Intranasal administration of a synthetic lipopeptide without adjuvant induces systemic immune responses

Abstract

Parenteral injection of a lipopeptide containing a human leucocyte antigen (HLA)-A*0201-restricted cytotoxic T-lymphocyte (CTL) epitope from the human cytomegalovirus (HCMV) immunodominant matrix protein pp65 efficiently induces systemic CTL responses in HLA-A*0201 transgenic mice. In this study, we demonstrate that intranasal (i.n.) administration of this lipopeptide, covalently linked to a universal T helper (Th) epitope (PADRE), also induces potent systemic CTL responses. Immune responses were substantially reduced when the unlipidated peptide analogue was used (P<0·01). The induced CTL were CD8⁺, major histocompatibility complex (MHC) class I-restricted and CMV specific. Moreover, i.n. administration of this lipidated peptide elicited both systemic and local mucosal CD4⁺ T-cell proliferative responses, as well as antigen-specific delayed type hypersensitivity (DTH) immune responses. In contrast, mice receiving the unlipidated peptide analogue developed substantially reduced Th or DTH responses (P<0·05). These results highlight the usefulness and potential of lipopeptides delivered via mucosal routes as painless, safe, and non-invasive vaccines.

Introduction

Synthetic lipopeptides have recently gained considerable interest and represent promising vaccine candidates. We and others have established that by using a variety of parenteral immunization protocols (i.e. subcutaneous, intravenous and intraperitoneal), lipopeptides can induce, without adjuvant, systemic B, T helper (Th) and cytotoxic T lymphocyte (CTL) responses.^1–7 However, the underlying mechanisms by which lipopeptides elicit immune responses are only partially understood.⁸ Among several hypotheses, the palmitoyl moiety of lipopeptides may be able to attach and then fuse to lipidic components of cell membranes and subsequently deliver peptide epitopes into the cytoplasm of antigen-presenting cells (APC).^8,9 Indeed, palmitoyl peptides of 8–40 residues have been shown to passively cross the cell membrane of non-phagocytic cells.⁹

Delivering inert peptide antigens to mucosal surfaces has been hampered in practical terms by poor peptide lipophobicity and adsorption to mucosal membranes, and by poor immunogenicity, coupled with a paucity of sufficiently potent adjuvants that can be tolerated by humans.^10,11 To overcome some of these obstacles, peptide antigens are often modified and usually need to be conjugated to lipophylic adhesive vehicles to permit and enhance their uptake by mucosal-associated lymphoid tissues.^12,13 The most representative approaches include encapsulation in [poly dl-lactide-coglycoide (PLG)] microparticles,^14,15 proteosome vesicles,^16,17 different formulations of liposomes,¹⁸ immune-stimulating complexes,¹⁹ immunopotentiating reconstituted influenza virosomes,²⁰ or association with the lipophylic and complex immunostimulating tripalmitoyl-s-glyceryl-cysteinyl-seryl-serine (P₃CSS) moiety.^15,21,22 Each of these approaches has its own unique advantages and disadvantages, which have been reviewed previously.^13,23,24

The use of adjuvants is often essential for the induction of effective immune responses, and for mucosal immunization they are particularly important as subunit vaccines may lead to a state of immunologic unresponsiveness, known as peripheral tolerance, when given via this route.^25–27 The most powerful and intensely studied mucosal adjuvants are cholera toxin (CT) produced by the bacterium Vibrio cholerae, and the closely related heat-labile enterotoxin (LT) of Escherichia coli, as well as their B subunits (CTB and LTB).^25–29 However, because of their severe diarrhoeagenic property when ingested by humans, in amounts as low as 0·5 µg, these toxins are unacceptable for human use.²⁵ Although recent progress in genetic detoxification have resulted in minimizing the residual toxicity of both CT and LT toxins by dissociating their diarrhoeagenicity and adjuvancity, the immunogenicity of these molecules may still prevent their widespread repeated use in humans, as pre-existing immunity to CT and LT in most populations reduces their adjuvant activity.^14,25,29,30 Given such limitations in current approaches, the successful development of alternative mucosal immunization regimens is highly desirable.

