Cell-Penetrating Peptide-Linked Polymers as Carriers for Mucosal

An equivalent dose of the cholera toxin B subunit (CTB), which was used as a positive control, was superior to the polymer in enhancing antibody produ...
0 downloads 0 Views 2MB Size
Article pubs.acs.org/molecularpharmaceutics

Cell-Penetrating Peptide-Linked Polymers as Carriers for Mucosal Vaccine Delivery Shinji Sakuma,*,† Masaya Suita,† Saki Inoue,† Yoko Marui,† Kazuhiro Nishida,† Yoshie Masaoka,† Makoto Kataoka,† Shinji Yamashita,† Noriko Nakajima,‡ Norihiro Shinkai,‡ Hitoshi Yamauchi,‡ Ken-ichiro Hiwatari,§ Hiroyuki Tachikawa,§ Ryoji Kimura,§ Tomofumi Uto,∥,⊥ and Masanori Baba*,∥ †

Faculty of Pharmaceutical Sciences, Setsunan University, 45-1, Nagaotoge-cho, Hirakata, Osaka 573-0101, Japan Research and Development Department, Nipropatch Co. Ltd., 8-1, Minamisakae-cho, Kasukabe, Saitama 344-0057, Japan § Advanced Materials Laboratory, Adeka Co., 7-2-34 Higashiogu, Arakawa-ku, Tokyo 116-8553, Japan ∥ Graduate School of Medical and Dental Sciences, Kagoshima University, 8-35-1, Sakuragaoka, Kagoshima, Kagoshima 890-8544, Japan ‡

ABSTRACT: We evaluated the potential of poly(N-vinylacetamide-co-acrylic acid) modified with D-octaarginine, which is a typical cell-penetrating peptide, as a carrier for mucosal vaccine delivery. Mice were nasally inoculated four times every seventh day with PBS containing ovalbumin with or without the D-octaarginine-linked polymer. The polymer enhanced the production of ovalbumin-specific immunoglobulin G (IgG) and secreted immunoglobulin A (IgA) in the serum and the nasal cavity, respectively. Ovalbumin internalized into nasal epithelial cells appeared to stimulate IgA production. Ovalbumin transferred to systemic circulation possibly enhanced IgG production. An equivalent dose of the cholera toxin B subunit (CTB), which was used as a positive control, was superior to the polymer in enhancing antibody production; however, dose escalation of the polymer overcame this disadvantage. A similar immunization profile was also observed when ovalbumin was replaced with influenza virus HA vaccines. The polymer induced a vaccine-specific immune response identical to that induced by CTB, irrespective of the antibody type, when its dose was 10 times that of CTB. Our cell-penetrating peptide-linked polymer is a potential candidate for antigen carriers that induce humoral immunity on the mucosal surface and in systemic circulation when nasally coadministered with antigens. KEYWORDS: mucosal vaccine delivery, cell-penetrating peptide, cell-penetrating peptide-linked polymer, oligoarginine, poly(N-vinylacetamide-co-acrylic acid)



INTRODUCTION

by conventional parenteral immunization. Secreted IgA represents the first immunological barrier to pathogens that infect the epithelial surface.6,7 However, because most antigens are poor immunogens when solely applied to the mucosa, adjuvants/antigen carriers, which are coadministered with vaccines, are essential for significant induction of immunity.

Intranasal vaccines have been studied as one of the most effective and adaptive immunization strategies for protecting hosts against infectious pathogens that invade epithelial cells at the mucosa such as the influenza virus.1−5 This strategy is expected to provide significant production of antigen-specific immunoglobulin G (IgG) and secreted immunoglobulin A (IgA) in the serum and at the mucosa, respectively. IgG, whose production is also enhanced by systemic immunization with subcutaneous or intramuscular vaccine administration, prevents severe infection-related symptoms. Mucosal immunity is mainly mediated by secreted IgA, whose production is not stimulated © 2012 American Chemical Society

Received: Revised: Accepted: Published: 2933

June 15, 2012 July 28, 2012 September 6, 2012 September 6, 2012 dx.doi.org/10.1021/mp300329r | Mol. Pharmaceutics 2012, 9, 2933−2941

Molecular Pharmaceutics

Article

Our previous studies18,19 revealed that oligoarginine-linked PNVA-co-AA significantly enhanced in vivo permeation of insulin through the nasal membrane in mice. The order of the penetration-enhancing function of oligoarginine anchored to PNVA-co-AA was 8 arginine residues ≥ 12 residues and Dconfiguration ≥ L-configuration. No penetration enhancement was observed when the polymer was substituted with its individual components, such as intact D-octaarginine; this demonstrated that only D-octaarginine anchored chemically to the polymeric platform enhanced membrane permeation of poorly membrane-permeable bioactive molecules. In vitro cell studies indicated that D-octaarginine-linked PNVA-co-AA remained on the cell membrane and the bioactive molecules were continuously internalized into cells mainly via macropinocytosis repeated for the individual peptidyl branches in the polymer backbone, as expected. The unique characteristics of oligoarginine-linked polymers potentially facilitate the intracellular delivery of bioactive molecules without the concomitant cellular uptake of the polymer themselves. Here, we report the potential of our polymer as a carrier for mucosal vaccine delivery.

