Live Bivalent Vaccine for Parainfluenza and Influe
http://www.100md.com
病菌学杂志 2005年第11期
Division of Virology, Department of Microbiology and Immunology, Institute of Medical Science, University of Tokyo, Tokyo 108-8639, Japan
Core Research for Evolutional Science and Technology (CREST), Japan Science and Technology Agency, Saitama 332-0021, Japan
Department of Pathobiological Sciences, University of Wisconsin-Madison, Madison, Wisconsin 53706
Institute for Molecular Virology, University of Wisconsin-Madison, Madison, Wisconsin 53706
ABSTRACT
Influenza and human parainfluenza virus infections are of both medical and economical importance. Currently, inactivated vaccines provide suboptimal protection against influenza, and vaccines for human parainfluenza virus infection are not available, underscoring the need for new vaccines against these respiratory diseases. Furthermore, to reduce the burden of vaccination, the development of multivalent vaccines is highly desirable. Thus, to devise a single vaccine that would elicit immune responses against both influenza and parainfluenza viruses, we used reverse genetics to generate an influenza A virus that possesses the coding region for the hemagglutinin/neuraminidase ectodomain of parainfluenza virus instead of the influenza virus neuraminidase. The recombinant virus grew efficiently in eggs but was attenuated in mice. When intranasally immunized with the recombinant vaccine, all mice developed antibodies against both influenza and parainfluenza viruses and survived an otherwise lethal challenge with either of these viruses. This live bivalent vaccine has obvious advantages over combination vaccines, and its method of generation could, in principle, be applied in the development of a "cocktail" vaccine with efficacy against several different infectious diseases.
INTRODUCTION
Influenza and human parainfluenza virus infections rank high among human respiratory diseases worldwide. Influenza accounts for 3 to 5 million cases of severe illness each year, 250,000 to 500,000 of which are fatal, while an estimated 16,300 to 96,500 people are hospitalized because of human parainfluenza virus infections (see the World Health Organization website at http://www.who.int/). Considerable effort has been expended to develop useful vaccines against influenza and human parainfluenza viruses. Inactivated influenza virus vaccines are safe and effective in preventing influenza virus-related hospitalization and death, but the immune response triggered by these vaccines is suboptimal and short lived, requiring annual vaccination (12). By contrast, live attenuated influenza virus vaccines elicit robust humoral and cellular responses that are predicted to be long lived (6). For human parainfluenza viruses, neither inactivated nor live vaccines are available. Although no data are available from studies with humans, formalin-inactivated vaccines for parainfluenza virus in animals aggravate the symptoms of infection upon challenge with homologous virus (23), as previously shown in humans and animal models given inactivated vaccines for respiratory syncytial virus (RSV) (a close relative of parainfluenza virus) (5, 9). This enhancement of pathogenicity was not observed for a live attenuated candidate vaccine, which proved to be effective in animal studies (22). Thus, for the prevention of both influenza and human parainfluenza virus infections, live attenuated vaccines are likely to be more effective than inactivated vaccines.
Multivalent vaccines containing protective antigens against several infectious agents have been developed as a means to reduce the burden of repeated childhood vaccinations (1). With the exception of the MMR (measles, mumps, and rubella) preparation, they are based on inactivated vaccines because propagation of one vaccine virus strain might interfere with the growth of the other vaccine component(s) (24). A possible solution to this problem would be to produce a single vaccine virus possessing several antigens that elicit protection against two or more infectious agents.
Influenza A virus (family Orthomyxoviridae, characterized by a segmented, negative-strand RNA genome) is considered an attractive vaccine vector candidate. It elicits strong humoral and cellular responses (6), its genome is amenable to genetic modification (7, 21), and the availability of 15 hemagglutinin (HA) and 9 neuraminidase (NA) subtypes (HA and NA are the major viral antigens) would allow repeated immunization. However, early attempts to express foreign proteins from recombinant influenza viruses failed due to the genetic instability of the foreign genes (15). This limitation was recently overcome when we identified regions in influenza viral RNA (vRNA) segments that govern the efficient incorporation of vRNA into virions (8, 33). These incorporation signals, which reside in the coding regions, have been exploited for the stable expression of foreign proteins from influenza viruses (27, 33).
Thus, using our knowledge of influenza vRNA incorporation signals, we aimed to create a bivalent live vaccine that would provide protection against both influenza and parainfluenza virus infections. This objective was pursued by generating an influenza A virus in which the ectodomain of the NA was replaced by that of the HN, a parainfluenza virus protective antigen, and then evaluating the protective potential of the vaccine in a mouse model.
MATERIALS AND METHODS
Cells. 293T human embryonic kidney cells (a derivative of the 293 line into which the gene for simian virus 40 T antigen was inserted) and Madin-Darby canine kidney (MDCK) cells were maintained in Dulbecco's modified Eagle medium supplemented with 10% fetal calf serum and in minimum essential medium containing 5% newborn calf serum, respectively. All cells were maintained at 37°C in 5% CO2.
Viruses. Influenza A/WSN/33 (H1N1) virus (WSN) was generated by plasmid-based reverse genetics as previously described (21) and propagated in 10-day-old embryonated chicken eggs. A murine parainfluenza virus type 1, Sendai virus (SeV; strain Enders) was propagated in 10-day-old embryonated chicken eggs.
Plasmid construction. Plasmid pPolINA(61)NA/HNect (Fig. 1), which was used for the production of a negative-sense vRNA containing the coding sequence for an NA-HN chimeric protein between the 3' and 5' NA vRNA incorporation signals (8), was generated as follows. The region corresponding to the WSN NA cytoplasmic tail, transmembrane domain, and a portion of the stalk (i.e., amino acids 1 to 37 of WSN NA) was amplified by PCR. Likewise, we amplified the region corresponding to the SeV HN ectodomain (i.e., amino acids 61 to 576 of SeV HN), followed by two consecutive stop codons. The resultant PCR products were inserted between the 3' and 5' NA vRNA incorporation signals (i.e., the 3' noncoding region plus the first 183 nucleotides [nt] of the coding region with replacement of the start codon, ATG by GCG, and the 5' noncoding region plus the last 157 nt of the coding region). The recombinant influenza viral cDNA thus produced was inserted between RNA polymerase I promoter and terminator sequences.
Generation of influenza virus expressing SeV HN antigen. To generate influenza virus expressing SeV HN protein, designated FluH/SeVHN, we performed reverse genetics as previously described (21). Briefly, eight RNA polymerase I-driven plasmids for transcription of all eight influenza vRNA segments, including the NA/HNect segment, were transfected into 293T cells as previously described (21). We then cotransfected cells with five eukaryotic protein expression plasmids encoding NA, as well as the PA, PB1, PB2, and NP proteins, which are required for replication and transcription of vRNAs. Two days posttransfection, virus was collected from the supernatant of transfected cells, grown once in 10-day-old embryonated chicken eggs, and stored at –80°C until used in experiments.
Experimental infection. The pathogenicity of FluH/SeVHN was tested by intranasal infection of 4-week-old female BALB/c mice (The Jackson Laboratory, Bar Harbor, ME) that were anesthetized with isoflurane. Survival times and body weights were monitored daily for 14 days after infection. Three days postinfection, three mice per group were euthanized to determine virus titers in nasal turbinates and lungs. The dose lethal to 50% of mice (MLD50) infected with wild-type WSN or SeV was determined by intranasally inoculation of 8-week-old female BALB/c mice with 10-fold serial dilutions of virus under anesthetization; infected mice were observed daily for 14 days.
