INFECTION AND IMMUNITY, Apr. 1998, p. 1334–1341 0019-9567/98/$04.0010 Copyright © 1998, American Society for Microbiology
Vol. 66, No. 4
Intranasal Administration of a Meningococcal Outer Membrane Vesicle Vaccine Induces Persistent Local Mucosal Antibodies and Serum Antibodies with Strong Bactericidal Activity in Humans BJØRN HANEBERG,1* ROLF DALSEG,1 ELISABETH WEDEGE,1 E. ARNE HØIBY,2 INGER LISE HAUGEN,1 FREDRIK OFTUNG,1 SVEIN RUNE ANDERSEN,1 LISBETH MEYER NÆSS,1 AUDUN AASE,1 TERJE E. MICHAELSEN,1 AND JOHAN HOLST1 Department of Vaccinology1 and Department of Bacteriology,2 National Institute of Public Health, N-0403 Oslo, Norway Received 21 October 1997/Returned for modification 26 November 1997/Accepted 7 January 1998
A nasal vaccine, consisting of outer membrane vesicles (OMVs) from group B Neisseria meningitidis, was given to 12 volunteers in the form of nose drops or nasal spray four times at weekly intervals, with a fifth dose 5 months later. Each nasal dose consisted of 250 mg of protein, equivalent to 10 times the intramuscular dose that was administered twice with a 6-week interval to 11 other volunteers. All individuals given the nasal vaccine developed immunoglobulin A (IgA) antibody responses to OMVs in nasal secretions, and eight developed salivary IgA antibodies which persisted for at least 5 months. Intramuscular immunizations did not lead to antibody responses in the secretions. Modest increases in serum IgG antibodies were obtained in 5 volunteers who had been immunized intranasally, while 10 individuals responded strongly to the intramuscular vaccine. Both the serum and secretory antibody responses reached a maximum after two to three doses of the nasal vaccine, with no significant booster effect of the fifth dose. The pattern of serum antibody specificities against the different OMV components after intranasal immunizations was largely similar to that obtained with the intramuscular vaccine. Five and eight vaccinees in the nasal group developed persistent increases in serum bactericidal titers to the homologous meningococcal vaccine strain expressing low and high levels, respectively, of the outer membrane protein Opc. Our results indicate that meningococcal OMVs possess the structures necessary to initiate systemic as well as local mucosal immune responses when presented as a nasal vaccine. Although the serum antibody levels were less conspicuous than those after intramuscular vaccinations, the demonstration of substantial bactericidal activity indicates that a nonproliferating nasal vaccine might induce antibodies of high functional quality.
systemic immune responses after application on mucosal surfaces. Outer membrane proteins from group B meningococci are clearly immunogenic in humans (31), and the OMVs which we used as a mucosal vaccine in mice were originally developed to be the main component of a parenteral vaccine against group B meningococcal disease (12). In a large-scale study of adolescents, this OMV vaccine was shown to protect against disease when given intramuscularly with aluminum hydroxide as adjuvant (6). In the present study, we used OMVs, suspended in saline without aluminum hydroxide, as a mucosal vaccine in the form of nasal drops or spray to human volunteers. The demonstration by others of M cells in the human nasopharyngeal area (30) forms the basis for an effect of such a vaccine when applied intranasally (19). Other researchers have recently also demonstrated that intranasal immunizations with either live influenza virus (18), the B subunit of CT (CTB) (4), or diphtheria-tetanus vaccines (2) can induce specific immune responses in humans. The results with our nasal OMV vaccine against meningococcal disease were compared with those obtained in another group of volunteers who received two intramuscular doses of the parenteral OMV vaccine formulation with aluminum hydroxide. The aim was to determine whether intranasal delivery of such particles might also induce immune responses in humans, and that they might serve as a model system for creating alternative mucosal vaccines against other bacterial diseases.
Vaccines administered directly onto mucosal surfaces may induce local mucosal as well as systemic immune responses (23, 24). Even nonproliferating mucosal vaccines may thus offer a challenging alternative to traditional parenteral vaccines, as has been shown for an oral cholera vaccine (15). It is required, however, that induction of tolerance to antigenic components of such vaccines be abrogated or that so-called mucosal adjuvants be added (10, 24). We have shown that in mice, the nasal mucosa is the preferred site for presentation of a vaccine consisting of whole killed pneumococci in suspension, with cholera toxin (CT) added as mucosal adjuvant (1). Even for intestinal immune responses, as measured by antibodies in feces, nasal immunizations were superior to administering the antigen by both the oral and gastric routes. Subsequently, we found that outer membrane vesicles (OMVs) from group B meningococci were also immunogenic in mice when given nasally (8). The antibody responses to OMVs in these experiments were largely independent of adding CT; i.e., the vesicles themselves possessed the necessary structures for induction of mucosal and
* Corresponding author. Mailing address: Department of Vaccinology, National Institute of Public Health, P.O. Box 4404 Torshov, N-0403 Oslo, Norway. Phone: 47 22 04 25 01. Fax: 47 22 04 23 01. E-mail:
[email protected]. 1334
VOL. 66, 1998
A NASAL OMV VACCINE IS IMMUNOGENIC IN HUMANS MATERIALS AND METHODS