Toxin-based adjuvants, such as the cholera toxin (CT) and Escherichia coli heat-labile toxin (HLT), were first used as adjuvants for mucosal vaccines.8,9 These enterotoxins enhanced the production of antigen-specific IgG in the serum and secreted IgA at the mucosa; however, severe diarrhea was simultaneously observed. Isolation of the B subunit of CT successfully removed the toxicity, while significant induction of mucosal and systemic immunity was retained.10 In 2000, healthy volunteers were nasally inoculated with a combination of inactivated influenza vaccines and the HLT adjuvant encapsulated into liposomes; however, a high incidence of Bell's palsy was reported in this preclinical trial.11−13 Safe and effective adjuvants/antigen carriers are needed for establishment of mucosal vaccines. Recent studies indicated that the synthetic double-stranded RNA poly(I:C) and modified pulmonary surfactants were safe and potent adjuvants for mucosal influenza vaccination.3,14 Biodegradable poly(γ-glutamic acid) nanoparticles that incorporate antigens are also expected to be sophisticated antigen carriers with unique immunological properties.15−17 We have been independently investigating cell-penetrating peptide-linked polymers as a new class of penetration enhancers.18,19 Cell-penetrating peptides are oligopeptides that are mainly composed of cationic amino acids such as arginine. They were discovered during studies on the human immunodeficiency virus (HIV), and the most typical cellpenetrating peptides are the HIV-1 Tat protein (48−60), oligoarginine, and penetratin. It is considered that cellpenetrating peptides are predominantly internalized into cells via macropinocytosis, which is a distinct endocytic pathway.20−22 Figure 1 shows the chemical structure of our



EXPERIMENTAL SECTION Materials. PNVA-co-AA sodium salts were gifted by Showa Denko Co. Ltd. (Tokyo, Japan). PNVA-co-AA [NVA units/AA units, 70/30; weight-average molecular weight (Mw), 1600 kDa], which is assigned as GE-160, was used in this study. Octaarginine (D-configuration) with amidated terminal carboxyl groups was obtained from Kokusan Chemical Co. Ltd. (Tokyo, Japan). All other chemicals were commercial products of analytical or reagent grade and were used without further purification. Dulbecco's phosphate-buffered saline, modified (PBS without calcium chloride and magnesium chloride), and ovalbumin (OVA) were purchased from Sigma-Aldrich (St. Louis, MO). Influenza virus hemagglutinin (HA) vaccines were obtained from Denka Seiken Co. Ltd. (Tokyo, Japan). Bovine serum albumin (BSA), cholera toxin B subunit (CTB), and polyoxyethylene (20) sorbitan monolaurate (Tween 20) were purchased from Wako Pure Chemical Industries Co. Ltd. (Osaka, Japan). Skim milk was obtained from Becton Dickinson Diagnostics (Sparks, MD). Horseradish peroxidase (HRP)-conjugated goat antimouse IgG and IgA antibodies were purchased from Southern Biotechnology Associates, Inc. (Birmingham, AL). TMB 1-Component Microwell Peroxidase Substrate, SureBlue Reserve, and TMB Stop Solution were furnished by Kirkegaard & Perry Laboratories, Inc. (Gaithersburg, MD), and 96-well plates (Immuno Modules, framed, 96 wells per frame, MaxiSorp surface, Nunc brand) were purchased from Thermo Fisher Scientific Inc. (Waltham, MA). Synthesis and Characterization of D-OctaarginineLinked PNVA-co-AA. D-Octaarginine-linked PNVA-co-AA was prepared in the same manner as previously described.18 Briefly, PNVA-co-AA was desalted, and its free form was dissolved in ice-cooled N,N′-dimethylformamide (DMF) containing an excess amount of N,N′-dicyclohexylcarbodiimide (DCC). N-Hydroxysuccinimide (HOSu) was added to the solution, and the DCC-activated carboxyl groups of PNVA-coAA were reacted with HOSu through an overnight incubation of the mixture at 60 °C. The resulting PNVA-co-AA Nhydroxysuccinimide ester (PNVA-co-AA-OSu) was collected as a precipitate by pouring the reaction solution into excess acetonitrile, and the precipitate was dried under reduced

Figure 1. Chemical structure of oligoarginine-linked PNVA-co-AA.

polymer: poly(N-vinylacetamide-co-acrylic acid) (PNVA-coAA) modified with oligoarginine. This biocompatible polymer was designed with a unique strategy that the polymers enable poorly membrane-permeable bioactive molecules physically mixed with them to effectively penetrate cell membranes without their concomitant cellular uptake. We expect that cellular uptake, which mainly acts via macropinocytosis, is initiated through the biorecognition of peptidyl branches in the backbone of the cell-penetrating peptide-linked polymer applied to the cell membrane along with the bioactive molecules. Because the uptake occurs simultaneously and competitively at multiple sites of the single-stranded polymer, cells fail to take up peptides anchored to the polymeric platform via a nonbiodegradable linker. However, the bioactive molecules that are present in the periphery of the peptidyl branches are taken up into the cells incidentally. Continuous internalization of bioactive molecules into cells is observed because the unaccomplished uptake of cell-penetrating peptides grafted onto the polymer backbone occurs repeatedly as long as the peptide-linked polymers remain on the cell membrane. 2934

