MicroRNA-21 Regulates PI3K/Akt/mTOR Signaling by Targeting TGFβI during Skeletal Muscle Development in Pigs

Abstract

MicroRNAs (miRNAs), which are short (22-24 base pairs), non-coding RNAs, play critical roles in myogenesis. Using Solexa deep sequencing, we detected the expression levels of 229 and 209 miRNAs in swine skeletal muscle at 90 days post-coitus (E90) and 100 days postnatal (D100), respectively. A total of 138 miRNAs were up-regulated on E90, and 31 were up-regulated on D100. Of these, 9 miRNAs were selected for the validation of the small RNA libraries by quantitative RT-PCR (RT-qPCR). We found that miRNA-21 was down-regulated by 17-fold on D100 (P<0.001). Bioinformatics analysis suggested that the transforming growth factor beta-induced (TGFβI) gene was a potential target of miRNA-21. Both dual luciferase reporter assays and western blotting demonstrated that the TGFβI gene was regulated by miRNA-21. Co-expression analysis revealed that the mRNA expression levels of miRNA-21 and TGFβI were negatively correlated (r = -0.421, P = 0.026) in skeletal muscle during the 28 developmental stages. Our results revealed that more miRNAs are expressed in prenatal than in postnatal skeletal muscle. The miRNA-21 is a novel myogenic miRNA that is involved in skeletal muscle development and regulates PI3K/Akt/mTOR signaling by targeting the TGFβI gene.

Fig 1. Deep-sequencing results and annotations of small RNAs from swine skeletal muscle.

A. Size distribution of sequenced small RNAs from E90. B. Size distribution of sequenced small RNAs from D100. C. Annotations of sequenced small RNAs.

Fig 2. The expression of muscle-related miRNAs during the skeletal muscle development of Tongcheng pigs was detected using real-time PCR.

The miRNAs obtained from the longissimus dorsi of the Tongcheng pigs between E90 and D100 were evaluated. At least 3 animals were used for each time point, and the expression of the miRNAs was normalized to that of U6.

Fig 3. TGFβI is a target for miR-21.

A. The predicted binding site between miR-21 and the TGFBI 3' UTR. B. Detection of luciferase activity in the PIEC cells after co-transfection with the luciferase reporter and miR-21/NC. The PIEC cells were co-transfected with the psi-check2-TGFβI 3’ UTR and miR-21 mimics/negative control duplex. The relative luciferase activity was measured after 48 h. The data represent the mean ± SD from three independent experiments performed in duplicate. C. RT-qPCR was performed to verify the decreased TGFβI expression following the overexpression of miR-21 and enhanced expression after miR-21 inhibition in the PIEC cells. The data represent the mean ± SD from three independent experiments performed in duplicate. D. Total cell extracts were harvested at 48 h after transfection, and the TGFβI protein was analyzed via immunoblotting. E. All quantitative results were normalized byβ-actin and were shown as mean ± SD from three independent experiments.

Fig 4. Expression profiles of miR-21 and TGFβI in Tongcheng pigs.

A. Tissue distribution of miR-21. Total RNAs were extracted from the heart, liver, spleen, lung, kidney, small intestine, uterus, ovary and longissimus dorsi tissues. B. Expression profiles of miR-21 at different developmental stages. Total RNAs were extracted from the longissimus dorsi of the Tongcheng pigs at different dpcs (E) and postnatal days (D). The relative expression of miR-21 was compared to the internal control reference U6 and was analyzed using ΔΔCt. The data represent the mean ± SD from four independent experiments performed in duplicate. C. Expression profiles of TGFβI at different developmental stages. Analyses were performed according to B. D. Correlation between miR-21 and TGFβI expression during myogenesis in pigs. E. TGFβI protein levels were analyzed in extracts from the longissimus dorsi on E90 and D100 via immunoblotting. F. All quantitative results were normalized byβ-actin and were shown as mean ± SD from three independent experiments.

Fig 5. The potentiation of the Akt/mTOR pathway at E90 and D100.

A. To detect the activation of the PI3K/Akt/mTOR signaling pathway, E90 and D100 longissimus dorsi were prepared and resolved by 10% SDS-PAGE. The PVDF membrane was probed with antibodies to phosphor-Akt and phosphor-mTOR, respectively. Antibodies against Akt, mTOR and β-actin were used to verify the equal protein loading. B. All quantitative results were normalized byβ-actin and were shown as mean ± SD from three independent experiments.

Fig 6. Crosstalk between miR-21, the target gene TGFβI and the PI3K/Akt/mTOR signaling pathway in myogenesis.