Pathogenesis of influenza: virus–host interactions

Abstract

Since their inception in March 1972, Keystone Symposia on Molecular and Cellular Biology have brought together scientists from across the globe to discuss key biological topics. Now in its 40th year, it is a completely independent, nonprofit organization devoted solely to providing outstanding scientific conferences in all areas of the biological and biomedical sciences. Towards the end of May 2011, over 200 virologists and immunologists came to Hong Kong, an appropriate setting given the emergence of H5N1, to discuss influenza virus and host interactions. The meeting, expertly organized by Siamon Gordon (University of Oxford, Oxofrd, UK), Malik Peiris (University of Hong Kong, Hong Kong, China) and Kanta Subbarao (NIAID, NIH, MD, USA), took place in the aftermath of the first pandemic in 40 years and provided great insight into both pandemic H1N1 and H5N1. This article focuses on some of the recurring themes that were discussed during the week.

Hemagglutinin, antigenic drift & viral tropism

The interaction between influenza hemagglutinin (HA), the antigenic glycoproteins that allow influenza to bind and infect cells, and host sialic acid (Sia)-linked glycoproteins was one of the key themes running throughout the symposium. In light of the now numerous solved HA structures, Sir John Skehel (MRC National Institute of Medical Research, London, UK) gave the opening keynote lecture on the differences in both structure and function between the HAs. After showing that the 16 known HA subtypes can be phylogenetically divided into two distinct groups, he went on to demonstrate that these groups exhibit differences in their membrane binding capabilities. He explained how, in order to mediate membrane fusion, the HA undergoes a pH dependent conformational change resulting in the insertion of the membrane fusion protein into the target membrane, adjacent to the membrane anchor region. Skehel explained how group 2 HAs (including H3 and H7) undergo this pH transition much faster than group 1 HAs, (which include H5 and H1) and emphasized how an understanding of detailed structural analysis may facilitate the development of cross-neutralizing antibodies [1].

Later in the week, the focus remained on HA, with the role and mechanism of antigenic drift being addressed by Jack Bennick (NIAID, NIH, MD, USA). He underlined the important role of receptor binding avidity in driving HA antigenic drift. Bennick further explained that although altering HA glycosylation dramatically changes antigenicity, this can be at the expense of receptor binding avidity, explaining why HA glycoslylation is not more regularly utilized as a means of immune escape [2]. He also stressed the close association between HA and neuraminidase (NA) evolution, showing that antibody driven HA escape mutants often had corresponding, and perhaps compensatory, alterations in NA [3].

The distribution of Sia on cell surfaces is one of the key determinants of host tropism and it has long been assumed that avian influenza viruses preferentially bind Sia2–3Gal and human viruses bind Sia2–6Gal. John Nicholls (University of Hong Kong, Hong Kong, China) discussed how the advent of lectin histochemistry, sophisticated glycan arrays and his own mass spectrometric data challenges this oversimplified paradigm. These techniques show that the human respiratory tract in fact contains a heterogeneous mixture of complex Sia structures, including both Sia2–6Gal and Sia2–3Gal, and that virus–receptor interaction is not simply dependent on the α2–3/α2–6 link, but also on O- and N- linked glycosylation and fucosylation [4]. Nicholls also demonstrated that it was possible to use recombinant sialidase fusion proteins, such as Das-181 (NexBio – currently in Phase II trials) to disrupt this viral–host interaction, by removing the Sia residues from host cells, and suggested that topical sialidase treatment may offer a valid therapeutic option [5].

Ecology & pathogenesis

Despite the ever advancing understanding of the factors that govern virulence, transmission and pathogenesis, the emergence of novel influenza viruses with pandemic potential continues to be unpredictable. Throughout the symposium the origins and evolution of both the recent 2009 H1N1 pandemic and the 1997 H5N1 outbreak was discussed.

Ron Fouchier (Erasmus Medical Centre, Rotterdam, The Netherlands) delivered an overview of the determinants of influenza virulence in mammals and presented data on the mechanisms for acquisition of H5N1 transmission potential. He commenced his talk by reminding the delegates of the devastating effects that result from the transmission of H5 or H7 low-pathogenic avian influenza viruses to poultry from wild birds with mutation into highly pathogenic avian influenza viruses through the acquisition of multiple basic amino acid residues at the HA cleavage site. Fouchier cited the H7N7 outbreak that occurred in The Netherlands in 2003, resulting in the culling of 30 million chickens and infection of 89 people. Of those 89 people, one developed severe pneumonia and acute respiratory distress syndrome (ARDS), and by sequence analysis, his group identified virulence determinants in PB2, PA, HA, NA and NS1 compared with the milder strain [6]. Following this he emphasized the importance of route of inoculation for primary disease presentation; his ferret data suggest that A/H5N1 spreads directly from the nasal cavity to the brain, and that CNS lesions contribute more than pulmonary lesions to the pathogenicity of H5N1 in ferrets [7].