To evaluate the protective efficacy of FluH/SeVHN against challenge with lethal doses of wild-type viruses, we infected each mouse intranasally with FluH/SeVHN (105.3, 106.3, or 107.3 50% egg infectious doses [EID50]/mouse). As controls, we infected mice intranasally with formalin-inactivated virus (equivalent to 106.3 EID50/mouse) or phosphate-buffered saline (PBS). Twenty-seven days later, serum samples as well as trachea-lung and nasal washes were collected from a subset of mice and examined for virus-specific immunoglobulin A (IgA) or IgG antibodies by an enzyme-linked immunosorbent assay (ELISA) (13). Briefly, wells of Immulon 2HB (Thermo Labsystems, Franklin, MA) plates were coated with purified and detergent-treated WSN or SeV and blocked with PBS containing 10 mg/ml bovine serum albumin. After incubation of virus-coated wells with test samples, bound antibodies were detected with goat anti-mouse IgA or IgG conjugated to horseradish peroxidase (Kirkegaard & Perry Laboratories Inc., Gaithersburg, MD). Statistical analyses were performed with Student's t test. Twenty-eight days after infection, the remaining mice were challenged intranasally with 10 MLD50 of wild-type WSN or SeV and monitored daily for survival and body weight for 28 days. Virus titers were determined in organs from five mice per group at 3 days postchallenge.
RESULTS
Generation and characterization of an influenza A virus expressing SeV HN antigen. To generate an influenza virus-based live, bivalent vaccine, we chose the influenza A/WSN/33 (H1N1) virus and the Enders strain of SeV, a murine type 1 parainfluenza virus, since both cause lethal infection in mice, providing a useful animal model with which to evaluate vaccine efficacy. Moreover, we recently mapped the sequences that mediate efficient WSN NA vRNA virion incorporation to 183 nt, corresponding to the amino-terminal coding region of NA, and to 157 nt, corresponding to the carboxy-terminal coding region of this protein, in addition to the 3' and 5' noncoding regions (8). With this information, we were able to generate a recombinant WSN virus expressing the ectodomain of SeV HN (designated FluH/SeVHN), as described in Materials and Methods. To determine if FluH/SeVHN expressed both the SeV HN and WSN HA glycoproteins, we infected MDCK cells with this virus or wild-type WSN (as a control) and stained cells 48 h later with antibodies to SeV HN, WSN HA, or influenza virus NP. Cells infected with influenza virus expressed HA and NP proteins (Fig. 2) but not SeV HN, whereas cells infected with FluH/SeVHN expressed both influenza virus HA and NP proteins and the SeV HN protein.
To determine whether the NA-HN chimeric protein was expressed on the cell surface, we performed a flow cytometry analysis of HN antigen on 293T cells transfected with pPolINA(61)NA/HNect plasmid, encoding the chimeric HN protein and those expressing the three polymerase and NP proteins. HN antigens were detected on the surface of these cells but not on control cells, which did not receive the polymerase I plasmid. Next, to determine whether FluH/SeVHN virus particles possess the HN antigen, we carried out ELISA. Briefly, wells of Immulon 2HB plates were coated with purified and detergent-treated FluH/SeVHN, WSN, or SeV; SeV HN and influenza virus NP were detected by monoclonal antibodies. The wells coated with FluH/SeVHN or SeV reacted with antibodies against SeV HN protein, while those coated with WSN did not (data not shown). By contrast, the wells coated with FluH/SeVHN or WSN, but not SeV, reacted with an antibody against influenza virus NP. These results demonstrate that SeV HN antigen is present in FluH/SeVHN virions.
For vaccine production, the virus must grow to high titers. We therefore evaluated the titers of the recombinant virus grown in MDCK cells or embryonated chicken eggs. In MDCK cells, FluH/SeVHN replication was attenuated compared with that of wild-type influenza virus (1.0 x 106 PFU/ml versus 1.6 x 108 PFU/ml), resulting in pinpoint plaques only (Fig. 2). In eggs, however, FluH/SeVHN replicated to titers comparable to those of wild-type influenza virus (109.3 EID50/ml for the former versus 108.3 EID50/ml for the latter).
The stability of FluH/SeVHN was assessed by passaging the virus 10 times in embryonated chicken eggs. During 10 serial passages, FluH/SeVHN virus grew well, with titers ranging from 108.5 EID50/ml to 109.8 EID50/ml. Even after 10 serial passages in eggs, 88% of the plaques expressed SeV HN when tested in MDCK cells, and only one mutation (Asn to Lys at amino acid position 229) was found in the NA/HNect vRNA. Since not all virions of even wild-type influenza virus contain all eight RNA segments (17), the finding that not all of the FluH/SeVHN plaques expressed the NA/HNect was anticipated. These results demonstrate that influenza vRNA packaging signals can be exploited to stably express foreign genes.
Pathogenicity of FluH/SeVHN virus in mice. Attenuation in animals is a critical requirement for live vaccines. We therefore determined the pathogenicity of FluH/SeVHN virus in mice by intranasally infecting them with 105.3, 106.3, or 107.3 EID50/mouse of recombinant virus that had been passed once in eggs. Mice inoculated with 107.3 EID50 or less showed no overt symptoms of illness. Thus, the MLD50 of FluH/SeVHN virus was calculated to be >107.3 EID50/mouse, while that of wild-type WSN virus in mice of the same age was 103.0 EID50/mouse (16). Virus titers in the lungs of mice (three mice/group) inoculated with 105.3, 106.3, or 107.3 EID50 of FluH/SeVHN virus ranged from 104.3 to 104.6 EID50/g at 3 days postinfection (Table 1), while those in nasal turbinates ranged from 103.9 to 104.0 EID50/g, lower than titers measured after inoculation with 105.9 EID50 of wild-type influenza virus (107.4 EID50/g for lung and 105.3 EID50/g for nasal turbinates) (16). Considered together, the results indicate that FluH/SeVHN virus is highly attenuated in mice.
Antibody responses of mice immunized with FluH/SeVHN virus. To determine if infection of mice with FluH/SeVHN induces antibodies against both SeV and WSN, we intranasally inoculated the animals with 105.3, 106.3, or 107.3 EID50 of virus. Mice intranasally inoculated with PBS or a dose equivalent to 106.3 EID50 of formalin-inactivated vaccine virus served as controls. Twenty-seven days after inoculation, mice were euthanized, and their sera, as well as trachea-lung and nasal washes, were tested for IgA and IgG antibody levels against both viruses by ELISA (Fig. 3). Compared to mice infected with PBS or formalin-inactivated virus, those infected with 106.3 or 107.3 EID50 of FluH/SeVHN virus had elevated antibody levels to SeV in trachea-lung washes (both IgA and IgG) and in sera (IgG), but not in nasal washes. For mice inoculated with 105.3 EID50 of vaccine virus, antibody titers to SeV did not increase significantly. For the antibody response to WSN, both IgA and IgG antibody titers were significantly elevated (P < 0.01) in all samples derived from mice infected with FluH/SeVHN. Mice inoculated with PBS or formalin-inactivated vaccine virus did not show any significant IgA or IgG antibody response to either WSN virus or SeV. These results indicate that 106.3 EID50/mouse of FluH/SeVHN virus induces significant antibody responses against both WSN and SeV in mice.
Protective efficacy of FluH/SeVHN. To evaluate the protective efficacy of FluH/SeVHN, we immunized mice with live virus (105.3, 106.3, or 107.3 EID50), formalin-inactivated vaccine virus (equivalent to 106.3 EID50), or PBS and challenged them with 10 MLD50 of wild-type SeV (106.5 PFU/mouse) or WSN (105.5 PFU/mouse) at 28 days postimmunization. Mice immunized with 106.3 or 107.3 EID50 of live FluH/SeVHN virus were completely protected from challenge with a lethal dose of SeV, while immunization with 105.3 EID50 provided only partial protection (Table 2). By contrast, control animals immunized with formalin-inactivated virus or PBS died on days 6 to 10 postchallenge. For mice immunized with 107.3 EID50 of live vaccine virus, SeV titers in lungs and nasal turbinates were lower than in mice immunized with PBS on day 3 postchallenge (Student's t test; P < 0.05). However, virus titers in the organs of mice immunized with lower doses were not significantly lower than those in the control group, in spite of the protection observed in the immunized groups (Table 2). Conceivably, viral clearance at later time points postchallenge might have been more efficient in the immunized animals. All mice that survived SeV challenge lost body weight with a maximum loss of approximately 26% occurring by day 6 or 7 postchallenge (Fig. 4). After challenge with lethal doses of WSN, none of the mice immunized with live vaccine virus died or showed weight loss; by contrast, mice immunized with formalin-inactivated virus or PBS succumbed to infection (Table 3). WSN titers in lungs and nasal turbinates of mice immunized with vaccine virus were below the detection limits (<100 PFU/g) on day 3 postchallenge. Taken together, these data indicate that FluH/SeVHN provides efficient protection of mice against challenge with lethal doses of SeV or WSN.
