Matthias Liniger

Chicken cells sense influenza A virus infection through MDA5 and CARDIF-signaling involving LGP2

ABSTRACT Avian influenza viruses (AIV) raise worldwide veterinary and public health concerns due to their potential for zoonotic transmission. While infection with highly pathogenic AIV results in high mortality in chickens, this is not necessarily the case in wild birds and ducks. It is known that innate immune factors can contribute to the outcome of infection. In this context, retinoic acid-inducible gene I (RIG-I) is the main cytosolic pattern recognition receptor known for detecting influenza A virus infection in mammalian cells. Chickens, unlike ducks, lack RIG-I, yet chicken cells do produce type I interferon (IFN) in response to AIV infection. Consequently, we sought to identify the cytosolic recognition elements in chicken cells. Chicken mRNA encoding the putative chicken analogs of CARDIF and LGP2 (chCARDIF and chLGP2, respectively) were identified. HT7-tagged chCARDIF was observed to associate with mitochondria in chicken DF-1 fibroblasts. The exogenous expression of chCARDIF, as well as of the caspase activation and recruitment domains (CARDs) of the chicken melanoma differentiation-associated protein 5 (chMDA5), strongly activated the chicken IFN-β (chIFN-β) promoter. The silencing of chMDA5, chCARDIF, and chIRF3 reduced chIFN-β levels induced by AIV, indicating their involvement in AIV sensing. As with mammalian cells, chLGP2 had opposing effects. While overexpression decreased the activation of the chIFN-β promoter, the silencing of endogenous chLGP2 reduced chIFN-β induced by AIV. We finally demonstrate that the chMDA5 signaling pathway is inhibited by the viral nonstructural protein 1. In conclusion, chicken cells, including DF-1 fibroblasts and HD-11 macrophage-like cells, employ chMDA5 for sensing AIV.

Hemagglutinin-Dependent Tropism of H5N1 Avian Influenza Virus for Human Endothelial Cells

ABSTRACT Although current H5N1 highly pathogenic avian influenza viruses (HPAIV) are inefficiently transmitted to humans, infected individuals can suffer from severe disease, often progressing rapidly to acute respiratory distress syndrome and multiorgan failure. This is in contrast with the situation with human influenza viruses, which in immunocompetent individuals usually cause only a respiratory disease which is less aggressive than that observed with avian H5N1 viruses. While the biological basis of inefficient transmission is well documented, the mechanisms by which the H5N1 viruses cause fatal disease remain unclear. In the present study, we demonstrate that human pulmonary microvascular endothelial cells (hPMEC) had a clearly higher susceptibility to infection by H5N1 HPAIV than to infection by human influenza viruses. This was measurable by de novo intracellular nucleoprotein production and virus replication. It was also related to a relatively higher binding capacity to cellular receptors. After infection of hPMEC, cell activation markers E-selectin and P-selectin were upregulated, and the proinflammatory cytokines interleukin-6 and beta interferon were secreted. H5N1 virus infection was also associated with an elevated rate of cell death. Reverse genetics analyses demonstrated a major role for the viral hemagglutinin in this cell tropism. Overall, avian H5N1 viruses have a particular receptor specificity targeting endothelial cells that is different from human influenza viruses, and this H5N1 receptor specificity could contribute to disease pathogenesis.

Overlapping sense and antisense transcription units in Trypanosoma brucei

Procyclins are the major surface glycoproteins of insect‐form Trypanosoma brucei. The procyclin expression sites are polycistronic and are transcribed by an α‐amanitin‐resistant polymerase, probably RNA polymerase I (Pol I). The expression sites are flanked by transcription units that are sensitive to α‐amanitin, which is a hallmark of Pol II‐driven transcription. We have analysed a region of 9.5 kb connecting the EP/PAG2 expression site with the downstream transcription unit. The procyclin expression site is longer than was previously realized and contains an additional gene, procyclin‐associated gene 4 (PAG4), and a region of unknown function, the T region, that gives rise to trans‐spliced, polyadenylated RNAs containing small open reading frames (ORFs). Two new genes, GU1 and GU2, were identified in the downstream transcription unit on the opposite strand. Unexpectedly, the 3′ untranslated region of GU2 and the complementary T transcripts overlap by several hundred base pairs. Replacement of GU2 by a unique tag confirmed that sense and antisense transcription occurred from a single chromosomal locus. Overlapping transcription is stage specific and may extend ≥ 10 kb in insect‐form trypanosomes. The nucleotide composition of the T. brucei genome is such that antisense ORFs occur frequently. If stable mRNAs can be derived from both strands, the coding potential of the genome may be substantially larger than has previously been suspected.