dx.doi.org/10.1021/mp300329r | Mol. Pharmaceutics 2012, 9, 2933−2941

Molecular Pharmaceutics

Article

method described previously.17 Briefly, 96-well plates were coated with OVA (10 μg/mL) dissolved in a carbonate− bicarbonate buffer (Na2CO3, 3.0 mg/mL; NaHCO3, 6.0 mg/ mL; pH 9.6) through an overnight incubation at 4 °C. After removal of the OVA solution, the plate was washed with PBS containing 0.05% (v/v) Tween 20 (T-PBS) and filled with a blocking buffer [T-PBS containing 1% (w/v) BSA and 1% (w/ v) skim milk] for 2 h at room temperature. The blocking buffer was removed, the plate was washed with T-PBS, and a 2-fold dilution series of samples (sera and nasal wash fluids), which were diluted with the blocking buffer, was prepared in wells. After overnight incubation at 4 °C, samples were removed, and the plate was washed with T-PBS. HRP-conjugated goat antimouse IgG and IgA antibodies, which were diluted 1:4000 and 1:2000 with the blocking buffer, respectively, were added to the corresponding wells. After a 2 h incubation at room temperature, the solution containing the antimouse antibodies was removed, and the plate was washed with T-PBS. TMB 1Component Microwell Peroxidase Substrate was added to each well, and the plate was incubated in the dark for 10 min at room temperature. The colorimetric reaction was terminated by adding a stop solution (TMB Stop Solution), and the absorbance of each well was measured at 450 nm. End point titers of OVA-specific antibodies were determined from the x-axis intercept of the dilution curve. Each value of end point titers is presented as the mean and standard deviation (SD). Statistical significance was assessed with the unpaired Student's t test, and p values of 0.05 or less were considered to be statistically significant. During experiments, mice inoculated with PBS containing OVA (a negative control) did not often exhibit an immune response. In this case, statistical significance was not assessed because of the lack of data variance in the control study. Influenza Virus HA Vaccine-Specific Antibodies. Titers of influenza virus HA vaccine-specific antibodies were assayed using a modification of the ELISA method reported by Mizuno et al.3 Briefly, 96-well plates were coated with influenza virus HA vaccines (1.0 μg/mL) and BSA (1.0 μg/mL) in PBS through an overnight incubation at 4 °C. After removal of PBS, the plate was washed with a Tris-HCl buffer (50 mM, pH 8.0) containing 0.14 M NaCl and 0.05% (v/v) Tween 20 (TTS) and filled with a blocking buffer [TTS containing 1% (w/v) BSA] for 1 h at room temperature. The blocking buffer was removed, the plate was washed with TTS, and a 2-fold dilution series of samples, which were diluted with the blocking buffer, was prepared in wells. After a 2 h incubation at room temperature, samples were removed, and the plate was washed with TTS. HRP-conjugated goat antimouse IgG and IgA antibodies (diluted 1:4000 and 1:2000 with the blocking buffer, respectively) were then added to the corresponding wells. After a 1 h incubation at room temperature, the solution containing the antimouse goat antibodies was removed, and the plate was washed with TTS. The plate was treated as described above, and the end point titers of influenza virus HA vaccine-specific antibodies were determined.

pressure. PNVA-co-AA-OSu and D-octaarginine were dissolved in ice-cooled DMF. The molar concentration of the former as a carboxyl group equivalent was identical to that of the latter as a terminal amino group equivalent. The terminal amino groups of D-octaarginine were coupled to the carboxyl groups of PNVA-co-AA by replacing the N-oxysuccinimide groups through an overnight incubation of the mixture at 40 °C. After the solution was poured into excess acetonitrile, the precipitate was collected and then dialyzed in purified water. After lyophilization, D-octaarginine-linked PNVA-co-AA was obtained. The degree of D-octaarginine grafted onto PNVA-co-AA was determined by the integral ratio of methine proton signals at the α-position of D-octaarginine to methylene proton signals of the N-vinylacetamide (NVA) unit on the 1H NMR spectrum of the polymer. The degree was expressed as the percentage of monomer units of acrylic acid (AA) grafting D-octaarginine to the total number of monomer units. D-Octaarginine-linked PNVA-co-AA with a grafting degree of 11%, whose performance was thoroughly examined in our previous study,19 was used in this study. Aggregation formation in the polymer solution and a mixture of polymers and below-mentioned antigens (OVA and influenza virus HA vaccines) was analyzed using dynamic and electrophoretic light-scattering spectrophotometry (ELSZ-2, Otsuka Electronics Co., Osaka, Japan). PBS was used as a solvent. Animal Studies. Immunization of Mice with OVA. All animal experiments were approved by the Ethical Review Committee of Setsunan University. Seven week old female BALB/c mice (weight, ca. 20 g; not fasted; n = 4) were used in each experiment. Four kinds of dosing solution, PBS, PBS containing OVA, PBS containing OVA and D-octaargininelinked PNVA-co-AA, and PBS containing OVA and CTB, were prepared. Mice were nasally inoculated with each dosing solution under ether anesthesia. Doses of OVA, D-octaargininelinked polymer, and CTB were set to 1−10, 5−40, and 10 μg, respectively/20 μL of dosing solution/mouse. The inoculation was performed on either days 0 and 7 (two times) or days 0, 7, 14, and 21 (four times). Seven days after the final inoculation, mice were anesthetized with sodium pentobarbital (DainipponSumitomo Pharma Co. Ltd., Osaka, Japan). The cervix was opened, and the trachea was observed. A tracheal cannula was inserted toward the nose by using a polyethylene tube (Hibiki; size, 3; inner diameter, 0.5 mm; and outer diameter, 1.0 mm). PBS (0.2 mL) was poured into the nasopharyngeal cavity through the cannula, and the fluid that drained from the nasal holes was collected. Blood samples were subsequently taken from the heart. The blood was centrifuged to obtain serum samples. Both serum samples and nasal wash fluids were frozen until analysis. Immunization of Mice with Influenza Virus HA Vaccines. BALB/c mice were nasally inoculated on days 0, 7, 14, and 21 with 20 μL of PBS containing 0.05−0.1 μg of influenza virus HA vaccines with or without 40−100 μg of D-octaargininelinked PNVA-co-AA. Seven days after the final inoculation, mice were sacrificed, and serum samples and nasal wash fluids were collected as described above. The vaccine alone and a mixture of the vaccine and CTB were used as negative and positive controls, respectively. The dose of CTB was set to 10 μg/mouse. Assay for Antigen-Specific Antibodies. OVA-Specific Antibodies. Titers of OVA-specific antibodies were measured using the enzyme-linked immunosorbent assay (ELISA)



