JOURNAL OF VIROLOGY, Feb. 1999, p. 930–938 0022-538X/99/$04.0010 Copyright © 1999, American Society for Microbiology. All Rights Reserved.
Vol. 73, No. 2
Chimeric Plant Virus Particles Administered Nasally or Orally Induce Systemic and Mucosal Immune Responses in Mice FRANK R. BRENNAN,1 TREVOR BELLABY,1 SHARON M. HELLIWELL,1 TIM D. JONES,1 SØREN KAMSTRUP,2 KRISTIAN DALSGAARD,2 JAN-INGMAR FLOCK,3 1 AND WILLIAM D. O. HAMILTON * Axis Genetics plc, Babraham, Cambridge CB2 4AZ, United Kingdom1; Danish Veterinary Institute for Virus Research, Lindholm, DK-4771 Kalvehave, Denmark2; and Department of Immunology, Microbiology, Pathology, and Infectious Diseases, Huddinge University Hospital, S-141 86 Huddinge, Sweden3 Received 30 July 1998/Accepted 20 October 1998
The humoral immune responses to the D2 peptide of fibronectin-binding protein B (FnBP) of Staphylococcus aureus, expressed on the plant virus cowpea mosaic virus (CPMV), were evaluated after mucosal delivery to mice. Intranasal immunization of these chimeric virus particles (CVPs), either alone or in the presence of ISCOM matrix, primed CPMV-specific T cells and generated high titers of CPMV- and FnBP-specific immunoglobulin G (IgG) in sera. Furthermore, CPMV- and FnBP-specific IgA and IgG could also be detected in the bronchial, intestinal, and vaginal lavage fluids, highlighting the ability of CVPs to generate antibody at distant mucosal sites. IgG2a and IgG2b were the dominant IgG subclasses in sera to both CPMV and FnBP, demonstrating a bias in the response toward the T helper 1 type. The sera completely inhibited the binding of human fibronectin to the S. aureus FnBP. Oral immunization of the CVPs also generated CPMV- and FnBP-specific serum IgG; however, these titers were significantly lower and more variable than those generated by the intranasal route, and FnBP-specific intestinal IgA was undetectable. Neither the ISCOM matrix nor cholera toxin enhanced these responses. These studies demonstrate for the first time that recombinant plant viruses have potential as mucosal vaccines without the requirement for adjuvant and that the nasal route is most effective for the delivery of these nonreplicating particles. derived from the VP2 protein of canine parvovirus (CPV) expressed on CPMV is immunogenic when administered to mink and subsequently protected the mink from a lethal challenge with the CPV-related mink enteritis virus (10). Most infectious viral and bacterial diseases involve colonization or invasion through mucosal surfaces by the pathogen, and hence it is important to develop vaccines that induce strong protective mucosal immune responses as a first line of defense. Where the organism, such as Vibrio cholerae and enterotoxigenic Escherichia coli, is restricted to the mucosa, strong mucosal immunity is often sufficient. However, when the organism disseminates from the mucosa into the bloodstream, a strong systemic response is also required to engender sterile immunity. Hence, the ideal mucosal vaccine should generate local immune responses at mucosal surfaces but also elicit generalized vaccine-specific immunity in the systemic lymphoid organs. The potential of CPMV-based vaccines for mucosal vaccination has not previously been determined. Oral immunization with particulate antigens, especially when presented as viable organisms, which can colonize the mucosa better than killed organisms, is effective at inducing local and generalized secretory and systemic immune responses (5, 43). However, the acidic pH and the presence of degradative enzymes in the gastrointestinal tract mean that when nonreplicating antigens are used, high concentrations are often required to elicit high levels of immunity (6). Another way to elicit mucosal immunity but circumvent the problems of oral immunization is to vaccinate via the intranasal route (2). Intranasal immunization requires up to 10-fold less immunogen for effective immunization and avoids the problems of low pH. Live vaccines (15, 37), virus-like particles (4, 25, 32), and synthetic peptides (17, 33, 44) in the absence of adjuvant have been shown to stimulate strong immunity when administered
Replicating vaccines such as live-attenuated bacterial (13) and virus (36, 40, 45) vaccines, as well as naked DNA vaccines (31), induce stronger and longer-lasting immune responses than conventional killed-subunit vaccines and also elicit protective cell-mediated immunity, often without the need for adjuvant. There are however, safety concerns over the use of these vaccines (24, 49), where persistence or reversion to virulence of the live vaccine strains and integration of the naked DNA vaccine into the host chromosome are of major concern. Recent technological advances, such as the use of more-effective adjuvants for both mucosal and systemic delivery (12, 16), liposome and ISCOM encapsulation of proteins and peptides (3, 19, 27), multiple antigenic peptides (35), and virus-like particles (VLPs) (1), have led to the development of moreeffective subunit vaccines. To circumvent the safety concerns of replicating vaccines and to avoid the need for peptide synthesis and chemical coupling to a carrier such as keyhole limpet hemocyanin, we have been examining the utility of the plant virus cowpea mosaic virus (CPMV) as a carrier of peptides for immune recognition. CPMV is composed of 2 subunits, the small (S) and large (L) coat proteins, of which there are 60 copies of each per virus particle (46). Foreign peptides up to 37 amino acids in length can be expressed on either the L or S proteins; hence, 60 to 120 copies of a peptide can be displayed on a single virus particle (4b, 34). A peptide from the human immunodeficiency virus (HIV) gp41 glycoprotein is highly immunogenic when displayed on CPMV, eliciting high titers of HIV neutralizing antibodies (28, 29). Furthermore, a peptide
* Corresponding author. Mailing address: Axis Genetics plc, Babraham, Cambridge CB2 4AZ, United Kingdom. Phone: 44-1223-837-611. Fax: 44-1223-837-604. E-mail:
[email protected]. 930
VOL. 73, 1999
MUCOSAL IMMUNIZATION WITH A CHIMERIC PLANT VIRUS
