Comparison of Immune Responses to Gonococcal PorB Delivered as Outer Membrane Vesicles, Recombinant Protein, or Venezuelan Equine Encephalit
http://www.100md.com
感染与免疫杂志 2005年第11期
Department of Medicine, Division of Infectious Diseases Carolina Vaccine Institute,Department of Microbiology and Immunology, School of Medicine, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina
ABSTRACT
Porin (PorB) is a major outer membrane protein produced by all Neisseria gonorrhoeae strains and has been a focus of intense interest as a vaccine candidate. In this study, the immunogenicity of PorB in mice was investigated after several immunization regimens. Outer membrane vesicles (OMV), recombinant renatured PorB (rrPorB), and PorB-expressing Venezuelan equine encephalitis (VEE) virus replicon particles (PorB VRP) were delivered intranasally (i.n.) or subcutaneously (s.c.) into the dorsal area or the hind footpad in three-dose schedules; the PorB VRP-immunized mice were given a single additional booster dose of rrPorB in Ribi adjuvant. Different delivery systems and administration routes induced different immune responses. Mice immunized s.c. with rrPorB in Ribi had the highest levels of PorB-specific serum immunoglobulin G (IgG) by enzyme-linked immunosorbent assay. Surprisingly, there was an apparent Th1 bias, based on IgG1/IgG2a ratios, after immunization with rrPorB in Ribi in the footpad while the same vaccine given in the dorsal area gave a strongly Th2-biased response. PorB VRP-immunized mice produced a consistent Th1 response with a high gamma interferon response in stimulated splenic lymphocytes and very low IgG1/IgG2a ratios. Immunization by OMV delivered i.n. was the only regimen that resulted in a serum bactericidal response, and it generated an excellent mucosal IgA response. Serum from mice immunized with rrPorB preferentially recognized the surface of whole gonococci expressing a homologous PorB, whereas serum from PorB VRP-immunized mice had relatively low whole-cell binding activity but recognized both heterologous and homologous PorB equally. The data resulting from this direct comparison suggested that important aspects of the immune response can be manipulated by altering the form of the antigen and its delivery. This information coupled with an understanding of protective antigonococcal immune responses will enable the design of the optimal vaccine for N. gonorrhoeae.
INTRODUCTION
Infection by Neisseria gonorrhoeae (the gonococcus [GC]) remains a serious problem in the United States and the world (10). Although GC infection has declined in recent years in the United States, with total reported infections of "only" 300,000 annually, prevalence may be as high as 5% in certain populations in inner cities and the rural Southeast, especially in African-Americans and socioeconomically deprived groups. Women suffer the bulk of the complications in the form of salpingitis, infertility, and ectopic pregnancy. Fetal death is inevitable in the case of ectopic pregnancy. Calculations of attributable risk show that GC is one of the major cofactors for human immunodeficiency virus (HIV) transmission (19), increasing risks of HIV transmission about threefold. Emergence of antibiotic-resistant GC including GC resistant to ciprofloxacin is threatening standard treatment modalities, just as earlier emergence of penicillin and tetracycline resistance relegated those classic therapies to history. Unless new oral therapies are developed, we will face a challenge that we have not experienced for decades. A safe and effective vaccine would be tremendously useful for all these reasons.
A vaccine for GC has seemed a difficult task based solely on natural history, which shows that repeat infections are common, including repeat infection by the same strain (17, 20, 25). However, in earlier days, before the advent of effective therapy, GC infections eventually underwent spontaneous resolution (38), with negative cultures and lack of risk to sexual partners (24). This showed that prolonged infection, although often associated with local complications, eventually resulted in immune clearance. Thus, an effective gonorrhea vaccine might be possible if the right immune response were directed at the right antigens.
Unlike the meningococcus, GC does not make capsules, and the search for a GC vaccine is aimed at outer membrane proteins rather than a capsule. Experience from the meningococcus group B outer membrane vesicle (OMV) vaccines showed that protection was correlated strongly with responses to porin protein antigens in the OMVs (13). PorB, which is expressed on all gonococci, is the major integral outer membrane protein on the surface of gonococci. It is essential to GC viability (9) and plays important roles in the pathogenesis of GC (3, 45). PorB is expressed constitutively and is relatively stable antigenically without undergoing phase or high-frequency antigenic variation (36, 45). Certain antibodies against PorB are known to be bactericidal (28), opsonic (23), or protective against toxicity in tissue culture (47, 48). Two reports have claimed PorB serovar-specific protection against reinfection in humans. In the first, female commercial sex workers in Nairobi, Kenya, were less likely to be reinfected by the same serovar, even though most were immunosuppressed by their HIV infection (40). The second, smaller study found no reinfections of the fallopian tubes (salpingitis) by GC with the same serovar as that for the index infection (7). These data suggest that developing an immune attack on PorB might be an effective vaccine strategy. However, GCs have developed several mechanisms to protect against host immune attack on PorB. These include the blocking antigen Rmp, which stimulates production of non-complement-fixing antibodies that block bactericidal activity of anti-PorB sera (14), and sialylation of terminal sugars on lipooligosaccharide, forming an "umbrella" that partially limits access of antibodies to PorB (15) and interferes with complement activation (15). A partially purified PorB vaccine isolated from GC apparently was not effective in early studies of experimentally inoculated human male volunteers (P. A. Rice, S. Gulait, and S. Ram, Abstr. 10th Int. Pathog. Neisseria Conf., abstr. 3, 1996). Subsequent analysis of this experiment suggested that failure was due to a vigorous immune response to contaminating blocking antigen Rmp (4). Thus, a successful PorB vaccine presumably would need to stimulate high titers of antibodies without eliciting anti-Rmp antibodies. Use of recombinant PorB from Escherichia coli avoids the problem of the immunogenicity of small amounts of contaminating Rmp when PorB is prepared from wild-type GC. Similarly, use of a Venezuelan equine encephalitis virus (VEE)-based viral vector, VEE replicon particle (VRP), expressing PorB completely avoids the problems of contamination by Rmp in a vaccine.
In addition to the choice of target antigen, the success of a vaccine also depends on the type, site, strength, and specificity of the immune responses that it induces. The route of administration of vaccines contributes to the effectiveness of the local and systemic immune responses. This prompted us to investigate the immunogenicity of PorB administered by different routes and in different forms. We utilized renatured recombinant PorB (rrPorB), as well as a VEE vaccine vector system expressing PorB. The VEE system is composed of an RNA replicon and a bipartite helper system for packaging the replicon into propagation-deficient VRPs (42). It is an efficient expression system for in vivo delivery of heterologous antigens to dendritic cells (34) and expresses very high levels of immunogen in cells in the draining lymph node. It can induce both systemic and mucosal immunity (12, 22), which presumably is important to host defense against pathogens such as N. gonorrhoeae that enter across a mucosal surface, and both humoral and cellular responses.
An improved mouse model of genital gonococcal infection has been developed recently (26). Intranasal (i.n.) immunization with GC OMVs reduced the duration of vaginal colonization of homologous strain MS11 in the female mouse genital tract (39). In this study, we tested the immunogenicity of rrPorB, OMV, and PorB expressed from a VEE-based vector (PorB VRP) in order to inform future vaccine studies of gonorrhea in the mouse challenge model.
MATERIALS AND METHODS
Strain, recombinant antigen, and rabbit anti-PorB serum. N. gonorrhoeae strain FA1090 (37) variant A23 was used throughout these experiments and was obtained from Marcia Hobbs (11). Passage of FA1090 and its derivatives was kept to a minimum to avoid phase or antigenic variation. Gonococcal strains were routinely maintained on GC base medium (Difco) supplemented with Kellogg's supplement I (29). All relevant strains and plasmids are shown in Table 1.
An individual isolate of pUNCH682 (porB) in BL21(DE3) (Invitrogen) expressing a high level of antigen was expanded to produce recombinant protein by inducing a culture with an optical density of 0.4 to 0.5 with 2 mM isopropyl--D-thiogalactopyranoside for 30 min and arresting normal cellular RNA synthesis with rifampin for an additional 2 h. We purified the hexa-His-tagged fusion rrPorB on Ni-nitrilotriacetic acid (NTA) agarose (QIAGEN) under denaturing conditions. Cell pellets were lysed for 10 min in 6 M guanidinium hydrochloride buffered with 100 mM NaH2PO4 and 10 mM Tris to pH 8.0. DNA was sheared by 10 passages through an 18- to 22-gauge needle; cellular debris was removed by pelleting at 10,000 x g for 30 min. His-tagged rrPorB was bound to Ni-NTA agarose in batch format for 30 min; the resin was washed with 8 M urea buffered as above to pH 8 in column format; the resin was washed again in 8 M urea buffered as above to pH 6.7. His-tagged rrPorB was eluted in 8 M urea buffered as above to pH 5.0. An attempt to renature the recombinant protein was made by slow dilution of the denaturing urea in the presence of low concentrations of detergents. The His-tagged rrPorB had assumed some native conformation as demonstrated by binding of conformation-dependent monoclonal antibodies (data not shown). The His-tagged rrPorB was used to immunize Elite rabbits at Covance Research Products. Rabbits were immunized with 250 μg of recombinant protein in Freund's complete adjuvant and boosted with 125 μg of recombinant protein in Freund's incomplete adjuvant three times at 3-week intervals. Rabbit sera were collected at 7, 10, and 13 weeks.
The renatured and purified rrPorB used in subsequent mouse immunization studies and immunoassays was a kind gift of Yuri Matsuka and Elizabeth Anderson of Wyeth Research. This material was prepared in a manner similar to a published method (35). In brief, rrPorB was expressed in Escherichia coli and refolded in vitro in the presence of a zwitterionic detergent. Its proper folding and subunit organization were confirmed by comparison with the native counterpart.
Preparation of OMVs. OMVs were prepared from the FA1090 rmp mutant strain FA6902 (rmp) as described elsewhere (5) with the following modifications. GCs were grown under iron-stressed conditions to fully express all iron-repressed proteins. The culture was collected by centrifugation at 10,000 x g (4°C) for 15 min and resuspended in 15 ml cold lithium acetate buffer (200 mM LiCl, 100 mM C2H3O2Li, pH 6.0). The bacterial suspension was passed through a 20-1/2-gauge needle for 20 times to shear off the outer membrane and centrifuged for 2 min to pellet whole bacteria, and the supernatant was ultracentrifuged at 100,000 x g (15°C) for 2 h. The pellets were resuspended in a small amount of lithium acetate buffer and frozen at –20°C until needed.
Production of PorB VRPs with and without a eukaryotic secretion signal (tPA). Gonococcal porB was amplified by PCR from FA1090 using primers GCV-3 and GCV-4 for porB. All oligonucleotide primers are shown in Table 2. The PCR product was then cloned using the TA cloning system (Invitrogen) to create pUNCH678 (porB). The eukaryotic secretion signal from tissue plasminogen activator (tPA) was placed upstream of the mature coding sequences of porB by cloning the insert from pUNCH678 (porB) downstream of the tPA signal sequence in pUNCH689 (tPA). The plasmid pUNCH689 is a derivative of VJns-tPA (33) with the BglII-AccI adapter inserted into the BglII site, creating a new unique AccI site which is compatible with the ClaI sites flanking the inserts of pUNCH678 (porB). The tPA-containing clone was designated pUNCH693 (tPA-porB), and this clone was used as the template for the corresponding replicon plasmid discussed below.
