AU2596699A - Vaccines comprising interleukin-12 and respiratory syncytial viral antigens - Google Patents

Description

WO 99/40937 PCT/US99/02848 VACCINES COMPRISING INTERLEUKIN-12 AND RESPIRATORY SYNCYTIAL VIRAL ANTIGENS BACKGROUND OF THE INVENTION The immune system uses many mechanisms for attacking pathogens; however, not all of these mechanisms are necessarily activated after immunization. Protective immunity induced by vaccination is dependent on the capacity of the vaccine to elicit the appropriate immune response to resist or eliminate the pathogen. Depending on the pathogen, this may require a cell-mediated and/or humoral immune response. The current paradigm for the role of helper T cells in the immune response is that T cells can be separated into subsets on the basis of the cytokines they produce, and that the distinct cytokine profile observed in these cells determines their function. This T cell model includes two major subsets: Thl cells that produce IL-2 and interferon-y (IFN-y) which augment both cellular and humoral immune responses, and Th2 cells that produce IL-4, IL-5 and IL-10 which augment humoral immune responses (Mosmann et al., J. Immunol. 126:2348 (1986)). It is often desirable to enhance the immunogenic potency of an antigen in order to obtain a stronger immune response in the organism being immunized and to strengthen host resistance to the antigen bearing agent. A substance that enhances the immunogenicity of an antigen with which it is administered is known as an adjuvant. For example, certain lymphokines have been shown to have adjuvant activity, thereby enhancing the WO 99/40937 PCT/US99/02848 -2 immune response to an antigen (Nencioni etal., J. Immunol. 139:800-804 (1987); EP285441 to Howard etal.). SUMMARY OF THE INVENTION 5 This invention pertains to vaccine compositions comprising a mixture of one or more respiratory syncytial virus (RSV) antigens, interleukin IL-12 and a mineral in suspension. The IL-12 may be either adsorbed onto the mineral suspension or simply mixed therewith. 0 In a particular embodiment of the invention, the IL-12 is adsorbed onto a mineral suspension such as alum (e.g., aluminum hydroxide or aluminum phosphate). These vaccine compositions modulate the protective immune response to the antigen; that is, the vaccine 5 composition is capable of quantitatively and qualitatively improving the vaccinated host's antibody response, and quantitatively increasing cell-mediated immunity for a protective response to a pathogen. In a particular embodiment of the invention, the RSV antigen 0 is an RSV F and/or G protein antigen. The invention also pertains to methods for preparing a vaccine composition comprising mixing an RSV antigen and IL-12 with a mineral in suspension. In particular, the IL-12 is adsorbed onto the mineral suspension. The 5 invention also pertains to methods for eliciting or increasing a vaccinate's humoral and/or cell-mediated immunity for a protective immune response, comprising administering to a vertebrate host an effective amount of a vaccine composition comprising a mixture of an RSV 0 antigen, IL-12 and a mineral in suspension in a WO 99/40937 PCT/US99/02848 -3 physiologically acceptable solution. In particular, the IL-12 is adsorbed onto the mineral suspension. BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a graph showing the proliferative 5 responses of splenic immunocytes from BALB/c mice vaccinated with F/AlOH plus ascending doses of IL-12. The dark shaded bars illustrate proliferation after in vitro stimulation with native F protein. Figures 2A and 2B are graphs showing the effect of D IL-12 on the ability of F/AlOH to generate IFN-y (Figure 2A) and IL-5 (Figure 2B) secreting spleen cells in BALB/c mice. Figure 3 is a graph showing the effects of IL-12 on the capacity of F/AlOH to induce or expand cell-mediated 5 immune responses, and the impact of IL-12 on the capacity of F/AlOH to elicit antigen-dependent killer cells after primary immunization. The solid and dashed lines denote the killer cell activities observed after incubation of bronchoalveolar lavage effector cells with ) syngeneic RSV-infected or control target cells, respectively, 5 days after challenge. Figure 4 is a graph showing the effect of IL-12 on the capacity of F/AlOH to boost the cell-mediated immune responses of seropositive BALB/c mice previously infected with RSV. The solid and dashed lines denote the killer cell activities observed after incubation of bronchoalveolar lavage effector cells with syngeneic RSV-infected or control target cells, respectively, 5 days after challenge.