Vaccinees. Twelve healthy volunteers, nine women and three men at 25 to 61 (median, 46) years of age, were included in the nasal vaccine study regardless of their prevaccination antibody levels. They had not previously received a meningococcal vaccine and did not receive other vaccines during the study. Another group of 11 healthy volunteers, seven women and four men at 24 to 49 (median, 38) years of age, were immunized intramuscularly with the regular vaccine formulation and served as controls for the nasally immunized volunteers. This group of volunteers was selected on the basis of low serum immunoglobulin G (IgG) antibody levels to meningococcal OMVs. The reason for different selection criteria in the two groups of volunteers is that the study was originally planned as two separate experiments. Even so, it happened that the prevaccination serum antibody levels in the nasal and intramuscular vaccine groups were not significantly different (P . 0.2), with median (range) levels of 12.4 (7.3 to 77.4) and 12.5 (3.7 to 30.8) kU/ml, respectively (see below for antibody measurements). Except for the known local and systemic adverse effects of the intramuscular vaccine, none of the vaccinees experienced clinical side effects which were attributed to the vaccinations; i.e., on questionnaires concerning possible side effects on days 1, 2, and 7 after each nasal dose, none reported moderate to marked signs of nasal irritation, nasal congestion or secretion, other airway symptoms, or fever or reduced sense of well-being. However, one had a sore throat on day 2, and one had a runny nose on day 7 after the fourth and third doses, respectively. Some felt a taste of the nasal vaccine, but there were no reports of swallowing or aspirating the vaccine. None of the vaccinees were carriers of meningococci by nasopharyngeal cultures taken immediately before or during the study. The study was approved by The Norwegian Medicines Control Authority and the regional Ethics Committee for Medical Sciences in Norway. Vaccines. The intramuscular vaccine contained OMVs from the group B meningococcal strain 44/76 (15:P1.7,16) adsorbed onto aluminum hydroxide (12). The OMVs were prepared by extraction of bacteria with 0.5% deoxycholate in 0.1 M Tris HCl buffer (pH 8.6) containing 10 mM EDTA and purified by differential centrifugation. Each intramuscular dose of 0.5 ml consisted of 25 mg of OMVs, measured as protein. The nasal vaccine was made from the original pool of OMVs used in the intramuscular vaccine formulation, but without aluminum hydroxide. Each nasal dose of 0.5 ml consisted of 250 mg OMVs, measured as protein. Immunizations. The nasal vaccine was given four times at weekly intervals, and a fifth dose was added 5 months later. Six of the volunteers received the vaccine as nasal drops; the other six received it as nasal spray. The drops were delivered by a regular pipette, 0.25 ml (125 mg of protein) into each nostril, with the head of the vaccinees tilted backward from a supine position to create a near vertical pathway to the upper nasal cavity, and the vaccinees remained in that position for 1 min after delivery. The spray was delivered, with the vaccinees seated, as repeated douches by Minigrip metered spray device (Apodan, Copenhagen, Denmark) to total premeasured volumes of 0.25 ml of vaccine into each nostril. Each spray was followed by a deep breath. The parenteral vaccine was given twice in the deltoid muscle at a 6-week interval. Collection of samples. Sera, separated from freshly drawn whole blood, oral secretions, and nasal fluid were obtained before each immunization and at 1, 2, 4, 8, and 21 weeks after the fourth dose and at 3 days and 1, 2, and 4 weeks after the fifth dose. Oral secretions (called saliva) were collected by four absorbent cylindrical wicks (2 by 25 mm; Polyfiltronics Group Inc., Rockland, Mass.), two of which were placed between the lower gum and buccal mucosa at each side after the volunteers had been using chewing gum for 1 min, and left in place for 1 min. Nasal fluid was collected by four similar absorbent wicks, two of which were used to pick up fluid at each nostril after spraying the nasal cavities with approximately 0.4 ml of lukewarm phosphate-buffered saline (PBS; pH 7.2) with use of Minigrip metered spray devices. The wicks with saliva or nasal fluid were placed into 1.5-ml microcentrifuge tubes, and the combined weights of the wicks and tubes were recorded. The weights of the captured secretions were calculated as the difference between the weight before and after collection. Net weights of captured saliva and nasal fluid were 74 to 310 mg (mean, 248 mg) and 147 to 306 mg (mean, 257 mg), respectively. All samples were stored at 220°C until used. Extraction of immunoglobulins from wicks. Proteins were extracted, largely as described before (13), by addition of 500 ml of PBS with the following protease inhibitors: 0.2 mM 4-(2-aminoethyl)-benzenesulfonylfluoride (Boehringer Mannheim GmbH, Mannheim, Germany), 1 mg of aprotinin (Sigma Chemical Company, St. Louis, Mo.) per ml, 10 mM leupeptin (Sigma), and 3.25 mM bestatin (Sigma). After vortexing for 1 min, a small hole was punched into the bottom of each tube, which were placed into another tube measuring 1.2 by 8 cm, and the extracts were collected into the outer tube by centrifugation at approximately 2,000 3 g for 5 min at 4°C. The extracts were stored at 220°C. Quantitation of antibodies and immunoglobulins. Levels of IgA, IgG, and IgM antibodies to OMVs, and total IgA, IgG, and IgM concentrations, were determined by enzyme-linked immunosorbent assay (ELISA) using Nunc immunoplates (MaxiSorp F96; A/S Nunc, Roskilde, Denmark). Plates for specific antibody assays were coated by incubation with OMVs, 4 mg per ml in Tris HCl buffer (pH 8.6), at 4°C for 1 week. Nonspecific protein binding sites were blocked with PBS (pH 7.2) containing 5% nonfat dry milk (Oxoid, Hampshire, United Kingdom) immediately before use. A sample of saliva from one donor with
1335
