- Email: [email protected]
A Novel Vaccine Delivery System Using Immunopotentiating Fusogenic Liposomes
Biochemical and Biophysical Research Communications 261, 824 – 828 (1999) Article ID bbrc.1999.1044, available online at http://www.idealibrary.com on
A Novel Vaccine Delivery System Using Immunopotentiating Fusogenic Liposomes Akira Hayashi,* ,1 Tsuyoshi Nakanishi,* ,1 Jun Kunisawa,* Masuo Kondoh,* Susumu Imazu,* Yasuo Tsutsumi,* Keiichi Tanaka,* Hiromi Fujiwara,† Toshiyuki Hamaoka,† and Tadanori Mayumi* ,2 *Graduate School of Pharmaceutical Sciences, Osaka University, 1-6, Yamadaoka, Suita, Osaka 565-0871, Japan; and †Biomedical Research Center, Osaka University Medical School, 2-2, Yamada-oka, Suita, Osaka 565-0871, Japan
Received April 20, 1999
We previously reported the preparation and characterization of fusogenic liposomes (FLs), which have two highly immunogenic glycoproteins of the Sendai virus on their surface. In this report, we investigated the capacity of FLs to enhance antigen-specific humoral immunity in mice. FLs function as a lymphocyte mitogen with high immunogenicity consistent with viral envelope proteins. Markedly increased levels of anti-ovalbumin (OVA) antibody were detected in serum from mice immunized with OVA encapsulated in FLs compared to sera from mice immunized with free OVA or OVA encapsulated in plain liposomes. An antiOVA antibody response was not observed in mice immunized with OVA simply mixed with empty FLs. These results indicate that FLs function as a novel immunoadjuvant in inducing antigen-specific antibody production. © 1999 Academic Press Key Words: fusogenic liposome; ovalbumin; nonreplicating vaccine; humoral immunity; Sendai virus.
Numerous nonreplicating vaccines composed of synthetic peptides, purified subunit antigens, or small inactivated viruses are currently undergoing evaluation (1–5). Although such vaccines offer the benefits of being antigenically defined and safe due to the absence of toxic contaminating substances, most of them are poorly immunogenic because of their innate molecular size and restricted epitope recognition (6). Currently available immunoadjuvants are effective against some antigens but they cause such adverse reactions as local 1
These authors contributed equally to this work. To whom correspondence should be addressed at Department of Biopharmaceutics, Graduate School of Pharmaceutical Sciences, Osaka University, 1-6, Yamadaoka Suita, Osaka 565-0871, Japan. Fax: 181-6-6879-8179. E-mail: [email protected]. Abbreviations used: FL, fusogenic liposome; OVA, ovalbumin; CTL, cytotoxic T lymphocyte; MHC, major histocompatibility complex; HAV, hepatitis A virus. 2
0006-291X/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.
granulomas, pain, fever, and possibly malignancies (7– 9). Therefore, a major goal in vaccine development is the design of new immunoadjuvants and delivery systems that induce potent immune responses without producing deleterious side effects. The Sendai virus, a member of the paramyxovirus family, produces two highly immunogenic envelope glycoproteins: fusion protein and hemagglutininneuraminidase protein. It has been demonstrated that inactivated Sendai virus retains the ability to fuse with host cells to elicit in vivo priming of cytotoxic T lymphocytes (CTLs) (10 –12). In addition, UV-inactivated Sendai virus and its envelope proteins act as lymphocyte mitogens, whereas infectious Sendai virus kills lymphocytes in vitro (13–15). These observations suggest that incorporating Sendai virus envelope glycoproteins into vaccine adjuvants and delivery systems may lead to an enhanced induction of immune responses. Previously, we reported that fusogenic liposomes (FLs), prepared by fusing simple liposomes with Sendai virus particles, can fuse with cell membranes and deliver their contents directly and efficiently into the cytoplasm in the same manner as the native virus (16 –19). In the present study, we assessed the potential of FLs to serve as a vaccine delivery system. MATERIALS AND METHODS Materials and mice. Egg phosphatidylcholine and L-adimyristoyl phosphatidic acid were purchased from Nippon Oil & Fats Co. (Tokyo, Japan). Cholesterol, stearylamine, ovalbumin (OVA), concanavalin A, and rabbit anti-OVA serum were purchased from Sigma Chemical Co. (St. Louis, MO, USA). Lipopolysaccharide was purchased from DIFCO Lab. Inc. (Detroit, MI, USA). Fluorescein isothiocyanate-conjugated anti-CD3 was purchased from PharMingen (San Diego, CA, USA). Phycoerythrin-conjugated antimouse IgG(H 1 L) was purchased from Kirkegaard & Perry Lab. Inc. (Gaithersburg, MO, USA). Horseradish peroxidase-conjugated antimouse IgG1 and horseradish peroxidase-conjugated anti-rabbit IgG were purchased from Zymed Laboratories Inc. (San Francisco, CA, USA). [ 3H]-thymidine was purchased from ICN Biomedicals Inc.
