|Home | About | Journals | Submit | Contact Us | Français|
Despite the reported efficacy of commercially available influenza virus vaccines, a considerable proportion of the human population does not respond well to vaccination. In an attempt to improve the immunogenicity of live influenza vaccines, an attenuated, cold-adapted (ca) influenza A virus expressing human interleukin-2 (IL-2) from the NS gene was generated. Intranasal immunization of young adult and aged mice with the IL-2-expressing virus resulted in markedly enhanced mucosal and cellular immune responses compared to those of mice immunized with the nonrecombinant ca parent strain. Interestingly, the mucosal immunoglobulin A (IgA) and CD8+ T-cell responses in the respiratory compartment could be restored in aged mice primed with the IL-2-expressing virus to magnitudes similar to those in young adult mice. The immunomodulating effect of locally expressed IL-2 also gave rise to a systemic CD8+ T-cell and distant urogenital IgA response in young adult mice, but this effect was less distinct in aged mice. Importantly, only mice immunized with the recombinant IL-2 virus were completely protected from a pathogenic wild-type virus challenge and revealed a stronger onset of virus-specific CD8+ T-cell recall response. Our findings emphasize the potential of reverse genetics to improve the efficacy of live influenza vaccines, thus rendering them more suitable for high-risk age groups.
Vaccination still remains the very best way to protect humans against influenza. Annual human influenza epidemics (caused by influenza type A or type B viruses) are manifested as highly infectious acute respiratory disease with high morbidity and significant mortality. Vaccination is accomplished at present with commercially available, chemically inactivated (killed) or live attenuated influenza virus vaccines (10, 31, 48).
The concept of the current live attenuated vaccine is based on the generation of a temperature-sensitive attenuated “master strain,” adapted to grow at 25°C (cold adaptation [ca]) in embryonated chicken eggs or cell cultures. The final vaccine strains contain a mixed genome consisting of six genome segments inherited from the master strain (ensuring the attenuation and ca phenotype of the vaccine virus) and two genome segments encoding the viral surface antigens hemagglutinin (HA) and neuraminidase, derived from the current epidemic viruses (30, 36, 56).
Live ca and inactivated vaccines stimulate the immune system differently. The inactivated virus vaccine is a better inducer of virus-specific serum antibody than the live ca vaccine. In contrast, live ca vaccines, apart from the advantage of painless nasal administration, also stimulate the induction of mucosal antibodies and cross-reactive cell-mediated immune responses, thus providing a broader and longer-lasting immunity (2, 3, 7, 12, 24). In children and healthy adults, serum hemagglutination inhibition antibodies and immunoglobulin G (IgG) and IgA antibodies to HA in respiratory secretions correlate with protection from infection (3, 11, 27). In healthy young adults, anti-influenza A virus cytotoxic T lymphocytes that are also capable of recognizing heterologous influenza A viruses (but not influenza B viruses), correlate with enhanced clearance and reduced shedding of wild-type (WT) influenza A virus (23, 35, 39).
The vaccine efficacies of the two types of vaccines have been reported to be comparable; however, about 30% of adults and 40 to 70% of the elderly (one of the high-risk groups) do not respond well to vaccination (7, 43, 44, 64). Moreover, recent reports indicate that the immunogenicity of the inactivated avian HA type 5 (H5)-based vaccines, a possible weapon against potential avian flu pandemics, is suboptimal in humans (45, 60). Therefore, attempts to improve the overall influenza vaccine immunogenicity are of paramount importance.
The previous investigation of a broad variety of adjuvants comprising immunomodulating cytokines as vaccine supplements has shown them to improve the immunogenicity of vaccines directed against various infectious diseases (1, 8, 47, 49, 53, 65). Utilization of human interleukin-2 (IL-2) as a vaccine supplement resulted, for example, in enhanced immunogenicity and protection efficacy of the inactivated influenza vaccine not only in young but also in aged hosts (1, 4, 5, 37). The beneficial immunomodulating effect of IL-2 on the immune system could also be demonstrated when IL-2 was encoded by different viral vectors, virus-like particles, or DNA candidate vaccines in approaches aimed at combating infectious diseases or in antitumor therapy programs (9, 16, 34, 38, 62).
