INFECTION AND IMMUNITY, Aug. 2005, p. 5031–5038 0019-9567/05/$08.00⫹0 doi:10.1128/IAI.73.8.5031–5038.2005
Vol. 73, No. 8
Intranasal Delivery of Group B Meningococcal Native Outer Membrane Vesicle Vaccine Induces Local Mucosal and Serum Bactericidal Antibody Responses in Rabbits David R. Shoemaker,† Nancy B. Saunders,‡ Brenda L. Brandt, E. Ellen Moran, Andrew D. LaClair, and Wendell D. Zollinger* Walter Reed Army Institute of Research, Silver Spring, Maryland 20910-7500 Received 9 February 2005/Returned for modification 25 February 2005/Accepted 11 March 2005
We have previously shown that intranasal immunization of mice with meningococcal native outer membrane vesicles (NOMV) induces both a good local mucosal antibody response and a good systemic bactericidal antibody response. However, in the intranasal mouse model, some of the NOMV entered the lung and caused an acute granulocytic response. We therefore developed an alternate animal model using the rabbit. This model reduces the probability of lung involvement and more closely mimics intranasal immunization of humans. Rabbits immunized intranasally with doses of 100 g of NOMV in 0.5 ml of saline developed serum bactericidal antibody levels comparable to those of rabbits immunized intramuscularly with 25-g doses, particularly when the primary intranasal immunization was given daily for 3 days. Intranasal immunization also induced a local mucosal response as evidenced by immunoglobulin A antibody in saliva, nasal washes, and lung lavage fluids. NOMV from a capsule-deficient mutant induced higher serum bactericidal antibody responses than NOMV from the encapsulated parent. Meningococcal NOMV could be administered intranasally at 400 g with no pyrogenic activity, but as little as 0.03 g/kg of body weight administered intravenously or 25 g administered intramuscularly induced a pyrogenic response. These data indicate that the rabbit is a useful model for preclinical testing of intranasal meningococcal NOMV vaccines, and this immunization regimen produces a safe and substantial systemic and local mucosal antibody response. of noncovalent complexes of purified outer membrane proteins (less than 1% LOS) and group C capsular polysaccharide. The relatively low bactericidal antibody response obtained from the volunteers was probably due in part to altered protein conformation and the exposure of immunogenic epitopes that were not exposed on the surface of the viable organism. In the younger children, who had most likely experienced little natural priming by carriage of Neisseria, the antibody response was predominantly to nonprotective epitopes. It may be particularly important to present the meningococcal OMPs to young children in the most native way possible in order to direct the immune response toward relevant protective epitopes. We identified three ways that the OMPs might be presented as a vaccine in a natural or artificial membrane environment. These three approaches included using native outer membrane vesicles (NOMV) as an intranasal vaccine, presenting purified OMPs and detoxified LOS in liposomes, and using NOMV from a ⌬lpxL2 strain as a parenteral vaccine. These approaches are being evaluated and compared. The present paper describes an animal model used to study the intranasal use of NOMV as a vaccine. Although the most extensively used approach to OMP vaccine preparation, deoxycholate extraction of outer membrane vesicles, has shown considerable promise, the detergent extraction does not preserve the natural membrane structure or composition. Much of the LOS and phospholipids are removed exposing new epitopes, potentially altering OMP conformation and removing most of a relatively conserved antigen (LOS) with protective potential. Our initial experiments in mice have demonstrated that i.n. immunization with meningococcal NOMV vaccines induces
Efforts to produce an efficacious vaccine against serogroup B Neisseria meningitidis have met with only moderate success. Efficacy studies with outer membrane protein-based vaccines have demonstrated efficacy in the range of 50 to 80% (3, 4, 10, 23). Those vaccines which have undergone recent phase 3 testing might be improved by two potential modifications. First, induction of a local mucosal response to supplement the systemic response may lead to more effective immunization against this organism. Intranasal (i.n.) immunization would be a direct approach to stimulating a mucosal response, since the human nasopharyngeal region is the natural habitat for meningococci, and it is believed that nasopharyngeal carriage leads to natural immunization (14). Secondly, a more effective serum bactericidal antibody response might be stimulated if the proteins were in a more native conformation, without any detergent treatment or extraction of lipooligosaccharide (LOS). In a large efficacy trial conducted in Iquique, Chile, strong anti-outer membrane protein (anti-OMP) antibody responses in young children, as measured by enzyme-linked immunosorbent assay (ELISA), did not correlate with protection or with levels of serum bactericidal activity (4). The vaccine consisted
* Corresponding author. Mailing address: Department of Bacterial Diseases, WRAIR, 503 Robert Grant Avenue, Silver Spring, MD 20910-7500. Phone: (301) 319-9901. Fax: (301) 319-9123. E-mail: [email protected]
† Present address: United States Army Medical Materiel Development Activity, Fort Detrick, Maryland. ‡ Present address: National Institutes of Health, Bethesda, Maryland. 5031
SHOEMAKER ET AL.
