Antipeptide Antibody Responses following Intranasal Immunization: Effectiveness of Mucosal Adjuvants

Abstract

Toxicity is a major factor limiting the development and use of potent adjuvants for human mucosally delivered vaccines. Novel adjuvant formulations have recently become available, and in the present study two have been used for intranasal immunization with a synthetic peptide immunogen (MAP-M2). This peptide represents a multiple antigenic peptide containing multiple copies of a mimotope M2, a peptide mimic of a conformational epitope of the fusion protein of measles virus. MAP-M2 was administered intranasally to experimental animals together with synthetic oligodeoxynucleotides containing unmethylated CpG motifs with or without a mutant of wild-type enterotoxin of Escherichia coli (LTR72). The combination of the mutant toxin LTR72 and the CpG repeats, codelivered with a peptide immunogen, induced both local and systemic peptide- and pathogen-specific humoral and cellular immune responses comparable to those obtained after intranasal immunization with the wild-type toxin LT. In addition, this combination of adjuvants induced a predominantly immunoglobulin G2a antibody response. If both the LTR72 and CpG adjuvants are shown to be safe for use in humans, this particular combination would appear to have potential as an adjuvant for mucosally delivered vaccines in humans.

Mucosal surfaces are uniquely structured for the development of effective immune responses against pathogens that invade via the mucosal route. Immunization via this route offers the potential for the induction of neutralizing antibodies and specific cellular responses, both systemically and locally, at site of pathogen entry. This is particularly important for development of immunity to diseases initiated at the mucosal surface (for instance, measles). Moreover, mucosal immunisation can be safe and effective even in young infants in the presence of maternally derived antibodies (40), and the elimination of the need for injection removes the risk of transmission of viral diseases such as hepatitis B and AIDS. The effectiveness of mucosal immunisation in humans has been demonstrated by the success of the oral polio vaccine (Sabin), which induces both local and systemic immune responses. There are also examples of successful measles vaccines in young children after administration as an aerosol or via the intranasal route (1, 35).

The choice of an appropriate adjuvant for mucosal vaccination is often the key for success since many antigens introduced via the mucosal route are poorly immunogenic and, in the absence of adjuvant, may induce a state of tolerance.

Bacterial toxins have been for a long time utilized as adjuvants in experimental models, and some chemically detoxified toxins have been employed to prevent bacterial infectious diseases (e.g., formalin inactivation of Corynebacterium diphtheriae or Clostridium tetani exotoxins). Although bacterial toxins possess excellent adjuvant properties, their high toxicity precludes their use in humans. At present, detoxified derivatives can be obtained by mutagenesis of the toxin genes and, since these modified genes encode different amino acid(s), their products no longer carry enzymatic activity. Such inactivated derivatives are safe and in the future could replace toxoids in existing vaccines as well as being used as mucosal adjuvants in new vaccination strategies.

The most powerful and most studied mucosal adjuvants are cholera toxin (CT) and heat-labile enterotoxin (LT) of Escherichia coli. Both toxins have a similar tertiary structure and share 80% homology in their primary sequence (10). Several mutants of both toxins have been produced and described in detail. Among them are CTK63 (Ser-63→Lys) (15), CTS106 (Pro-106→Ser) (15), CTF61 (Ser-61→Phe) (41), LTK63 (Ser-63→Lys) (12, 32), LTR72 (Ala-72→Arg) (32), and LTG192 (Arg-192→Gly) (13). In the murine model of measles virus (MV) infection, mucosal coimmunization with LKT63 and a synthetic peptide representing a cytotoxic-T-lymphocyte (CTL) epitope from measles N protein resulted in effective in vivo priming of peptide- and MV-specific CTLs (30). LTR72 has been demonstrated to be a strong adjuvant in comparison to LTK63, and in animals it has been shown to be 100,000-fold less toxic but 20 times less effective than LT (17). Moreover, CD4⁺ lymphocytes from animals immunized with ovalbumin together with LTR72 exhibited very strong proliferative responses, which were very similar to those induced by wild-type LT (17).