DISCUSSION
Here, we describe the development of a live bivalent vaccine against two major respiratory infections: influenza virus and parainfluenza virus. Using reverse genetics, we produced a recombinant influenza virus, FluH/SeVHN, that expresses the ectodomain of SeV HN (an antigen that elicits protection against SeV infection) instead of the influenza virus NA, which is not as important as the HA for induction of a protective immune response (19). FluH/SeVHN protected mice against lethal challenge with influenza and murine parainfluenza viruses. It also replicated efficiently in eggs, stably expressed the HN antigen, was highly attenuated in mice, and induced protective antibodies against both parainfluenza and influenza viruses in these animals. Hence, it meets the basic requirements for a live attenuated vaccine.
The development of viral vector systems for vaccine delivery is being vigorously pursued. DNA viruses and positive-sense RNA viruses have attracted the most attention as possible vaccine vectors, although with the advent of reverse genetics, the potential of negative-sense RNA viruses can be explored as well. In contrast to DNA viruses, the genetic material of RNA viruses (with the exception of members of the Retroviridae family) cannot be integrated into the host cell genome, adding to their biological safety. Among the group of nonsegmented negative-sense RNA viruses, Newcastle disease virus (20), vesicular stomatitis virus (14, 26), rinderpest virus (32), and Sendai virus (18, 30) have been tested as vaccine vectors. The use of segmented negative-sense RNA viruses as vaccine vectors has been impeded by the difficulty of artificially generating viruses that contain multiple gene segments (eight for influenza A virus). Moreover, early attempts to express foreign genes from a ninth segment were not encouraging, due to genetic instability (15). These limitations have been overcome with the establishment of a plasmid-based reverse genetics system for influenza virus (21) and our recent identification of segment-specific virion incorporation signals in the influenza vRNA segments (8, 33). Here, we extended our earlier findings by generating an influenza A virus that contains a parainfluenza virus HN ectodomain and by demonstrating the protective efficacy of this new vaccine candidate in a mouse model.
The high mutation rate and the potential reassortment of influenza viruses have raised some concern over their use as live vaccines. The recently approved live attenuated influenza virus vaccine relies on a limited number of amino acid replacements to confer the requisite cold-adapted, temperature-sensitive, and attenuated phenotypes; hence, point mutations could result in the loss of such phenotypes. By contrast, the vaccine virus described in this report is attenuated through replacement of the NA ectodomain with that of SeV HN, so that point mutations are highly unlikely to result in a loss of attenuation.
Combined vaccination, in which recipients are immunized with vaccines for multiple pathogens at once, is highly desirable in the clinic. However, simultaneous immunization with two or more live vaccine viruses can result in immunization failure if one virus affects the replication and/or viral pathogenicity of the other viruses (24). With vector systems, immune responses against vector components can present major obstacles, prohibiting repeated immunization with the same vector backbone, as noted for adenoviral vectors (25). In our approach, the immunogenicity of the influenza virus vector is considered beneficial, since it would elicit immune responses against a societally and economically important pathogen. Indeed, we found that an influenza virus expressing the SeV HN ectodomain induced significant antibody levels against both influenza and parainfluenza viruses and fully protected immunized mice from challenge with lethal doses of the parental viruses. Swayne et al. (29), using a strategy similar to ours, generated a recombinant Newcastle disease virus that expressed the HA protein of an avian influenza virus; however, the vaccine failed to fully protect chickens from challenge with lethal doses of Newcastle disease or influenza virus. This discrepancy may reflect differences in the immunogenic potential of the two viruses, the doses used for protection and/or challenge, or the type of animal model used.
The known protective effect of the SeV HN ectodomain during viral infection led us to choose it as an SeV antigen. In addition, we anticipated that its sialidase activity would compensate for the lack of this enzyme in the neuraminidase-deficient influenza A virus vector. Efficient replication of WSN NA/HNect virus in eggs and poor replication in eggs of a recombinant virus expressing green fluorescent protein instead of NA (hence lacking sialidase activity; data not shown) substantiated this concept. The effect of balanced HA and NA activities on influenza virus replication depends on culture system (i.e., mammalian cell culture and embryonated chicken eggs) (3, 31). For example, WSN viruses with deletions in the stalk region of the NA protein replicated as well as wild-type WSN virus in cultured cells; however, some of these mutant viruses replicated poorly in embryonated chicken eggs (2). For FluH/SeVHN virus, we replaced the NA ectodomain with that of SeV HN, likely affecting the balance between the receptor-binding and -destroying activities of the viral glycoproteins and resulting in attenuated phenotypes in cell culture. By contrast, the imbalanced HA and NA activities were overcome in eggs, allowing efficient FluH/SeVHN replication.
The FluH/SeVHN virus we generated appears to have the essential properties of an ideal live vaccine. It is safe, genetically stable, and strongly immunogenic, and its growth profile in vitro is compatible with the requirements for efficient vaccine production. Hence, our results could open a new avenue for the production of multivalent vaccines. For example, the live influenza virus vaccine currently in use is a mixture of two type A viruses (subtypes H1N1 and H3N2) and a type B virus. Thus, by replacing the NA of these influenza viruses with three different genes encoding the HNs of human parainfluenza virus types 1 and 3 (the two most common human parainfluenza viruses) and the G or F of RSV, one could generate a protective cocktail vaccine against six types of respiratory tract infections. Human parainfluenza virus and RSV infection primarily occur in the first year of life, and influenza can also occur in patients under 6 months of age (10). Although live influenza virus vaccine is not currently recommended for children under 5 years of age, the safety and efficacy of live attenuated influenza virus vaccines in infants under 1 year of age have been shown (4, 11, 28). Moreover, the safety and efficacy of the current live attenuated influenza virus vaccine can be further improved by using reverse genetics. Thus, we believe that influenza virus-based vaccines for younger children will be used in the future. Ultimately, by exploiting the known packaging signals of influenza vRNAs, it may be possible to add genes encoding the protective antigens of any infectious agent.
ACKNOWLEDGMENTS
We thank Yutaka Fujii for scientific advice, Martha McGregor and Krisna Wells for excellent technical assistance, the animal care staff at the University of Wisconsin-Madison for the professional care of mice, and John Gilbert for editing the manuscript.
This study was supported by National Institute of Allergy and Infectious Diseases Public Health Service research grants; by CREST (Japan Science and Technology Agency); by grants-in-aid from the Ministry of Education, Culture, Sports, Science, and Technology of Japan; by the Ministry of Health, Labor, and Welfare of Japan; and by research fellowships of the Japan Society for the Promotion of Science for Young Scientists.
REFERENCES
Capiau, C., J. Poolman, B. Hoet, H. Bogaerts, and F. Andre. 2003. Development and clinical testing of multivalent vaccines based on a diphtheria-tetanus-acellular pertussis vaccine: difficulties encountered and lessons learned. Vaccine 21:2273-2287.
Castrucci, M. R., and Y. Kawaoka. 1993. Biologic importance of neuraminidase stalk length in influenza A virus. J. Virol. 67:759-764.
Clavijo, A., D. B. Tresnan, R. Jolie, and E. M. Zhou. 2002. Comparison of embryonated chicken eggs with MDCK cell culture for the isolation of swine influenza virus. Can J. Vet. Res. 66:117-121.
Clements, M. L., M. K. Makhene, R. A. Karron, B. R. Murphy, M. C. Steinhoff, and K. Subbarao. 1996. Effective immunization with live attenuated influenza A virus can be achieved in early infancy. Pediatric Care Center. J. Infect. Dis. 173:44-51.