both local mucosal and systemic antibody responses (21). However, we observed that when 25 l of vaccine was delivered i.n. to anesthetized mice, some of it reached the lungs, where it caused an acute granulocytic response. Preliminary experiments in this laboratory involving intranasal immunization of unanesthetized rabbits have shown that excess liquid was swallowed rather than inhaled, suggesting that the larger animal model would more closely mimic human i.n. immunization. The rabbit has been previously demonstrated to be a good model of shock caused by bacterial endotoxin (6). Safety as demonstrated in such an animal is an important element in the present studies, since a significant component of the vaccine is LOS. We have shown that NOMV can be safely used as an intranasal vaccine in human volunteers and that it induces a high quality antibody response characterized by persistent serum bactericidal antibodies (11, 17). It is not known whether endotoxin present in an intranasal vaccine acts as a mucosal adjuvant. The L3,7 LOS in the NOMV intranasal vaccine was, however, effective in inducing bactericidal antibodies (11). The meningococcal NOMV vaccine used in this study and in previous studies consists of the outer membrane blebs of the meningococcus and contains about 20 to 25% LOS relative to protein. They are extracted without the use of detergent or denaturing substances and are thus in their native configuration. The goal of this study was to evaluate the rabbit as a model for i.n. immunization with future meningococcal NOMV vaccines as well as other intranasal vaccines. The model has utility for studying both immune response and safety. Our results indicate that intranasal immunization with NOMV is nonpyrogenic at high doses (400 g/rabbit) and induces serum bactericidal antibodies as well as immunoglobulin A (IgA) antibodies in mucosal secretions. Moreover, the vaccine administered i.n. to rabbits did not reach the lungs. The immune response thus appeared to be a result of contact with cells in the nasopharyngeal region. Additional experiments in which the vaccine was administered by a peroral or intragastric route demonstrated that the immune response was not a result of contact with gut-associated lymphoid tissue. MATERIALS AND METHODS Bacterial strains and vaccines. The vaccine seed strain was derived from Neisseria meningitidis strain 9162 (B:15:P1.7-2,3:L3,7), which was isolated from a case of meningococcal meningitis in Iquique, Chile, in October 1990. This parent strain 9162 was used to produce NOMV for use in comparative immunogenicity studies and ELISAs and was also used as the target strain in bactericidal assays. Strain 9162 was genetically modified by partial deletion of synX (30) and insertion of a kanamycin resistance gene for selection. The gene synX (25), also called siaA (15), is essential for sialic acid biosynthesis. The resulting mutant, 9162 ⌬synX, is phenotypically capsule negative and cannot sialylate its LOS. The NOMV vaccines consisted of native vesicles extracted from whole cells without the use of detergents or denaturing agents (21, 30, 32). These vesicles were referred to in earlier publications as outer membrane complex or OMC (29). NOMV were extracted in buffer containing 0.05 M Tris-HCl, 0.15 M NaCl, and 0.001 M EDTA by warming to 56°C for 30 min and shearing in a Waring blender for 3 min and were isolated by differential centrifugation (30). Some of the experiments employed a batch of NOMV vaccine (lot no. 0123) prepared from the mutant 9162 ⌬synX strain under cGMP conditions at the Walter Reed Army Institute of Research Pilot Vaccine Production Facility. This lot of vaccine was also used in two clinical studies of intranasal vaccination (11, 17). This vaccine was prepared from cells grown under iron-limiting conditions in order to induce the iron-regulated uptake proteins, and it contained about 25% LOS
INFECT. IMMUN. relative to protein. The vaccine consisted of purified NOMV. Purity was verified by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, Western blotting, UV spectrophotometry, and analysis by negative-stain electron microscopy. The vaccine was bottled in sterile normal saline at 0.8 mg protein/ml and stored at 4°C prior to use. The same lot of vaccine has subsequently been used in two clinical studies (11, 17). Rabbits. New Zealand White rabbits (Hazleton Research Products, Denver, PA, or Charles Rivers Laboratories, Wilmington, MA) were used in all experiments. Intranasal immunization was accomplished with unanesthetized rabbits by using a two-person procedure. Rabbits were held in a supine position and a flexible micropipettor was used to drip 0.5 ml of vaccine in the nares, with about half the volume in each naris. Peroral immunizations were performed by allowing the rabbit to drink 0.5 ml vaccine from a flexible micropipette inserted in the mouth. Gastric immunization with 0.5 ml vaccine was performed by intubation of a restrained rabbit (0.5 ml vaccine followed by 1.0 ml saline to wash the gavage tube). Rabbits immunized intramuscularly also received 0.5 ml vaccine in the middle of the quadriceps muscle. Animal use. Research was conducted in compliance with the Animal Welfare Act and other federal statutes and regulations relating to animals and experiments involving animals and adheres to principles stated in the Guide for the Care and Use of Laboratory Animals (19a). Nasal washes were conducted in a similar manner on unanesthetized supine rabbits with the head tilted to the side and slightly downward. A flexible micropipettor was used to deliver 1.0 ml sterile injectable saline into the upper naris. Nasal washes were recovered in a sterile petri plate after dripping out of the lower naris. Typical recovery volumes were 0.5 to 0.7 ml. Saliva was collected immediately following euthanasia by rotating a sterile cotton-tipped applicator in the mouth for 15 seconds and then placing it in 0.5 ml phosphate-buffered saline (PBS) containing 1% bovine serum albumin and 10 g/ml gentamicin and was frozen immediately on dry ice. Lung lavage fluids were collected following euthanasia by opening the trachea just below the larynx and injecting and aspirating 10 ml PBS containing 1% bovine serum albumin and 10 g/ml gentamicin. The typical recovery was 1 to 2 ml. Any samples contaminated by blood, as determined by hemolysis, were discarded. Serum bactericidal