by this route. Furthermore, stimulation of the nasal mucosa, like stimulation of the intestinal mucosa, has been shown to be effective at generating protective immunity at distant mucosal sites (reviewed in reference 2). To assess the potential of CVPs as mucosal vaccines, mice were immunized intranasally or orally with CPMV expressing a peptide derived from the fibronectin-binding protein B (FnBP) D2 motif of Staphylococcus aureus (14, 42). The three fibronectin-binding domains, termed D1, D2, and D3, of FnBP have been shown to be immunogenic in mice and rats (7, 41). The CVPs were shown to be more immunogenic when administered (without adjuvant) via the intranasal route than when administered by the oral route, generating high titers of D2specific antibody in serum and mucosa, and the serum antibody inhibited fibronectin binding to FnBP. MATERIALS AND METHODS Experimental animals. Female C57BL/6 mice (H-2b) mice aged 6 to 8 weeks at the start of the experiment were obtained from Harlan-Olac, Bicester, United Kingdom, and were housed at the Department of Pathology, University of Cambridge. All procedures were performed according to the United Kingdom Home Office guidelines for animals in medical research. Construction, propagation, and purification of CPMV-MAST1 virions. The genome of CPMV consists of two molecules of single-stranded plus-sense RNA, both of which have been cloned on separate plasmids as full-length cDNAs termed pCP1 and pCP2 (10). The construction of vector pCP2-0.51 has been previously described (10). pCP2-0.51 was digested with NheI and AatII, and the excised fragment was replaced by overlapping oligonucleotides coding for the first 30 amino acids of the D2 domain of FnBP-B (1GQNNGNQSFEEDTEKD KPKYEQGGNIIDID30) and CPMV flanking sequences such that the epitope was inserted between amino acids 22 and 23 of the S coat protein (26). The resultant plasmid was termed pCP2-MAST1. Plasmid pCP2-MAST1 was linearized and mixed with linearized pCP1 and inoculated onto young cowpea plants as described earlier (11). Particles were purified from the first inoculation by PEG precipitation and differential centrifugation of leaf homogenate as described earlier (10) and then passaged a second time to produce virus working stocks. Working stocks were purified and characterized by reverse transcriptase PCR, sequencing, and sodium dodecyl sulfate-polyacrylamide gel electrophoresis and then quantified, also as described previously (10). Working-stock yields were 1.2 mg/g of leaf tissue. Each virion expresses 60 copies of the D2 peptide, and 1 mg of CPMV-MAST1 contains approximately 40 ng of D2 peptide. The presence of a functional D2 epitope on the surface of CPMV-MAST1 was verified by enzyme-linked immunosorbent assay (ELISA) by using a D2-specific monoclonal antibody, V6F2 (a gift from Aquila Biopharmaceuticals, Worcester, Mass.). Immunization of mice with CVPs. For both intranasal and oral delivery of CVPs, mice (n 5 5 per group) received four immunizations on days 0, 7, 14, and 21. For each intranasal immunization, mice were lightly anesthetized with halothane and then given 100 mg of CPMV-MAST1 or CPMV expressing an unrelated peptide (control CVP) either alone or with 10 mg of ISCOM matrix (prepared as described previously [9]) in a total volume of 20 ml. Mice were immunized orally with the same amounts of CVPs and adjuvant in a total volume of 100 ml by using a gavage needle. For parenteral delivery of the CVPs, mice (n 5 5) were immunized subcutaneously with 10 mg of CPMV-MAST1 together with 10 mg of QS-21 on days 0 and 21. Blood was collected by tail bleeding or after exsanguination, and sera were collected and stored at 220°C. Collection of bronchial, intestinal, and vaginal lavage fluids. Lavage fluids were collected from mice as described previously (5). Briefly, mice were culled by exsanguination and the lungs, intestines, and vaginas were washed out with 0.5, 3, and 0.05 ml, respectively, of ice-cold 50 mM EDTA containing soybean trypsin inhibitor. The lavage fluids were centrifuged at 13,000 3 g for 10 min to remove large debris, and then 10 ml of 0.2 M phenylmethylsulfonyl fluoride in ethanol (95% [vol/vol]) and 10 ml of sodium azide (2% [wt/vol]) per ml were added to the clarified supernatant. Fetal calf serum (Gibco) was added at 3%, and the samples were stored at 280°C. ELISA for detection of specific serum and mucosal antibody. To detect anti-D2 peptide antibody, the wells of 96-well ELISA plates (Dynatech Immunol-4) were coated for 1 h at 37°C at 0.5 mg/well with a glutathione S-transferase (GST) fusion protein incorporating the D1, D2, and D3 peptides of FnBP (GSTD1-3). The GSTD1-3 construct was made by introducing a DNA fragment from nucleotide 2352 to nucleotide 2647 (42) of the Fnb gene, after appropriate modifications in the ends, into a pGEX plasmid (Pharmacia Biotech, Uppsala, Sweden) encoding GST. The fragment encodes the D1, D2, and most of the D3 peptide of the Fnb gene. Purification of the GSTD1-3 fusion protein was carried out by using glutathione affinity chromatography according to the recommendations of the supplier of the system. For the detection of anti-CPMV antibody, the wells were coated at 0.1 mg/well with CPMV for 1 h at 37°C. A series of doubling dilutions of serum and lavage, starting at 1:100 and 1:4, respectively, were
931
incubated on the antigen-coated plates for 1 and 4 h, respectively, at 37°C. Bound serum immunoglobulin G (IgG) was detected either with a 1:1 mixture of alkaline phosphatase (AP)-conjugated goat anti-mouse IgG1 and IgG2a or with AP-conjugated goat anti-mouse IgG1, IgG2a, IgG2b, or IgG3 conjugates tested individually. Bound IgA in lavages was detected with AP-conjugated goat antimouse IgA (all conjugates from Southern Biotechnologies, Inc.) and with pnitrophenyl phosphate (Sigma) as the substrate. The products were quantified with an automated ELISA reader at 405 nm. For CPMV-specific titers, the results are expressed as the endpoint titer, which was calculated as the inverse of the dilution that gave a mean optical density at 405 nm (OD405) higher than the mean OD405 obtained with a 1:50 dilution of pooled sera (OD405 range, 0.098 to 0.143) or a 1:2 dilution of pooled lavage fluid (OD405 range, 0.074 to 0.092 [for bronchial and vaginal lavage fluids]; OD405 range, 0.120 to 0.160 [for intestinal lavage fluids]) from unimmunized mice (n 5 6). For FnBP-specific titers, endpoint titers are the inverse of the dilution that gave a mean OD405 higher than the mean OD405 obtained with a 1:50 dilution of pooled sera (OD405 range, 0.090 to 0.146) and a 1:2 dilution of pooled lavage fluid (OD405 range, 0.074 to 0.080 [for