Collins, P. L., and B. R. Murphy. 2002. Respiratory syncytial virus: reverse genetics and vaccine strategies. Virology 296:204-211.
Cox, R. J., K. A. Brokstad, and P. Ogra. 2004. Influenza virus: immunity and vaccination strategies. Comparison of the immune response to inactivated and live, attenuated influenza vaccines. Scand. J. Immunol. 59:1-15.
Fodor, E., L. Devenish, O. G. Engelhardt, P. Palese, G. G. Brownlee, and A. Garcia-Sastre. 1999. Rescue of influenza A virus from recombinant DNA. J. Virol. 73:9679-9682.
Fujii, Y., H. Goto, T. Watanabe, T. Yoshida, and Y. Kawaoka. 2003. Selective incorporation of influenza virus RNA segments into virions. Proc. Natl. Acad. Sci. USA 100:2002-2007.
Fulginiti, V. A., J. J. Eller, O. F. Sieber, J. W. Joyner, M. Minamitani, and G. Meiklejohn. 1969. Respiratory virus immunization. I. A field trial of two inactivated respiratory virus vaccines; an aqueous trivalent parainfluenza virus vaccine and an alum-precipitated respiratory syncytial virus vaccine. Am. J. Epidemiol. 89:435-448.
Glezen, W. P., L. H. Taber, A. L. Frank, W. C. Gruber, and P. A. Piedra. 1997. Influenza virus infections in infants. Pediatr. Infect. Dis. J. 16:1065-1068.
Gruber, W. C., K. Kirschner, S. Tollefson, J. Thompson, G. Reed, K. M. Edwards, and P. F. Wright. 1993. Comparison of monovalent and trivalent live attenuated influenza vaccines in young children. J. Infect. Dis. 168:53-60.
Harper, S. A., K. Fukuda, N. J. Cox, and C. B. Bridges. 2003. Using live, attenuated influenza vaccine for prevention and control of influenza: supplemental recommendations of the Advisory Committee on Immunization Practices (ACIP). Morb. Mortal. Wkly. Rep. Recomm. Rep. 52(RR-13):1-8.
Kida, H., L. E. Brown, and R. G. Webster. 1982. Biological activity of monoclonal antibodies to operationally defined antigenic regions on the hemagglutinin molecule of A/Seal/Massachusetts/1/80 (H7N7) influenza virus. Virology 122:38-47.
Kretzschmar, E., L. Buonocore, M. J. Schnell, and J. K. Rose. 1997. High-efficiency incorporation of functional influenza virus glycoproteins into recombinant vesicular stomatitis viruses. J. Virol. 71:5982-5989.
Luytjes, W., M. Krystal, M. Enami, J. D. Pavin, and P. Palese. 1989. Amplification, expression, and packaging of foreign gene by influenza virus. Cell 59:1107-1113.
Maeda, Y., H. Goto, T. Horimoto, A. Takada, and Y. Kawaoka. 2004. Biological significance of the U residue at the –3 position of the mRNA sequences of influenza A viral segments PB1 and NA. Virus Res. 100:153-157.
Martin, K., and A. Helenius. 1991. Nuclear transport of influenza virus ribonucleoproteins: the viral matrix protein (M1) promotes export and inhibits import. Cell 67:117-130.
Matano, T., M. Kobayashi, H. Igarashi, A. Takeda, H. Nakamura, M. Kano, C. Sugimoto, K. Mori, A. Iida, T. Hirata, M. Hasegawa, T. Yuasa, M. Miyazawa, Y. Takahashi, M. Yasunami, A. Kimura, D. H. O'Connor, D. I. Watkins, and Y. Nagai. 2004. Cytotoxic T lymphocyte-based control of simian immunodeficiency virus replication in a preclinical AIDS vaccine trial. J. Exp. Med. 199:1709-1718.
Mozdzanowska, K., K. Maiese, M. Furchner, and W. Gerhard. 1999. Treatment of influenza virus-infected SCID mice with nonneutralizing antibodies specific for the transmembrane proteins matrix 2 and neuraminidase reduces the pulmonary virus titer but fails to clear the infection. Virology 254:138-146.
Nakaya, T., J. Cros, M. S. Park, Y. Nakaya, H. Zheng, A. Sagrera, E. Villar, A. Garcia-Sastre, and P. Palese. 2001. Recombinant Newcastle disease virus as a vaccine vector. J. Virol. 75:11868-11873.
Neumann, G., T. Watanabe, H. Ito, S. Watanabe, H. Goto, P. Gao, M. Hughes, D. R. Perez, R. Donis, E. Hoffmann, G. Hobom, and Y. Kawaoka. 1999. Generation of influenza A viruses entirely from cloned cDNAs. Proc. Natl. Acad. Sci. USA 96:9345-9350.
Newman, J. T., J. M. Riggs, S. R. Surman, J. M. McAuliffe, T. A. Mulaikal, P. L. Collins, B. R. Murphy, and M. H. Skiadopoulos. 2004. Generation of recombinant human parainfluenza virus type 1 vaccine candidates by importation of temperature-sensitive and attenuating mutations from heterologous paramyxoviruses. J. Virol. 78:2017-2028.
Ottolini, M. G., D. D. Porter, V. G. Hemming, and G. A. Prince. 2000. Enhanced pulmonary pathology in cotton rats upon challenge after immunization with inactivated parainfluenza virus 3 vaccines. Viral Immunol. 13:231-236.
Pichichero, M. E. 2000. New combination vaccines. Pediatr. Clin. North Am. 47:407-426.
Ritter, T., M. Lehmann, and H. D. Volk. 2002. Improvements in gene therapy: averting the immune response to adenoviral vectors. BioDrugs 16:3-10.
Roberts, A., E. Kretzschmar, A. S. Perkins, J. Forman, R. Price, L. Buonocore, Y. Kawaoka, and J. K. Rose. 1998. Vaccination with a recombinant vesicular stomatitis virus expressing an influenza virus hemagglutinin provides complete protection from influenza virus challenge. J. Virol. 72:4704-4711.
Shinya, K., Y. Fujii, H. Ito, T. Ito, and Y. Kawaoka. 2004. Characterization of a neuraminidase-deficient influenza a virus as a potential gene delivery vector and a live vaccine. J. Virol. 78:3083-3088.
Steinhoff, M. C., N. A. Halsey, M. H. Wilson, B. A. Burns, R. K. Samorodin, L. F. Fries, B. R. Murphy, and M. L. Clements. 1990. Comparison of live attenuated cold-adapted and avian-human influenza A/Bethesda/85 (H3N2) reassortant virus vaccines in infants and children. J. Infect. Dis. 162:394-401.
Swayne, D. E., D. L. Suarez, S. Schultz-Cherry, T. M. Tumpey, D. J. King, T. Nakaya, P. Palese, and A. Garcia-Sastre. 2003. Recombinant paramyxovirus type 1-avian influenza-H7 virus as a vaccine for protection of chickens against influenza and Newcastle disease. Avian Dis. 47:1047-1050.
Takimoto, T., J. L. Hurwitz, C. Coleclough, C. Prouser, S. Krishnamurthy, X. Zhan, K. Boyd, R. A. Scroggs, B. Brown, Y. Nagai, A. Portner, and K. S. Slobod. 2004. Recombinant Sendai virus expressing the G glycoprotein of respiratory syncytial virus (RSV) elicits immune protection against RSV. J. Virol. 78:6043-6047.
Wagner, R., M. Matrosovich, and H. D. Klenk. 2002. Functional balance between haemagglutinin and neuraminidase in influenza virus infections. Rev. Med. Virol. 12:159-166.
Walsh, E. P., M. D. Baron, L. F. Rennie, P. Monaghan, J. Anderson, and T. Barrett. 2000. Recombinant rinderpest vaccines expressing membrane-anchored proteins as genetic markers: evidence of exclusion of marker protein from the virus envelope. J. Virol. 74:10165-10175.