824
Vol. 261, No. 3, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
(Costa Mesa, CA, USA). Female C57BL/6 mice, 8 weeks old, were purchased from SLC Inc (Hamamatsu, Japan). Preparation of liposomes. Unilaminar liposomes were prepared as described by Oku et al. (20). Cholesterol, egg phosphatidylcholine, and L-a-dimyristoryl phosphatidic acid were mixed in a molar ratio of 5:4:1. The lipid mixture (15 mg) in chloroform was subsequently evaporated to obtain a thin lipid film. Liposome suspensions were prepared by dispersing the thin lipid film in 400 ml of buffered salt solution (10 mM Tris-HCl, 150 mM NaCl, pH 7.6) or antigen solution (20 mg/ml). The liposomes were frozen rapidly in liquid nitrogen and left to thaw at 37°C for 15 min. After three cycles of freezing and thawing, the liposomes were sized by extrusion through a 0.2-mm polycarbonate membrane and pelleted by ultracentrifugation. FLs were prepared as described elsewhere (17–19). Briefly, unilaminar liposomes were mixed with Sendai virus and incubated at 37°C for 2 h with shaking. FLs were separated from free liposomes and Sendai virus by sucrose step centrifugation (24,000 rpm, 2 h). The viral protein concentration of FLs was determined by a protein assay kit (Bio-Rad Laboratories, Hercules, CA, USA) using bovine serum albumin as a standard. The amount of OVA encapsulated within liposomes was determined by ELISA. SDS-PAGE and Western blotting. Gel electrophoresis was performed in 10% (w/v) polyacrylamide gels under reducing conditions. Western blotting analysis of OVA in FLs was performed using rabbit anti-OVA serum and horseradish peroxidase-conjugated anti-rabbit IgG. Determination of OVA-specific IgG1 antibody. For induction of systemic antibody responses, C57BL/6 mice were injected subcutaneously in the back with 50 mg of OVA in various formulations. Serum samples were collected from the tail vein at different time intervals after the primary immunization. Control serum was obtained from nonimmunized mice. Anti-OVA and anti-Sendai virus IgG1 serum levels were determined by ELISA. Cell culture. Primary murine spleen cells were cultured as follows: aseptically removed spleen cells were teased into single cell suspensions in RPMI 1640 medium supplemented with 10% heatinactivated fetal calf serum and 50 mM 2-mercaptoethanol. The cells were tested for viability and diluted to 1 3 10 6 viable cells/ml. The cell suspension was dispensed as 0.1 ml/well into microtiter plates and an equal volume of stimulator or medium was added to each well. Each culture was pulsed with 20 kBq of [ 3H]-thymidine for 8 h before harvesting. Flow cytometry analysis. For detection of B cells and T cells, spleen cells cultured with various stimulators for 72 h were stained with fluorescein isothiocyanate-conjugated anti-CD3 and phycoerythrinconjugated anti-mouse IgG for 30 min at 4°C. After washing, stained cells were analyzed with a flow cytometry system (Becton Dickinson Microbiology Systems, Sparks, MD, USA).
RESULTS To investigate whether FL-mediated adjuvancy or delivery of soluble protein antigen primes an efficient immune response, FLs encapsulating OVA (OVA-FLs) were prepared as a model of soluble protein antigen. Figure 1 shows the analysis by gel electrophoresis of OVA-FL, Sendai virus particle, and empty FL. The OVA-FL and empty FL samples are composed of fusion protein, hemagglutin-neuraminidase protein, matrix protein, nucleoprotein, and polymerase similar to those found in the Sendai virus particle sample. Figure 1 also shows that OVA-FL contains OVA while empty FL does not. Hemagglutinating/hemolytic titers of FLs
FIG. 1. Entrapment of ovalbumin into fusogenic liposomes. (A) SDS-PAGE of OVA-FL. Total protein (10 mg) from Sendai virus (lane 2), empty FLs (lane 3), free OVA (lane 4), or OVA-FLs (lane 5) was analyzed by SDS-PAGE followed by staining with Coomassie brilliant blue. Molecular weight markers are shown in lane 1. (B) Western blotting analysis of OVA-FLs. Empty FLs (lane 1), OVA-FLs (lane 2), or free OVA (lane 3) was analyzed by immunoblotting with anti-OVA serum. Abbreviations used in this figure: P, polymerase; HN, hemagglutin-neuraminidase protein; NP, nucleoprotein; F, fusion protein; M, matrix protein.
were similar to those of intact viral particles (data not shown). Previous studies have shown that UV-inactivated Sendai virus and viral envelope glycoproteins are mitogenic for spleen cells (13–15). To study the potential of FLs to act as an immunoadjuvant, the ability of empty FLs to stimulate mitogenesis was examined in mouse spleen cells. Table I presents data from three experiments which show empty FLs can stimulate mitogenesis in mouse splenocytes. The mitogenic response increased depending on the viral protein dose of empty FLs. The optimum time for mitogenesis was 72–96 h after the initiation of cultures. At all time points examined, the mitogenic response induced by empty FLs was less than those induced by lipopolysaccharide or concanavalin A, which is consistent with the findings in a UV-inactivated Sendai virus mitogen system (13). To determine which populations of splenocytes are targeted in FL-mediated mitogenesis, spleen cells stimulated with empty FLs were stained with lymphocyte subpopulation-specific antibodies labeled with fluorescein isothiocyanate or phycoerythrin and examined by flow cytometry analysis. The results show that the proportion of T cells decreases in cultures exposed to empty FLs compared to that of T cells in unstimulated cultures (Fig. 2). An increase in the percentage of B cells was also observed in cultures exposed to empty FLs compared to that of B cells in unstimulated cultures (Fig. 2). Similar results were obtained using OVA-FLs instead of empty FLs (data not shown). These results indicate that FLs directly prime B cells for mitogenesis and are consistent with a previous report (13) which demonstrated that hemagglutinneuraminidase and fusion glycoproteins of Sendai virus stimulate B cells but not T cells.