bronchial and vaginal lavage fluids]; OD405 range, 0.140 to 0.189 [for intestinal lavage fluids]) from control CVP-immunized mice (n 5 6). Removal of fibronectin from mouse sera. IgG was purified from the pooled sera of mice immunized intranasally with CPMV-MAST1 by using immobilized protein A (Affinitypak; Pierce, Rockford, Ill.) according to the manufacturer’s instructions. As a positive control, IgG was purified from the sera of mice immunized subcutaneously with CPMV-MAST1 in QS-21 (day 42 sera). As negative controls, IgG was purified from the sera of mice immunized subcutaneously with CPMV expressing an unrelated peptide in QS-21 and from normal mouse sera. Purification of the IgG was done to remove the serum fibronectin that causes interference in the fibronectin-binding assays (see below). For each group of mice, IgG was purified from 1 ml of pooled sera. Each IgG sample was then diluted until the A280 values for all of the samples were identical and hence contained the same concentration of IgG. No changes in the ratios of the isotypes of FnBP-specific IgG were observed as a result of the purification procedure. Inhibition of fibronectin binding to GSTD1-3. This assay was performed as described previously (7). Briefly, plates were coated with a 1:250 dilution of GSTD1-3 (0.1 mg/well in 100 ml) for 1 h at 37°C, washed, and then blocked with 100 ml of 2% bovine serum albumin per well. After the washing, serial twofold dilutions of mouse IgG were added to all the wells (50 ml/well) and incubated for 1 h at 37°C. Human plasma fibronectin (Sigma) was then added to the wells (50 ml/well) at a final concentration of 5 mg/ml and incubated for 1 h at 37°C. After the washing, rabbit anti-human fibronectin antibodies conjugated to HRP (1: 4,000; Dako, Ltd., Ely, United Kingdom) were added and, after 1 h at 37°C, o-phenylenediamine dihydrochloride substrate was added. After incubation at 37°C for 30 min, the reaction was stopped by the addition at 50 ml/well of 2.5 M H2SO4, and the A492 value was determined. T-cell proliferation assays. Spleens from mice immunized either intranasally or orally with CPMV-MAST1 were removed, and single cell suspensions were made. After the washing and lysis of erythrocytes with ice-cold 0.85% NH4Cl, the splenocytes from five mice were pooled and dispensed into round-bottom 96-well plates (2 3 105/100 ml; Nunclon Delta Surface; Nunc, Roskilde, Denmark). Cells were with cultured with concanavalin A (ConA; 2.5 mg/ml; Sigma) or with wild-type CPMV (50 and 5 mg/ml) for 5 days at 37°C in 5% CO2 with the addition of 0.5 mCi/well of [methyl-3H]thymidine (Amersham Life Science, Little Chalfont, United Kingdom) in the last 6 h of culture. Cells were harvested onto filter mats (Wallac Oy, Turku, Finland), and the incorporation of label was measured by using a 1450 Microbeta Trilux liquid scintillation and luminescence counter (Wallac). Data are expressed both as mean counts per minute (cpm) and as a stimulation index (SI), where the mean cpm in the presence of antigen were divided by the mean cpm measured in the absence of antigen (medium only). A value of 3 or greater was considered significant. Statistics. Differences between the groups were evaluated by using the Student t test, where P , 0.05 was considered statistically significant.
RESULTS Intranasal immunization with CPMV-MAST1 without adjuvant induces FnBP D2-specific antibody in serum, lung, intestine, and vagina. Preliminary studies have shown that cholera toxin (CT) fails to enhance the immunogenicity of a number of CVPs after intranasal delivery (4a); hence, CT was not used for the intranasal delivery of CPMV-MAST1. Mice intranasally immunized with four doses of 100 mg of CPMVMAST1 without adjuvant elicited high titers of CPMV-specific IgG in sera (all mice had titers of .512,000) and IgA in bronchial, intestinal, and vaginal lavage fluids (Fig. 1A to D) when examined on day 50. Furthermore, we also detected high levels of FnBP D2-specific IgG in sera (Fig. 1A) and D2specific IgA in bronchial (Fig. 1B), intestinal (four of five mice)
932
BRENNAN ET AL.
J. VIROL.
FIG. 1. Intranasal immunization with CPMV-MAST1 without adjuvant induces FnBP-specific IgG in serum and variable levels of IgA in mucosa. Mice were immunized intranasally with 100 mg of CPMV-MAST1 without adjuvant on days 0, 7, 14, and 21. Sera (A) and bronchial (B), intestinal (C), and vaginal (D) lavage fluids were collected on day 50 and tested for GSTD1-3- and CPMV-specific IgG1 and IgG2a (a mixture of the two conjugates) (sera) and IgA (lavage fluids) by ELISA. Titers are expressed as endpoint titers obtained with the sera or lavage fluids from individual mice.
(Fig. 1C), and vaginal lavage fluids (three of five mice) (Fig. 1D). CPMV- and FnBP-specific IgG was also detectable in bronchial and intestinal lavage fluids and mirrored the IgA responses in individual mice (results not shown). When the time course of the serum responses to the D2 peptide were examined, it was found that variable titers were detected in all mice as early as day 14, rising by days 21 and 28, and reaching high levels on days 42 and 50 (Fig. 2). No FnBP-specific antibody was detected in the sera or lavage fluids of wild-type CPMV-immunized mice at any time (see the OD ranges given in Materials and Methods). The ISCOM matrix fails to enhance the systemic and mucosal responses to CPMV-MAST1 after intranasal delivery. Since CT failed to enhance the intranasal responses to CPMVMAST1, an adjuvant was sought that could improve the mu-
cosal delivery of the CVPs. Mice were immunized with four doses of 100 mg of CPMV-MAST1 together with 10 mg of ISCOM matrix as adjuvant. On day 50 after immunization (Fig. 3A to D), high titers of CPMV- and FnBP-specific IgG were detected in sera (Fig. 3A) and high titers of CPMV- and FnBP-specific IgA were detected in the bronchial (Fig. 3B), intestinal (Fig. 3C), and vaginal (Fig. 3D) lavage fluids of all the mice. However, the titers achieved by using the ISCOM matrix were not significantly greater than those observed in the absence of adjuvant (Fig. 1). Indeed, mean titers generated to CPMV-MAST1 with ISCOM matrix were significantly lower, particularly in sera (see Table 1 and Fig. 2) and in bronchial lavage fluids (P , 0.05). Mean titers in the vaginal lavage fluids, however, were somewhat higher, but these differences were not statistically significant.