The replicon plasmids were assembled essentially as previously described (2) by a process of overlapping PCR amplification to create an authentic fusion between the 26S VEE viral structural gene promoter and the synthetic start codon of the bacterial gene. The insert from pUNCH678 (porB) was amplified by PCR using primers GCV-7/GCV-8 to incorporate an upstream overlap with the 26S VEE promoter and a downstream AscI site. In a similar fashion the insert from pUNCH693 (tPA-porB) was amplified with GCV-12/GCV-8. The 26S promoter was amplified with primers GCV-9/GCV-10 and combined with each of the two previous PCR products in an overlapping PCR fusion using GCV-9 as the upstream amplimer and GCV-8 as the downstream amplimer. These two PCR fusion products were digested with ApaI (cuts the 26S promoter) and AscI (cuts the downstream end of each of the PCR products) and ligated into similarly cut replicon vector pVR21 (2), to give rise to pUNCH697 (tPA-porB) and pUNCH699 (porB). The inserts of the two final constructs were sequenced to confirm fidelity, and any errors were corrected by screening and/or subcloning (Table 2 shows amplimer sequences). RNase-free NotI-linearized replicon plasmids were transcribed using the mMESSAGE mMACHINE T7 kit (Ambion), and the capped runoff transcripts were electroporated into baby hamster kidney (BHK) cells using the conditions of Liljestrom and Garoff (32) (0.4-cm gap cuvette; three pulses, 0.85 kV, 25 μF). VRPs were generated by coelectroporating the transcript containing porB or tPA-porB with two helper mRNAs encoding the VEE capsid and glycoproteins. Glycoprotein from the V3000 molecular clone (21) was used in these constructs. VRPs were harvested between 20 and 27 h after transfection and purified from cell culture supernatants by pelleting through a 20% sucrose cushion. The pelleted VRPs were resuspended in phosphate-buffered saline (PBS) and stored at –80°C. Efficiency of transfection was measured by indirect immunofluorescence. An aliquot of each electroporation was plated onto a chamber slide and incubated along with the rest of the electroporation for 18 to 24 h. The chamber slides were rinsed with PBS, fixed for 10 min in ice-cold acetone, dried, and stored at –20°C. Slides were later rehydrated in PBS with 0.2% normal goat serum (NGS) and then blocked in 5% NGS. Antigen was detected using the rabbit antiserum (1:1,000 anti-His-PorB), biotin-SP-AffiniPure goat anti-rabbit immunoglobulin G (IgG) (heavy plus light) (1:100, catalog no. 111-065-003), and cyanine Cy2 streptavidin (1:100, catalog no. 016-220-084) (Jackson ImmunoResearch Laboratories) sequentially, each diluted in PBS-5% NGS. Each VRP preparation was shown to be free of propagation-competent virus by a sensitive cytopathic effect assay following serial passage on BHK cells. The titer of VRP in each preparation was determined by indirect immunofluorescence on serial dilutions as described above.
Western blot analysis of replicon-based expression. Efficiency of expression was measured by subjecting a crude lysate of the transfected BHK cells to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and Western blotting (6, 30). The crude lysate consisted of rinsed BHK cells solubilized in 1% SDS for 5 min at 37°C and diluted to equal total protein concentration determined by bicinchoninic acid protein assay (Pierce, Rockford, IL). SDS-PAGE samples were routinely boiled for 2 min under reducing conditions (5% - mercaptoethanol) prior to loading. Antigen was detected in standard immunoblot format using 1:8,000 rabbit anti-His-PorB; 1:5,000 alkaline phosphatase-conjugated sheep anti-rabbit IgG F(ab')2 Frag (Sigma, St. Louis, MO), and LumiPhos WB (Pierce, Rockford, IL).
Mouse inoculation and sample collection. Four- to five-week-old female BALB/c mice were purchased from Harlan Laboratory Animals and housed in the University of North Carolina animal facility according to federal and institutional animal care and use committee guidelines. rrPorB was formulated with Ribi R-700 (Corixa, Hamilton, MT), which consists of monophosphoryl lipid A and synthetic trehalose dicorynomycolate in a squalene-Tween 80 emulsion. This adjuvant was prepared at a 2x concentration in PBS according to manufacturer's instructions. Mice in groups of six were immunized as described in Table 3. All the mice were anesthetized with methoxyflurane (Medical Developments, Australia) before inoculation. PorB VRP groups were immunized with 20 μl PBS containing a total of 105 to 106 PorB VRPs or 105 tPA PorB VRPs in the left hind footpad (FP). No adverse effects of VRP inoculation were observed. Recombinant proteins were inoculated by two delivery routes: subcutaneously (s.c.) via FP with 10 μg rrPorB in 20 μl of R-700 Ribi adjuvant system (Corixa) or s.c. via dorsal area with 5 μg rrPorB in 50 μl of R-700 Ribi at each of two sites. OMV-immunized mice received 10 μl PBS containing 10 μg of OMV in each nostril. Mice were boosted with immunizations identical to their prime immunization at 3, 6, and 9 weeks postprime. For some of the PorB VRP animals, a final boost with a total of 10 μg of rrPorB in Ribi adjuvant was given s.c. in the dorsal area at 9 weeks. A single dose of 10 μg rrPorB in Ribi adjuvant ("1° rrPorB") was given in the dorsal area at 9 weeks to an unimmunized control group to compare the results to those obtained after the booster dose of the same immunogen given to the PorB VRP group. A mock-immunized group received 50 μl of R-700 Ribi adjuvant s.c. at each of two dorsal sites without added protein at 0, 3, 6, and 9 weeks.
Venous tail blood was collected 2 weeks after the third and fourth immunizations. Blood was collected in Microtainer serum separator tubes (Becton Dickinson), allowed to clot for 30 min at room temperature, and stored at 4°C overnight. The following morning tubes were centrifuged according to manufacturer's specifications. Serum was filtered through a sterile 0.45-μm cellulose acetate centrifuge tube filter (Costar). Vaginal mucosal samples were collected at 2 weeks after the fourth immunization by washing the mouse vaginal cavity several times with 50 μl PBS, followed by addition of 10 μl 5x stabilization buffer (protease inhibitor cocktail P8340 [Sigma, St. Louis, MO] in PBS with 0.1% sodium azide). Samples were stored at –20°C until needed.
Serum and vaginal mucosal antibody analysis. The quantity of antibody present in immunized mouse serum and washed from mucosal surfaces was measured by indirect enzyme-linked immunosorbent assay (ELISA) as previously described by Zhu et al. (49). The quantity of PorB-specific antibody bound to the plates was determined by comparing the optical density of the PorB-coated wells to that obtained by coating the wells with goat anti-mouse IgG-Fc-affinity-purified coating antibody at 1:100 (or IgA-affinity-purified coating antibody at 1:100 for IgA ELISA), comparing them to a standard curve of a mouse reference serum with known quantities of immunoglobulins (IgG, IgA, IgG1, or IgG2a, all purchased from Bethyl Laboratories), and developing them in parallel with the PorB-specific wells.
Cytokine assays. Splenocytes were collected from immunized mice at 2 weeks after the final immunization and tested for gamma interferon (IFN-) and interleukin-4 (IL-4) secretion by ELISPOT assay. Viable splenocytes were isolated with Lympolyte-M (Accurate, Westbury, NY). Thirty-two synthetic PorB peptides of 20 amino acids in length with a 10-amino-acid overlap spanning PorB were synthesized by Mimotopes (Victoria, Australia). The individual peptides were dissolved in dimethyl sulfoxide and mixed to make a master pool of equal concentrations. Aliquots of the master pool were further diluted with AIM V (Invitrogen) immediately before addition to the splenocytes to reach a final concentration of 1.5 μg/ml in the wells. For each ELISPOT assay, a 96-well filtration plate (Millipore, Bedford, MA) was coated overnight with either antibody to mouse IFN- (Mabtech, Nacka, Sweden; 5 μg/ml) or antibody to mouse IL-4 (Mabtech; 2.5 μg/ml) at 4°C. The plates were washed with PBS and blocked with RPMI medium containing 10% fetal calf serum. After being washed with PBS, each well was loaded with 6 x 105 cells. The plates were incubated at 37°C, 5% CO2, for 20 (IFN-) or 40 (IL-4) hours. Production of cytokines was detected by addition of biotinylated anti-mouse IFN- or anti-mouse IL-4 antibodies (Mabtech). After the wells were washed, peroxidase-labeled streptavidin was added. The spots were developed with the Sigma AEC Chromogen kit. Plates were sent to Zellnet Consulting Inc. (Fort Lee, NJ), and results were reported as spot-forming cells per million splenocytes.
Whole-cell binding assay. The relative amount of antibodies that recognized the surface of the bacteria was measured by indirect whole-cell radioimmunoassay. This assay consisted of exposing dilutions of serum to whole bacteria differing only in the antigen of interest and detecting bound antibody using 125I-labeled goat anti-mouse immunoglobulin (8, 27). Immune sera were diluted 1:100 in 50 μl Bacto GC medium base broth (GCB) and then mixed with 150 μl GCB containing 2 x 108/ml CFU of gonococci (final dilution of serum, 1:400) for 15 min at room temperature. The gonococci were collected by filtration in a multiscreen plate (Millipore, Bedford, MA). Filters were washed five times with PBS-azide (0.1%) buffer before addition of 106 cpm of 125I-labeled secondary antibody (goat anti-mouse IgG, IgA, and IgM; ICN Biomedicals, Inc., Aurora, OH) in 100 μl to each well. After incubation at room temperature for 15 min, the filters were washed five times with PBS-azide (0.1%) and air dried, and the filter bottom was punched out into borosilicate glass tubes. Antibody binding was assessed by counting gonococcus-associated radioactivity in a 1271 RIAGAMMA counter (LKB Wallac). Controls included either no antibody or monoclonal anti-PorB antibodies that bind to only one strain. The antibodies recognizing the surface-exposed epitopes of PorB were measured by comparing binding to FA1090-A23 (porB serovar PIB-3) and an isogenic derivative, FA7224, containing FA19 porB (serovar PIA-1). In FA7224 the porB gene of FA1090 was replaced with the PIA-1 serovar porB gene of FA19. The complete replacement of the porB gene was confirmed by PCR and sequencing.
Bactericidal assay. Bactericidal activity against FA1090 and FA7224 was measured as previously described by Elkins (15), with fresh pooled normal human serum (NHS) as a complement source. Briefly, immunized mouse sera were heat inactivated for 30 min at 56°C before being added to 104 CFU/ml gonococci in 100 μl GCB (final dilution, 1:50) and incubated at 37°C for 15 min. NHS was added to a final concentration of 10%, and incubation was continued at 37°C for 30 additional minutes. Samples were plated, and viable counts were determined. Rabbit serum was used at a 1:100 dilution as a positive control. Other controls included bacteria incubated with 10% NHS alone and 10% NHS inactivated at 56°C for 30 min. Significant killing was defined as a greater than 50% decrease in viable count compared to the PBS-immunized group.
Statistical analysis. The difference between immunization groups was analyzed by analysis of variance on ranks. Significant results of analysis of variance were further analyzed by the Mann-Whitney rank sum test for two-group comparison. Statistical analyses were performed using SigmaStat 3.0 (Systat Software Inc., Richmond, Calif.). Differences were considered significant at P < 0.05.
RESULTS
Expression of gonococcal antigens in BHK cells. Because of the risk that the bacterial codon usage would be less than optimal in a eukaryotic expression system, the replicon expression plasmids were tested in their earliest form, that is, RNA transfected by electroporation into BHK cells. We assessed antigen expression by both indirect immunofluorescence and Western blotting of whole-cell lysates. Transfection frequencies, as measured by percentage of BHK cells expressing PorB by immunofluorescence, were high (>80%) and reasonably consistent between expression plasmids (data not shown). Western blot assays of whole-cell lysates from these electroporations (Fig. 1) showed that the replicon expression plasmids were capable of expressing full-length antigen (35 kDa). Earlier replicon expression plasmids containing porB fused to the viral promoter with an intervening short insert of irrelevant sequence showed very poor expression and poor immune responses compared to these constructs (data not shown). This indicated the importance of direct fusion of the viral sequence just upstream of the start codon in achieving maximum expression of these bacterial gene products. Despite the inherent difficulties in quantifying the levels of expression by Western blotting, the tPA-containing replicons consistently expressed more gonococcal antigen than identical replicons lacking this eukaryotic signal sequence. Faster-migrating immunoreactive bands could represent breakdown products of the primary expression product or alternative translation products that remain immunoreactive and therefore could still serve as antigen.
Development of anti-PorB serum IgG. All delivery systems and routes induced PorB-specific serum IgG responses after immunization (Fig. 2). Mice immunized s.c. with rrPorB in Ribi either dorsally or in the FP exhibited the highest serum specific IgG levels, which ranged from 1 mg/ml to 2 mg/ml. Mice immunized intranasally with OMV generated a similar response, with 0.6 mg/ml of PorB-specific serum IgG.
All mice immunized with PorB VRPs developed specific IgG after inoculation. At 14 days after the third immunization, PorB VRP-immunized mice produced 20 μg/ml specific IgG antibody, which was 200-fold greater than background levels in the mock group but only about 1% of that induced by rrPorB in Ribi immunization. Inclusion of the tPA signal sequence in the PorB VRP vaccine constructs improved the amount of specific IgG produced by PorB VRP-immunized mice by about fivefold compared to the non-tPA PorB VRP-immunized mice. A final boost with rrPorB in Ribi resulted in about a 15-fold increase in PorB-specific IgG in the PorB VRP-immunized animals.
