Integrated Analysis of Transcriptome mRNA and miRNA Profiles Reveals Self-Protective Mechanism of Bovine MECs Induced by LPS

Abstract

Many studies have investigated the molecular crosstalk between mastitis-pathogens and cows by either miRNA or mRNA profiles. Here, we employed both miRNA and mRNA profiles to understand the mechanisms of the response of bovine mammary epithelial cells (bMECs) to lipopolysaccharide (LPS) by RNA-Seq. The total expression level of miRNAs increased while mRNAs reduced after LPS treatment. About 41 differentially expressed mRNAs and 45 differentially expressed miRNAs involved in inflammation were screened out. We found the NFκB-dependent chemokine, CXCL1, CXCL3, CXCL6, IL8, and CX3CL1 to be strongly induced. The anti-apoptosis was active because BCL2A1 and BIRC3 significantly increased with a higher expression. The effects of anti-microbe and inflammation were weakly activated because TNF, IL1, CCL20, CFB, S100A, MMP9, and NOS2A significantly increased but with a low expression, IL6 and β-defensin decreased. These activities were supervised by the NFKBIA to avoid excessive damage to bMECs. The bta-let-7a-5p, bta-miR-30a-5p, bta-miR-125b, and bta-miR-100 were essential to regulate infection process in bMECs after LPS induction. Moreover, the lactation potential of bMECs was undermined due to significantly downregulated SOSTDC1, WNT7B, MSX1, and bta-miR-2425-5p. In summary, bMECs may not be good at going head-to-head with the pathogens; they seem to be mainly charged with sending out signals for help and anti-apoptosis for maintaining lives after LPS induction.

Keywords: LPS; bMEC; mRNA profile; mastitis; miRNA profile; self-protection.

Figure 1

Overview of transcriptome sequencing data. (A) The relative expression density of transcripts; (B) The relative expression level of transcripts; (C) The genome distribution of clean reads in lipopolysaccharide (LPS)-induced bovine mammary epithelial cells at 0, 4, 8, and 12 h, respectively.

Figure 2

Gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis of DEGs. (A) The significantly enriched GO terms in the biological process at 4, 8, and 12 h, respectively. (B) The significantly enriched GO terms in cellular components at 4, 8, and 12 h, respectively. (C) The significantly enriched GO terms in molecular function at 4, 8, and 12 h, respectively. (D) The significantly enriched KEGG pathway at 4, 8, and 12 h, respectively. (E) The heatmap of differentially expressed genes (DEGs) involved in immune response, inflammatory response, chemokine activity, cytokine activity, NOD-like receptor (NLR) signaling pathway, phagosome, RIG-I-like receptor signaling pathway, and COP9 signalosome. In (E), the dark green represents down-regulationand the red represents upregulation.

Figure 3

The relative expression level of mRNA by qPCR. The bar graph shows the qPCR result, and the stair graph shows the sequencing result; 0, 4, 8, and 12 h represent the LPS-induced 0, 4, 8, and 12 h bMECs; “*” represents P ≤ 0.05; “**” represents P ≤ 0.01.

Figure 4

The 29 top-expressed miRNAs at 0, 4, 8, and 12 h LPS-induced bMECs. “n” represents the number of identified miRNAs at 0, 4, 8, and 12 h LPS induced bMECs.

Figure 5

The significantly up/downregulated miRNAs along with significantly down/upregulated targeted mRNAs were screened out at 4, 8, and 12 h LPS-induced bMECs compared to 0 h, respectively. (A) The results were at 4 h; (B) the results were at 8 h; (C) the results were at 12 h. Green represents downregulation and red represents upregulation.

Figure 6

The differentially expressed miRNAs involved in inflammation and immunity. (A) Forty-five significantly up- or down regulated miRNAs involved in inflammation and immunity and their targeted mRNAs. Diamond represents mRNAs and the oval represents miRNAs; (B) The heatmap of relative expression for 45 significantly up/downregulated miRNAs in 4, 8, and 12 h bMECs. Green represents the downregulation and red represents the upregulation.