825
Vol. 261, No. 3, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS TABLE I
Stimulation of Mouse Spleen Cells by FL [ 3H]thymidine uptake (cpm) a Treatment None FL
LPS b Con A b a b
Protein conc. (mg/ml)
24 h
48 h
72 h
96 h
120 h
— 45 15 5 1.5 0.5 0.15 2.5 0.25
528 6 41 1488 6 5 1408 6 15 1331 6 111 1239 6 32 944 6 18 930 6 53 20668 6 922 10424 6 532
1396 6 41 4804 6 256 4366 6 62 3761 6 163 3138 6 14 2917 6 115 2446 6 44 115047 6 7532 171245 6 3282
1231 6 41 6834 6 131 4113 6 176 3896 6 190 3081 6 174 2415 6 235 1978 6 87 158381 6 3553 131541 6 433
1648 6 110 7025 6 721 5258 6 639 4140 6 566 3910 6 248 3316 6 417 2963 6 216 62944 6 230 16534 6 1272
422 6 30 4286 6 357 3401 6 414 3212 6 91 2142 6 324 1379 6 199 668 6 57 22504 6 760 9580 6 399
Results are means 6 S.D. for triplicate. Abbreviations used in this table; LPS, lipopolysaccharide; Con A, concanavalin A.
The capacity of FLs to induce B cell mitogenesis led us to examine whether the FL vector might elicit an antigen-specific antibody response in vivo. Mice (strain C57BL/6) were immunized with 50 mg of OVA in various formulations and serum samples were collected from the tail vein at various time points after the primary immunization. The serum levels of anti-OVA antibody were determined using ELISA. Virtually no primary IgG1 antibody was found in serum from mice immunized with free OVA or OVA encapsulated within plain liposomes. In contrast, immunization with OVA-FL resulted in strikingly high levels of antibody production, although the levels were appreciably lower than those observed for the alum adsorbed OVA (Fig. 3A). Immunization with empty FLs simply added to free OVA failed to induce OVA-specific IgG1 responses, whereas it produced Sendai virus-specific IgG1 responses at levels comparable to those induced using OVA-FLs (Fig. 3B). Similar results to those in Fig. 3
FIG. 2. Flow cytometry analysis of mouse spleen cells stimulated by FLs. Spleen cells from C57BL/6 mice were cultured without (A) or with (B) FLs at 15 mg protein/ml for 72 h. Cells were labeled with fluorescein isothiocyanate-conjugated anti-CD3 antibodies and phycoerythrin-conjugated anti-IgG(H 1 L) antibodies and were analyzed by flow cytometry. The mean percentage of cells is given for each subpopulation.
were observed using strains of mice other than C57BL/6. These results indicate that FLs function as an antigen carrier rather than as an immunomodulator. DISCUSSION Our results demonstrate that FLs function as a novel immunoadjuvant in inducing antigen-specific antibody production. However, when FLs were simply mixed with antigen, the adjuvancy of FLs was not observed despite their B cell-mitogenic capacity. It has been reported that UV-inactivated Sendai virus can induce detectable levels of interleukin-6 (IL-6), interferons (IFNs), and tumor necrosis factors a and b (TNF-a and TNF-b) in human peripheral blood mononuclear cells and lymphoblastoid cells (14, 15). However, we did not observe the production of cytokines (IL-2, IL-5, IL-6, IFN-g, and TNF-a) during stimulation of mouse splenocytes with FLs (data not shown). Together, these observations indicate that the adjuvancy of FLs is ascribed not simply to an immunopotentiating ability but rather to the capacity to deliver an encapsulated antigen to the immune system and stimulate antigen-specific immune responses in a highly effective manner. An approach similar to that used here was reported by Gluck et al. (4, 5). The authors prepared a virosomeformulated hepatitis A virus (HAV) vaccine by incorporating biologically active influenza virus glycoprotein into the liposome and by adsorbing inactivated HAV. Their results showed that healthy seronegative volunteers developed HAV-specific antibody responses after immunization with their virosome vaccine (4), thus demonstrating the immunoadjuvancy of influenza virosomes in the induction of humoral immunity. However, it is unknown whether these virosome vaccines can prime antigen-specific cell-mediated immunity, including cytotoxic T lymphocyte (CTL) responses.
826
Vol. 261, No. 3, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 3. Production of OVA-specific and Sendai virus-specific IgG1 following immunization with various ovalbumin formulations. The C57BL/6 mice were immunized with 50 mg of free OVA (■), OVA encapsulated in FLs (7,500 HAU) (E), OVA simply mixed with empty FLs (7,500 HAU) (F), OVA encapsulated in plain liposomes (h), or OVA in alum (Œ); serum samples were collected from the tail vain at various time points after immunization. Serum levels of anti-OVA IgG1 (A) and anti-Sendai virus IgG1 (B), diluted 1:500, were determined by ELISA. Results are means 6 SE for 5 or 6 mice per group.
