A meta-analysis of transcriptomic characterization revealed extracellular matrix pathway involved in the H5N1 and H7N9 infections

Avian-origin H5N1 and H7N9 influenza A viruses are capable of causing lethal infection in humans, with serious lung pathology and leading to acute respiratory distress syndrome. The contribution of host response associated with the poor prognosis of H5N1 and H7N9 infections remains unclear. The aim of this study was to identify the host factors involved in the high pathogenicity of H5N1 and H7N9 by a systematical meta-analysis. The RNA-seq datasets related to H5N1, H7N9, and H1N1 infections with time series were retrieved from GEO. After merging the data from different series, ComBat was used to adjust the known variances from different batches. The transcription factors binding the genes in each cluster were predicted by PASTAA. We figured out the genes that were differentially expressed at any time point in samples infected with H5N1, H7N9, or H1N1. The analysis of biological function showed that genes related with cytokine were up-regulated in all three viruses. However, genes associated with carbon metabolism were found exclusively down-regulated in H7N9 and the extracellular matrix pathway were only enriched in H5N1 and H7N9. To summary, our study suggested that the extracellular matrix might be associated with the high fatality of H5N1 and H7N9 viruses in humans.