WO 99/40937 PCT/US99/02848 -4 Figure 5 is a graph illustrating the effect of recombinant IL-12 on the protective immune responses induced in BALB/c mice vaccinated with F/AlOH. The bars are one standard deviation of the geometric mean. An asterisk denotes that infectious virus -was below detectable levels. DETAILED DESCRIPTION OF THE INVENTION IL-12 is produced by a variety of antigen-presenting cells, principally macrophages and monocytes. It is a critical element in the induction of Thl cells from naive T cells. Production of IL-12 or the ability to respond to it has been shown to be critical in the development of protective Thl-like responses, for example, during parasitic infections, most notably Leishmaniasis (Scott et al., U.S. Patent No. 5,571,515). The effects of IL-12 are mediated by IFN-y produced by NK cells and T helper cells. IFN-y is critical for the induction of IgG2a antibodies to T-dependent protein antigens (Finkelman and Holmes, Annu. Rev. Immunol. 8:303-33 (1990) and IgG3 responses to T-independent antigens (Snapper et al., J. Exp. Med. 175:1367-1371 (1992). Interleukin-12 (IL-12), originally called natural killer cell stimulatory factor, is a heterodimeric cytokine (Kobayashi et al., J. Exp. Med. 170:827 (1989). The expression and isolation of IL-12 protein in recombinant host cells is described in International Patent Application WO 90/05147. The studies described herein relate to the utility of IL-12 as an adjuvant in a respiratory syncytial virus (RSV) vaccine. Accordingly, this invention pertains to WO 99/40937 PCTIUS99/02848 -5 vaccine compositions comprising a mixture of an RSV antigen, IL-12 and a mineral in suspension. In a particular embodiment of the invention, the IL-12 is adsorbed onto a mineral suspension such as alum (e.g., 5 aluminum hydroxide or aluminum phosphate). These vaccine compositions modulate the protective immune response to the antigen; that is, the vaccine composition is capable of eliciting the vaccinated host's cell-mediated immunity for a protective response 0 to the pathogenic antigen. In particular embodiments, the antigen is the RSV F protein and/or G protein. IL-12 can be obtained from several suitable sources. It can be produced by recombinant DNA methodology; for example, the gene encoding human IL-12 has been cloned 5 and expressed in host systems, permitting the production of large quantities of pure human IL-12. Also useful in the present invention are biologically active subunits or fragments of IL-12. Further, certain T lymphocyte lines produce high levels of IL-12, thus providing a 0 readily available source. Commercial sources of recombinant human and murine IL-12 include Genetics Institute, Inc. (Cambridge, MA). The antigen of this invention, e.g., an RSV antigen, can be used to elicit an immune response to the antigen in a vertebrate such 5 as a mammalian host. For example, the antigen can be an RSV F protein (Collins etal., Proc. Natl. Acad. Sci. USA 81:7683-7687 (1984) or G protein (Satake et al., Nuc. Acids Res. 13:7795-7812 (1985) antigen or a portion thereof which retains the ability to stimulate an immune 0 response. Examples of such immunogenic portions are polypeptides comprising amino acid positions 283-315, WO 99/40937 PCTIUS99/02848 -6 289-315 and 294-299 of the RSV F protein. These regions include an epitope of the RSV F protein which elicits both neutralizing and antifusion antibodies (Paradiso et al., U.S. Patent 5,639,853). Alternatively, an RSV F 5 protein in its native dimeric form (140 kD) may be used (Paradiso et al., U.S. Patent 5,223,254). The method of the present invention comprises administering to a vertebrate an immunologically effective dose of a vaccine composition comprising a ) mixture of an antigen, e.g., an RSV antigen such as the F and/or G protein, an adjuvant amount of IL-12 and a mineral in suspension. In particular, the IL-12 is adsorbed onto the mineral suspension. As used herein, an "adjuvant amount" of IL-12 is intended to mean a quantity of IL-12 which is sufficient to enhance or modify the immune response to the vaccine antigen, e.g., an RSV antigen such as the F and/or G protein. As used herein, an "immunologically effective" dose of the vaccine composition is a dose which is suitable to ) elicit an immune response. The particular dosage will depend upon the age, weight and medical condition of the vertebrate to be treated, as well as on the method of administration. Suitable doses will be readily determined by the skilled artisan. The vaccine composition can be optionally administered in a pharmaceutically or physiologically acceptable vehicle, such as physiological saline or ethanol polyols such as glycerol or propylene glycol. The vaccine composition may optionally comprise additional adjuvants such as vegetable oils or emulsions thereof, surface active substances, e.g., hexadecylamin, WO 99/40937 PCT/US99/02848 -7 octadecyl amino acid esters, octadecylamine, lysolecithin, dimethyl- dioctadecylammonium bromide, N,N-dicoctadecyl-N'-Nbis (2-hydroxyethyl-propane diamine), methoxyhexadecylglycerol, and pluronic 5 polyols; polyamines, e.g., pyran, dextransulfate, poly IC, carbopol; peptides, e.g., muramyl dipeptide, dimethylglycine, tuftsin; immune stimulating complexes; oil emulsions; lipopolysaccharides such as MPL® (3-0 deacylated monophosphoryl lipid A; RIBI ImmunoChem 0 Research, Inc., Hamilton, Montana); and mineral gels. The antigens of this invention can also be incorporated into liposomes, cochleates, biodegradable polymers such as poly-lactide, poly-glycolide and poly-lactide-co glycolides, or ISCOMS (immunostimulating complexes), and 5 supplementary active ingredients may also be employed. The antigens of the present invention can also be administered in combination with bacterial toxins and their attenuated derivatives. The antigens of the invention can also be administered in combination with 0 other lymphokines, including, but not limited to, IL-2, IL-3, IL-15, IFN-y and GM-CSF. The vaccines can be administered to a human or animal by a variety of routes, including, but not limited to, parenteral, intrarterial, intradermal, 5 transdermal (such as by the use of slow release polymers), intramuscular, intraperitoneal, intravenous, subcutaneous, oral and intranasal routes of administration. The amount of antigen employed in such vaccines will vary depending upon the identity of the 0 antigen. Adjustment and manipulation of established dosage ranges used with traditional carrier antigens for WO 99/40937 PCT/US99/02848 -8 adaptation to the present vaccine is well within the ability of those skilled in the art. The vaccines of the present invention are intended for use in the treatment of both immature and adult warm-blooded 5 animals, and in particular, humans. Typically, the IL-12 and the antigen will be co-administered; however, in some instances the skilled artisan will appreciate that the IL-12 can be administered close in time but prior to or after vaccination with the antigen. 0 The RSV antigen of the present invention can be coupled to another molecule in order to modulate or enhance the immune response. Suitable carrier proteins include bacterial toxins which are safe for administration to mammals and immunologically effective 5 as carriers. Examples include pertussis, diphtheria, and tetanus toxoids and non-toxic mutant proteins (cross-reacting materials (CRM)), such as the non-toxic variant of diphtheria toxoid, CRM 197 . Fragments of the native toxins or toxoids, which contain at least one D T-cell epitope, are also useful as carriers for antigens. Methods for preparing conjugates of antigens and carrier molecules are well known in the art and can be found, for example, in Dick and Burret, Contrib Microbial Immunol. 10:48-114 (Cruse JM, Lewis RE Jr, 5 eds; Based, Krager (1989) and U.S. Patent No. 5,360,897 (Anderson et al.). The adjuvant action of IL-12 has a number of important implications. The adjuvanticity of IL-12 can increase the concentration of protective antibodies produced against the antigen in the vaccinated organism. As a result, effective (i.e., protective) vaccination WO 99/40937 PCT/US99/02848 -9 can be achieved with a smaller quantity of antigen than would be normally required. This reduction in the required amount of antigen may lead to more widespread use of vaccines which are difficult and costly to prepare. Additionally, the use of IL-12 as an adjuvant can enhance the ability of antigens which are weakly antigenic or poorly immunogenic to elicit an immune response. It may also provide for safer vaccination when the antigen is toxic at the concentration normally required for effective immunization. By reducing the amount of antigen, the risk of toxic reaction is reduced. Furthermore, the adjuvant action can reduce the antigen load of a subject being immunized with a large number of vaccines in a short time. Typically, vaccination regimens call for the administration of antigen over a period of weeks or months in order to stimulate a "protective" immune response. A protective immune response is an immune response sufficient to protect the immunized organism from productive infection by a particular pathogen or pathogens to which the vaccine is directed. IL-12, when administered with an antigen, such as an RSV antigen including, but not limited to the F protein and the G protein, and mixed with or adsorbed onto a mineral alum in suspension, can accelerate the generation of a protective immune response. This may reduce the time course of effective vaccination regimens. In some instances, it may result in the generation of a protective response in a single dose. The goal of the work described herein was to determine the feasibility of using recombinant IL-12 as WO 99/40937 PCT/US99/02848 -10 an immune response modifier for vaccines against RSV. To that end, groups of BALB/c mice were immunized with the native fusion (F) protein of the A2 strain of RSV and ascending amounts of IL-12. The F protein and IL-12 5 were adsorbed to aluminum hydroxide (AlOH, Alu-gel-Sm, Serva Fine Biochemicals, Westbury, NY) adjuvant. Thereafter, the capacities of the vaccines to induce systemic cell-mediated and humoral immune responses were compared. The results presented herein demonstrate that 0 IL-12 is a powerful modifier of both systemic humoral and cell-mediated immune responses. Significant increases in complement-assisted and anti-F protein IgG2a antibody titers were observed after primary and secondary vaccination. In addition, 0.01 and 0.1 ig 5 IL-12 had profound effects on the ability of F/AlOH to elicit cell-mediated immune responses. Five days after challenge with the A2 strain of RSV, the lungs of seronegative and seropositive mice contained augmented antigen-dependent killer cell activities. o To illustrate the mode of action of IL-12 on the immune responses generated by the AlOH adjuvanted vaccines, the supernatants from bulk spleen cells cultured with native F protein were analyzed for the cytokines associated with helper T cell subsets. The 5 results presented herein suggest that IL-12 increases the ability of the vaccines components to elicit immune responses governed by type 1 helper T cells (Thl). The presence of IL-12 in the vaccine appears to be associated with increased amounts of IFN-y in the 0 culture supernatant. Moreover, the results imply that increased amounts of IL-12 in the formulation diminish WO 99/40937 PCT/US99/02848 -11 the capacity of the vaccine to generate type 2 helper T cells (Th2). That is, as the amount of IL-12 in the vaccine is increased, the culture supernatants contain lower quantities of IL-5. However at doses of IL-12 5 above 1.0 pg, both Thl and Th2 cytokines appear diminished. Work described herein also examined the immunomodulatory effects of IL-12 on the immunogenicity of RSV native attachment (G) protein and D formalin-inactivated RSV (FI-RSV). Both vaccines have been shown in rodents to generate immune responses associated with the induction of Th2 helper T cells. The FI-RSV vaccine was a facsimile of the original Lot-100 vaccine formulated by Pfizer (Fulginiti et al., 5 Am. J. Epidemiol. 89:435-448 (1969) and Chin, et al., Am. J. Epidemiol. 89:449-463 (1969)). Both vaccines were adsorbed to AlOH and formulated alone or in the presence of 10-fold ascending doses of IL-12. Several criteria were assessed to determine the D immunoregulatory properties of IL-12. The ability of IL-12 to alter the capacity of G/AlOH and FI-RSV to bias recipients for atypical pulmonary inflammatory responses after challenge was particularly significant. Mice twice immunized with either FI-RSV alone or G/AlOH 5 - alone, and challenged with the A2 strain of RSV, developed atypical pulmonary inflammatory responses; these inflammatory responses were characterized by statistically significant increased percentages of eosinophils and quantities of IL-5 in the D bronchoalveolar lavage (BAL) fluids 5 days after challenge. In contrast, pulmonary eosinophilia was not WO 99/40937 PCTIUS99/02848 -12 observed in control mice experimentally infected with the A2 strain of RSV. The addition of 10-fold ascending doses of IL-12 to FI-RSV significantly reduced both the relative 5 percentage of eosinophils and the amount of IL-5 detected in the BAL fluids after challenge. Concomitant with the diminution in IL-5 and eosinophils was the significant transformation in serum anti-F and anti-G protein-specific IgG1 to IgG2a antibody. That is, 0 vaccination with FI-RSV containing IL-12 resulted in significant decreases in IgG1 and increases in IgG2a protein-specific antibody titers. The data, therefore, imply that IL-12 limits the induction of F and G protein-specific Th2 helper T cells and their 5 concomitant IL-5 secretion, and, instead, generates IFN-y- secreting Th1 helper T cells. Thus, the results suggest that the presence of IL-12 in FI-RSV, and ultimately at the site of the immune response, directs the induction of distinct Thl helper cell subsets. The 0 data also support the idea that the Thl helper T cells, facilitated by IL-12 injection, hinder the ability of FI-RSV to elicit Th2 helper T cells. Although the addition of 10-fold ascending doses of IL-12 to G/AlOH resulted in significant reductions in 5 pulmonary IL-5 concentrations and alterations in anti-G and anti-F protein IgG subclass antibody ratios, it appears that IL-12 is unable to modify the capacity of G/AlOH to predispose mice for pulmonary eosinophilia after challenge. Thus, the data suggest that distinct 0 pathways exist for the control of systemic humoral immune responses and the mobilization and replication of WO 99/40937 PCTIUS99/02848 -13 eosinophils. Alternatively, it may be that cytokines other than IL-5 play a role in generating pulmonary eosinophilia. The addition of IL-12 to G/AlOH resulted in significant reductions of IL-5 in the BAL fluids, but had no significant effect on the relative percentage of eosinophils in the pulmonary tissues after challenge. However, IL-5 may be present in the lungs at levels that are non-detectable in the assays employed. It is also possible that IL-12's inability to limit the D predilection for eosinophils results from the relatively large amounts of G protein in G/AlOH when compared to FI-RSV, or that native G protein may have several epitopes with potential to bias for eosinophilia. Some of these epitopes may be destroyed by formalin 5 treatment. Thus, the discrepancy between FI-RSV and G/AlOH with respect to their capacity to bias for eosinophilia may be a quantitative phenomenon. The inability of IL-12 to transform the atypical pulmonary inflammatory responses predisposed by G/AlOH, on the ) other hand, may be related to the uniqueness of the heavily glycosylated protein. As noted above, the results presented herein demonstrate that IL-12 is a potent regulator of systemic humoral immune responses generated against both purified 5 native G and F proteins of RSV. Nonetheless, the presence of IL-12 does not appear to enhance the complement fixing neutralizing antibody titers attributed to the IgG2a subclass with all RSV vaccines. Augmented serum complement-fixing neutralizing antibody D titers were not observed 2 weeks after secondary immunization with either G/AlOH or FI-RSV. With respect WO 99/40937 PCT/US99/02848 -14 to the FI-RSV vaccine, the failure of IL-12 to positively influence neutralizing antibody titers may reflect the destruction by formalin treatment of the F and G protein epitopes responsible for the generation of IgG2a complement fixing antibodies. As the quantity of IL-12 is increased in the vaccine, neutralizing and IgG1 antibodies are decreased. In contrast, IL-12 reproducibly augments the IgG2a and complement enhanced neutralizing titers in response to F/AlOH (see Tables 1 D and 2, below). The explanation for the inability of IL-12 to influence the generation of IgG2a complement-fixing neutralizing antibodies after vaccination with native G and F protein appears more complex. Without wishing to be bound by theory, one explanation is that the majority of the protein-specific IgG2a antibodies which are elevated following administration of IL-12 are directed against non-neutralizing epitopes. Another possibility is that the contaminating F protein in G/AlOH elicited complement-fixing neutralizing antibodies that obscured the effects of IL-12 on the generation of G protein-specific IgG2a neutralizing antibodies. The data suggest that the optimal dose of IL-12 is between 0.01 and 1.0 pg; higher doses of IL-12 (10 ig) appear counterproductive. This conclusion is based on the capacity of the vaccine to generate protective humoral and cell-mediated immune responses. Taken together, the results support the use of recombinant IL-12 in vaccine formulations to regulate systemic humoral and cell-mediated immune responses. Furthermore, no untoward effects on the general health of the WO 99/40937 PCT/US99/02848 -15 recipients were observed after the administration of IL-12. In summation, the results presented herein show that IL-12 is useful as an immune response modifier for RSV 5 vaccines containing a mineral in suspension. Moreover, the transformation of the immune responses occurs at IL-12 doses appropriate for human use. Thus, it is clear that RSV F and G proteins, alone or in combination with other viral antigens, adjuvanted with an alum gel plus 0 IL-12, are particularly suitable for RSV vaccine preparation. The following Examples are offered for the purpose of illustrating the present invention and are not to be construed to limit the scope of this invention. The 5 teachings of all references cited herein are hereby incorporated herein by reference. EXAMPLES EXAMPLE 1: Effect of IL-12 on immunogenicity of RSV F protein adsorbed to aluminum hydroxide 0 adjuvant STUDY DESIGN The purpose of the study was to determine the effects of recombinant murine IL-12 on the immunogenicity of the fusion protein of RSV formulated with aluminum hydroxide adjuvant. Naive female BALB/c mice (8-10 weeks of age) were vaccinated intramuscularly at weeks 0 and 4 with purified native fusion (F) protein. The F protein was WO 99/40937 PCT/US99/02848 -16 adsorbed to Alu-Gel-Sm (aluminum hydroxide at 2%, Serva Fine Chemicals, Westbury, NY). The vaccines were prepared such that each mouse received 3.0 ig F protein/dose, 100 ig of aluminum 5 hydroxide (AlOH) per dose, and 0, 0.01-, 0.1, or 1.0 ig IL-12/dose. Control mice were injected with 100 ig Alu-Gel-Sm in PBS alone. Four weeks after primary and 2 weeks after secondary vaccination, serum was collected for the determination by ELISA of geometric mean endpoint 0 antibody titers. The microwells were coated with highly purified ion exchange purified F protein. In addition, neutralizing antibody titers were determined by the plaque reduction neutralization test in the presence and absence of complement against the A2 strain of virus. 5 To determine the effects of IL-12 on the capacity of the vaccines to generate distinct helper T cell subsets, bulk spleen cell suspensions were obtained 2 weeks after secondary immunization and cultured for 6 days in the presence of increasing concentrations of purified F 0 protein, purified UV-inactivated RSV A2, CRM.