Finally, liposome-associated vaccines have the capacity to stimulate CTL responses. CTL plays a critical role in the protection against viral infections (21, 22) and malignancy (23). The responses against antigens are generated when antigen short peptides processed intracellularly are presented to CD8 1 CTL precursors in association with the class I major histocompatibility complex (MHC) (24, 25). Key to this process is the intracytosolic presence of antigens, as has been observed for endogenously synthesized viral proteins. Exogenous proteins and peptides are internalized by phagocytosis and are processed within the vacuolar compartment for presentation by class II MHC molecules (26, 27). They fail to enter the cytosolic compartment and therefore are not efficiently presented via the class I MHC pathway (26). We have recently observed that FLs can deliver an encapsulated soluble protein directly into the cytosol of cultured cells and introduce it into the class I MHC antigen-presentation pathway (T. Nakanishi et al., manuscript in preparation). Moreover, a single immunization with OVA encapsulated in FLs but not in plain liposomes results in the potent priming of OVA-specific CTLs (T. Nakanishi et al., manuscript in preparation). Our combined observations indicate that FLs prime both humoral and cellmediated immune responses. This suggests that FLs deliver an encapsulated antigen efficiently into not only class II but also class I MHC antigen-presentation pathways. In this study, we showed that FLs, a nonreplicating vector, are capable of stimulating the production of antibody specific to an encapsulated antigen. A vaccine delivery system using FLs could prove useful for the
development of nonreplicating vaccines against a variety of antigens. ACKNOWLEDGMENTS This study was supported in part by Grants-in Aid for Scientific Research from the Ministry of Education, Science, Sports, and Culture of Japan, by Health Sciences Research Grants for Special Research, Research on Human Genome and Gene Therapy from Ministry of Health and Welfare, and by the Uehara Memorial Foundation.
REFERENCES 1. Flehmig, B., Heinricy, U., and Pfisterer, M. (1989) Lancet 1, 1039 –1041. 2. Miller, L. H., Howard, R. J., Carter, R., Good, M. F., Nussenzweig, V., and Nussenzweig, R. S. (1986) Science 234, 1349 –1356. 3. Esparza, J., Osmanov, S., and Heyward, W. L. (1995) Drugs 50, 792– 804. 4. Gluck, R., Mischler, R., Branstchen, S., Just, M., Althaus, B., and Cryz, S. J., Jr. (1992) J. Clin. Invest. 90, 2491–2495. 5. Mengiardi, B., Berger, R., Just, M., and Gluck, R. (1995) Vaccine 13, 1306 –1315. 6. Good, M. F., Kumar, S., and Miller, L. H. (1988) Immunol. Today 9, 351–355. 7. Freund, J. (1951) Am. J. Clin. Pathol. 21, 645– 656. 8. Beebe, G. W., Simon, A. H., and Vivona, S. (1972) Am. J. Epidemiol. 95, 337–346. 9. Gupta, R. K., and Siber, G. R. (1995) Vaccine 13, 1263–1276. 10. Palmer, J. C., Lewandowski, L. J., and Waters, D. (1977) Nature 269, 595–597. 11. McGee, M., Hale, A. H., and Panetti, M. (1980) Eur. J. Immunol. 10, 923–928. 12. Koszinowski, U. H., and Simon, M. M. (1979) Eur. J. Immunol. 9, 715–722.
827
Vol. 261, No. 3, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
13. Kizaka, S., Goodman-Snitkoff, G., and McSharry, J. J. (1983) Infect. Immun. 40, 592– 600. 14. Fukuda, S., Ando, S., Sanou, O., Taniai, M., Fujii, M., Masaki, N., Nakamura, K., Ando, O., Torigoe, K., Sugimoto, T., and Kurimoto, M. (1988) Lymphokine Res. 7, 175–185. 15. Neustock, P., Brand, J. M., Kruse, A., and Kirchner, H. (1993) Immunobiology 188, 293–302. 16. Nakanishi, M., Mizuguchi, H., Ashihara, K., Senda, T., Akuta, T., Okabe, J., Nagoshi, E., Masago, A., Eguchi, A., Suzuki, Y., Inokuchi, H., Watabe, A., Ueda, S., Hayakawa, T., and Mayumi, T. (1998) J. Control. Release 54, 61– 68. 17. Mizuguchi, H., Nakanishi, M., Nakanishi, T., Nakagawa, T., Nakagawa, S., and Mayumi, T. (1996) Br. J. Cancer 73, 472– 476. 18. Mizuguchi, H., Nakagawa, T., Nakanishi, M., Imazu, S., Nakagawa, S., and Mayumi, T. (1996) Biochem. Biophys. Res. Commun. 218, 402– 407. 19. Mizuguchi, H., Nakagawa, T., Toyosawa, S., Nakanishi, M., Imazu, S., Nakanishi, T., Tsutsumi, Y., Nakagawa, S., Haya-
20. 21. 22. 23. 24. 25. 26. 27.
828
kawa, T., Ijuhin, N., and Mayumi, T. (1998) Cancer Res. 58, 5725–5730. Oku, N., Kendall, D. A., and MacDonald, R. C. (1982) Biochim. Biophys. Acta. 691, 332–340. Rouse, B. T., Norley, S., and Martin, S. (1988) Rev. Infect. Dis. 10, 16 –33. Lukacher, A. E., Braciale, V. L., and Braciale, T. J. (1984) J. Exp. Med. 160, 814 – 826. Cerundolo, V., Lahaye, T., Horvath, C., Zanovello, P., Collavo, D., and Engers, H. D. (1987) Eur. J. Immunol. 17, 173–178. Townsend, A. R., Rothbard, J., Gotch, F. M., Bahadur, G., Wraith, D., and McMichael, A. J. (1986) Cell 44, 959 –968. Van-Bleek, G. M., and Nathenson, S. G. (1990) Nature 348, 213–216. Germain, R. N. (1986) Nature 322, 687– 689. Harding, C. V., and Geuze, H. J. (1993) J. Immunol. 151, 3988 – 3998.