Highly pathogenic avian influenza virus H5N1 controls type I IFN induction in chicken macrophage HD-11 cells: a polygenic trait that involves NS1 and the polymerase complex

BackgroundInfluenza A viruses are well characterized to antagonize type I IFN induction in infected mammalian cells. However, limited information is available for avian cells. It was hypothesised that avian influenza viruses (AIV) with distinct virulence may interact differently with the avian innate immune system. Therefore, the type I IFN responses induced by highly virulent and low virulent H5N1 AIV and reassortants thereof were analysed in chicken cells.ResultsThe highly pathogenic (HP) AIV A/chicken/Yamaguchi/7/04 (H5N1) (Yama) did not induce type I IFN in infected chicken HD-11 macrophage-like cells. This contrasted with an NS1 mutant Yama virus (Yama-NS1A144V) and with the attenuated H5N1 AIV A/duck/Hokkaido/Vac-1/04 (Vac) carrying the haemagglutinin (HA) of the Yama virus (Vac-Yama/HA), that both induced type I IFN in these cells. The substitution of the NS segment from Yama with that from Vac in the Yama backbone resulted in induction of type I IFN secretion in HD-11 cells. However, vice versa, the Yama NS segment did not prevent type I IFN induction by the Vac-Yama/HA virus. This was different with the PB1/PB2/PA segment reassortant Yama and Vac-Yama/HA viruses. Whereas the Yama virus with the Vac PB1/PB2/PA segments induced type I IFN in HD-11 cells, the Vac-Yama/HA virus with the Yama PB1/PB2/PA segments did not. As reported for mammalian cells, the expression of H5N1 PB2 inhibited the activation of the IFN-β promoter in chicken DF-1 fibroblast cells. Importantly, the Yama PB2 was more potent at inhibiting the IFN-β promoter than the Vac PB2.ConclusionsThe present study demonstrates that the NS1 protein and the polymerase complex of the HPAIV Yama act in concert to antagonize chicken type I IFN secretion in HD-11 cells. PB2 alone can also exert a partial inhibitory effect on type I IFN induction. In conclusion, the control of type I IFN induction by H5N1 HPAIV represents a complex phenotype that involves a particular viral gene constellation rather than a single viral protein. Collectively, these findings contribute to understand the high virulence of HPAIV H5N1 viruses observed in the chicken host.

Cleavage of trypanosome surface glycoproteins by alkaline trypsin-like enzyme(s) in the midgut of Glossina morsitans

EP and GPEET procyclin, the major surface glycoproteins of procyclic forms of Trypanosoma brucei, are truncated by proteases in the midgut of the tsetse fly Glossina morsitans morsitans. We show that soluble extracts from the midguts of teneral flies contain trypsin-like enzymes that cleave the N-terminal domains from living culture-derived parasites. The same extract shows little activity against a variant surface glycoprotein on living bloodstream form T. brucei (MITat 1.2) and none against glutamic acid/alanine-rich protein, a major surface glycoprotein of Trypanosoma congolense insect forms although both these proteins contain potential trypsin cleavage sites. Gel filtration of tsetse midgut extract revealed three peaks of tryptic activity against procyclins. Trypsin alone would be sufficient to account for the cleavage of GPEET at a single arginine residue in the fly. In contrast, the processing of EP at multiple sites would require additional enzymes that might only be induced or activated during feeding or infection. Unexpectedly, the pH optima for both the procyclin cleavage reaction and digestion of the trypsin-specific synthetic substrate Chromozym-TRY were extremely alkaline (pH 10). Direct measurements were made of the pH within different compartments of the tsetse digestive tract. We conclude that the gut pH of teneral flies, from the proventriculus to the hindgut, is alkaline, in contradiction to previous measurements indicating that it was mildly acidic. When tsetse flies were analysed 48 h after their first bloodmeal, a pH gradient from the proventriculus (pH 10.6+/-0.6) to the posterior midgut (pH 7.9+/-0.4) was observed.