VOL. 73, 1999
MUCOSAL IMMUNIZATION WITH A CHIMERIC PLANT VIRUS
933
FIG. 2. Time course of the serum antibody responses after intranasal immunization. Mice were immunized intranasally with 100 mg of CPMV-MAST1 either alone or with the ISCOM matrix on days 0, 7, 14, and 21. Sera were collected on days 14, 21, 28, 42, and 50 and were tested for GSTD1-3-specific IgG1 and IgG2a (a mixture of the two conjugates) by ELISA. Titers are expressed as mean endpoint titers 6 the standard deviation (SD) for each time point for the two groups.
The IgG in the sera of mice immunized intranasally with CPMV-MAST1 is predominantly of the IgG2a and IgG2b subclasses. Intranasal immunization of mice with CPMV-MAST1 either alone (Fig. 4A) or with the ISCOM matrix (Fig. 4B) induces a highly polarized T-helper type 1 (Th1)-type antibody response to FnBP D2, since all of the mice produced high titers of FNBP D2-specific IgG2a and IgG2b. Levels of FnBP-specific IgG1 and IgG3 were generally much lower or absent, although two mice given CPMV-MAST1 in ISCOMs produced significant quantities of IgG1 (Fig. 4B). This Th1 bias in the response to CPMV-MAST1 was not merely due to the intranasal route of delivery of the particles, since subcutaneous immunization of mice with CPMV-MAST1 in QS-21 adjuvant also led to the generation of predominantly FnBP-specific IgG2a and IgG2b (Fig. 4C). The response to the CPMV carrier after both intranasal and subcutaneous immunization was also predominantly of the IgG2a and IgG2b isotypes (results not shown). IgG in the sera of mice immunized intranasally with CPMVMAST1 inhibits the binding of fibronectin to immobilized FnBP. IgG purified from the sera of mice immunized intranasally with CPMV-MAST1 (collected 50 days after primary immunization) was shown to strongly inhibit the binding of 5 mg of human fibronectin per ml to GSTD1-3 (Fig. 5). The inhibition, however, was slightly less than that achieved with IgG from mice immunized subcutaneously with CPMV-MAST1 in QS-21 (Fig. 5). No inhibition of fibronectin binding was achieved with IgG purified from the sera of mice immunized subcutaneously (in QS-21) with CPMV expressing an unrelated peptide or with IgG purified from normal mouse sera (Fig. 5). Oral immunization with CPMV-MAST1 with or without adjuvant induces FnBP D2-specific antibody in serum but not in intestine. Mice orally immunized with four doses of 100 mg of CPMV-MAST1 without adjuvant elicited CPMV-specific IgG in sera and IgA in intestinal lavages when examined on day 42 (Fig. 6). However, these titers were significantly lower (P , 0.001) than those achieved after intranasal delivery (Table 1). FnBP-specific serum IgG was detected in those mice which had generated high levels of CPMV-specific serum IgG but was low in those mice with low CPMV-specific titers (Fig. 6A). FnBPspecific IgA was not detected in the intestinal lavages of any of
the orally immunized mice (Fig. 6B). The serum antibody responses to FnBP after oral immunization could not be enhanced by either CT or ISCOM matrix (Table 1), and FnBPspecific antibody responses were absent (results not shown). Strong CPMV-specific T cell responses are generated after nasal but not after oral immunization with CPMV-MAST1. CPMV provides T-cell epitopes for the generation of B-cell responses to the expressed peptide; hence, it is important to determine the levels of CPMV-specific T-cell priming after both nasal and oral delivery of CVPs. Spleen cells from mice immunized intranasally with CPMV-MAST1 without adjuvant were shown to proliferate strongly in response to CPMV in vitro (Table 2). By comparison, spleen cells from mice orally immunized with CPMV-MAST1 without adjuvant proliferated weakly in response to CPMV; however, these responses were significantly (P , 0.05) greater than those of cells from unimmunized mice (Table 2). DISCUSSION Although the nasal cavity is already established as a potential alternative to the parenteral route for administering peptide drugs, the full potential of both the nasal and pulmonary routes in the administration of immunogens has not been exploited. Nasal administration of antigens could produce effective vaccines for the protection of both the upper respiratory tract and other distant mucosal sites, as well as systemic immunity (2, 23, 38, 39, 43, 44). To investigate the feasibility of using a plant virus as a carrier of immunogenic peptides to the mucosal immune system, we immunized mice intranasally or orally with CPMV expressing a peptide derived from the FnBP of S. aureus. Four intranasal doses of CPMV-MAST1 without adjuvant and containing a total of 16 mg of peptide were sufficient to prime carrier CPMV-specific T cells and to induce high titers of both CPMV-specific and FnBP D2-specific IgG in serum and IgA in the lung, intestine, and vagina. The IgA in the lavage was most likely secretory IgA generated locally, as levels of peptide-specific serum IgA were very low (mean titer, ,100 6 24.3), ruling out the possibility that the IgA detected in the lavages was derived from serum. Thus, it appears that CVPs can effectively target the nasal-tract-associated lymphoid
934
BRENNAN ET AL.