7 , concanavalin A (Con A, a T lymphocyte mitogen), or medium alone. The supernatants from these cultures were pooled and tested by capture-ELISA for the presence of interferon-y (IFN-y) and interleukin-5 (IL-5). 5 RESULTS The results shown in Table 1 are the geometric mean endpoint antibody titers determined by ELISA. The neutralizing antibody titers are the geometric mean neutralizing antibody titers and were determined by the 0 plaque reduction neutralization test in the presence (+) WO 99/40937 PCT/US99/02848 -17 and absence (-) of 5% serum to supply complement. The antibody titers were determined 4 and 2 weeks after primary and secondary vaccination, respectively. IL-12 augmented systemic humoral immune responses generated by F/AlOH 4 weeks (Upper Panel, Table 1) and 2 weeks (Lower Panel, Table 1) after primary and secondary vaccination respectively. For example, 4 weeks after primary immunization with F/AlOH formulated with either 0.1 or 1.0 pg IL-12 per dose, the total IgG endpoint antibody titers were significantly different and enhanced 10 and 18 times, respectively, when compared to the endpoint titers in the sera of mice vaccinated with F/AlOH alone. In similar fashion, the total IgG antibody titers were elevated 3 and 8 times respectively, 2 weeks after secondary vaccination (Lower Panel, Table 1). The data further implied that the increase in IgG antibody titers was dependent on the dose of IL-12 in the vaccine. This was best exemplified after the determination of F protein-specific IgG2a titers. When contrasted with the serum anti-F protein IgG2a antibody titers that were elicited 4 weeks after primary immunization with F/AlOH alone, the vaccines formulated with F/AlOH plus either 0.01, 0.1 or 1.0 pg IL-12 were significantly elevated 4-, 48-, and 158-fold, respectively (Upper Panel, Table 1). Elevations in IgG2a antibody titers were also observed 2 weeks after secondary vaccination with F/AlOH plus IL-12 (Lower Panel, Table 1). Most importantly, the data suggested that the addition of IL-12 could increase the capacity of F/AlOH to generate complement assisted serum neutralizing antibody titers. The serum complement assisted WO 99/40937 PCT/US99/02848 -18 neutralizing antibody titers of mice injected with F/AlOH plus 1.0 ig IL-12 were heightened at least 7 times 4 weeks after primary immunization (Upper Panel, Table 1). Following secondary vaccination with F/AlOH formulated 5 with either 0.1 or 1.0 ig IL-12, the neutralizing antibody titers were increased 5 and 10 times, respectively (Lower Panel, Table 1). The results implied that immune cells from the spleens of mice injected with IL-12 had more potential to 0 replicate when presented with antigen than those of mice vaccinated with aluminum gel alone (Figure 1). For example, the stimulation index of the splenic immunocytes from mice twice immunized with F/AlOH plus 0.01 ig IL-12 was nearly twice that of mice immunized with F/AlOH alone 5 after in vitro culture with native F protein. The results further suggested that employing doses of IL-12 greater than 0.01 ig were counterproductive. The stimulation indices of mice vaccinated with F/AlOH plus either 0.1 or 1.0 pg IL-12 were 5-fold less. 0 To estimate the effect of various doses of IL-12 on the capacity of the aluminum adjuvants to generate distinct antigen dependent helper T cell subsets, bulk spleen cell suspensions were prepared and cultured in the presence or absence of native F protein. The data 5 suggested that when F/AlOH was formulated in the absence of IL-12, type 2-like helper T cell responses were generated. IFN-y was barely detectable in the supernatants of spleen cells cultured 6 days with native F protein (Figure 2A). Alternatively, the same 0 supernatant contained 6 ng IL-5 per ml culture supernatant (Figure 2B).

