Adjuvant Quil Research Articles

Oral vaccination with immune stimulating complexes

There is a need for non-living adjuvant vectors which will induce a full range of local and systemic immune responses to orally administered purified antigens. Here we describe our experience with lipophilic immune stimulating complexes (ISCOMS) containing the saponin adjuvant Quil A. When given orally, ISCOMS containing the model protein antigen ovalbumin (OVA) induce a wide range of systemic immune responses, including Th1 and Th2 CD4 dependent activity, class I MHC restricted cytotoxic T-cell responses and local production of secretory IgA antibodies. More recent results indicate that ISCOMS may act partly by enhancing the uptake of protein from the gut. In addition, intraperitoneal injection of ISCOMS recruits and activates many components of the innate immune system, including neutrophils, macrophages, and dendritic cells. In parallel, there is increased production of nitric oxide (NO), reactive oxygen intermediates (ROI), interleukins (IL) 1, 6, 12, and γ interferon ( γIFN). Of these factors, only IL12 is essential for the immunogenicity of ISCOMS in vivo, as mucosal and systemic responses to ISCOMS are reduced in IL12KO mice, but not in IL4KO, IL6KO, inducible NO synthase (iNOS) KO, or γIFN receptor KO mice. We propose that ISCOMS act by targetting antigen and adjuvant to macrophages and/or dendritic cells. This pathway may be amenable to exploitation for vaccine development, especially if combined with another vector with a different mucosal adjuvant profile, such as cholera toxin.

Vaccine-induced Th1-type responses are dominant over Th2-type responses in the short term whereas pre-existing Th2 responses are dominant in the longer term.

The effect of adjuvant on induction of human papillomavirus type 16 E7 protein-specific cytotoxic T lymphocytes (CTL) and immunoglobulin G (IgG)2a antibody was studied in C57BL/6 J mice immunized with various adjuvants and E7 protein. Quil-A adjuvant, but not complete Freund's adjuvant (CFA) or Algammulin, induced a T-helper 1 (Th1)-type response to E7, which was characterized by CTL activity against a tumour cell line transfected with E7 protein and by E7-specific IgG2a. All tested adjuvants elicited comparable levels of E7-specific IgG1. The longest duration and greatest magnitude of CTL response was seen following two immunizations with the highest dose of E7 and Quil-A. Simultaneous immunization with a Th1 and a T helper 2 (Th2)-promoting adjuvant gave a Th1-type response. However, E7 and Quil-A were unable to induce a Th1-type response (as measured by the inability to generate anti-E7 IgG2a antibody) in animals with a pre-existing Th2-type response to E7. These results suggest that saponin adjuvants may be suitable for immunotherapy in humans where a Th1-type response is sought, provided that there is no pre-existing Th2-type response to the antigen.

Immune stimulating complexes as mucosal vaccines.

There is a need for non-living adjuvant vectors that will allow a full range of local and systemic immune responses to orally administered purified antigens. Here we describe our experience with lipophilic immune-stimulating complexes (ISCOMs) containing the saponin adjuvant Quil A. When given orally, ISCOMs containing the model protein antigen ovalbumin (OVA) induce a wide range of systemic immune responses, including Th1 and Th2 CD4-dependent activity, serum IgG antibodies and class I MHC-restricted cytotoxic T cell responses. In addition, there is local production of secretory IgA antibodies in the intestine itself, as well as priming of CD4 and CD8 T cell responses in the draining lymphoid tissues. Preliminary results indicate that the mucosal adjuvant properties of ISCOMs may reflect their ability to deliver antigen combined with the pro-inflammatory properties of Quil A in a particulate form. Of the many inflammatory mediators induced, interleukin-12, derived from dendritic cells and/or macrophages, appears to be of central importance. These results indicate that ISCOMs may prove to be useful mucosal vaccine vectors with functions which are distinct from existing vectors of this type.