High interferon type I responses in the lung, plasma and spleen during highly pathogenic H5N1 infection of chicken.

This study shows that high pathogenic H5N1 influenza virus infection of chicken induced high levels of bioactive interferon type I in the lung (4.3 × 105 U/mg tissue), plasma (1.1 × 105 U/mL), and spleen (9.1 × 105 U/mg tissue). In contrast, a low pathogenic attenuated H5N1 vaccine strain only induced approximately 24 times less IFN in the lung, 441 times less in the spleen and 649 less in the plasma. This was in the same range as a reassortant carrying the HA from the vaccine strain and the remaining genes from the high pathogenic virus. On the other hand, a reassortant virus with the HA from the high pathogenic H5N1 with the remaining genes from the vaccine strain had intermediate levels of IFN. The level of interferon responses related to the viral load, and those in the spleen and blood to the spread of virus to lymphoid tissue, as well as disease severity. In vitro, the viruses did not induce interferon in chicken embryonic fibroblasts, but high levels in splenocytes, with not clear relationship to pathogenicity and virulence. This, and the responses also with inactivated viruses imply the presence of plasmacytoid dendritic cell-like leukocytes within the chicken immune system, possibly responsible for the high interferon responses during H5N1 infection. Our data also indicate that the viral load as well as the cleavability of the HA enabling systemic spread of the virus are two major factors controlling systemic IFN responses in chicken.

MDA5 Can Be Exploited as Efficacious Genetic Adjuvant for DNA Vaccination against Lethal H5N1 Influenza Virus Infection in Chickens

Chickens lack the retinoic acid-inducible gene I (RIG-I) and sense avian influenza virus (AIV) infections by means of the melanoma differentiation-associated gene 5 product (chMDA5). Plasmid-driven expression of the N-terminal half of chMDA5 containing the caspase activation and recruitment domains [chMDA5(1-483)] triggers interferon-β responses in chicken cells. We hypothesized that mimicking virus infection by chMDA5(1-483) expression may enhance vaccine-induced adaptive immunity. In order to test this, the potential genetic adjuvant properties of chMDA5(1-483) were evaluated in vivo in combination with a suboptimal quantity of a plasmid DNA vaccine expressing haemagglutinin (HA) of H5N1 AIV. Co-administration of the HA plasmid with plasmid DNA for chMDA5(1-483) expression resulted in approximately 10-fold higher HA-specific antibody responses than injection of the HA plasmid mixed with empty vector DNA as control. Accordingly, compared with HA DNA vaccination alone, the chMDA5(1-483)-adjuvanted HA DNA vaccine mediated enhanced protection against a lethal H5N1 challenge infection in chickens, with reduced clinical signs and cloacal virus shedding. These data demonstrate that innate immune activation by expression of signaling domains of RIG-I-like receptors can be exploited to enhance vaccine efficacy.

Efficient Sensing of Avian Influenza Viruses by Porcine Plasmacytoid Dendritic Cells

