Regulators of Salmonella-host interaction identified by peripheral blood transcriptome profiling: roles of TGFB1 and TRP53 in intracellular Salmonella replication in pigs

Abstract

Peripheral blood transcriptome is an important intermediate data source for investigating the mechanism of Salmonella invasion, proliferation, and transmission. We challenged 4-week old piglets with Salmonella enterica serovar Typhimurium LT2 and investigated the peripheral blood gene expression profile before treatment (d0) and at 2 and 7 days post-inoculation (dpi) using deep sequencing. Regulator pathways were first predicted in silico and validated by wet-lab experiments. In total, 1255, 765, and 853 genes were differentially expressed between 2 dpi/d0, 7 dpi/d0, and 7 dpi/2 dpi, respectively. Additionally, 1333 genes showed a time effect during the investigated Salmonella infection period. Clustering analysis showed that the differentially expressed genes fell into six distinct expression clusters. Pathway annotation of these gene clusters showed that the innate immune system was first significantly upregulated at 2 dpi and then attenuated at 7 dpi. Toll-like receptor cascades, MyD88 cascade, phagosome pathway, cytokine signaling pathway, and lysosome pathway showed a similar expression pattern. Interestingly, we found that the ribosome pathway was significantly inhibited at 2 and 7 dpi. Gene expression regulation network enrichment analysis identified several candidate factors controlling the expression clusters. Further in vitro study showed that TGFB1 can inhibit Salmonella replication whereas TRP53 can promote Salmonella replication in porcine peripheral blood mononuclear cells and murine macrophages. These results provide new insights into the molecular mechanism of Salmonella-host interactions and clues for the genetic improvement of Salmonella infection resistance in pigs.

Figure 1

Numbers of differentially expressed genes. A Summary of differentially expressed transcripts responding to Salmonella inoculation (2 dpi/d0, 7 dpi/d0, 7 dpi/2 dpi), or expression showing time effect. B Comparison of differentially expressed gene counts in LS/PS animals as reported by Huang et al. [25]. These genes were identified using the linear mixed model, and the false discovery rate (FDR) was controlled at less than 0.05, with the fold change between time points required to be higher than 1.5 or less than 0.67.

Figure 2

Hierarchical cluster analysis of differentially expressed genes. We performed data adjustment (median center and normalization) in the cluster analysis. The color codes red, white, black, and dark green represent high, average, low, and absent expression, respectively. A detailed view of gene expression levels in clustering patterns is shown in plot areas from A–F.

Figure 3

Gene expression regulator enrichment analysis using the genes in Cluster A. The significantly over-represented regulators were plotted based on the ratio of targets to the total number of transcripts in the differentially expressed gene list (observed) and all genes presented in the transcriptome (expected). Red color indicates that the regulator is statistically significant whereas blue indicates no significance. The size of the triangle indicates the number of targets for the regulator.

Figure 4

Peripheral blood concentrations in Salmonella-challenged pigs. Serum protein levels were measured prior to challenge and at 2 and 7 dpi using ELISA. Results are presented as box plots plus scatter plots for each sample. Plots A–D show the results of TNF-α, IFN-γ, TGFB1, and TRP53 respectively.

Figure 5

Hierarchical clustering of gene expression data from peripheral blood measured by real-time PCR. Peripheral blood samples were treated in vitro with two different doses of recombinant TGFB1 and p53 protein (1 ng/mL and 10 ng/mL, panel A) and TGFB1 and TRP53 inhibitors (0.1 µmol/mL and 0.5 µmol/mL, panel B). Color codes of yellow, black, and blue represent high, average, and low expression levels, respectively, across the treatments shown.

Figure 6

Intracellular growth of Salmonella enterica serovar Typhimurium strain LT2 in macrophages. Macrophages were first treated with recombinant protein or protein inhibitor, followed by infection with Salmonella bacteria, incubation for 2 h, treatment with gentamicin (100 µg/mL) for 2 h to kill the extracellular bacteria, and immediate lysis or incubation for an additional 8, 12, or 24 h. Line graphs show the intracellular Salmonella counts after the macrophages were treated with Salmonella bacteria. The filled areas represent the standard deviation of three duplicate samples. A PBMCs treated with recombinant TGFB1 or TGFB1 inhibitor; B murine macrophages (RAW 264.7) treated with recombinant TGFB1 or TGFB1 inhibitor; C PBMCs treated with recombinant TRP53 or TRP53 inhibitor; D murine macrophages (RAW 264.7) treated with recombinant TRP53 or TRP53 inhibitor.