J. VIROL.
FIG. 3. Intranasal immunization with CPMV-MAST1 with the ISCOM matrix induces FnBP-specific IgG in serum and FnBB-specific IgA in mucosa. Mice were immunized intranasally with 100 mg of CPMV-MAST1 with the ISCOM matrix (10 mg/dose) on days 0, 7, 14, and 21. Sera (A) and bronchial (B), intestinal (C), and vaginal (D) lavage fluids were collected on day 50 and were tested for GSTD1-3- and CPMV-specific IgG1 and IgG2a (a mixture of the two conjugates) (sera) and IgA (lavage fluids) by ELISA. Titers are expressed as endpoint titers obtained with the sera or lavage fluids from individual mice.
tissue (NALT) and generate specific immune responses without the requirement for the targeting and adjuvant properties of CT. By comparison, however, oral immunization with CPMVMAST1 proved to be far less effective than the intranasal route at generating mucosal and systemic humoral immune responses to either the virus carrier or the D2 peptide. These low-level antibody responses were reflected in the lower level of priming of CPMV-specific systemic (splenic) T cells after oral immunization compared to that achieved by intranasal immunization. The T-cell proliferation to CPMV by spleen cells from unimmunized mice was most likely nonspecific, since no gamma interferon or interleukin-4 was detected in the cell supernatants, whereas high levels of gamma interferon are routinely produced by CPMV-stimulated cells from CVP-immu-
nized mice (4a). Despite the low level of the T-cell responses to the carrier, some orally immunized mice did produce moderate titers of both CPMV-specific and FnBP-specific serum IgG, but FnBP-specific intestinal IgA was undetectable. Other mice produced only low levels of CPMV-specific serum IgG, and consequently FnBP-specific serum IgG was low or absent in these mice. Thus, intranasal immunization is more effective and consistent than the oral route at generating specific IgA in the intestine, even though the oral route may be expected to deliver antigen directly to this site. Intranasal immunization with nonreplicating rotavirus VLPs, which are in some ways analogous to CVPs, is also more effective than oral immunization (32). It is possible that the poor and variable responses achieved after oral immunization are a result of degradation of the CVPs following passage through the stomach. Variable and
VOL. 73, 1999
MUCOSAL IMMUNIZATION WITH A CHIMERIC PLANT VIRUS
935
FIG. 4. Intranasal and parenteral immunization with CPMV-MAST1 induces FnBP-specific IgG predominantly of the IgG2a and IgG2b subclasses. Mice were immunized intranasally with 100 mg of CPMV-MAST1 either alone (A) or with the ISCOM matrix on days 0, 7, 14, and 21 (B). (C) Mice were immunized subcutaneously with 10 mg of CPMV-MAST1 in QS-21 on days 0 and 21. Sera were collected on day 50 (intranasal) or day 42 (subcutaneous) and tested for GSTD1-3-specific IgG1, IgG2a, IgG2b, and IgG3 by ELISA. Titers are expressed as endpoint titers obtained with sera from individual mice.
low responses have also been reported after oral immunization with Norwalk virus VLPs (4), and the oral administration of chimeric alfalfa mosaic virus expressing rabies virus epitopes elicited only very weak serum and intestinal responses to both the virus particle and the expressed peptide (30). In the latter case, these low titers were achieved despite the administration of a total of 1 mg of purified virus, which is a dose 2.5 times greater than we administered in the present study. Preliminary studies had shown that CT did not enhance the responses to CVPs after intranasal or oral delivery. CT, however, has been shown to enhance the titers to both rotavirus (32) and Norwalk virus (4) VLPs to various degrees, although, as with CVPs, CT was not a requirement for the induction of these responses. We therefore looked for another mucosally active adjuvant that might enhance the responses to the CVPs. The ISCOM matrix is a suspension of cage-like structures formed when the saponin adjuvant QuilA is mixed with phospholipids and cholesterol (9). Antigens incorporated within the ISCOM matrix forming ISCOMs have been shown to be effective immunogens when administered both parenterally and intranasally (9, 27). The ISCOM matrix has been shown to enhance the immunogenicity of orally administered inactivated rotavirus (49) and the immunogenicity of recombinant CT-B delivered intranasally (12a). Since CVPs are too large to be incorporated within ISCOMs, the CVPs were mixed with the
ISCOM matrix and administered intranasally and orally in an attempt to augment the peptide-specific responses. The ISCOM matrix failed to enhance the responses achieved with CPMV-MAST1 administered alone when given by the intranasal or oral route. We are currently examining alternative approaches to enhance the immunogenicity of oral CVP after delivery. When the isotype of the anti-D2 IgG from mice immunized intranasally with CPMV-MAST1 was examined, either alone or with the adjuvant ISCOM matrix, it was found to be predominantly of the IgG2a and IgG2b subclasses. It has been shown that the route of immunization can affect the isotypes of peptide-specific IgG that are generated (18). However, the Th1 bias in the FnBP-specific responses was not a result of the intranasal route of administration, since subcutaneous immunization with CPMV-MAST1 in QS-21, an adjuvant thought to favor Th2-type responses (8, 20), also elicited a Th1-type response. This strong polarization toward a Th1 response to D2 in mice may be a consequence of the CPMV carrier eliciting predominantly CPMV-specific Th1-type responses which also govern the response to the expressed peptide. The Th1 responses elicited by CPMV-MAST1 are of major consequence for vaccine design, since IgG2a and IgG2b are the major IgG isotypes in mice that facilitate both antibody-dependent pha-
936
BRENNAN ET AL.
J. VIROL.
FIG. 5. Intranasal immunization with CPMV-MAST1 without adjuvant elicits FnBP-specific antibody that inhibits binding of fibronectin to GSTD1-3. Mice were immunized intranasally (i.n.) with 100 mg of CPMV-MAST1 on days 0, 7, 14, and 21. Serum was collected on day 50, and the IgG fraction was tested for the ability of the GSTD1-3-specific antibody to inhibit the binding of 5 mg of human plasma fibronectin per ml to immobilized GSTD1-3. Sera from mice taken 42 days after subcutaneous (s.c.) immunization with 100 mg of CPMV-MAST1 in QS-21 on days 0, 14, and 28 were tested for comparison. The IgG from mice immunized with CPMV expressing an unrelated peptide (CVP control) and normal mouse IgG were tested as negative controls. Results are expressed as the percent inhibition of binding achieved with the different IgG samples compared to that in the absence of IgG (0%).
