Genetic basis of pathogenicity of influenza: 1. Genes in humans associated with more severe influenza 2. A new genetic virological factor in influenza viruses that modulates the host response.Archived

Genetic basis of pathogenicity of influenza: 1. Genes in humans associated with more severe influenza 2. A new genetic virological factor in influenza viruses that modulates the host response.

1. IFITM3 restricts the morbidity and mortality associated with influenzaEveritt AR, Clare S, Pertel T, et al.Nature (2012) doi:10.1038/nature10921; Published online 25 March 2012

The Role of Host Genetics in Susceptibility to Influenza: A Systematic Review. Horby P, Nguyen NY, Dunstan SJ, Baillie JK (2012) PLoS ONE 7(3):e33180.doi:10.1371/journal.pone.0033180; Published online 15 March 2012

2. An overlapping protein-coding region in influenza A virus segment 3 modulates the host responseJagger BW, Wise HM, Kash JC et al. Science (2012) 337 199 doi 10.1126/science.122213; Published online 28 June 2012

Although generally a mild infection, influenza can cause severe morbidity and mortality in a subset of the population, even in those without recognised risk factors. Little is known about any host/human genetic factors that may play a role. Equally the whole mechanism for how the viruses effect their pathogenicity and the genetics behind it is becoming increasingly complex. These two studies (the first accompanied by a systematic review) address these two genetic issues. respectively.

1. Genes in humans associated with more severe influenza

The first study concerned host factors. Everitt et al investigated whether interferon-inducible transmembrane (IFITM) protein 3, previously identified as a strong (and broad, non-selective) inhibitor of viral replication in vitro, has a role in intrinsic resistance to influenza A infection in vivo in mice and humans. This work combined three distinct approaches.

Studies in Mice

First, Ifitm 3 “knockout mice” (mice in which a specific gene has been inactivated, or "knocked out," replacing it or disrupting it with an artificial piece of DNA) were challenged with various low-pathogenicity influenza strains. The clinical, pathological and immunological aspects of the response were characterised and compared with wild-type littermates. Once infected with influenza the Ifitm-deficient mice (Ifitm3−/−) developed fulminant viral pneumonia with viral penetration deep into the lung tissue (but no systemic viraemia), severe signs of clinical illness and weight loss. In contrast, in the normal (Ifitm3+/+) mice, the virus was restricted mainly to the bronchioles, weight loss was less severe and the mice recovered fully. Examining the lung tissue of Ifitm3−/− mice, it was found that virus persisted and was not cleared as quickly as by the wild-type mice, resulting in a 10 times higher viral load. In lung tissue of infected wild-type mice, Ifitm3 expression was shown to be increasing over time. The Ifitm3−/− mice also had an imbalanced immune response: reduced proportions of CD4+ and CD8+ T-cells and NK cells, but an elevated proportion of neutrophils in the lungs; systemic lymphopenia; and enhanced inflammatory chemokine and cytokine (TNF-α, IL-6, G-CSF, MCP-1) production. Taken together, the signs that these Ifitm3−/− mice experienced were reminiscent of reactions expected to a much higher pathogenicity influenza. After challenge with such high pathogenicity influenza strain, knock-out mice lost weight faster and at lower infectious doses than wild-type. However, when expression of the viral interferon antagonist NS1 was blocked, infection was attenuated in both knock-out and wild-type mice, indicating that Ifitm3−/− mice can mount an interferon response. This suggests that Ifitm3 blocks viral replication occurs before NS1 mediated interferon antagonism, and that the profound damage in infected Ifitm3−/− mice is due to unchecked viral replication in the lungs combined with an exaggerated inflammatory response.

Studies in Humans

In order to find out whether IFITM3 also plays a role in susceptibility to influenza in humans, the authors sequenced 1.8 kilobases of the IFITM3 locus from 53 severely affected hospitalised influenza cases from the Mechanisms of Severe Acute Influenza Consortium (MOSAIC) study, and looked at single nucleotide polymorphisms (SNPs) within this population. In particular, one SNP rs12252, resulting in a truncated IFITM3 protein missing the first 21 amino acids, was three times as common in the hospitalised cases, compared to ethnically matched Europeans from the 1000 Genomes project. Patients’ genotypes also departed from Hardy–Weinberg equilibrium (the probability of the divergence happening by chance was p = 0.003), showing an excess of the minority allele, encoding for truncated IFITM3, in this population. Principal components analysis showed no evidence of hidden population structure differences between controls and a subset of the cases. The study also compared allele frequencies between the various populations from the 1000 Genomes project: the allele for truncated IFITM3, though ancestral, was rare in people of African or European decent, but appeared to be more frequent in other populations.