We have hypothesized that if lipopeptides attach and cross cell membranes, then they should be able to bind and pass across mucosal membranes, allowing delivery of functional epitopes to the systemic immune system. To test this hypothesis, a model of a dipalmitoylated lipopeptide containing the pp65_495–503 CTL epitope from the human cytomegalovirus (HCMV) immunodominant matrix protein pp65,^3,31 covalently linked to a universal Th epitope,³² was administered intranasally (i.n.) to mice, and systemic cellular immunity was investigated. The data obtained revealed that application of a lipopeptide to the nasal epithelial surfaces can efficiently prime systemic CTL, Th and delayed-type hypersensitivity (DTH) immune responses.

Materials and methods

Peptides and lipopeptides

The human leucocyte antigen (HLA)-A*0201-restricted CTL epitope derived from the CMV immunodominant matrix protein pp65, referred to as pp65_495–503 peptide (NLVPMVATV), was used in this study (Fig. 1).^3,31 As a Th epitope, the Pan DR promiscuous peptide (PADRE), which was engineered to bind major histocompatibility complex (MHC) class II molecules with high affinity, was employed.³² Peptides were synthesized on an ABI 432 instrument using standard F-moc technology (PE Applied Biosystems, Foster City, CA). Peptides were cleaved from the resin using trifluoroacetic acid : ethanedithiol : thioanisole : anisole (90 : 5 : 3 : 2%) followed by ether (methyl-t-butyl ether) extraction and lyophilization. The lipopeptide construct, referred to here as (PAM)₂-K25V lipopeptide, is shown in Fig. 1. An unlipidated version, also purchased from Peninsula Laboratories (Peninsula Laboratories Inc, Belmont, CA), was used as control for i.n. immunization. The TT_615–629 peptide from tetanus toxoid (TT) protein,³¹ and the p53_149–157 peptide from the p53 protein,³³ were used as control peptides for Th and CTL responses, respectively. Purity of peptides and lipopeptides was >90% and >80%, respectively, as established by high-performance liquid chromatography (HPLC), mass spectrometry and amino acid analysis. Stock solutions were made at 5 mm in 10% dimethyl sulphoxide (DMSO) and 0·1% acetic acid for pp65_495–503, 1 mg/ml in phosphate-buffered saline (PBS) for PADRE, and 2 mm in 4% formic acid and 96% DMSO for the PAM₂-K25V lipopeptide and K25V unlipidated peptides. All peptides and lipopeptides were aliquoted and stored at −20° until assayed.

Figure 1.

A schematic representation of PAM₂-K25V dipalmitoylated peptide containing the amino acid sequence 495–503 from the human cytomegalovirus (HCMV) matrix protein pp65 (pp65_495–503) and the T helper (Th) epitope PADRE. The pp65_495–503 cytotoxic T-lymphocyte (CTL) epitope was covalently linked to the C-terminal end of the PADRE Th epitope, and the peptide was then modified with two palmitic acids moieties at the N-terminus by automated synthesis on ABI 432. X, cyclohexylalanine.

Animals and immunization procedures

The HLA-A*0201 transgenic (Tg) mice (6–8 weeks of age) used in this study were maintained as a breeding colony in the vivarium at the City of Hope (Duarte, CA). The Institutional Animal Care and Use Committee approved all animal procedures performed. Groups of three to five mice were immunized i.n. on days 0 and 20 with 25 nmol/dose of PAM₂-K25V, as previously described.^28,34 To investigate whether the palmitoyl moiety is required for immunogenicity of the lipid-tailed peptide, equimolar amounts of the unlipidated version (i.e. 25 nmol of the CTL-Th unlipidated peptide) were also administered i.n. A third group of mice was injected subcutaneously (s.c.) with 25 nmol of PAM₂-K25V. For i.n. administration, while mice were under a light Metofane® anaesthesia (Pitman-Moore, Inc. Mundelein, IL) a total volume of 4×5 µl of sterile PBS and 10% DMSO containing the peptide or the lipopeptide solution was distributed equally to the entrance of both nares using a 10-µl sterile pipette tip. To prevent localized trauma, the pipette tips were not placed into the nares. To avoid swallowing and oral delivery of lipopeptides, drops of 5 µl/nare were gently delivered each time into the nasal cavities, for 20–30 min. For parenteral immunization, mice were injected once s.c. at the base of the tail, as previously described.^3,31