gocytosis (21) and complement fixation leading to complement-mediated phagocytosis and potentially bacteriolysis (22). IgG purified from the sera of mice immunized intranasally with CPMV-MAST1 completely inhibited the binding of human fibronectin to the S. aureus FnBP. However, the blocking titers were lower than those achieved with IgG from mice subcutaneously immunized with CPMV-MAST1 in QS-21. Hence, although intranasal immunization with CVPs is extremely effective at generating peptide-specific serum antibody, subcutaneous vaccination with adjuvant may elicit a higher titer of blocking antibody. Since it is unlikely that murine epithelial cells possess receptors for CPMV, presumably the size and particulate nature of the CVPs allow them to be efficiently taken up and retained by M cells in the NALT which present them to the lymphoid cells in the underlying tissue. The ability to stimulate strong mucosal and systemic immune responses after intranasal delivery is not an intrinsic quality of plant viruses. The rod-shaped potato virus X expressing the same D2 peptide generates strong FnBPspecific responses after subcutaneous immunization (4b), but it is poorly immunogenic after intranasal delivery (4a). This sug-
TABLE 1. Comparison of the intranasal and oral routes for the generation of D2-specific serum antibody to CPMV-MAST1a Route
Nasal Oral
FIG. 6. Oral immunization with CPMV-MAST1 without adjuvant induces FNBP-specific antibody in serum but not in mucosa. Mice were immunized orally with 100 mg of CPMV-MAST1 without adjuvant on days 0, 7, 14, and 21. Sera (A) and intestinal lavage fluids (B) were collected on day 42 and tested for GSTD1-3- and CPMV-specific IgG1 and IgG2a (a mixture of the two conjugates) (sera) and IgA (lavage fluids) by ELISA. Titers are expressed as endpoint titers obtained with the sera or lavage fluids from individual mice.
Mean endpoint titer 6 SD with: No adjuvant
ISCOM matrix b,c
81,920 6 28,043 9,200 6 10,460
CT b
37,120 6 20,439 6,320 6 10,831
NT 7,800 6 10,407
a Mice were immunized intranasally or orally with 100 mg of CPMV-MAST1 either without or with the ISCOM matrix and CT (oral route only) on days 0, 7, 14, and 21. On day 42 (oral) and day 49 (nasal) sera were collected and tested for GSTD1-3 IgG1 and IgG2a (a mixture of the two conjugates) by ELISA. Titers are expressed as the mean endpoint titers 6 the SD obtained with sera from the mice in each group. NT, not tested. b Comparison between nasal and oral administrations for each group. P , 0.001. c Comparison between nasal administration, with or without adjuvant. P , 0.05.
VOL. 73, 1999
MUCOSAL IMMUNIZATION WITH A CHIMERIC PLANT VIRUS
937
TABLE 2. Nasal immunization of CPMV-MAST1 without adjuvant better primes carrier CPMV-specific T cells than oral immunizationa Mean cpm 6 SD (SI) with: Group (route)
CPMV
CPMV-MAST1 (intranasal) CPMV-MAST1 (oral) Unimmunized
50 mg/ml
5 mg/ml
8,288 6 730b (22.4) 3,156 6 287c (10.3) 2,079 6 296 (7.4)
5,291 6 719b (14.3) 2,203 6 321c (6.8) 1,208 6 291 (4.3)
ConA (2.5 mg/ml)
19,462 6 3,177 (52.6) 17,859 6 2,965 (48.5) 14,500 6 3,405 (51.3)
a Mice were immunized intranasally or orally with 100 mg of CPMV-MAST1 without adjuvant on days 0, 7, 14, and 21. On day 42 (oral) and day 49 (nasal) spleen cells were removed and cultured either alone (medium), with 2.5 mg of ConA per ml, or with 50 or 5 mg of CPMV per ml for 5 days. Spleen cells from unimmunized mice were tested as a control for nonspecific proliferation. Incorporation of [3H]thymidine was measured in the last 6 h of culture. b P , 0.001. c P , 0.05.
gests that the icosahedral structure of CPMV is important in facilitating mucosal uptake and immune recognition. The fourdose regime totaling 400 mg of CVP represents only 16 mg of peptide, which was sufficient to elicit peptide-specific serum and mucosal antibody in the absence of adjuvant. In previous studies much higher doses of soluble peptide were required to generate effective peptide-specific responses after intranasal vaccination (17, 33, 44). In these studies, 30 to 200 mg of antigen administered nasally in the presence of CT or CT-B was required to elicit significant quantities of serum antibody, and responses without CT were substantially lower. In conclusion, we have shown here that CPMV is an efficient carrier of foreign peptides to the mucosal immune system, generating mucosal and serum antibody after intranasal vaccination with only very small quantities of peptide. The oral route is less effective under the conditions tested. Since these plant virus-derived vaccines have the potential for cost-effective manufacture and are expected to be extremely safe, as they are not known to infect mammalian cells, they have the potential as vaccines for the intranasal delivery of antigens to which the production of serum and/or mucosal antibody provides a protective response. ACKNOWLEDGMENTS This work was supported in part by the European Union, DGXII contract no. FAIR-CT95-0720. The CVPs were produced under UK MAFF licence PHL 91/2275 (06/1997). We also thank Aquila Biopharmaceuticals, Inc., for supplying QS-21 and monoclonal antibody V6F2. REFERENCES 1. Adams, S. E., K. M. Dawson, K. Gulli, S. M. Kingsman, and A. J. Kingsman. 1987. The expression of hybrid HIV:ty virus-like particles in yeast. Nature 329:68–70. 2. Almeida, A. J., and H. O. Alpar. 1996. Nasal delivery of vaccines. J. Drug Targeting 3:455–467. 3. Alving, C. R. 1993. Lipopolysaccharide, lipid A, and liposomes containing lipid A as immunologic adjuvants. Immunobiology 187:430–446. 4. Ball, J. M., M. E. Hardy, R. L. Atmar, M. E. Conner, and M. E. Estes. 1998. Oral immunization with recombinant Norwalk virus-like particles induces a systemic and mucosal immune response in mice. J. Virol. 72:1345–1353. 4a.Brennan, F. R., et al. Unpublished observations. 4b.Brennan, F. R., et al. Submitted for publication. 5. Brennan, F. R., J. J. Oliver, and G. D. Baird. 1994. Differences in the immune responses of mice and sheep to an aromatic-dependent mutant of Salmonella typhimurium. J. Med. Microbiol. 40:11–19. 6. Bukawa, H., K.-I. Sekigawa, K. Hamajima, J. Fukushima, Y. Yamada, H. Kiyono, and K. Okuda. 1995. Neutralization of HIV-1 by secretory IgA induced by oral immunization with a new macromolecular multicomponent peptide vaccine candidate. Nat. Med. 1:681–685. 7. Ciborowski, P., J.-I. Flock, and T. Wadstrom. 1992. Immunological response to a Staphylococcus aureus fibronectin-binding protein. J. Med. Microbiol. 37:376–381. 8. Cox, J. C., and A. R. Coulter. 1997. Adjuvants—a classification and review of their modes of action. Vaccine 15:248–256.