assay. A standard bactericidal assay as previously described (19) was used to measure the level of serum bactericidal activity. Briefly, a fresh log phase culture of meningococci was diluted to a final concentration of 4 ⫻ 104 organisms/ml. Serial dilutions of test sera diluted in Gey’s balanced salt solution with 0.2% gelatin (50 l), plus extrinsic human complement (human serum lacking bactericidal activity against the test strain) (25 l), plus bacterial suspension (25 l) were combined in a 96-well plate. This mixture was incubated with shaking for 1 h at 37°C and plated in duplicate on GC agar medium with defined supplement along with appropriate controls. The number of colonies formed after 16 h of incubation was counted, and the endpoint titer was determined as the greatest dilution of serum that killed ⱖ50% of the organisms. ELISA procedures. An ELISA was performed to determine IgG and IgM levels as previously described (22) except that peroxidase-labeled goat anti-rabbit IgG and IgM antibodies (Organon Teknika-Cappel, Durham, NC) were used. ELISA plates were developed with peroxidase substrate (Kirkegaard & Perry Laboratories, Gaithersburg, MD), and the reaction was stopped by the addition of 1% sodium dodecyl sulfate. Detection of rabbit IgA was done using a four-layer sandwich ELISA. Briefly, 96-well plates (Costar Corp., Cambridge, MA) were coated with NOMV for 2 h at 37°C, washed once, and coated with blocking buffer (0.5% bovine albumin and 0.5% casein) for 1 h at 37°C. The plates were washed twice with PBS, incubated with serial dilutions of test sera overnight at room temperature (RT), and washed four times with PBS. The plates were then incubated overnight at RT with a 1:1,000 dilution of mouse ascites containing a monoclonal IgG anti-rabbit IgA (cell line NRBA; kindly provided by A. Louis Bourgeois, Naval Medical Research Institute, Bethesda, MD) (7a) and washed four times. The plates were then incubated overnight at RT with phosphatase-labeled goat anti-mouse IgG (Kirkegaard & Perry Laboratories, Inc., Gaithersburg, MD), developed using Sigma 104 phosphatase substrate (Sigma Diagnostics, St. Louis, MO), and stopped with 3N NaOH. Absorbances were read at 405 nm. For measurement of antibodies to LOS, purified L3,7 or L8 LOS was noncovalently complexed 1:1 (wt/wt) with bovine serum albumin prior to use as antigen. Complexing was done by combining the LOS and bovine serum albumin in Tris-buffered saline containing 1% Empigen BB (Calbiochem, La Jolla, CA). The mixture was then precipitated with four volumes of cold ethanol and centrifuged to pellet the precipitate. The precipitate was washed once with ethanol, and the final pellet was dissolved in water. Plates were sensitized with 10 g/ml of LOS in Dulbecco’s PBS. Quantitation was done by running standard plates using rabbit IgG (Sigma, St.
VOL. 73, 2005
INTRANASAL MENINGOCOCCAL VACCINE IN RABBITS
FIG. 1. Rabbit antibody responses to 100 g intranasally administered group B meningococcal NOMV vaccines made from the wild-type 9162 strain (closed circles) and a 9162 ⌬synX mutant strain (open circles). A control group of animals was vaccinated with normal saline (closed triangles). (A) Serum bactericidal activity; (B) serum IgG as determined by ELISA with homologous NOMV as the antigen; (C) serum IgA as determined by ELISA with homologous NOMV as the antigen; (D) nasal wash IgA as determined by an ELISA with homologous NOMV as the antigen.
Louis, MO), IgM (Rockland, Inc., Gilbertsville, PA), or IgA (Organon TeknikaCappel, Durham, NC) standard captured by anti-rabbit IgG, IgM (Organon Teknika-Cappel, Durham, NC), or IgA (Sigma, St. Louis, MO) bound to the plate (31). Intranasal dye experiment. One rabbit was sham-immunized with 0.5 ml Coomassie blue R-250 (Bio-Rad Laboratories, Richmond, CA). This animal was euthanized 5 min later and necropsied to determine the disposition of the dye. Pyrogen testing. Pyrogen testing of NOMV was performed according to the Code of Federal Regulations, Title 21, section 610.13b (http://www.accessdata .fda.gov/scripts/cdrh/cfdocs/cfcfr/CFRSearch.cfm?fr⫽610.13), by contract with BioReliance, Rockville, MD. Modifications of the pyrogen test were made to allow for intranasal and intramuscular administration of meningococcal NOMV. Intranasal immunizations were given using a 0.5-ml volume administered to the nares of unanesthetized rabbits as described above, and intramuscular immunizations were given in a 0.5-ml volume in the hindquarter musculature of the animal. Western blotting. Western blots were performed following resolution of strain 9162 ⌬synX NOMV by sodium dodecyl sulfate-polyacrylamide gel electrophoresis on 10% acrylamide–bis-acrylamide gels using the method of Laemmli (18). Western blotting was done by a modification of the method of Burnette (5). Papers were incubated with primary antibody both in the presence and in the absence of 0.15% Empigen BB (Calbiochem, La Jolla, CA). Blocking buffer and antibody diluent contained 1% casein in place of bovine serum albumin. Sera were diluted 1:100 for incubation with the nitrocellulose strips. Alkaline phosphatase labeled anti-rabbit IgG was used as the second antibody, and the papers were developed using the Fast Red TR/Naphthol AS MX substrate (Sigma, St. Louis, MO).
RESULTS Immune response to two vaccine formulations. Systemic and mucosal immune responses of rabbits were measured at days 0, 28, and 56, after intranasal administration of NOMV vaccine prepared from wild-type 9162 or a 9162 ⌬synX mutant. Coomassie blue-stained polyacrylamide gels (data not shown) confirmed a similar banding pattern of proteins between these vaccines. Serum bactericidal activity was significantly higher (P ⫽ 0.002) at days 28, 42, 56, and 70 in rabbits receiving the sialic acid-deficient NOMV than in rabbits receiving NOMV from the encapsulated strain (Fig. 1A). No differences in levels of serum IgG or IgA (Fig. 1B and C) or nasal wash IgA (Fig. 1D) were observed, however. The mucosal antibody response was determined by measuring meningococcus-specific IgA, IgG, and IgM in lung lavage fluid and saliva at day 70 (Table 1). Substantial amounts of IgA and IgG had been induced and were detected in both the saliva and lung lavage fluid. No IgM was detected. Intranasal versus intramuscular immunization. NOMV vaccine was administered to rabbits via different routes and
SHOEMAKER ET AL.
