Effect of lipids on the infectivity of influenza A viruses

Influenza is an enveloped, single stranded negative sense RNA virus that subverts host cell factors like lipids for its own replication. Airway epithelium contributes towards the host defence against pulmonary pathogens by a number of mechanisms including the production of cytokines, chemokines and surfactant phospholipids and proteins. Phosphatidylcholine (PC) and phosphatidylglycerol (PG) make up the major of the composition of pulmonary surfactant. Therefore, this project aimed to explore the potential antiviral activity of exogenous phospholipids against influenza A viruses in vitro. The specific aims were to investigate the potential of these lipids to reduce the infectivity of influenza A virus and to measure the potential toxicity of specific exogenous phospholipids, using human adenocarcinomic alveolar basal epithelial (A549) and Madin Darby canine kidney (MDCK) cell lines and chicken red blood cells. Two types of phospholipids were used at concentrations of 5, 50 and 500 μM to treat avian influenza virus H2N3, equine influenza virus H3N8 or pandemic influenza virus H1N1. The first type are phospholipids composed of two aliphatic chain including 1,2-dipalmitoyl-sn-glycero-3- phospho-L-serine (DPPS), 1,2-dipalmitoyl-sn-glycero-3-phospho-(1'-rac-glycerol) (DPPG), 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC). The second type are lyso-derived lipids including 1-palmitoyl-2-hydroxy-sn-glycero-3-phospho-L-serine (LPS), 1-palmitoyl-2-hydroxy-sn- glycero-3 -phospho-(1'-rac-glycerol) (LPG), 1-palmitoyl-2-hydroxy-sn-glycero-3-phosphocholine (LPC). Lipids make up the bilayer membranes of both virus particles and host cells. The potential toxicity of these lipids was assessed using chicken RBCs; incubation with DPPC deformed the RBC as they became round and nuclear loss occurred after 5-30 mins of incubation with 500μM DPPC However, no changes were observed when excess lipids were removed by ultrafiltration prior to incubation using two sequential filtration steps with an Amicon Ultra-15 Filter (50KDa MWCO; molecular weight cut-off). This result was further confirmed through ultrafiltration of phospholipids DPPG, LPG, DPPC, LPC, DPPS, or LPS prior to incubating them with cells as no significant changes were measured in both cell size and granularity of suspension MDCK cells or on the cell viability of adherent MDCK or A549 cells. Therefore, the use of ultrafiltration to remove excess lipids using the Amicon filtration system prior to incubating them with cells was used for further experiments to avoid any direct toxicity of the lipids on the cells. Modifying the lipid envelope of influenza A virus by cholesterol depletion using methyl-β-cyclodextrin (MβCD) reduced H2N3 infectivity of MDCK cells in a dose-dependent manner suggesting that it disrupts the natural packing of influenza viral envelope lipids. Regarding the importance of phospholipid head charge and size, pre-treated H2N3 with charged and small lipid heads, such as DPPG reduced virus infectivity more than when DPPS or DPPC were used. However, treating H2N3 virus with various lyso-lipids including LPS, LPC, or LPG produced a significant inhibition to < 15 % of virus infection in MDCK cells, suggesting that the lipid head and the number of aliphatic chains are an essential factors in the modulation of virus infectivity by phospholipids. Similarly, pre-treatment of H1N1 and H3N8 with phosphatidylglycerol (DPPG or LPG) significantly reduced the virus infectivity in a dose-dependent manner, although, the use of DPPG on the filamentous H3N8 virus showed less reduction than that of H2N3 and H1N1. In a similar manner, pre-treatment of LPAI H2N3 with LPG and DPPG reduced virus infectivity in both A549 and MDCK cells, but it was more noticeable in A549 than MDCK cells suggesting some differences in the receptor/binding characteristics between these cell lines. Transmission electron microscopy (TEM) revealed changes in influenza virus H2N3 and H3N8 morphology upon treatment of H2N3 or H3N8 with exogenous lipid treatment, H2N3 displayed extensive fusions between virions and liposomes and formed giant particles with an uneven surface distribution of viral glycoprotein. The lyso-lipids had a greater impact on morphology of both virus strains, suggesting that the exogenous lipids may have been inserted into only one leaflet of the virion, leading to disruption of the viral envelope and therefore reduction in influenza virus infectivity. The physicochemical structure of the lipid head has an impact on influenza virus adhesion and fusion to liposomes. The results showed that the two aliphatic chains, negatively charged lipid head domain DPPG and DPPS blocked virus haemagglutinin unit (HAU) 16 and 32 at ≥ 500 μM or 250 μM, respectively. However, flow cytometry results showed that only LPG, LPC, LPS, DPPC and DPPG pre-treatment significantly blocked the binding of H2N3 to MDCK cells, in a dose-dependent manner, while DPPS pre-treatment did not reduce virus binding. These flow cytometry results were confirmed by immunofluorescence staining of virus protein using MDCK and A459 cells infected with H2N3 or H3N8, pre-treated with 50 and 500 μM of LPG and DPPG at 4°C. This suggested that pre-incubating the virus with phospholipids seemed to have an impact on the ability of the virus to bind either by inhibiting virus binding to cellular receptors or by causing disruption of the viral envelope. Expression of cytokine mRNA for IL-8 and TNF-α were significantly reduced in MDCK or A549 cells infected with H2N3 or MDCK cells infected with H1N1 and H3N8 following pre-treatment of virus with DPPG or LPG. However, no significant reduction of TNF-α was observed under the same conditions following pre-treatment of H3N8 virus. Impact of pre-treatment of virus with either 500μM of DPPG or LPG in reducing H2N3 infectivity correlated with reduction of IFN-β mRNA expression at 6 and 4 hrs pi when compared to un-treated virus. This suggested that the virus may have been altered upon lipid pre-treatment leading to reduced cytokine expression levels due to a reduced binding and entry of virus to cells. A significant reduction in virus M gene RNA expression was also observed following pre-treatment of H2N3 with phospholipid in both MDCK (DPPG: 42%, LPG 37%) and A549 cells (DPPG: 13%, LPG 58%). In conclusion, the type of polar head and the number of aliphatic chains are the two variables to consider when modulating virus infectivity with lipids. Altering the lipid composition or structure/shape of the virus using exogenous lipids can be envisaged as a potential treatment for influenza