IgG1 and IgG2a responses. Measurement of IgG1 and IgG2a isotypes in serum was performed as a surrogate for Th1 and Th2 responses (Fig. 3). Surprisingly, after immunization with rrPorB in Ribi in the FP, the IgG1/IgG2a ratio ranged from 0.2 to 0.7, indicating a Th1 bias. The same vaccine and adjuvant given s.c. in the dorsal area gave a strongly Th2-biased response, based on IgG1/IgG2a ratios ranging from 0.7 to 17. The OMV-immunized mice gave a more balanced response (IgG1/IgG2a ratios, 0.2 to 2.3). All PorB VRP vaccine regimens consistently resulted in low IgG1/IgG2a ratios, usually less than 0.1 to 0.2, indicative of a predominantly Th1-biased response. Interestingly, boosting with rrPorB in Ribi increased the amplitude but did not alter the bias of the response.
Mucosal antibody responses. In order to understand whether different immunization regimens result in differences in the production of antigen-specific mucosal antibody, we examined the vaginal wash samples from individual mice to determine their total IgG and IgA and PorB-specific IgG and IgA antibody levels (Table 4).
In most of the immunization groups, the PorB-specific mucosal IgG responses followed a pattern similar to that of the serum PorB-specific IgG responses (Fig. 4A). Mice immunized s.c. with rrPorB in Ribi generated the highest amount of PorB IgG in mucosal secretions, ranging from 1 to 10 μg/ml. PorB VRP-immunized mice generated only a modest mucosal IgG response, compared to OMV immunization by the i.n. route. The mucosal/serum IgG ratio ranged from 0.01% to 0.78% for all regimens. Either the presence of tPA signal sequence in the constructs or boosting with rrPorB in Ribi improved the amount of specific mucosal PorB-specific IgG without any change of mucosal/serum IgG ratio, reflecting increased serum IgG responses. These results suggested that mucosal results generally were proportional to serum IgG levels and therefore that a significant fraction of the measured mucosal IgG may represent transudation of serum IgG onto the mucosal surface. However, after boosting with rrPorB in Ribi, mice immunized with tPA PorB VRPs developed significantly increased mucosal PorB-specific IgG (P = 0.017) and a higher mucosal/serum IgG ratio (P = 0.03), indicating a better specific mucosal response. This response was elicited by inoculation using only parenteral, nonmucosal routes.
Specific mucosal IgA antibody responses differed according to the immunization regimen (Fig. 4B). Mucosal PorB-specific IgA was highest after immunization with OMVs (up to 100 μg/ml). The rrPorB in Ribi s.c. and the PorB VRP regimens did not induce substantial mucosal IgA production.
Effect of vaccination on cytokine production. Since IgG isotype responses suggested that different vaccine regimens induced either a predominantly Th1- or a predominantly Th2-type response, we wanted to know whether this was also true for T-cell responses. ELISPOT assays were used to monitor the number of PorB-specific IFN-- and IL-4-producing cells after different immunization regimens at 7, 14, and 28 days after the final immunization. For IL-4, the only significant production was observed 7 days after the last immunization, with about 137 spot-forming cells per million stimulated splenocytes from mice immunized with rrPorB in Ribi by the dorsal s.c. route. The number of IL-4-producing cells was significantly higher in this group than in the mock group (PBS in Ribi, 15 spot-forming cells per million stimulated splenocytes). Mice immunized with rrPorB in Ribi by the FP route exhibited 77 spot-forming cells per million splenocytes, a value which was not significantly different from that for the mock group. These numbers demonstrated a greater Th2 bias after vaccination with rrPorB in Ribi by the dorsal s.c. route, consistent with the serum IgG1/IgG2a data.
For IFN- production, similar IFN- responses were observed 7 days after the last immunization with rrPorB in Ribi s.c. administered either in the dorsal or in the FP area, 367 positive cells per million splenocytes for mice vaccinated by the dorsal s.c. route and 435 positive cells per million for those vaccinated by the FP route. Both were significantly higher than values for the mock group. At 14 days after the last immunization, IFN- responses to vaccination by OMV delivered i.n. and PorB VRP delivered in the FP were significantly higher than those in mock groups (Fig. 5A). Positive cells from stimulated splenocytes were 101 per million for the PorB VRP group and 17 per million for the OMV group, consistent with a stronger Th1 bias in the PorB VRP group. To examine the long-term cytokine effect, spleen cells were harvested at 28 days after the third immunization. We compared the IFN- responses of splenic lymphocytes to PorB mixed peptides, after immunization with either PorB VRP (three times), PorB VRP (twice) boosted with rrPorB in Ribi, or rrPorB in Ribi (three times) administered in the FP (Fig. 5B). Results showed that IFN- responses were low in the rrPorB in Ribi in FP group, higher in the PorB VRP (twice) plus rrPorB in Ribi boost group, and highest in the PorB VRP (three times) group. Only PorB VRP (three times)-immunized mice showed a significantly higher number of IFN--producing cells than did the mock group, indicating a stronger Th1-biased memory in the PorB VRP group. Interestingly, at this time point, when specific responses probably were decreasing to memory levels, the IFN- responses to the irrelevant peptide (influenza virus hemagglutinin [HA]) were also high in both of the VRP groups, suggesting a higher frequency of T cells able to respond to nonvaccine peptides after VRP immunization.
Surface binding. The relative amount of surface binding antibody produced by each immunization group was measured in a whole-cell binding assay (Fig. 6). Assays of surface binding antibodies were complicated by the inability to construct a negative control lacking PorB, due to the essential role of PorB. However, we utilized an isogenic strain expressing an immunologically distinct porB gene as a measure of specificity. The porB gene of FA1090 is of the PIB-3 serotype, while the porB gene of FA7224 is PIA-1 (from FA19). We did not subtract antibodies bound to the surface of FA7224 because some of those antibodies could represent cross-reactions to epitopes conserved between the two serotypes. Therefore, comparisons were made with sera from mock-immunized controls. Only the rrPorB in Ribi- and OMV-immunized groups had significantly more surface binding antibodies than the mock group. Levels of surface binding antibodies were higher in the mock-immunized group (which included Ribi adjuvant) than most of the PorB VRP groups (which did not include Ribi adjuvant). The only exception was the PorB VRP group boosted with rrPorB in Ribi adjuvant. This may reflect an immunological response to Ribi adjuvant, which could have induced antibodies that cross-reacted with antigens on the bacterial surface other than PorB. The relatively low levels of surface-bound antibodies observed in the PorB VRP groups were higher than the no-primary-antibody controls. The amount of antibodies recognizing the surface of FA1090 increased in the PorB VRP-immunized group after the final rrPorB in Ribi boost. Immunization with rrPorB in Ribi significantly biased the response toward the homologous strain as judged by comparing the amount of antibody bound to the surface of FA1090 to that bound to FA7224, with the exception of the group immunized with tPA PorB VRP plus boost with rrPorB. Although levels of surface binding antibodies were low in the PorB VRP- and tPA PorB VRP-immunized groups, they were equal for both FA1090 and FA7224, suggesting that they might recognize common PIA and PIB epitopes.
Sera from mice intranasally inoculated with OMV exhibited bactericidal activity against FA1090 and its porB isogenic derivative FA7224 at a serum dilution of 1:80. Despite the presence of antibodies in all the immunized groups that recognized the surface of FA1090, none of the other vaccine regimens produced significant killing compared to the mock group. All sera including PBS-immunized control sera killed FA1090 at a dilution of 1:20.
DISCUSSION
This study investigated the immunogenicity of PorB administered by different delivery systems and routes in mice. We successfully manipulated different aspects of immune responses in mice by using different immunization regimens. Mice immunized s.c. with rrPorB elicited a vigorous serum antibody response. PorB VRP-immunized mice generated a strong Th1-type response characterized by a very low IgG1/IgG2a ratio and high IFN- production. OMV delivered i.n. generated an excellent genital IgA response and also serum bactericidal activity. This information will help us to discover correlates of immunity in the mouse GC infection and protection model.
Mice subcutaneously immunized with rrPorB in Ribi produced high amounts of PorB-specific IgG in serum and the immune sera showed significant specific binding to whole cells, but the immune sera did not exhibit bactericidal activity against FA1090. Several explanations for this negative result are possible. Although PorB is a target for bactericidal antibodies (8), the strain that we used (FA1090) originally was isolated from a patient with disseminated gonococcal infection and is relatively resistant to killing by NHS. Results also were constrained by the limited amount of available mouse antiserum. Use of greater volumes of sera or different target gonococci might affect the results of these experiments.
Curiously, administration of rrPorB in Ribi in the dorsal area gave an apparently Th2-biased response, whereas with administration in the FP there was an apparently Th1-biased response. Reasons for the different responses to rrPorB in Ribi delivered either in the dorsal or in the FP region are unknown. This could relate to lower doses of Ribi adjuvant in the FP regimen but also could reflect other unmeasured differences including local populations of particular dendritic cells, cytokines, or other factors.
Vaccination by outer membrane vesicles delivered intranasally generated good levels of PorB-specific serum IgG, although slightly less than that observed after rrPorB in Ribi immunization. The anti-OMV sera showed high binding to the surface of both FA1090 and FA7224, perhaps reflecting binding mediated by non-PorB-specific immunoglobulins. In addition, the OMV immune serum was bactericidal, and the OMV-immunized mice generated an excellent vaginal mucosal IgA response. These data are consistent with successful i.n. OMV vaccination against strain MS11 infection of the female mouse genital tract (39). It was unclear whether the bactericidal activity and previously reported protection after OMV vaccination were due to antibodies to PorB or to other antigens on OMV. Use of OMV as a vaccine has been problematic because of the likely toxicity of OMV, which contains lipopolysaccharide. However, it is possible to reduce the toxicity of OMV preparations by use of detergent-treated OMV (43), preparation of OMV from meningococcal lipooligosaccharide mutant strains (18), or intranasal administration (1).
A recently appreciated aspect of infection and immune evasion is the ability of GC to invade and grow within epithelial cells (41, 46). It is known that there is a T-cell response after natural gonorrhea (16, 44). Although the relative importance of intraepithelial growth to the life cycle of GC is unknown, it could provide an important means to evade polymorphonuclear neutrophils, antibodies, and complement on mucosal surfaces. A VEE VRP-based gonococcal vaccine has many features that make it an attractive system for studies of potentially protective immune responses. The ability to generate both substantial Th1 responses (31) and mucosal responses (22) with a VRP-based vaccine might be advantageous. We successfully expressed PorB VRP in BHK cells with and without a eukaryotic secretion signal (tPA). The VRP-vaccinated groups tended toward a Th1-type response, based both on serum IgG1/IgG2a ratios and on results of IFN- ELISPOT assays. The tPA-fused antigens consistently were more highly expressed and produced more serum IgG antibodies but retained the same pattern of a Th1-type response. The basis for the greater immunogenicity of the tPA-fused antigens is unknown but presumably reflects different intracellular processing of the expressed antigen in target dendritic cells. The booster effect of the rrPorB in Ribi after PorB VRP immunization suggested that primary immunization with PorB VRPs elicited a good memory response.
Different immune responses generated by different antigen delivery methods provide a fresh opportunity to understand which regimens are protective and to dissect the mechanisms of protection. The recently developed GC infection model in female mice provides a powerful tool to evaluate whether PorB immune responses generated by different delivery systems are protective.
ACKNOWLEDGMENTS
This work was supported by grants 5 R01 AI26837 and 5 U19 AI31496 from the National Institutes of Health to P.F.S.
We thank Christopher Elkins for helpful comments on the design of our immunological assays and Annice Rountree for technical assistance. Ann Jerse provided many helpful comments. We are grateful to Yury Matsuka and Elizabeth Anderson for supplying the FA1090 rrPorB and to numerous members of Robert Johnston's lab for sharing their time, facilities, and expertise with us during the course of these experiments, especially Chad Williamson, Kevin Brown, Martha Collier, and Ande West. We acknowledge the Gonococcal Genome Sequencing Project supported by USPHS/NIH grant AI38399 and B. A. Roe, L. Song, S. P. Lin, X. Yuan, S. Clifton, Tom Ducey, Lisa Lewis, and D. W. Dyer at the University of Oklahoma.
REFERENCES
1. Bakke, H., K. Lie, I. L. Haugen, G. E. Korsvold, E. A. Hoiby, L. M. Naess, J. Holst, I. S. Aaberge, F. Oftung, and B. Haneberg. 2001. Meningococcal outer membrane vesicle vaccine given intranasally can induce immunological memory and booster responses without evidence of tolerance. Infect. Immun. 69:5010-5015.