TABLE 1. Geometric mean IgA and IgG levels in saliva and lung lavage fluid for immunized rabbitsa Geometric mean (SEM range) for:b IgA
Wild-type 9162 NOMV 9162 ⌬synX NOMV Saline
Saliva (n ⫽ 4)
Lung lavage fluid (n ⫽ 4)
Saliva (n ⫽ 4)
Lung lavage fluid (n ⫽ 4)
18 (11.9–27.4) 17.7 (12.8–24.6) 0.4 (0.27–0.59)
71.9 (48.2–107) 210 (140–316)c 0.79 (0.54–1.15)
24.5 (22.7–26.4) 21.0 (11.9–37.2) 0.14 (0.11–0.17)
217 (123–380) 327 (150–714)c 1.62 (1.23–2.14)
a Rabbits were immunized intranasally on days 0, 28, and 56 with 100 g of group B meningococcal NOMV vaccine prepared from either N. meningitidis 9162 wild-type NOMV or 9162 ⌬synX NOMV, and saliva and lung lavage fluids were taken on day 70. b ELISA units (optical density at 405 nm ⫻ reciprocal serum dilution) as determined by enzyme-linked immunosorbent assay. SEM, standard error of the mean. c n ⫽ 3 for this group of rabbits.
using different primary immunization schedules. Similar high levels of serum bactericidal activity were induced regardless of route or number of immunizations (Fig. 2A). Serum IgG appeared to peak following the first boost with intramuscular delivery of NOMV, whereas serum IgG levels in the intranasal groups continued to rise following the second boost (Fig. 2B). Lower levels of specific serum IgM were observed following intranasal immunization (Fig. 2C), in contrast with intramuscular immunization. Interestingly, intramuscular immunization and immunization by the intranasal route induced similar high levels of serum IgA (Fig. 2D). Intranasal immunization induced a local mucosal response, as evidenced by meningococcus-specific IgA (Fig. 2E), and meningococcus-specific IgG (Fig. 2F) in nasal washes. Intramuscular immunization induced IgG antibodies only in nasal washes (Fig. 2F). Saliva and lung lavage fluid collected at day 70 contained high levels of IgA only after i.n. immunizations (Table 2). Both i.n. and intramuscular (i.m.) immunization induced substantial amounts of IgG in lung lavage fluids and detectable but low levels of IgG in the saliva samples (Table 2). Virtually no IgM was detected in saliva or lung lavage fluid of any group (data not shown). Delivery of vaccine to nasopharyngeal region. Necropsy of a rabbit sham-immunized with dye suggested that intranasally delivered vaccine was presented to the nasopharyngeal region of the rabbit. Serial sections of the rostrum indicated good coverage of the nasal region with the dye (data not shown). Dye was found in the oropharynx, esophagus, and stomach, indicating that some of it had been swallowed. No dye was found in the lungs or trachea. Because the vaccine came in contact with gut-associated lymphoid tissue as well as the nasopharynx, it could not be determined whether the subsequent
immune response was due to stimulation of nasal-associated tissue. Therefore, rabbits were immunized intranasally, orally, or by gastric gavage. Serum bactericidal antibody responses to oral or gastric immunization were minimal compared to the responses found after three i.n. or i.m. immunizations (Table 3). Serum and nasal wash IgA levels also increased only after i.n. or i.m. immunizations (Table 3). Pyrogen testing. Pyrogen testing indicated that only very low levels (ⱕ0.1 g/rabbit) of meningococcal NOMV could be given intravenously (i.v.) without inducing fever (Table 4). Intramuscular delivery of 25 g NOMV also caused a spike in temperature. Large doses (400 g) of native NOMV, however, could be given to rabbits i.n. without inducing fever. Analysis of immune response by Western blotting. Analysis of the specificity of the antibody response of rabbits to intranasal vaccination with the NOMV was done by Western blotting using the NOMV vaccine as the antigen. The results of Western blots revealed that intranasal vaccination resulted in an antibody response against a broad range of outer membrane antigens, including both outer membrane proteins and LOS. Figure 3 shows the results for a representative rabbit that received intranasal vaccination with NOMV from the ⌬synX strain. Strips were incubated with diluted serum in the presence (lanes 7, 8, and 9) or absence (lanes 4, 5, and 6) of 0.15% Empigen BB to aid in renaturing denatured OMPs. Antibodies specific for both PorA and PorB bound better in the presence of the Empigen BB, whereas LOS, H8, and certain other proteins bound antibodies less efficiently when Empigen BB was present. There was no evidence that intranasal immunization induced antibodies of different specificity than intramuscular immunization. The specific antibody response to LOS was determined by
TABLE 2. Geometric mean salivary and lung lavage IgA and IgG levels in ELISA units at day 70 Geometric mean (SEM range) for: IgA
i.n. 3⫻a i.n. 5⫻b i.m. 3⫻c Saline 5⫻d
7.8 (5.7–10.8) 12.8 (9.1–18) 0.4 (0.35–0.48) 0.0
IgG Lung lavage fluid
35 (15–83) 107 (70–166) 6.3 (5.6–7.2) 1.4 (0.8–2.3)
