Exosomal miR-29b of Gut Origin in Patients With Ulcerative Colitis Suppresses Heart Brain-Derived Neurotrophic Factor

Abstract

Background and Aims: While the interplay between heart and gut in inflammatory bowel disease (IBD) has previously been noted, how the inflamed gut impairs heart function remain elusive. We hypothesized that exosomal miRNAs of gut origin induce cardiac remodeling in IBD. Our aim was to identify plasma exosomal miRNAs that not only are of diagnostic value but also contribute to cardiac remodeling in patients with ulcerative colitis (UC). Methods: Plasma exosomes were isolated from UC patients and healthy control subjects and exosomal miRNAs were profiled by next-generation sequencing. Exosomal miR-29b levels in CCD841 CoN colon epithelial cells were detected by RT-qPCR. Exosomes packaged with miR-29b were incubated with H9c2 cells or administered to live mice. Results: The plasma exosomal miRNA profiles of the UC patients were significantly different from that of the controls and 20 miRNAs including miR-29b were differentially expressed. In CCD841 CoN cells, TNFα, IL-1β, and H₂O₂ significantly elevated miR-29b in both the cells and their secreted exosomes (p < 0.01), suggesting that intestinal epithelium secrets exosomes rich in miR-29b in IBD. In H9c2 myoblast cells, miR-29b modulated multiple genes including brain-derived neurotrophic factor (BDNF). Epithelial cell-derived exosomes packaged with miR-29b also attenuated BDNF and increased cleaved caspase 3, suggestive of apoptosis. Furthermore, tail vein injection of engineered exosomes with high levels of miR-29b suppressed BDNF and augmented cleaved caspase 3 in the heart of adult mouse (p < 0.01). Conclusion: Plasma exosomal miRNA profile could be a novel diagnostic approach for IBD. Excessive plasma exosomal miR-29b suppresses critical proteins like BDNF in IBD, leading to cardiac impairment.

Keywords: cardiac remodeling; exosome; inflammatory bowel disease; microRNA; pro-inflammatory cytokine.

FIGURE 1

Genome-wide survey of human small RNAs in plasma exosomes. (A) Stacked bar charts showing relative distribution of sequencing reads corresponding to 8 types of non-coding RNAs. (B) Principal component analysis (PCA) plot showing the first two principal components using normalized read counts. Each dot indicates a sample. C = control; UC = ulcerative colitis.

FIGURE 2

Identification of dysregulated exosomal miRNAs in patients with UC by statistical analysis of small RNA-seq data. (A) Volcano plot representation of differential expression analysis of miRNAs in patients with UC vs healthy control subjects. The miRNAs are colored if they pass the thresholds for p value and log fold change, red if they are upregulated and green if they are downregulated. (B) Hierarchical clustering of normalized exosomal miRNA levels for the 20 miRNAs having statistical significance for differential expression between UC patients and control subjects.

FIGURE 3

KEGG pathway analyses of the dysregulated exosomal miRNAs. (A) KEGG pathway analysis of upregulated miRNAs. (B) KEGG pathway analysis of downregulated miRNAs.

FIGURE 4

GO enrichment analyses of the exosomal miRNAs dysregulated in UC patients compared to the control subjects. (A) Upregulated miRNAs. (B) Downregulated miRNAs.

FIGURE 5

Upregulation of miR-29b gene (Mir29b) expression by TNFα, IL-1β, and H₂O₂ in CCD841 CoN colon epithelial cells. CCD841 CoN cells were incubated with serial doses of TNFα, IL-1β, H₂O₂ or their combination for 24 h, or with 10 ng/ml TNFα, 10 ng/ml IL-1β, or 10 µM of H₂O₂ for various times as indicated. Mir29b gene expression was quantitated by TaqMan-based RT-qPCR. (A) TNFα dose- and time-dependently elevated Mir29b levels. (B) IL-1β increased Mir29b in a dose- and time-dependent manner. (C) H₂O₂ dose- and time-dependently up-regulated Mir29b. (D) Combination of TNFα, IL-1β, and H₂O₂ augmented Mir29b expression. (E) Exosomal levels of Mir29b were elevated by 10 ng/ml TNFα, 10 ng/ml IL-1β, or 10 µM H₂O₂ after 24 h incubation. n = 3 independent experiments. *p < 0.01 vs. control (concentration or time zero).

FIGURE 6

miR-29b modulates the expression levels of multiple genes related to heart function in H9c2 cells. miR-29b or negative control miRNA was overexpressed in H9c2 cells by transient transfection. TaqMan-based RT-qPCR was performed to determine the expression levels of Bdnf (A), Col1a1 (B), Col3a1 (C), Fox O 4 (D), Myl7 (E), Bcl2 (F), Cacna1g (G), Mcf2 (H), Mylpf (I), Myo5b (J), Dnmt3a (K), and Dnmt3b (L). Gapdh served as an internal control. n = 3 independent experiments. *p < 0.01 vs. negative control miRNA (Ctl miR).

FIGURE 7

Exosomes rich in miR-29b suppress BDNF and enhance cleaved caspase 3 in vitro (cell culture) and in vivo (mouse tail vein injection). (A) The medium from IL-1β-treated CCD841 CoN cells suppressed BDNF mRNA (left panel) and protein levels (right panel) in H9c2 cells. CCD841 CoN cells were transfected with miR-29b inhibitor (inh) and treated with 10 ng/ml IL-1β. The medium without IL-1β was collected and used to treat H9c2 cells for 24 h. (B) Representative images of Western blots showing exosomal miR-29b decreased BDNF in vitro and in vivo. Bar graphs show the relative band densities of BDNF (C) and cleaved caspase 3 (D) after normalization to the housekeeping gene GAPDH. n = 3 independent experiments for H9c2 cells. n = 5 for in vivo experiments. *p < 0.01 vs. controls. CCD841 CoN-derived or mouse plasma exosomes were transfected with miR-29b or control miRNA. The CCD841 CoN-derived exosomes loaded with rno-miR-29b-3p mimics were added to H9c2 cells and incubated for 24 h, followed by protein isolation and Western blot analysis. Adult mice were given the mouse exosomes transfected with mmu-miR-29b-3p mimics or control miRNA through tail vein injection, and the animals were euthanized 24 h later. Individual hearts were collected for Western blot analysis of proteins.