9. Dalsgaard, K., K. Lovgren, and D. E. S. Steward-Tull. 1995. Immune-stimulating complexes with QuilA, p. 129–143. In D. E. S. Stewart-Tull (ed.), The theory and practical application of adjuvants. John Wiley & Sons, Ltd., London, United Kingdom. 10. Dalsgaard, K., A. Uttenthal, T. D. Jones, F. Xu, A. M. Merryweather, W. D. O. Hamilton, J. P. Langeveld, R. S. Boshuizen, S. Kamstrup, G. P. Lomonossoff, C. Porta, C. Vela, J. I. Casal, R. H. Meloen, and P. B. Rodgers. 1997. Plant-derived vaccine protects target animals against a viral disease. Nat. Biotechnol. 15:248–252. 11. Dessens, J. T., and G. P. Lomonossoff. 1993. Cauliflower mosaic virus 35S promoter-controlled DNA copies of cowpea mosaic virus RNAs are infectious on plants. J. Gen. Virol. 74:889–892. 12. Elson, C. O., and M. T. Dertzbaugh. 1994. Mucosal adjuvants, p. 391–402. In P. L. Ogra, J. Mestecky, M. E. Lamm, W. Strober, J. R. McGhee, and J. Bienenstock (ed.), Handbook of mucosal immunology. Academic Press, Inc., New York, N.Y. 12a.Elstrom, J., et al. Submitted for publication. 13. Everest, P., P. Griffiths, and G. Dougan. 1995. Live Salmonella vaccines as a route towards oral immunization. Biologicals 23:119–124. 14. Flock, J.-I., G. Froman, K. Jonsson, B. Guss, C. Signas, B. Nilsson, G. Raucci, M. Hook, T. Wadstrom, and M. Lindberg. 1987. Cloning and expression of the gene for a fibronectin-binding protein of Staphylococcus aureus. EMBO J. 6:2351–2357. 15. Gilleland, H. E., Jr., L. B. Gilleland, J. Staczek, R. N. Harty, A. GarciaSastre, O. G. Engelhardt, and P. Palese. 1997. Chimeric influenza viruses incorporating epitopes of outer membrane protein F as vaccine against pulmonary infection with Pseudomonas aeruginosa. Behring Inst. Mitt. 98: 291–301. 16. Gupka, R. K., and G. R. Siber. 1995. Adjuvants for human vaccines—current status, problems and future prospects. Vaccine 13:1263–1276. 17. Hathaway, L. J., C. D. Partidos, P. Vohra, and M. W. Steward. 1995. Induction of systemic immune responses to measles virus synthetic peptides administered intranasally. Vaccine 13:1495–1500. 18. Hocart, M. J., J. S. Mackenzie, and G. A. Stewart. 1989. The IgG subclass response to influenza virus haemagglutinin in the mouse: effect of route of inoculation. J. Gen. Virol. 70:809–818. 19. Hsu, S.-C., E. B. Schadeck, A. Delmas, M. Shaw, and M. W. Steward. 1996. Linkage of a fusion peptide to a CTL epitope from the nucleoprotein of measles virus enables incorporation into ISCOMS and induction of CTL responses following intranasal immunization. Vaccine 14:1159–1166. 20. Kensil, C. R. 1996. Saponins as vaccine adjuvants. Crit. Rev. Ther. Drug Carrier Syst. 13:1–55. 21. Kipps, T. J., J. Parnham, J. Punt, and L. A. Herzemberg. 1985. Importance of immunoglobulin isotype in human antibody-dependent, cell-mediated cytotoxicity directed by murine monoclonal antibodies. J. Exp. Med. 161:1–17. 22. Klaus, G. G. B., M. B. Pepys, K. Kitajina, and B. A. Asconas. 1979. Activation of mouse complement by different classes of mouse antibody. J. Immunol. 38:687–695. 23. Kleanthous, H., G. A. Myers, K. M. Georgakopoulos, T. J. Tibbitts, J. W. Ingrassia, H. L. Gray, R. Ding, Z.-Z. Zhang, W. Lei, R. Nichols, C. K. Lee, T. H. Ermak, and T. P. Monath. 1998. Rectal and intranasal immunizations with recombinant urease induce distinct local and serum immune responses in mice and protect against Helicobacter pylori infection. Infect. Immun. 66: 2879–2886. 24. Klinman, D. M., M. Takeno, M. Ichino, M. Gu, G. Yamshchikov, G. Mor, and J. Conover. 1997. DNA vaccines: safety and efficacy issues. Springer Semin. Immunopathol. 19:245–256. 25. Lehner, T., L. A. Bergmeier, C. Panagiotidi, L. Tao, R. Brookes, L. S. Klavinskis, P. Walker, J. Walker, R. G. Ward, L. Hussain, A. J. H. Gearing, and S. E. Adams. 1992. Induction of mucosal and systemic immunity to a recombinant simian immunodeficiency viral protein. Science 25:1365–1369. 26. Lin, T., C. Porta, G. Lomonossoff, and J. E. Johnson. 1996. Structure-based design of peptide presentation on a viral surface: the crystal structure of a
938
BRENNAN ET AL.