Watanabe, T., S. Watanabe, T. Noda, Y. Fujii, and Y. Kawaoka. 2003. Exploitation of nucleic acid packaging signals to generate a novel influenza virus-based vector stably expressing two foreign genes. J. Virol. 77:10575-10583.(Yasuko Maeda, Masato Hatt)
Core Research for Evolutional Science and Technology (CREST), Japan Science and Technology Agency, Saitama 332-0021, Japan
Department of Pathobiological Sciences, University of Wisconsin-Madison, Madison, Wisconsin 53706
Institute for Molecular Virology, University of Wisconsin-Madison, Madison, Wisconsin 53706
ABSTRACT
Influenza and human parainfluenza virus infections are of both medical and economical importance. Currently, inactivated vaccines provide suboptimal protection against influenza, and vaccines for human parainfluenza virus infection are not available, underscoring the need for new vaccines against these respiratory diseases. Furthermore, to reduce the burden of vaccination, the development of multivalent vaccines is highly desirable. Thus, to devise a single vaccine that would elicit immune responses against both influenza and parainfluenza viruses, we used reverse genetics to generate an influenza A virus that possesses the coding region for the hemagglutinin/neuraminidase ectodomain of parainfluenza virus instead of the influenza virus neuraminidase. The recombinant virus grew efficiently in eggs but was attenuated in mice. When intranasally immunized with the recombinant vaccine, all mice developed antibodies against both influenza and parainfluenza viruses and survived an otherwise lethal challenge with either of these viruses. This live bivalent vaccine has obvious advantages over combination vaccines, and its method of generation could, in principle, be applied in the development of a "cocktail" vaccine with efficacy against several different infectious diseases.
INTRODUCTION
Influenza and human parainfluenza virus infections rank high among human respiratory diseases worldwide. Influenza accounts for 3 to 5 million cases of severe illness each year, 250,000 to 500,000 of which are fatal, while an estimated 16,300 to 96,500 people are hospitalized because of human parainfluenza virus infections (see the World Health Organization website at http://www.who.int/). Considerable effort has been expended to develop useful vaccines against influenza and human parainfluenza viruses. Inactivated influenza virus vaccines are safe and effective in preventing influenza virus-related hospitalization and death, but the immune response triggered by these vaccines is suboptimal and short lived, requiring annual vaccination (12). By contrast, live attenuated influenza virus vaccines elicit robust humoral and cellular responses that are predicted to be long lived (6). For human parainfluenza viruses, neither inactivated nor live vaccines are available. Although no data are available from studies with humans, formalin-inactivated vaccines for parainfluenza virus in animals aggravate the symptoms of infection upon challenge with homologous virus (23), as previously shown in humans and animal models given inactivated vaccines for respiratory syncytial virus (RSV) (a close relative of parainfluenza virus) (5, 9). This enhancement of pathogenicity was not observed for a live attenuated candidate vaccine, which proved to be effective in animal studies (22). Thus, for the prevention of both influenza and human parainfluenza virus infections, live attenuated vaccines are likely to be more effective than inactivated vaccines.
Multivalent vaccines containing protective antigens against several infectious agents have been developed as a means to reduce the burden of repeated childhood vaccinations (1). With the exception of the MMR (measles, mumps, and rubella) preparation, they are based on inactivated vaccines because propagation of one vaccine virus strain might interfere with the growth of the other vaccine component(s) (24). A possible solution to this problem would be to produce a single vaccine virus possessing several antigens that elicit protection against two or more infectious agents.
Influenza A virus (family Orthomyxoviridae, characterized by a segmented, negative-strand RNA genome) is considered an attractive vaccine vector candidate. It elicits strong humoral and cellular responses (6), its genome is amenable to genetic modification (7, 21), and the availability of 15 hemagglutinin (HA) and 9 neuraminidase (NA) subtypes (HA and NA are the major viral antigens) would allow repeated immunization. However, early attempts to express foreign proteins from recombinant influenza viruses failed due to the genetic instability of the foreign genes (15). This limitation was recently overcome when we identified regions in influenza viral RNA (vRNA) segments that govern the efficient incorporation of vRNA into virions (8, 33). These incorporation signals, which reside in the coding regions, have been exploited for the stable expression of foreign proteins from influenza viruses (27, 33).
Thus, using our knowledge of influenza vRNA incorporation signals, we aimed to create a bivalent live vaccine that would provide protection against both influenza and parainfluenza virus infections. This objective was pursued by generating an influenza A virus in which the ectodomain of the NA was replaced by that of the HN, a parainfluenza virus protective antigen, and then evaluating the protective potential of the vaccine in a mouse model.
MATERIALS AND METHODS
Cells. 293T human embryonic kidney cells (a derivative of the 293 line into which the gene for simian virus 40 T antigen was inserted) and Madin-Darby canine kidney (MDCK) cells were maintained in Dulbecco's modified Eagle medium supplemented with 10% fetal calf serum and in minimum essential medium containing 5% newborn calf serum, respectively. All cells were maintained at 37°C in 5% CO2.
Viruses. Influenza A/WSN/33 (H1N1) virus (WSN) was generated by plasmid-based reverse genetics as previously described (21) and propagated in 10-day-old embryonated chicken eggs. A murine parainfluenza virus type 1, Sendai virus (SeV; strain Enders) was propagated in 10-day-old embryonated chicken eggs.
Plasmid construction. Plasmid pPolINA(61)NA/HNect (Fig. 1), which was used for the production of a negative-sense vRNA containing the coding sequence for an NA-HN chimeric protein between the 3' and 5' NA vRNA incorporation signals (8), was generated as follows. The region corresponding to the WSN NA cytoplasmic tail, transmembrane domain, and a portion of the stalk (i.e., amino acids 1 to 37 of WSN NA) was amplified by PCR. Likewise, we amplified the region corresponding to the SeV HN ectodomain (i.e., amino acids 61 to 576 of SeV HN), followed by two consecutive stop codons. The resultant PCR products were inserted between the 3' and 5' NA vRNA incorporation signals (i.e., the 3' noncoding region plus the first 183 nucleotides [nt] of the coding region with replacement of the start codon, ATG by GCG, and the 5' noncoding region plus the last 157 nt of the coding region). The recombinant influenza viral cDNA thus produced was inserted between RNA polymerase I promoter and terminator sequences.
Generation of influenza virus expressing SeV HN antigen. To generate influenza virus expressing SeV HN protein, designated FluH/SeVHN, we performed reverse genetics as previously described (21). Briefly, eight RNA polymerase I-driven plasmids for transcription of all eight influenza vRNA segments, including the NA/HNect segment, were transfected into 293T cells as previously described (21). We then cotransfected cells with five eukaryotic protein expression plasmids encoding NA, as well as the PA, PB1, PB2, and NP proteins, which are required for replication and transcription of vRNAs. Two days posttransfection, virus was collected from the supernatant of transfected cells, grown once in 10-day-old embryonated chicken eggs, and stored at –80°C until used in experiments.
Experimental infection. The pathogenicity of FluH/SeVHN was tested by intranasal infection of 4-week-old female BALB/c mice (The Jackson Laboratory, Bar Harbor, ME) that were anesthetized with isoflurane. Survival times and body weights were monitored daily for 14 days after infection. Three days postinfection, three mice per group were euthanized to determine virus titers in nasal turbinates and lungs. The dose lethal to 50% of mice (MLD50) infected with wild-type WSN or SeV was determined by intranasally inoculation of 8-week-old female BALB/c mice with 10-fold serial dilutions of virus under anesthetization; infected mice were observed daily for 14 days.