Here, we suggest an approach based on a modern technology, reverse genetics, directed towards the generation of live influenza virus vaccines profiting from the benefits of the immunomodulating IL-2 cytokine adjuvant concept. By introducing an additional reading frame into the NS gene, we were able to generate a ca influenza virus expressing biologically active IL-2 in a backbone of the ca influenza A/Singapore/1/57 virus (ca Sing-IL-2) (32). Significantly enhanced virus-specific mucosal IgA and CD8+ T-cell responses were detected in young adult and aged mice primed intranasally (i.n.) with the ca Sing-IL-2 virus compared with the responses in mice immunized with the cold-adapted parent strain A/Singapore/1/57 (ca Sing) virus. Moreover, young adult mice immunized with the vaccine virus expressing IL-2 were completely protected against a homologous influenza virus challenge infection.
Vero cells (ATCC CCL-81) were adapted to and further cultivated in 1:1 Dulbecco's modified Eagle's medium (DMEM)-Ham's F-12 (Biochrom F4815) with 4 mM l-glutamine and protein-free supplement (proprietary formulation; Polymun Scientific GmbH, Austria). MDCK cells (ATCC CCL-34) were cultivated in DMEM-Ham's F-12 medium containing 2% heat-inactivated fetal calf serum (HyClone SH30071) and 4 mM l-glutamine.
Influenza virus A/Singapore/1/57 (Sing-WT) was propagated on 10-day-old embryonated hen eggs, and the viral titer was determined by a plaque assay on MDCK cells with agar overlay containing DMEM-Ham's F-12 medium, 4 mM l-glutamine, 5 μg/ml of trypsin, 0.01% DEAE-dextran (Pharmacia), and 0.6% agar (Sigma). ca Sing was prepared by serial passages in Vero cells at a suboptimal temperature (56). During this procedure, another spontaneous mutant virus appeared, ca A/Sing/57-NS1Δ87 (ca Sing-NS1Δ87). This virus contains an 87-amino-acid deletion in the RNA binding domain of the NS1 protein (amino acids 34 to 120). ca Sing-NS1Δ87 and ca Sing viruses were propagated on Vero cells at 33°C, and the viral titer was determined by a limiting dilution assay of the 50% tissue culture infective dose (TCID50). A hemagglutination assay was performed with a 0.5% suspension of chicken red blood cells at 4°C.
Recombinant viruses were rescued as described previously (32), with minor modifications. Briefly, Vero cells were cotransfected with plasmid pPol-NS1-125IL2StSt-HDV, expressing viral RNA of the recombinant NS gene, in addition to plasmids expressing RNP of influenza A/PR/8/34 and subsequently infected with the interferon-sensitive helper virus ca Sing-NS1Δ87. By using recombinant alpha interferon (IFN-α; human leukocyte derived [NIBSC First International Standard, 1999]) to select for recombinants, a pure virus expressing IL-2 was obtained after several rounds of plaque-to-plaque purification and was designated ca A/Sing/NS1-125IL2StSt (ca Sing-IL-2). The correct sequence was confirmed by nucleotide sequence analysis. IL-2 synthesis within the cells infected with ca Sing-IL-2 was confirmed by a functional test utilizing an IL-2-sensitive detector cell line, as described previously (32).
Young adult (8-week-old) and aged (18-month-old) BALB/c mice (Charles River, Germany) were infected i.n. with the ca Sing-IL-2 virus (2 × 105 PFU/mouse), the ca Sing virus (2 × 105 PFU/mouse), or phosphate-buffered saline (PBS; controls) under ether anesthesia. In a separate set of experiments, 8-week-old BALB/c mice were primed in the same way and were challenged i.n. 4 weeks later with a mouse-adapted Sing-WT virus (2 × 105 PFU/mouse).
Primed mice (three mice per group) were sacrificed at days 2, 4, and 6 after intranasal inoculation of the viral stocks. In the challenge experiment, primed mice were sacrificed 3 days following the i.n. challenge immunization with the Sing-WT virus. Lungs were aseptically removed and pooled, and 10% (wt/vol) tissue suspensions were prepared (17). The suspensions were centrifuged at 3,000 × g for 5 min and the supernatants assayed for infectious viral particles by a limiting dilution assay of the TCID50 on MDCK cells.