Viruses and cell lines

The AD 169 strain of human CMV [American Type Culture Collection (ATCC), Manassas, VA] was propagated and titred as previously described.³ Briefly, the virus was propagated in MRC-5 (ATCC), a human fibroblast cell line, by three serial passages. Infectious supernatant was harvested for preparation of viral stocks when MRC-5 cells showed a 100% cytopathic effect. Dermal HLA-matched (A*0201, B51 and B55) and HLA-mismatched (A1, A30, B13 and B18) human fibroblast lines were generated by expanded culture of skin biopsies. Fibroblasts were maintained in Dulbecco's modified Eagle's minimal essential medium (DMEM) supplemented with 10% heat-inactivated fetal calf serum (FCS) containing 2 mm glutamine, 100 U/ml penicillin and 100 µg/ml streptomycin, and were used as target cells between the 4th and 12th passages. T2 cells,³⁵ HLA-matched Epstein–Barr virus-transformed lymphocyte cell lines (EBV-LCL), expressing A*0201, B51 and B55, as well as mismatched EBV-LCL expressing A1, A30, B13 and B18, were used to assay the MHC restriction of effector cells. EBV-LCL and T2 cell lines were maintained in RPMI-1640 with 10% FCS and were split 1 : 2 in fresh medium 1 day before use.

Lymphocyte proliferation assays

Twelve days after immunization, spleen, inguinal or draining cervical lymph nodes (LN) were removed from the mice and placed into ice-cold DMEM. Spleen and LN single-cell suspensions were washed twice with DMEM supplemented with 15 mm HEPES, 5·5×10⁻⁵ m β-mercaptoethanol (Gibco-BRL, Grand Island, NY), 2 mm glutamine, 50 U/ml of penicillin and 50 µg of streptomycin (termed CM). Spleen and LN single-cell suspensions were adjusted to 4×10⁶ cells/ml in CM supplemented with 10% heat-inactivated FCS and used as previously described.³⁶ Equal volumes of cells and CM, or CM with PADRE or control TT_615–629 peptide, were mixed to give a final concentration of 2×10⁶ cells/ml in medium alone or in medium with peptide at a concentration of 60, 20, 6, 2 or 0·2 µg/ml. The cell suspensions were incubated for 72 hr at 37° in an incubator with 5% CO₂. [³H]Thymidine ([³H]TdR; 1 µCi) (Dupont NEN, Boston, MA) was added to each well (16 hr before the cells were harvested onto glass-fibre filters), and wells were counted on a Matrix 96 ionization counter (Packard Instruments, Meriden, CT). The incorporated radioactivity was determined as described previously.^4,36,37 Cultures were set up in triplicate. Results are expressed as the mean counts per minute (c.p.m.) of cell-associated [³H]TdR recovered from wells containing antigen subtracted from the mean c.p.m. of cell-associated [³H]TdR recovered from wells without Ag (Δc.p.m.) (average of quadruplicate samples). Concanavalin A (Con A) (Sigma, St. Louis, MO) was used as a positive control. T-cell proliferative responses were considered positive when the Δc.p.m. was >1000 and the stimulation index (SI) was >2.^5,31,36,37

Generation of CTL and chromium release assays (CRA)