2. Balasuriya, U. B., H. W. Heidner, J. F. Hedges, J. C. Williams, N. L. Davis, R. E. Johnston, and N. J. MacLachlan. 2000. Expression of the two major envelope proteins of equine arteritis virus as a heterodimer is necessary for induction of neutralizing antibodies in mice immunized with recombinant Venezuelan equine encephalitis virus replicon particles. J. Virol. 74:10623-10630.
3. Bauer, F. J., T. Rudel, M. Stein, and T. F. Meyer. 1999. Mutagenesis of the Neisseria gonorrhoeae porin reduces invasion in epithelial cells and enhances phagocyte responsiveness. Mol. Microbiol. 31:903-913.
4. Blake, M. S., and L. M. Wetzler. 1995. Vaccines for gonorrhea: where are we on the curve Trends Microbiol. 3:469-474.
5. Brodeur, B. R., Y. Larose, P. Tsang, J. Hamel, F. Ashton, and A. Ryan. 1985. Protection against infection with Neisseria meningitidis group B serotype 2b by passive immunization with serotype-specific monoclonal antibody. Infect. Immun. 50:510-516.
6. Bszewczyki, A., and L. M. Koloft. 1985. A method for efficient blotting of basic proteins from SDS-PAGE to nitrocellulose. Anal. Biochem. 150:403-407.
7. Buchanan, T. M., D. A. Eschenbach, J. S. Knapp, and K. K. Holmes. 1980. Gonococcal salpingitis is less likely to recur with Neisseria gonorrhoeae of the same principal outer membrane protein antigenic type. Am. J. Obstet. Gynecol. 138:978-980.
8. Carbonetti, N., V. Simnad, C. Elkins, and P. F. Sparling. 1990. Construction of isogenic gonococci with variable porin structure: effects on susceptibility to human serum and antibiotics. Mol. Microbiol. 4:1009-1018.
9. Carbonetti, N. H., V. I. Simnad, H. S. Seifert, M. So, and P. F. Sparling. 1988. Genetics of protein I of Neisseria gonorrhoeae: construction of hybrid porins. Proc. Natl. Acad. Sci. USA 85:6841-6845.
10. Centers for Disease Control and Prevention. 2002. Discontinuation of cefixime tablets—United States. Morb. Mortal. Wkly. Rep. 51:1052.
11. Cohen, M. S., J. G. Cannon, A. E. Jerse, L. M. Charniga, S. F. Isbey, and L. G. Whicker. 1994. Human experimentation with Neisseria gonorrhoeae: rationale, methods, and implications for the biology of infection and vaccine development. J. Infect. Dis. 169:532-537.
12. Davis, N. L., A. West, E. Reap, G. MacDonald, M. Collier, S. Dryga, M. Maughan, M. Connell, C. Walker, K. McGrath, C. Cecil, L. H. Ping, J. Frelinger, R. Olmsted, P. Keith, R. Swanstrom, C. Williamson, P. Johnson, D. Montefiori, and R. E. Johnston. 2002. Alphavirus replicon particles as candidate HIV vaccines. IUBMB Life 53:209-211.
13. Delvig, A. A., T. E. Michaelsen, A. Aase, E. A. Hoiby, and E. Rosenqvist. 1997. Vaccine-induced IgG antibodies to the linear epitope on the PorB outer membrane protein promote opsonophagocytosis of Neisseria meningitidis by human neutrophils. Clin. Immunol. Immunopathol. 84:27-35.
14. Elkins, C., K. B. Barkley, N. H. Carbonetti, A. J. Coimbre, and P. F. Sparling. 1994. Immunobiology of purified recombinant outer membrane porin protein I of Neisseria gonorrhoeae. Mol. Microbiol. 14:1059-1075.
15. Elkins, C., N. H. Carbonetti, V. A. Varela, D. Stirewalt, D. G. Klapper, and P. F. Sparling. 1992. Antibodies to N-terminal peptides of gonococcal porin are bactericidal when gonococcal lipopolysaccharide is not sialylated. Mol. Microbiol. 6:2617-2628.
16. Elkins, C., and P. F. Sparling. 1992. The immunobiology of Neisseria gonorrhoeae, p. 113-139. In T. C. Quinn (ed.), Sexually transmitted diseases; advances in host defense mechanisms, vol. 8. Raven Press, New York, N.Y.
17. Faruki, H., R. N. Kohmescher, W. P. McKinney, and P. F. Sparling. 1985. Community-based outbreak of infection with penicillin-resistant Neisseria gonorrhoeae not producing penicillinase (chromosomally mediated resistance). N. Engl. J. Med. 313:607-610.
18. Fisseha, M., P. Chen, B. Brandt, T. Kijek, E. Moran, and W. Zollinger. 2005. Characterization of native outer membrane vesicles from lpxL mutant strains of Neisseria meningitidis for use in parenteral vaccination. Infect. Immun. 73:4070-4080.
19. Fleming, D. T., and J. N. Wasserheit. 1999. From epidemiological synergy to public health policy and practice: the contribution of other sexually transmitted diseases to sexual transmission of HIV infection. Sex. Transm. Infect. 75:3-17.
20. Fox, K. K., J. C. Thomas, D. H. Weiner, R. H. Davis, P. F. Sparling, and M. S. Cohen. 1999. Longitudinal evaluation of serovar-specific immunity to Neisseria gonorrhoeae. Am. J. Epidemiol. 149:353-358.
21. Grieder, F. B., N. L. Davis, J. F. Aronson, P. C. Charles, D. C. Sellon, K. Suzuki, and R. E. Johnston. 1995. Specific restrictions in the progression of Venezuelan equine encephalitis virus-induced disease resulting from single amino acid changes in the glycoproteins. Virology 206:994-1006.
22. Harrington, P. R., B. Yount, R. E. Johnston, N. Davis, C. Moe, and R. S. Baric. 2002. Systemic, mucosal, and heterotypic immune induction in mice inoculated with Venezuelan equine encephalitis replicons expressing Norwalk virus-like particles. J. Virol. 76:730-742.
23. Heckels, J. E., M. Virji, K. Zak, and J. N. Fletcher. 1987. Immunobiology of gonococcal outer membrane protein I. Antonie Leeuwenhoek 53:461-464.
24. Hill, J. 1942. Experimental infection with Neisseria gonorrhoeae. Am. J. Syph. Gonorr. Vener. Dis. 27:733-771.
25. Hobbs, M. M., T. M. Alcorn, R. H. Davis, W. Fischer, J. C. Thomas, I. Martin, C. Ison, P. F. Sparling, and M. S. Cohen. 1999. Molecular typing of Neisseria gonorrhoeae causing repeated infections: evolution of porin during passage within a community. J. Infect. Dis. 179:371-381.
26. Jerse, A. E. 1999. Experimental gonococcal genital tract infection and opacity protein expression in estradiol-treated mice. Infect. Immun. 67:5699-5708.
27. Joiner, K. A., R. Scales, K. A. Warren, M. M. Frank, and P. A. Rice. 1985. Mechanism of action of blocking immunoglobulin G for Neisseria gonorrhoeae. J. Clin. Investig. 76:1765-1772.
28. Joiner, K. A., K. A. Warren, M. Tam, and M. M. Frank. 1985. Monoclonal antibodies directed against gonococcal protein I vary in bactericidal activity. J. Immunol. 134:3411-3419.
29. Kellogg, D. S., Jr., W. L. Peacock, Jr., W. E. Deacon, L. Brown, and C. I. Pirkle. 1963. Neisseria gonorrhoeae. I. Virulence genetically linked to clonal variation. J. Bacteriol. 85:1274-1279.
30. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685.
31. Lee, J. S., B. K. Dyas, S. S. Nystrom, C. M. Lind, J. F. Smith, and R. G. Ulrich. 2002. Immune protection against staphylococcal enterotoxin-induced toxic shock by vaccination with a Venezuelan equine encephalitis virus replicon. J. Infect. Dis. 185:1192-1196.
32. Liljestrom, P., and H. Garoff. 1991. A new generation of animal cell expression vectors based on the Semliki Forest virus replicon. Biotechnology (New York) 9:1356-1361.
33. Lozes, E., K. Huygen, J. Content, O. Denis, D. L. Montgomery, A. M. Yawman, P. Vandenbussche, J. P. Van Vooren, A. Drowart, J. B. Ulmer, and M. A. Liu. 1997. Immunogenicity and efficacy of a tuberculosis DNA vaccine encoding the components of the secreted antigen 85 complex. Vaccine 15:830-833.
34. MacDonald, G. H., and R. E. Johnston. 2000. Role of dendritic cell targeting in Venezuelan equine encephalitis virus pathogenesis. J. Virol. 74:914-922.
35. Matsuka, Y. V., D. A. Dilts, S. Hoiseth, and R. Arumugham. 1998. Characterization of a subunit structure and stability of the recombinant porin from Neisseria gonorrhoeae. J. Protein Chem. 17:719-728.
36. McDade, R. L., Jr., and K. H. Johnston. 1980. Characterization of serologically dominant outer membrane proteins of Neisseria gonorrhoeae. J. Bacteriol. 141:1183-1191.
37. Nachamkin, I., J. G. Cannon, and R. S. Mittler. 1981. Monoclonal antibodies against Neisseria gonorrhoeae: production of antibodies directed against a strain-specific cell surface antigen. Infect. Immun. 32:641-648.
38. Ober, W. B. 1969. Boswell's gonorrhea. N. Y. Acad. Med. 45:587-636.
39. Plante, M., A. Jerse, J. Hamel, F. Couture, C. R. Rioux, B. R. Brodeur, and D. Martin. 2000. Intranasal immunization with gonococcal outer membrane preparations reduces the duration of vaginal colonization of mice by Neisseria gonorrhoeae. J. Infect. Dis. 182:848-855.
40. Plummer, F. A., J. N. Simonsen, H. Chubb, L. Slaney, J. Kimata, M. Bosire, J. O. Ndinya-Achola, and E. N. Ngugi. 1989. Epidemiologic evidence for the development of serovar-specific immunity after gonococcal infection. J. Clin. Investig. 83:1472-1476.
41. Post, D. M., N. J. Phillips, J. Q. Shao, D. D. Entz, B. W. Gibson, and M. A. Apicella. 2002. Intracellular survival of Neisseria gonorrhoeae in male urethral epithelial cells: importance of a hexaacyl lipid A. Infect. Immun. 70:909-920.
42. Pushko, P., M. Parker, G. V. Ludwig, N. L. Davis, R. E. Johnston, and J. F. Smith. 1997. Replicon-helper systems from attenuated Venezuelan equine encephalitis virus: expression of heterologous genes in vitro and immunization against heterologous pathogens in vivo. Virology 239:389-401.
43. Quakyi, E. K., H. D. Hochstein, and C. M. Tsai. 1997. Modulation of the biological activities of meningococcal endotoxins by association with outer membrane proteins is not inevitably linked to toxicity. Infect. Immun. 65:1972-1979.
44. Simpson, S. D., Y. Ho, P. A. Rice, and L. M. Wetzler. 1999. T lymphocyte response to Neisseria gonorrhoeae porin in individuals with mucosal gonococcal infections. J. Infect. Dis. 180:762-773.
45. Tam, M. R., T. M. Buchanan, E. C. Sandstrom, K. K. Holmes, J. S. Knapp, A. W. Siadak, and R. C. Nowinski. 1982. Serological classification of Neisseria gonorrhoeae with monoclonal antibodies. Infect. Immun. 36:1042-1053.
46. van Putten, J. P., T. D. Duensing, and J. Carlson. 1998. Gonococcal invasion of epithelial cells driven by P.IA, a bacterial ion channel with GTP binding properties. J. Exp. Med. 188:941-952.
47. Virji, M., J. N. Fletcher, K. Zak, and J. E. Heckels. 1987. The potential protective effect of monoclonal antibodies to gonococcal outer membrane protein IA. J. Gen. Microbiol. 133:2639-2646.
48. Virji, M., K. Zak, and J. E. Heckels. 1986. Monoclonal antibodies to gonococcal outer membrane protein IB: use in investigation of the potential protective effect of antibodies directed against conserved and type-specific epitopes. J. Gen. Microbiol. 132:1621-1629.