Figure 7

The relative expression level of microRNA by qPCR. The bar graph is the qPCR result, and the stair graph is the sequencing result; 0, 4, 8, and 12 h represent the LPS-induced 0, 4, 8, and 12 h bMECs; “*” represents P ≤ 0.05; “**” represents P ≤ 0.01; “***” represents P ≤ 0.001.

Figure 8

The expression network of the nutrition metabolism pathway-related genes. (A) The light blue triangle represents GO term related to the nutrition metabolism; the pentagon represents the differentially expressed mRNAs related to the nutrition metabolism pathway; (B) The expression level in fragment per kilobase of exon (FPKM), of 4 genes in LPS-induced bMECs; (C) The stair graphic of the relative differential relationship for 4 genes between different timepoints, such as log2(12/0 h), log2(8/0 h), log2(4/0 h), log2(12/8 h), log2(12/4 h), and log2(8/4 h) values. 0, 4, 8, and 12 h represent the LPS-induced 0, 4, 8, and 12 h bMECs; Green represents downregulation and red represents upregulation in (A,C).

Figure 9

The TLR4 signaling cascades in LPS-induced bMECs. (A) The expression level (FPKM) of all members involved in myeloid differentiation factor 88 (MyD88)-dependent and independent pathway in LPS-induced bMECs; (B) TLR4 signaling following MyD88- dependent and independent pathways. First, signaling via MyD88 subsequently recruits IRAK1, IRAK4 (IL1R associated kinase family), and TNF receptor-associated factor 6 (TRAF6), forms a complex with TAK1 (mitogen-activated protein kinase kinase kinase 7, MAP3K7), then activate TAK1, which phosphorylates IκB kinases (IKKs, IKK1/IKK2/NEMO complex) and MAP kinases (e.g., JNK, p38). TAK1 can also be induced by receptor-interacting protein 1 (RIP1). IKKs phosphorylate the IκBα, then leads to IκBα degradation, enabling the nuclear translocation of NFκB, ultimately beginning the transcription activation of pro-inflammatory cytokines (TNFα, IL1, and IL6), chemokines (GRO1, CXCL3, CXCL6, CXCL8, and CX3CL1), anti-apoptotic genes (BCL2A1 and BIRC2/3), anti-microbial genes (β-defensins, S100A, CFB, and CCL20) and inflammatory genes (COX2, MMP9, SOD, and NOS2A). Another pathway AP-1 transcription factor formed by c-Fos and c-Jun can be activated by JNK1/p38 complex, then initiate the production of cytokines and chemokines like NF-kB signaling. Second, the TRIF-dependent cascade through the bridging adaptor, TRIF-related adaptor molecule (TRAM) by the delayed mode in either a TRAF6 or RIP1-dependent manner activate TAK1 and trigger NFκB and AP-1 pathway. TRIF also interacts with TRAF-associated NFκB activator-binding kinase 1 (TBK1) and IκB kinase epsilon (IKKε), which phosphorylate IRF3, leading to induction of typeIIFN genes (IFNβ/IL6). The IL6 or IFNβ binds to its receptor, IFNAR1/2 causing the activation of JAKs, eventually activating STAT1. The STAT1 binds as a homodimer to the gamma interferon-activating sequence or trimerizes with STAT2 and IRF9 to form interferon-stimulated transcription factor 3 (ISGF3), which can bind to interferon-stimulated response element (ISRE)-mediated expression of secondary response genes, such as MX1, OASZ1, ISG15, CXCL10, CCL2, CCL5, and IRF7. Binding IFNAR1/2 also activates MAPKs, then trigger C/EBPβ to induce the expression of acute-phase proteins, such as CCL5, SAA3, and HP. Red represents increased expression; green represents decreased expression in (B).