Respiratory immunisation and mucosal immunity against influenza

Influenza A virus (IAV) is a considerable threat to human and animal health. Current immunisation strategies rely primarily on inducing strain-specific neutralising antibodies with a limited breadth of protection. Increasing evidence indicates the importance of local immune responses and tissue resident memory T cells (TRM) for broad protection against respiratory infections, however, a vaccine that can be safely delivered to the lung is yet to be approved. In addition, there is little information on influenza A virus (IAV) specific TRM in the pig, a large animal model for IAV research, and a natural reservoir for the virus. The first part of this thesis describes the characterisation of porcine CD8 TRM in the lungs and upper respiratory tract following H1N1pmd09 infection. The phenotype of CD8 T cells specific for three influenza nucleoprotein (NP) epitopes, detected using MHC class I tetramers, was determined. This analysis revealed that the hierarchy of response differed with time in different tissues. The frequency of NP-specific cells declined over 63 days in all tissues, but it was best maintained in the lung. Most tetramer+ CD8 T cells isolated from broncho-alveolar lavage (BAL) displayed a TRM phenotype, expressing CD69 and an effector memory phenotype. NP-specific T cells isolated from BAL presented genes characteristic of human TRM, but gene expression differed at 7, 21 and 63 days post-infection. It is essential to assess vaccine deposition to provide optimal delivery of vaccines to the respiratory tract. The distribution of droplets generated by a commonly used intranasal device (MAD) and two commercially available vibrating mesh nebulizers (VMNs) was therefore compared. Droplets generated with the drug albuterol or a radiolabel (99mTc-DTPA), or a model vaccine (S-FLU) presented similar aerosol characteristics. A scintigraphic study using 99mTc-DTPA showed that VMNs uniformly distributed the radiolabel in the porcine lung and upper respiratory tract. The intranasal administration of a large volume (1ml/nostril) by MAD also delivered a high proportion of the dose to the lungs, although in a concentrated and restricted area. Recombinant single-cycle IAV viruses, such as S-FLU, may offer a promising solution to currently available vaccines as they can be delivered safely to the lung and elicit heterotypic immunity. However, previous studies in pigs showed a limited reduction in viral replication following challenge. In the last chapter, two strategies to improve S-FLU immunogenicity and protection are presented: 1) implementing a second-generation version of S-FLU expressing HA (CLEARFLU) and 2) administering S-FLU simultaneously by aerosol and intramuscular injection (SIM). CLEARFLU did not generate neutralising antibodies, however, S-FLU administered by SIM produced a high titer of neutralising antibody, robust local cellular responses and conferred protection against homologous challenge. On the other hand, intramuscular injection alone elicited humoral responses in the periphery, but failed to do so locally, and did not stimulate cellular responses. Aerosol S-FLU did not protect despite high T cell responses in the respiratory tract, probably due to the lack of neutralising antibody in the periphery. These data showed for the first time that porcine TRM present similar properties to those of humans, further establishing the pig as a useful model for investigation of local memory responses following respiratory infection. In addition, aerosol delivery can uniformly target the lungs but only the SIM immunisation regime successfully protected pigs from challenge, warranting further studies of this immunisation strategy against IAV and other respiratory pathogens.