Rabbits (n ⫽ 4) immunized intranasally with 100 g NOMV vaccine at days 0, 28, and 56. Rabbits (n ⫽ 4) immunized intranasally with 100 g NOMV vaccine at days 0, 1, 2, 28, and 56. Rabbits (n ⫽ 4) immunized intramuscularly with 25 g NOMV vaccine at days 0, 28, and 56. d Rabbits (n ⫽ 2) immunized with saline intranasally and intramuscularly at days 0, 1, 2, 28, and 56. a b c
Lung lavage fluid
6.4 (4.9–8.5) 9.1 (4.8–17.2) 3.3 (1.8–6.0) 0.0
87.2 (55–139) 299 (246–365) 172 (141–210) 0.8 (0.5–1.2)
VOL. 73, 2005
INTRANASAL MENINGOCOCCAL VACCINE IN RABBITS
FIG. 2. Antibody responses to meningococcal 9162 ⌬synX NOMV. Groups of four rabbits were given three i.n. (closed circles) or three i.m. (open circles) immunizations of 100 g and 25 g NOMV, respectively, at days 0, 28, and 56, or were immunized i.n. five times (closed triangles) with 100 g NOMV at days 0, 1, 2, 28, and 56. Negative control animals (n ⫽ 2) were sham-immunized i.n. and i.m. with saline (open triangles). (A) Serum bactericidal activity; (B) serum IgG by ELISA with homologous NOMV as the antigen; (C) serum IgM by ELISA with homologous NOMV as the antigen; (D) serum IgA by ELISA with homologous NOMV as the antigen; (E) nasal wash IgA by ELISA with homologous NOMV as the antigen; and (F) nasal wash IgG by ELISA with homologous NOMV as the antigen.
SHOEMAKER ET AL.
TABLE 3. Geometric mean serum bactericidal activity and serum and nasal wash IgA levels Result (SEM range) Group
Bactericidal titer (geometric mean)
i.m.a i.n.b Oralc Gastricd
256 (172–382) 22.6 (16–32) 2.0 (1.3–3.0) 1.7 (1.2–2.3)
Serum IgA (ELISA units)
Nasal wash IgA (ELISA units)
93.7 (80.5–109) 77.3 (47.6–125) 4.13 (2.77–6.15) 3.59 (3.07–4.20)
0.48 (0.34–0.69) 7.5 (3.1–18.3) 0.85 (0.52–1.38) 0.21 (0.04–1.06)
a Rabbits (n ⫽ 4) immunized intramuscularly with 25 g NOMV vaccine on days 0, 28 and 56. b Rabbits (n ⫽ 4) immunized intranasally with 100 g NOMV vaccine at days 0, 28, and 56. c Rabbits (n ⫽ 4) immunized orally with 100 g NOMV vaccine at days 0, 28, and 56. d Rabbits (n ⫽ 4) immunized gastrically with 100 g NOMV vaccine at days 0, 28, and 56.
ELISA using purified L3,7 and L8 LOS as antigens (Table 5). Interestingly, the IgG antibodies induced by the L3,7 LOS in the NOMV vaccines bound equally well or better to the L8 antigen than to the homologous L3,7 antigen. This suggests that the epitope recognized by these antibodies was in the core region of the LOS rather than the lacto-N-neotetraose portion of the L3,7 LOS. NOMV from the ⌬synX strain induced higher levels of antibody to L8 than the NOMV derived from the wild-type strain (P ⬍ 0.001). DISCUSSION In recent years, intranasal immunization has received considerable attention both at the basic science level and at the practical level as a potentially effective, noninvasive method of vaccination which can induce mucosal as well as systemic immunity. The most promising results have been obtained with live attenuated viral vaccines, particularly influenza virus vaccine (2, 7, 8, 12, 26). Recently, the first intranasal vaccine (FluMist; MedImmune, Inc.) was licensed in the United States (1). The first application of intranasal vaccination to meningococcal disease was in the early 1970s when Wenzel et al. conducted a study that examined the intranasal delivery of a single dose of meningococcal group C polysaccharide as a method to immunize against meningococcal meningitis (27). For the polysaccharide vaccine, the approach was not particularly effective and did not appear to have an advantage over parenteral vaccination. Intranasal vaccination of volunteers with a meningococcal group B vaccine was first undertaken by Haneberg et al. (16), who used the same deoxycholate-extracted outer mem-
TABLE 4. Results of pyrogenicity testing of group B meningococcal NOMV vaccines in rabbits Dose per rabbit (g)a
0.05 0.1 0.1 0.5 25 100 400
i.v. i.v. i.v. i.v. i.m. i.n. i.n.
Pass Pass Fail Fail Fail Pass Pass
Body weight of each rabbit was ⬃3 kg.
Temperature increases (°C)
0.0, 0.3, 0.6, 0.8, 1.4, 0.1, 0.0,
0.3, 0.3, 0.6, 1.2, 1.5, 0.1, 0.0,
0.4 0.4 0.7 1.7 1.5 0.2 0.0
FIG. 3. Western blot of representative sera from rabbit intranasally immunized with 9162 ⌬synX NOMV vaccine. NOMV from strain 9162 was run on the polyacrylamide gel and blotted onto nitrocellulose paper, which was then cut into strips; the individual strips were incubated with different antibody solutions. Lanes 7, 8, and 9 were incubated with the indicated rabbit sera in the presence of 0.15% Empigen BB added to the diluted serum solution. Lane 1, mixture of monoclonal antibodies (MAbs) specific for PorA (12-1-P1.3) and Opa (23-1P5.10); lane 2, MAb specific for PorB (2-1-P15); lane 3, mixture of MAbs specific for RmpM (9F5) and L3,7 LOS (9-2-L379); lanes 4 and 7, prevaccination rabbit serum; lanes 5 and 8, 6-week postvaccination rabbit serum; lanes 6 and 9, 10-week postvaccination rabbit serum.