Blood was collected from the murine retro-orbital venous plexus 8, 16, and 24 days following priming, and the sera were prepared and stored at −20°C. To obtain murine nasal secretions, salivation was induced 3 weeks after immunization by intraperitoneal injection of 0.1 mg of pilocarpine-HCl (Sigma). Concurrent with the salivation, small amounts of nasal secretions in the nasal openings were observed and immediately absorbed with the aid of sterile wicks (Spectrum Laboratories, Inc., Houston, TX), which were then placed into 50 μl of ice-cold sterile PBS supplemented with protease inhibitors (Roche) for 2 h. Vaginal secretions were obtained from individual mice by the introduction of 50 μl sterile PBS containing protease inhibitors into the vaginal lumen and subsequent collection of the effluent. Group-specific nasal or vaginal secretion pools were prepared, vigorously vortexed, and centrifuged at 15,000 × g for 5 min prior to assay or storage at −20°C (17).
A modified enzyme-linked immunosorbent assay (ELISA) protocol was performed as described previously (18). Briefly, sucrose-purified and UV-inactivated ca Sing virus (adjusted to 20 hemagglutination units/well in carbonate buffer [pH 9.6]) was used as a coating antigen. Serial dilutions of individual mouse sera and pooled nasal and vaginal wash specimens in PBS containing 1% skim milk (Serva) were added to the coated plates, and the mixtures were incubated for 1.5 h at room temperature. Bound antibodies were detected with goat anti-mouse IgG1, IgG2a, or IgA conjugated with horseradish peroxidase (Zymed). The plates were stained with 3,3′,5,5′-tetramethylbenzidine (MBI Fermentas) as a substrate, and the absorbance was measured (wavelength, 450 nm). The cutoff value was defined as the mean value of absorption for the negative control sera plus 2 standard deviations.
Spleens and lymph nodes draining the murine respiratory tracts (mediastinal and tracheobronchial lymph nodes) of immunized mice (three mice per group) were collected at day 10 following priming (spleens and lymph nodes) or 60 h after the challenge with the mouse-adapted Sing-WT virus (lymph nodes). Spleens and draining lymph nodes were mechanically dissociated into single-cell suspensions by means of cell strainers (Falcon). The erythrocytes present in the cell suspensions were lysed with Tris-buffered ammonium chloride, washed several times, and finally resuspended in DMEM containing 10% fetal calf serum, penicillin, streptomycin, IL-2 (30 U/ml), and 50 μM 2-mercaptoethanol (18).
An immediate ex vivo CD8+ IFN-γ enzyme-linked immunospot (ELISPOT) assay utilizing the synthetic peptide TYQRTRALV, an H-2Kd-restricted immunodominant cytotoxic T-lymphocyte epitope of the conserved influenza A virus nucleoprotein (NP147-155), was performed as described previously (19). Briefly, twofold serial dilutions of cell populations derived from murine spleens or lymph nodes draining the respiratory tracts were transferred to wells coated with anti-IFN-γ monoclonal antibody (R4-6A2; BD PharMingen). Cells were incubated for 24 h at 37°C under 5% CO2 in DMEM containing 10% fetal calf serum, IL-2 (30 U/ml), penicillin, streptomycin, and 50 μM 2-mercaptoethanol in the presence of the peptide. A biotinylated anti-IFN-γ monoclonal antibody (XMG1.2; BD PharMingen) was utilized as a conjugate antibody, followed by incubation of plates with streptavidin peroxidase (0.25 U/ml; Boehringer Mannheim Biochemica). Spots representing IFN-γ-secreting CD8+ cells were developed utilizing the substrate 3-amino-9-ethylcarbazole (Sigma) in the presence of hydrogen peroxide in 0.1 M sodium acetate, pH 5.0. The spots were counted with the help of a dissecting microscope, and the results were expressed as the mean number of IFN-γ-secreting cells ± the standard error of the mean of triplicate cultures from two independent experiments. Cells of control mice or cells incubated in the absence of synthetic peptides developed <15 spots/106 cells. Since depletion of CD8+ T cells usually resulted in a 92% reduction of spot formation, cell separation was omitted in most assays (data not shown).
Endpoint serum ELISA titers are presented as geometric mean titers. The differences in numbers of IFN-γ-secreting CD8+ cells were analyzed using Student's t test. A P value of <0.01 was considered significant.