49. Zhu, W., C. E. Thomas, and P. F. Sparling. 2004. DNA immunization of mice with a plasmid encoding Neisseria gonorrhoeae PorB protein by intramuscular injection and epidermal particle bombardment. Vaccine 22:660-669.(Weiyan Zhu, Christopher E)
ABSTRACT
Porin (PorB) is a major outer membrane protein produced by all Neisseria gonorrhoeae strains and has been a focus of intense interest as a vaccine candidate. In this study, the immunogenicity of PorB in mice was investigated after several immunization regimens. Outer membrane vesicles (OMV), recombinant renatured PorB (rrPorB), and PorB-expressing Venezuelan equine encephalitis (VEE) virus replicon particles (PorB VRP) were delivered intranasally (i.n.) or subcutaneously (s.c.) into the dorsal area or the hind footpad in three-dose schedules; the PorB VRP-immunized mice were given a single additional booster dose of rrPorB in Ribi adjuvant. Different delivery systems and administration routes induced different immune responses. Mice immunized s.c. with rrPorB in Ribi had the highest levels of PorB-specific serum immunoglobulin G (IgG) by enzyme-linked immunosorbent assay. Surprisingly, there was an apparent Th1 bias, based on IgG1/IgG2a ratios, after immunization with rrPorB in Ribi in the footpad while the same vaccine given in the dorsal area gave a strongly Th2-biased response. PorB VRP-immunized mice produced a consistent Th1 response with a high gamma interferon response in stimulated splenic lymphocytes and very low IgG1/IgG2a ratios. Immunization by OMV delivered i.n. was the only regimen that resulted in a serum bactericidal response, and it generated an excellent mucosal IgA response. Serum from mice immunized with rrPorB preferentially recognized the surface of whole gonococci expressing a homologous PorB, whereas serum from PorB VRP-immunized mice had relatively low whole-cell binding activity but recognized both heterologous and homologous PorB equally. The data resulting from this direct comparison suggested that important aspects of the immune response can be manipulated by altering the form of the antigen and its delivery. This information coupled with an understanding of protective antigonococcal immune responses will enable the design of the optimal vaccine for N. gonorrhoeae.
INTRODUCTION
Infection by Neisseria gonorrhoeae (the gonococcus [GC]) remains a serious problem in the United States and the world (10). Although GC infection has declined in recent years in the United States, with total reported infections of "only" 300,000 annually, prevalence may be as high as 5% in certain populations in inner cities and the rural Southeast, especially in African-Americans and socioeconomically deprived groups. Women suffer the bulk of the complications in the form of salpingitis, infertility, and ectopic pregnancy. Fetal death is inevitable in the case of ectopic pregnancy. Calculations of attributable risk show that GC is one of the major cofactors for human immunodeficiency virus (HIV) transmission (19), increasing risks of HIV transmission about threefold. Emergence of antibiotic-resistant GC including GC resistant to ciprofloxacin is threatening standard treatment modalities, just as earlier emergence of penicillin and tetracycline resistance relegated those classic therapies to history. Unless new oral therapies are developed, we will face a challenge that we have not experienced for decades. A safe and effective vaccine would be tremendously useful for all these reasons.
A vaccine for GC has seemed a difficult task based solely on natural history, which shows that repeat infections are common, including repeat infection by the same strain (17, 20, 25). However, in earlier days, before the advent of effective therapy, GC infections eventually underwent spontaneous resolution (38), with negative cultures and lack of risk to sexual partners (24). This showed that prolonged infection, although often associated with local complications, eventually resulted in immune clearance. Thus, an effective gonorrhea vaccine might be possible if the right immune response were directed at the right antigens.
Unlike the meningococcus, GC does not make capsules, and the search for a GC vaccine is aimed at outer membrane proteins rather than a capsule. Experience from the meningococcus group B outer membrane vesicle (OMV) vaccines showed that protection was correlated strongly with responses to porin protein antigens in the OMVs (13). PorB, which is expressed on all gonococci, is the major integral outer membrane protein on the surface of gonococci. It is essential to GC viability (9) and plays important roles in the pathogenesis of GC (3, 45). PorB is expressed constitutively and is relatively stable antigenically without undergoing phase or high-frequency antigenic variation (36, 45). Certain antibodies against PorB are known to be bactericidal (28), opsonic (23), or protective against toxicity in tissue culture (47, 48). Two reports have claimed PorB serovar-specific protection against reinfection in humans. In the first, female commercial sex workers in Nairobi, Kenya, were less likely to be reinfected by the same serovar, even though most were immunosuppressed by their HIV infection (40). The second, smaller study found no reinfections of the fallopian tubes (salpingitis) by GC with the same serovar as that for the index infection (7). These data suggest that developing an immune attack on PorB might be an effective vaccine strategy. However, GCs have developed several mechanisms to protect against host immune attack on PorB. These include the blocking antigen Rmp, which stimulates production of non-complement-fixing antibodies that block bactericidal activity of anti-PorB sera (14), and sialylation of terminal sugars on lipooligosaccharide, forming an "umbrella" that partially limits access of antibodies to PorB (15) and interferes with complement activation (15). A partially purified PorB vaccine isolated from GC apparently was not effective in early studies of experimentally inoculated human male volunteers (P. A. Rice, S. Gulait, and S. Ram, Abstr. 10th Int. Pathog. Neisseria Conf., abstr. 3, 1996). Subsequent analysis of this experiment suggested that failure was due to a vigorous immune response to contaminating blocking antigen Rmp (4). Thus, a successful PorB vaccine presumably would need to stimulate high titers of antibodies without eliciting anti-Rmp antibodies. Use of recombinant PorB from Escherichia coli avoids the problem of the immunogenicity of small amounts of contaminating Rmp when PorB is prepared from wild-type GC. Similarly, use of a Venezuelan equine encephalitis virus (VEE)-based viral vector, VEE replicon particle (VRP), expressing PorB completely avoids the problems of contamination by Rmp in a vaccine.
In addition to the choice of target antigen, the success of a vaccine also depends on the type, site, strength, and specificity of the immune responses that it induces. The route of administration of vaccines contributes to the effectiveness of the local and systemic immune responses. This prompted us to investigate the immunogenicity of PorB administered by different routes and in different forms. We utilized renatured recombinant PorB (rrPorB), as well as a VEE vaccine vector system expressing PorB. The VEE system is composed of an RNA replicon and a bipartite helper system for packaging the replicon into propagation-deficient VRPs (42). It is an efficient expression system for in vivo delivery of heterologous antigens to dendritic cells (34) and expresses very high levels of immunogen in cells in the draining lymph node. It can induce both systemic and mucosal immunity (12, 22), which presumably is important to host defense against pathogens such as N. gonorrhoeae that enter across a mucosal surface, and both humoral and cellular responses.
An improved mouse model of genital gonococcal infection has been developed recently (26). Intranasal (i.n.) immunization with GC OMVs reduced the duration of vaginal colonization of homologous strain MS11 in the female mouse genital tract (39). In this study, we tested the immunogenicity of rrPorB, OMV, and PorB expressed from a VEE-based vector (PorB VRP) in order to inform future vaccine studies of gonorrhea in the mouse challenge model.
MATERIALS AND METHODS
Strain, recombinant antigen, and rabbit anti-PorB serum. N. gonorrhoeae strain FA1090 (37) variant A23 was used throughout these experiments and was obtained from Marcia Hobbs (11). Passage of FA1090 and its derivatives was kept to a minimum to avoid phase or antigenic variation. Gonococcal strains were routinely maintained on GC base medium (Difco) supplemented with Kellogg's supplement I (29). All relevant strains and plasmids are shown in Table 1.
An individual isolate of pUNCH682 (porB) in BL21(DE3) (Invitrogen) expressing a high level of antigen was expanded to produce recombinant protein by inducing a culture with an optical density of 0.4 to 0.5 with 2 mM isopropyl--D-thiogalactopyranoside for 30 min and arresting normal cellular RNA synthesis with rifampin for an additional 2 h. We purified the hexa-His-tagged fusion rrPorB on Ni-nitrilotriacetic acid (NTA) agarose (QIAGEN) under denaturing conditions. Cell pellets were lysed for 10 min in 6 M guanidinium hydrochloride buffered with 100 mM NaH2PO4 and 10 mM Tris to pH 8.0. DNA was sheared by 10 passages through an 18- to 22-gauge needle; cellular debris was removed by pelleting at 10,000 x g for 30 min. His-tagged rrPorB was bound to Ni-NTA agarose in batch format for 30 min; the resin was washed with 8 M urea buffered as above to pH 8 in column format; the resin was washed again in 8 M urea buffered as above to pH 6.7. His-tagged rrPorB was eluted in 8 M urea buffered as above to pH 5.0. An attempt to renature the recombinant protein was made by slow dilution of the denaturing urea in the presence of low concentrations of detergents. The His-tagged rrPorB had assumed some native conformation as demonstrated by binding of conformation-dependent monoclonal antibodies (data not shown). The His-tagged rrPorB was used to immunize Elite rabbits at Covance Research Products. Rabbits were immunized with 250 μg of recombinant protein in Freund's complete adjuvant and boosted with 125 μg of recombinant protein in Freund's incomplete adjuvant three times at 3-week intervals. Rabbit sera were collected at 7, 10, and 13 weeks.
The renatured and purified rrPorB used in subsequent mouse immunization studies and immunoassays was a kind gift of Yuri Matsuka and Elizabeth Anderson of Wyeth Research. This material was prepared in a manner similar to a published method (35). In brief, rrPorB was expressed in Escherichia coli and refolded in vitro in the presence of a zwitterionic detergent. Its proper folding and subunit organization were confirmed by comparison with the native counterpart.
Preparation of OMVs. OMVs were prepared from the FA1090 rmp mutant strain FA6902 (rmp) as described elsewhere (5) with the following modifications. GCs were grown under iron-stressed conditions to fully express all iron-repressed proteins. The culture was collected by centrifugation at 10,000 x g (4°C) for 15 min and resuspended in 15 ml cold lithium acetate buffer (200 mM LiCl, 100 mM C2H3O2Li, pH 6.0). The bacterial suspension was passed through a 20-1/2-gauge needle for 20 times to shear off the outer membrane and centrifuged for 2 min to pellet whole bacteria, and the supernatant was ultracentrifuged at 100,000 x g (15°C) for 2 h. The pellets were resuspended in a small amount of lithium acetate buffer and frozen at –20°C until needed.
Production of PorB VRPs with and without a eukaryotic secretion signal (tPA). Gonococcal porB was amplified by PCR from FA1090 using primers GCV-3 and GCV-4 for porB. All oligonucleotide primers are shown in Table 2. The PCR product was then cloned using the TA cloning system (Invitrogen) to create pUNCH678 (porB). The eukaryotic secretion signal from tissue plasminogen activator (tPA) was placed upstream of the mature coding sequences of porB by cloning the insert from pUNCH678 (porB) downstream of the tPA signal sequence in pUNCH689 (tPA). The plasmid pUNCH689 is a derivative of VJns-tPA (33) with the BglII-AccI adapter inserted into the BglII site, creating a new unique AccI site which is compatible with the ClaI sites flanking the inserts of pUNCH678 (porB). The tPA-containing clone was designated pUNCH693 (tPA-porB), and this clone was used as the template for the corresponding replicon plasmid discussed below.