WO 99/40937 PCT/US99/02848 -19 The addition of IL-12 to F/AlOH appeared to augment the capacity of splenic immunocytes to secrete IFN-y (Figure 2A). After 6 days culture with native F protein, 1, 9, 16, and 16 units IFN-y were secreted per ml supernatant by the spleen cells from mice vaccinated with either F/AlOH alone or plus 0.01, 0.1, or 1.0 pg IL-12, respectively (Figure 2A). However, the data show that the presence of 0.1 or 1.0 pag of IL-12 in the vaccine was associated with decreased amounts of IL-5 in the culture supernatants. Stated otherwise, the same supernatants from mice vaccinated with F/AlOH plus 10-fold ascending doses of IL-12 also secreted 6, 13, 2, and 0.4 ng IL-5 per ml supernatant, respectively (Figure 2B).

WO 99/40937 PCT/US99/02848 -20 Table 1: The ability of recombinant murine IL-12 to modify the systemic humoral immune responses of BALB/c mice immunized with F/AlOH ANTIBODY TITERS F Protein (X1000)* Neutralizingt Vaccine IL-12 IgG IgGi IgG2a (+) (-) F/AlOH 1.0 Ig 4,621.2 272.9 1,057.4 132' <20 F/AlOH 0.1 pag 2,554.2 263.4 318.9 27 <20 F/AlOH 0.01 pg 586.8a 268.9 25.3a 22 <20 F/AlOH NONE 258.3a 120.3 6 .7d <20 <20 PBS/AlOH NONE <0.05 NT NT <20 <20 F/AlOH 1.0 ig 1 2

,

8 6 2 .2b 889.4 1,730.0 5 1 9 9 55 F/A1OH 0.1 jig 4,524.9 752.7 521.3 2 2 4 h 32 F/AlOH 0.01 pg 3,052.2 1,247.1 205.4* 96 22 F/AlOH NONE 1,521.5c 630.4 4 8

.

0 d 49 32 PBS/A1OH NONE <1.0 NT NT, <20 <20 BALB/c mice were vaccinated intramuscularly on weeks 0 and 4 with native F protein (3 ig/dose) adsorbed to aluminum hydroxide (AlOH, 100 ig/dose). IL-12 was added to the vaccines at the indicated doses. Control mice were injected with PBS plus AlOH. The upper and lower panels depict antibody titers 4 and 2 weeks after primary and secondary vaccination respectively. * The numbers are the geometric mean endpoint antibody titers determined by ELISA. t The numbers are the geometric mean neutralizing antibody titers and were determined by the plaque reduction neutralization test in the presence (+) or absence (-) of 5% complement. There were 5 mice per group. a P<0.05 vs. the IgG titers elicited after vaccination with F/AlOH plus 0.1 or 1.0 pig IL-12.

WO 99/40937 PCT/US99/02848 -21 b P<0.05 vs. the IgG titers elicited after vaccination with F/AlOH plus 0.0, 0.01, or 0.1 pg IL-12. c P<0.05 vs. the IgG titers elicited after vaccination with F/AlOH plus 0.1 pg IL-12. d P<0.05 vs. the IgG titers elicited after vaccination with F/AlOH plus 0.01, 0.1, or 1.0 pg IL-12. * P<0.05 vs. the IgG titers elicited after vaccination with F/AlOH plus 1.0 pg IL-12. P<0.05 vs. the complement dependent neutralizing antibody titers elicited after vaccination with F/AlOH plus 0.0, 0.01, or 0.1 pag IL-12. g P<0.05 vs. the complement dependent neutralizing antibody titers elicited after vaccination with F/AlOH plus 0.0 or 0.01 lag IL-12. h P<0.05 vs. the complement dependent neutralizing antibody titers elicited after vaccination with F/AlOH without IL-12.

WO 99/40937 PCT/US99/02848 -22 EXAMPLE 2: The effect of administration of IL-12 at a distal site on the ability of F/AlOH to induce systemic humoral immune responses. STUDY DESIGN 5 The purpose of the study was to determine the biological effects of recombinant murine IL-12 on systemic humoral immune responses when injected at a site distal to that of F/AlOH. To that end, naive female BALB/c mice (8-10 weeks of age) were primed 10 intramuscularly (IM) with ion exchange purified native fusion (F) protein from the A2 strain of RSV. The F protein (3 pg/dose) was adsorbed to aluminum hydroxide adjuvant (AlOH, 100 pg/dose, Alu-gel-S-). F/AlOH was administered in combination with 10-fold ascending doses 15 of recombinant murine IL-12 (1, 10, 100 ng IL-12/dose). In brief, 2 immunization protocols with F/AlOH plus IL-12 were employed. In the first scenario, groups of mice were injected IM with F/AlOH in one thigh and received 10-fold ascending doses of IL-12 in the contralateral 20 thigh. In the second instance, groups of mice were injected once with a vaccine composed of F/AlOH formulated with 1 of 3 10-fold ascending doses of IL-12. The vaccine was incubated overnight at 4 0 C to allow maximum time for adsorption of IL-12 to AlOH. Additional 25 control mice were vaccinated either with F/AlOH alone, F protein in PBS alone, or PBS plus AlOH alone. Four weeks after primary vaccination, sera were collected for the determination of geometric mean endpoint anti-F protein total and subclass IgG antibody titers by ELISA.

WO 99/40937 PCTIUS99/02848 -23 RESULTS The results depicted in Table 2 confirmed the capacity of IL-12 to augment the systemic humoral immune responses induced after vaccination with F/AlOH. When 5 compared to F/AlOH alone, the anti-F protein total IgG antibody titers elicited by F/AlOH plus either 10 or 100 ng IL-12 were significantly greater 4 weeks after primary immunization. More importantly, the presence of either 10 or 100 ng IL-12 in the vaccines was associated with 10 statistically enhanced protein specific IgG2a antibody titers (Table 2). However, the data also suggested that in a single dose protocol, IL-12 must be present at the local site of injection. When mice were primed with F/AlOH alone and IL-12 was injected at a distal site, 15 statistically lower anti-F protein antibody titers were obtained (Table 2). The protein specific IgG2a antibody titers of mice injected with F/AlOH plus 100 ng IL-12 were 10 times greater than those of cohort mice immunized with F/AlOH alone plus 100 ng IL-12 administered in the 20 contralateral thigh.

WO 99/40937 PCTIUS99/02848 -24 Table 2. The effect of distal administration of IL-12 on the ability of F/AlOH to induce systemic humoral immune responses 4 weeks after primary vaccination.t ANTIBODY TITERS (LOG 10) 5 Vaccine na IL-12 #In1. #Sites IqG IgG1 IqG2a F/A1OH 100 2 2 5.6±0.2' 5.

3 ±0.

2 b 4

.

7 ±0.

3 b F/A1OH 100 1 1 6.9±0.2a 6

.

0

±

0

.

1 b 6.2±0.2a F/A1OH 10 2 2 5 .8±0.2b 5 .5±0.2b 5 .0± 0 . 3 b F/A1OH 10 1 1 6.6±0.3a 5 .9±0.2b 6.0±0.3a 10 F/A1OH 1 2 2 6.0±0.2b 5 .5±0.2b 5 .2±0.4b F/A1OH 1 1 1 6.1±0.2b 5 .6±0.2b 5 .4±0.3b F/A1OH None 1 1 6.1±0.2 5.5±0.2 5.2±0.3 F/PBS None 1 1 3.5±0.5 3.3±0.4 <3.0 PBS None 1 1 <3.0 NT NT 15 t Naive female BALB/c mice were primed intramuscularly (IM) with native fusion (F) protein (3 ig/dose) adsorbed to aluminum hydroxide adjuvant (AlOH). F/AlOH was administered in combination with 10-fold ascending doses of recombinant murine IL-12 (100, 10, 1 ng IL-12/dose). 20 Two immunization strategies with F/AlOH plus IL-12 were employed: 2 inj., 2 sites denotes mice which were injected IM with F/AlOH in one thigh and received 10-fold ascending doses of IL-12 in the contralateral thigh; 1 inj., 1 site indicates that mice were injected once with 25 a vaccine composed of F/AlOH plus 1 of 3 10-fold ascending doses of IL-12. Additional control mice were vaccinated either with F/AlOH alone, F protein in PBS alone, or PBS plus AlOH alone. The numbers are geometric endpoint titers of 5 mice per group. 30 a P<0.05 vs. the serum anti-F protein antibody titers from mice injected with IL-12 at a distal site, and the titers of mice immunized with F/AlOH alone. b P>0.05 vs. the serum anti-F protein antibody titers from mice immunized with F/AlOH alone.

