Microarray analysis of differential gene expression profiles in blood cells of naturally BLV-infected and uninfected Holstein-Friesian cows

Abstract

The aim of the present study was to examine gene expression changes in response to bovine leukemia virus (BLV) infection, in an effort to determine genes that take a part in molecular events leading to persistent lymphocytosis (PL), and to better define genes involved in host response to BLV infection. Using bovine 70-mer oligonucleotide spotted microarrays (BLOPlus) and qRT-PCR validation, we studied global gene expression profiles in blood cells in vivo of 12 naturally BLV-infected Polish Holstein cows, and 12 BLV non-infected controls of the same breed and reared in herds with high BLV seroprevalence. With an arbitrary cut-off value of 1.5-fold change in gene expression, we identified the down-regulation of 212 genes (M value ≤-0.585) and the up-regulation of 158 genes (M value of ≥0.585) at 1% false discovery rate in BLV-positive animals in comparison to the BLV-negative group. The gene set enrichment analysis demonstrated that the differentially expressed (DE) genes could be classified to diverse biological processes, including immune response of host blood cells. Interestingly, our data indicated the potential involvement of the innate immunity, including complement system activation, NK-cell cytotoxicity and TREM-1 signaling, during the BLV-induced pathogenesis. We showed the occurrence of numerous regulatory processes that are targeted by BLV-infection. We also suggest that a complex network of interrelated pathways is disturbed, causing the interruption of the control of B-cell proliferation and programmed cell death.

Keywords: BLV; Cattle; Gene expression; Host response; Microarrays.

Fig. 1

Real-time qRT-PCR validation of microarray results. a Whisker-box plots of relative expression values for 14 genes in BLV-infected group in comparison to non-infected control. The median of gene expression is depicted by the dotted line. Whiskers and box show the minimum and maximum values and the interquartile range of observations, respectively. b Correlation plot between fold-change (FC) differences revealed by gene expression measurement methods (microarray vs. qRT-PCR). The regression line among the results from different platforms is described by equation: log2FC-qRT-PCR = −0.39 + 1.02 × log2FC-microarray, with Pearson’s correlation coefficient (r) equals 0.98

Fig. 2

Enrichment analysis of differentially expressed genes identified in this study. Bar length indicates the significance and equals to the negative logarithm of enrichment p value. a Top 20 significant biological processes from GO ontology. b Top 20 significant process networks from MetaCore ontology. c Top 20 significant molecular functions from GO ontology. d Top 20 significant localization (cellular compartments) from GO ontology

Fig. 3

Enrichment analysis of differentially expressed genes identified by disease biomarkers (GeneGO). Bar length indicates the significance and equals to the negative logarithm of enrichment p value. a Diseases associated with the genes down-regulated in BLV-infected cattle in comparison to non-infected controls. b Diseases associated with the genes up-regulated in BLV-infected cattle in comparison to non-infected controls

Fig. 4

CREB Network. The CREB network shows the CREB transcription factor (enclosed by black circle) as a hub controlling the expression of proteins which are encoded by genes from the DE gene list. The various symbols used in the network have been described in detail in MetaCore Quick Reference Guide file publicly available at: https://portal.genego.com/help/MC_legend.pdf

Fig. 5

HIF1 Network. The HIF1 network shows the HIF1 transcription factor (enclosed by black circle) as a hub controlling the expression of proteins which are encoded by genes from the DE gene list. The various symbols used in the network have been described in detail in MetaCore Quick Reference Guide file publicly available at: https://portal.genego.com/help/MC_legend.pdf