high-titered IgA antibodies to OMVs was used as reference standard for specific IgA antibodies in secretions, and sera from different donors with high-titered IgA, IgG, and IgM antibodies were used as reference standards for immunoglobulin in serum and for IgG and IgM antibodies in secretions. Twofold dilutions of both test samples and standard solutions were made, and sample volumes of 100 ml were applied to ELISA plates and incubated overnight at 4°C. After being washed with PBS containing 0.05% Tween (PBS-Tween), plates were incubated for 1 h at room temperature with horseradish peroxidaseconjugated goat antibodies specific for either human IgA, IgG, or IgM (Sigma) and developed with o-phenylenediamine (Sigma). Optical densities were read at 492 nm with Titertek Multiscan MK II (Labsystems, Helsinki, Finland). Standard curves were generated, and arbitrary units were determined based on reference standards (16). To avoid the diluting influence on antibody concentrations in nasal fluid by the various amounts of buffer sprayed into the nose and by the variations in flow of saliva, concentrations of specific antibodies in secretions were related to the total concentrations of the respective immunoglobulin isotype (11). Such corrected antibody concentrations were expressed as the ratio of specific antibodies (units) per weight unit of the corresponding immunoglobulin. Concentrations of total IgA, IgG, and IgM in samples of secretions and sera were determined by ELISA as described above except that the plates were coated with affinity-purified goat antibodies directed against human IgA (a-chain specific), IgG (g-chain specific), or IgM (m-chain specific) (all from Sigma). After incubation with standard and unknown samples, bound immunoglobulins were detected with peroxidase-conjugated goat antibodies to human IgA, IgG, or IgM (Sigma). Purified human IgA, IgG, and IgM (DAKO A/S, Glostrup, Denmark) were used as standards. Immunoblot analyses of antibody specificities. IgA and IgG antibodies to OMV antigens were analyzed by immunoblotting as described previously (26, 28). Electroblots from 12% polyacrylamide gels (7 by 8 cm), loaded with 45 mg of the same OMV preparation as for the ELISAs, were cut into about 25 strips and incubated overnight at room temperature with 1:200 dilutions of sera taken before nasal vaccination and 2 weeks after the fourth and fifth doses, respectively. The corresponding nasal fluids and saliva samples were diluted 1:10. All sera and extracts were incubated in the presence and absence of 0.15% Empigen BB (Albright and Wilson, Cumbria, United Kingdom) to increase renaturation of outer membrane proteins (27). Binding of IgA and IgG in samples was detected after 2 h of incubation with a 1:1,000 dilution of peroxidase-conjugated goat anti-human IgA (Sigma) and 1:500 dilution of peroxidase-conjugated rabbit anti-human IgG (DAKO), respectively. Blots were stained for 10 min with 3-amino-9-ethylcarbazole and hydrogen peroxide (26). Intensity of antibody binding to the different antigens was determined both visually and by scanning image analysis with a video camera and Cream 1-D software system (Kem-EnTec A/S, Copenhagen, Denmark). Two or more guide strips from each blot served to identify the major antigens after incubation with monoclonal antibodies directed against class 1, class 4, and Opc proteins. SBA. The serum bactericidal activity (SBA) assay was performed with an agar overlay method on microtiter plates as described previously (14). Briefly, twofold dilutions of sera, starting at 1:2, were inoculated with about 80 to 100 CFU per well of meningococci in logarithmic growth phase, which was obtained with the 44/76-SL (SL) strain grown for 4 h in 5% CO2 atmosphere on brain heart infusion agar with 1% normal horse serum. Human serum, obtained by plasmapheresis and conferring no reduction in bacterial survival after 60 min of incubation, was used at 25% (final dilution) as a complement source (20). Agar was added to the plates after a 30-min incubation of the reaction mixture in air at 37°C. The number of surviving CFU was counted after overnight incubation in 5% CO2 at 37°C, and the titers are given as the highest reciprocal serum dilutions killing more than 50% of the inoculum. The SL strain of the inoculum, although containing the opc gene, expressed the Opc antigen only weakly (21, 22). This was controlled with the use in every assay of a monoclonal antibody (154, D11) which never showed any bactericidal effect on this inoculum, whereas it always killed a variant 44/76 strain expressing more of the Opc antigen (22). SBA against this variant strain was also tested for. Statistics. Differences of significance between groups of vaccinees, or values obtained at various times, were determined by the two-tailed Mann-Whitney U test and the Wilcoxon signed rank test, respectively. Simple linear regressions and correlation coefficients were calculated with use of StatView 5121 program for Macintosh computers.
RESULTS Nasal vaccine induced strong mucosal ELISA antibody responses. After the first one or two doses of nasal vaccine, at least twofold increases in IgA antibody levels to OMVs were observed in nasal secretions from 9 of the 12 vaccinees. The mean levels of such antibodies, which reached about 10 times the prevaccination levels, remained high until at least 3 months after the start of immunizations (Fig. 1). This was markedly different from the constant low levels of nasal IgA antibodies in the group of individuals receiving the intramuscular vaccine.
1336
HANEBERG ET AL.
FIG. 1. Increases of IgA antibodies (corrected for by total IgA) to meningococcal OMVs in nasal fluid and saliva after immunizations (marked by arrows) of 12 volunteers with a nasal OMV vaccine with no adjuvant and of 11 volunteers with an intramuscular (I.M.) OMV vaccine adsorbed to aluminum hydroxide. The means 6 standard errors, measured in ELISA, are given as percentages of prevaccination levels.
