Example 1.

The antibody was dissociated from the St.A as described in Example 1, part (b) and subjected to SDS-PAGE. Well separated heavy and light chain bands were obtained.

This product can be incubated with a tetanus toxin extract to produce an SMAA complex useful for anti-tetanus inoculation. EXAMPLE 7

Preparation and testing of a St.A-anti-DNP-DNP(BSA) SMAA complex

Much research is being undertaken to use chemically synthesised peptides as immunogens for vaccine development. We have been developing a "universal" immunization scheme for use with peptides and proteins. The basis of the method is to attach a hapten [e.g. dinitrophenol (DNP)] to specific amino acid residues in synthetic peptides or to proteins. Monoclonal antibodies can be complexed with these modified peptides and proteins to form SMAA complexes which are then used as immunogens. (a) Preparation of the SMAA complexes

To establish the above principle, the hapten DNP was coupled to bovine serum albumin (BSA) by reacting 2,4-dinitrofluoro- benzene with BSA at a 10:1 molar ratio using standard methods. Monoclonal antibodies against DNP were then bound to solid support (St.A) and the bound antibodies were reacted with BSA-DNP by the method described in Example, 1 part (a), using 250μl of 10% w/v St.A-bound antibody per 200μg of DNP-BSA. St.A-anti-DNP- DNP-(BSA) complexes were thus prepared. (b) Dissociation of the antigen from the complex

Proceeding as in Example 1, part (b), the antigen and antibody were dissociated from the SMAA complex and the dissociated polypeptides were subject to SDS-PAGE analysis and the bands stained with Coomassie Brilliant Blue. Referring to Figure 6, lane 1 shows BSA/DNP and antibody (IgH, IgL) bands. The DNP is of too small a molecular weight to alter the mobility of the BSA. Lane 2 shows for comparison St.A-bound anti-DNP reacted with BSA alone (BSA not coupled to DNP). Since the BSA does not react with the anti-DNP Mab, the washed complexes contain no BSA.

(c) Immunization of mice with the SMAA complex and testing of the resultant mouse sera

Proceeding as in Example 1, part (c), mice were immunized with the SMAA complexes containing BSA-DNP. Sera from immunized mice were diluted initially 1 in 20 and thereafter by doubling dilution in PBS and reacted with BSA coupled to the wells of a 96 well microtitre plate. Bound antibody was detected with 1²⁵I-labelled Protein A and autoradiography. As shown in the autoradiograph of Figure 7, slots 1-3 relating to the sera from SMAA-immunized mice, a strong reaction to BSA was obtained. The control unimmunized mice, slots 4-6, showed scarcely any reaction. EXAMPLE 8

Preparation and testing of a St.A-anti-DNP-(DNP-spacer arm)-peptide SMAA complex The principle of this Example is similar to that of Example 7 except that (1) the DNP hapten is attached to a peptide during the chemical synthesis of the peptide and (2) it is attached via a spacer arm to distance the anti-DNP antibody from the peptide. (a) Preparation of the peptide Lyslne (DNP)-OH was synthesised by the method of R.R. Porter and F. Sanger, Blochem. J. 42, 287-294 (1948). The N-α-fluorenyl-methoxycarbonyl derivative Fmoc-L-Lysine (DNP)-OH was synthesised by method "B" of R.E. Shute and D.H. Rich, Synthesis, April 1987, 346-348, such that this DNP modified amino acid could be incorporated at any point during the synthesis of a peptide via a prepared symmetrical anhydride by the method of A. Dryland and R.C. Sheppard, J. Chem. Soc. Perkin I 125-137 (1986). Initially, to characterise this method of antigen presentation for peptides the following peptide was constructed using standard peptide synthesis methods, A. Dryland and R.C. Sheppard Tetrahedron 44, 859-876 (1988):

N terminal KYALADASLKMADPNRFRGKDLPVDNEOPTTRAOKLFAM

This is a chimeric peptide consisting of a sequence present in the VPl protein of poliovirus type 3 (underlined) coupled to the sequence found at the N terminal end of the gD protein of herpes simplex virus, i.e. of gD which has lost its signal peptide. Different monoclonal antibodies which either recognise DNP, the gD sequence or the polio sequence all react with this peptide.