A Novel Vaccine Delivery System Using Immunopotentiating Fusogenic Liposomes Akira Hayashi,* ,1 Tsuyoshi Nakanishi,* ,1 Jun Kunisawa,* Masuo Kondoh,* Susumu Imazu,* Yasuo Tsutsumi,* Keiichi Tanaka,* Hiromi Fujiwara,† Toshiyuki Hamaoka,† and Tadanori Mayumi* ,2 *Graduate School of Pharmaceutical Sciences, Osaka University, 1-6, Yamadaoka, Suita, Osaka 565-0871, Japan; and †Biomedical Research Center, Osaka University Medical School, 2-2, Yamada-oka, Suita, Osaka 565-0871, Japan
Received April 20, 1999
We previously reported the preparation and characterization of fusogenic liposomes (FLs), which have two highly immunogenic glycoproteins of the Sendai virus on their surface. In this report, we investigated the capacity of FLs to enhance antigen-specific humoral immunity in mice. FLs function as a lymphocyte mitogen with high immunogenicity consistent with viral envelope proteins. Markedly increased levels of anti-ovalbumin (OVA) antibody were detected in serum from mice immunized with OVA encapsulated in FLs compared to sera from mice immunized with free OVA or OVA encapsulated in plain liposomes. An antiOVA antibody response was not observed in mice immunized with OVA simply mixed with empty FLs. These results indicate that FLs function as a novel immunoadjuvant in inducing antigen-specific antibody production. © 1999 Academic Press Key Words: fusogenic liposome; ovalbumin; nonreplicating vaccine; humoral immunity; Sendai virus.
Numerous nonreplicating vaccines composed of synthetic peptides, purified subunit antigens, or small inactivated viruses are currently undergoing evaluation (1–5). Although such vaccines offer the benefits of being antigenically defined and safe due to the absence of toxic contaminating substances, most of them are poorly immunogenic because of their innate molecular size and restricted epitope recognition (6). Currently available immunoadjuvants are effective against some antigens but they cause such adverse reactions as local 1
These authors contributed equally to this work. To whom correspondence should be addressed at Department of Biopharmaceutics, Graduate School of Pharmaceutical Sciences, Osaka University, 1-6, Yamadaoka Suita, Osaka 565-0871, Japan. Fax: 181-6-6879-8179. E-mail: [email protected]. Abbreviations used: FL, fusogenic liposome; OVA, ovalbumin; CTL, cytotoxic T lymphocyte; MHC, major histocompatibility complex; HAV, hepatitis A virus. 2
0006-291X/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.
granulomas, pain, fever, and possibly malignancies (7– 9). Therefore, a major goal in vaccine development is the design of new immunoadjuvants and delivery systems that induce potent immune responses without producing deleterious side effects. The Sendai virus, a member of the paramyxovirus family, produces two highly immunogenic envelope glycoproteins: fusion protein and hemagglutininneuraminidase protein. It has been demonstrated that inactivated Sendai virus retains the ability to fuse with host cells to elicit in vivo priming of cytotoxic T lymphocytes (CTLs) (10 –12). In addition, UV-inactivated Sendai virus and its envelope proteins act as lymphocyte mitogens, whereas infectious Sendai virus kills lymphocytes in vitro (13–15). These observations suggest that incorporating Sendai virus envelope glycoproteins into vaccine adjuvants and delivery systems may lead to an enhanced induction of immune responses. Previously, we reported that fusogenic liposomes (FLs), prepared by fusing simple liposomes with Sendai virus particles, can fuse with cell membranes and deliver their contents directly and efficiently into the cytoplasm in the same manner as the native virus (16 –19). In the present study, we assessed the potential of FLs to serve as a vaccine delivery system. MATERIALS AND METHODS Materials and mice. Egg phosphatidylcholine and L-adimyristoyl phosphatidic acid were purchased from Nippon Oil & Fats Co. (Tokyo, Japan). Cholesterol, stearylamine, ovalbumin (OVA), concanavalin A, and rabbit anti-OVA serum were purchased from Sigma Chemical Co. (St. Louis, MO, USA). Lipopolysaccharide was purchased from DIFCO Lab. Inc. (Detroit, MI, USA). Fluorescein isothiocyanate-conjugated anti-CD3 was purchased from PharMingen (San Diego, CA, USA). Phycoerythrin-conjugated antimouse IgG(H 1 L) was purchased from Kirkegaard & Perry Lab. Inc. (Gaithersburg, MO, USA). Horseradish peroxidase-conjugated antimouse IgG1 and horseradish peroxidase-conjugated anti-rabbit IgG were purchased from Zymed Laboratories Inc. (San Francisco, CA, USA). [ 3H]-thymidine was purchased from ICN Biomedicals Inc.