Immunization with P. carinii gpA is Immunogenic But Not Efficacious in a Mouse Model of P. carinii Pneumonia † 840

Recent studies have shown that P. carinii (Pc) pneumonia (PCP) can be prevented by immunization with whole organisms in an immunosuppressed mouse model of disease. This raises the interesting possibility of trying to use this strategy to prevent disease in “at-risk” humans. However, because Pc cannot be cultured, immunization with whole organisms is not feasible. Therefore, we have undertaken studies to more precisely define the protective immune response to Pc with the hope of identifying the antigens responsible for confirming protection after immunization with whole organisms. Passive immunization with either monoclonal antibodies or sensitized lymphocytes have suggested that Pc gpA may be one such antigen. To directly assess the role of gpA in eliciting a protective immune response, we immunized mice with gpA, either in the form of nitrocellulose antigen bearing particles, or administered with the adjuvant Quil-A. Control animals received either immunization with whole organisms or with BSA. A gpA-specific response induced by either form of isolated gpA was equivalent to or exceeded that measured in the animals immunized with whole Pc. Immunization with gpA in the form of nitrocellulose antigen bearing particles induced a Th2-like antibody response, while Quil-A produced a mixed Th1/Th2-like antibody response. Table

Biodegradable implants for the delivery of veterinary vaccines: design, manufacture and antibody responses in sheep

Biodegradable implants made from cholesterol and lecithin (C:L) were used to deliver a recombinant antigen (recombinant Dichelobacter nodosus pili) and adjuvant (Quil A) to sheep. Implants (5.5-×1.8-mm) were placed subcutaneously and compared to a conventional vaccination regime (2 injections, 4 weeks apart) for antibody responses and tissue compatibility. Release profiles of antigen and adjuvant were also studied in vitro and in vivo. The presence of Quil A in vaccine implants had a marked effect on the rate at which antigen was released with 29 and 44% being released in the first 24 h from implants containing pili alone and pili with Quil A, respectively. Sheep produced significant levels of antibody when immunized with implants, however the response was short-lived and of significantly lower intensity than the response stimulated by two injections of antigen with Quil A ( P<0.01). A second implant system was developed where implants coated with C:L to delay antigen release, were used in combination with uncoated implants to deliver a priming dose and boosting dose of antigen. Antibody titres stimulated by the double implant system were equivalent to those stimulated by a conventional regime of two injections (four weeks apart) for the first six weeks of the experiment. After this time, antibody levels in the groups which received implants dropped significantly. In vitro studies revealed that some of the implant coatings had caused a delay in the release of antigen (the rate of release peaked at 72 h), however this was not long enough to provide a significant boosting effect. In all cases, implants were well tolerated by sheep and caused less local reaction than injected vaccines.

Effects of carbohydrate modification of Quillaja saponaria Molina QH-B fraction on adjuvant activity, cholesterol-binding capacity and toxicity

The iscom is an efficient antigen-presenting system for various antigens inducing both MHC class I and class II restricted immune responses. Protective immunity has been evoked against a variety of infectious agents. The saponin adjuvant Quil A, which was originally used to form iscoms, is composed of a mixture of structurally similar triterpenoids from Quillaja saponaria Molina having different biological activities. A purified, toxic Quillaja triterpenoid fraction with strong adjuvant activity, designated QH-B, was used to study whether modification of the carbohydrate moiety with sodium periodate would alter the toxicity without harming adjuvant activity and cholesterol-binding capacity. Most sugars, and in particular Api, Gal and Xyl, were modified by periodate treatment with only minor changes of the molecular weights indicating no loss of sugar residues. The adjuvant activity of QH-B was reduced in a dose-related manner, and at a concentration of 25 mM sodium periodate a significant reduction in toxicity was observed. The differences in both toxicity and adjuvant activity of the periodate-treated QH-B could be derived from alternations in the structure of the sugars Gal and Xyl, while modification of Api may influence adjuvant activity but not toxicity in vivo. The cholesterol-binding capacity, a prerequisite for iscom formation, was not affected by periodate oxidation at the doses tested. However, the use of modified QH-B as described in the present study for iscom-matrix formation resulted in “saponin-lipid complexes” which, to a various degree or totally, deviated from the characteristic iscom morphology.

The potential of polymeric nanoparticles to increase the immunogenicity of the hapten clenbuterol

Micro‐ and nanoparticles prepared from the biodegradable and biocompatible polymers poly(lactide‐co‐glycolide) (PLGA) and polymethylmethacrylate (PMMA) have been successfully used as immunopotentiating antigen delivery systems. In our study, this approach was used to improve polyclonal antibody production to clenbuterol (CBL), a model hapten. PLGA and PMMA nanoparticles were loaded with either CBL alone or with a clenbuterol‐transferrin conjugate (CBL—Tfn) and administered subcutaneously to mice. PLGA nanoparticles were administered with or without the saponin adjuvant Quil A. The anti‐CBL titres present in experimental sera were determined by an enzyme immunoassay (ELISA). CBL‐Tfn‐loaded PLGA nanoparticles co‐administered with Quil A had obvious advantages immunologically over the currently used method of raising antibodies to CBL (the positive control). The combined adjuvanticity of Quil A and PLGA nanoparticles resulted in a positive response in all four of the mice tested and in higher antibody titres than were seen in the positive control group. Furthermore, the sustained release of immunogen from the nanoparticles permitted a reduction in immunizing frequency over the 15‐week study period.