Although the concentrations of nasal antibodies in four of those who received the nasal vaccine were still at least double the prevaccination levels after 6 months, the mean level was then not significantly raised. In some individuals, increases in nasal mucosal IgA antibodies were found after the fifth nasal dose, given at 6 months from the start, but this difference was likewise not significant. We could therefore not demonstrate any local mucosal booster effect of the nasal vaccine. Increased levels of IgA antibodies to OMVs, although less pronounced than in nasal secretions, were also observed in saliva after nasal immunizations. The mean postvaccination concentrations reached almost three times the prevaccination levels (Fig. 1), and unlike the findings in nasal fluid, saliva IgA antibodies after 6 months were significantly higher than the prevaccination levels (P , 0.05). We have thus demonstrated that a mucosal vaccine can initiate a continuous production of mucosal antibodies for that period of time. From comparison of IgA antibody levels in nasal fluid before and 3 months after start of the nasal immunizations, it was evident that all vaccinees had responded with at least twofold increases (Fig. 2). The magnitude of these responses was approximately the same in those with high preexisting local mucosal antibodies as in those with low antibody levels, and it did not seem to be influenced by the way the vaccine had been administered, i.e., as nasal spray or as drops. Concerning the antibodies in saliva, however, only 8 of the 12 vaccinees were considered responders with at least twofold increases at 3 months from the start of vaccinations (Fig. 2). In saliva, moreover, none of those who had relatively high prevaccination antibody levels were considered responders. On the other hand, these nonresponders were also among those who had received the nasal vaccine as spray and not as drops. Since the antibody responses in nasal fluid did not seem to have been influenced by preexisting local mucosal antibodies, the relatively poor salivary antibody responses might be due to the way the nasal vaccine had been administered. However,
INFECT. IMMUN.
this difference in saliva antibody responses after immunizations with spray rather than drops was not significant (P 5 0.2). Only low levels of IgG or IgM antibodies to OMVs could be demonstrated in nasal secretions and saliva from either group of vaccinees (results not shown). Generally, the absolute concentrations in nasal secretions of these antibody isotypes were only 1 to 3% of the corresponding serum concentrations, with exceptional values reaching 9 and 8% of serum IgG and IgM antibodies, respectively. The IgG antibodies in nasal secretions were thus most likely the result of passive leakage from serum and not the result of local production in the mucosa. Nasal vaccine induced modest serum ELISA antibody responses. Antibody responses in serum after nasal immunizations were much less pronounced than in secretions. Approximately twofold increases in the mean levels of IgG antibodies to OMVs were attained after two to three doses of the nasal vaccine (Fig. 3). The corresponding mean IgA antibody levels increased almost threefold, whereas no significant increase in serum IgM antibodies was observed. This was markedly different from the responses in those who received the intramuscular vaccine, with maximal 20-, 10-, and 3-fold increases in mean levels of IgG, IgA, and IgM, respectively. However, after the nasal vaccine, the mean IgG antibody levels remained constant up to 6 months after the start of the experiment. As in secretions, we did not observe any significant booster effect of the fifth dose intranasally. Analyses of serum antibody concentrations 3 months after the start of immunizations showed that only 5 of 12 individuals
FIG. 2. Comparisons of IgA antibody concentrations to meningococcal OMVs in nasal fluid and saliva from 12 volunteers, before and 3 months after the start of immunizations with a nasal OMV vaccine and no adjuvant and from 11 volunteers with an intramuscular (I.M.) OMV vaccine adsorbed to aluminum hydroxide. Individual pre- and postimmunization values, measured in ELISA, are corrected for by total IgA and given as ratios of specific anti-OMV IgA, in arbitrary units per milliliter, to total IgA in micrograms per milliliter. The diagonal on each graph indicates the same post- as preimmunization values, and two lateral lines indicate twofold increases and decreases of concentrations.
VOL. 66, 1998
A NASAL OMV VACCINE IS IMMUNOGENIC IN HUMANS
1337
were less distinct on the picture), whereas the binding to the class 4 protein did not appear to increase after immunizations. Some individuals who had received the nasal vaccine responded with antibodies in secretions directed against the same antigens as the serum antibodies, e.g., against class 1 protein in vaccinee 011M (Fig. 5). In others, as in vaccinee 005M, antibodies against class 1 and 4 proteins were found in secretions, whereas class 1 and 3 protein antibodies were demonstrated in serum. Nasal immunizations may thus induce mucosal immune responses somewhat different from systemic responses. Nasal vaccine induced serum antibodies with strong bactericidal activity. Increases of in vitro bactericidal activity were demonstrated in sera from several individuals who received the nasal vaccine (Fig. 6). This was evident both with the meningococcal SL strain that had been used for the vaccine production and with the homologous strain expressing higher levels of the Opc outer membrane protein (at the time of vaccine production, there was little knowledge about the possible role of Opc as an antigen). The sera with the highest bactericidal titers all had distinct IgG responses on blots against class 1, class 5 (including Opc), and/or LPS antigens.
