Traditional methods of vaccination using intramuscular or subcutaneous routes have been successful but have limitations and safety issues. Additionally, these routes of vaccination often do not generate adequate immunity at the sites of infection particularly for mucosally transmitted viruses [29
]. Aerosol vaccination provides a method to overcome these obstacles, and offers advantages that include noninvasive delivery, reduced risk of transfer of blood-borne pathogens, elimination of needles, reduced medical waste, and potential dose-sparing [30
]. The need to improve methods of vaccine delivery has been recognized by the World Health organization (WHO), particularly for measles virus (MV) vaccination [20
], and recent studies evaluating aerosol administration of MV vaccine has shown aerosol vaccination to be a promising non-invasive alternative to subcutaneous (s.c.) injection [20
In this study, we examined aerosol delivery of influenza virus by means of a nebulizer that uses ultrasonic vibrating mesh to generate aerosol particles. We tested aerosol parameters that included median particle size, dose volume, and virus titer in the aerosol suspension, delivery time, and flow rate for optimal deposition of influenza virus in the airways. Assessing these parameters in mice can be difficult because the nares and airways are small, and the level of anesthesia can affect the outcome. These difficulties are well-documented for i.n. instillation in mice where the relative distribution of the instillation agent between the upper and lower respiratory tract and lungs is heavily influenced by delivery volume and the level of anesthesia [33
]. For example it has been shown that as the volume of fluid is increased during nasal instillation in anesthetized mice, there is a concomitant increase in relative dosing to the lungs [33
]. Moreover, any volume beyond 0.05 ml that is i.n. instilled results in a level lower respiratory tract deposition [33
]. Thus, it is not surprising that the distribution of aerosolized particles in the airways is dependent in part on volume delivered, particle size and flow rate.
In this study, we show no statistical difference in the level of infectious virus isolated from nasal washes or lungs of mice administered X31 using aerosol delivery of 20 or 30 micron median diameter particles, and as expected, the viral titers in nasal washes and lungs decreased with decreasing virus dose titers. The aerosolization rate required for effective delivery of live virus was evaluated at ranges of 0.1 cc/min. to 2.0 cc/min.; however, there was no significant difference in the level of infectious virus recovered from nasal washes at any flow rate tested. Interestingly, no infectious virus was detected in the lungs using a 0.1 cc/min suggesting this flow rate preferentially delivers virus to the upper airways. Of the three delivery routes examined in this study, i.n. and aerosol delivery were the most effective for eliciting high serum antibody titers against HA and reduced virus replication in the respiratory tract of mice challenged with homologous or heterologous virus, and aerosol delivery of X31 evoked robust immune responses even at very low doses (102 TCID50/ml).
Several important findings regarding the induction of immune responses following vaccination or challenge are reported here. One is that i.m. inoculation with X31 induced lower mean serum anti-HA antibody titers and no detectable cross-protection to PR8 challenge, while both i.n. and aerosol delivery protected against homologous and heterologous challenge. These findings are consistent with a previous study that showed nasal instillation or small particle aerosol vaccination of mice with an attenuated, temperature-sensitive recombinant influenza A virus induced similar levels of virus neutralizing antibodies and provided similar levels of protection as measured by recovery from a sub-lethal challenge with a virulent virus [34
]. Another finding is that lung IgA levels were higher for i.n. immunized mice compared to aerosol or i.m. inoculated mice prior to challenge, but at day 7 post-challenge, lung IgA titers were considerably higher for i.m vaccinated mice compared to the other inoculation routes. It has previously been reported in mice that maximal secondary IgA responses in the respiratory tract are achieved by a combination of intranasal priming and boosting with vaccine [35
]. However, our finding that i.m. inoculation boosts for higher mucosal IgA levels following challenge has also been observed in a different influenza virus study that examined parenteral DNA vaccination and boosting with antigen-matched recombinant adenovirus which induced strong IgA responses and better protection against morbidity following H1N1 and H5N1 challenge [36
]. In the challenge study shown in this report, i.n. or aerosol inoculation and challenge also induced substantial increases in lung IgG1 and IgG2b. While induction of IgG1 has a role in protection against influenza, the major serum isotype present in mice that survive lethal virus challenge is IgG2a even in the presence of low levels of IgG1 [37
]. Studies have shown that generation of IgG2a antibodies is associated with increased influenza vaccine efficacy as this isotype is more efficient at viral clearance [40
]. In the studies reported here, all inoculation routes induced similar levels of IgG1 both after vaccination and challenge; however, the aerosol exposed group had a higher increase in IgG2a levels after challenge compared to the i.n. inoculation group. These results further suggest aerosol vaccination may be a strategy to improve vaccine efficacy for influenza.
Existing licensed influenza vaccines include inactivated and live attenuated influenza virus formulations. Aerosol delivery is being developed for vaccinating against influenza virus infection using live attenuated influenza A virus. We evaluated aerosol delivery of influenza in mice, the most common model for preliminary vaccine efficacy studies. In these studies a mouse-adapted influenza A virus was used because live attenuated human influenza viruses do not replicate in mice; in general human influenza A viruses do not replicate in mice without prior adaptation [42
]. The ferret is considered to be the most suitable animal model for preclinical evaluation of human influenza vaccines. Influenza infection in ferrets closely mimics that in humans with respect to clinical signs, pathogenesis, and immunity and ferrets are naturally susceptible to infection with human influenza A viruses [43
]. Ferrets also share marked similarities to humans in terms of lung physiology, airway morphology and cell types present in the respiratory tract, including the distribution of α-2,6-linked sialic acids, the receptor for human influenza viruses [44
]. The further development of aerosol immunization using live attenuated influenza virus will use ferrets because it is the more relevant animal model for influenza vaccine studies.
With the burgeoning need for the development safe and effective vaccination strategies for influenza, this study shows that aerosol delivery induces robust and protective immunity. Further, the results from aspects of the study suggest that reducing aerosol flow rate may allow for upper airway targeted vaccine delivery, and that exceedingly low titers of virus can be used to vaccinate for robust antibody responses to HA. Clearly, aerosol vaccination strategies that induce robust and protective immunity and obviate the issues associated with parenteral vaccination offer an advantage, but perhaps the greatest is that aerosol vaccination offers a natural route of infection leading to immunity at the site of natural infection.