RESULTS Aggregation Formation of D -Octaargine-Linked PNVA-co-AA Solution and a Mixture of the Polymers and Antigens. D-Octaargine-linked PNVA-co-AA was dissolved in PBS at a concentration of 1.0 mg/mL. The solution was clear visually, and aggregation formation such as particles was not detected when the solution was analyzed using 2935

dx.doi.org/10.1021/mp300329r | Mol. Pharmaceutics 2012, 9, 2933−2941

Molecular Pharmaceutics

Article

Figure 2. Levels of OVA-specific IgG in sera (a−d) and OVA-specific secreted IgA in nasal wash fluids (e and f) of individual mice. Mice were nasally inoculated with PBS (a), PBS containing OVA (b and e), PBS containing OVA and D-octaarginine-linked PNVA-co-AA (c and f), or PBS containing OVA and CTB (d). The inoculation was performed on either days 0 and 7 (a−d) or days 0, 7, 14, and 21 (e and f). Doses of OVA, D-octaargininelinked PNVA-co-AA, and CTB were all set to 10 μg/20 μL of dosing solution/mouse.

Figure 3. Levels of OVA-specific IgG in sera (a) and OVA-specific secreted IgA in nasal wash fluids (b) of mice that were nasally inoculated with each dosing solution (PBS, white bar; PBS containing OVA, left hashed bar; PBS containing OVA and D-octaarginine-linked PNVA-co-AA, black bar; and PBS containing OVA and CTB, right hashed bar). Each bar represents end point titers of the antibody (mean ± s.d., n = 4). The number of inoculations was set to 2 (days 0 and 7) or 4 (days 0, 7, 14, and 21). Doses of OVA, D-octaarginine-linked PNVA-co-AA, and CTB were all set to 10 μg/20 μL of dosing solution/mouse.

2936

dx.doi.org/10.1021/mp300329r | Mol. Pharmaceutics 2012, 9, 2933−2941

Molecular Pharmaceutics

Article

Figure 4. Levels of OVA-specific IgG in sera (a) and OVA-specific secreted IgA in nasal wash fluids (b) of mice that were nasally inoculated four times (days 0, 7, 14, and 21) with each dosing solution (PBS, white bar; PBS containing OVA, left hashed bar; PBS containing OVA and Doctaarginine-linked PNVA-co-AA, black bar; and PBS containing OVA and CTB, right hashed bar). Each bar represents end point titers of the antibody (mean ± SD, n = 4). Doses of OVA, D-octaarginine-linked PNVA-co-AA, and CTB were set to 1, 10, and 10 μg, respectively/20 μL of dosing solution/mouse.

dynamic light-scattering spectrophotometry. OVA was separately dissolved in PBS at a concentration of 1.0 mg/mL, and it was mixed with an equivalent volume of the polymer solution. The mixture, which was used as a dosing solution in belowmentioned animal studies, was slightly cloudy, and instrumental analysis revealed that there were particles with an average diameter and ζ-potential of 460 nm and 3.9 mV, respectively. The OVA concentration was next reduced to 0.1 mg/mL, and it was mixed with an equivalent volume of the polymer solution. The mixture, which corresponded to a dosing solution used in below-mentioned animal studies, was clear visually; however, particles with an average diameter and ζ-potential of 245 nm and 8.5 mV, respectively, were instrumentally detected. The OVA concentration was further reduced to 0.025 mg/mL, and a mixture was prepared in a similar manner described above. The polymer/OVA weight ratio (40/1) was equal to that of a dosing solution used in below-mentioned animal studies (the highest dose of the polymer), although the concentration was adjusted to one-fourth of the dosing solution. No aggregation formation in the mixture was observed visually and instrumentally. OVA was replaced with influenza virus HA vaccines, and the polymer/the vaccine weight ratio was set to 800/1 and 2000/1 on the basis of animal experiments described below. The polymer concentration in the mixture was adjusted to 1.0 mg/mL. Aggregation formation was not detected when the mixture was analyzed using dynamic light-scattering spectrophotometry, irrespective of the weight ratio. Immunization with a High Dose of OVA in the Presence of D-Octaargine-Linked PNVA-co-AA. We first examined whether a humoral immune response was induced by the intranasal administration of a mixture of OVA and Doctaargine-linked PNVA-co-AA. Figures 2 and 3 show the levels of OVA-specific antibodies in sera and nasal wash fluids of mice inoculated with each dosing solution. The absorbance at 450 nm (the values in the y-axis of Figure 2) increases with an elevation of antibody levels. No production of serum OVAspecific IgG was observed when PBS, which was a vehicle, was administered (Figure 2a). As is evident from an increase in the absorbance (Figure 2b), a quarter of the mice were immunized when OVA alone was administered twice at a dose of 10 μg/ mouse (a negative control). Similar results were observed when the test was repeated (data not shown). An elevation of OVAspecific IgG levels was observed in all mice inoculated twice with PBS containing OVA and CTB, which was used as a

