|dc.description.abstract||Humans are subject to infection with a wide range of commensal and pathogenic organisms.
Each pathogen requires an appropriate immune response to eliminate or control the invading
organism and minimise pathology. Many pathogens have evolved strategies to subvert or
manipulate the immune response and establish on-going infections. Similarly acute respiratory
infection with virulent strains of influenza A virus are often poorly controlled by the immune
system and can cause severe immunopathology and even fatality as a result of an inappropriate
and excessive inflammatory response called a ‘cytokine storm’.
Morbidity due to influenza infection and exacerbation by the immune response can vary greatly
between individuals. The effect of underlying infection on the immune system could contribute
to the variation in response. The aim of this project was therefore to determine if co-infection
with two pathogens that establish on-going infections could alter the immune response to
influenza A and impact the outcome of infection.
Persistent infections with filarial helminths can cause debilitating disease and significantly
impact the immune response toward a skewed TH2 or regulatory phenotype in order to control
pathology. In contrast, infection with gammaherpesviruses in an immunocompetent host causes
an initial inflammatory ‘anti-viral’ response before becoming an asymptomatic, latent infection.
In an immunocompromised host, gammaherpesviruses can reactivate and lead to clinical
presentation of disease. This suggests that these viruses require an on-going immune response to
control all stages of infection. Both filarial helminths and gammaherpesviruses are common
infections in human populations and therefore mouse models of these infections provide
relevant systems to study their potential role in influenza virus infections.
In a BALB/c murine co-infection model, latent infection with the rodent gammaherpesvirus
MHV-68 led to significantly decreased weight loss and clinical signs following high dose
infection with A/WSN/33, (a H1N1 influenza A virus). This was coupled with decreased
immunopathology in the lung and fewer infiltrating lymphocytes in the alveolar spaces and
around larger airways, although infectious virus titres were not significantly reduced. This
response was coupled with a decreased production of inflammatory cytokines and chemokines
in co-infected mice 6 days post infection which correlated with the amelioration of pathogenesis
in these animals.
A repeat of the study in 129Sv/Ev IFNγR knock out mice showed the same protective effect in
the co-infected mice, suggesting IFNγ is not critical for the protective phenotype. Mice infected
with latent MHV-68 alone showed a significant increase in expression of T cell chemokines in
the lung and alveolar macrophages had a significantly increased production of suppressor of
cytokine signalling (SOCS-1) suggesting latent MHV-68 infection may impact the phenotype of
macrophages in the lung, modulating the response to influenza co-infection.
A co-infection model with a persistent rodent filarial helminth, Litomosoides sigmodontis and
A/WSN/33 was also established in BALB/c mice. The L4 developmental stage of L.
sigmodontis infection had no impact on co-infection with A/WSN/33. Adult stage worms,
however, appeared to have a protective effect against A/WSN/33 pathogenesis. Co-infected
mice had significantly delayed weight loss and clinical signs 3-5 days post infection. CD4+ and
CD8+ T cells in the lung draining lymph nodes had significantly reduced TH1 and TH2
phenotypes (measured by cytokine production) compared with singly infected controls. IFNγ
secreting CD4+ T cells in the lungs of co-infected mice also secreted increased levels of IL-10,
suggesting an increase in regulation of the inflammatory response to A/WSN/33.
At the full patent stage of L. sigmodontis infection, co-infection with A/WSN/33 led to
increased clinical signs and significantly exacerbated weight loss. CD4+ and CD8+ T cells in the
lung draining lymph nodes were inflammatory in L. sigmodontis infected mice alone as well as
co-infected mice and there were no differences in the percentage of CD4+ T cells in the lung
secreting IL-10 and IFNγ between co-infected and influenza infected mice. A loss in regulatory
responses during the patent stage of L. sigmodontis infection may therefore contribute to the
loss of protection against A/WSN/33 at this time point within the co-infection model.
Understanding the impact of an underlying infection on the immune system could provide
immune mechanisms that could be exploited to increase vaccine efficacy against influenza and
similarly help to provide better treatment for individuals infected with influenza A. These
results may also help predict the outcome of influenza A infection in individuals already
infected with highly immunogenic, on-going infections.||en