Synthetic oligodeoxynucleotides (ODNs) that contain unmethylated CpG motifs (CpG ODNs) are also novel candidates as adjuvants for mucosal immunization. Initially, it was reported that these motifs could induce in vitro production of interleukin-6 (IL-6) and gamma interferon (IFN-γ) by CD4⁺ T cells, IL-6 and IL-12 by B cells, and IFN-γ by NK cells (20). Such properties led to the use of CpG ODNs as adjuvants in several experimental models (11, 21, 22, 23) and, indeed, studies published thus far support the view that Th1-type responses dominate after CpG coadministration with an immunogen. The nature of the immune response developed depends on the age of the animals (2, 21), the route of antigen delivery (2), and the nature of the antigen (22); nevertheless, the potential of CpG motifs as adjuvants for delivery via the mucosal surfaces is particularly promising (3, 25, 26, 27).

In the work presented here, intranasal immunization with a multiple antigenic peptide construct (MAP) containing a peptide mimic (M2) of a conformational B-cell epitope from measles F protein (MAP-M2) was studied. The mimotope M2 induces specific antipeptide and antimeasles neutralizing antibodies when administered intraperitoneally (39) and, when presented as a MAP (MAP-M2), is very immunogenic and induces high titers of high-affinity antibodies (29). The immunogenicity of MAP-M2 when delivered to the mucosal surface in the presence of the novel adjuvants LTR72 and CpG ODN was studied. The results show that the combination of the mutant toxin LTR72 and the CpG repeats, codelivered with MAP-M2, induced both local and systemic peptide- and pathogen-specific immune responses. These responses were comparable to those obtained after intranasal immunization with the wild-type toxin LT.

MATERIALS AND METHODS

Selection and synthesis of peptides.

M2 (NIIRTKKQ) represents a peptide mimic of an epitope from the MVF protein (39), which was selected from a solid-phase 8-mer random combinatorial peptide library by screening with a monoclonal antibody (F7-21) to the F protein of MV. Immunization with M2 induces antibodies that cross-react with MV and confers protection against fatal encephalitis induced following challenge with MV (39). MAP-M2 is a MAP with eight M2 sequences added using Fmoc (9-fluorenylmethoxy carbonyl) chemistry. All peptides were synthesized using commercially available resin with a polylysine backbone (Nova Biochem). The purity of peptides was assessed by high-pressure liquid chromatography and mass spectrometry.

Mice.

Inbred female BALB/c (H-2^d) mice were purchased from the National Institute of Medical Research, Mill Hill, United Kingdom, and maintained at the London School of Hygiene and Tropical Medicine, London, United Kingdom.

Measles virus.

The Edmondson strain of MV was used for the determination of anti-MV antibody titers and antibody avidity to the virus in enzyme-linked immunosorbent assay (ELISA) assays. The virus was grown in Vero cells in 199 Earl's medium (Life Technologies) supplemented with 5% fetal calf serum (FCS), 5% HEPES, and 0.2% penicillin-streptomycin. When the cytopathic effect was extensive, the cell monolayers were removed and clarified by centrifugation (250 × g, 4°C, 10 min). The supernatant was concentrated by ultrafiltration (Amicon) using a membrane with a molecular weight cutoff of 100,000 at an N₂ pressure of 20 lb/in². The resulting suspension was overlaid onto a 20%–60% sucrose gradient and spun (30,000 × g, 4°C, 90 min); the pellet was then resuspended in phosphate-buffered saline (PBS) and spun again.

Adjuvants.

E. coli LT toxin was purchased from Sigma. LTR72 is a mutant of LT toxin and was a kind gift of R. Rappuoli (Chiron S.p.A., Siena, Italy). CpG repeats with nucleotide sequence TCCATGACGTTCCTGACGTT (ODN 1826, originally published by Davis et al. [11]) were synthesized by Pharmacia Biotech.

Immunization of mice.