plant/animal virus chimera at 2.8 Å resolution. Folding Design 1:179–187. 27. Lovgren, K., H. Kaberg, and B. Morein. 1990. An experimental influenza subunit vaccine (ISCOM): induction of protective immunity to challenge infection in mice after intranasal or subcutaneous administration. Clin. Exp. Immunol. 82:435–440. 28. McLain, L., C. Porta, G. P. Lomonossoff, Z. Durrani, and N. J. Dimmock. 1995. Human immunodeficiency virus type 1 neutralizing antibodies raised to a gp41 peptide expressed on the surface of a plant virus. AIDS Res. Hum. Retroviruses 11:327–334. 29. McLain, L., Z. Durrani, L. A. Wisniewski, C. Porta, G. P. Lomonossoff, and N. J. Dimmock. 1996. Stimulation of neutralizing antibodies to human immunodeficiency virus type 1 in three strains of mice immunized with a 22-mer amino acid peptide expressed on the surface of a plant virus. Vaccine 14:799–810. 30. Modelska, A., B. Dietzschold, N. Sleysh, Z. F. Fu, K. Steplewsk, D. C. Hooper, H. Koprowski, and Y. Yusibov. 1998. Immunization against rabies with plant-derived antigen. Proc. Natl. Acad. Sci. USA 95:2481–2485. 31. Montgomery, D. L., J. J. Donnelly, J. W. Shiver, M. A. Liu, and J. B. Ulmer. 1994. Protein expression in vivo by injection of polynucleotides. Curr. Opin. Biotechnol. 5:505–510. 32. O’Neal, C. M., S. E. Crawford, M. K. Estes, and M. E. Conner. 1997. Rotavirus virus-like particles administered mucosally induce protective immunity. J. Virol. 71:8707–8717. 33. Partidos, C. D., P. Vohra, and M. W. Steward. 1996. Induction of measles virus-specific cytotoxic T-cell responses after intranasal immunization with synthetic peptides. Immunology 87:179–185. 34. Porta, C., V. E. Spall, J. E. Loveland, J. E. Johnson, P. J. Barker, and G. P. Lomonossoff. 1994. Development of cowpea mosaic virus as a high-yielding system for the presentation of foreign peptides. Virology 202:949–955. 35. Reed, R. C., V. Louis-Wileman, E. V. Cosmai, S. Fang, D. L. Jue, R. M. Wohlhueter, R. L. Hunter, and A. A. Lal. 1997. Multiple antigen constructs (MACs): induction of sterile immunity against sporozoite stage of rodent malaria parasites, Plasmodium berghei and Plasmodium yoelii. Vaccine 15: 482–488. 36. Resnick, D. A., A. D. Smith, S. C. Geisler, A. Zhang, A. Arnold, and G. F. Arnold. 1995. Chimeras from a human rhinovirus 14-human immunodeficiency virus type 1 (HIV-1) V3 loop seroprevalence library induce neutralizing responses against HIV-1. J. Virol. 69:2406–2411. 37. Richards, C. M., T. J. Hill, and N. A. Williams. 1997. Enhancement of the immune response to non-replicating herpes simplex virus type-1 prepara-
J. VIROL.
38.
39.
40. 41. 42.
43. 44.
45. 46. 47. 48. 49.
tions by mucosal administration in the presence of cholera toxin. Vaccine 15: 1065–1069. Rudin, A., E.-L. Johansson, C. Berquist, and J. Holmgren. 1998. Differential kinetics and distribution of antibodies in serum and nasal and vaginal secretions after nasal and oral vaccination of humans. Infect. Immun. 66:3390– 3396. Russell, M. W., Z. Moldoveanu, P. L. White, G. G. Sibert, J. Mestecky, and S. M. Michalek. 1996. Salivary, nasal, genital and systemic antibody responses in monkeys immunized intranasally with a bacterial protein antigen and the cholera toxin B subunit. Infect. Immun. 64:1272–1283. Rose, C. S. P., and D. J. Evans. 1991. Poliovirus antigen chimeras. Trends Biotechnol. 9:415–421. Schennings, T., A. Heimdahl, K. Coster, and J.-I. Flock. 1993. Immunization with fibronectin binding protein from Staphylococcus aureus protects against experimental endocarditis in rats. Microb. Pathog. 15:227–236. Signas, C., G. Raucci, K. Jonsson, P.-E. Lindgren, G. M. Anantharamaiiah, M. Hook, and M. Lindgren. 1989. Nucleotide sequence of the gene for a fibronectin-binding protein from Staphylococcus aureus: use of this peptide sequence in the synthesis of biologically active peptides. Proc. Natl. Acad. Sci. USA 86:699–703. Staats, H. F., R. J. Jackson, M. Marinaro, I. Takahashi, H. Kiyono, and J. R. McGhee. 1994. Mucosal immunity to infection with implications for vaccine development. Curr. Opin. Immunol. 6:572–583. Staats, H. F., W. G. Nichols, and T. J. Palker. 1996. Mucosal immunity to HIV-1-systemic and vaginal antibody-responses after intranasal immunization with the HIV-1 C4/V3 peptide T1SP10 MN (A). J. Immunol. 157: 462–472. Sutter, G., and B. Moss. 1992. Non-replicating vaccinia vector efficiently expresses recombinant genes. Proc. Natl. Acad. Sci. USA 89:10847–10851. Usha, R., J. Rohll, V. E. Spall, M. Shanks, A. J. Maule, J. E. Johnson, and G. P. Lomonossoff. 1993. Expression of an animal virus antigenic site on the surface of a plant virus particle. Virology 197:366–374. van Kammen, A., and C. P. de Jager. 1978. Cowpea mosaic virus. CMI descriptions of plant viruses, no. 197. Commonwealth Mycological Institute, Kew, Richmond, Surrey, United Kingdom. van Pinxteren, L. A., I. Campbell, C. J. Clarke, D. R. Snodgrass, and M. G. Bruce. 1997. A single oral dose of inactivated rotavirus and ISCOM matrices induces partial protection in lambs. Biochem. Soc. Trans. 25:340S. Vlacha, V., E. N. Forman, D. Miron, and G. Peter. 1996. Recurrent thrombocytopenic purpura after repeated measles-mumps-rubella vaccination. Pediatrics 97:738–739.