positive control (Figure 2d). When OVA was coadministered twice with D-octaargine-linked PNVA-co-AA, three mice were immunized; however, the production of OVA-specific IgG was not enhanced in sera of the remaining mouse (Figure 2c). End point titers of OVA-specific IgG were determined to qualify the strength of the immune response (the titer increases as the immune response becomes stronger). As shown in Figure 3a, the levels of OVA-specific IgG elevated significantly when OVA was coadministered twice with CTB. A similar elevation was observed when CTB was replaced with D-octaarginine-linked PNVA-co-AA. However, the elevation was not statistically significant due to large variations in the OVA-specific IgG levels in sera of mice inoculated with PBS containing OVA, irrespective of coadministration of the polymer. The levels of OVA-specific secreted IgA in nasal wash fluids were preliminary examined. As shown in Figure 3b, the antibody was not detected when OVA alone was administered twice at a dose of 10 μg/mouse. The antibody levels tended to elevate through coadministration of OVA with D-octaarginelinked PNVA-co-AA on days 0 and 7. When the number of inoculations was doubled, a clear elevation of OVA-specific secreted IgA levels was observed in half the number of mice inoculated with PBS containing OVA and the polymer (Figure 2f). However, because of large variations in the antibody levels, the immune response did not significantly differ from that of its counterpart whose OVA-specific secreted IgA levels barely elevated even when the number of inoculations was doubled (Figure 2e). Immunization with a Low Dose of OVA in the Presence of D-Octaargine-Linked PNVA-co-AA. The OVA dose was reduced to 1 μg/mouse to avoid the immune response of mice inoculated with OVA alone. The number of inoculations was set to four because the immune response at the nasal mucosa tended to be stimulated through an increase in the number of inoculations. As shown in Figure 4a, a 90% reduction of the OVA dose resulted in a decrease of the serum OVA-induced immune response in the absence of additives (Doctaargine-linked PNVA-co-AA and CTB). The elevation of the OVA-specific IgG levels, which was observed in the negative control with a high dose of OVA (Figure 2b), disappeared, and the variation in the antibody levels was small. The OVA-specific IgG levels clearly elevated when OVA was coadministered with D-octaargine-linked PNVA-co-AA and CTB. 2937

dx.doi.org/10.1021/mp300329r | Mol. Pharmaceutics 2012, 9, 2933−2941

Molecular Pharmaceutics

Article

Figure 5. Effect of D-octaarginine-linked PNVA-co-AA dose on the levels of OVA-specific IgG in sera (a) and OVA-specific secreted IgA in nasal wash fluids (b) of mice. Mice were nasally inoculated four times (days 0, 7, 14, and 21) with PBS containing OVA (left hashed bar), PBS containing OVA and D-octaarginine-linked PNVA-co-AA (black bar), or PBS containing OVA and CTB (right hashed bar). Each bar represents end point titers of the antibody (mean ± SD, n = 4). Doses of OVA, D-octaarginine-linked PNVA-co-AA, and CTB were set to 1, 5−40, and 10 μg, respectively/20 μL of dosing solution/mouse.

Figure 6. Levels of influenza virus HA vaccine-specific IgG in sera (a) and the vaccine-specific secreted IgA in nasal wash fluids (b) of mice that were nasally inoculated four times (days 0, 7, 14, and 21) with each dosing solution (PBS, white bar; PBS containing the vaccine, left hashed bar; PBS containing the vaccine and D-octaarginine-linked PNVA-co-AA, black bar; and PBS containing the vaccine and CTB, right hashed bar). Each bar represents end point titers of the antibody (mean ± SD, n = 4). Doses of the vaccine, D-octaarginine-linked PNVA-co-AA, and CTB were set to 0.05, 40−100, and 10 μg, respectively/20 μL of dosing solution/mouse.

specific IgG dramatically increased, and the variation decreased when the polymer dose was doubled (10 μg/mouse); the immune response significantly differed from that of the negative control. The antibody levels and the variation increased and decreased, respectively, as the polymer dose further increased. The production of serum IgG reached a level comparable to that observed in the positive control. The levels of OVA-specific secreted IgA in the nasal wash fluids of mice inoculated with PBS containing OVA and Doctaarginine-linked PNVA-co-AA clearly elevated with an increase in the polymer dose (Figure 5b). At a polymer dose of 40 μg/mouse, the average antibody level (end point titer) reached approximately 29. This maximal level still differed significantly from that observed in the positive control; however, the difference was also comparable, as seen in the case of the serum antibody. Immunization with Influenza Virus HA Vaccines in the Presence of D-Octaargine-Linked PNVA-co-AA. To examine the effect of the antigen type on the D-octaarginelinked PNVA-co-AA-induced immune response, influenza virus HA vaccines were substituted for OVA. Four inoculations were performed as we predicted that this would be preferable to two

Intranasal OVA-specific secreted IgA was not detected when OVA was nasally administered four times at a dose of 1 μg/ mouse (Figure 4b). The antibody levels elevated when OVA was coadministered with D-octaargine-linked PNVA-co-AA (statistical significance was not assessed due to the lack of data variance in the negative control). However, the antibody levels were considerably lower than those observed in the nasal wash fluids of mice inoculated with PBS containing OVA and CTB. Effect of D-Octaarginine-Linked PNVA-co-AA Dose on Antibody Levels. Because the D-octaargine-linked PNVA-coAA-induced immune response was poorer than that induced by an equivalent dose of CTB, the effect of the polymer dose on antibody levels was examined. The OVA dose and the number of inoculations were set to 1 μg/mouse and 4, respectively. As shown in Figure 5a, the average level of OVA-specific IgG (end point titer) in sera of mice, to which OVA was administered with D-octaarginine-linked PNVA-co-AA whose dose was 5 μg/ mouse, was approximately 28 times that without the polymer. However, because a quarter of the mice in the former group were not immunized (detailed data not shown), the variation in the antibody levels was large. The average level of serum OVA2938

