Overexpression of miRNA-25-3p inhibits Notch1 signaling and TGF-β-induced collagen expression in hepatic stellate cells - PubMed (original) (raw)
doi: 10.1038/s41598-019-44865-1.
Berit Genz 1 2 3, Katharine M Irvine 2 3, Jamie R Kutasovic 3 4, Mariska Miranda 4, Francis D Gratte 5 6, Janina E E Tirnitz-Parker 5, John K Olynyk 7 8, Diego A Calvopina 1, Anna Weis 1, Nicole Cloonan 9, Harley Robinson 10, Michelle M Hill 10, Fares Al-Ejeh 4, Grant A Ramm 11 12
Affiliations
- PMID: 31189969
- PMCID: PMC6561916
- DOI: 10.1038/s41598-019-44865-1
Overexpression of miRNA-25-3p inhibits Notch1 signaling and TGF-β-induced collagen expression in hepatic stellate cells
Berit Genz et al. Sci Rep. 2019.
Abstract
During chronic liver injury hepatic stellate cells (HSCs), the principal source of extracellular matrix in the fibrotic liver, transdifferentiate into pro-fibrotic myofibroblast-like cells - a process potentially regulated by microRNAs (miRNAs). Recently, we found serum miRNA-25-3p (miR-25) levels were upregulated in children with Cystic Fibrosis (CF) without liver disease, compared to children with CF-associated liver disease and healthy individuals. Here we examine the role of miR-25 in HSC biology. MiR-25 was detected in the human HSC cell line LX-2 and in primary murine HSCs, and increased with culture-induced activation. Transient overexpression of miR-25 inhibited TGF-β and its type 1 receptor (TGFBR1) mRNA expression, TGF-β-induced Smad2 phosphorylation and subsequent collagen1α1 induction in LX-2 cells. Pull-down experiments with biotinylated miR-25 revealed Notch signaling (co-)activators ADAM-17 and FKBP14 as miR-25 targets in HSCs. NanoString analysis confirmed miR-25 regulation of Notch- and Wnt-signaling pathways. Expression of Notch signaling pathway components and endogenous Notch1 signaling was downregulated in miR-25 overexpressing LX-2 cells, as were components of Wnt signaling such as Wnt5a. We propose that miR-25 acts as a negative feedback anti-fibrotic control during HSC activation by reducing the reactivity of HSCs to TGF-β-induced collagen expression and modulating the cross-talk between Notch, Wnt and TGF-β signaling.
Conflict of interest statement
The authors declare no competing interests.
Figures
Figure 1
In vitro analysis of the effect of miR-25 overexpression on the activation status of human hepatic stellate cells (HSCs). (A) In situ hybridization of LX-2 cell line with DIG-labeled miR-25 specific probes (red). A scrambled miR-25 probe was used as negative control (scale bar: left panel 100 µm, right panel 10 µm). (B) Relative quantification of miR-25 expression 24, 48, 72 and 96 h after transfection of LX-2 cells with miR-25 mimics (n = 3–4). (C) Analysis of relative mRNA expression of different HSC marker for quiescence (PPAR-γ (PPARG), E-cadherin (CDH1)) and activation (vimentin (VIM), αSMA (ACTA2), collagen 1a1 (COL1a1), TGF-β1 (TGFB)) as well as TGF-β receptor type 1 (TGFBR1) in miR-25 over-expressing compared to control miR transfected cells 48 h after transfection (n = 5–7). Proliferation analysis using Incucyte confluency measurement (n = 3) (D) or MTT assay (n = 5) (E) in control and miR-25 over-expressing LX-2 cells. (F) Relative quantification of miR-25 expression in untreated and TGF-β treated (5 ng/ml for 24 h) LX-2 cells (n = 5). (*p < 0.05 vs control).