We previously generated a recombinant influenza A/PR8/NS1-125IL-2StSt virus expressing human IL-2 from its NS gene by using an overlapping stop-start codon expression strategy and showed an increased potency of this virus to induce a cytotoxic T-lymphocyte immune response (32). However, this viral vector, being not temperature sensitive, was capable of growing to high titers in mouse respiratory tracts and was therefore not suitable as a vaccine model. In order to investigate the effect of IL-2 in the context of an attenuated ca influenza virus vaccine, we introduced the recombinant NS-IL-2 genome segment into the backbone of ca Sing. The ca Sing-IL-2 vector and its parent ca Sing strain could replicate to high titers (106 to 107 TCID50s) in Vero and MDCK cells at 33°C but were attenuated at higher culture temperatures, growing on average at 39°C to titers 4 orders of magnitude lower. Both viruses possess similar replication potentials in mouse lungs, not exceeding titers of 2 × 102 TCID50s/g of tissue on day 2 postinfection (p.i.) (mean titers of 2 × 102 and 1.1 × 102 TCID50s/g of tissue for ca Sing and ca Sing-IL-2, respectively). No viral titer was detectable at days 4 and 6 postinfection.
In order to check whether the coexpression of IL-2 might enhance the immunogenic potential of this attenuated vaccine, young adult mice (10 mice/group) and aged mice (3 mice/group) were immunized once i.n. with 2 × 105 PFU per animal of the ca Sing or ca Sing-IL-2 virus, respectively. The peak virus-specific serum IgG1 and IgG2a titers of young adult and old mice were detected at day 24 p.i. by ELISA. The virus-specific IgG1 and IgG2a titers detected in sera of aged mice (ELISA titer range, 6,400 to 25,600) were significantly lower than those detected in sera of young adult mice (ELISA titer range, 51,200 to 204,800) (Fig. (Fig.1).1). The virus-specific serum IgG1 and IgG2 titers of young adult mice immunized with either the ca Sing or the ca Sing-IL-2 virus were indistinguishable. In aged mice, slightly lower serum IgG1 and IgG2a titers were detected in animals immunized with the ca Sing-IL-2 virus than in the group of mice immunized with the ca Sing virus (Fig. (Fig.1).1). No significant differences were observed in serum hemagglutination inhibition titers in either the old or the young adult mice (data not shown).
Since secretory IgA in the respiratory tract has been found previously to be critical in protection from influenza infection (11, 12), we were interested in analyzing virus-specific mucosal IgA in mouse nasal secretions. Surprisingly, we could detect an increase of at least 3.5-fold in virus-specific nasal IgA titers for both young adult and aged mice primed with ca Sing-IL-2 compared with titers of mice primed with the ca Sing virus (Fig. (Fig.2a2a).
Due to the concept of the common mucosal immune system (40), we also analyzed whether mice primed with ca Sing-IL-2 developed secretory antibodies at a site distal to the site of immunization, the urogenital tract. We could detect an increase in virus-specific vaginal-wash IgA titers in young adult mice. However, although aged mice primed with the ca Sing-IL-2 virus developed markedly increased virus-specific IgA titers compared with mice of the same age group primed with the parent ca Sing vaccine virus, these titers were still lower than those of the young adult mice primed with the ca Sing virus (Fig. (Fig.2b2b).
Since it is known that a deficiency in IL-2 and its receptor in aged hosts impairs the T-cell-mediated immune system (14, 25, 26), we investigated whether the ca Sing-IL-2 virus is capable of inducing an enhanced virus-specific cellular immune response in aged mice. Mice of both age groups primed with the ca Sing-IL-2 virus developed significant higher (P < 0.01) numbers of virus-specific CD8+ T cells in the systemic compartment, represented by single-cell populations derived from spleens, than did those immunized with the ca Sing virus (Fig. (Fig.3a).3a). The highest number (about 580 IFN-γ-secreting cells/million cells) of virus-specific CD8+ T cells was detected in single-cell populations derived from the spleens of young adult mice immunized with the ca Sing-IL-2 virus. The immunization of aged mice with the ca Sing-IL-2 virus resulted in about the same number of IFN-γ-spot-forming cells per million cells as those determined in young adult mice immunized with the ca Sing virus.