brane vesicle vaccine that was demonstrated to be efficacious as a parenteral vaccine in a large efficacy trial (3). This vaccine was also found to be immunogenic and safe when given by the intranasal route. We became interested in intranasal vaccination for meningococcal group B disease because we were seeking methods for presentation of the OMPs and LOS in their natural membrane environment, where their surface exposure and conformation would be the same as on the whole viable organism. The NOMV, extracted from cells without exposure to detergents or denaturing solvents, or simply recovered from the liquid growth medium, approximates this ideal antigen presentation. Although the NOMV contain too much endotoxin for use as a parenteral vaccine, we found they could be safely administered as a vaccine by the intranasal route. We have now conducted two clinical studies with NOMV used as an intranasal vaccine. These studies have shown that NOMV
VOL. 73, 2005
INTRANASAL MENINGOCOCCAL VACCINE IN RABBITS
TABLE 5. Serum antibody responses to LOS of rabbits immunized intranasally with NOMVa Vaccine
Mean (log2 [g/ml]) ⫾ SEM for ELISA antibody vs:
Serum sample time
9162 wild-type NOMV
9162 ⌬synX NOMV
6 wk 10 wk
6 wk 10 wk
ND ⫾ 0.84 ⫾ 0.51
⫾ 0.75 ⫾ 0.52
ND ⫾ 0.52 ⫾ 0.18
⫾ 0.43 ⫾ 0.16
a ND, none detected; SEM, standard error of the mean; Pre, sample taken before immunization.
are safe and immunogenic in humans when administered by the intranasal route and induce both a mucosal and a serum antibody response (11, 17). In other studies, we have shown that meningococcal NOMV administered intranasally in mice induce high levels of serum bactericidal antibody as well as local mucosal antibody responses (21). Dalseg et al. (9) found similar results and showed that use of cholera toxin as a mucosal adjuvant boosted overall serum antibody response but failed to increase bactericidal antibody levels. We were not fully satisfied with the mouse as a model for intranasal vaccination of humans, since it was difficult not to introduce some of the vaccine into the lung, which from our point of view is not desirable and does not occur to a significant extent in normal human intranasal vaccination protocols. We found we could easily vaccinate unanesthetized rabbits intranasally and avoid the problem of vaccine entering the lung. In spite of the availability of fewer immunological reagents for the rabbit, we felt the rabbit could be a useful model for intranasal vaccination. In the present study, we demonstrated that intranasally administered meningococcal NOMV induced a good mucosal response in the rabbit, as indicated by the presence of IgA in nasal wash, mouth swabs, and lung lavage fluid, whereas virtually no mucosal IgA was detected with samples from animals immunized intramuscularly. Notably, intranasal delivery of NOMV induced high levels of serum bactericidal antibody, which is an important correlate of protection against meningococcal disease (13) and is consistent with the ability of intranasally administered NOMV to induce serum bactericidal antibodies in human volunteers (11, 17). The intranasal route of vaccine delivery offers several advantages over traditional intramuscular immunizations for candidate vaccines against group B meningococcal disease. First, intranasal immunization mimics the process of natural immunization by carriage of meningococci in the nasopharyngeal region and induces serum bactericidal antibodies as well as mucosal antibodies. Mucosal antibodies may provide a second line of defense against meningococcal disease, but the relative importance of mucosal antibodies in protection against meningococcal disease has not been elucidated. Secondly, native antigens such as NOMV, which have not been depleted of endotoxin, can be given safely by the intranasal route. Our meningococcal NOMV vaccine, which con-
tains about 20% to 25% LOS relative to protein, was nonpyrogenic in rabbits when given intranasally at a 4,000-fold higher dose than the highest nonpyrogenic intravenous dose. Parenterally administered vaccines against group B meningococcal disease have focused largely on delivering LOS and phospholipid-depleted outer membrane vesicles derived from the outbreak strain or purified recombinant outer membrane proteins (3, 10, 20, 23, 28). While several of these vaccines have induced adequate levels of serum bactericidal antibodies in human trials and show promise for wider use in pediatric and adult populations, use of a more native antigen, such as NOMV, may have significant advantages. Additional advantages may result from the presence of higher levels of LOS, which is a relatively conserved antigen, and more surface lipoproteins that are largely removed by deoxycholate extraction. The NOMV also provide a more selective presentation of surface-exposed OMP epitopes resulting from the undisturbed LOS-phospholipid membrane environment. Potential drawbacks to the intranasal route of NOMV vaccine delivery also exist. Higher doses of vaccine are required intranasally to induce levels of serum antibodies equivalent to those obtained with parenterally delivered vaccines. Unlike the rabbit studies reported here, the antibody response following intranasal vaccination of human volunteers with NOMV vaccine (11, 17) was, on the average, not as robust as the typical responses to parenteral vaccination with similar antigens. It appears that a mucosal adjuvant may be required for human subjects to obtain an adequately strong immune response to intranasal vaccination. Although rabbits had a more robust response to intranasal vaccination than human beings, some aspects of the antibody response were similar. In both systems, antibodies to a wide range of antigens were induced, as determined by Western blotting, and there was a significant response to the LOS component of the NOMV. Also, we found rabbits to be more consistent in their antibody responses than mice (21). We are not aware of any animal model that can accurately predict the human antibody response to vaccination with meningococcal group B vaccines, but studies of animals can be useful in comparing different candidate vaccines. Stephens et al. (24) have shown that N. meningitidis expressing (␣238)-linked polysialic acid did not adhere to human buccal cells or nasopharyngeal organ cultures as well as capsule-defective mutants. We examined whether the presence of capsule influenced the immunogenicity of intranasally delivered meningococcal NOMV by immunizing rabbits i.n. with NOMV made from an encapsulated parent or from a ⌬synX mutant of the same strain. Our data showed a stronger bactericidal antibody response with the ⌬synX mutant NOMV, suggesting that sialic acid on the surface of the vesicles may diminish interaction of the NOMV with the mucosal surface or block epitopes capable of inducing bactericidal antibodies. The ⌬synX NOMV, however, did not induce a higher overall serum antibody response as measured by ELISA or a higher mucosal antibody response. The reason for the higher bactericidal antibody response to the ⌬synX NOMV is unclear. Stronger bactericidal activity may have resulted from induction of higher avidity antibodies or of antibodies with different epitope specificities. We believe the intranasal rabbit model described in this study provides a useful tool for the study of nasal vaccines. Intranasal administration of vaccine was shown to result in effective coating of the intranasal region, and excess liquid was swallowed and was
SHOEMAKER ET AL.