BALB/c mice (5 to 8 weeks old; four animals per group) were immunized intranasally under halothane anesthesia. Animals were given 50 μg of MAP-M2 (i) in normal saline, (ii) coimmunized with 10 μg of LTR72, (iii) coimmunized with 10 μg of CpG ODN, (iv) coimmunized with 10 μg of LTR72 and 10 μg of CpG ODN, and (v) coimmunized with 10 μg of LT (2, 17). Immunization was performed on days 0, 7, 14, and 28 with a total volume of 30 μl per mouse per inoculation (17).

Antibody ELISA.

Anti-peptide and anti-MV antibody titers in serum and saliva samples were assessed by a solid-phase ELISA on microtiter plates (Nunc, Roskilde, Denmark). Plates were coated overnight at 4°C with 50 μl of a 5-μg/ml solution of MAP-M2 per well or with 50 μl of a 5-μg/ml of purified MV in 0.1 M carbonate-bicarbonate buffer (pH 9.6) per well. The plates were blocked with 1% bovine serum albumin (BSA) in PBS (pH 7.3). Serial twofold dilutions of sera or saliva in PBS–0.05% Tween 20–1% BSA (final volume, 50 μl) were added to the plates, which were incubated at 37°C for 1 h and then washed. Then, 50 μl of a 1:2,000 dilution of peroxidase-conjugated rabbit anti-mouse immunoglobulin G (IgG; heavy and light chains), IgG1, IgG2a, IgG2b, or IgA (Nordic) was added to each well, and the plates were incubated for 1 h at 37°C. Unbound conjugate was removed by washing, and 50 μl of 0.04% o-phenylenediamine–hydrogen peroxidase in citrate-phosphate buffer was added to detect bound enzyme. The reaction was stopped after 10 min by the addition of 25 μl of 2 M sulfuric acid per well, and the absorbance (A₄₉₂) was determined in an automatic plate reader (Dynex MRX). Titers of anti-peptide and anti-virus antibodies are expressed as the log₁₀ of the reciprocal of the serum dilution giving an absorbance of 0.2.

Antibody affinity measurement.

The affinity of anti-peptide antibodies for M2 was assessed by a solid-phase enzyme inhibition assay (33). ELISA plates were coated with MAP-M2 at 5 μg/ml, and doubling dilutions of sera were added. The assay was continued, as described above, to assess the dilution of antibody giving an A₄₉₂ of 0.6, which was then used in the second stage of the assay. Serial dilutions of a 3 mM solution of M2 were used to inhibit the binding of homologous antibody to the solid-phase antigen. The relative affinity of antibody was calculated as the reciprocal of the concentration of peptide giving 50% inhibition (I_0.5) of the binding in the absence of the peptide. These values represent an estimation of “average” antibody affinity.

ELISPOT.

Bone marrow cell suspensions were assayed for the number of specific antibody-secreting cells (ASC) by a modification of the original enzyme-linked immunospot (ELISPOT) method (8, 9, 37). Plates with cellulose at the bottom (Millipore) were coated overnight with 100 μl of a 5-μg/ml solution of MAP-M2 or a 5-μg/ml solution of MV in carbonate-bicarbonate buffer. Plates were washed with PBS and blocked with 200 μl of RPMI supplemented with 5% FCS and 0.2% penicillin-streptomycin (complete medium) per well for 30 min at 37°C. The contents of the wells were replaced with 100 μl of complete medium containing various numbers of lymphocytes isolated earlier by Histopaque (Sigma Diagnostics) gradient centrifugation of fresh bone marrow cells. May-Grünwald staining was performed to determine the morphology of isolated cells. On average, 90% of all cells were lymphocytes, and cell viability was >95% (as determined by eosin dye exclusion). For each determination, duplicates of three different cell concentrations (10⁴ to 10⁶ per well) were assayed. Plates were incubated undisturbed for 4 h at 37°C in 5% CO₂ and rinsed three times with PBS and four times with PBS-Tween 20. Next, 100 μl of PBS-Tween containing 1% FCS and 1/1,000-diluted goat anti-mouse alkaline phosphatase (AP)-conjugated IgA (Sigma ImmunoChemicals) or rabbit anti-mouse horseradish peroxidase (HRP)-conjugated IgG (Nordic) was added. Plates were incubated at 4°C overnight and washed three times with PBS-Tween, twice with PBS, and twice with 0.05 M Tris-buffered saline (pH 8.0). Wells were then exposed to 100 μl of AP chromogen substrate (Sigma Fast 5-bromo-4-chloro-3-indolylphosphatide–nitroblue tetrazolium [Sigma] for 10 to 20 min or to 100 μl of HRP chromogen substrate (3,3′-diaminobenzidine [Sigma]) for 10 min. Plates were thoroughly rinsed with tap water, dried, and examined for the presence of blue (AP) or brown (HRP) spots. The spots were counted under low magnification (×40).