824
Vol. 261, No. 3, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
(Costa Mesa, CA, USA). Female C57BL/6 mice, 8 weeks old, were purchased from SLC Inc (Hamamatsu, Japan). Preparation of liposomes. Unilaminar liposomes were prepared as described by Oku et al. (20). Cholesterol, egg phosphatidylcholine, and L-a-dimyristoryl phosphatidic acid were mixed in a molar ratio of 5:4:1. The lipid mixture (15 mg) in chloroform was subsequently evaporated to obtain a thin lipid film. Liposome suspensions were prepared by dispersing the thin lipid film in 400 ml of buffered salt solution (10 mM Tris-HCl, 150 mM NaCl, pH 7.6) or antigen solution (20 mg/ml). The liposomes were frozen rapidly in liquid nitrogen and left to thaw at 37°C for 15 min. After three cycles of freezing and thawing, the liposomes were sized by extrusion through a 0.2-mm polycarbonate membrane and pelleted by ultracentrifugation. FLs were prepared as described elsewhere (17–19). Briefly, unilaminar liposomes were mixed with Sendai virus and incubated at 37°C for 2 h with shaking. FLs were separated from free liposomes and Sendai virus by sucrose step centrifugation (24,000 rpm, 2 h). The viral protein concentration of FLs was determined by a protein assay kit (Bio-Rad Laboratories, Hercules, CA, USA) using bovine serum albumin as a standard. The amount of OVA encapsulated within liposomes was determined by ELISA. SDS-PAGE and Western blotting. Gel electrophoresis was performed in 10% (w/v) polyacrylamide gels under reducing conditions. Western blotting analysis of OVA in FLs was performed using rabbit anti-OVA serum and horseradish peroxidase-conjugated anti-rabbit IgG. Determination of OVA-specific IgG1 antibody. For induction of systemic antibody responses, C57BL/6 mice were injected subcutaneously in the back with 50 mg of OVA in various formulations. Serum samples were collected from the tail vein at different time intervals after the primary immunization. Control serum was obtained from nonimmunized mice. Anti-OVA and anti-Sendai virus IgG1 serum levels were determined by ELISA. Cell culture. Primary murine spleen cells were cultured as follows: aseptically removed spleen cells were teased into single cell suspensions in RPMI 1640 medium supplemented with 10% heatinactivated fetal calf serum and 50 mM 2-mercaptoethanol. The cells were tested for viability and diluted to 1 3 10 6 viable cells/ml. The cell suspension was dispensed as 0.1 ml/well into microtiter plates and an equal volume of stimulator or medium was added to each well. Each culture was pulsed with 20 kBq of [ 3H]-thymidine for 8 h before harvesting. Flow cytometry analysis. For detection of B cells and T cells, spleen cells cultured with various stimulators for 72 h were stained with fluorescein isothiocyanate-conjugated anti-CD3 and phycoerythrinconjugated anti-mouse IgG for 30 min at 4°C. After washing, stained cells were analyzed with a flow cytometry system (Becton Dickinson Microbiology Systems, Sparks, MD, USA).
RESULTS To investigate whether FL-mediated adjuvancy or delivery of soluble protein antigen primes an efficient immune response, FLs encapsulating OVA (OVA-FLs) were prepared as a model of soluble protein antigen. Figure 1 shows the analysis by gel electrophoresis of OVA-FL, Sendai virus particle, and empty FL. The OVA-FL and empty FL samples are composed of fusion protein, hemagglutin-neuraminidase protein, matrix protein, nucleoprotein, and polymerase similar to those found in the Sendai virus particle sample. Figure 1 also shows that OVA-FL contains OVA while empty FL does not. Hemagglutinating/hemolytic titers of FLs
FIG. 1. Entrapment of ovalbumin into fusogenic liposomes. (A) SDS-PAGE of OVA-FL. Total protein (10 mg) from Sendai virus (lane 2), empty FLs (lane 3), free OVA (lane 4), or OVA-FLs (lane 5) was analyzed by SDS-PAGE followed by staining with Coomassie brilliant blue. Molecular weight markers are shown in lane 1. (B) Western blotting analysis of OVA-FLs. Empty FLs (lane 1), OVA-FLs (lane 2), or free OVA (lane 3) was analyzed by immunoblotting with anti-OVA serum. Abbreviations used in this figure: P, polymerase; HN, hemagglutin-neuraminidase protein; NP, nucleoprotein; F, fusion protein; M, matrix protein.
were similar to those of intact viral particles (data not shown). Previous studies have shown that UV-inactivated Sendai virus and viral envelope glycoproteins are mitogenic for spleen cells (13–15). To study the potential of FLs to act as an immunoadjuvant, the ability of empty FLs to stimulate mitogenesis was examined in mouse spleen cells. Table I presents data from three experiments which show empty FLs can stimulate mitogenesis in mouse splenocytes. The mitogenic response increased depending on the viral protein dose of empty FLs. The optimum time for mitogenesis was 72–96 h after the initiation of cultures. At all time points examined, the mitogenic response induced by empty FLs was less than those induced by lipopolysaccharide or concanavalin A, which is consistent with the findings in a UV-inactivated Sendai virus mitogen system (13). To determine which populations of splenocytes are targeted in FL-mediated mitogenesis, spleen cells stimulated with empty FLs were stained with lymphocyte subpopulation-specific antibodies labeled with fluorescein isothiocyanate or phycoerythrin and examined by flow cytometry analysis. The results show that the proportion of T cells decreases in cultures exposed to empty FLs compared to that of T cells in unstimulated cultures (Fig. 2). An increase in the percentage of B cells was also observed in cultures exposed to empty FLs compared to that of B cells in unstimulated cultures (Fig. 2). Similar results were obtained using OVA-FLs instead of empty FLs (data not shown). These results indicate that FLs directly prime B cells for mitogenesis and are consistent with a previous report (13) which demonstrated that hemagglutinneuraminidase and fusion glycoproteins of Sendai virus stimulate B cells but not T cells.