The replicon plasmids were assembled essentially as previously described (2) by a process of overlapping PCR amplification to create an authentic fusion between the 26S VEE viral structural gene promoter and the synthetic start codon of the bacterial gene. The insert from pUNCH678 (porB) was amplified by PCR using primers GCV-7/GCV-8 to incorporate an upstream overlap with the 26S VEE promoter and a downstream AscI site. In a similar fashion the insert from pUNCH693 (tPA-porB) was amplified with GCV-12/GCV-8. The 26S promoter was amplified with primers GCV-9/GCV-10 and combined with each of the two previous PCR products in an overlapping PCR fusion using GCV-9 as the upstream amplimer and GCV-8 as the downstream amplimer. These two PCR fusion products were digested with ApaI (cuts the 26S promoter) and AscI (cuts the downstream end of each of the PCR products) and ligated into similarly cut replicon vector pVR21 (2), to give rise to pUNCH697 (tPA-porB) and pUNCH699 (porB). The inserts of the two final constructs were sequenced to confirm fidelity, and any errors were corrected by screening and/or subcloning (Table 2 shows amplimer sequences). RNase-free NotI-linearized replicon plasmids were transcribed using the mMESSAGE mMACHINE T7 kit (Ambion), and the capped runoff transcripts were electroporated into baby hamster kidney (BHK) cells using the conditions of Liljestrom and Garoff (32) (0.4-cm gap cuvette; three pulses, 0.85 kV, 25 μF). VRPs were generated by coelectroporating the transcript containing porB or tPA-porB with two helper mRNAs encoding the VEE capsid and glycoproteins. Glycoprotein from the V3000 molecular clone (21) was used in these constructs. VRPs were harvested between 20 and 27 h after transfection and purified from cell culture supernatants by pelleting through a 20% sucrose cushion. The pelleted VRPs were resuspended in phosphate-buffered saline (PBS) and stored at –80°C. Efficiency of transfection was measured by indirect immunofluorescence. An aliquot of each electroporation was plated onto a chamber slide and incubated along with the rest of the electroporation for 18 to 24 h. The chamber slides were rinsed with PBS, fixed for 10 min in ice-cold acetone, dried, and stored at –20°C. Slides were later rehydrated in PBS with 0.2% normal goat serum (NGS) and then blocked in 5% NGS. Antigen was detected using the rabbit antiserum (1:1,000 anti-His-PorB), biotin-SP-AffiniPure goat anti-rabbit immunoglobulin G (IgG) (heavy plus light) (1:100, catalog no. 111-065-003), and cyanine Cy2 streptavidin (1:100, catalog no. 016-220-084) (Jackson ImmunoResearch Laboratories) sequentially, each diluted in PBS-5% NGS. Each VRP preparation was shown to be free of propagation-competent virus by a sensitive cytopathic effect assay following serial passage on BHK cells. The titer of VRP in each preparation was determined by indirect immunofluorescence on serial dilutions as described above.
Western blot analysis of replicon-based expression. Efficiency of expression was measured by subjecting a crude lysate of the transfected BHK cells to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and Western blotting (6, 30). The crude lysate consisted of rinsed BHK cells solubilized in 1% SDS for 5 min at 37°C and diluted to equal total protein concentration determined by bicinchoninic acid protein assay (Pierce, Rockford, IL). SDS-PAGE samples were routinely boiled for 2 min under reducing conditions (5% - mercaptoethanol) prior to loading. Antigen was detected in standard immunoblot format using 1:8,000 rabbit anti-His-PorB; 1:5,000 alkaline phosphatase-conjugated sheep anti-rabbit IgG F(ab')2 Frag (Sigma, St. Louis, MO), and LumiPhos WB (Pierce, Rockford, IL).
Mouse inoculation and sample collection. Four- to five-week-old female BALB/c mice were purchased from Harlan Laboratory Animals and housed in the University of North Carolina animal facility according to federal and institutional animal care and use committee guidelines. rrPorB was formulated with Ribi R-700 (Corixa, Hamilton, MT), which consists of monophosphoryl lipid A and synthetic trehalose dicorynomycolate in a squalene-Tween 80 emulsion. This adjuvant was prepared at a 2x concentration in PBS according to manufacturer's instructions. Mice in groups of six were immunized as described in Table 3. All the mice were anesthetized with methoxyflurane (Medical Developments, Australia) before inoculation. PorB VRP groups were immunized with 20 μl PBS containing a total of 105 to 106 PorB VRPs or 105 tPA PorB VRPs in the left hind footpad (FP). No adverse effects of VRP inoculation were observed. Recombinant proteins were inoculated by two delivery routes: subcutaneously (s.c.) via FP with 10 μg rrPorB in 20 μl of R-700 Ribi adjuvant system (Corixa) or s.c. via dorsal area with 5 μg rrPorB in 50 μl of R-700 Ribi at each of two sites. OMV-immunized mice received 10 μl PBS containing 10 μg of OMV in each nostril. Mice were boosted with immunizations identical to their prime immunization at 3, 6, and 9 weeks postprime. For some of the PorB VRP animals, a final boost with a total of 10 μg of rrPorB in Ribi adjuvant was given s.c. in the dorsal area at 9 weeks. A single dose of 10 μg rrPorB in Ribi adjuvant ("1° rrPorB") was given in the dorsal area at 9 weeks to an unimmunized control group to compare the results to those obtained after the booster dose of the same immunogen given to the PorB VRP group. A mock-immunized group received 50 μl of R-700 Ribi adjuvant s.c. at each of two dorsal sites without added protein at 0, 3, 6, and 9 weeks.
Venous tail blood was collected 2 weeks after the third and fourth immunizations. Blood was collected in Microtainer serum separator tubes (Becton Dickinson), allowed to clot for 30 min at room temperature, and stored at 4°C overnight. The following morning tubes were centrifuged according to manufacturer's specifications. Serum was filtered through a sterile 0.45-μm cellulose acetate centrifuge tube filter (Costar). Vaginal mucosal samples were collected at 2 weeks after the fourth immunization by washing the mouse vaginal cavity several times with 50 μl PBS, followed by addition of 10 μl 5x stabilization buffer (protease inhibitor cocktail P8340 [Sigma, St. Louis, MO] in PBS with 0.1% sodium azide). Samples were stored at –20°C until needed.
Serum and vaginal mucosal antibody analysis. The quantity of antibody present in immunized mouse serum and washed from mucosal surfaces was measured by indirect enzyme-linked immunosorbent assay (ELISA) as previously described by Zhu et al. (49). The quantity of PorB-specific antibody bound to the plates was determined by comparing the optical density of the PorB-coated wells to that obtained by coating the wells with goat anti-mouse IgG-Fc-affinity-purified coating antibody at 1:100 (or IgA-affinity-purified coating antibody at 1:100 for IgA ELISA), comparing them to a standard curve of a mouse reference serum with known quantities of immunoglobulins (IgG, IgA, IgG1, or IgG2a, all purchased from Bethyl Laboratories), and developing them in parallel with the PorB-specific wells.
Cytokine assays. Splenocytes were collected from immunized mice at 2 weeks after the final immunization and tested for gamma interferon (IFN-) and interleukin-4 (IL-4) secretion by ELISPOT assay. Viable splenocytes were isolated with Lympolyte-M (Accurate, Westbury, NY). Thirty-two synthetic PorB peptides of 20 amino acids in length with a 10-amino-acid overlap spanning PorB were synthesized by Mimotopes (Victoria, Australia). The individual peptides were dissolved in dimethyl sulfoxide and mixed to make a master pool of equal concentrations. Aliquots of the master pool were further diluted with AIM V (Invitrogen) immediately before addition to the splenocytes to reach a final concentration of 1.5 μg/ml in the wells. For each ELISPOT assay, a 96-well filtration plate (Millipore, Bedford, MA) was coated overnight with either antibody to mouse IFN- (Mabtech, Nacka, Sweden; 5 μg/ml) or antibody to mouse IL-4 (Mabtech; 2.5 μg/ml) at 4°C. The plates were washed with PBS and blocked with RPMI medium containing 10% fetal calf serum. After being washed with PBS, each well was loaded with 6 x 105 cells. The plates were incubated at 37°C, 5% CO2, for 20 (IFN-) or 40 (IL-4) hours. Production of cytokines was detected by addition of biotinylated anti-mouse IFN- or anti-mouse IL-4 antibodies (Mabtech). After the wells were washed, peroxidase-labeled streptavidin was added. The spots were developed with the Sigma AEC Chromogen kit. Plates were sent to Zellnet Consulting Inc. (Fort Lee, NJ), and results were reported as spot-forming cells per million splenocytes.
Whole-cell binding assay. The relative amount of antibodies that recognized the surface of the bacteria was measured by indirect whole-cell radioimmunoassay. This assay consisted of exposing dilutions of serum to whole bacteria differing only in the antigen of interest and detecting bound antibody using 125I-labeled goat anti-mouse immunoglobulin (8, 27). Immune sera were diluted 1:100 in 50 μl Bacto GC medium base broth (GCB) and then mixed with 150 μl GCB containing 2 x 108/ml CFU of gonococci (final dilution of serum, 1:400) for 15 min at room temperature. The gonococci were collected by filtration in a multiscreen plate (Millipore, Bedford, MA). Filters were washed five times with PBS-azide (0.1%) buffer before addition of 106 cpm of 125I-labeled secondary antibody (goat anti-mouse IgG, IgA, and IgM; ICN Biomedicals, Inc., Aurora, OH) in 100 μl to each well. After incubation at room temperature for 15 min, the filters were washed five times with PBS-azide (0.1%) and air dried, and the filter bottom was punched out into borosilicate glass tubes. Antibody binding was assessed by counting gonococcus-associated radioactivity in a 1271 RIAGAMMA counter (LKB Wallac). Controls included either no antibody or monoclonal anti-PorB antibodies that bind to only one strain. The antibodies recognizing the surface-exposed epitopes of PorB were measured by comparing binding to FA1090-A23 (porB serovar PIB-3) and an isogenic derivative, FA7224, containing FA19 porB (serovar PIA-1). In FA7224 the porB gene of FA1090 was replaced with the PIA-1 serovar porB gene of FA19. The complete replacement of the porB gene was confirmed by PCR and sequencing.
Bactericidal assay. Bactericidal activity against FA1090 and FA7224 was measured as previously described by Elkins (15), with fresh pooled normal human serum (NHS) as a complement source. Briefly, immunized mouse sera were heat inactivated for 30 min at 56°C before being added to 104 CFU/ml gonococci in 100 μl GCB (final dilution, 1:50) and incubated at 37°C for 15 min. NHS was added to a final concentration of 10%, and incubation was continued at 37°C for 30 additional minutes. Samples were plated, and viable counts were determined. Rabbit serum was used at a 1:100 dilution as a positive control. Other controls included bacteria incubated with 10% NHS alone and 10% NHS inactivated at 56°C for 30 min. Significant killing was defined as a greater than 50% decrease in viable count compared to the PBS-immunized group.
Statistical analysis. The difference between immunization groups was analyzed by analysis of variance on ranks. Significant results of analysis of variance were further analyzed by the Mann-Whitney rank sum test for two-group comparison. Statistical analyses were performed using SigmaStat 3.0 (Systat Software Inc., Richmond, Calif.). Differences were considered significant at P < 0.05.
RESULTS
Expression of gonococcal antigens in BHK cells. Because of the risk that the bacterial codon usage would be less than optimal in a eukaryotic expression system, the replicon expression plasmids were tested in their earliest form, that is, RNA transfected by electroporation into BHK cells. We assessed antigen expression by both indirect immunofluorescence and Western blotting of whole-cell lysates. Transfection frequencies, as measured by percentage of BHK cells expressing PorB by immunofluorescence, were high (>80%) and reasonably consistent between expression plasmids (data not shown). Western blot assays of whole-cell lysates from these electroporations (Fig. 1) showed that the replicon expression plasmids were capable of expressing full-length antigen (35 kDa). Earlier replicon expression plasmids containing porB fused to the viral promoter with an intervening short insert of irrelevant sequence showed very poor expression and poor immune responses compared to these constructs (data not shown). This indicated the importance of direct fusion of the viral sequence just upstream of the start codon in achieving maximum expression of these bacterial gene products. Despite the inherent difficulties in quantifying the levels of expression by Western blotting, the tPA-containing replicons consistently expressed more gonococcal antigen than identical replicons lacking this eukaryotic signal sequence. Faster-migrating immunoreactive bands could represent breakdown products of the primary expression product or alternative translation products that remain immunoreactive and therefore could still serve as antigen.
Development of anti-PorB serum IgG. All delivery systems and routes induced PorB-specific serum IgG responses after immunization (Fig. 2). Mice immunized s.c. with rrPorB in Ribi either dorsally or in the FP exhibited the highest serum specific IgG levels, which ranged from 1 mg/ml to 2 mg/ml. Mice immunized intranasally with OMV generated a similar response, with 0.6 mg/ml of PorB-specific serum IgG.
All mice immunized with PorB VRPs developed specific IgG after inoculation. At 14 days after the third immunization, PorB VRP-immunized mice produced 20 μg/ml specific IgG antibody, which was 200-fold greater than background levels in the mock group but only about 1% of that induced by rrPorB in Ribi immunization. Inclusion of the tPA signal sequence in the PorB VRP vaccine constructs improved the amount of specific IgG produced by PorB VRP-immunized mice by about fivefold compared to the non-tPA PorB VRP-immunized mice. A final boost with rrPorB in Ribi resulted in about a 15-fold increase in PorB-specific IgG in the PorB VRP-immunized animals.
