Pig defences against respiratory virusesM. Bailey, M.A. Birchall, K. Haverson, K. Gorti and S. Wilson
Department of Clinical Veterinary Sciences, University of Bristol, Langford House, Langford, Bristol BS40 5DU, United Kingdom
Abstract - Mucosal surfaces are exposed both to pathogens and to a wide range of environmental antigens which pose no obvious threat. Recent evidence clearly demonstrates that antigen-specific immune mechanisms are not necessary for controlling most commensal organisms. Indeed, expression of active immune responses to commensal flora at mucosal surfaces is associated with damaging inflammatory responses. Consistent with this is the observation that the normal intestinal mucosa is committed to expression of Th2-type immune responses (primarily antibody), rather than the more damaging Th1-type (primarily macrophages and cytotoxic T cells). Indeed, there is compelling evidence that the default immune response at mucosal surfaces is to generate immunological tolerance. In the respiratory tract, non-replicating antigens given intranasally have been used to generate both active immune response and tolerance. It is clear that the system is a dynamic immunological tissue with the potential for active immune response and for tolerance induction: however, the extent to which the normal respiratory mucosa is committed to Th1 or Th2 responses is unclear. Recognition of viral proteins by the immune system takes place in specialised inductive sites and is followed by proliferation, differentiation and spread of effector and memory cells. On relocalisation to the mucosa, these cells must then recognise antigen locally in order to express their effector function. The respiratory tract of the pig includes a number of distinct immunological architectures in which these sequential interactions with viral infections may take place. The large palatine, lingual and nasal tonsils are major sites of B-cell proliferation and immunoglobulin secretion. By analogy with Peyers patches, these tissues are probably sites of antigen sampling and induction of immune responses. However, the observation that several pathogenic organisms can persist within the tonsils of humans and pigs (e.g. Streptococcus suis) and that active, cellular immune response in these organised sites can be harmful, suggests that they may also be involved in maintaining functional tolerance, or at least a predominantly Th2/antibody response. The tonsils are not the only sites of antigen sampling in the upper respiratory tract: small, isolated, organised B cell follicles, flanked by T cell areas, also occur within the mucosa of the upper respiratory tract of the pig and are also likely to be capable of primary, inductive responses. The mucosa of the upper respiratory tract also contains a heavy component of dendritic cells, a cell type absolutely required for induction of primary immune responses. These dendritic cells lie at the dermal/epidermal junction and are closely associated with CD3 + T cells, presumably of memory phenotype. In other sites, comparable dendritic cells acquire antigens from epithelial cells and transport them to the lymph nodes, where presentation to T cells can occur. By analogy, initial presentation of viral antigens derived from infection of upper respiratory epithelial surfaces is likely to occur in the retropharyngeal and cervical lymph nodes and may generate mixed mucosal and systemic immune responses. The importance of antigen sampling and inductive mechanisms is paralleled by a heavy presence of effector mechanisms in the upper respiratory mucosa. Plasma cells secreting IgA contribute to classical mucosal defence, but the mucosa also contains a large, resident T cell component, presumably engaged in surveillance and the provision of rapid responses to recall antigen, although studies in the intestine have suggested that similar cells at that site may be heavily regulated. The immunological architecture of the lower respiratory tract presents several distinct features. Organised lymphoid structures comparable to the tonsils, Peyers patches or isolated follicles do not occur in the normal pig, although unstructured lymphoid form in response to antigenic challenge (Pabst, 1996). Studies in normal rats have also shown dendritic cells underneath the epithelial basement membrane which rapidly mobilise to local lymph nodes on exposure to luminal antigen. The interstitium of the lower respiratory tract does contain resident T cells with memory phenotype which may provide rapid local immune responses to previously experienced antigen. Recent studies have also demonstrated the presence of a large, intra-vascular pool of lymphocytes and macrophages in the lungs of pigs. Both alveolar and intra-vascular macrophages are effective at phagocytosis, intracellular and secretion of proinflammatory cytokines. These populations of macrophages and T cells within the pulmonary capillaries is likely to provide a rapidly recruitable source of effector cells in the event of viral infection. However, their ability to secrete cytokines such as TNF presents a potential problem: in models of septic shock, TNF is associated with recruitment of neutrophils into the lung, progressive tissue damage and respiratory collapse. The respiratory tract contains a large immunological component which is potentially capable of rapid, local responses to invading micro-organisms. However, this ability to mount massive and potentially damaging immune responses must be carefully controlled to avoid inappropriate reactions to microbial antigens. Vaccine strategies will need to determine whether targeting tonsillar or lymph node-based responses produce the most appropriate protection against specific viruses.
Corresponding author: M. Bailey Tel.: (44) 117 9289262; fax: (44) 117 9289505;
© INRA, EDP Sciences 2000