not observed in the trachea or lungs, as frequently occurs with mice. When vaccine enters the lung, the model is not truly an intranasal model. Rabbits responded to intranasal vaccination with both a serum and a mucosal antibody response. Nasal washes for measurement of mucosal immune response were easily obtained from rabbits using noninvasive procedures. The rabbit was also useful for studying pyrogenicity of vaccines given by the intranasal route. Although there are also some disadvantages to the model, such as reduced availability of immunological reagents, we believe the rabbit can be a useful animal for evaluation of intranasal vaccines prior to human studies. ACKNOWLEDGMENTS We thank A. Louis Bourgeois for supplying the anti-rabbit IgA monoclonal antibody. We thank J. McLeod Griffiss for the gift of monoclonal antibody 9F5. This work was supported by the United States Army Medical Research and Materiel Command through the Military Infectious Disease Research Program office. The opinions and assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the Department of the Army or the Department of Defense. REFERENCES 1. Anonymous. 2003. Influenza virus vaccine live intranasal—MedImmune vaccines: CAIV-T, influenza vaccine live intranasal. Drugs R D 4:312–319. 2. Belshe, R. B., W. C. Gruber, P. M. Mendelman, I. Cho, K. Reisinger, S. L. Block, J. Wittes, D. Iacuzio, P. Piedra, J. Treanor, J. King, K. Kotloff, D. I. Bernstein, F. G. Hayden, K. Zangwill, L. Yan, and M. Wolff. 2000. Efficacy of vaccination with live attenuated, cold-adapted, trivalent, intranasal influenza virus vaccine against a variant (A/Sydney) not contained in the vaccine. J. Pediatr. 136:168–175. 3. 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. 4. Boslego, J., J. Garcia, C. Cruz, W. Zollinger, B. Brandt, S. Ruiz, M. Martinez, J. Arthur, P. Underwood, W. Silva, E. Moran, W. Hankins, J. Gilly, J. Mays, and the Chilean National Committee for Meningococcal Disease. 1995. Efficacy, safety, and immunogenicity of a meningococcal group B (15:P1.3) outer membrane protein vaccine in Iquique, Chile. Vaccine 13: 821–829. 5. Burnette, W. N. 1981. “Western blotting”: electrophoretic transfer of proteins from sodium dodecyl sulfate-polyacrylamide gels to unmodified nitrocellulose and radiographic detection with antibody and radioiodinated protein A. Anal. Biochem. 112:195–203. 6. Caputo, G. L., G. Baldwin, G. Alpert, J. Parsonnet, Z. A. Gillis, G. Siber, and G. Fleisher. 1992. Effect of meningococcal endotoxin in a rabbit model of shock. Circ. Shock 36:104–112. 7. Clements, M. L., and B. R. Murphy. 1986. Development and persistence of local and systemic antibody responses in adults given live attenuated or inactivated influenza A virus vaccine. J. Clin. Microbiol. 23:66–72. 7a.Cole, Carole. 1989. Monoclonal antibody anti-rabbit IgA, NRbA. Monoclonal Antibody News 7:23. 8. D’Allessandri, D., G. Plotkin, R. M. Kluge, M. K. Wittner, E. N. Fox, A. Dorfman, and R. H. Waldman. 1978. Protective studies with group A streptococcal M protein vaccine. III. Challenge of volunteers after systemic or intranasal immunization with type 3 or type 12 group A Streptococcus. J. Infect. Dis. 138:712–718. 9. Dalseg, R., E. Wedege, J. Holst, I. L. Haugen, E. A. Hoiby, and B. Haneberg. 1999. Outer membrane vesicles from group B meningococci are strongly immunogenic when given intranasally to mice. Vaccine 17:2336–2345. 10. de Moraes, J. C., B. A. Perkins, M. C. C. Camargo, N. T. R. Hidalgo, H. A. Barbosa, C. T. Sacchi, I. M. L. Gral, V. L. Gattas, H. De G. Vasconcelos, B. D. Plikaytis, J. D. Wenger, and C. V. Broome. 1992. Protective efficacy of a serogroup B meningococcal vaccine in Sao Paulo, Brazil. Lancet 340:1074–1078. 11. Drabick, J. J., B. L. Brandt, E. E. Moran, N. B. Saunders, D. R. Shoemaker,
Editor: J. N. Weiser