Evaluation of protective efficacy of an Actinobacillus pleuropneumoniae serotype 1 lipopolysaccharide-protein conjugate in mice

Actinobacillus pleuropneumoniae is the causative agent of porcine pleuropneumonia. The major adhesin of A. pleuropneumoniae has previously been identified as a lipopolysaccharide (LPS), and more recently, we demonstrated that high molecular mass LPS were involved in A. pleuropneumoniae adherence to porcine respiratory tract cells. We postulated that immunization with a LPS-based vaccine may confer a protective immunity. The high molecular mass O-polysaccharides obtained after acid hydrolysis and chromatographic separation were conjugated to bovine serum albumin (BSA) as a protein carrier. Groups of mice were injected twice with the following antigen preparations: whole-cell preparation, outer membrane preparation, O-polysaccharide-BSA conjugate, hydrolyzed LPS and phenol/water extracted LPS. A combination of different adjuvants was also used during these immunization procedures to induce a stronger immunological response to the polysaccharide antigen. Two weeks after the second injection, the mice were challenged intranasally with either homologous A. pleuropneumoniae serotype 1 strain or a serotype 5 strain. The highest survival rate, up to 80%, compared to the control groups ( P < 0.05), was recorded when the mice were injected twice with 15 μg of carbohydrates of O-polysaccharide-BSA conjugate mixed with the saponin-derived adjuvant Quil A. Survival rates of between 60 and 70%, twice those observed in the control groups immunized with PBS, were recorded in mice injected with the O-polysaccharide-BSA conjugate mixed with other adjuvant preparations such as alhydrogel, peanut oil and Freund's incomplete adjuvant. However, the protection induced by the conjugate antigen preparation was serotype specific, because mice challenged with a serotype 5 strain were killed. Taken together, these results confirm the important role of A. pleuropneumoniae LPS in pathogenesis.

The effects of continuous and intermittent delivery of antigens of Boophilus microplus on the development of murine antibodies

Protective antigens solubilised from the membranes of the midgut (LI-GM) of the adult cattle tick Boophilus microplus were delivered to mice either continuously for up to 4 weeks from osmotic pumps implanted subcutaneously or intermittently in a pulsatile fashion by injections or by a combination of both modes of delivery. The effects of delivery profile on antigen specific antibody levels and avidity were compared. LI-GM delivered either by three injections (weeks 0, 2 and 4), or continuously from osmotic pumps (over 4 weeks) induced similar levels of antibodies in mice. The mode of delivery of LI-GM when in the presence of the adjuvant Quil A did not generally affect either the level or the avidity of the antibody response; indeed a single injection of LI-GM in the presence of Quil A stimulated an immune response similar to that induced by several combinations of pulsatile and of continuous delivery where mice were exposed to antigen and adjuvant for up to 4 weeks. LI-GM incubated at 37°C in vitro and in vivo for periods from 4 h to 14 days was partly degraded into low molecular weight (less than 29 kDa) components. The immunogenicity of LI-GM incubated in vitro for 4 h was significantly decreased, although its antigenicity was not affected after incubation for up to 14 days. In conclusion, delivery of LI-GM continuously from osmotic pumps demonstrated that single-step immunisation of mice with tick antigens was feasible. However, it was also demonstrated that the continuous delivery of antigen was only advantageous (i.e potential for a decrease in the number of times an animal must be handled) compared with delivery by injections when no adjuvant was used. Further work is necessary to establish the effect antigen degradation has in limiting the immune response resulting from continuous delivery of antigen from osmotic pumps.

Field trials of ephemeral fever vaccines

Various bovine ephemeral fever (BEF) vaccines were tested between 1982 and 1984 in 24 Queensland herds; neutralising antibody responses were monitored and six of the herds were closely observed following natural challenge with ephemeral fever. A vaccine regime of two consecutive vaccinations with attenuated virus combined with the adjuvant Quil A provided excellent protection against BEF for at least 12 months, whereas one vaccination with the Quil A vaccine or two vaccinations with vaccine containing the adjuvant aluminium hydroxide gel did not provide significant protection. Antibody responses were highest for the vaccine incorporating Quil A when it was given as two consecutive injections. There was an apparent relationship between neutralising antibody response and the level of protection.