dx.doi.org/10.1021/mp300329r | Mol. Pharmaceutics 2012, 9, 2933−2941

Molecular Pharmaceutics

Article

production of serum IgG. The production of intranasal secreted IgA was also stimulated when OVA was coadministered with the polymer. However, the polymer was inferior to an equivalent dose of CTB, particularly in the induction of mucosal immunity. Dose escalation of the polymer overcame this disadvantage (Figure 5). The OVA-specific immune response reached a level comparable to that observed in the positive control, irrespective of the antibody type, when the polymer dose was four times that of CTB. We have separately confirmed that fluorescein isothiocyanate-dextran with a molecular weight of 40 kDa, whose magnitude is similar to that of OVA, is taken up into cultured cells when incubated with D-octaarginine-linked PNVA-co-AA (data not shown). This finding and the penetration-enhancing mechanism of the polymer indicate that OVA applied to the nasal mucosa was internalized into epithelial cells in the presence of the polymer. It appears that the internalized OVA stimulated the production of intranasal secreted IgA. OVA transferred to systemic circulation possibly enhanced the production of serum IgG. It appears that the aggregation formation between the polymers and the OVA did not affect the strength of the immune response. Details of induction processes of mucosal and systemic immunity will be examined in future studies. OVA, which is a typical model antigen, has relatively high immunogenicity. We next examined whether D-octaarginelinked PNVA-co-AA induced mucosal and systemic immunity, irrespective of the antigen type. OVA was replaced with influenza virus HA vaccines, which are clinically used. Spectrophotometric studies suggested that there was no association between the polymers and the vaccines. The vaccine dose was set to 0.05 μg/mouse to avoid significant production of vaccine-specific antibodies in mice that were nasally inoculated four times with the vaccine alone. The production of serum IgG and intranasal secreted IgA was clearly enhanced when the vaccine was coadministered with the polymer whose dose was 40 μg/mouse (Figure 6), as seen in the case of OVA. Surprising immunization was observed through further dose escalation of the polymer. The vaccinespecific immune response reached a level identical to that induced by CTB, irrespective of the antibody type, when the polymer dose was 10 times that of CTB. The currently available influenza virus vaccines are administered subcutaneously or intramuscularly. The parenteral vaccines effectively prevent severe influenza virus infection because they predominantly induce IgG-mediated protection in systemic immune compartments. The number of patients who suffer from respiratory diseases due to secondary bacterial infections, which occasionally result in death, is factually reduced.24 However, this systemic immunization offers insufficient protection at the mucosal surface, which is the viral infection site. Additionally, parenteral vaccines are less protective against antigenic drift variants within a subtype of the influenza virus and do not provide protection against viruses from different subtypes. In contrast, mucosal IgA is the immunoglobulin that is primarily involved in the crossprotection of the mucosal surface against variant virus infection.25 One of the future studies on our polymer will be designed from the standpoint of cross-protection. Many attempts have been made to develop promising adjuvants/antigen carriers with potent immunostimulatory activities. However, most of them have not been used in the clinical setting because of unacceptable toxicities. At present,

inoculations for mucosal immunity. On the basis of the literature,3 a vaccine dose of 0.1 μg/mouse was first used. When mice were nasally inoculated with the vaccine alone, an elevation of vaccine-specific IgG levels in sera was observed in half the number of mice (data not shown). Because the IgG production was barely enhanced when the vaccine dose was halved, the vaccine dose was reset to 0.05 μg/mouse. As shown in Figure 6, D-octaargine-linked PNVA-co-AA clearly induced both mucosal and systemic immune responses against influenza virus HA vaccines, as did CTB. No statistical significance was observed for the levels of both serum IgG and intranasal secreted IgA at a polymer dose of 100 μg/mouse.



DISCUSSION The development of technologies that enable bioactive molecules with low membrane permeability to effectively penetrate biological membranes is one of the greatest challenges in the pharmaceutical field.18,19,23 A study on oligoarginine-linked polymers was initiated with the aim of developing these kinds of DDS technologies. The in vivo performance of oligoarginine-linked polymers was first examined, and we found that D-octaarginine-linked PNVA-coAA significantly enhanced the nasal absorption of insulin in mice.18 The in vitro performance of the polymer on cell membranes was subsequently evaluated thoroughly to validate the appropriateness of the proposed mechanism.19 The polymer dramatically enhanced the membrane permeation of poorly membrane-permeable molecules in a charge-independent manner, suggesting that intermolecular binding between the molecules and the peptidyl branches in the polymer backbone was not a prerequisite for penetration enhancement. None of the individual components (e.g., D-octaarginine) had any influence on membrane permeation, as seen in the case of in vivo studies. Polymer-induced cellular uptake of the molecules was constantly high, irrespective of the incubation time. Confocal laser scanning microphotographs of cells incubated with D-octaarginine-linked PNVA-co-AA bearing rhodamine red demonstrated that the cell outline was stained with red fluorescence. These experiments strongly indicated that Doctaarginine-linked PNVA-co-AA remained on the cell membrane and continuously enhanced the membrane permeation of the molecules that were present in the periphery of the peptidyl branches. The polymer-induced cellular uptake was significantly suppressed by 5-(N-ethyl-N-isopropyl)amiloride, which is an inhibitor of macropinocytosis; this suggested that the molecules were taken up into cells mainly via macropinocytosis through the biorecognition of the peptidyl branches. The present study demonstrated that D-octaarginine-linked PNVA-co-AA is a potential candidate for antigen carriers that induce mucosal and systemic immunity when nasally coadministered with antigens. OVA was used as a model antigen, and based on the literature,16 two inoculations were performed. Doses of OVA, D-octaarginine-linked PNVA-co-AA, and CTB were all set to 10 μg/mouse. This first trial indicated that the polymer enhanced the production of OVA-specific IgG in sera and that the inoculation schedule was insufficient for stimulating the production of OVA-specific secreted IgA in nasal wash fluids (Figures 2 and 3). Data also indicated that a part of mice inoculated with OVA alone were immunized due to the high dose of OVA. As shown in Figure 4, an increase in the number of inoculations and a reduction of the OVA dose resulted in a clear observation of the polymer-enhanced 2939