Mice primed with the ca Sing-IL-2 virus developed, irrespective of age, significantly increased (P < 0.01) numbers of virus-specific CD8+ T cells in the local respiratory compartment (lymph nodes draining the respiratory tract) compared with those of mice immunized with the ca Sing virus. Importantly, the ca Sing-IL-2 virus restores the virus-specific CD8+ T-cell-mediated immunity at this site in aged mice, developing virus-specific CD8+ T cells at magnitudes comparable to those of young adult mice immunized with the same virus (Fig. (Fig.3b3b).
Next, we investigated whether mice immunized with the ca Sing-IL-2 virus are capable of mounting an enhanced memory immune response following a challenge infection with a homotypic WT virus. Groups of young adult mice were primed with 2 × 105 PFU/animal of either ca Sing, ca Sing-IL-2, or PBS. Four weeks postimmunization, mice were challenged with 2 × 105 PFU/mouse of the Sing-WT virus. The virus-specific CD8+ T-cell-mediated recall response was investigated in murine lymph nodes draining the respiratory tract, a site of virus entry, 60 h postchallenge. Again, a significant increase (P < 0.01) in response was observed for the group of mice primed with the ca Sing-IL-2 virus, which developed numbers of virus-specific CD8+ T cells clearly surpassing those detected for the group of mice immunized with the ca Sing virus (Fig. (Fig.44).
The protection efficacy of the primary immunization was determined in relation to the potential of the WT virus to grow in the lungs of young adult mice challenged as described above. The Sing-WT virus titer in lungs of mice 3 days postchallenge has been determined. Whereas a single immunization with ca Sing-IL-2 resulted in the complete protection of all mice from the homologous viral challenge, only about 50% of mice were protected when immunized with the ca Sing virus (Table (Table1).1). As expected, none of the control mice was protected from the viral challenge.
Reverse genetics systems, developed in the 1990s, still remain a challenging and promising technology, enabling the creation of influenza viruses with modified replicative, pathogenic, antigenic, and immunogenic properties (20, 42). We and others have recently shown that it is possible to generate replication-deficient and thus safe, but still highly immunogenic, influenza A viruses by genetic engineering of the NS gene segment (15, 18, 32, 48, 51, 61). The NS genome segment encodes two proteins: the NS1 and NEP, both carrying out essential functions in the viral life cycle. The NS1 protein, which is about 230 amino acids long, was identified as a type I IFN antagonist and possesses several domains and sites for the binding of the diverse proteins required, e.g., for the processing and cellular transport of cellular mRNAs and the translation of viral mRNAs (22, 33, 50). Targeted modification of its specific domains resulted in decreased, or even absent, virus replication in vivo, mainly due to the abrogation of the NS1 protein's function to suppress the IFN response machinery or other innate signaling pathways within the infected cells (6, 13, 33, 46, 59, 63). Moreover, infected cells produced high levels of type I IFNs, chemokines, and other immunomodulating cytokines, thus stimulating the development of the innate and adaptive immune responses (18, 29, 59).
In the present work, we addressed the question of whether the efficacy of a cold-adapted influenza virus vaccine can be improved by generating a virus vector encoding human IL-2 from the NS gene segment. IL-2 plays a crucial role in the stimulation and maturation of the immune system and is frequently used as a vaccine supplement or genetic adjuvant (1, 4, 9, 16, 34, 37, 38, 52, 62). For this reason, we compared the immunogenicities of the attenuated influenza ca Sing virus and a modified variant which expresses biologically active IL-2 in young adult and aged mice.
Due to their ca phenotypes, both viruses were highly attenuated in mice, since the maximal viral load detected at day 2 p.i. did not exceed 2 × 102 PFU/g tissue. We assume that the truncation of the NS1 protein and thus alteration of its biological functions resulted in an additional attenuation of the ca Sing-IL-2 virus (titer of 1.1 × 102 TCID50s/g tissue) compared with that of the ca Sing virus (titer of 2 × 102 × TCID50s/g tissue).