To evaluate the protective efficacy of FluH/SeVHN against challenge with lethal doses of wild-type viruses, we infected each mouse intranasally with FluH/SeVHN (105.3, 106.3, or 107.3 50% egg infectious doses [EID50]/mouse). As controls, we infected mice intranasally with formalin-inactivated virus (equivalent to 106.3 EID50/mouse) or phosphate-buffered saline (PBS). Twenty-seven days later, serum samples as well as trachea-lung and nasal washes were collected from a subset of mice and examined for virus-specific immunoglobulin A (IgA) or IgG antibodies by an enzyme-linked immunosorbent assay (ELISA) (13). Briefly, wells of Immulon 2HB (Thermo Labsystems, Franklin, MA) plates were coated with purified and detergent-treated WSN or SeV and blocked with PBS containing 10 mg/ml bovine serum albumin. After incubation of virus-coated wells with test samples, bound antibodies were detected with goat anti-mouse IgA or IgG conjugated to horseradish peroxidase (Kirkegaard & Perry Laboratories Inc., Gaithersburg, MD). Statistical analyses were performed with Student's t test. Twenty-eight days after infection, the remaining mice were challenged intranasally with 10 MLD50 of wild-type WSN or SeV and monitored daily for survival and body weight for 28 days. Virus titers were determined in organs from five mice per group at 3 days postchallenge.
RESULTS
Generation and characterization of an influenza A virus expressing SeV HN antigen. To generate an influenza virus-based live, bivalent vaccine, we chose the influenza A/WSN/33 (H1N1) virus and the Enders strain of SeV, a murine type 1 parainfluenza virus, since both cause lethal infection in mice, providing a useful animal model with which to evaluate vaccine efficacy. Moreover, we recently mapped the sequences that mediate efficient WSN NA vRNA virion incorporation to 183 nt, corresponding to the amino-terminal coding region of NA, and to 157 nt, corresponding to the carboxy-terminal coding region of this protein, in addition to the 3' and 5' noncoding regions (8). With this information, we were able to generate a recombinant WSN virus expressing the ectodomain of SeV HN (designated FluH/SeVHN), as described in Materials and Methods. To determine if FluH/SeVHN expressed both the SeV HN and WSN HA glycoproteins, we infected MDCK cells with this virus or wild-type WSN (as a control) and stained cells 48 h later with antibodies to SeV HN, WSN HA, or influenza virus NP. Cells infected with influenza virus expressed HA and NP proteins (Fig. 2) but not SeV HN, whereas cells infected with FluH/SeVHN expressed both influenza virus HA and NP proteins and the SeV HN protein.
To determine whether the NA-HN chimeric protein was expressed on the cell surface, we performed a flow cytometry analysis of HN antigen on 293T cells transfected with pPolINA(61)NA/HNect plasmid, encoding the chimeric HN protein and those expressing the three polymerase and NP proteins. HN antigens were detected on the surface of these cells but not on control cells, which did not receive the polymerase I plasmid. Next, to determine whether FluH/SeVHN virus particles possess the HN antigen, we carried out ELISA. Briefly, wells of Immulon 2HB plates were coated with purified and detergent-treated FluH/SeVHN, WSN, or SeV; SeV HN and influenza virus NP were detected by monoclonal antibodies. The wells coated with FluH/SeVHN or SeV reacted with antibodies against SeV HN protein, while those coated with WSN did not (data not shown). By contrast, the wells coated with FluH/SeVHN or WSN, but not SeV, reacted with an antibody against influenza virus NP. These results demonstrate that SeV HN antigen is present in FluH/SeVHN virions.
For vaccine production, the virus must grow to high titers. We therefore evaluated the titers of the recombinant virus grown in MDCK cells or embryonated chicken eggs. In MDCK cells, FluH/SeVHN replication was attenuated compared with that of wild-type influenza virus (1.0 x 106 PFU/ml versus 1.6 x 108 PFU/ml), resulting in pinpoint plaques only (Fig. 2). In eggs, however, FluH/SeVHN replicated to titers comparable to those of wild-type influenza virus (109.3 EID50/ml for the former versus 108.3 EID50/ml for the latter).
The stability of FluH/SeVHN was assessed by passaging the virus 10 times in embryonated chicken eggs. During 10 serial passages, FluH/SeVHN virus grew well, with titers ranging from 108.5 EID50/ml to 109.8 EID50/ml. Even after 10 serial passages in eggs, 88% of the plaques expressed SeV HN when tested in MDCK cells, and only one mutation (Asn to Lys at amino acid position 229) was found in the NA/HNect vRNA. Since not all virions of even wild-type influenza virus contain all eight RNA segments (17), the finding that not all of the FluH/SeVHN plaques expressed the NA/HNect was anticipated. These results demonstrate that influenza vRNA packaging signals can be exploited to stably express foreign genes.
Pathogenicity of FluH/SeVHN virus in mice. Attenuation in animals is a critical requirement for live vaccines. We therefore determined the pathogenicity of FluH/SeVHN virus in mice by intranasally infecting them with 105.3, 106.3, or 107.3 EID50/mouse of recombinant virus that had been passed once in eggs. Mice inoculated with 107.3 EID50 or less showed no overt symptoms of illness. Thus, the MLD50 of FluH/SeVHN virus was calculated to be >107.3 EID50/mouse, while that of wild-type WSN virus in mice of the same age was 103.0 EID50/mouse (16). Virus titers in the lungs of mice (three mice/group) inoculated with 105.3, 106.3, or 107.3 EID50 of FluH/SeVHN virus ranged from 104.3 to 104.6 EID50/g at 3 days postinfection (Table 1), while those in nasal turbinates ranged from 103.9 to 104.0 EID50/g, lower than titers measured after inoculation with 105.9 EID50 of wild-type influenza virus (107.4 EID50/g for lung and 105.3 EID50/g for nasal turbinates) (16). Considered together, the results indicate that FluH/SeVHN virus is highly attenuated in mice.
Antibody responses of mice immunized with FluH/SeVHN virus. To determine if infection of mice with FluH/SeVHN induces antibodies against both SeV and WSN, we intranasally inoculated the animals with 105.3, 106.3, or 107.3 EID50 of virus. Mice intranasally inoculated with PBS or a dose equivalent to 106.3 EID50 of formalin-inactivated vaccine virus served as controls. Twenty-seven days after inoculation, mice were euthanized, and their sera, as well as trachea-lung and nasal washes, were tested for IgA and IgG antibody levels against both viruses by ELISA (Fig. 3). Compared to mice infected with PBS or formalin-inactivated virus, those infected with 106.3 or 107.3 EID50 of FluH/SeVHN virus had elevated antibody levels to SeV in trachea-lung washes (both IgA and IgG) and in sera (IgG), but not in nasal washes. For mice inoculated with 105.3 EID50 of vaccine virus, antibody titers to SeV did not increase significantly. For the antibody response to WSN, both IgA and IgG antibody titers were significantly elevated (P < 0.01) in all samples derived from mice infected with FluH/SeVHN. Mice inoculated with PBS or formalin-inactivated vaccine virus did not show any significant IgA or IgG antibody response to either WSN virus or SeV. These results indicate that 106.3 EID50/mouse of FluH/SeVHN virus induces significant antibody responses against both WSN and SeV in mice.
Protective efficacy of FluH/SeVHN. To evaluate the protective efficacy of FluH/SeVHN, we immunized mice with live virus (105.3, 106.3, or 107.3 EID50), formalin-inactivated vaccine virus (equivalent to 106.3 EID50), or PBS and challenged them with 10 MLD50 of wild-type SeV (106.5 PFU/mouse) or WSN (105.5 PFU/mouse) at 28 days postimmunization. Mice immunized with 106.3 or 107.3 EID50 of live FluH/SeVHN virus were completely protected from challenge with a lethal dose of SeV, while immunization with 105.3 EID50 provided only partial protection (Table 2). By contrast, control animals immunized with formalin-inactivated virus or PBS died on days 6 to 10 postchallenge. For mice immunized with 107.3 EID50 of live vaccine virus, SeV titers in lungs and nasal turbinates were lower than in mice immunized with PBS on day 3 postchallenge (Student's t test; P < 0.05). However, virus titers in the organs of mice immunized with lower doses were not significantly lower than those in the control group, in spite of the protection observed in the immunized groups (Table 2). Conceivably, viral clearance at later time points postchallenge might have been more efficient in the immunized animals. All mice that survived SeV challenge lost body weight with a maximum loss of approximately 26% occurring by day 6 or 7 postchallenge (Fig. 4). After challenge with lethal doses of WSN, none of the mice immunized with live vaccine virus died or showed weight loss; by contrast, mice immunized with formalin-inactivated virus or PBS succumbed to infection (Table 3). WSN titers in lungs and nasal turbinates of mice immunized with vaccine virus were below the detection limits (<100 PFU/g) on day 3 postchallenge. Taken together, these data indicate that FluH/SeVHN provides efficient protection of mice against challenge with lethal doses of SeV or WSN.