Cell-line Studies

The role of murine and human Ifitm3 in fighting influenza infection was confirmed by in vitro infection studies in various cell lines. These experiments showed that for different cell lines and influenza strains, cells expressing truncated Ifitm3, or no such protein at all, were more readily infected after challenge than cells expressing full-length Ifitm3. In murine embryonic fibroblast cell line, the Ifitm3−/− genotype was rescued completely by stable expression of Ifitm3 from a vector. In lymphoblastoid cell lines, homozygosity for the minority human Ifitm3 allele, coding for the truncated protein, correlated with overall lower levels of Ifitm3 protein expression.

2. A new virological factor in influenza viruses that modulates the host response

The second study looked at a specific influenza virus factor using a combination of genomic, biochemical, functional and in-vivo studies of both ordinary A(H1N1) human influenza and the more pathogenic reconstructed 1918 A(H1N1). In the paper the researchers report how segment 3 of the virus contains a second open reading frame termed “X-ORF” which operates through ribosomal frame-shifting, a known genetic mechanism but not previously described in influenza viruses. The frameshift protein product, termed PA-X, comprises the endonuclease domain of the viral PA protein with a C-terminal domain encoded by the X-ORF. It functions to repress cellular gene expression in the host and PA-X also modulates influenza A virulence in a mouse infection model, acting to decrease pathogenicity. When the PA-X was not produced there were changes in the kinetics of the global host response, which notably includes increases in inflammatory, apoptotic, and T lymphocyte–signaling pathways. Thus this suggests that X-ORF is another important mechanism through which the influenza viruses evade the host response to allow their successful replication.

ECDC Comment (17 August 2012):

It has been recognised for many years that human genetic composition can influence the susceptibility of humans to infectious disease with the prime example being sickle cell trait and severe malaria where some of the mechanisms are now being established.(1) Other more recently recognised examples are with West Nile Virus and also HIV transmission to and disease progression in children.(2,3) Knowledge of host genetics in susceptibility to influenza is well reviewed in a recent systematic review which relies mostly on the extensive knowledge of mouse genetics and influenza and studies of unusual human familial susceptibility to influenza.(4) On the virological side there is much more complexity and knowledge about how viruses replicate and evade the human immunological defences. Of late particular interest has arisen over a non-structural protein, NS1 and its ability to inhibit host immune responses, especially the limitation of both interferon (IFN) production and the antiviral effects of interferon induced proteins.(5) In humans there are also observations of particular genetic traits in viruses that are associated with more severe disease.(6) However, where these non-experimental studies must be interpreted cautiously is that it is unclear whether the particular association is through the genetic difference or whether the association is some epi-phenomenon due to both the genetic trait and the high pathogenicity is due to some other truly causative factor that they both happen to be associated with.(6)What are the public health and clinical benefit of studies like these Presently we are a long way off from much public health benefit. For example there would have to be a strong association between severe disease and particular human genetic traits, and/or those traits would need to account for a major population accountable risk before screening the population for those genes would be contemplated. Even then there would be many ethical dilemmas that arose.(7) There may be more clinical benefit in the short term if it was possible to identify on admission to hospital those individuals who were more likely to progress to the most severe disease. That is why studies like (MOSAIC) when combined with the like of the 1000 Genomes project for comparisons can be useful for clinical purposes. Though a problem with MOSAIC is that its data are almost entirely derived from the 2009 pandemic (influenza A(H1N1)pdm09) and its impossible to predict whether its findings will extend to the more diverse seasonal influenza viruses.(8) The studies of the mechanisms for how influenza evades the human body immune systems and the viral genetic basis for this are especially important if they suggest new approaches for drug design since the number of drug types licensed in Europe is confined to one group, the neuraminidases.