The method used for generation of CTL lines has been described previously.^3,31 Immune spleen or cervical LN cells were restimulated in vitro (IVS) for 5–6 days on autologous spleen feeder cells, which were first lipopolysaccharide (LPS) stimulated for 72 hr then acid treated and loaded with pp65_495–503 peptide. An additional IVS was performed on day 5–6 following the first stimulation, using peptide-pulsed and irradiated T2 cells as APC and irradiated spleen cells as feeders. Growth factors in the form of 5% (vol/vol) Rat T-Stim (Collaborative Biomedical Products, Bedford, MA) were added to the culture 2 days after each IVS. Six days after the second IVS, cytotoxic activity was assessed against peptide-sensitized T2 or HLA-A*0201-positive B-LCL target cells. Targets were incubated with either pp65_495–503 or control (p53_149–157) peptide, as described previously.^3,31 After Na-⁵¹CrO₄ (Amersham Corp., Arlington Heights, IL) loading, cells were rinsed three times and resuspended at a concentration of 10⁵/ml, and 0·1 ml of the cell preparation was added to round-bottomed microtitre wells. Different numbers of effector cells were added in triplicate in 0·1 ml of medium to achieve the desired effector : target (E : T) ratio. Autologous and mismatched human fibroblast (0·5×10⁶ cells) were split into 25-cm² flasks overnight, followed by incubation with 400 IU of human interferon-γ (IFN-γ) for 48 hr, as described previously.^3,31 Cells were then infected with CMV (AD 169 strain) at a multiplicity of infection (MOI) of 5 for ≈12 hr. Cells were labelled with Na-⁵¹CrO₄ for 45 min and washed twice prior to use as targets (2000/well) in a standard 4-hr CRA in a 96-well U-bottom plate. For spontaneous ⁵¹Cr release, 0·1 ml of medium was added to the labelled target cells in place of effector cells. For maximum ⁵¹Cr release, 0·1 ml of 2% sodium dodecyl sulphate (SDS) (United States Biochemical Co., Cleveland, OH) was added to the labelled target cells. After incubation for 4 hr at 37° in a 5% CO₂ incubator, the microtitre plates were centrifuged at 1120 g for 2 min (RT-2000D; Sorvall Instruments, Newtown, CT). Supernatants (0·1 ml) were removed, radioactivity in the supernatants was determined using a Cobra II™ auto γ-counter (Packard, Downers Grove, IL) and the per cent specific lysis was determined as described previously.³¹ Experimental determinations were performed in triplicate, and assay data were taken into consideration only if the background was <30%. The calculated spontaneous release was <15% for B-LCL cells and <25% for fibroblasts. In some experiments, monoclonal antibody (mAb) blocking of CTL activity was performed by adding serial dilutions of anti-CD8 mAb 53-6.7 (PharMingen, San Diego, CA), anti-CD4 mAb GK1.5 (PharMingen, San Diego, CA) or an isotype-matched control mAb, at 10 µg/ml, during the lysis part of the assay.

DTH assay and footpad histology

For the measurement of DTH responses, footpad swelling was employed as previously described.²⁹ Briefly, using a Hamilton microlitre 702 syringe (Hamilton Company, Reno, NV), 25 µl of sterile PBS containing 5 µg of PADRE, pp65_495–503, or control peptide (TT_615–629), was injected into the right hind footpad of mice immunized (s.c. or i.n.) with the (PAM)₂-Th-CTL lipopeptide or unlipidated peptide analogue. The left hind footpad was injected with 25 µl of sterile PBS as a control. Using a vernier caliper, footpad swelling was determined 24–72 hr later. Peptide-specific footpad swelling was calculated by subtracting the footpad swelling of the PBS-injected footpad from the swelling of the peptide-injected footpad. In some experiments, 72 hr after the boost, footpad tissues were excised immediately after killing the mice, washed in sterile PBS, fixed in 4% paraformaldehyde for 7–10 days and footpad histology was performed as previously described.³⁸ For light microscopy, footpad tissues were fixed in 10% neutral-buffered formalin and processed for paraffin embedding. Serial 4-µm sections were stained with haematoxylin & eosin and examined using a Leitz DM-RB light microscope (Leica Microsystems Inc., Deerfield, IL). Preparations were photographed using a digitizing SPOT 1.1.0 camera (Diagnostic Instruments Inc., Sterling Heights, MI) and analyzed using Image Pro 3.0 software (Media Cybernetics, Silver Spring, MD) on a PC (P5-100; Gateway, Lake Forest, CA).

Statistical analysis

All graphs represent data from at least two independent experiments. Data are expressed as mean values ±standard deviation (except Fig. 5a), which were compared using the Student's t-test with the statview ii statistical program (Abacus Concepts, Berkeley, CA).

Figure 5.

Induction of an antigen-specific delayed-type hypersensitivity (DTH) response following intranasal administration of the PAM₂-K25V peptide. (a) PAM₂-K25V was administered to groups of human leucocyte antigen (HLA)-A*0201 transgenic (Tg) mice (n=3), and DTH immune responses were recalled 2 weeks later by injecting PADRE, pp65_495–503 or the control TT_615–629 peptide into the footpad. Footpad swelling was measured 48 hr after the booster immunization. The standard deviation (SD) for each group is indicated by a bar. (b) Histological analysis of footpads from HLA-A*0201 Tg mice primed intranasally (i.n.) (A and C) or subcutaneously (s.c.) (B and D) with PAM₂-K25V and boosted with PADRE (A and B) or control peptide (C and D). Mice were killed 48 hr after the second immunization, and tissues from the distal footpad were stained with haematoxylin and eosin. Original magnification: ×400 for all sections. The data are similar and representative of two independent experiments.