Figure 2
In vitro analysis of the effect of miR-25 overexpression on TGF-β signaling in LX-2 cells. (A) qRT-PCR analysis of collagen1a1 (COL1A1) and αSMA (ACTA2) mRNA expression (n = 3) and (B) Western Blot analysis of phosphorylation of SMAD2 (n = 4) and SMAD3 (n = 8) proteins after TGF-β stimulation (5 ng/ml for 24 h (mRNA) and 6 h (protein)) in miR-25- or control-transfected LX-2 cells. (C) Western Blot analysis of TGF-β signaling components SMAD2 and SMAD3 in miR-25 overexpressing LX-2 cells compared to control cells (n = 7–8). Cropped Western Blot images originate from the same (target vs. loading control) or different membranes (phosphorylates vs. whole protein; time points). (D) qRT-PCR analysis of collagen1a1 (COL1A1) and αSMA (ACTA2) mRNA expression after TGF-β stimulation (5 ng/ml for 24 h) in miR-25 antagomir- (inhibitor) or control antagomir-transfected LX-2 cells. (*p < 0.05 vs control; #p < 0.05 vs. PBS; ##p < 0.01 vs. PBS; ###p < 0.001 vs. PBS).
Figure 3
Identification of miR-25 biologically relevant targets using biotin pull-down assay. (A) Hierarchical clustering of Illumina sequencing data using the lumi package to demonstrate sample relationship between whole mRNA lysates (control, samples D, H, L) and enriched miRNA target mRNA lysates (pull-down, samples C, G, K). (B) Volcano plot depicting log2-transformed fold-change against log10-transformed adjusted p-value of genes detected above background level of the array in pull-down compared to whole cell lysate samples (fold change ≤−2; adj. p-value < 0.05). (C) Venn diagram showing the overlap between miR-25 predicted targets by TargetScan, miRDB and biotin pull-down assay. (D) Selected Putative miR-25 target mRNA enriched by biotin-pulldown assay (E) qRT-PCR expression analysis of selected predicted target mRNA regulated by miR-25 (n = 4–6). (F) Validation of miR-25 binding to predicted target mRNA by luciferase-assay (n = 6–7). (G) Analysis of relative mRNA expression of target genes ADAM-17 and FKBP14 in LX-2 cells treated with miR-25 antagomir (inhibitor) or control antagomir (n = 9–10). (*p < 0.05 vs control; **p < 0.01 vs. control; ***p < 0.001 vs. control).
Figure 4
NanoString mRNA analysis of stem cell-related signaling pathways in miR-25 overexpressing LX-2 cells. (A) Heat map of the mRNA expression ratio of the measured genes in miR-25 vs. control transfected LX-2 cells 24 and 48 h after transfection. Each column represents the average of triplicates per time point (data for each replicate is in Supplementary Data 2). Statistical evaluation was carried out with two-way ANOVA corrected for multiple comparisons by false discovery rate (FDR) using two-stage linear step-up procedure of Benjamini, Krieger and Yekutieli (GraphPad® Prism). The q values (FDR) and individual P values are shown in Supplementary Data 2. The genes shown in the heatmap were statistically different (q < 0.05) between miR-25-transfected cells and control-transfected cells; non-statistically significant differences are marked as n.s. Genes downregulated in miR-25 overexpressing cells are presented in blue, upregulated genes in red. Colour scale represents fold change. String pathway analysis of genes downregulated (B) and upregulated (C) in miR-25 transfected LX-2 cells compared to control cells.
Figure 5
Effect of miR-25 overexpression on Notch signaling in LX-2 cells. (A) qRT-PCR analysis of mRNA expression of Notch signaling components in miR-25 transfected LX-2 cells compared to control 48 h after transfection (n = 7–8). Western Blot analysis of (B) ADAM-17 (n = 18), (C) NOTCH1 full length (FL) (n = 10) and (D) N-terminal protein (NTM) (n = 13), in miR-25 overexpressing LX-2 48 and 72 h after transfection. (E) Western Blot analysis of NOTCH1 intracellular domain (NICD1) protein expression in nuclear fraction of LX-2 cells 48 h after transfection with miR-25 or control mimics (n = 8). Western Blot analysis of (F) Jagged-1 (JAG1) (n = 7), (G) NOTCH3 full length (FL) (n = 13) and (H) N-terminal protein (NTM) (n = 12) in miR-25 overexpressing LX-2 48 and 72 h after transfection. Cropped Western Blot images originate from the same (target vs. loading control) or different membranes (time points). (cytopl. – cytoplasmatic fraction; *p < 0.05 vs control; **p < 0.01 vs. control; ***p < 0.001 vs. control).