WO 99/40937 PCT/US99/02848 -25 EXAMPLE 3: The effect of IL-12 on the ability of F/AlOH to induce cell-mediated immune responses in seronegative recipients. STUDY DESIGN 5 The purpose of the experiment was to ascertain if the addition of IL-12 to F/AlOH could augment cell-mediated immune responses. Naive female BALB/c mice (8-10 weeks of age) were primed intramuscularly with one of three F protein based vaccines plus recombinant murine IL-12. 10 The native F protein was obtained from the A2 strain of RSV after ion exchange purification. The vaccines were composed of F protein (3 pg/dose) adsorbed to aluminum hydroxide (AlOH, 100 pg/dose) plus one of three 10-fold ascending doses of IL-12 (0.01, 0.1, and 1.0 pg 15 IL-12/dose). The aluminum hydroxide was prepared at Wyeth-Lederle Vaccines and Pediatrics, Pearl River, NY. Control mice were vaccinated with F/AlOH alone, or were intranasally administered mock infected Hep2 cell lysate. Four weeks after primary vaccination the mice were 20 challenged intranasally with RSV A2 (~106 PFU) and bronchoalveolar lavage (BAL) was performed 5 days later. The cytolytic capacity of the inflammatory cells was determined directly in a standard 4-hour 51 Cr release assay after incubation with syngeneic RSV-infected and 25 control target cells. RESULTS The data depicted in Figure 3 demonstrated that IL-12 enhanced the ability of F/AlOH to generate cell-mediated immune responses. The cytolytic activity of the BAL WO 99/40937 PCTIUS99/02848 -26 cells from mice primed with F/AlOH plus 0.01 pg IL-12 (filled triangles) was 39% against syngeneic RSV-infected (solid lines) target cells 5 days after challenge. When the BAL cells from mice primed with F/AlOH alone (filled 5 circles) were incubated with the same syngeneic RSV-infected target cells, cytotoxicity was not detected. The results further implied that the killer cell activity was antigen dependent. The BAL cells from mice vaccinated with F/AlOH plus 0.01 pg IL-12 did not lyse 10 syngeneic control targets (dashed lines) not infected with virus. The data also suggested that doses of IL-12 greater than 0.01 pg were counterproductive for augmenting cell-mediated immune responses (Figure 3). Primary 15 immunization with F/AlOH plus either 0.1 (inverted filled triangles) or 1.0 (right filled triangle) pg IL-12 resulted in BAL cells 5 days after challenge that, when compared to those of mice primed with F/AlOH plus 0.01 pg IL-12, were 3 and 8 times less cytolytic, respectively.

WO 99/40937 PCTIUS99/02848 -27 EXAMPLE 4: The effect of IL-12 on the ability of F/AlOH to induce cell-mediated immune responses in seropositive recipients. STUDY DESIGN 5 The goal of the study was to determine the effect of IL-12 on the capacity of F/AlOH to expand cell-mediated immune responses in recipients previously infected with RSV. Naive female BALB/c mice (8-10 weeks of age) were primed by experimental infection with the A2 strain of 10 RSV. Four weeks later, the mice were injected intramuscularly with one of three F protein based vaccines plus recombinant murine IL-12. The vaccines were composed of F protein (3 pg/dose) adsorbed to aluminum hydroxide (AlOH, 100 pg/dose) plus one of three 15 10-fold ascending doses of IL-12 (0.01, 0.1, and 1.0 pg IL-12/dose). The aluminum hydroxide was prepared at Wyeth-Lederle Vaccines and Pedatrics. Control mice were primed by infection with RSV and secondarily vaccinated with F/AlOH alone, or were intranasally administered mock 20 infected Hep2 cell lysate. Two weeks after secondary vaccination the mice were challenged intranasally with RSV A2 (~106 PFU) and bronchoalveolar lavage (BAL) was performed 5 days later. The cytolytic capacity of the inflammatory cells was determined directly in a standard 25 4-hour "Cr release assay after incubation with syngeneic RSV-infected and control target cells. RESULTS The data depicted in Figure 4 demonstrated that all three doses of IL-12 amplified the ability of F/AlOH to WO 99/40937 PCT/US99/02848 -28 expand the cell-mediated immune responses of mice previously infected with RSV. The cytolytic activity of the BAL cells from mice secondarily vaccinated with F/AlOH plus 0.1 ig IL-12 (inverted filled triangles) was 5 58% (effector:target ratio = 54:1) against syngeneic RSV-infected (solid lines) target cells 5 days after challenge. When the BAL cells from mice primed by infection and secondarily vaccinated with F/AlOH alone (filled circles) were incubated with the same syngeneic 10 RSV-infected target cells, cytolytic activity was 35%. When the killer cell activities were examined at the 18:1 effector:target ratio, the cytolytic activities were 52% and 27%, respectively. The results further implied that the killer cell activity was antigen dependent. The BAL 15 cells from mice primed with RSV and secondarily vaccinated with F/AlOH plus IL-12 did not lyse control targets (dashed lines) not infected with virus. The data further suggested that the ability of F/AlOH to expand cell-mediated immune responses was dependent on 20 the dose of IL-12 in the vaccine (Figure 4). In seropositive animals, the greatest level of antigen dependent killer cell activity appeared to occur in the lungs of mice secondarily immunized with F/AlOH plus 0.1 ig IL-12 5 days after challenge (inverted filled 25 triangles, Figure 4).

WO 99/40937 PCT/US99/02848 -29 EXAMPLE 5: Effect of IL-12 on Predisposition for Atypical Pulmonary Inflammatory Response STUDY DESIGN The purpose of this study was to determine the effect 5 of IL-12 on the capacities of either highly purified native G protein or a facsimile of the Pfizer Lot 100 formalin inactivated vaccine to predispose BALB/c mice for atypical pulmonary inflammatory responses after challenge. Both vaccines were adsorbed to aluminum 10 hydroxide adjuvant. Naive female BALB/c mice (8-10 weeks of age) were vaccinated intramuscularly (IM) at weeks 0 and 4 with either 1 ig purified native attachment glycoprotein (G) adsorbed to aluminum hydroxide (AlOH, Alu-gel-S", Serva, 15 100 .ig dose) adjuvant or 0.1 ml formalin-inactivated RSV (FI-RSV). The FI-RSV vaccine was a facsimile of the original Lot-100 vaccine formulated by Pfizer and was adsorbed to AlOH (1600 ig dose). This vaccine was used as a benchmark for atypical pulmonary inflammatory 20 response, an undesirable immune response for subunit vaccines. IL-12 was added to G/AlOH and FI-RSV in 10-fold ascending doses (0.1 to 10.0 pig IL-12/dose). Additional companion groups of mice were infected (0.05 ml 25 intranasally) with the A2 strain of RSV, injected IM with 0.1 ml formalin-inactivated parainfluenza virus type 3 (FI-PIV3) vaccine adsorbed to AlOH (1600 pig/dose), or were intranasally administered 50 pl of mock-infected Hep2 cell lysate (MOCK). Two weeks after secondary 30 vaccination, serum was collected for the determination of WO 99/40937 PCT/US99/02848 -30 geometric mean endpoint antibody titers by ELISA in microwells coated with either affinity-purified G protein or ion exchange purified F protein. Geometric mean neutralizing antibody titers were also determined by the 5 plaque reduction neutralization test in the presence and absence of complement against the A2 strain of the virus. To characterize the effect of IL-12 on the local pulmonary inflammatory responses, bronchoalveolar lavage (BAL) was performed 5 days after challenge with 10 infectious virus (RSV A2, ~106 PFU). Inflammatory cell morphology was determined after staining of cytopreps with DIF-QIK@ (a reagent which stains leukocytes; Baxter International, Inc., Deerfield, Illinois) and the enumeration of at least 400 cells. The BAL fluids were 15 also examined for the presence of IFN-y and IL-5 by capture-ELISA. Statistical differences between the groups in IL-5 and IFN-y secretion were determined after comparing the OD 490 by ANOVA (analysis of variance by JMP Software; SAS Institute, Cary, North Carolina). 20 RESULTS The data suggest that parenteral immunization with the facsimile of the Lot-100 FI-RSV vaccine predisposed BALB/c mice to generate F and G protein-dependent helper T cell responses that were predominantly of the type 2 25 (Th2) phenotype. The serum anti-G (Table 3) and serum anti-F (Table 4) protein IgG1 to IgG2a antibody ratios 2 weeks after secondary immunization supported this conclusion. The ratio of serum anti-F protein IgGl to IgG2a antibody titers observed after secondary 30 vaccination with FI-RSV was 75.3 (Table 4). Likewise, WO 99/40937 PCT/US99/02848 -31 the ratio of serum anti-G protein IgGl to IgG2a antibody titers observed 2 weeks after secondary vaccination was greater than 184.2 (Table 3). Moreover, immunization with FI-RSV was associated 5 with an elevation in IL-5 production and the induction of atypical pulmonary inflammatory immune responses in the BAL fluids 5 days after challenge (Table 6). In Table 6, the results in the "% EOS" column are the geometric mean relative percentage of eosinophils (EOS) 10 enumerated in the BAL fluids 5 days after challenge with virus. "ND" denotes not determined. IL-5 was detected by capture ELISA and quantified from a standard curve. The results in the "IL-5 (OD)" column are the geometric mean optical density

(OD

490 ). 15 The geometric mean relative percentage of eosinophils (38.5%) observed in the lungs of mice vaccinated with FI-RSV 5 days after challenge was significantly elevated when compared with that of control mice vaccinated with FI-PIV3 (6.5%) and undergoing primary infection, or mice 20 immunized by experimental infection (<1.0%) (Table 6). In addition, the amount of IL-5 secreted into the lavage fluids of mice vaccinated with FI-RSV (106 pg/ml) was significantly elevated when contrasted with those fluids from recipients of either FI-PIV3 (<8 pg/ml) or 25 infectious virus (<35 pg/ml) 5 days after infection (Table 6). The results indicate that the G protein employed in this study contained concentrations of F protein that were immunogenic for BALB/c mice. Noteworthy were the 30 serum anti-F protein total IgG antibody titers observed 2 weeks after secondary vaccination with G/AlOH (Table 4).