Anti-polysaccharide immunoglobulin isotype levels and opsonic activity of antisera: relationships with protection against Streptococcus pneumoniae infection in mice.

Relationships between in vitro parameters (opsonic activity and anti-pneumococcal polysaccharide [PS] antibody subclasses) and in vivo mouse protection were established by logistic regression analysis. Data were from 158 mice challenged with pneumococci after vaccination with synthetic oligosaccharide- and PS-protein conjugates in combination with the adjuvant Quil A. The hypothesis that serum opsonic activity has predictive value for protection against pneumococcal infection was tested. Serum opsonic activity was well correlated with protection (chi 2 = 35.5, P < 0.001), although a stronger correlation was observed for anti-PS IgM and IgG. The combined use of IgG and opsonic activity as predictor variables yielded the best fitting model for predicting protection (chi 2 = 74.1, P < 0.001). When opsonic activity data were added to models that included various antibody isotypes, the statistical significance of the models was enhanced. Thus, the opsonic activity of antisera induced by pneumococcal vaccines can predict mouse protection.

Age-dependent immune response in Merino sheep

The numerical and functional attributes of populations of lymphocytes were compared in the blood, lymph and skin of young and mature sheep. Young sheep, four to eight months old, had a lower proportion of CD4 + cells in blood, lymph and skin than mature sheep three to six years old. In contrast, B cells and T19 + cells were as prevalent or more prevalent in young sheep as in mature sheep. Blood lymphocytes from young sheep, cultured in vitro produced less interferon-γ, both spontaneously and in the presence of concanavalin A than did lymphocytes from older sheep. The serum antibody responses of adult sheep to the T cell-independent antigen Brucella abortus lipopolysaccharide (LPs) were greater over a range of antigen doses, suggesting that an apparent excess of antigen could not overcome the relative immune deficiency of young sheep. The adjuvant Quil A corrected the depressed antibody response of young sheep to B abortus LPs, but dextran sulphate did not. The skin contact hypersensitivity of mature sheep to dinitrochlorobenzene was greater. However, the T cell phenotypes present in infiltrates of lymphocytes elicited by the intradermal injection of tetanus and diphtheria, but not tuberculin antigens, were comparable for the two age groups. The capacity of Quil A to raise the antibody responses of both young and mature sheep to a similar titre suggests that it may be possible to overcome the immunological hyporesponsiveness that may contribute to the disease susceptibility of young sheep.

Removal of the cleavage site of recombinant feline immunodeficiency virus envelope protein facilitates incorporation of the surface glycoprotein in immune-stimulating complexes.

Recombinant vaccinia viruses were constructed that expressed the complete env gene of feline immunodeficiency virus with or without the nucleotide sequence encoding the cleavage site between the surface (SU) protein and the transmembrane (TM) protein. The removal of this cleavage site resulted in the expression of a 150K protein that was processed into a 130K protein and was not cleaved into the SU and the TM proteins. Removal of the cleavage site also facilitated incorporation of the SU protein in immune-stimulating complexes (iscoms). Antibody responses to both an SU and a TM peptide representing two immunodominant B cell epitopes were measured. These were higher in cats immunized with iscoms prepared from the cleavage site-deleted envelope protein than in cats immunized with iscoms prepared from the native envelope protein or immunized with the envelope protein and the adjuvant Quil A.

Adjuvant Quil A improves protection in mice and enhances opsonic capacity of antisera induced by pneumococcal polysaccharide conjugate vaccines

The adjuvant effect of Quil A on the primary antibody response of mice to pneumococcal capsular polysaccharide conjugates was examined. Quil A increased the anti-capsular polysaccharide antibody titres, the protection against Streptococcus pneumoniae, and the opsonic capacity of the antibodies as measured in a newly developed in vitro phagocytosis assay, using the mouse macrophage cell line J774.