INFECT. IMMUN. and W. D. Zollinger. 1999. Safety and immunogenicity testing of an intranasal group B meningococcal native outer membrane vesicle vaccine in healthy volunteers. Vaccine 18:160–172. 12. Ganguly, R., P. L. Ogra, S. Regas, and R. H. Waldman. 1973. Rubella immunization of volunteers via the respiratory tract. Infect. Immun. 8:497– 502. 13. Goldschneider, I., E. C. Gotschlich, and M. S. Artenstein. 1969. Human immunity to the meningococcus. I. The role of humoral antibodies. J. Exp. Med. 129:1307–1326. 14. Goldschneider, I., E. C. Gotschlich, and M. S. Artenstein. 1969. Human immunity to the meningococcus. II. Development of natural immunity. J. Exp. Med. 129:1327–1348. 15. Hammerschmidt, S., R. Hilse, J. P. van Putten, R. Gerardy-Schahn, A. Unkmeir, and M. Frosch. 1996. Modulation of cell surface sialic acid expression in Neisseria meningitidis via a transposable genetic element. EMBO J. 15:192–198. 16. Haneberg, B., R. Dalseg, E. Wedege, E. A. Hoiby, I. L. Haugen, F. Oftung, S. R. Andersen, L. M. Naess, A. Aase, T. E. Michaelsen, and J. Holst. 1998. Intranasal administration of a meningococcal outer membrane vesicle vaccine induces persistent local mucosal antibodies and serum antibodies with strong bactericidal activity in humans. Infect. Immun. 66:1334–1341. 17. Katial, R. K., B. L. Brandt, E. E. Moran, S. Marks, V. Agnello, and W. D. Zollinger. 2002. Immunogenicity and safety testing of a group B intranasal meningococcal native outer membrane vesicle vaccine. Infect. Immun. 70: 702–707. 18. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680–685. 19. Moran, E. E., B. L. Brandt, and W. D. Zollinger. 1994. Expression of the L8 lipopolysaccharide determinant increases the sensitivity of Neisseria meningitidis to serum bactericidal activity. Infect. Immun. 62:5290–5295. 19a. National Research Council. 1996. Guide for the care and use of laboratory animals. National Academy Press, Washington, D.C. 20. Peeters, C. C., H. C. Rumke, L. C. Sundermann, E. M. Rouppe van der Voort, J. Meulenbelt, M. Schuller, A. J. Kuipers, P. van der Ley, and J. T. Poolman. 1996. Phase I clinical trial with a hexavalent PorA containing meningococcal outer membrane vesicle vaccine. Vaccine 14:1009–1015. 21. Saunders, N. B., D. R. Shoemaker, B. L. Brandt, E. E. Moran, T. Larsen, and W. D. Zollinger. 1999. Immunogenicity of intranasally administered meningococcal native outer membrane vesicles in mice. Infect. Immun. 67:113–119. 22. Saunders, N. B., D. R. Shoemaker, B. L. Brandt, and W. D. Zollinger. 1997. Confirmation of suspicious cases of meningococcal meningitis by PCR and enzyme-linked immunosorbent assay. J. Clin. Microbiol. 35:3215–3219. 23. Sierra, G. V., H. C. Campa, N. M. Varcacel, I. L. Garcia, P. L. Izquierdo, P. F. Sotolongo, G. V. Casanueva, C. O. Rico, C. R. Rodriguez, and M. H. Terry. 1991. Vaccine against group B Neisseria meningitidis: protection trial and mass vaccination results in Cuba. NIPH Ann. 14:195–207. 24. Stephens, D. S., P. A. Spellman, and J. S. Swartley. 1993. Effect of the (␣238)-linked polysialic acid capsule on adherence of Neisseria meningitidis to human mucosal cells. J. Infect. Dis. 167:475–479. 25. Swartley, J. S., and D. S. Stephens. 1994. Identification of a genetic locus involved in the biosynthesis of N-acetyl-D-mannosamine, a precursor of the (␣238)-linked polysialic acid capsule of serogroup B Neisseria meningitidis. J. Bacteriol. 176:1530–1534. 26. Waldman, R. H., S. H. Wood, E. J. Torres, and P. A. Small, Jr. 1970. Influenza antibody response following aerosol administration of inactivated virus. Am. J. Epidemiol. 91:575–584. 27. Wenzel, R. P., J. R. Mitzel, J. A. Davies, E. A. Edwards, C. Berling, D. P. McCormick, and W. E. Beam, Jr. 1973. Antigenicity of a polysaccharide vaccine from Neisseria meningitidis administered intranasally. J. Infect. Dis. 128:31–40. 28. Zollinger, W. D. 1997. New and improved vaccines against meningococcal disease, p. 469–488. In M. M. Levine, et al. (ed.), New generation vaccines. Marcel Dekker, Inc., New York, N.Y. 29. Zollinger, W. D., D. L. Kasper, B. J. Veltri, and M. S. Artenstein. 1972. Isolation and characterization of a native cell wall complex from Neisseria meningitidis. Infect. Immun. 6:835–851. 30. Zollinger, W. D., D. R. Shoemaker, N. B. Saunders, and B. L. Brandt. May 2003. Vaccine against gram negative bacteria. U.S. patent 6,558,677 B2. 31. Zollinger, W. D., and J. W. Boslego. 1981. A general approach to standardization of the solid-phase radioimmunoassay for quantitation of class-specific antibodies. J. Immunol. Methods 46:129–140. 32. Zollinger, W. D., R. E. Mandrell, J. M. Griffiss, P. Altieri, and S. Berman. 1979. Complex of meningococcal group B polysaccharide and type 2 outer membrane protein immunogenic in man. J. Clin. Investig. 63:836–848.