WO 99/40937 PCT/US99/02848 -32 The data implied that both G and the contaminating F protein in G/AlOH, like the FI-RSV vaccine, induced primarily Th2 helper T cell subsets. The serum anti-F protein IgGl to IgG2a antibody ratios after secondary 5 vaccination were greater than 617 (Table 4). In similar fashion, the serum G protein-specific IgGl to IgG2a antibody ratios after secondary vaccination with G/AlOH were greater than 1251 (Table 3). In addition, immunization with G/AlOH was associated with the 10 induction of atypical pulmonary inflammatory immune responses and the presence of IL-5 in the BAL fluids 5 days after challenge (Table 6). The geometric mean relative percentage of eosinophils (35.0%) observed in the lungs of mice twice vaccinated with G/AlOH 5 days 15 after challenge was significantly elevated when compared with that of control mice vaccinated with FI-PIV3 (6.5%) and undergoing primary infection. Moreover, the amount of IL-5 secreted into the lavage fluids of mice vaccinated with G/AlOH (167 pg/ml) was significantly 20 elevated when contrasted with those fluids from recipients of either mock infected Hep-2 cell lysates (<35 pg/ml), infectious virus (<35 pg/ml), or FI-PIV3 (<8 pg/ml) 5 days after infection (Table 6). In contrast to vaccination with either FI-RSV or 25 G/AlOH, infection with the wild-type A2 strain of RSV elicited F protein dependent type 1 (Thl) helper T cell responses. The systemic humoral immune responses elicited by the F protein contained in infectious virus was characterized by secondary serum anti-F protein IgG1 30 to IgG2a antibody ratios less than 1.0 (Table 4). Most importantly, atypical pulmonary inflammatory responses WO 99/40937 PCTIUS99/02848 -33 were not associated With previous RSV infection (Table 6). Pulmonary eosinophilia was not observed in naive mice undergoing primary infection (Table 6). The results suggest that IL-12 had a profound impact 5 on the humoral immune responses generated after vaccination with either the facsimile vaccine or G/AlOH. The results implied that the dose of IL-12 was important. For example, the data suggested that the addition of either 0.1 or 1.0 ig IL-12 to the F protein contaminated 10 G/AlOH resulted in serum anti-F protein IgG2a antibody titers 2 weeks after secondary immunization that were augmented 4 and 1,300-fold respectively when compared to titers from mice vaccinated with G/AlOH alone (Table 4). The inhibitory effect of IL-12 was also observed 15 following addition of 10 pag to G/AlOH. The data implied that IL-12 limited the ability of G/AlOH to generate anti-F protein IgG1 antibody titers. When compared to the titers generated by G/AlOH alone, secondary immunization with the G/AlOH plus 10 pag IL-12 generated 20 IgG1 antibody titers that were 5-fold and significantly less (Table 4). The data indicated that 10 pig IL-12 inhibited the capacity of the Lot-100 facsimile vaccine to elicit its characteristic anti-F protein total IgG antibody titers. The serum antibody titers were 17 fold 25 less than those of mice twice immunized with FI-RSV plus 0.1 ig IL-12 (Table 4). The results shown in Tables 3 and 4 are the geometric mean endpoint antibody titers determined by ELISA. 1/2a is the ratio of geometric mean IgG1 to IgG2a subclass 30 antibody titers. "NT" denotes not tested; "ND" denotes not determined.

WO 99/40937 PCT/US99/02848 -34 The addition of IL-12 did not appear to dramatically alter the magnitude of anti-G protein total IgG antibody titers generated after secondary vaccination with either FI-RSV or G/AlOH (Table 3). However, the presence of 5 IL-12 in the vaccines significantly diminished IgG1 antibody titers, while the IgG2a protein specific antibody titers were significantly elevated (Table 3). The results imply that IL-12 modified the capacity of the vaccines to induce Th2 and Thl helper T cell 10 responses. The transformations of serum G and F protein specific IgGl and IgG2a antibody titers (Tables 3 and 4) support this hypothesis. For example, 2 weeks after secondary immunization the F protein-specific antibody subclass ratio was reduced from 617.1 to 92.1 with the 15 addition of 0.1 ig IL-12 to G/AlOH (Table 4). Moreover, a 10-fold increase in IL-12 from 0.1 to 1.0 jig per dose resulted in a reduction in the F protein-specific antibody subclass ratio to 0.1. However, the data suggested that doses of IL-12 greater than 1.0 ig were 20 counterproductive. For example, the F protein-specific IgG2a antibody subclass titers elicited after secondary vaccination with G/AlOH plus 1.0 jig IL-12 were 1,300 times greater when compared to those generated by G/AlOH alone (Table 4). However, immunization with G/AlOH plus 25 10 jig IL12 resulted in F protein-specific IgG2a subclass antibody titers that were comparable to those generated after vaccination with G/AlOH plus 1.0 jig IL-12. The data imply that IL-12 has the ability, via the induction of distinct helper T cell subsets, to modify 30 the infiltration and/or replication of eosinophils in the pulmonary tissues after challenge. This was exemplified WO 99/40937 PCT/US99/02848 -35 by the effect of IL-12 on the capacity of the vaccines to predispose mice for increased amounts of IL-5 and relative percentages eosinophils in the lungs 5 days after challenge (Table 6). When compared with the BAL 5 fluids from mice vaccinated with FI-RSV alone, the addition of 0.1 or 1.0 ig IL-12 to the FI-RSV vaccine significantly reduced the amount of IL-5 and relative number of eosinophils, respectively (Table 6). The addition of 1.0 pg IL-12 per dose to G/AlOH also resulted 10 in the diminution in IL-5 in the BAL fluids after challenge. However, IL-12 did not appear to have any transforming effect on the capacity of G/AlOH to predispose BALB/c mice to pulmonary eosinophilia after challenge (Table 6). It was noteworthy that atypical 15 pulmonary inflammatory responses were not observed in mice immunized by infection with the A2 strain of virus. Furthermore, IL-5 and eosinophilia were not observed in the lungs of control mice 5 days after primary infection (Table 6). IFN-y was at or near baseline for all groups 20 examined (data not shown). The results shown in Table 5 are the geometric mean neutralizing antibody titers. The titers were determined 4 and 2 weeks after primary and secondary vaccination respectively by the plaque reduction neutralization test 25 and in the presence (+) or absence (-) of 5% serum as a source of complement. Although IL-12 transformed the ratio of F and G protein-specific IgG1 to IgG2a antibody subclasses, there was no statistical augmentation in complement-assisted 30 neutralizing antibody titers 4 and 2 weeks after primary and secondary vaccination respectively (Table 5).

WO 99/40937 PCT/US99/02848 -36 Indeed, the results implied that IL-12 diminished serum neutralizing antibody titers 2 weeks after secondary vaccination. The geometric mean serum neutralizing antibody titers of mice vaccinated with FI-RSV alone were 5 at least 4 times greater than that of recipients of FI-RSV plus 10 pg IL-12. The presence of 10 lpg IL-12 in G/AlOH was also associated with a statistically significant 10-fold reduction in complement-independent neutralizing antibodies 2 weeks after secondary 10 vaccination. However, the magnitude of the complement-dependent neutralizing antibody titers elicited by the F protein contaminated G/AlOH were equivalent to that of mice vaccinated by experimental infection (Table 5).