Results

Intranasal administration with a synthetic lipopeptide is effective in stimulating both mucosal and systemic Th-cell responses

Immunogenicity of PAM₂-K25V (Fig. 1) comprised the universal Th epitope (PADRE), and the HLA-A*0201-restricted CMV pp65_495–503 CTL epitope was assessed following i.n. administration of HLA-A*0201 Tg mice in saline/DMSO without mucosal adjuvant. The effectiveness of i.n. administration of PAM₂-K25V was evaluated in stimulating systemic Th responses. In a conventional [³H]TdR-uptake proliferation assay, spleen cells were examined from mice immunized i.n. with either the lipopeptide or with an equimolar amount of the unlipidated peptide analogue. For comparative purposes, Th-cell responses induced by the lipopeptide using the s.c. route were also examined. As shown in Fig. 2(a), spleen cells recovered from mice immunized i.n. with the lipopeptide, but not the unlipidated analogue, proliferated in response to PADRE peptide (P<0·01 for Th-cell responses comparing lipopeptide with the unlipidated form). These T-cell responses were antigen specific, as no response was detected against the control TT_615–629 peptide (Fig. 2a). Splenic Th-cell responses were of similar magnitude in mice receiving lipopeptide following the i.n. (Fig. 2a) or s.c. (Fig. 2b) route of administration (P<0·05 comparing the s.c. with the i.n. route using lipopeptide).

Figure 2.

Intranasal immunization with a lipopeptide induces systemic T helper (Th) cell responses. (a) Human leucocyte antigen (HLA)-A*0201 transgenic (Tg) mice (n=4) were immunized intranasally (i.n.) with either PAM₂-K25V lipopeptide or an equimolar amount of the unlipidated version. Th responses were stimulated in vitro utilizing either PADRE (filled squares) or TT_615–629 (filled triangles) for PAM₂-K25V, and equivalently for the unlipidated version (open squares, triangles). Comparison between vaccine groups was significant at P<0·01. (b) Same nomenclature as (a), except that peptides were introduced using the subcutaneous (s.c.) route. Spleen cells were isolated and PADRE-specific Th-cell responses were determined as described in the Materials and methods. The TT_615–629 peptide was used as a control for measurement of Th responses. The results are reported as average of triplicate cultures and proliferation is indicated as stimulation indices (SI) [mean counts per minute (c.p.m.) in experimental wells ÷ mean c.p.m. in control wells without antigen). The standard deviation for each group is indicated by a bar. The c.p.m. of cells without antigen was 1566–1921. [³H]TdR, [³H]thymidine.

Next, it was investigated whether i.n. immunization with PAM₂-K25V was effective in inducing local mucosal Th responses. Significant PADRE-specific Th-cell responses were found in the draining cervical LNs following a prime-boost i.n. immunization with the lipopeptide (Fig. 3a). These responses were at a similar level compared to those detected in the inguinal LNs from the same animals (Fig. 3b) (P<0·01 for the comparison of Th-cell responses between cervical and inguinal LNs). PADRE-specific Th-cell responses induced by i.n. immunization with the lipopeptide were CD4 dependent. Proliferation was abrogated when responding cells were incubated in vitro with anti-CD4 antibodies, but were not affected by anti-CD8 antibodies (SI=16·1 without antibodies, and SI=3·5 and 14·9 when antibodies against CD4 and CD8 were used, respectively, data not shown).

Figure 3.

Intranasal administration with a lipopeptide results in local mucosal T helper (Th) cell responses. Groups of human leucocyte antigen (HLA)-A*0201 transgenic (Tg) mice (n=4) were immunized intranasally with either PAM₂ K25V (filled squares and triangles) or with equimolar amounts of the unlipidated version (open squares and triangles). PADRE-specific (filled and open squares) Th cell responses were assessed (a) in local cervical lymph node (LN) cells and (b) in remote inguinal LN cells. The TT_615–629 peptide was used as a control for measurement of Th responses (open triangles and squares). Data represent an average of triplicate cultures of a representative experiment and proliferation is indicated as stimulation indices (SI). The standard deviation for each group is indicated by a bar. Counts per minute (c.p.m.) of cells without antigen was 2743–2987.