(b) Preparation of the SMAA complex

St.A-anti-DNP-DNP-spacer arm (lysine)-chimeric peptide SMAA complexes were prepared using the monoclonal antibody to DNP as the coupling antibody St.A was saturated with the monoclonal antibody to DNP as described In Example 1. The resulting St.A-antibody complexes were in turn saturated with the DNP-peptide by mixing 250μl of the 10% w/v suspension of the St.A-antibody complex with 200μg of the DNP-peptide for 2-4h at 4ºC. Unbound peptide was removed from the SMAA complexes by sedimentation and resuspension of the complexes 1n phosphate buffered saline in a manner that has been described in Example 1.

(c) Dissociation of the antigen (chimeric peptide) from the complex

Proceeding as in Example 1 part (b), the antigen and antibody were dissociated from the SMAA complex and the dissociated polypeptides were subjected to SDS-PAGE analysis and the bands stained with Coomassie Brilliant Blue. Referring to Figure 8, lane 1 shows that the SMAA complexes of the invention which contain a monoclonal antibody to DNP give a low molecular weight DNP-peptide band. This is absent from a St.A-bound control monoclonal antibody (lane 2). The purified DNP-peptide was also electrophoresed through the gel (lane 3) as a marker. (d) Immunization of mice with the SMAA complex and testing of the resultant mouse sera

This test is of particular interest because it compares the immunostimulatory effect of the peptide alone with an SMAA complex containing it. Mice were immunized by the method of Example 1, part (c) with the SMAA complex or with the chimeric peptide alone. Those mouse sera and, for comparison, uncomplexed monoclonal antibodies to the HSV gD protein and to the DNP were reacted with a soluble antigenic extract of HSV-infected cells. Figure 10 is an autoradiograph of an RIA of a Western blot of HSV-infected cell polypeptides, separated through a 12% SDS-polyacrylamide gel, and reacted with uncomplexed monoclonal antibodies LP14 (lane 1) and K3 (lane 2) or with serum from mice immunized with antibody (K3)-DNP chimeric peptide complexes (lane 3) or with chimeric peptide alone (lane 4). Monoclonal antibody LP14 recognises the gD protein of HSV, while K3 recognises DNP. As DNP is not present in HSV-infected cells, these results clearly demonstrate that while the peptide alone does not give rise to an immune response the antibody-peptide complexes does. Bound antibody was detected with ¹²⁵I Protein A and autoradiography.

It will be appreciated that the poliovirus protein element of the chimeric peptide can be regarded as an extension of the spacer arm in relation to the anti-HSV activity conferred by the gD protein of HSV. Another peptide residue could clearly be substituted for the poliovirus protein residue.

Claims

CLAIMS 1. A support material-anti body-antigen (SMAA) complex for use in vaccination, comprising:

(2) antibodies of at least one kind bound to the support material, and

(3) immunostimulating antigens of at least one kind bound to the antibodies.

2. A SMAA complex according to claim 1, wherein antigens are immunologically directly bound to antibodies.

3. A SMAA complex according to claim 2, wherein the antigens are saturation-bound to the antibody so that the antigen:antibody molecular ratio is about 1:1.

4. A SMAA complex according to claim 1, 2 or 3 wherein different antigens of the same infective agent are bound to their specific antibodies.

5. A SMAA complex according to any preceding claim wherein the immunostimulating antigens are bound to other, Intermediate antigens or haptens which are in turn bound to the antibodies.

6. A SMAA complex according to claim 5 wherein the immunostimulating and intermediate antigens comprise a fusion polypeptide.

7. A SMAA complex according to claim 5 wherein haptens are bound to the immunostimulating antigens by a spacer arm effective to prevent steric Inhibition by the anti-hapten antibody of the immunostimulating function of the antigen.

8. A SMAA complex according to claim 5 wherein at least two kinds of immunostimulating antigen are covalently bound to a single molecule of hapten.

9. A SMAA complex according to any preceding claim wherein the support material comprises biologically inert beads, killed cells of Staphylococcus aureus Cowan Strain A or protein A.

10. A sterile pharmacological composition comprising a SMAA complex according to any preceding claim and a physiologically acceptable carrier or diluent.

11. A support material-antibody-antigen (SMAA) complex comprising: (1) a support material capable of being taken up by antigen-presenting cells or attached to antigen-responding cells of the immune system,

(2) antibodies of at least one kind bound to the support material,

(3) intermediate haptens or antigens directly bounds to the antibodies, and

(4) immunostimulating antigens of at least one kind bound to the intermediate antigens or haptens.