Lymphocyte proliferation assay.

Mice were killed 2 weeks after the final immunization. Spleens were removed aseptically, transferred to transport medium, teased, pooled within each group, and passed through a sieve. Cells were spun at 1,200 rpm for 5 min, the supernatant was discarded, and red blood cells were lysed using lysing buffer (0.14 M ammonium chloride, 20 mM Tris; pH 7.5). B lymphocytes were removed by nylon wool chromatography as described earlier (7, 19). Eluted cells are referred to as the T-cell-enriched population. After three washes in RPMI 1640, cells were resuspended in complete medium (2% FCS, 0.2% penicillin-streptomycin, 1 mM glutamine, and 1 mM HEPES buffer in RPMI 1640). Three 10-fold dilutions of MAP-M2 and concanavalin A were dispensed into the wells of 96 round-bottom tissue culture plates. Negative controls were included. All tests were done in triplicate. A total of 2 × 10⁵ cells were added to each well, and plates were incubated for 3 days at 37°C in 5% CO₂. Cells were pulsed on the fourth day with tritiated thymidine (1 μCi/well) and harvested 18 h later on glass microfiber papers. Thymidine incorporation was assessed by liquid scintillation spectrometry. Results are expressed as stimulation indices (SI) of the mean counts per minute (cpm) from triplicate cultures in the presence of antigen divided by the mean cpm of triplicate cultures obtained with medium only. Values equal to or higher than 2 were considered positive.

IFN-γ ELISA.

Immulon-4 plates were coated overnight with 50 μl of 5-μg/ml concentration of capture antibody (IFN-γ; Pharmingen) per well diluted in carbonate-bicarbonate buffer (pH 9.6). Plates were kept at 4°C, washed with PBS-Tween, and blocked with 2% BSA for 1 h at 37°C. Supernatants obtained 48 h from the beginning of cultures (described above) were added in triplicate at three dilutions in RPMI 1640. The IFN-γ standards and negative wells were included. After 2 h of incubation at room temperature, plates were washed with PBS-Tween, and biotinylated rat anti-mouse IFN-γ in a 1:4,000 dilution (Pharmingen) was added at 50 μl per well. Plates were further incubated for 1 h and washed as before, and anti-biotin antibodies conjugated to peroxidase were added to each well at a concentration 1:500 (Sigma) for 1 h. The remaining steps of the assay were performed as described above.

Statistical analysis.

Antibody titers and values of affinity and avidity were analyzed by using the Tukey-Kramer multiple comparisons test.

RESULTS

Peptide-specific and MV-specific antibodies.