IgG1 and IgG2a responses. Measurement of IgG1 and IgG2a isotypes in serum was performed as a surrogate for Th1 and Th2 responses (Fig. 3). Surprisingly, after immunization with rrPorB in Ribi in the FP, the IgG1/IgG2a ratio ranged from 0.2 to 0.7, indicating a Th1 bias. The same vaccine and adjuvant given s.c. in the dorsal area gave a strongly Th2-biased response, based on IgG1/IgG2a ratios ranging from 0.7 to 17. The OMV-immunized mice gave a more balanced response (IgG1/IgG2a ratios, 0.2 to 2.3). All PorB VRP vaccine regimens consistently resulted in low IgG1/IgG2a ratios, usually less than 0.1 to 0.2, indicative of a predominantly Th1-biased response. Interestingly, boosting with rrPorB in Ribi increased the amplitude but did not alter the bias of the response.
Mucosal antibody responses. In order to understand whether different immunization regimens result in differences in the production of antigen-specific mucosal antibody, we examined the vaginal wash samples from individual mice to determine their total IgG and IgA and PorB-specific IgG and IgA antibody levels (Table 4).
In most of the immunization groups, the PorB-specific mucosal IgG responses followed a pattern similar to that of the serum PorB-specific IgG responses (Fig. 4A). Mice immunized s.c. with rrPorB in Ribi generated the highest amount of PorB IgG in mucosal secretions, ranging from 1 to 10 μg/ml. PorB VRP-immunized mice generated only a modest mucosal IgG response, compared to OMV immunization by the i.n. route. The mucosal/serum IgG ratio ranged from 0.01% to 0.78% for all regimens. Either the presence of tPA signal sequence in the constructs or boosting with rrPorB in Ribi improved the amount of specific mucosal PorB-specific IgG without any change of mucosal/serum IgG ratio, reflecting increased serum IgG responses. These results suggested that mucosal results generally were proportional to serum IgG levels and therefore that a significant fraction of the measured mucosal IgG may represent transudation of serum IgG onto the mucosal surface. However, after boosting with rrPorB in Ribi, mice immunized with tPA PorB VRPs developed significantly increased mucosal PorB-specific IgG (P = 0.017) and a higher mucosal/serum IgG ratio (P = 0.03), indicating a better specific mucosal response. This response was elicited by inoculation using only parenteral, nonmucosal routes.
Specific mucosal IgA antibody responses differed according to the immunization regimen (Fig. 4B). Mucosal PorB-specific IgA was highest after immunization with OMVs (up to 100 μg/ml). The rrPorB in Ribi s.c. and the PorB VRP regimens did not induce substantial mucosal IgA production.
Effect of vaccination on cytokine production. Since IgG isotype responses suggested that different vaccine regimens induced either a predominantly Th1- or a predominantly Th2-type response, we wanted to know whether this was also true for T-cell responses. ELISPOT assays were used to monitor the number of PorB-specific IFN-- and IL-4-producing cells after different immunization regimens at 7, 14, and 28 days after the final immunization. For IL-4, the only significant production was observed 7 days after the last immunization, with about 137 spot-forming cells per million stimulated splenocytes from mice immunized with rrPorB in Ribi by the dorsal s.c. route. The number of IL-4-producing cells was significantly higher in this group than in the mock group (PBS in Ribi, 15 spot-forming cells per million stimulated splenocytes). Mice immunized with rrPorB in Ribi by the FP route exhibited 77 spot-forming cells per million splenocytes, a value which was not significantly different from that for the mock group. These numbers demonstrated a greater Th2 bias after vaccination with rrPorB in Ribi by the dorsal s.c. route, consistent with the serum IgG1/IgG2a data.
For IFN- production, similar IFN- responses were observed 7 days after the last immunization with rrPorB in Ribi s.c. administered either in the dorsal or in the FP area, 367 positive cells per million splenocytes for mice vaccinated by the dorsal s.c. route and 435 positive cells per million for those vaccinated by the FP route. Both were significantly higher than values for the mock group. At 14 days after the last immunization, IFN- responses to vaccination by OMV delivered i.n. and PorB VRP delivered in the FP were significantly higher than those in mock groups (Fig. 5A). Positive cells from stimulated splenocytes were 101 per million for the PorB VRP group and 17 per million for the OMV group, consistent with a stronger Th1 bias in the PorB VRP group. To examine the long-term cytokine effect, spleen cells were harvested at 28 days after the third immunization. We compared the IFN- responses of splenic lymphocytes to PorB mixed peptides, after immunization with either PorB VRP (three times), PorB VRP (twice) boosted with rrPorB in Ribi, or rrPorB in Ribi (three times) administered in the FP (Fig. 5B). Results showed that IFN- responses were low in the rrPorB in Ribi in FP group, higher in the PorB VRP (twice) plus rrPorB in Ribi boost group, and highest in the PorB VRP (three times) group. Only PorB VRP (three times)-immunized mice showed a significantly higher number of IFN--producing cells than did the mock group, indicating a stronger Th1-biased memory in the PorB VRP group. Interestingly, at this time point, when specific responses probably were decreasing to memory levels, the IFN- responses to the irrelevant peptide (influenza virus hemagglutinin [HA]) were also high in both of the VRP groups, suggesting a higher frequency of T cells able to respond to nonvaccine peptides after VRP immunization.
Surface binding. The relative amount of surface binding antibody produced by each immunization group was measured in a whole-cell binding assay (Fig. 6). Assays of surface binding antibodies were complicated by the inability to construct a negative control lacking PorB, due to the essential role of PorB. However, we utilized an isogenic strain expressing an immunologically distinct porB gene as a measure of specificity. The porB gene of FA1090 is of the PIB-3 serotype, while the porB gene of FA7224 is PIA-1 (from FA19). We did not subtract antibodies bound to the surface of FA7224 because some of those antibodies could represent cross-reactions to epitopes conserved between the two serotypes. Therefore, comparisons were made with sera from mock-immunized controls. Only the rrPorB in Ribi- and OMV-immunized groups had significantly more surface binding antibodies than the mock group. Levels of surface binding antibodies were higher in the mock-immunized group (which included Ribi adjuvant) than most of the PorB VRP groups (which did not include Ribi adjuvant). The only exception was the PorB VRP group boosted with rrPorB in Ribi adjuvant. This may reflect an immunological response to Ribi adjuvant, which could have induced antibodies that cross-reacted with antigens on the bacterial surface other than PorB. The relatively low levels of surface-bound antibodies observed in the PorB VRP groups were higher than the no-primary-antibody controls. The amount of antibodies recognizing the surface of FA1090 increased in the PorB VRP-immunized group after the final rrPorB in Ribi boost. Immunization with rrPorB in Ribi significantly biased the response toward the homologous strain as judged by comparing the amount of antibody bound to the surface of FA1090 to that bound to FA7224, with the exception of the group immunized with tPA PorB VRP plus boost with rrPorB. Although levels of surface binding antibodies were low in the PorB VRP- and tPA PorB VRP-immunized groups, they were equal for both FA1090 and FA7224, suggesting that they might recognize common PIA and PIB epitopes.
Sera from mice intranasally inoculated with OMV exhibited bactericidal activity against FA1090 and its porB isogenic derivative FA7224 at a serum dilution of 1:80. Despite the presence of antibodies in all the immunized groups that recognized the surface of FA1090, none of the other vaccine regimens produced significant killing compared to the mock group. All sera including PBS-immunized control sera killed FA1090 at a dilution of 1:20.
DISCUSSION
This study investigated the immunogenicity of PorB administered by different delivery systems and routes in mice. We successfully manipulated different aspects of immune responses in mice by using different immunization regimens. Mice immunized s.c. with rrPorB elicited a vigorous serum antibody response. PorB VRP-immunized mice generated a strong Th1-type response characterized by a very low IgG1/IgG2a ratio and high IFN- production. OMV delivered i.n. generated an excellent genital IgA response and also serum bactericidal activity. This information will help us to discover correlates of immunity in the mouse GC infection and protection model.
Mice subcutaneously immunized with rrPorB in Ribi produced high amounts of PorB-specific IgG in serum and the immune sera showed significant specific binding to whole cells, but the immune sera did not exhibit bactericidal activity against FA1090. Several explanations for this negative result are possible. Although PorB is a target for bactericidal antibodies (8), the strain that we used (FA1090) originally was isolated from a patient with disseminated gonococcal infection and is relatively resistant to killing by NHS. Results also were constrained by the limited amount of available mouse antiserum. Use of greater volumes of sera or different target gonococci might affect the results of these experiments.
Curiously, administration of rrPorB in Ribi in the dorsal area gave an apparently Th2-biased response, whereas with administration in the FP there was an apparently Th1-biased response. Reasons for the different responses to rrPorB in Ribi delivered either in the dorsal or in the FP region are unknown. This could relate to lower doses of Ribi adjuvant in the FP regimen but also could reflect other unmeasured differences including local populations of particular dendritic cells, cytokines, or other factors.
Vaccination by outer membrane vesicles delivered intranasally generated good levels of PorB-specific serum IgG, although slightly less than that observed after rrPorB in Ribi immunization. The anti-OMV sera showed high binding to the surface of both FA1090 and FA7224, perhaps reflecting binding mediated by non-PorB-specific immunoglobulins. In addition, the OMV immune serum was bactericidal, and the OMV-immunized mice generated an excellent vaginal mucosal IgA response. These data are consistent with successful i.n. OMV vaccination against strain MS11 infection of the female mouse genital tract (39). It was unclear whether the bactericidal activity and previously reported protection after OMV vaccination were due to antibodies to PorB or to other antigens on OMV. Use of OMV as a vaccine has been problematic because of the likely toxicity of OMV, which contains lipopolysaccharide. However, it is possible to reduce the toxicity of OMV preparations by use of detergent-treated OMV (43), preparation of OMV from meningococcal lipooligosaccharide mutant strains (18), or intranasal administration (1).
A recently appreciated aspect of infection and immune evasion is the ability of GC to invade and grow within epithelial cells (41, 46). It is known that there is a T-cell response after natural gonorrhea (16, 44). Although the relative importance of intraepithelial growth to the life cycle of GC is unknown, it could provide an important means to evade polymorphonuclear neutrophils, antibodies, and complement on mucosal surfaces. A VEE VRP-based gonococcal vaccine has many features that make it an attractive system for studies of potentially protective immune responses. The ability to generate both substantial Th1 responses (31) and mucosal responses (22) with a VRP-based vaccine might be advantageous. We successfully expressed PorB VRP in BHK cells with and without a eukaryotic secretion signal (tPA). The VRP-vaccinated groups tended toward a Th1-type response, based both on serum IgG1/IgG2a ratios and on results of IFN- ELISPOT assays. The tPA-fused antigens consistently were more highly expressed and produced more serum IgG antibodies but retained the same pattern of a Th1-type response. The basis for the greater immunogenicity of the tPA-fused antigens is unknown but presumably reflects different intracellular processing of the expressed antigen in target dendritic cells. The booster effect of the rrPorB in Ribi after PorB VRP immunization suggested that primary immunization with PorB VRPs elicited a good memory response.
Different immune responses generated by different antigen delivery methods provide a fresh opportunity to understand which regimens are protective and to dissect the mechanisms of protection. The recently developed GC infection model in female mice provides a powerful tool to evaluate whether PorB immune responses generated by different delivery systems are protective.
ACKNOWLEDGMENTS
This work was supported by grants 5 R01 AI26837 and 5 U19 AI31496 from the National Institutes of Health to P.F.S.
We thank Christopher Elkins for helpful comments on the design of our immunological assays and Annice Rountree for technical assistance. Ann Jerse provided many helpful comments. We are grateful to Yury Matsuka and Elizabeth Anderson for supplying the FA1090 rrPorB and to numerous members of Robert Johnston's lab for sharing their time, facilities, and expertise with us during the course of these experiments, especially Chad Williamson, Kevin Brown, Martha Collier, and Ande West. We acknowledge the Gonococcal Genome Sequencing Project supported by USPHS/NIH grant AI38399 and B. A. Roe, L. Song, S. P. Lin, X. Yuan, S. Clifton, Tom Ducey, Lisa Lewis, and D. W. Dyer at the University of Oklahoma.
REFERENCES
1. Bakke, H., K. Lie, I. L. Haugen, G. E. Korsvold, E. A. Hoiby, L. M. Naess, J. Holst, I. S. Aaberge, F. Oftung, and B. Haneberg. 2001. Meningococcal outer membrane vesicle vaccine given intranasally can induce immunological memory and booster responses without evidence of tolerance. Infect. Immun. 69:5010-5015.