825
Vol. 261, No. 3, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS TABLE I
Stimulation of Mouse Spleen Cells by FL [ 3H]thymidine uptake (cpm) a Treatment None FL
LPS b Con A b a b
Protein conc. (mg/ml)
24 h
48 h
72 h
96 h
120 h
— 45 15 5 1.5 0.5 0.15 2.5 0.25
528 6 41 1488 6 5 1408 6 15 1331 6 111 1239 6 32 944 6 18 930 6 53 20668 6 922 10424 6 532
1396 6 41 4804 6 256 4366 6 62 3761 6 163 3138 6 14 2917 6 115 2446 6 44 115047 6 7532 171245 6 3282
1231 6 41 6834 6 131 4113 6 176 3896 6 190 3081 6 174 2415 6 235 1978 6 87 158381 6 3553 131541 6 433
1648 6 110 7025 6 721 5258 6 639 4140 6 566 3910 6 248 3316 6 417 2963 6 216 62944 6 230 16534 6 1272
422 6 30 4286 6 357 3401 6 414 3212 6 91 2142 6 324 1379 6 199 668 6 57 22504 6 760 9580 6 399
Results are means 6 S.D. for triplicate. Abbreviations used in this table; LPS, lipopolysaccharide; Con A, concanavalin A.
The capacity of FLs to induce B cell mitogenesis led us to examine whether the FL vector might elicit an antigen-specific antibody response in vivo. Mice (strain C57BL/6) were immunized with 50 mg of OVA in various formulations and serum samples were collected from the tail vein at various time points after the primary immunization. The serum levels of anti-OVA antibody were determined using ELISA. Virtually no primary IgG1 antibody was found in serum from mice immunized with free OVA or OVA encapsulated within plain liposomes. In contrast, immunization with OVA-FL resulted in strikingly high levels of antibody production, although the levels were appreciably lower than those observed for the alum adsorbed OVA (Fig. 3A). Immunization with empty FLs simply added to free OVA failed to induce OVA-specific IgG1 responses, whereas it produced Sendai virus-specific IgG1 responses at levels comparable to those induced using OVA-FLs (Fig. 3B). Similar results to those in Fig. 3
FIG. 2. Flow cytometry analysis of mouse spleen cells stimulated by FLs. Spleen cells from C57BL/6 mice were cultured without (A) or with (B) FLs at 15 mg protein/ml for 72 h. Cells were labeled with fluorescein isothiocyanate-conjugated anti-CD3 antibodies and phycoerythrin-conjugated anti-IgG(H 1 L) antibodies and were analyzed by flow cytometry. The mean percentage of cells is given for each subpopulation.
were observed using strains of mice other than C57BL/6. These results indicate that FLs function as an antigen carrier rather than as an immunomodulator. DISCUSSION Our results demonstrate that FLs function as a novel immunoadjuvant in inducing antigen-specific antibody production. However, when FLs were simply mixed with antigen, the adjuvancy of FLs was not observed despite their B cell-mitogenic capacity. It has been reported that UV-inactivated Sendai virus can induce detectable levels of interleukin-6 (IL-6), interferons (IFNs), and tumor necrosis factors a and b (TNF-a and TNF-b) in human peripheral blood mononuclear cells and lymphoblastoid cells (14, 15). However, we did not observe the production of cytokines (IL-2, IL-5, IL-6, IFN-g, and TNF-a) during stimulation of mouse splenocytes with FLs (data not shown). Together, these observations indicate that the adjuvancy of FLs is ascribed not simply to an immunopotentiating ability but rather to the capacity to deliver an encapsulated antigen to the immune system and stimulate antigen-specific immune responses in a highly effective manner. An approach similar to that used here was reported by Gluck et al. (4, 5). The authors prepared a virosomeformulated hepatitis A virus (HAV) vaccine by incorporating biologically active influenza virus glycoprotein into the liposome and by adsorbing inactivated HAV. Their results showed that healthy seronegative volunteers developed HAV-specific antibody responses after immunization with their virosome vaccine (4), thus demonstrating the immunoadjuvancy of influenza virosomes in the induction of humoral immunity. However, it is unknown whether these virosome vaccines can prime antigen-specific cell-mediated immunity, including cytotoxic T lymphocyte (CTL) responses.
826
Vol. 261, No. 3, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 3. Production of OVA-specific and Sendai virus-specific IgG1 following immunization with various ovalbumin formulations. The C57BL/6 mice were immunized with 50 mg of free OVA (■), OVA encapsulated in FLs (7,500 HAU) (E), OVA simply mixed with empty FLs (7,500 HAU) (F), OVA encapsulated in plain liposomes (h), or OVA in alum (Œ); serum samples were collected from the tail vain at various time points after immunization. Serum levels of anti-OVA IgG1 (A) and anti-Sendai virus IgG1 (B), diluted 1:500, were determined by ELISA. Results are means 6 SE for 5 or 6 mice per group.