Figure 6
Effect of miR-25 overexpression on Wnt signaling in LX-2 cells. (A) qRT-PCR analysis of mRNA expression of Wnt signaling components in miR-25 transfected LX-2 cells compared to control 48 h after transfection (n = 7–8). Western Blot analysis of (B) Wnt5 (n = 7) and (C) β-catenin (n = 8) in miR-25 overexpressing LX-2 48 and 72 h after transfection. Cropped Western Blot images originate from the same (target vs. loading control) or different membranes (time points) (*p < 0.05 vs control; **p < 0.01 vs. control; ***p < 0.001 vs. control).
Figure 7
miR-25 expression in murine models of hepatic fibrosis. (A) In situ hybridization of DIG-labeled miR-25 specific probes (red) in cultured mouse primary HSCs 1 day (quiescent) and 7 day (activated) in culture. Cells were counterstained with HSC-marker GFAP (glial fibrillary protein) or αSMA (green), respectively. A scrambled miR-25 probe was used as negative control (scale bar: 20 µm). (B) qRT-PCR expression analysis of miR-25 in primary murine HSCs 1 day (1d) and 7 days (7d) in culture (n = 5). qRT-PCR expression analysis of miR-25 expression in liver tissue of mice (C) treated with TAA in the drinking water for 12 weeks (n = 9 per group) or (D) fed with a CDE diet for 1 week (n = 7–9 per group). Control animals were kept on standard chow and water. (E) qRT-PCR expression analysis of miR-25 in liver tissue from mice treated with TAA for 8 weeks, following fibrosis regression for 2 or 4 weeks after cessation of the toxin (n = 6 per group). (F) Analysis of Pearson correlation between liver mRNA expression of the HSC activation marker αSMA (Acta2) and miR-25 in livers of animals regressing from 8 weeks of TAA treatment. The line represents orthogonal regression. qRT-PCR expression analysis of (G) miR-25 and (H) Fkbp14, Adam-17 and Tgfbr1 mRNA in HSCs isolated from liver tissue of healthy mice (control) or mice treated with TAA for 6 weeks (n = 3–5). (*p < 0.05 vs. control/8 wk TAA; ***p < 0.001 vs. control/8 wk TAA).
Figure 8
Schematic representation of the putative effect of miR-25 overexpression on TGF-β-induced collagen expression in HSCs. (A) In untreated HSCs Notch1 receptor is stepwise cleaved by ADAM-17 sheddase and the γ-secretase complex, which is stabilised by FKBP14 protein, releasing the Notch intracellular domain (NICD1) into the cytoplasm. The NICD1 translocates into the nucleus and induces gene expression of TGF-βR1 by binding to the CSL (CBF1/RBPJ-κ, Suppressor of Hairless, Lag-1) transcription factors. In the cell membrane TGF-β receptor 1 (TGF-βRI) protein dimerises with TGF-βRII to bind TGF-β, resulting in phosphorylation of Smad proteins 2 and 3. The p-Smad2/3 complex then initiates expression of collagen1a1 (COL1A1) in the nucleus. (B) HSCs overexpressing miR-25 downregulate expression of ADAM-17 and FKBP14, therefore cleavage of Notch1 receptor is decreased. As a result less NICD1 translocates into the nucleus followed by reduced expression of TGF-βRI in those cells. A reduction of TGF-βRI on the cell surface leads to diminished sensitivity to TGF-β, decreased Smad2 phosphorylation and subsequent COL1A1 expression after TGF-β stimulation.
References
Publication types
MeSH terms
Substances
LinkOut - more resources
Full Text Sources
Molecular Biology Databases
Miscellaneous