Effective induction of neutralizing antibodies with the amino terminus of VP2 of canine parvovirus as a synthetic peptide

Fourteen synthetic peptides corresponding to previously mapped antigenic sites in VP2 of canine parvovirus (CPV) were used for immunization of rabbits to identify antiviral properties favourable for inclusion into a vaccine. Most antipeptide antisera obtained were reactive with viral protein, and with one of them it was possible to locate the hypothetical amino terminus of VP3 within positions 15–31 of VP2. Virus-neutralizing antibodies were only obtained with two overlapping 15-mer peptides corresponding in sequence to the amino terminus of VP2 (MSDGAVQPDGGQPAVRNERAT). Antibodies in the neutralizing sera bound most strongly to amino acids of the sequence DGGQPAV within the N-terminus of VP2, indicating that efforts to develop a synthetic vaccine against CVP should be focused on this stretch of amino acids. The two peptides induced long-lasting immunity (at least 8 months) using either Freund's adjuvant or aluminium hydroxide plus Quil A. Thus, this approach delineated the exact peptide sequence useful for vaccines applied to the amino-terminal region of VP2. These findings in experimental animals form a solid basis for exploration of a synthetic peptide vaccine against parvovirus infection in dogs, minks or cats.

Detection by a solid phase independent enzyme immunoassay of monoclonal anti-idiotypic antibodies against monoclonal antibodies neutralizing Semliki Forest virus and encephalomyocarditis virus

Neutralization inhibition enzyme immunoassay (NI-EIA) was evaluated for its usefulness to detect monoclonal anti-idiotypic antibodies (ab2mAbs) against idiotypic monoclonal antibodies (ab1mAbs) neutralizing either Semliki Forest virus (SFV) or encephalomyocarditis virus (EMCV). Purified ab1mAbs were coupled to keyhole limpet hemocyanin (KLH) mixed with the adjuvant Quil A and then injected intracutaneously into homologous BALB/c mice. Successful fusions were performed 5, 6 and 7 days after intracutaneous booster immunizations of these mice. Ab2mAbs in hybridoma supernatant fluids were detected by their capacity to block virus neutralization by ab1mAbs in NI-EIA. Two stable ab2mAb producing hybridomas were obtained against SFV neutralizing mAb UM 1.13, and twelve against EMCV neutralizing mAb UM 21.1.

Comparison of conventional adjuvants and ‘adjuvant-free’ monoclonal antibody targeting for stimulating antibody responses against a conjugate of luteinizing hormone releasing hormone and avidin

A conjugate of luteinizing hormone releasing hormone (LHRH) and avidin was used for immunizing mice and sheep. In mice, a range of adjuvants co-administered with the conjugate was shown to affect antibody titre and isotype. The commercially available Montanide and Ribi adjuvant emulsions, polyadenylic-polyuridylic acid, Quil A and Freund's complete adjuvant all produced high antibody titres to LHRH (reciprocal titre range 8000–20000) and avidin (range 9000–35000). Alhydrogel, Pluronic gel, Immunostim, DEAE-Dextran and muramyl dipeptide were weak to moderately effective (reciprocal titre ranges of 1300–5000 for LHRH and 3000–6000 for avidin, respectively). With Quil A, the response to avidin in mice was found to consist of roughly equal proportions of the IgG1, IgG2a and IgG2b isotypes. In contrast, the response to avidin using the other adjuvants and to LHRH consisted predominantly of the IgG1 isotype. An ‘adjuvant-free’ immunization strategy was attempted by targeting the LHRH-avidin conjugate to class II major histocompatibility complex (MHC) determinants on cells of the immune system. Using an immunoconjugate formed by linking LHRH-avidin to an antibody specific for mouse and sheep class II molecules, a twofold augmentation of titres relative to controls was obtained in mice. These antibodies were almost exclusively of the IgG1 isotype. In contrast, monoclonal antibody targeting in sheep resulted in a significantly enhanced immune response (reciprocal titres of 15 000 and 20 000 for LHRH and avidin, respectively) which was comparable to that achieved by immunizing with Quil A as adjuvant. Both the monoclonal antibody targeting and Quil A treatment tended to favour the production of antibody isotype IgG1 over IgG2. Our results suggest that monoclonal antibody targeting has significant potential as an alternative to the use of conventional adjuvants in animals and also in human subjects.

Meningococcal lipopolysaccharide (LPS)-derived oligosaccharide-protein conjugates evoke outer membrane protein- but not LPS-specific bactericidal antibodies in mice: influence of adjuvants.