Finally, liposome-associated vaccines have the capacity to stimulate CTL responses. CTL plays a critical role in the protection against viral infections (21, 22) and malignancy (23). The responses against antigens are generated when antigen short peptides processed intracellularly are presented to CD8 1 CTL precursors in association with the class I major histocompatibility complex (MHC) (24, 25). Key to this process is the intracytosolic presence of antigens, as has been observed for endogenously synthesized viral proteins. Exogenous proteins and peptides are internalized by phagocytosis and are processed within the vacuolar compartment for presentation by class II MHC molecules (26, 27). They fail to enter the cytosolic compartment and therefore are not efficiently presented via the class I MHC pathway (26). We have recently observed that FLs can deliver an encapsulated soluble protein directly into the cytosol of cultured cells and introduce it into the class I MHC antigen-presentation pathway (T. Nakanishi et al., manuscript in preparation). Moreover, a single immunization with OVA encapsulated in FLs but not in plain liposomes results in the potent priming of OVA-specific CTLs (T. Nakanishi et al., manuscript in preparation). Our combined observations indicate that FLs prime both humoral and cellmediated immune responses. This suggests that FLs deliver an encapsulated antigen efficiently into not only class II but also class I MHC antigen-presentation pathways. In this study, we showed that FLs, a nonreplicating vector, are capable of stimulating the production of antibody specific to an encapsulated antigen. A vaccine delivery system using FLs could prove useful for the
development of nonreplicating vaccines against a variety of antigens. ACKNOWLEDGMENTS This study was supported in part by Grants-in Aid for Scientific Research from the Ministry of Education, Science, Sports, and Culture of Japan, by Health Sciences Research Grants for Special Research, Research on Human Genome and Gene Therapy from Ministry of Health and Welfare, and by the Uehara Memorial Foundation.
REFERENCES 1. Flehmig, B., Heinricy, U., and Pfisterer, M. (1989) Lancet 1, 1039 –1041. 2. Miller, L. H., Howard, R. J., Carter, R., Good, M. F., Nussenzweig, V., and Nussenzweig, R. S. (1986) Science 234, 1349 –1356. 3. Esparza, J., Osmanov, S., and Heyward, W. L. (1995) Drugs 50, 792– 804. 4. Gluck, R., Mischler, R., Branstchen, S., Just, M., Althaus, B., and Cryz, S. J., Jr. (1992) J. Clin. Invest. 90, 2491–2495. 5. Mengiardi, B., Berger, R., Just, M., and Gluck, R. (1995) Vaccine 13, 1306 –1315. 6. Good, M. F., Kumar, S., and Miller, L. H. (1988) Immunol. Today 9, 351–355. 7. Freund, J. (1951) Am. J. Clin. Pathol. 21, 645– 656. 8. Beebe, G. W., Simon, A. H., and Vivona, S. (1972) Am. J. Epidemiol. 95, 337–346. 9. Gupta, R. K., and Siber, G. R. (1995) Vaccine 13, 1263–1276. 10. Palmer, J. C., Lewandowski, L. J., and Waters, D. (1977) Nature 269, 595–597. 11. McGee, M., Hale, A. H., and Panetti, M. (1980) Eur. J. Immunol. 10, 923–928. 12. Koszinowski, U. H., and Simon, M. M. (1979) Eur. J. Immunol. 9, 715–722.
827
Vol. 261, No. 3, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
13. Kizaka, S., Goodman-Snitkoff, G., and McSharry, J. J. (1983) Infect. Immun. 40, 592– 600. 14. Fukuda, S., Ando, S., Sanou, O., Taniai, M., Fujii, M., Masaki, N., Nakamura, K., Ando, O., Torigoe, K., Sugimoto, T., and Kurimoto, M. (1988) Lymphokine Res. 7, 175–185. 15. Neustock, P., Brand, J. M., Kruse, A., and Kirchner, H. (1993) Immunobiology 188, 293–302. 16. Nakanishi, M., Mizuguchi, H., Ashihara, K., Senda, T., Akuta, T., Okabe, J., Nagoshi, E., Masago, A., Eguchi, A., Suzuki, Y., Inokuchi, H., Watabe, A., Ueda, S., Hayakawa, T., and Mayumi, T. (1998) J. Control. Release 54, 61– 68. 17. Mizuguchi, H., Nakanishi, M., Nakanishi, T., Nakagawa, T., Nakagawa, S., and Mayumi, T. (1996) Br. J. Cancer 73, 472– 476. 18. Mizuguchi, H., Nakagawa, T., Nakanishi, M., Imazu, S., Nakagawa, S., and Mayumi, T. (1996) Biochem. Biophys. Res. Commun. 218, 402– 407. 19. Mizuguchi, H., Nakagawa, T., Toyosawa, S., Nakanishi, M., Imazu, S., Nakanishi, T., Tsutsumi, Y., Nakagawa, S., Haya-
20. 21. 22. 23. 24. 25. 26. 27.
828
kawa, T., Ijuhin, N., and Mayumi, T. (1998) Cancer Res. 58, 5725–5730. Oku, N., Kendall, D. A., and MacDonald, R. C. (1982) Biochim. Biophys. Acta. 691, 332–340. Rouse, B. T., Norley, S., and Martin, S. (1988) Rev. Infect. Dis. 10, 16 –33. Lukacher, A. E., Braciale, V. L., and Braciale, T. J. (1984) J. Exp. Med. 160, 814 – 826. Cerundolo, V., Lahaye, T., Horvath, C., Zanovello, P., Collavo, D., and Engers, H. D. (1987) Eur. J. Immunol. 17, 173–178. Townsend, A. R., Rothbard, J., Gotch, F. M., Bahadur, G., Wraith, D., and McMichael, A. J. (1986) Cell 44, 959 –968. Van-Bleek, G. M., and Nathenson, S. G. (1990) Nature 348, 213–216. Germain, R. N. (1986) Nature 322, 687– 689. Harding, C. V., and Geuze, H. J. (1993) J. Immunol. 151, 3988 – 3998.