Since H5N1 first emerged in 1996 it has spread to 15 countries, infecting over 550 people and killing more than 320. In the first part of his lecture, Yoshihiro Kawaoka (University of Wisconsin, WI, USA and University of Tokyo, Tokyo, Japan) concentrated on three H5N1 cases complicated by CNS involvement, asking the question: ‘Are long-term neurological sequelae seen in H5N1 survivors?’ In ferrets, he showed evidence of encephalitis and brain hemorrhage up to 6 months post-infection. He went on to present elegant cross sectional images of the brain, mapping the route of entry of H5N1, suggesting that it may invade the CNS primarily via the olfactory route in addition to via the vasculature [8]. For the latter half of his talk, he addressed the question of how H5N1 viruses reach the lower respiratory tract, postulating whether humans can be infected by eating virus-infected poultry. Using a novel inoculation method in ferrets, involving ingestion of petroleum-coated gelatin capsules containing homogenates of virus-infected chicken embryos, he went on to demonstrate how an ingested virus can disseminate to the nondigestive organs, possibly through the lymphatic system of the GI tract [9].

Immunology & interventions

The role of innate immunity in the pathogenesis of influenza was a recurring theme throughout the symposium and Tracy Hussell’s group (Imperial College London, London, UK) presented data suggesting that influenza infection causes long-term remodeling of lung immune homeostasis, resetting the ‘innate immune rheostat’, thereby increasing susceptibility to secondary bacterial pneumonia, one of the major causes of mortality in pandemic influenza. Using murine models they have shown that the alveolar microenvironment is kept in a constant state of immune equilibrium and that, even in the absence of stimulatory antigen, inflammation can result if negative inflammatory regulators, such as CD200R and TREM2, are absent. They presented data showing that CD200 receptor is expressed on the surface of alveolar macrophages and receives a unidirectional negative regulatory signal when bound to CD200 ligand expressed on the alveolar epithelium. When influenza damages the epithelium, this suppressive signal is lost, CD200R is upregulated, and alveolar macrophages are allowed to perform their defensive role. However, on resolution of the infection CD200R levels remain high and the immune regulatory mechanisms within the alveoli appear to ‘overshoot’, leading to a state of relative immune suppression and, thus, increased susceptibility to secondary bacterial infection [10]. They proposed that a greater understanding of the interplay between alveolar macrophages and the epithelium will no doubt have important implications for novel antiviral treatments.

On a similar theme, Stephan Ludwig (Institute of Molecular Virology, Muenster, Germany) emphasized the urgent need for novel anti-inflammatory agents in light of H5N1 and expanding drug resistance. He elegantly described how shifting the focus of therapy from targeting viral components to host signal transduction pathways may pave the way for the development of new treatments. Supporting this position, he went on to explain how influenza viruses hijack host cellular signaling pathways, including IKK/NFκB, to promote efficient replication and how the novel NFκB inhibitor, SC75741, may exploit this. He presented data derived from murine models demonstrating how SC75741 blocks the replication of both influenza A and B, in addition to reducing clinical symptoms and improving survival in SC75741 treated mice. Ludwig also demonstrated how SC75741 suppresses the ‘cytokine storm’ induced by H5N1 in the lung without showing any tendency to induce resistant virus variants [11].

Tao Dong (MRC Human Immunology Unit, University of Oxford, Oxford, UK), showed that memory T cells established by seasonal human influenza A infection can cross-react with avian influenza A (H5N1) in healthy individuals, and highlighted the potential for designing a universal T-cell vaccine. They also presented data showing evidence of underestimated asymptomatic H5N1 infection in populations at high risk of H5N1 exposure [12].

Despite the many advances in our understanding of the pathogenesis of influenza shown in this symposium and the development of novel therapies, the humbling fact that our therapeutic options remain limited when influenza causes diffuse alveolar damage and the ARDS was highlighted by the intensivist, Arthur Slutsky (University of Toronto, Toronto, Canada). ARDS and acute lung injury (a variant of ARDS with less severe hypoxemia) caused by 2009 H1N1 pandemic put considerable strain on the intensive care community around the world and Slutsky emphasized the fact that all pharmaceutical trials to date in this area have been negative. He pointed the audience to the literature from the late 1990s on the ‘biotrauma’ caused by overzealous ventilation to underline the importance of maintaining low tidal volumes and encouraged the delegates to consider including the effects of ventilation when assessing virus–host interactions [13].

This Keystone symposia event was 18 months in the making and succeeded in providing a superb, comprehensive update on the pathogenesis of influenza.

Financial & competing interests disclosure

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.