FIG. 3. Increases of serum IgA, IgG, and IgM antibodies to meningococcal OMVs after immunizations (marked by arrows) of 12 volunteers with a nasal OMV vaccine and no adjuvant and of 11 volunteers with an intramuscular (I.M.) OMV vaccine adsorbed to aluminum hydroxide. The means 6 standard errors, measured in ELISA, are given as percentages of prevaccination levels.
responded to the nasal vaccine with at least twofold increases in IgG antibodies, whereas 10 of 11 individuals responded in this way to the intramuscular vaccine (Fig. 4). Intramuscular vaccinations also induced marked increases in serum IgA antibodies, and the increases were more pronounced than after nasal immunizations. The vaccinees receiving the nasal vaccine were not selected on the basis of their prevaccination serum antibody levels, which also included high values. Serum IgG responses, however, were seen in vaccinees with high as well as low prevaccination levels. Moreover, since there were three responders among those who had received the nasal vaccine as drops and two responders after the spray vaccine, we could reach no conclusion as to whether the systemic antibody responses depended on the way the nasal vaccine had been given. Nasal vaccine induced a mucosal antibody pattern partially different from that of serum. On immunoblots, serum IgG antibody responses to the nasal vaccine were mainly directed against the class 1 (PorA) and class 5 (including Opc) outer membrane proteins, as well as lipopolysaccharide (LPS) and higher-molecular-mass (70- to 80-kDa) proteins (Fig. 5), which are also the main immunogens after intramuscular vaccinations (22). In nasal fluid and saliva, however, no reaction of antibodies to LPS or to the high-molecular-mass components was observed after intranasal immunizations (Fig. 5). The IgA antibodies in the secretions were mainly directed against the class 1 and class 5 proteins (antibodies to the class 5 protein
FIG. 4. Comparisons of serum IgA, IgG, and IgM antibody concentrations to meningococcal OMVs, before and 3 months after the start of immunizations, in 12 volunteers with a nasal OMV vaccine and no adjuvant and in 11 volunteers with an intramuscular (I.M.) OMV vaccine adsorbed to aluminum hydroxide. Individual pre- and postimmunization values, measured in ELISA, are given in arbitrary units or kilounits per milliliter. The diagonal on each graph indicates the same post- as preimmunization values, and two lateral lines indicate twofold increases and decreases of concentrations.
1338
HANEBERG ET AL.
INFECT. IMMUN.
FIG. 5. Immunoblots of serum IgG and nasal fluid IgA antibody responses against OMV antigens for vaccinees 011M and 005M, who received the nasal vaccine as spray and drops, respectively. Samples were taken before vaccination (lanes 1), 2 weeks after the fourth dose (lanes 2), and 2 weeks after the fifth dose (lanes 3). Arrows indicate the positions of the high-molecular-weight (HMW) protein, class 1, 3, 4, and 5 proteins, and LPS. On this reproduction, the increase in IgA binding to the class 5 protein is less distinct.
Three months after start of the study, 5 of 11 vaccinees immunized nasally (one vaccinee was excluded because of antibiotic therapy) had at least fourfold increases in serum bactericidal titers against the SL strain, whereas all 10 who had been immunized intramuscularly (one vaccinee had left the study) had similar increases (Fig. 7). When the strain expressing higher levels of the Opc protein was used in the assay, 8 of 11 in the nasal group and all 10 in the intramuscular group had similar bactericidal activities. Sera from those who responded to the nasal vaccine attained levels of bactericidal activity in the same range of magnitude as in vaccinees given the intramuscular vaccine. In the following months, serum bactericidal activity after
FIG. 6. SBA against meningococci after immunizations (marked by arrows) of 11 volunteers with a nasal OMV vaccine and no adjuvant and of 10 volunteers with an intramuscular (I.M.) OMV vaccine adsorbed to aluminum hydroxide. One vaccinee in each group was left out or had dropped out of the study (see text). Individual SBA values are given in titers, corresponding to reciprocal dilutions of serum, reducing by more than 50% the number of CFU of either the SL strain of bacteria expressing low levels of Opc (Opc1/2) or the strain expressing high levels (Opc11).
FIG. 7. Comparisons of SBA against meningococci, before and 3 months after the start of immunizations of 11 volunteers with a nasal OMV vaccine and no adjuvant and of 10 volunteers with an intramuscular (I.M.) OMV vaccine adsorbed to aluminum hydroxide. Individual pre- and postimmunization values are given in titers, corresponding to reciprocal dilutions of serum, reducing by more than 50% the number of CFU of either the SL strain of bacteria expressing low levels of Opc (Opc1/2) or the strain expressing high levels (Opc11). The diagonal on each graph indicates the same post- as preimmunization values, and two lateral lines indicate twofold increases and decreases of concentrations.
nasal vaccinations seemed well preserved (Fig. 6). All vaccinees who responded to the nasal vaccine at 3 months were still considered as responders before the fifth vaccine dose at approximately 6 months from the start of the experiment; i.e., 5 and 8 of 11 nasal vaccinees had persistent bactericidal activity against the SL and Opc strains, respectively. After the fifth vaccine dose, however, there was no consistent increase in serum bactericidal activity (Fig. 6). This result confirmed our observations with antibody levels, as measured in ELISA, that a nasal vaccine might not easily induce a detectable booster effect. Surprisingly, only two of the five individuals who had responded to the nasal vaccine with bactericidal activity to the SL strain 3 months after start of the study were also serum IgG responders as determined by at least twofold increases in ELISA. Even so, there was a linear correlation (r 5 0.68, P 5 0.02) between bactericidal activity and IgG antibodies in ELISA (Fig. 8). A stronger correlation, however, was found for the bactericidal activity against the strain expressing higher levels of the Opc protein and serum IgG antibody levels (r 5 0.86, P 5 0.0008). The Opc protein may thus be important for induction of immunity via the mucous membranes. Only four vaccinees had responded with at least twofold increases in serum IgA antibodies to OMVs 3 months after the start of the study (Fig. 4). As for the IgG antibodies, a linear correlation was found between the serum IgA antibody levels and the corresponding bactericidal titers, both against the SL strain (r 5 0.77, P 5 0.005) and against the strain expressing
VOL. 66, 1998
A NASAL OMV VACCINE IS IMMUNOGENIC IN HUMANS
FIG. 8. Correlation between SBA against meningococci and serum concentrations of IgG antibodies to meningococcal OMVs, at 3 months after start of immunizations in 11 volunteers with a nasal OMV vaccine with no adjuvant. Individual SBA values are given in titers, corresponding to reciprocal dilutions of serum, reducing by more than 50% the number of CFU of either the SL strain of bacteria expressing low levels of Opc (Opc1/2) or the strain expressing high levels (Opc11). IgG antibody values, measured in ELISA, are given in arbitrary kilounits per milliliter. Simple regression lines and correlation coefficients are shown after logarithmic transformation of the scales.