WO 99/40937 PCT/US99/02848 -37 0 02Lf H- 0'% C\N m C) C) > 02 z CdC) C) CD 41 C) C) H H C V co Lfl u AI 0 m C' ~ N E-i co r- c) DD E u~ H 1:i C) z t-- mf C) H- z (N cq V;4 -y %D-~ UHH C) H C) rj-j H H 4j rX E- C) 3 cW cc N E-q 0 - C) kD I- ~ 2 > ( Ir 04 H Hi - CD cl m 04 v I (N 1 1 (N 0 N >N r) c) H Hf - C rH 0~ a) 41 H H u 0 ( L) m~ ry) (N C'1 C) M m' rf) H co a) (NI t- '0i H za > V 4 u l M Ln I I If I >d M H CY) N O' 41 r0 L 0~"0( co >n 0i in H 0 04- a) C) H- H- CD CD C) H-- H- CD C) C) O;0 H H rI C C) 4-4 '44 '4-4 C0 a) H a)) E-i a) ri2 wl wl ml H 0 00 0 H- Q) 0 0 0 Pr Z 4 4 E-1 41)- _ _ _ _ WO 99/40937 PCT/US99/02848 -38 Table 3 Legend a The numbers are the geometric mean endpoint antibody titers. The titers were determined by ELISA on sera collected 2 weeks after secondary vaccination. NT 5 and ND denote not tested and not determined respectively. There were 5 mice per group. b P<0.05 vs. total serum IgG antibody titers from mice vaccinated with FI-RSV alone or plus 0.1 or 1.0 pig L-12. 10 C P<0.05 vs. total serum IgG antibody titers from mice vaccinated with FI-RSV plus 0.1 pg IL-12. d P<0.05 vs. serum IgG1 antibody titers from mice vaccinated with FI-RSV plus 1.0 or 10.0 pg IL-12. e P<0.05 vs. serum IgG2a antibody titers from mice 15 vaccinated with FI-RSV plus 0.1 pg IL-12. P<0.05 vs. serum IgG2a antibody titers from mice vaccinated with FI-RSV plus 0.1 or 1.0 pg IL-12. g P<0.05 vs. serum IgG1 antibody titers from mice vaccinated with G/AlOH plus 10.0 pg IL-12. 20 h P<0.05 vs. serum IgG2a antibody titers from mice vaccinated with G/AlOH plus 0.1, 1.0, or 10.0 lg IL-12.

WO 99/40937 PCTIUS99/02848 -39 (N (NC\ ~1 ( Cl1 r-4 rn H rHi mY a) r d 0 0 0 Lf 0 0) N >- C0 Q) Cq *4 a) >1 t- c 0 ~ r) CN CN E 4 0 rH 00m . - k0 z .dH m(No 0 ~ 10 r- -HH U) H 04~ E-4 ":: 00C l rT4 I- co r-i L tv L n U rl i CC) v C) -l m - m wd H H - rl 0 C , 0 C F- H H 1;3 (Ni C-) "0 l Ln C)' c' Vn oo 0l CN E co. IZ n r o 41 1 H 0o0 CD 00 C i c n C (d H 'Ln w " m ') IZJ 00 m ;li 0 CN N (NI ~ C E-1H '4-4 - :1 0) ri iH C0 c0 0 i-A -4 0 CD 0 o o Cdi i U)4 r. U) a) a) U) U) M rq > > > > U ) U) M MOC OC H 0 0 0 0 0-. 124 pr. PQZ P -i iH -H rH rHi >d H H H H H "'. " - " O 0 Ln' C0 Ln' ri i- WO 99/40937 PCT/US99/02848 -40 Table 4 Legend a The numbers are the geometric mean endpoint antibody titers. The titers were determined by ELISA on sera collected 2 weeks after secondary vaccination. NT 5 and ND denote not tested and not determined respectively. There were 5 mice per group. b P<0.05 vs. total serum IgG antibodies from mice vaccinated with FI-RSV plus 0.1 or 1.0 pg IL-12. c P<0.05 vs. total serum IgG antibodies from mice 10 vaccinated with FI-RSV plus 0.1, 1.0, or 10.0 pg IL-12. d P<0.05 vs. serum IgG1 antibodies from mice vaccinated with FI-RSV alone or plus 0.1 or 1.0 pg IL-12. e P<0.05 vs. serum IgG2a antibodies from mice 15 vaccinated with FI-RSV alone or plus 0.1 pg IL-12. P<0.05 vs. serum IgG2a antibodies from mice vaccinated with FI-RSV plus 0.1 or 1.0 pg IL-12. g P<0.05 vs. serum IgG1 antibodies from mice vaccinated with G/AlOH alone. 20 h P<0.05 vs. serum IgG2a antibodies from mice vaccinated with G/AlOH alone or plus 0.1 pg IL-12.

WO 99/40937 PCT/US99/02848 -41 Table 5. The effect of IL-12 on the ability of native G protein adsorbed to AlOH and FI-RSV to induce serum neutralizing antibodies. NEUTRALIZING ANTIBODY TITERSa PRIMARY SECONDARY VACCINE IL-100 <) -)10 < FI-RSV 10.0 <10 <10 <20 <20 FI-RSV 1.0 13 <10 34 <20 FI-RSV 0.1 27 <10 90C <20 FI-RSV NONE 11 <10 85c <20 FI-PIV3 NONE <10 <10 <20 <20 G/A1OH 10.0 38 <10 1,116 35 d G/A1OH 1.0 150 13 2,155 67 d G/A1OH 0.1 99 22 1,935 117 G/A1OH NONE 41 12 3,278 382 RSV NONE 194 14 2,794 113 MOCK NONE <10 <10 <20 <20 a The numbers are the geometric mean neutralizing antibody titers. The titers were determined 4 and 2 weeks after primary and secondary vaccination respectively by the plaque reduction neutralization test and in the presence (+) or absence (-) of 5% complement. There were 5 mice per group. b BALB/c mice were immunized intramuscularly with either FI-RSV or G/AlOH alone, or plus 10-fold ascending doses of recombinant murine IL-12. C P<0.05 vs. serum neutralizing antibody titers from mice vaccinated with FI-RSV plus 1.0 or 10.0 pg IL-12, or FI-PIV3. d P<0.05 vs. serum neutralizing antibody titers from mice vaccinated with G/AlOH alone, 1 pig IL-12 or 10 ig IL-12.

WO 99/40937 PCT/US99/02848 -42 Table 6. The effect of IL-12 on the pulmonary inflammatory responses of BALB/c mice vaccinated with either native G protein adsorbed to AlOH or formalin-inactivated RSV 5 days after challenge with the 5 A2 strain of RSV. IL-5 IL-5 VACCINEa IL-12 9EOSb (pg/ml) (OD)* FI-RSV 10 ND <7 0.029 FI-RSV 1 10.9 <7 0.035 FI-RSV 0.1 17.8 <10 0.044 0 FI-RSV 0 3 8 .5d 106 0

.

1 6 0 f FI-PIV3 0 6.5 <8 0.036 G/AlOH 10 31.1 50 0.115 G/AlOH 1 24.7 24 0.074 G/AlOH 0.1 32.8 37 0.096 5 G/AlOH 0 35.0 167 0.1919 RSV 0 <1.Oe <35 0.028 MOCK 0 ND <35 0.035 a BALB/c mice were immunized intramuscularly on weeks 0 and 4 with formalin-inactivated RSV (FI-RSV) plus 10 0 fold ascending doses of recombinant murine IL-12. Control mice were vaccinated with formalin-inactivated parainfluenza virus type 3 (FI-PIV3), natural fusion protein admixed with QS-21 (F/QS-21) or infected with the A2 strain of RSV. 5 Additional control mice received an intranasal administration of mock-infected Hep2 cell lysate (MOCK) or were injected intramuscularly with PBS/QS-21. b The numbers are the geometric mean relative 0 percentage of eosinophils (EOS) enumerated in the BAL fluids 5 days after challenge with virus. ND denotes not determined.

WO 99/40937 PCT/US99/02848 -43 C IL-5 was detected by capture ELISA and quantified from a standard curve. The numbers are the geometric mean optical density (OD 49 0 ) . d P<0.05 vs. the eosinophils detected in mice vaccinated with either FI-PIV3 or FI-RSV plus 1.0 ig IL-12. e P<0.05 vs. the eosinophils detected in mice vaccinated with either G/AlOH or FI-RSV alone. P<0.05 vs. the IL-5 (OD 4 90 ) detected in mice vaccinated with FI-RSV plus 0.1, 1.0, or 10.0 pg IL-12. g P<0.05 vs. the IL-5 (OD 4 90 ) detected in mice vaccinated with FI-RSV plus 1.0 pg IL-12.