Anti-MAP-M2 and anti-MV antibody titers were assessed in serum samples from mice immunized intranasally with (i) MAP-M2 in PBS, (ii) MAP-M2 with LTR72, (iii) MAP-M2 with CpG, (iv) MAP-M2 with LTR72 and CpG, and (v) MAP-M2 with LT (Fig. 1). The highest anti-peptide antibody levels were obtained in animals immunized with MAP-M2 and wild-type toxin LT (group E) or MAP-M2 and mutant toxin LTR72 (group B). In the group where CpG was used (group C), anti-MAP-M2 antibody titers were not significantly different from those in the group receiving no adjuvant (group A, P = 0.47), but the addition of LTR72 and CpG (group D) significantly increased antibody levels (P = 0.01, compared with group A, and P = 0.005 compared with group C). Anti-MV antibody titers were significantly higher in all groups of mice receiving MAP-M2 with each of the adjuvants compared to mice receiving peptide in saline only (P < 0.001). Animals that were given MAP-M2 and CpG plus LTR72 (group D) generated the highest anti-MV antibody levels, and this was significantly different from those obtained from animals immunized with MAP-M2 and CpG (P = 0.005). The effect of the use of adjuvant could also be observed when the affinity of anti-mimotope antibodies was assessed. The data presented in Fig. 2 indicate that the use of each of the adjuvants significantly increases the affinity of anti-mimotope antibodies compared to that seen following immunization without adjuvant (P < 0.05). The IgG subclass distribution of anti-peptide antibodies is shown in Fig. 3. Mice immunized with MAP-M2 and CpG (group C), MAP-M2 and CpG plus LTR72 (group D), or MAP-M2 and LT (group E) generated more IgG2a antibodies than IgG1 antibodies, with the highest IgG2a/IgG1 ratio (4.07) in group C. Immunization with MAP-M2 without an adjuvant or with mutant toxin LTR72 (groups A and B) resulted in the production of predominantly IgG1 anti-peptide antibodies.

FIG. 1.

Anti-MV (A) and anti-MAP-M2 (B) antibody titers in mice immunized intranasally with MAP-M2 and the indicated adjuvant. Female BALB/c mice at 6 to 8 weeks of age (four animals per group) were immunized on days 0, 7, 14, and 28 with 50 μg of MAP-M2 and 10 μg of adjuvant in 30 μl. Blood was withdrawn for analysis 2 weeks after the last immunization. Results are presented as the means ± the standard deviation (SD).

FIG. 2.

Affinity of anti-mimotope antibodies for M2. Results are presented as the means of four values ± the SD.

FIG. 3.

IgG subclasses of anti-MAP-M2 antibodies induced following intranasal immunization with MAP-M2 and the indicated adjuvant. Female BALB/c mice (6 to 8 weeks old; four animals per group) were immunized on days 0, 7, 14, and 28 with 50 μg of MAP-M2 and 10 μg of adjuvant in 30 μl. Blood was withdrawn for analysis 2 weeks after the last immunization. Results are presented as the means ± the SD. The ratios of IgG2a to IgG1 titers were as follows: A, 0.19; B, 0.4; C, 4.07; D, 2.45; and E, 2.04.

Peptide-specific IgA levels were measured in saliva obtained from animals 2 weeks after the last intranasal immunization with MAP-M2, and animals coimmunized with wild-type toxin LT or its mutant LTR72 produced the highest amounts of IgA (Fig. 4).

FIG. 4.

Anti-MAP-M2 mucosal IgA antibody responses. Mice were immunized intranasally with the indicated constructs. Results are shown as the mean titers and the SD values.

Assessment of antibody secreting cells by ELISPOT assay.

Bone marrow cell preparations were assayed for the number of specific ASC by adding serial dilutions of freshly isolated cells to plates coated with MAP-M2 (Fig. 5A) or MV (Fig. 5B). The highest numbers of cells producing anti-MAP-M2 IgG antibodies were obtained from animals immunized with MAP-M2 and LTR72, MAP-M2 and LT, or MAP-M2 and LTR72 plus CpG (Fig. 5A). Although animals immunized with MAP-M2 and CpG developed significantly fewer anti-peptide IgG-secreting cells, their numbers were very close to the numbers of cells secreting anti-MV antibodies (Fig. 5B).

FIG. 5.

(A) Anti-MAP-M2 ASC in bone marrow after intranasal immunization with MAP-M2 and the indicated adjuvant. (B) Anti-MV ASC in bone marrow after intranasal immunization with MAP-M2 and the indicated adjuvant.

When IgA-secreting cells were assessed, the highest numbers of peptide-specific and MV-specific cells were observed in mice coimmunized with MAP-M2 and a combination of the two adjuvants, CpG and LTR72. However, this was not significantly different from the group immunized with MAP-M2 and LT. Moreover, the numbers of peptide-specific and MV-specific cells producing IgA within each group were similar.