the Opc protein (r 5 0.80, P 5 0.003). However, the increases in serum IgA antibodies coincided with those of IgG antibodies (r 5 0.80, P 5 0.002). A possible negative influence of IgA antibodies on the bactericidal activity could therefore not be ascertained. DISCUSSION In addition to the simplicity of administration, the commonly recognized advantage with vaccines applied directly onto mucosal surfaces is their ability to induce mucosal antibodies which might act as a barrier to the invasion of pathogenic microorganisms through the mucosal membranes (7). In this study, we have demonstrated that OMVs from group B meningococci, suspended in saline and given as nasal drops or spray were indeed able to initiate mucosal immunity with transfer into nasal secretions and saliva of specific IgA antibodies. The most marked effect, however, was seen in the local mucosal area which had been exposed to the vaccine, i.e., in secretions from the nasal mucosal area as opposed to saliva or secretions obtained from the adjacent oral cavity. Although others have found that nasal immunizations with CTB lead to antibody responses in vaginal secretions (4), we did not study a possible induction by the nasal OMV vaccine of antibodies at distant mucosal sites. As opposed to concentrations of serum antibodies, which depend on a balance between production and degradation, the persistence of antibodies in secretions depends more on the ability of the local mucosal immune system to keep up a de novo synthesis of antibodies which are continuously secreted (7). It is not expected, therefore, that antibodies in secretions will persist in the same way as antibodies in serum. Our dem-
1339
onstration of elevated antibody levels to meningococcal OMVs in secretions for up to 6 months after the start of nasal immunizations indicated that nonproliferating nasal vaccines might eventually be made to induce a protective barrier for a prolonged time. The demonstration in this study of only low levels of IgG antibodies in secretions, which seemed to mirror serum antibodies, contrasts with the findings by others of relatively high IgG antibody concentrations in nasal secretions after intranasal immunizations with live attenuated influenza vaccine (18) or with CTB (4). Possibly this discrepancy can be explained by differences in effects on the mucosa by the antigens used. Also, the different methods for sampling of secretions from the mucosal surfaces may have influenced the results. From previous experience, we know that intramuscular administration of the aluminum-adsorbed OMV vaccine induces high levels of serum IgG antibodies (22). In comparison OMVs administered intranasally without any adjuvant induced only low levels of serum IgG antibodies in our volunteers. But despite the fact that we in this study on humans used the same nasal doses as some of us previously used in mice (8), we obtained significant increases in serum IgG and IgA antibodies. Moreover, the modestly raised serum antibodies were persistent for the whole observation period. Similar to the continued transfer of antibodies into secretions, this finding suggests that OMVs presented as a nasal vaccine can lead to prolonged systemic immune stimulation. The pattern of antibody responses after nasal immunizations, as revealed by immunoblots, showed that serum IgG antibodies were largely directed against the same antigens (70to 80-kDa high-molecular-mass proteins, class 1 and 5 proteins, and LPS) as were immunogenic by intramuscular administration of the OMV vaccine (22, 29). This finding might indicate that the nasal OMV vaccine is able to induce serum antibodies with at least some protective power. However, the IgA antibody pattern in secretions was more restricted than in serum, as no activity against LPS and the high-molecular-mass components was observed. It was also demonstrated that antibodies specific for a meningococcal immunogen can be induced in secretions and not in serum of that same individual. This finding seems to confirm previous observations that the mucosal immune system can operate independently of the systemic one (7). Antibodies in secretions, with specificities which are not found in serum, might also add to the potential beneficial systemic effects of nasal vaccines. Clinical studies with the OMV vaccine, given intramuscularly with aluminum hydroxide, have shown that the serum bactericidal activity may represent a reasonable in vitro correlate to protection against invasive meningococcal disease (20). Since we found that the nasal OMV vaccine in many of the vaccinees induced serum bactericidal activity in the same range of magnitude as after intramuscular immunizations, it seems likely that this vaccine would also confer protection. Similarly to the modest levels of serum ELISA antibodies which were induced by the nasal vaccine, the bactericidal activity was also remarkably persistent over the whole observation period. Thus, the findings so far indicate that outer membrane particles, without an added adjuvant, possess the antigens and conformation necessary to initiate sustained and strong local mucosal as well as systemic immune responses. Studies in animals have shown that this might also be the case with several airway pathogens presented intranasally as whole heat-inactivated bacteria in suspension (1, 5, 16). The discrepancy between the low IgG antibody responses and the high bactericidal activity in sera after nasal immunizations made us question the identity of the factor responsible
1340
HANEBERG ET AL.