Intranasal immunization with PAM₂-K25V generates systemic CMV-specific CTL responses

To evaluate whether mucosal immunization with PAM₂-K25V could induce systemic CTL responses, separate groups of HLA-A*0201 Tg mice were immunized i.n. either with the lipopeptide or with an equimolar amount of the unlipidated version. Peptide-specific CTL responses were examined in the spleens of all groups of mice. As shown in Fig. 4(a), i.n. administration of PAM₂-K25V induced pp65_495–503-specific CTL responses in spleen cells more efficiently compared to mice immunized with the unlipidated peptide (P<0·01, comparison of CTL responses stimulated by lipidated or unlipidated peptide vaccine). The specificity of these responses was confirmed by the lack of CTL activity against the control p53_149–157 peptide³³ (Fig. 4a). As shown in Fig. 4(b), the CTLs induced were of the CD8 phenotype, as the specific lysis was abrogated by antibodies binding to CD8 but only slightly affected with antibodies against CD4 (P<0·05). We also investigated whether the CTL generated using PAM₂-K25V administered i.n. would recognize CMV-infected human fibroblasts (Fig. 4c). A prime-boost i.n. immunization induced CTLs that were able to lyse CMV-infected fibroblasts expressing HLA-A*0201 (Fig. 4c). These responses were both virus- and HLA-specific, as far less activity was detected against either mock-infected HLA-A*0201 or HLA mismatched, CMV-infected fibroblasts (Fig. 4c).

Figure 4.

Systemic CD8⁺ cytotoxic T lymphocyte (CTL) responses induced following intranasal (i.n.) administration of a lipopeptide. (a) Systemic pp65_495–503-specific (open and filled squares) CTL responses were assessed in the spleens of human leucocyte antigen (HLA)-A*0201 transgenic (Tg) mice (n=3) after i.n. administration of PAM₂-K25V (filled squares and triangles) or the unlipidated peptide version (open squares and triangles) using a chromium release assay (CRA), as described in the Materials and methods. The p53_149–157 peptide was used as a control (open and filled triangles). (b) At an effector : target (E : T) ratio of 30, monoclonal antibody (mAb) blocking of CTL activity was performed by adding serial dilutions of anti-CD8 or anti-CD4 during the lysis part of the CRA assay. (c) Cytomegalovirus (CMV)-specific and HLA-restricted CTL responses determined against either HLA matched or mismatched CMV-infected fibroblasts in a CRA assay. The standard deviation for each group is indicated by a bar.

Mucosal route results in a DTH immune response

A DTH immune response was assessed after in vivo challenge with recall (PADRE) or control (TT_615–629) peptide administered into the footpad, following i.n. immunization with the PAM₂-K25V lipopeptide. As shown in Fig. 5(a), a PADRE-specific DTH response was observed in mice immunized i.n. with PAM₂-K25V lipopeptide. These cellular responses were comparable to those induced by s.c. injection (Fig. 5a) (P<0·01 comparing the s.c. with the i.n. route). The development of a DTH response after challenge with recall PADRE peptide (Fig. 5b, panels a and b), but not with the control peptide (TT_615–629, Fig. 5b, panels c and d), was accompanied by a mononuclear cellular infiltrate, as determined by footpad histology. No DTH response was observed in animals that received the unlipidated peptide (data not shown).

Discussion

Strategies that avoid the use of needles and adjuvants are highly desirable for human vaccines, as such techniques will greatly reduce barriers to immunization. We and others have obtained promising results in recent years showing that parenteral injections of synthetic lipopeptides, without adjuvant, are highly effective in inducing systemic B-cell, CTL and Th responses in both experimental animal models and humans.^1–6,39 Taken together, the data in this report demonstrate that i.n. immunization with a lipopeptide, without a mucosal adjuvant, is effective in generating systemic antigen-specific Th responses (i.e. in cells taken at a remote distance from the antigen-delivery site) and in the mucosa. The lipid moiety was absolutely essential as these T-cell responses were substantially reduced when the unlipidated peptide was employed.