2. Balasuriya, U. B., H. W. Heidner, J. F. Hedges, J. C. Williams, N. L. Davis, R. E. Johnston, and N. J. MacLachlan. 2000. Expression of the two major envelope proteins of equine arteritis virus as a heterodimer is necessary for induction of neutralizing antibodies in mice immunized with recombinant Venezuelan equine encephalitis virus replicon particles. J. Virol. 74:10623-10630.
3. Bauer, F. J., T. Rudel, M. Stein, and T. F. Meyer. 1999. Mutagenesis of the Neisseria gonorrhoeae porin reduces invasion in epithelial cells and enhances phagocyte responsiveness. Mol. Microbiol. 31:903-913.
4. Blake, M. S., and L. M. Wetzler. 1995. Vaccines for gonorrhea: where are we on the curve Trends Microbiol. 3:469-474.
5. Brodeur, B. R., Y. Larose, P. Tsang, J. Hamel, F. Ashton, and A. Ryan. 1985. Protection against infection with Neisseria meningitidis group B serotype 2b by passive immunization with serotype-specific monoclonal antibody. Infect. Immun. 50:510-516.
6. Bszewczyki, A., and L. M. Koloft. 1985. A method for efficient blotting of basic proteins from SDS-PAGE to nitrocellulose. Anal. Biochem. 150:403-407.
7. Buchanan, T. M., D. A. Eschenbach, J. S. Knapp, and K. K. Holmes. 1980. Gonococcal salpingitis is less likely to recur with Neisseria gonorrhoeae of the same principal outer membrane protein antigenic type. Am. J. Obstet. Gynecol. 138:978-980.
8. Carbonetti, N., V. Simnad, C. Elkins, and P. F. Sparling. 1990. Construction of isogenic gonococci with variable porin structure: effects on susceptibility to human serum and antibiotics. Mol. Microbiol. 4:1009-1018.
9. Carbonetti, N. H., V. I. Simnad, H. S. Seifert, M. So, and P. F. Sparling. 1988. Genetics of protein I of Neisseria gonorrhoeae: construction of hybrid porins. Proc. Natl. Acad. Sci. USA 85:6841-6845.
10. Centers for Disease Control and Prevention. 2002. Discontinuation of cefixime tablets—United States. Morb. Mortal. Wkly. Rep. 51:1052.
11. Cohen, M. S., J. G. Cannon, A. E. Jerse, L. M. Charniga, S. F. Isbey, and L. G. Whicker. 1994. Human experimentation with Neisseria gonorrhoeae: rationale, methods, and implications for the biology of infection and vaccine development. J. Infect. Dis. 169:532-537.
12. Davis, N. L., A. West, E. Reap, G. MacDonald, M. Collier, S. Dryga, M. Maughan, M. Connell, C. Walker, K. McGrath, C. Cecil, L. H. Ping, J. Frelinger, R. Olmsted, P. Keith, R. Swanstrom, C. Williamson, P. Johnson, D. Montefiori, and R. E. Johnston. 2002. Alphavirus replicon particles as candidate HIV vaccines. IUBMB Life 53:209-211.
13. Delvig, A. A., T. E. Michaelsen, A. Aase, E. A. Hoiby, and E. Rosenqvist. 1997. Vaccine-induced IgG antibodies to the linear epitope on the PorB outer membrane protein promote opsonophagocytosis of Neisseria meningitidis by human neutrophils. Clin. Immunol. Immunopathol. 84:27-35.
14. Elkins, C., K. B. Barkley, N. H. Carbonetti, A. J. Coimbre, and P. F. Sparling. 1994. Immunobiology of purified recombinant outer membrane porin protein I of Neisseria gonorrhoeae. Mol. Microbiol. 14:1059-1075.
15. Elkins, C., N. H. Carbonetti, V. A. Varela, D. Stirewalt, D. G. Klapper, and P. F. Sparling. 1992. Antibodies to N-terminal peptides of gonococcal porin are bactericidal when gonococcal lipopolysaccharide is not sialylated. Mol. Microbiol. 6:2617-2628.
16. Elkins, C., and P. F. Sparling. 1992. The immunobiology of Neisseria gonorrhoeae, p. 113-139. In T. C. Quinn (ed.), Sexually transmitted diseases; advances in host defense mechanisms, vol. 8. Raven Press, New York, N.Y.
17. Faruki, H., R. N. Kohmescher, W. P. McKinney, and P. F. Sparling. 1985. Community-based outbreak of infection with penicillin-resistant Neisseria gonorrhoeae not producing penicillinase (chromosomally mediated resistance). N. Engl. J. Med. 313:607-610.
18. Fisseha, M., P. Chen, B. Brandt, T. Kijek, E. Moran, and W. Zollinger. 2005. Characterization of native outer membrane vesicles from lpxL mutant strains of Neisseria meningitidis for use in parenteral vaccination. Infect. Immun. 73:4070-4080.
19. Fleming, D. T., and J. N. Wasserheit. 1999. From epidemiological synergy to public health policy and practice: the contribution of other sexually transmitted diseases to sexual transmission of HIV infection. Sex. Transm. Infect. 75:3-17.
20. Fox, K. K., J. C. Thomas, D. H. Weiner, R. H. Davis, P. F. Sparling, and M. S. Cohen. 1999. Longitudinal evaluation of serovar-specific immunity to Neisseria gonorrhoeae. Am. J. Epidemiol. 149:353-358.
21. Grieder, F. B., N. L. Davis, J. F. Aronson, P. C. Charles, D. C. Sellon, K. Suzuki, and R. E. Johnston. 1995. Specific restrictions in the progression of Venezuelan equine encephalitis virus-induced disease resulting from single amino acid changes in the glycoproteins. Virology 206:994-1006.
22. Harrington, P. R., B. Yount, R. E. Johnston, N. Davis, C. Moe, and R. S. Baric. 2002. Systemic, mucosal, and heterotypic immune induction in mice inoculated with Venezuelan equine encephalitis replicons expressing Norwalk virus-like particles. J. Virol. 76:730-742.
23. Heckels, J. E., M. Virji, K. Zak, and J. N. Fletcher. 1987. Immunobiology of gonococcal outer membrane protein I. Antonie Leeuwenhoek 53:461-464.
24. Hill, J. 1942. Experimental infection with Neisseria gonorrhoeae. Am. J. Syph. Gonorr. Vener. Dis. 27:733-771.
25. Hobbs, M. M., T. M. Alcorn, R. H. Davis, W. Fischer, J. C. Thomas, I. Martin, C. Ison, P. F. Sparling, and M. S. Cohen. 1999. Molecular typing of Neisseria gonorrhoeae causing repeated infections: evolution of porin during passage within a community. J. Infect. Dis. 179:371-381.
26. Jerse, A. E. 1999. Experimental gonococcal genital tract infection and opacity protein expression in estradiol-treated mice. Infect. Immun. 67:5699-5708.
27. Joiner, K. A., R. Scales, K. A. Warren, M. M. Frank, and P. A. Rice. 1985. Mechanism of action of blocking immunoglobulin G for Neisseria gonorrhoeae. J. Clin. Investig. 76:1765-1772.
28. Joiner, K. A., K. A. Warren, M. Tam, and M. M. Frank. 1985. Monoclonal antibodies directed against gonococcal protein I vary in bactericidal activity. J. Immunol. 134:3411-3419.
29. Kellogg, D. S., Jr., W. L. Peacock, Jr., W. E. Deacon, L. Brown, and C. I. Pirkle. 1963. Neisseria gonorrhoeae. I. Virulence genetically linked to clonal variation. J. Bacteriol. 85:1274-1279.
30. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685.
31. Lee, J. S., B. K. Dyas, S. S. Nystrom, C. M. Lind, J. F. Smith, and R. G. Ulrich. 2002. Immune protection against staphylococcal enterotoxin-induced toxic shock by vaccination with a Venezuelan equine encephalitis virus replicon. J. Infect. Dis. 185:1192-1196.
32. Liljestrom, P., and H. Garoff. 1991. A new generation of animal cell expression vectors based on the Semliki Forest virus replicon. Biotechnology (New York) 9:1356-1361.
33. Lozes, E., K. Huygen, J. Content, O. Denis, D. L. Montgomery, A. M. Yawman, P. Vandenbussche, J. P. Van Vooren, A. Drowart, J. B. Ulmer, and M. A. Liu. 1997. Immunogenicity and efficacy of a tuberculosis DNA vaccine encoding the components of the secreted antigen 85 complex. Vaccine 15:830-833.
34. MacDonald, G. H., and R. E. Johnston. 2000. Role of dendritic cell targeting in Venezuelan equine encephalitis virus pathogenesis. J. Virol. 74:914-922.
35. Matsuka, Y. V., D. A. Dilts, S. Hoiseth, and R. Arumugham. 1998. Characterization of a subunit structure and stability of the recombinant porin from Neisseria gonorrhoeae. J. Protein Chem. 17:719-728.
36. McDade, R. L., Jr., and K. H. Johnston. 1980. Characterization of serologically dominant outer membrane proteins of Neisseria gonorrhoeae. J. Bacteriol. 141:1183-1191.
37. Nachamkin, I., J. G. Cannon, and R. S. Mittler. 1981. Monoclonal antibodies against Neisseria gonorrhoeae: production of antibodies directed against a strain-specific cell surface antigen. Infect. Immun. 32:641-648.
38. Ober, W. B. 1969. Boswell's gonorrhea. N. Y. Acad. Med. 45:587-636.
39. Plante, M., A. Jerse, J. Hamel, F. Couture, C. R. Rioux, B. R. Brodeur, and D. Martin. 2000. Intranasal immunization with gonococcal outer membrane preparations reduces the duration of vaginal colonization of mice by Neisseria gonorrhoeae. J. Infect. Dis. 182:848-855.
40. Plummer, F. A., J. N. Simonsen, H. Chubb, L. Slaney, J. Kimata, M. Bosire, J. O. Ndinya-Achola, and E. N. Ngugi. 1989. Epidemiologic evidence for the development of serovar-specific immunity after gonococcal infection. J. Clin. Investig. 83:1472-1476.
41. Post, D. M., N. J. Phillips, J. Q. Shao, D. D. Entz, B. W. Gibson, and M. A. Apicella. 2002. Intracellular survival of Neisseria gonorrhoeae in male urethral epithelial cells: importance of a hexaacyl lipid A. Infect. Immun. 70:909-920.
42. Pushko, P., M. Parker, G. V. Ludwig, N. L. Davis, R. E. Johnston, and J. F. Smith. 1997. Replicon-helper systems from attenuated Venezuelan equine encephalitis virus: expression of heterologous genes in vitro and immunization against heterologous pathogens in vivo. Virology 239:389-401.
43. Quakyi, E. K., H. D. Hochstein, and C. M. Tsai. 1997. Modulation of the biological activities of meningococcal endotoxins by association with outer membrane proteins is not inevitably linked to toxicity. Infect. Immun. 65:1972-1979.
44. Simpson, S. D., Y. Ho, P. A. Rice, and L. M. Wetzler. 1999. T lymphocyte response to Neisseria gonorrhoeae porin in individuals with mucosal gonococcal infections. J. Infect. Dis. 180:762-773.
45. Tam, M. R., T. M. Buchanan, E. C. Sandstrom, K. K. Holmes, J. S. Knapp, A. W. Siadak, and R. C. Nowinski. 1982. Serological classification of Neisseria gonorrhoeae with monoclonal antibodies. Infect. Immun. 36:1042-1053.
46. van Putten, J. P., T. D. Duensing, and J. Carlson. 1998. Gonococcal invasion of epithelial cells driven by P.IA, a bacterial ion channel with GTP binding properties. J. Exp. Med. 188:941-952.
47. Virji, M., J. N. Fletcher, K. Zak, and J. E. Heckels. 1987. The potential protective effect of monoclonal antibodies to gonococcal outer membrane protein IA. J. Gen. Microbiol. 133:2639-2646.
48. Virji, M., K. Zak, and J. E. Heckels. 1986. Monoclonal antibodies to gonococcal outer membrane protein IB: use in investigation of the potential protective effect of antibodies directed against conserved and type-specific epitopes. J. Gen. Microbiol. 132:1621-1629.
49. Zhu, W., C. E. Thomas, and P. F. Sparling. 2004. DNA immunization of mice with a plasmid encoding Neisseria gonorrhoeae PorB protein by intramuscular injection and epidermal particle bombardment. Vaccine 22:660-669.(Weiyan Zhu, Christopher E)