for this bactericidal activity. Others, who observed a similar difference between low levels of specific antibodies and high degree of protection against infection after mucosal immunizations with a live rotavirus vaccine, suggested that the protective effect might be ascribed to a hitherto unknown factor (25). However, the positive correlation that we observed between serum IgG antibody levels to OMVs and the bactericidal activity, especially to the 44/76 meningococcal strain expressing high levels of the Opc protein, indicated that the bactericidal activity was probably conferred by the antibodies. It is likely, therefore, that the antibodies measured by ELISA after nasal immunizations were of higher functional quality than those initiated by intramuscular immunizations. It has been claimed that serum IgA antibodies, which do not normally bind complement, may bind to the microbial antigens and thus inhibit the complement-dependent bactericidal activity (17). Compared to the results after intramuscular immunizations, however, the nasal vaccine in the present study induced only negligible serum IgA antibody increases. Moreover, the demonstration of a positive correlation between serum bactericidal activity and IgA (as well as IgG) antibody levels after nasal immunizations does not support the notion that a nasal vaccine might be more of a hazard in this respect than a vaccine given parenterally. Our observations that neither the local nor the serum antibody responses to the nasal OMV vaccine increased with the third to fourth doses could indicate that further responses are hampered by the presence of local mucosal antibodies. This could also explain the lack of a booster effect, or even a primary response, to the single fifth nasal dose given months later when local mucosal antibodies were still present. If that is the case, our success in animals with OMVs as a presumed vaccine carrier or mucosal adjuvant for killed influenza virus given nasally (9) may have only limited applicability. The limitations of potent immunogens as mucosal adjuvants have also been addressed by others (3). Anyhow, the present results indicate that with improved formulations or methods of delivery, efficient nonproliferating mucosal vaccines may soon be a reality. To further study the functional effects of such experimental vaccines, however, there is a need for good animal models or in vitro correlates to protection. ACKNOWLEDGMENTS We are grateful to Marian R. Neutra and Per Brandtzaeg for valuable discussions and to L. Oddvar Frøholm and Edgar Rivedal for kind assistance. This research project received financial support from the WHO Global Programme for Vaccines and Immunization. Support to B.H. was provided by The Research Council of Norway. REFERENCES 1. Aaberge, I. S., E.-C. Groeng, I. L. Haugen, R. Dalseg, M. Løvik, and B. Haneberg. 1995. Induction of systemic and mucosal antibodies to pneumococcal antigens by mucosal immunizations. J. Cell. Biochem. Suppl. 19A:242. 2. Aggerbeck, H., S. Gizurarson, J. Wantzin, and I. Heron. 1997. Intranasal booster vaccination against diphtheria and tetanus in man. Vaccine 15: 307–316. 3. Bergquist, C., T. Lagergård, and J. Holmgren. 1997. Anticarrier immunity suppresses the antibody response to polysaccharide antigens after intranasal immunization with the polysaccharide-protein conjugate. Infect. Immun. 65: 1579–1583. 4. Bergquist, C., E. L. Johansson, T. Lagergård, J. Holmgren, and A. Rudin. 1997. Intranasal vaccination of humans with recombinant cholera toxin B subunit induces systemic and local antibody responses in the upper respiratory tract and the vagina. Infect. Immun. 65:2676–2684. 5. Berstad, A. K. H., J. Holst, I. L. Haugen, B. Møgster, and B. Haneberg. 1997.
INFECT. IMMUN.
6.
7. 8. 9.
10. 11. 12.
13.
14.
15.
16.
17. 18. 19. 20.
21. 22.
23.
24. 25. 26. 27.
A nasal whole-cell Bordetella pertussis vaccine can induce strong systemic and mucosal antibody responses which are not enhanced by cholera toxin. Vaccine 15:1473–1478. Bjune, G., E. A. Høiby, J. K. Grønnesby, Ø. Arnesen, J. H. Fredriksen, A. Halstensen, E. Holten, A. K. Lindbak, H. Nøkleby, E. Rosenqvist, L. K. Solberg, O. Closs, J. Eng, L. O. Frøholm, A. Lystad, L. S. Bakketeig, and B. Hareide. 1991. Effect of outer membrane vesicle vaccine against group B meningococcal disease in Norway. Lancet 338:1093–1096. Brandtzaeg, P. 1992. Humoral immune response patterns of human mucosa: induction and relation to bacterial respiratory infections. J. Infect. Dis. 165: s167–s176. Dalseg, R., E. Wedege, I. L. Haugen, J. Holst, and B. Haneberg. 1995. Group B meningococcal OMV vaccine as a mucosal immunogen. Clin. Immunol. Immunopathol. 76:S93. Dalseg, R., J. Holst, T. Tangen, G. Stabbetorp, P. Simonsen, E. Jantzen, B. Haneberg, E. Namork, and L. R. Haaheim. 