Proliferative in vitro responses to immunogens.

The ability of antigen-primed lymphocytes to respond in vitro to the antigen was assessed by conventional lymphocyte proliferation assay. Immunization with MAP-M2 and wild-type LT toxin led to the development of in vitro lymphocyte proliferative responses to MAP-M2 (Fig. 6A). Greater SI were observed with lymphocytes from animals immunized with MAP-M2 and a combination of two adjuvants, CpG and LTR72. However, using either of these adjuvants separately (MAP-M2 and CpG or MAP-M2 and LTR72) induced notably lower proliferative responses.

FIG. 6.

Lymphoproliferative responses of splenocytes from mice immunized intranasally with MAP-M2 and the indicated adjuvant. (A) Whole-lymphocyte population. (B) T-cell-enriched population. Cells were restimulated in vitro with MAP-M2. Data are presented as the mean SI values. Concanavalin A in vitro restimulation gave SI values in the range of 4.5 to 5.82.

T-cell-enriched populations were tested under the same conditions as for the unfractionated cells, and the proliferative results obtained show that the removal of B lymphocytes did not significantly affect the SI (Fig. 6B).

IFN-γ in cell cultures.

The levels of IFN-γ in supernatants obtained from lymphocyte proliferation assays varied significantly (Fig. 7). The highest levels were observed in supernatants from cells restimulated in vitro with MAP-M2 obtained from animals immunized with MAP-M2 plus LT or from animals that received MAP-M2 plus LTR72.

FIG. 7.

IFN-γ in cell cultures restimulated in vitro with MAP-M2. Results represent the means of three replications. Background values from wells without antigen were subtracted. Concanavalin A restimulation of cells gave a value of 3.4 × 10⁴ ± 2.5 × 10⁴ pg/ml.

DISCUSSION

The use of adjuvants is often essential for the induction of effective immune responses but for mucosal immunization they are particularly important since most antigens are poorly immunogenic when given via this route. It is also important to note that potent adjuvants often exert high toxicity, a fact that needs to be addressed in terms of the design of vaccine strategies.

In experiments presented here, a synthetic peptide immunogen (MAP-M2) was used to study the influence of different adjuvant formulations on the immune responses induced. MAP-M2 was immunogenic even when administered with saline alone, but the addition of different adjuvants significantly enhanced the peptide-specific humoral and cellular immune responses observed.

Toxin LT of E. coli was used throughout this study as a positive control since, when coimmunized with MAP-M2, it induced high serum antibody titers (Fig. 1) and a high number of ASC in bone marrow (Fig. 5). Serum antibodies were shown to be of high affinity (Fig. 2), and significant levels of secretory IgA in saliva were observed (Fig. 4). Furthermore, MAP-M2 plus LT primed lymphocytes were able to respond to MAP-M2 in vitro (Fig. 6A).

Similar results were obtained when the mutant of LT, LTR72, was used as an adjuvant. However, differences between the responses induced in the presence of the two toxins were demonstrated: (i) immunization with MAP-M2 plus LT induced predominantly IgG2a antibodies, whereas immunization with MAP-M2 plus LTR72 induced mainly IgG1 antibodies (Fig. 3), and (ii) immunization with MAP-M2 plus LT primed for in vitro lymphocyte restimulation with MAP-M2 more effectively than did immunization with MAP-M2 plus LTR72 (Fig. 6). These findings are consistent with recent observations (17, 30, 31).

Intranasal coadministration of CpG ODN with MAP-M2 peptide had no marked adjuvant effect on serum IgG antibody levels and lymphocyte proliferative responses to the peptide. However, it did enhance the salivary IgA responses and increased the number of IgG and IgA anti-virus-specific antibody-secreting bone marrow cells. The numbers of anti-peptide and anti-MV IgG-secreting cells were very similar, which may suggest that synthetic ODNs are important in the recruitment of pathogen-specific cells (Fig. 5). The lack of proliferative responses of splenocytes primed with MAP-M2 plus CpG could be explained by the recent demonstration by Chu et al. (5) that CpG ODNs downregulate antigen processing and presentation functions of macrophages (which are abundant in the spleen) by decreasing the synthesis of major histocompatibility complex (MHC) class II molecules. However, CpG ODNs do stimulate dendritic cells to express high levels of class II MHC and costimulator molecules and increase their antigen-processing activity (18, 38). That this was occurring in the experiments described here is suggested by the numbers of IgA- and IgG-secreting bone marrow cells observed (Fig. 5).

