Elucidation of the Host Bronchial Lymph Node miRNA Transcriptome Response to Bovine Respiratory Syncytial Virus

Abstract

Bovine respiratory disease (BRD) causes substantial morbidity and mortality, affecting cattle of all ages. One of the main causes of BRD is an initial inflammatory response to bovine respiratory syncytial virus (BRSV). MicroRNAs are novel and emerging non-coding small RNAs that regulate many biological processes and are implicated in various inflammatory diseases. The objective of the present study was to elucidate the changes in the bovine bronchial lymph node miRNA transcriptome in response to BRSV following an experimental viral challenge. Holstein-Friesian calves were either administered a challenge dose of BRSV (10^3.5 TCID₅₀/ml × 15 ml) (n = 12) or were mock inoculated with sterile phosphate buffered saline (n = 6). Daily scoring of clinical signs was performed and calves were euthanized at day 7 post-challenge. Bronchial lymph nodes were collected for subsequent RNA extraction and sequencing (75 bp). Read counts for known miRNAs were generated using the miRDeep2 package using the UMD3.1 reference genome and the bovine mature miRNA sequences from the miRBase database (release 22). EdgeR was used for differential expression analysis and Targetscan was used to identify target genes for the differentially expressed (DE) miRNAs. Target genes were examined for enriched pathways and gene ontologies using Ingenuity Pathway Analysis (Qiagen). Multi-dimensional scaling (MDS) based on miRNA gene expression changes, revealed a clearly defined separation between the BRSV challenged and control calves, although the clinical manifestation of disease was only mild. One hundred and nineteen DE miRNAs (P < 0.05, FDR < 0.1, fold change > 1.5) were detected between the BRSV challenged and control calves. The DE miRNAs were predicted to target 465 genes which were previously found to be DE in bronchial lymph node tissue, between these BRSV challenged and control calves. Of the DE predicted target genes, 455 had fold changes that were inverse to the corresponding DE miRNAs. There were eight enriched pathways among the DE predicted target genes with inverse fold changes to their corresponding DE miRNA including: granulocyte and agranulocyte adhesion and diapedesis, interferon signalling and role of pathogen recognition receptors in recognition of bacteria and viruses. Functions predicted to be increased included: T cell response, apoptosis of leukocytes, immune response of cells and stimulation of cells. Pathogen recognition and proliferation of cytotoxic T cells are vital for the recognition of the virus and its subsequent elimination.

Keywords: BRSV challenge; bovine respiratory disease; dairy calves; miRNA; pneumonia; small RNA-Seq.

FIGURE 1

An MDS plot generated in EdgeR illustrating the similarity of the samples based on the Log₂ miRNA gene expression covariance matrix among individuals. Samples from BRSV challenged calves are coloured in red and samples from control calves are coloured in blue. The numbers (1–18) refer to the calf identifier and the letter S in front of the numbers refers to the word “sample.”

FIGURE 2

The enriched canonical pathways found in the IPA analysis (P < 0.05, FDR < 0.1) among the DE predicted target genes with inverse fold changes to those of the corresponding DE miRNAs. The pathways are shown on the x-axis and the –Log₁₀ Benjamini–Hochberg adjusted p values are displayed on the y-axis. The threshold is set to one which equals a Benjamini–Hochberg adjusted p value of 0.1. Pathways with a positive z-score are predicted by IPA to have increased activity and pathways with a negative z-score are predicted to have decreased activity.

FIGURE 3

A Venn diagram (Husen et al., 2008) portraying the DE (up-regualted) genes causing the up-regulation of the interferon signalling pathway in the BRSV challenged calves and the number of their targeting DE (down-regulated) miRNAs.