WO 99/40937 PCT/US99/02848 -44 EXAMPLE 6: The Effect of IL-12 on the Capacity of F/AlOH to Generate Protective Immune Responses in BALB/c Mice. STUDY DESIGN The purpose of the study was to investigate the capacity of IL-12 to improve the ability of F/AlOH to generate protective immune responses in the lungs. Naive female BALB/c mice (8-10 weeks of age) were primed intramuscularly (IM) with ion exchange purified natural fusion (F) protein from the A2 strain of RSV. The F protein (30 ng/dose) was administered in combination with PBS alone, or with one of two doses of recombinant murine IL-12 (10 or 100 ng IL-12/dose). The F protein and IL-12 were adsorbed to aluminum hydroxide adjuvant (AlOH, 100 pig/dose) overnight at 4 0 C. Additional control mice were immunized by experimental infection with the A2 strain of RSV. Four weeks after primary immunization, all mice were challenged with RSV A2 (50 pil, -5 X 106 PFU). The level of virus replication in the pulmonary tissues was assessed four days later. Briefly, the lung and tracheal tissues were removed en bloc, homogenized, clarified, snap frozen, and stored at -70 0 C until assayed for infectious virus. The level of virus replication in the respiratory tract tissues was assessed in a plaque assay employing Hep-2 cell monolayers. Sera were also collected four weeks after primary vaccination for the determination of geometric mean endpoint anti-F protein total and subclass IgG antibody titers by ELISA. Geometric mean serum neutralizing antibody titers were also revealed by the WO 99/40937 PCT/US99/02848 -45 plaque reduction neutralization test against the A2 strain of virus in the presence or absence of 5% complement. RESULTS The results shown in Figure 5 are the geometric mean 5 plaque forming units (PFU) of virus per gram of pulmonary tissue determined four days after challenge. The data depicted in Table 7 are the geometric mean endpoint anti-F protein IgG antibody titers determined by ELISA. The neutralizing antibody titers are the geometric mean 0 neutralizing antibody titers and are determined by the plaque reduction neutralization test in the presence (+) or absence (-) of 5% complement. The antibody titers were determined four weeks after primary vaccination. The addition of either 10 or 100 ng IL-12 to F/AlOH 5 elicited immune responses that were significantly more efficacious than those generated four weeks after primary immunization with F/AlOH in PBS alone. Infectious virus was not detected in the pulmonary tissues of mice primed with F/AlOH plus either 10 or 100 ng IL-12 and challenged 0 with RSV A2 (Figure 5). In contrast, the lungs of mice primed with either F/AlOH in PBS alone, or PBS/AlOH alone contained greater than 3 log 0 PFU virus. An examination of the sera for F protein-specific endpoint IgG and complement-assisted neutralizing antibody 5 titers suggested that the increased efficacy of the IL-12 formulated vaccines was related to heightened systemic humoral immune responses (Table 7). The mean complement assisted neutralizing antibody titers generated after immunization with F/AlOH plus 100 ng IL-12 were 0 significantly greater than those of mice primed with WO 99/40937 PCT/US99/02848 -46 F/AlOH in PBS alone. Nonetheless, immunization with F/AlOH plus 10 ng IL-12 did not result in elevated F protein-specific endpoint IgG and complement-assisted neutralizing antibody titers (Table 7). However, the lungs of these mice inhibited virus replication (Figure 5). Thus, a correlation between IL-12, increased systemic humoral immune responses, and improved efficacy could not be established. In all likelihood, the lack of association was related to the less than optimal 30 ng D dose of F protein in the vaccine. Regardless, the addition of as little as 10 ng IL-12 enhanced the ability of F/AlOH to elicit systemic immune responses in BALB/c mice that were significantly more efficacious than those generated after vaccination with F/AlOH in PBS alone.

WO 99/40937 PCT/US99/02848 -47 Table 7. The effect of recombinant murine interleukin-12 on the capacity of fusion (F) protein adsorbed to aluminum hydroxide (AlOH) adjuvant to generate protective humoral immune responses in BALB/c mice SERUM ANTIBODY TITERS (Log,,)* Anti-F Protein Neutralizing Antigen IL-12 (ng) IgG IgG1 IgG2a + F/AlOH NONE 5.7±0.1 5.3±0.2 3.2±1.1 1.5±0.2 <1.3 F/AlOH 10 6 .2± 0 .2 b 5 .7±0.3b 4

.

7 ±1.

3 b 2 .2± 0

.

9 b 1.3 F/AlOH 100 6 .3 0 .8b 5.

7 ±0.

7 b 5.0±1.7a 2.5±0.8a 1.3 PBS NONE <3.0 <3.0 <3.0 <1.3 <1.3 RSV NONE 5.6±0.1 4.6±0.2 5.9±0.9 2.3±0.3 1.4 * BALB/c mice were primed intramuscularly (IM) with ion exchange purified F protein (30 ng/dose) adsorbed to aluminum hydroxide adjuvant (AlOH, 100 pg/dose) adjuvant. The F/AlOH was administered with PBS alone, or in combination with 100 or 10 ng recombinant murine IL 12/dose. Additional control mice were immunized by experimental infection with the A2 strain of RSV. The numbers are geometric endpoint IgG and neutralizing antibody titers (log,,) ±1 standard deviation of the mean of 5 mice per group. The neutralizing antibody titers (logo) were determined in the presence (+) or absence (-) of 5% complement. a P<0.05 vs. the serum antibody titers from mice vaccinated with F/AlOH in PBS alone. b P>0.05 vs. the serum antibody titers from mice vaccinated with F/AlOH in PBS alone.

WO 99/40937 PCT/US99/02848 -48 EQUIVALENTS Those skilled in the art will recognize, or be able to ascertain, using no more than routine experimentation, many equivalents to the specific embodiments of the 5 invention described herein. Such equivalents are intended to be encompassed by the scope of the present invention.

Claims (27)

1. A vaccine composition comprising a mixture of a respiratory syncytial virus antigen, an adjuvant amount of interleukin-12 and a mineral in suspension, and optionally comprising a physiologically acceptable vehicle.

2. A vaccine composition according to Claim 1, wherein the interleukin-12 is adsorbed onto the mineral suspension.

3. A vaccine composition according to Claim 1, wherein the interleukin-12 is human interleukin-12.

4. A vaccine composition according to Claim 1, wherein the mineral in suspension is an aqueous suspension of alum.

5. A vaccine composition according to Claim 4, wherein the alum is aluminum hydroxide or aluminum phosphate.

6. A vaccine composition according to Claim 1, wherein the respiratory syncytial virus antigen is selected from the group consisting of the RSV F protein, the RSV G protein and combinations thereof.

7. A vaccine composition according to Claim 1, wherein the respiratory syncytial virus antigen is conjugated to a carrier molecule. WO 99/40937 PCT/US99/02848 -50

8. A vaccine composition according to Claim 7, wherein the carrier molecule is selected from the group consisting of tetanus toxin, diphtheria toxin, pertussis toxin and non-toxic variants thereof.

9. A vaccine composition according to Claim 1, wherein the adjuvant amount of interleukin-12 is from about 0.01 ig to about 1.0 ig.

10. A method of eliciting an immune response to a respiratory syncytial virus antigen, comprising administering to a vertebrate host an effective amount of a vaccine composition comprising a mixture of a respiratory syncytial virus antigen, an adjuvant amount of interleukin-12 and a mineral in suspension, and optionally comprising a physiologically acceptable vehicle.

11. A method according to Claim 10, wherein the interleukin-12 is adsorbed onto the mineral suspension.

12. A method according to Claim 10, wherein the interleukin-12 is human interleukin-12.

13. A method according to Claim 10, wherein the mineral in suspension is an aqueous suspension of alum.

14. A method according to Claim 13, wherein the alum is aluminum hydroxide or aluminum phosphate. WO 99/40937 PCT/US99/02848 -51

15. A method according to Claim 10, wherein the adjuvant amount of interleukin-12 is from about 0.01 ig to about 1.0 pg.

16. A method according to Claim 10, wherein the respiratory syncytial virus antigen is conjugated to a carrier molecule.

17. A method according to Claim 16, wherein the carrier molecule is selected from the group consisting of tetanus toxin, diphtheria toxin, pertussis toxin and non-toxic variants thereof.

18. A method according to Claim 10, wherein the respiratory syncytial virus antigen is selected from the group consisting of the RSV F protein, the RSV G protein and combinations thereof.

19. An immunogenic composition comprising a mixture of a respiratory syncytial virus antigen, an adjuvant amount of interleukin-12 and a mineral in suspension, and optionally comprising a physiologically acceptable vehicle.

20. An immunologic composition according to Claim 19, wherein the interleukin-12 is adsorbed onto the mineral suspension.

21. An immunogenic composition according to Claim 19, wherein the interleukin-12 is human interleukin-12. WO 99/40937 PCT/US99/02848 -52

22. An immunogenic composition according to Claim 19, wherein the mineral in suspension is an aqueous suspension of alum.

23. An immunogenic composition according to Claim 22, wherein the alum is aluminum hydroxide or aluminum phosphate.

24. An immunogenic composition according to Claim 19, wherein the respiratory syncytial virus antigen is selected from the group consisting of the RSV F protein, the RSV G protein and combinations thereof.

25. An immunogenic composition according to Claim 19, wherein the respiratory syncytial virus antigen is conjugated to a carrier molecule.

26. An immunogenic composition according to Claim 25, wherein the carrier molecule is selected from the group consisting of tetanus toxin, diphtheria toxin, pertussis toxin and non-toxic variants thereof.

27. An immunogenic composition according to Claim 19, wherein the adjuvant amount of interleukin-12 is from about 0.01 ig to about 1.0 ig.