DISCUSSION
Here, we describe the development of a live bivalent vaccine against two major respiratory infections: influenza virus and parainfluenza virus. Using reverse genetics, we produced a recombinant influenza virus, FluH/SeVHN, that expresses the ectodomain of SeV HN (an antigen that elicits protection against SeV infection) instead of the influenza virus NA, which is not as important as the HA for induction of a protective immune response (19). FluH/SeVHN protected mice against lethal challenge with influenza and murine parainfluenza viruses. It also replicated efficiently in eggs, stably expressed the HN antigen, was highly attenuated in mice, and induced protective antibodies against both parainfluenza and influenza viruses in these animals. Hence, it meets the basic requirements for a live attenuated vaccine.
The development of viral vector systems for vaccine delivery is being vigorously pursued. DNA viruses and positive-sense RNA viruses have attracted the most attention as possible vaccine vectors, although with the advent of reverse genetics, the potential of negative-sense RNA viruses can be explored as well. In contrast to DNA viruses, the genetic material of RNA viruses (with the exception of members of the Retroviridae family) cannot be integrated into the host cell genome, adding to their biological safety. Among the group of nonsegmented negative-sense RNA viruses, Newcastle disease virus (20), vesicular stomatitis virus (14, 26), rinderpest virus (32), and Sendai virus (18, 30) have been tested as vaccine vectors. The use of segmented negative-sense RNA viruses as vaccine vectors has been impeded by the difficulty of artificially generating viruses that contain multiple gene segments (eight for influenza A virus). Moreover, early attempts to express foreign genes from a ninth segment were not encouraging, due to genetic instability (15). These limitations have been overcome with the establishment of a plasmid-based reverse genetics system for influenza virus (21) and our recent identification of segment-specific virion incorporation signals in the influenza vRNA segments (8, 33). Here, we extended our earlier findings by generating an influenza A virus that contains a parainfluenza virus HN ectodomain and by demonstrating the protective efficacy of this new vaccine candidate in a mouse model.
The high mutation rate and the potential reassortment of influenza viruses have raised some concern over their use as live vaccines. The recently approved live attenuated influenza virus vaccine relies on a limited number of amino acid replacements to confer the requisite cold-adapted, temperature-sensitive, and attenuated phenotypes; hence, point mutations could result in the loss of such phenotypes. By contrast, the vaccine virus described in this report is attenuated through replacement of the NA ectodomain with that of SeV HN, so that point mutations are highly unlikely to result in a loss of attenuation.
Combined vaccination, in which recipients are immunized with vaccines for multiple pathogens at once, is highly desirable in the clinic. However, simultaneous immunization with two or more live vaccine viruses can result in immunization failure if one virus affects the replication and/or viral pathogenicity of the other viruses (24). With vector systems, immune responses against vector components can present major obstacles, prohibiting repeated immunization with the same vector backbone, as noted for adenoviral vectors (25). In our approach, the immunogenicity of the influenza virus vector is considered beneficial, since it would elicit immune responses against a societally and economically important pathogen. Indeed, we found that an influenza virus expressing the SeV HN ectodomain induced significant antibody levels against both influenza and parainfluenza viruses and fully protected immunized mice from challenge with lethal doses of the parental viruses. Swayne et al. (29), using a strategy similar to ours, generated a recombinant Newcastle disease virus that expressed the HA protein of an avian influenza virus; however, the vaccine failed to fully protect chickens from challenge with lethal doses of Newcastle disease or influenza virus. This discrepancy may reflect differences in the immunogenic potential of the two viruses, the doses used for protection and/or challenge, or the type of animal model used.
The known protective effect of the SeV HN ectodomain during viral infection led us to choose it as an SeV antigen. In addition, we anticipated that its sialidase activity would compensate for the lack of this enzyme in the neuraminidase-deficient influenza A virus vector. Efficient replication of WSN NA/HNect virus in eggs and poor replication in eggs of a recombinant virus expressing green fluorescent protein instead of NA (hence lacking sialidase activity; data not shown) substantiated this concept. The effect of balanced HA and NA activities on influenza virus replication depends on culture system (i.e., mammalian cell culture and embryonated chicken eggs) (3, 31). For example, WSN viruses with deletions in the stalk region of the NA protein replicated as well as wild-type WSN virus in cultured cells; however, some of these mutant viruses replicated poorly in embryonated chicken eggs (2). For FluH/SeVHN virus, we replaced the NA ectodomain with that of SeV HN, likely affecting the balance between the receptor-binding and -destroying activities of the viral glycoproteins and resulting in attenuated phenotypes in cell culture. By contrast, the imbalanced HA and NA activities were overcome in eggs, allowing efficient FluH/SeVHN replication.
The FluH/SeVHN virus we generated appears to have the essential properties of an ideal live vaccine. It is safe, genetically stable, and strongly immunogenic, and its growth profile in vitro is compatible with the requirements for efficient vaccine production. Hence, our results could open a new avenue for the production of multivalent vaccines. For example, the live influenza virus vaccine currently in use is a mixture of two type A viruses (subtypes H1N1 and H3N2) and a type B virus. Thus, by replacing the NA of these influenza viruses with three different genes encoding the HNs of human parainfluenza virus types 1 and 3 (the two most common human parainfluenza viruses) and the G or F of RSV, one could generate a protective cocktail vaccine against six types of respiratory tract infections. Human parainfluenza virus and RSV infection primarily occur in the first year of life, and influenza can also occur in patients under 6 months of age (10). Although live influenza virus vaccine is not currently recommended for children under 5 years of age, the safety and efficacy of live attenuated influenza virus vaccines in infants under 1 year of age have been shown (4, 11, 28). Moreover, the safety and efficacy of the current live attenuated influenza virus vaccine can be further improved by using reverse genetics. Thus, we believe that influenza virus-based vaccines for younger children will be used in the future. Ultimately, by exploiting the known packaging signals of influenza vRNAs, it may be possible to add genes encoding the protective antigens of any infectious agent.
ACKNOWLEDGMENTS
We thank Yutaka Fujii for scientific advice, Martha McGregor and Krisna Wells for excellent technical assistance, the animal care staff at the University of Wisconsin-Madison for the professional care of mice, and John Gilbert for editing the manuscript.
This study was supported by National Institute of Allergy and Infectious Diseases Public Health Service research grants; by CREST (Japan Science and Technology Agency); by grants-in-aid from the Ministry of Education, Culture, Sports, Science, and Technology of Japan; by the Ministry of Health, Labor, and Welfare of Japan; and by research fellowships of the Japan Society for the Promotion of Science for Young Scientists.
REFERENCES
Capiau, C., J. Poolman, B. Hoet, H. Bogaerts, and F. Andre. 2003. Development and clinical testing of multivalent vaccines based on a diphtheria-tetanus-acellular pertussis vaccine: difficulties encountered and lessons learned. Vaccine 21:2273-2287.
Castrucci, M. R., and Y. Kawaoka. 1993. Biologic importance of neuraminidase stalk length in influenza A virus. J. Virol. 67:759-764.
Clavijo, A., D. B. Tresnan, R. Jolie, and E. M. Zhou. 2002. Comparison of embryonated chicken eggs with MDCK cell culture for the isolation of swine influenza virus. Can J. Vet. Res. 66:117-121.
Clements, M. L., M. K. Makhene, R. A. Karron, B. R. Murphy, M. C. Steinhoff, and K. Subbarao. 1996. Effective immunization with live attenuated influenza A virus can be achieved in early infancy. Pediatric Care Center. J. Infect. Dis. 173:44-51.
Collins, P. L., and B. R. Murphy. 2002. Respiratory syncytial virus: reverse genetics and vaccine strategies. Virology 296:204-211.
