Transcriptomic profiling of a chicken lung epithelial cell line (CLEC213) reveals a mitochondrial respiratory chain activity boost during influenza virus infection

Abstract

Avian Influenza virus (AIV) is a major concern for the global poultry industry. Since 2012, several countries have reported AIV outbreaks among domestic poultry. These outbreaks had tremendous impact on poultry production and socio-economic repercussion on farmers. In addition, the constant emergence of highly pathogenic AIV also poses a significant risk to human health. In this study, we used a chicken lung epithelial cell line (CLEC213) to gain a better understanding of the molecular consequences of low pathogenic AIV infection in their natural host. Using a transcriptome profiling approach based on microarrays, we identified a cluster of mitochondrial genes highly induced during the infection. Interestingly, most of the regulated genes are encoded by the mitochondrial genome and are involved in the oxidative phosphorylation metabolic pathway. The biological consequences of this transcriptomic induction result in a 2.5- to 4-fold increase of the ATP concentration within the infected cells. PB1-F2, a viral protein that targets the mitochondria was not found associated to the boost of activity of the respiratory chain. We next explored the possibility that ATP may act as a host-derived danger signal (through production of extracellular ATP) or as a boost to increase AIV replication. We observed that, despite the activation of the P2X7 purinergic receptor pathway, a 1mM ATP addition in the cell culture medium had no effect on the virus replication in our epithelial cell model. Finally, we found that oligomycin, a drug that inhibits the oxidative phosphorylation process, drastically reduced the AIV replication in CLEC213 cells, without apparent cellular toxicity. Collectively, our results suggest that AIV is able to boost the metabolic capacities of its avian host in order to provide the important energy needs required to produce progeny virus.

Fig 1. Microarray analysis of gene expression in CLEC213 cells infected with H6N2 virus.

(A) Schematic representation of the experimental design and classification of differentially expressed genes. (B) Volcano plot representing the distribution of the expressed genes during the infection as a fold change (x axis) and as significance (y axis). (C) Canonical pathways associated to DEG in H6N2-infected CLEC213. Pathways were identified using IPA software and ranked with the p-value (negative logarithm of significance) obtained using the right-tailed Fisher’s exact test. Red and blue bars indicate predicted pathway activation or inhibition, respectively (z-score). Gray bars indicate pathways where no prediction can currently be made.

Fig 2. Functional relationships of pathways regulated during infection of CLEC213 with H6N2 virus.

IPA software was used to build a biological network using the DEG in the H6N2-infected cells compared with mock-infected controls (twofold change, p-value<0.01). Significance of the pathways are indicated at the bottom of each node. The diagram shows the interactions between pathways. Pathways are connected through common genes; the number of shared genes between pathways is indicated in red.

Fig 3. H6N2 infection impacts oxidative phosphorylation.

(A) Schematic representation of the “oxidative phosphorylation” metabolic pathway, constituted of 5 multiprotein complexes. The 4 complexes induced by AIV infection are colored in pink. The detail of the proteins composing the complexes is indicated, each induced gene is highlighted in pink and the fold induction is indicated. (B-C) Heat map showing the normalized level of expression of mitochondrial DEG encoded by the mitochondrial genome (B) or encoded by the cell genome (C). The four replicates of each condition (mock- and H6N2-infected) are represented.

Fig 4. H6N2 induces an inflammatory response and a mitochondrial response at the transcriptomic level.

(A) H6N2-induced inflammatory signature observed by microarray was controlled using independent samples through monitoring the level of expression of IL6 by qRT-PCR. (B) Mitochondrial genes (MT-) and mitochondria-addressed nuclear genes expression was assessed in H6N2-infected cells by qRT-PCR. (C) Mitochondrial gene expression was monitored in H7N1- and ΔF2 H7N1 infected cells. Results are expressed as means ± SEM (* p-value < 0.01).

Fig 5. AIV infection of CLEC213 cells induces an increase in ATP production.

(A) Electron microscopy analysis of H7N1-infected CLEC213 cells. Experiments were done two times independently using new batches of infected cells. Pictures are representative of 30 cell sections analyzed for mitochondria ultrastructure and number in infected or mock cells. (B) Mitochondrial DNA was quantified by qPCR in mock-, H6N2- and H7N1-infected cells. Cells were infected for 16 h at an MOI of 1. Results are expressed as fold-change vs. mock. (C) ATP production was quantified in mock-, H6N2- and H7N1-infected cells and expressed as a ratio of ATP signal on cell number. Results are shown as means ± SEM.

Fig 6. Investigation of extracellular ATP as a danger signal in CLEC213 cells.

(A) P2X7R expression was measured in H6N2-infected cells by qRT-PCR. Cells were infected for 16 h at an MOI of 1. Data are means ± SEM of triplicate qRT-PCR, ** p-value<0.01. (B) CLEC213 cells were cotransfected with NF-κB -luciferase and pRSV-β-Galactosidase reporter plasmids. Cells were then stimulated with 1mM ATP for 16 h. Cell lysates were prepared and assayed for luciferase activity. Results are expressed as mean relative luminescence units normalized to β-galactosidase activity of 3 independent samples. One representative experiment of 3 is shown, * p-value< 0.05. (C) H6N2 AIV Replication was measured in presence or absence of 1mM extracellular ATP. CLEC213 cells were infected at an MOI of 1. Total RNA was extracted 16 h pi and segment 7 vRNA was quantified by qRT-PCR. Results are expressed as copy numbers per μμg of total RNA.

Fig 7. Inhibition of oxidative phosphorylation reduces AIV replication.

(A) Replication of the H6N2 and H7N1 AIV was measured in presence or absence of 125μμM oligomycin. CLEC213 cells were infected at an MOI of 1. Total RNA was extracted 16 h pi and segment 7 vRNA was quantified by qRT-PCR. Results are expressed as copy numbers per μμg of total RNA, *: p-value<0.05, ***: p-value<0.001. (B) Minigenome activity in H6N2-infected cells in the presence or absence of oligomycin (125μμM). Results are expressed as a ratio of RLU on β-Gal activity ± SEM, ***: p-value<0.001.