Meningococcal lipopolysaccharide (LPS)-derived oligosaccharides (OS) were coupled to tetanus toxoid (TT) and purified P1.7,16 outer membrane proteins (OMP). The immunogenicities of the conjugates with and without the addition of the adjuvant Quil A or the nonionic block polymer L121 were studied in mice. Immunotype L2 and L3,7,9 OS-TT conjugates induced immunoglobulin G (IgG) responses that were strongly augmented by Quil A and L121. These adjuvants not only enhanced the amount of IgG evoked but also shifted the IgG subclass distribution from mainly IgG1 toward the complement-activating subclasses IgG2a and IgG2b. The antibodies induced were directed against the OS part of meningococcal LPS. They were not bactericidal for group B meningococci. Both the L3,7,9 OS-P1.7,16 OMP conjugate and purified P1.7,16 OMP evoked a strong IgG response against the P1.7,16 OMP but not against the L3,7,9 LPS. These anti-OMP IgG responses were comparable to the IgG OMP-specific responses induced by the H44/76 or HIII-5 outer membrane vesicles but still did not lyse group B meningococcal strains. The IgG response evoked with OS-OMP or purified OMP consisted mainly of the IgG1 subclass, whereas the H44/76 or HIII-5 outer membrane vesicles induced high amounts of bactericidal IgG2a and IgG2b antibodies next to the IgG1 antibodies. The addition of the adjuvant Quil A or L121 to OS-OMP or OMP resulted in the induction of high levels of bactericidal anti-P1.7,16-specific OMP antibodies, as reflected by the presence of substantial amounts of IgG2a and IgG2b antibodies. These results indicate that (i) mouse anti-LPS antibodies evoked by LPS-derived OS-protein conjugates are not bactericidal for group B meningococci, (ii) extensive purification of P1.7,16 OMP can lead to the loss of the intrinsic adjuvant properties of outer membrane vesicle preparations, and (iii) the addition of suitable adjuvants restores the ability of these purified P1.7,16 OMP to induce bactericidal antibodies.

Quantitative serological aspects of the enhanced enteric uptake of human gamma globulin by Quil-A saponin in Oreochromis mossambicus

Orally and anally delivered human gamma globulin (HGG) is rapidly absorbed into the plasma of tilapia, Oreochromis mossambicus, paralleling its intestinal appearance. The levels of HGG absorbed were quantified using enzyme-linked immunosorbent assay (ELISA) and temporally complex patterns were observed. Maximal levels of HGG in the plasma were found 6 h after oral delivery and 1 h after anal delivery. Delivery of HGG with the saponin adjuvant Quil-A resulted in a significant increase in the absorption of HGG to the plasma, when administered before or with the antigen or as a defined tertiary complex. Administration of Quil-A also had a variable effect on the absorptive kinetics of the antigen to the plasma. HGG was found degraded into a complexity of fragments in the plasma following detection by Western blotting. These fragments also had respective temporal maxima which were affected by the administration of the antigen with Quil-A. HGG was voided into the tank water in signficant amounts. These may potentially, be immunologically significant if re-absorbed into the fish.

The influence of the adjuvant quil a on the epitope specificity of meningococcal lipopolysaccharide anti-carbohydrate antibodies

Rabbits were immunized with immunotype L3,7,9 phosphoethanolamine (PEA) group containing oligosaccharide-tetanus toxoid conjugates both with and without the addition of the adjuvant Quil A. The epitope specificity of the antibodies present in these antisera was analysed in an immunotype L2 and L3,7,9 specific inhibition ELISA using the homologous and heterologous lipopolysaccharide, oligosaccharide and partial dephosphorylated oligosaccharide as inhibitors. Two groups of antisera could be identified. In one group of antisera, at least two antibody populations are present, namely directed against the PEA group containing determinants on immunotype L3,7,9 lipopolysaccharide and against immunotype L2 specific epitopes in which no PEA group is present. In the second group of antisera, one but probably more antibody populations are detected with a similar specificity towards the conserved epitopes of both immunotypes. In general, immunization with the conjugates only resulted in the induction of antibodies against the PEA group containing epitopes on the L3,7,9 lipopolysaccharide (80%). Antibodies directed against the conserved epitopes of both immunotypes are mainly evoked with the conjugates in combination with the adjuvant Quil A (80%). Although these results suggest that the epitope specificity of the antibodies induced depends on the use of Quil A, the influence of genetic factors cannot be excluded. At the moment it is not known whether the differences in epitope specificities are reflected in biological function of these antibodies. However, the induction of antibodies with clearly different epitope specificities after immunization of different rabbits with the same antigen stresses the importance of this kind of analysis when developing a vaccine based on oligosaccharide-protein conjugates.