Immunization via an application to mucosal surfaces without physical penetration by needles, and without adjuvant, would greatly simplify vaccination programmes. However, mucosal immunization by subunit vaccines has proven to be a major challenge.^12,23,24,40 Toxicity is a major factor limiting the development and use of adjuvants for human mucosally delivered vaccines.^11,41 Recent progress in genetic detoxification have produced both CT and LT mutants that exhibit reduced toxicity.^25,28,30 A number of reports have evaluated the efficacy of these mutants as mucosal adjuvants.^{26,34,42–45} Although CT and LT mutants show reduced toxicity, their immunogenicity may prevent widespread repeated use in humans, because pre-existing immunity to CT and LT reduces adjuvant activity.^25,29,30

The approach presented in this report benefits from the fact that mucosal administration with a synthetic lipopeptide in the absence of mucosal adjuvants, such as CT or LT, is sufficient to achieve a wide range of systemic immune responses (i.e. CTL, Th and DTH responses). The lipid moiety was absolutely required to generate an effective immune response. In addition, after i.n. administration of lipopeptides, no general or local adverse effects were observed following close examination of the nasal cavity. For technical reasons, we did not detect any mucosal CTL response following i.n. administration of the lipopeptide. The low number of cells recovered from the local draining cervical LNs prevented us from stimulating and maintaining a sufficient number of effector cells, and to generate enough of them would represent a challenge beyond the scope of the present study.

Newly devised mucosal immunization methods feature i.n. application of monomeric, multimeric (lipo-MAP^15,21), or polyoxime peptides²² conjugated to the relatively complex lipopeptide chain P₃CSS as a means to stimulate systemic immune responses. However, the P₃CSS moiety may have limited clinical utility because it has been shown to be mitogenic and somewhat toxic, even without conjugation to immunogenic peptides.^46,47 Nonetheless, P₃CSS only confers T-cell responsiveness to a peptide when it is covalently linked.^7,48 In contrast, the straight-chain palmitic acid derivatives we have used are not immunogenic by themselves, but they can confer immunogenic properties to covalently attached peptide epitopes.^2,3,49 We report one of the most probable and promising immunization procedures using i.n. administration of a relatively simple and non-toxic lipopeptide construct, confirming and extending the above findings. i.n. administration with lipopeptides containing B-cell epitopes also induces strong serum immunoglobulin G (IgG) antibody responses, indicating that the mucosal immunogenicity of lipopeptides can be extended to humoral immunity (L. BenMohamed et al., manuscript in preparation). In addition, the observation that i.n. administration of lipopeptides has immunogenicity comparable to that of the s.c. route is of particular interest. However, the significance of this approach in the development of mucosal vaccines will hinge upon the clinical effectiveness of the strategy.

The underlying mechanisms by which lipopeptides elicit both mucosal and systemic immune responses without adjuvant are only partially understood. It is assumed to be based upon a complex multistep phenomenon resulting from direct interactions of lipopeptides with different cell types at the presumed site of uptake and immune induction.^4,8,49,50 Our results revealed that a lipopeptide, but not the unlipidated version, administered mucosally is able to induce systemic immune responses. It is probable that lipidation of peptides is critical for directing mucosally administered peptide epitopes across the epithelial barrier and gaining access to systemic compartments. In addition to the enhancement of lipopeptide penetration via mucosal membranes, the excellent immunogenicity of lipopeptides may be a result of the fact that the lipid moiety protects them from degradation by mucosal enzymes.

In summary, the present study provides the first evidence that i.n. administration with a synthetic lipopeptide is effective in eliciting antigen-specific systemic cellular immunity, including CTL, Th and DTH, with mucosal Th also induced in the local cervical LNs. Although the underlying mechanism of mucosal immunogenicity of lipopeptides is not yet clear, the observations that the vaccine formulation used can induce such a wide range of immune responses, without a toxic mucosal adjuvant, is encouraging. The vaccination strategy described here requires no special equipment and may ultimately lower the barriers to immunization associated with current injection methods. Application of lipopeptides via mucosal routes is a highly promising concept that should be a fruitful avenue for the development of mucosal vaccines.

Abbreviations

CTL

cytotoxic T lymphocyte

DTH

delayedtype hypersensitivity

i.n

intranasal

i.p

intraperitoneal

(PAM)₂-Th-CTL

dipalmitoylated lipopeptide

s.c.

subcutaneous

SI

stimulation index

Th

T helpers