Cox, R. J., K. A. Brokstad, and P. Ogra. 2004. Influenza virus: immunity and vaccination strategies. Comparison of the immune response to inactivated and live, attenuated influenza vaccines. Scand. J. Immunol. 59:1-15.
Fodor, E., L. Devenish, O. G. Engelhardt, P. Palese, G. G. Brownlee, and A. Garcia-Sastre. 1999. Rescue of influenza A virus from recombinant DNA. J. Virol. 73:9679-9682.
Fujii, Y., H. Goto, T. Watanabe, T. Yoshida, and Y. Kawaoka. 2003. Selective incorporation of influenza virus RNA segments into virions. Proc. Natl. Acad. Sci. USA 100:2002-2007.
Fulginiti, V. A., J. J. Eller, O. F. Sieber, J. W. Joyner, M. Minamitani, and G. Meiklejohn. 1969. Respiratory virus immunization. I. A field trial of two inactivated respiratory virus vaccines; an aqueous trivalent parainfluenza virus vaccine and an alum-precipitated respiratory syncytial virus vaccine. Am. J. Epidemiol. 89:435-448.
Glezen, W. P., L. H. Taber, A. L. Frank, W. C. Gruber, and P. A. Piedra. 1997. Influenza virus infections in infants. Pediatr. Infect. Dis. J. 16:1065-1068.
Gruber, W. C., K. Kirschner, S. Tollefson, J. Thompson, G. Reed, K. M. Edwards, and P. F. Wright. 1993. Comparison of monovalent and trivalent live attenuated influenza vaccines in young children. J. Infect. Dis. 168:53-60.
Harper, S. A., K. Fukuda, N. J. Cox, and C. B. Bridges. 2003. Using live, attenuated influenza vaccine for prevention and control of influenza: supplemental recommendations of the Advisory Committee on Immunization Practices (ACIP). Morb. Mortal. Wkly. Rep. Recomm. Rep. 52(RR-13):1-8.
Kida, H., L. E. Brown, and R. G. Webster. 1982. Biological activity of monoclonal antibodies to operationally defined antigenic regions on the hemagglutinin molecule of A/Seal/Massachusetts/1/80 (H7N7) influenza virus. Virology 122:38-47.
Kretzschmar, E., L. Buonocore, M. J. Schnell, and J. K. Rose. 1997. High-efficiency incorporation of functional influenza virus glycoproteins into recombinant vesicular stomatitis viruses. J. Virol. 71:5982-5989.
Luytjes, W., M. Krystal, M. Enami, J. D. Pavin, and P. Palese. 1989. Amplification, expression, and packaging of foreign gene by influenza virus. Cell 59:1107-1113.
Maeda, Y., H. Goto, T. Horimoto, A. Takada, and Y. Kawaoka. 2004. Biological significance of the U residue at the –3 position of the mRNA sequences of influenza A viral segments PB1 and NA. Virus Res. 100:153-157.
Martin, K., and A. Helenius. 1991. Nuclear transport of influenza virus ribonucleoproteins: the viral matrix protein (M1) promotes export and inhibits import. Cell 67:117-130.
Matano, T., M. Kobayashi, H. Igarashi, A. Takeda, H. Nakamura, M. Kano, C. Sugimoto, K. Mori, A. Iida, T. Hirata, M. Hasegawa, T. Yuasa, M. Miyazawa, Y. Takahashi, M. Yasunami, A. Kimura, D. H. O'Connor, D. I. Watkins, and Y. Nagai. 2004. Cytotoxic T lymphocyte-based control of simian immunodeficiency virus replication in a preclinical AIDS vaccine trial. J. Exp. Med. 199:1709-1718.
Mozdzanowska, K., K. Maiese, M. Furchner, and W. Gerhard. 1999. Treatment of influenza virus-infected SCID mice with nonneutralizing antibodies specific for the transmembrane proteins matrix 2 and neuraminidase reduces the pulmonary virus titer but fails to clear the infection. Virology 254:138-146.
Nakaya, T., J. Cros, M. S. Park, Y. Nakaya, H. Zheng, A. Sagrera, E. Villar, A. Garcia-Sastre, and P. Palese. 2001. Recombinant Newcastle disease virus as a vaccine vector. J. Virol. 75:11868-11873.
Neumann, G., T. Watanabe, H. Ito, S. Watanabe, H. Goto, P. Gao, M. Hughes, D. R. Perez, R. Donis, E. Hoffmann, G. Hobom, and Y. Kawaoka. 1999. Generation of influenza A viruses entirely from cloned cDNAs. Proc. Natl. Acad. Sci. USA 96:9345-9350.
Newman, J. T., J. M. Riggs, S. R. Surman, J. M. McAuliffe, T. A. Mulaikal, P. L. Collins, B. R. Murphy, and M. H. Skiadopoulos. 2004. Generation of recombinant human parainfluenza virus type 1 vaccine candidates by importation of temperature-sensitive and attenuating mutations from heterologous paramyxoviruses. J. Virol. 78:2017-2028.
Ottolini, M. G., D. D. Porter, V. G. Hemming, and G. A. Prince. 2000. Enhanced pulmonary pathology in cotton rats upon challenge after immunization with inactivated parainfluenza virus 3 vaccines. Viral Immunol. 13:231-236.
Pichichero, M. E. 2000. New combination vaccines. Pediatr. Clin. North Am. 47:407-426.
Ritter, T., M. Lehmann, and H. D. Volk. 2002. Improvements in gene therapy: averting the immune response to adenoviral vectors. BioDrugs 16:3-10.
Roberts, A., E. Kretzschmar, A. S. Perkins, J. Forman, R. Price, L. Buonocore, Y. Kawaoka, and J. K. Rose. 1998. Vaccination with a recombinant vesicular stomatitis virus expressing an influenza virus hemagglutinin provides complete protection from influenza virus challenge. J. Virol. 72:4704-4711.
Shinya, K., Y. Fujii, H. Ito, T. Ito, and Y. Kawaoka. 2004. Characterization of a neuraminidase-deficient influenza a virus as a potential gene delivery vector and a live vaccine. J. Virol. 78:3083-3088.
Steinhoff, M. C., N. A. Halsey, M. H. Wilson, B. A. Burns, R. K. Samorodin, L. F. Fries, B. R. Murphy, and M. L. Clements. 1990. Comparison of live attenuated cold-adapted and avian-human influenza A/Bethesda/85 (H3N2) reassortant virus vaccines in infants and children. J. Infect. Dis. 162:394-401.
Swayne, D. E., D. L. Suarez, S. Schultz-Cherry, T. M. Tumpey, D. J. King, T. Nakaya, P. Palese, and A. Garcia-Sastre. 2003. Recombinant paramyxovirus type 1-avian influenza-H7 virus as a vaccine for protection of chickens against influenza and Newcastle disease. Avian Dis. 47:1047-1050.
Takimoto, T., J. L. Hurwitz, C. Coleclough, C. Prouser, S. Krishnamurthy, X. Zhan, K. Boyd, R. A. Scroggs, B. Brown, Y. Nagai, A. Portner, and K. S. Slobod. 2004. Recombinant Sendai virus expressing the G glycoprotein of respiratory syncytial virus (RSV) elicits immune protection against RSV. J. Virol. 78:6043-6047.
Wagner, R., M. Matrosovich, and H. D. Klenk. 2002. Functional balance between haemagglutinin and neuraminidase in influenza virus infections. Rev. Med. Virol. 12:159-166.
Walsh, E. P., M. D. Baron, L. F. Rennie, P. Monaghan, J. Anderson, and T. Barrett. 2000. Recombinant rinderpest vaccines expressing membrane-anchored proteins as genetic markers: evidence of exclusion of marker protein from the virus envelope. J. Virol. 74:10165-10175.
Watanabe, T., S. Watanabe, T. Noda, Y. Fujii, and Y. Kawaoka. 2003. Exploitation of nucleic acid packaging signals to generate a novel influenza virus-based vector stably expressing two foreign genes. J. Virol. 77:10575-10583.(Yasuko Maeda, Masato Hatt)