We found that coexpression of human IL-2 markedly improved the immunogenicity of the ca Sing virus. The ca Sing-IL-2 virus induced significantly higher numbers of virus-specific CD8+ T cells and mucosal IgA in young adult and aged mice. It is noteworthy that the stimulation capacity of the ca Sing-IL-2 virus was observed in the cellular and mucosal arms of the immune system compared with the parent ca Sing virus but not in the humoral systemic compartment. Both viruses induced almost identical virus-specific serum IgG1 and IgG2 responses in mice, irrespective of age. Interestingly, when young adult and aged mice were primed with the ca Sing-IL-2 virus, increased virus-specific mucosal IgA was also detected at a site distal to the site of immunization, the urogenital tract. This observation supports the concept of the common mucosal immune system playing a critical role in protecting hosts against mucosal pathogens (40). The central role of secretory IgA in the protection of the upper respiratory tract from influenza infection has been repeatedly reported elsewhere (11, 12, 54, 55, 66). Since the mucosal respiratory tissue provides the first barrier to influenza virus penetration, an induction of enhanced secretory IgA levels by new-generation vaccines is highly desirable.
Interestingly, ca Sing-IL-2 induces the local mucosal IgA and CD8+ T-cell immune response to markedly higher magnitudes than does the ca Sing virus in both aged and young adult mice. At distant sites devoid of the virus-encoded IL-2 production, e.g., the urogenital tract (mucosal IgA) or the spleen (systemic CD8+ response), this immune response was clearly weaker, especially for aged mice.
These observations are most probably related to the decline of immune system functions repeatedly observed in aged hosts. Impairments in humoral and cellular immunities in aged hosts include, e.g., the development of suboptimal immunogen-specific mucosal IgA titers; decreased immunogen-specific IgG avidity; disorders in the proliferation of memory cells, which are unable to secrete IL-2 on stimulation by antigens; and lack of the CD28-costimulatory molecules on expanded oligoclonal populations of activated T cells, etc. (14, 21, 28, 41, 57, 58). Nevertheless, by priming with the ca Sing-IL-2 virus, it was possible to restore the distant immune response of aged mice almost to the levels in young adult mice immunized with ca Sing. These findings may partly explain the suboptimal efficacy of vaccines and the dramatically increased mortality in the influenza virus-infected elderly compared to those in young adults (41) but show simultaneously the high immunogenic potential of the IL-2-expressing influenza virus as a vaccine candidate especially suitable for the elderly.
It has to be noted that the ca Sing-IL-2 virus also positively stimulates the development of immunological memory. We observed a rapid onset of the recall CD8+ T-cell response 60 h postchallenge in lymph nodes draining the respiratory tracts of young adult mice primed with the ca Sing-IL-2 virus, surpassing the numbers of CD8+ T cells detected in the group of mice immunized with the ca Sing virus. Most importantly, young adult mice primed with the ca Sing-IL-2 virus were completely protected from a homotypic WT virus challenge, in contrast to ca Sing virus-primed animals.
This study demonstrates a reverse-genetics-based modeling approach to improve the immunogenicity of live influenza virus vaccines in adult hosts and especially in aged hosts who do not respond well to vaccination. Since the attenuation phenotype of existing live ca influenza vaccines has been shown to be determined by mutations in the polymerase genome segments but not in the NS gene segment, the modified NS gene encoding human IL-2 (responsible also for the cellular production of other immunomodulating cytokines) could be introduced, in theory, into any ca vaccine viral genome, thus replacing the vaccine virus's own NS gene. The partial truncation of the recombinant NS protein in the context of a live ca influenza virus master strain will ensure an enhanced local immunogenicity and, at the same time, an additional attenuation resulting in a high safety standard of live ca influenza vaccine strains. This technology also seems to be especially suitable for the development of more-efficient vaccines directed against new potential pandemic viruses, such as avian H5N1 influenza viruses. Moreover, this approach also eliminates the potential side effects observed upon systemic administration of IL-2 to hosts. Thus, we hope that this study provides the basis for the future development of safe and highly immunogenic live influenza virus vaccines.
We thank I. Wilson for comments on the paper and orthographical corrections. We kindly thank E. Jensen-Jarolim and E. Untersmayr (Medical University of Vienna) for providing us with aged mice. We also acknowledge the efforts of H. Fekete for the routine cell culture work.
This study received funding from Polymun Scientific GmbH, Vienna, Austria.
Published ahead of print on 13 September 2006.