H5N1 influenza A virus (IAV) infections in human remain rare events but have been associated with severe disease and a higher mortality rate compared to infections with seasonal strains. An excessive release of pro-inflammatory cytokine together with a greater virus dissemination potential have been proposed to explain the high virulence observed in human and other mammalian and avian species. Among the cells involved in the cytokine storm, plasmacytoid dendritic cells (pDC) could play an important role considering their unique capacity to secrete massive amounts of type I interferon (IFN). Considering the role of IFN as a major component of antiviral responses as well as in priming inflammatory responses, we aimed to characterize the induction of IFN-α release upon infection with IAV originating from various avian and mammalian species in a comparative way. In our porcine pDC model, we showed that the viral components triggering IFN responses related to the ability to hemagglutinate, although virosomes devoid of viral RNA were non-stimulatory. Heat-treatment at 65 °C but not chemical inactivation destroyed the ability of IAV to stimulate pDC. All IAV tested induced IFN-α but at different levels and showed different dose-dependencies. H5 and H7 subtypes, in particular H5N1, stimulated pDC at lower doses when compared to mammalian IAV. At high viral doses, IFN-α levels reached by some mammalian IAV surpassed those induced by avian isolates. Although sialic acid-dependent entry was demonstrated, the α-2,3 or α-2,6 binding specificity alone did not explain the differences observed. Furthermore, we were unable to identify a clear role of the hemagglutinin, as the IFN-α doses-response profiles did not clearly differ when viruses with all genes of identical avian origin but different HA were compared. This was found with IAV bearing an HA derived from either a low, a high pathogenic H5N1, or a human H3. Stimulation of pDC was associated with pDC depletion within the cultures. Taken together and considering the efficient sensing of H5N1 at low dose, pDC on one side may play a role in the cytokine storm observed during severe disease, on the other hand could participate in early antiviral responses limiting virus replication.

Intracellular membrane association of the N-terminal domain of classical swine fever virus NS4B determines viral genome replication and virulence.

Classical swine fever virus (CSFV) causes a highly contagious disease in pigs that can range from a severe haemorrhagic fever to a nearly unapparent disease, depending on the virulence of the virus strain. Little is known about the viral molecular determinants of CSFV virulence. The nonstructural protein NS4B is essential for viral replication. However, the roles of CSFV NS4B in viral genome replication and pathogenesis have not yet been elucidated. NS4B of the GPE-  vaccine strain and of the highly virulent Eystrup strain differ by a total of seven amino acid residues, two of which are located in the predicted trans-membrane domains of NS4B and were described previously to relate to virulence, and five residues clustering in the N-terminal part. In the present study, we examined the potential role of these five amino acids in modulating genome replication and determining pathogenicity in pigs. A chimeric low virulent GPE- -derived virus carrying the complete Eystrup NS4B showed enhanced pathogenicity in pigs. The in vitro replication efficiency of the NS4B chimeric GPE-  replicon was significantly higher than that of the replicon carrying only the two Eystrup-specific amino acids in NS4B. In silico and in vitro data suggest that the N-terminal part of NS4B forms an amphipathic α-helix structure. The N-terminal NS4B with these five amino acid residues is associated with the intracellular membranes. Taken together, this is the first gain-of-function study showing that the N-terminal domain of NS4B can determine CSFV genome replication in cell culture and viral pathogenicity in pigs.

A Japanese Encephalitis Virus Vaccine Inducing Antibodies Strongly Enhancing In Vitro Infection Is Protective in Pigs

The Japanese encephalitis virus (JEV) is responsible for zoonotic severe viral encephalitis transmitted by Culex mosquitoes. Although birds are reservoirs, pigs play a role as amplifying hosts, and are affected in particular through reproductive failure. Here, we show that a lentiviral JEV vector, expressing JEV prM and E proteins (TRIP/JEV.prME), but not JEV infection induces strong antibody-dependent enhancement (ADE) activities for infection of macrophages. Such antibodies strongly promoted infection via Fc receptors. ADE was found at both neutralizing and non-neutralizing serum dilutions. Nevertheless, in vivo JEV challenge of pigs demonstrated comparable protection induced by the TRIP/JEV.prME vaccine or heterologous JEV infection. Thus, either ADE antibodies cause no harm in the presence of neutralizing antibodies or may even have protective effects in vivo in pigs. Additionally, we found that both pre-infected and vaccinated pigs were not fully protected as low levels of viral RNA were found in lymphoid and nervous system tissue in some animals. Strikingly, the virus from the pre-infection persisted in the tonsils throughout the experiment. Finally, despite the vaccination challenge, viral RNA was detected in the oronasal swabs in all vaccinated pigs. These latter data are relevant when JEV vaccination is employed in pigs.