1996. Outer membrane vesicles from group-B meningococci can act as mucosal adjuvant for influenza virus antigens, p. 177–182. In E. Brown, E. Norrby, D. Burton, and J. Mekalanos (ed.), Vaccines 96. Molecular approaches to the control of infectious diseases. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. Elson, C. O. 1996. Cholera toxin as a mucosal adjuvant, p. 59–72. In H. Kiyono, P. L. Ogra, and J. R. McGhee (ed.), Mucosal vaccines. Academic Press, San Diego, Calif. Forrest, B. D. 1992. Effects of sample processing on the measurement of specific intestinal IgA immune responses. Vaccine 10:802–805. Fredriksen, J. H., E. Rosenqvist, E. Wedege, K. Bryn, G. Bjune, L. O. Frøholm, A.-K. Lindbak, B. Møgster, E. Namork, U. Rye, G. Stabbetorp, R. Winsnes, B. Aase, and O. Closs. 1991. Production, characterization and control of MenB-vaccine “Folkehelsa”: an outer membrane vesicle vaccine against group B meningococcal disease. NIPH Ann. 14:67–80. Haneberg, B., D. Kendall, H. M. Amerongen, F. M. Apter, J.-P. Kraehenbuhl, and M. R. Neutra. 1994. Induction of specific immunoglobulin A in the small intestine, colon-rectum, and vagina measured by a new method for collection of secretions from local mucosal surfaces. Infect. Immun. 62: 15–23. Høiby, E. A., E. Rosenqvist, L. O. Frøholm, G. Bjune, B. Feiring, H. Nøkleby, and E. Rønnild. 1991. Bactericidal antibodies after vaccination with the Norwegian meningococcal serogroup B outer membrane vesicle vaccine: a brief survey. NIPH Ann. 14:147–156. Holmgren, J., and A.-M. Svennerholm. 1996. Oral vaccines against cholera and enterotoxigenic Escherichia coli diarrhea, p. 241–253. In H. Kiyono, P. L. Ogra, and J. R. McGhee (ed.), Mucosal vaccines. Academic Press, San Diego, Calif. Hordnes, K., T. Tynning, T. A. Brown, B. Haneberg, and R. Jonsson. 1997. Nasal immunization with group B streptococci can induce high levels of specific IgA antibodies in cervicovaginal secretions of mice. Vaccine 15: 1244–1251. Jarvis, G. A., and J. M. Griffiss. 1991. Human IgA1 blockade of IgG-initiated lysis of Neisseria meningitidis is a function of antigen-binding fragment binding to the polysaccharide capsule. J. Immunol. 147:1962–1967. Moldoveanu, Z., M. L. Clements, S. J. Prince, B. R. Murphy, and J. Mestecky. 1995. Human immune responses to influenza virus vaccines administered by systemic or mucosal routes. Vaccine 13:1006–1012. Neutra, M. R., A. Frey, and J.-P. Kraehenbuhl. 1996. Epithelial M cells: gateways for mucosal infection and immunization. Cell 86:345–348. Perkins, B. A., K. Jonsdottir, H. Briem, E. Griffiths, B. D. Plikaytis, E. A. Høiby, E. Rosenqvist, J. Holst, H. Nøkleby, F. Sotolongo, G. Sierra, C. Huergo, G. M. Carlone, D. Williams, J. Dykes, D. Kapczynski, E. Tikhomirov, J. D. Wenger, and C. V. Broome. Immunogenicity of two efficacious outer membrane protein-based serogroup B meningococcal vaccines among young adults in Iceland. J. Infect. Dis., in press. Rosenqvist, E., E. A. Høiby, E. Wedege, B. Kusecek, and M. Achtmann. 1993. The 5c protein of Neisseria meningitidis is highly immunogenic in humans and induces bactericidal antibodies. J. Infect. Dis. 167:1065–1073. Rosenqvist, E., E. A. Høiby, E. Wedege, K. Bryn, J. Kolberg, A. Klem, E. Rønnild, G. Bjune, and H. Nøkleby. 1995. Human antibody responses to meningococcal outer membrane antigens after three doses of the Norwegian group B meningococcal vaccine. Infect. Immun. 63:4642–4652. Staats, H. F., and J. R. McGhee. 1996. Application of basic principles of mucosal immunity to vaccine development, p. 17–39. In H. Kiyono, P. L. Ogra, and J. R. McGhee (ed.), Mucosal vaccines. Academic Press, San Diego, Calif. Walker, R. I. 1994. New strategies for using mucosal vaccination to achieve more effective immunization. Vaccine 12:387–400. Ward, R. L. 1996. Mechanism of protection against Rotavirus in humans and mice. J. Infect. Dis. 174:S51–S58. Wedege, E., and L. O. Frøholm. 1986. Human antibody response to a group B serotype 2a meningococcal vaccine determined by immunoblotting. Infect. Immun. 51:571–578. Wedege, E., K. Bryn, and L. O. Frøholm. 1988. Restoration of antibody binding to blotted meningococcal outer membrane proteins using various
VOL. 66, 1998
A NASAL OMV VACCINE IS IMMUNOGENIC IN HUMANS
detergents. J. Immunol. Methods 113:51–59. 28. Wedege, E., E. A. Høiby, E. Rosenqvist, and L. O. Frøholm. 1990. Serotyping and subtyping of Neisseria meningitidis isolated by co-agglutination, dotblotting and ELISA. J. Med. Microbiol. 31:195–201. 29. Wedege, E., G. Bjune, L. O. Frøholm, E. A. Høiby, and E. Rosenqvist. 1991. Immunoblotting studies of vaccinee and patient sera from a Norwegian
Editor: J. R. McGhee
1341
serogroup B meningococcal vaccination trial. NIPH Ann. 14:183–186. 30. Winther, B., and D. J. Innes. 1994. The human adenoids. A morphologic study. Arch. Otolaryngol. Head Neck Surg. 120:144–149. 31. Zollinger, W. D. 1991. Meningococcal meningitis, p. 113–126. In S. J. Cryz (ed.), Vaccines and immunotherapy. Pergamon Press, New York, N.Y.
