A landscape of mouse mitochondrial small non-coding RNAs

Abstract

Small non-coding RNAs (ncRNAs), particularly miRNAs, play key roles in a plethora of biological processes both in health and disease. Although largely operative in the cytoplasm, emerging data indicate their shuttling in different subcellular compartments. Given the central role of mitochondria in cellular homeostasis, here we systematically profiled their small ncRNAs content across mouse tissues that largely rely on mitochondria functioning. The ubiquitous presence of piRNAs in mitochondria (mitopiRNA) of somatic tissues is reported for the first time, supporting the idea of a strong and general connection between mitochondria biology and piRNA pathways. Then, we found groups of tissue-shared and tissue-specific mitochondrial miRNAs (mitomiRs), potentially related to the "basic" or "cell context dependent" biology of mitochondria. Overall, this large data platform will be useful to deepen the knowledge about small ncRNAs processing and their governed regulatory networks contributing to mitochondria functions.

Fig 1. Rationale of the study.

Illustration created with Servier Medical Art, provided by Servier, licensed under a Creative Commons Attribution 3.0 Unported license.

Fig 2. ncRNAs identified in the current study.

The light-blue indicates RNA biotypes percentages detected in indicated mitochondria tissue on the annotated ones in the different databases, i.e. RNA-Central, Ensembl (GRCm38_p5) and miRBase (v22). Misc_RNA: miscellaneous RNA; otherRNA: snRNA, snoRNA and RNA fragments are included.

Fig 3. mitopiRNAs expression analyses.

(A) Normalized expression unsupervised heatmap relative to piRNAs in the mitochondria of the indicated tissues, obtained by using the software ComplexHeatmap, a package of R. (B) The Venn diagram displaying the numbers of overlapping mitopiRNAs (Raw Count Mean ≥1) among the different tissues.

Fig 4. mitomiRNAs expression analyses.

(A) Normalized expression unsupervised heatmaps relative to mitomiRs of the indicated tissues, obtained by using the software ComplexHeatmap, a package of R. (B) Distribution of mitomiR levels with respect to frequency; numbers of sequence reads (showed on the bars) were taken as miRNA levels and the values are represented in the form of range of values in mitochondrial sRNA libraries of each tissue. (C) The Venn diagram displaying the numbers of overlapping mitomiRs among the different tissues.

Fig 5. Expression patterns of miRNAs in cytosol and mitochondria fractions from the different tissues.

Expression analyses of the indicated miRNAs were performed by using Taqman assays on the RNA extracted from cytosol and mitochondria fractions of the indicated tissues. All analyses were performed at least in triplicate on each sample and mean + sd is reported. p-values at Student’s t-test were *p < 0.05, **p < 0.01, or ***p < 0.001.

Fig 6. Biological pathways potentially governed by tissue-shared and tissue-specific mitomiR targetomes.

The whole targetomes (predicted and validated) of tissue-shared or tissue-specific mitomiRs were analyzed by the program DAVID. Significantly enriched pathways according to KEGG are showed for tissue-shared (A; top-20 p values), liver-specific (B), testis-specific (top-20 p values) and WAT-specific mitomiRs; “PI3K-Akt signaling pathway” was the only one significantly enriched for gastrocnemius-specific mitomiRs targetome; full lists are given as Supplementary material. Biological pathways were considered statistically significant if p value was less than 0.05 (Benjamini–Hochberg procedure for multiple correction).

Fig 7. Mitochondrial-related biological pathways potentially governed by tissue-shared and tissue-specific mitomiRs.

The targetomes of tissue-shared (A) and tissue-specific mitomiRs (B, gastrocnemius muscle; C, liver; D, testis; E, WAT) were further processed as described in Materials and Methods and then analyzed by MitoCarta3.0 for assignment to the seven broad functional categories relevant to mitochondria (“MitoPathways”).