dx.doi.org/10.1021/mp300329r | Mol. Pharmaceutics 2012, 9, 2933−2941

Molecular Pharmaceutics



only aluminum-based mineral salts (alum) are used in humans. Alum is relatively safe, but its induction of antigen-specific immunity is still insufficient.17,26 The cytotoxicity of Doctaarginine-linked PNVA-co-AA was previously evaluated by monitoring lactose dehydrogenase (LDH), which a typical marker in cytotoxicity, released in the nasal cavity of mice after incubation with the polymer.18 The LDH activity in the nasal cavity after 60 min of incubation with the polymer at a dose of 10 μg/mouse was 102 ± 21.6 unit/L (mean ± SD, n = 4, the polymer was dissolved in saline solution of pH 3.0). This value was constant when the polymer solution was replaced with saline solution (92.0 ± 35.6 unit/L). Because the maximal dose of D-octaarginine-linked PNVA-co-AA in this study was 10 times that in the previous study, an additional experiment was performed under the same conditions. No significant elevation of the LDH activity was observed (112 ± 31.4 unit/L, n = 4, PBS was used as a solvent) even when the polymer dose jumped to 100 μg/mouse. Cationic polymers generally show dose-dependent toxicity. Because of the strong hydrophilicity of PNVA, it seems that water layers are formed on the surface of PNVA-co-AA used as a platform of D-octaarginine.23,27 The safety of the polymer should be thoroughly evaluated prior to its practical use; however, it appears that the water layers reduced cationic moieties-induced toxicity. We expect that the use of D-octaarginine-linked PNVA-co-AA will become one of the strategies for mucosal vaccine delivery.

REFERENCES

(1) Yuki, Y.; Kiyono, H. New generation of mucosal adjuvants for the induction of protective immunity. Rev. Med. Virol. 2003, 13, 293−310. (2) Okamoto, S.; Matsuura, M.; Akagi, T.; Akashi, M.; Tanimoto, T.; Ishikawa, T.; Takahashi, M.; Yamanishi, K.; Mori, Y. Poly(γ-glutamic acid) nano-particles combined with mucosal influenza virus hemagglutinin vaccine protects against influenza virus infection in mice. Vaccine 2009, 27, 5896−5905. (3) Mizuno, D.; Kurihara, M. I.; Ichinomiya, T.; Kubo, I.; Kido, H. Modified pulmonary surfactant is a potent adjuvant that stimulates the mucosal IgA production in response to the influenza virus antigen. J. Immunol. 2006, 176, 1122−1130. (4) Greenbaum, E.; Furst, A.; Kiderman, A.; Stewart, B.; Levy, R.; Schlesinger, M.; Morag, A.; Zakay-Rones, Z. Serum and mucosal immunologic responses in children following the administration of a new inactivated intranasal anti-influenza vaccine. J. Med. Virol. 2001, 65, 178−184. (5) Tamura, S. I.; Asanuma, H.; Ito, Y.; Hirabayashi, Y.; Suzuki, Y.; Nagamine, T.; Aizawa, C.; Kurata, T.; Oya, A. Superior crossprotective effect of nasal vaccination to subcutaneous inoculation with influenza hemagglutinin vaccine. Eur. J. Immunol. 1992, 22, 477−481. (6) Takada, A.; Matsushita, S.; Ninomiya, A.; Kawaoka, Y.; Kida, H. Intranasal immunization with formalin-inactivated virus vaccine induces a broad spectrum of heterosubtype immunity against influenza A virus infection in mice. Vaccine 2003, 21, 3212−3218. (7) Liew, F. Y.; Russell, S. M.; Appleyard, G.; Brand, C. M.; Beale, J. Cross-protection in mice infected with influenza A virus by the respiratory route is correlated with local IgA antibody rather than serum antibody or cytotoxic T cell reactivity. Eur. J. Immunol. 1984, 14, 350−356. (8) Glück, R.; Mischler, R.; Durrer, P.; Fürer, E.; Lang, A. B.; Herzog, C.; Cryz, S. J., Jr. Safety and immunogenicity of intranasally administered inactivated trivalent virosome-formulated influenza vaccine containing Escherichia coli heat-labile toxin as a mucosal adjuvant. J. Infect. Dis. 2000, 181, 1129−1132. (9) Tamura, S.; Yamanaka, A.; Shimohara, M.; Tomita, T.; Komase, K.; Tsuda, Y.; Suzuki, Y.; Nagamine, T.; Kawahara, K.; Danbara, H. Synergistic action of cholera toxin B subunit (and Escherichia coli heatlabile toxin B subunit) and a trace amount of cholera whole toxin as an adjuvant for nasal influenza vaccine. Vaccine 1994, 12, 419−426. (10) Hashigucci, K.; Ogawa, H.; Ishidate, T.; Yamashita, R.; Kamiya, H.; Watanabe, K.; Hattori, N.; Sato, T.; Suzuki, Y.; Nagamine, T.; Aizawa, C.; Tamura, S.; Kurata, T.; Oya, A. Antibody responses in volunteers induced by nasal influenza vaccine combined with Escherichia coli heat-labile enterotoxin B subunit containing a trace amount of the holotoxin. Vaccine 1996, 14, 113−119. (11) Mutsch, M.; Zhou, W.; Rhodes, P.; Bopp, M.; Chen, R. T.; Linder, T.; Spyr, C.; Steffen, R. Use of the inactivated internasal influenza vaccine and the risk of Bell’s palsy in Switzerland. N. Engl. J. Med. 2004, 350, 896−903. (12) Durrer, P.; Glück, U.; Spyr, C.; Lang, A. B.; Zurbriggen, R.; Herzog, C.; Glück, R. Mucosal antibody response induced with a nasal virosome-based influenza vaccine. Vaccine 2003, 21, 4328−4334. (13) de Bernardi di Valserra, M.; Zanasi, A.; Ragusa, S.; Glück, R.; Herzog, C. An open-label comparison of the immunogenicity and tolerability of intranasal and intramuscular formulations of virosomal influenza vaccine in healthy adults. Clin. Ther. 2002, 24, 100−111. (14) Ichinohe, T.; Watanabe, I.; Ito, S.; Fujii, H.; Moriyama, M.; Tamura, S.; Takahashi, H.; Sawa, H.; Chiba, J.; Kurata, T.; Sata, T.; Hasegawa, H. Synthetic double-stranded RNA poly(I:C) combined with mucosal vaccine protects against influenza virus infection. J. Virol. 2005, 79, 2910−2919. (15) Uto, T.; Akagi, T.; Hamasaki, T.; Akashi, M.; Baba, M. Modulation of innate and adaptive immunity by biodegradable nanoparticles. Immunol. Lett. 2009, 125, 46−52. (16) Uto, T.; Wang, X.; Akagi, T.; Zenkyu, R.; Akashi, M.; Baba, M. Improvement of adaptive immunity by antigen-carrying biodegradable nanoparticles. Biochem. Biophys. Res. Commun. 2009, 379, 600−604.