The mucosal adjuvanticity of the CpG motif was more pronounced when it was coadministered with the LTR72 mutant. Both antibody and T-cell responses were potentiated particularly at the level of proliferative T-cell responses, which were even higher than those induced by coimmunization with the LT (Fig. 6).

It is well recognized that the in vivo cytokine network can determine the profile of antibodies induced during an immune response (14). The results presented here show that the synergistic effect of LTR72+CpG alters the ratio of IgG2a and IgG1 antibody responses to the coadministered peptide, a finding which is suggestive of a Th1 type of response, whereas the result of peptide administration with the LTR72 mutant is suggestive of a Th2 type of response. This finding is consistent with recent observations by McCluskie and Davis (26), who have shown that intranasal coadministration of HBsAg with CT and the CpG ODN alters the anti-HBsAg subclass profile from Th2 to Th1. This shift in antibody subclasses suggests that the presence of the CpG ODN (which is a potent Th1 inducer [5, 34]) might downregulate the expression of the Th2 cell phenotype. Indeed, several studies have demonstrated that selected cytokines produced by Th1- or Th2-type cells can downregulate the expression of the opposite Th cell phenotype (6, 28, 36). For example, IFN-γ produced by Th1-type cells downregulates IL-4 produced by Th2-type cells (6, 28, 36). However, in the present study, when levels of IFN-γ from cultured immune splenocytes were measured, no direct correlation was observed between antibody subclasses and the concentration of IFN-γ in groups of mice in which LTR72 and LTR72+CpG were used as adjuvants. It could be argued that the observed shift in antibody subclasses might be a result of the regulatory effect mediated by IL-12 induced by the CpG motif (5, 34). This possibility is supported by recent findings by Marinaro et al., who have demonstrated the regulatory role of mucosally administered IL-12 on the phenotype of T-cell responses to a mucosally administered vaccine (24). However, since full cytokine profiles of the supernatants in the present study were not determined, it would be premature to confidently conclude that the use of these adjuvants induced a particular Th response.

MAP-M2 was originally designed as an immunogen to stabilize the secondary conformation of the mimotopes attached to the polylysine backbone. However, the results of preliminary experiments indicated that MAP-M2 might also be able to induce T-helper cell responses, since the construct induced good anti-mimotope and anti-MV antibodies, although it did not contain any known MV T-helper epitope. The nature of this T-cell help is not known, although in recent studies (29) the evidence suggests that the processing of the MAP-M2 molecule results in the formation of a new epitope that can be recognized by a T-cell receptor. The possibility exists that the new structure generated is a mimic of a T-cell epitope (16). Further studies would appear warranted to determine the detailed nature of this phenomenon.

We have presented here data showing that MAP-M2 is a powerful immunogen to induce specific anti-peptide and anti-MV antibodies and can be successfully delivered via the intranasal route. We have also demonstrated that the choice of mucosal adjuvant is crucial for the development of important parameters of both systemic and local immunity. The administration of CpG repeats together with another adjuvant skewing immune responses toward a Th2-type response (i.e., mutant LTR72) might redirect the immune responses toward a Th1 type (Fig. 3). Such an outcome may be beneficial for protection against measles but may also prevent hypersensitivity reactions after mucosal immunization. The combination of adjuvants appears to induce good T- and B-cell reactivity, and the presence of CpG ODNs provides improved specificity of IgA and IgG responses.

The observation that the combination of LTR72 and CpG has adjuvanticity comparable to that of wild-type LT toxin is of particular interest, but its significance in the development of mucosal vaccines will hinge upon these adjuvants being shown to be nontoxic in humans.