CONCLUSIONS The potential of D-octaarginine-linked PNVA-co-AA as a carrier for mucosal vaccine delivery was evaluated. Repeated nasal inoculation with PBS containing OVA and the polymer enhanced the production of OVA-specific IgG and secreted IgA in sera and nasal wash fluids, respectively. A similar immunization profile was also observed for influenza virus HA vaccines. The vaccine-specific humoral immunity was identical to that induced by CTB as a positive control, when the polymer dose was 10 times that of CTB. Data demonstrate that our polymer is a potential candidate for antigen carriers that induce mucosal and systemic immunity when nasally coadministered with antigens.



Article

AUTHOR INFORMATION

Corresponding Author

*Address: Faculty of Pharmaceutical Sciences, Setsunan University, 45-1, Nagaotoge-cho, Hirakata, Osaka 573-0101, Japan. Tel: +81-72-866-3124. Fax: +81-72-866-3126. E-mail: [email protected] (S.S.). Address: Graduate School of Medical and Dental Sciences, Kagoshima University, 8-35-1, Sakuragaoka, Kagoshima, Kagoshima 890-8544, Japan. Tel: +81-99-275-5930. Fax: +81-99-275-5932. E-mail: [email protected] (M.B.). Present Address ⊥

Graduate School of Engineering, Osaka University, 2-1 Yamadaoka, Suita, Osaka 565-0871, Japan.

Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS We thank Showa Denko Co. Ltd. for the gift of PNVA-co-AA. 2940

dx.doi.org/10.1021/mp300329r | Mol. Pharmaceutics 2012, 9, 2933−2941

Molecular Pharmaceutics

Article

(17) Uto, T.; Wang, X.; Sato, K.; Haraguchi, M.; Akagi, T.; Akashi, M.; Baba, M. Targeting of antigen to dendritic cells with poly(γglutamic acid) nanoparticles induces antigen-specific humoral and cellular immunity. J. Immunol. 2007, 178, 2979−2986. (18) Sakuma, S.; Suita, M.; Masaoka, Y.; Kataoka, M.; Nakajima, N.; Shinkai, N.; Yamauchi, H.; Hiwatari, K.; Tachikawa, H.; Kimura, R.; Yamashita, S. Oligoarginine-linked polymers as a new class of penetration enhancers. J. Controlled Release 2010, 148, 187−196. (19) Sakuma, S.; Suita, M.; Yamamoto, T.; Masaoka, Y.; Kataoka, M.; Yamashita, S.; Nakajima, N.; Shinkai, N.; Yamauchi, H.; Hiwatari, K.; Hashizume, A.; Tachikawa, H.; Kimura, R.; Ishimaru, Y.; Kasai, A.; Maeda, S. Performance of cell-penetrating peptide-linked polymers physically mixed with poorly membrane-permeable molecules on cell membranes. Eur. J. Pharm. Biopharm. 2012, 81, 64−73. (20) Futaki, S. Membrane-permeable arginine-rich peptides and the translocation mechanisms. Adv. Drug Delivery Rev. 2005, 57, 547−558. (21) Nakase, I.; Takeuchi, T.; Tanaka, G.; Futaki, S. Methodological and cellular aspects that govern the internalization mechanisms of arginine-rich cell-penetrating peptides. Adv. Drug Delivery Rev. 2008, 60, 598−607. (22) Jones, A. T. Macropinocytosis: Searching for an endocytic identity and a role in the uptake of cell penetrating peptides. J. Cell. Mol. Med. 2007, 11, 670−684. (23) Sakuma, S.; Hayashi, M.; Akashi, M. Design of nanoparticles composed of graft copolymers for oral peptide delivery. Adv. Drug Delivery Rev. 2001, 47, 21−37. (24) Nichol, K. L. The efficacy, effectiveness and cost-effectiveness of inactivated influenza virus vaccines. Vaccine 2003, 21, 1769−1775. (25) Belshe, R. B. Current status of live attenuated influenza virus vaccine in the US. Virus Res. 2004, 103, 177−185. (26) Petrovsky, N.; Aguilar, J. C. Vaccine adjuvants: Current state and future trends. Immunol. Cell Biol. 2004, 82, 488−496. (27) Sakuma, S.; Sudo, R.; Suzuki, N.; Kikuchi, H.; Akashi, M.; Ishida, Y.; Hayashi, M. Behavior of mucoadhesive nanoparticles having hydrophilic polymeric chains in the intestine. J. Controlled Release 2002, 81, 281−290.

2941

dx.doi.org/10.1021/mp300329r | Mol. Pharmaceutics 2012, 9, 2933−2941